From 6d8d2ecc8536cd46df3bbd747f651e1d4f64b930 Mon Sep 17 00:00:00 2001 From: Open vSwitch CI Date: Sep 21 2022 13:06:32 +0000 Subject: Import openvswitch2.17-2.17.0-44 from Fast DataPath --- diff --git a/.openvswitch.metadata b/.openvswitch.metadata index 9189888..aae65b4 100644 --- a/.openvswitch.metadata +++ b/.openvswitch.metadata @@ -1,6 +1,6 @@ 002450621b33c5690060345b0aac25bc2426d675 SOURCES/docutils-0.12.tar.gz 722b63cd114c21041abda7b38d7f14e46338e3e0 SOURCES/openvswitch-2.17.0.tar.gz +8509a716f9f936526f64fb23f313c5a9baf2f123 SOURCES/pyelftools-0.27.tar.gz d34f96421a86004aa5d26ecf975edefd09f948b1 SOURCES/Pygments-1.4.tar.gz 3a11f130c63b057532ca37fe49c8967d0cbae1d5 SOURCES/Sphinx-1.2.3.tar.gz 17331a86759beba4b6635ed530ce23b0b73c0744 SOURCES/dpdk-21.11.tar.xz -8509a716f9f936526f64fb23f313c5a9baf2f123 SOURCES/pyelftools-0.27.tar.gz diff --git a/SOURCES/openvswitch-2.17.0.patch b/SOURCES/openvswitch-2.17.0.patch index 193d288..f7c6c71 100644 --- a/SOURCES/openvswitch-2.17.0.patch +++ b/SOURCES/openvswitch-2.17.0.patch @@ -1,5 +1,5 @@ diff --git a/.ci/linux-build.sh b/.ci/linux-build.sh -index 6cd38ff3ef..c4ec93a398 100755 +index 6cd38ff3ef..2dabd3d0a4 100755 --- a/.ci/linux-build.sh +++ b/.ci/linux-build.sh @@ -220,7 +220,7 @@ fi @@ -11,8 +11,19 @@ index 6cd38ff3ef..c4ec93a398 100755 fi install_dpdk $DPDK_VER fi +@@ -244,9 +244,7 @@ fi + if [ "$ASAN" ]; then + # This will override default option configured in tests/atlocal.in. + export ASAN_OPTIONS='detect_leaks=1' +- # -O2 generates few false-positive memory leak reports in test-ovsdb +- # application, so lowering optimizations to -O1 here. +- CFLAGS_ASAN="-O1 -fno-omit-frame-pointer -fno-common -fsanitize=address" ++ CFLAGS_ASAN="-fno-omit-frame-pointer -fno-common -fsanitize=address" + CFLAGS_FOR_OVS="${CFLAGS_FOR_OVS} ${CFLAGS_ASAN}" + fi + diff --git a/.cirrus.yml b/.cirrus.yml -index a7ae793bc4..a4d2a5bbcd 100644 +index a7ae793bc4..29be50029b 100644 --- a/.cirrus.yml +++ b/.cirrus.yml @@ -2,8 +2,8 @@ freebsd_build_task: @@ -22,14 +33,45 @@ index a7ae793bc4..a4d2a5bbcd 100644 - image_family: freebsd-12-2-snap - image_family: freebsd-11-4-snap + image_family: freebsd-12-3-snap -+ image_family: freebsd-13-0-snap ++ image_family: freebsd-13-1-snap cpu: 4 memory: 4G +diff --git a/.github/workflows/build-and-test.yml b/.github/workflows/build-and-test.yml +index eac3504e48..6c9dddbb59 100644 +--- a/.github/workflows/build-and-test.yml ++++ b/.github/workflows/build-and-test.yml +@@ -6,7 +6,7 @@ jobs: + build-linux: + env: + dependencies: | +- automake libtool gcc bc libjemalloc1 libjemalloc-dev \ ++ automake libtool gcc bc libjemalloc2 libjemalloc-dev \ + libssl-dev llvm-dev libelf-dev libnuma-dev libpcap-dev \ + ninja-build selinux-policy-dev + deb_dependencies: | +@@ -25,7 +25,7 @@ jobs: + TESTSUITE: ${{ matrix.testsuite }} + + name: linux ${{ join(matrix.*, ' ') }} +- runs-on: ubuntu-18.04 ++ runs-on: ubuntu-20.04 + timeout-minutes: 30 + + strategy: diff --git a/Documentation/faq/releases.rst b/Documentation/faq/releases.rst -index af524251ff..319ee38c7d 100644 +index af524251ff..33a0d5d2d4 100644 --- a/Documentation/faq/releases.rst +++ b/Documentation/faq/releases.rst +@@ -32,7 +32,7 @@ Q: What does it mean for an Open vSwitch release to be LTS (long-term support)? + If a significant bug is identified in an LTS release, we will provide an + updated release that includes the fix. Releases that are not LTS may not + be fixed and may just be supplanted by the next major release. The current +- LTS release is 2.13.x. ++ LTS release is 2.17.x. + + For more information on the Open vSwitch release process, refer to + :doc:`/internals/release-process`. @@ -208,9 +208,9 @@ Q: What DPDK version does each Open vSwitch release work with? 2.12.x 18.11.9 2.13.x 19.11.10 @@ -121,9 +163,58 @@ index c10e9bfacc..7c71284f97 100644 --------------------- - Userspace datapath: diff --git a/acinclude.m4 b/acinclude.m4 -index 0c360fd1ef..61e88105f5 100644 +index 0c360fd1ef..c981f90bc7 100644 --- a/acinclude.m4 +++ b/acinclude.m4 +@@ -19,7 +19,7 @@ dnl This enables automatically running all unit tests with all MFEX + dnl implementations. + AC_DEFUN([OVS_CHECK_MFEX_AUTOVALIDATOR], [ + AC_ARG_ENABLE([mfex-default-autovalidator], +- [AC_HELP_STRING([--enable-mfex-default-autovalidator], ++ [AS_HELP_STRING([--enable-mfex-default-autovalidator], + [Enable MFEX autovalidator as default + miniflow_extract implementation.])], + [autovalidator=yes],[autovalidator=no]) +@@ -38,7 +38,7 @@ dnl This enables automatically running all unit tests with all DPCLS + dnl implementations. + AC_DEFUN([OVS_CHECK_DPCLS_AUTOVALIDATOR], [ + AC_ARG_ENABLE([autovalidator], +- [AC_HELP_STRING([--enable-autovalidator], ++ [AS_HELP_STRING([--enable-autovalidator], + [Enable DPCLS autovalidator as default subtable + search implementation.])], + [autovalidator=yes],[autovalidator=no]) +@@ -57,7 +57,7 @@ dnl Set OVS DPIF default implementation at configure time for running the unit + dnl tests on the whole codebase without modifying tests per DPIF impl + AC_DEFUN([OVS_CHECK_DPIF_AVX512_DEFAULT], [ + AC_ARG_ENABLE([dpif-default-avx512], +- [AC_HELP_STRING([--enable-dpif-default-avx512], ++ [AS_HELP_STRING([--enable-dpif-default-avx512], + [Enable DPIF AVX512 implementation as default.])], + [dpifavx512=yes],[dpifavx512=no]) + AC_MSG_CHECKING([whether DPIF AVX512 is default implementation]) +@@ -89,7 +89,7 @@ dnl OVS_ENABLE_WERROR + AC_DEFUN([OVS_ENABLE_WERROR], + [AC_ARG_ENABLE( + [Werror], +- [AC_HELP_STRING([--enable-Werror], [Add -Werror to CFLAGS])], ++ [AS_HELP_STRING([--enable-Werror], [Add -Werror to CFLAGS])], + [], [enable_Werror=no]) + AC_CONFIG_COMMANDS_PRE( + [if test "X$enable_Werror" = Xyes; then +@@ -118,10 +118,10 @@ dnl + dnl Configure linux kernel source tree + AC_DEFUN([OVS_CHECK_LINUX], [ + AC_ARG_WITH([linux], +- [AC_HELP_STRING([--with-linux=/path/to/linux], ++ [AS_HELP_STRING([--with-linux=/path/to/linux], + [Specify the Linux kernel build directory])]) + AC_ARG_WITH([linux-source], +- [AC_HELP_STRING([--with-linux-source=/path/to/linux-source], ++ [AS_HELP_STRING([--with-linux-source=/path/to/linux-source], + [Specify the Linux kernel source directory + (usually figured out automatically from build + directory)])]) @@ -305,6 +305,13 @@ AC_DEFUN([OVS_CHECK_LINUX_TC], [ ])], [AC_DEFINE([HAVE_TCA_SKBEDIT_FLAGS], [1], @@ -138,7 +229,25 @@ index 0c360fd1ef..61e88105f5 100644 ]) dnl OVS_CHECK_LINUX_SCTP_CT -@@ -1424,7 +1431,7 @@ AC_DEFUN([OVS_ENABLE_SPARSE], +@@ -348,7 +355,7 @@ dnl + dnl Check both Linux kernel AF_XDP and libbpf support + AC_DEFUN([OVS_CHECK_LINUX_AF_XDP], [ + AC_ARG_ENABLE([afxdp], +- [AC_HELP_STRING([--enable-afxdp], [Enable AF-XDP support])], ++ [AS_HELP_STRING([--enable-afxdp], [Enable AF-XDP support])], + [], [enable_afxdp=no]) + AC_MSG_CHECKING([whether AF_XDP is enabled]) + if test "$enable_afxdp" != yes; then +@@ -390,7 +397,7 @@ dnl + dnl Configure DPDK source tree + AC_DEFUN([OVS_CHECK_DPDK], [ + AC_ARG_WITH([dpdk], +- [AC_HELP_STRING([--with-dpdk=static|shared|yes], ++ [AS_HELP_STRING([--with-dpdk=static|shared|yes], + [Specify "static" or "shared" depending on the + DPDK libraries to use])], + [have_dpdk=true]) +@@ -1424,11 +1431,11 @@ AC_DEFUN([OVS_ENABLE_SPARSE], : ${SPARSE=sparse} AC_SUBST([SPARSE]) AC_CONFIG_COMMANDS_PRE( @@ -147,6 +256,11 @@ index 0c360fd1ef..61e88105f5 100644 AC_ARG_ENABLE( [sparse], +- [AC_HELP_STRING([--enable-sparse], [Run "sparse" by default])], ++ [AS_HELP_STRING([--enable-sparse], [Run "sparse" by default])], + [], [enable_sparse=no]) + AM_CONDITIONAL([ENABLE_SPARSE_BY_DEFAULT], [test $enable_sparse = yes])]) + diff --git a/configure.ac b/configure.ac index 4e9bcce272..5cc3f4801e 100644 --- a/configure.ac @@ -363,6 +477,18 @@ index 3e0d3a66e3..5ddd655d6c 100644 openvswitch (2.17.0-1) unstable; urgency=low * New upstream version +diff --git a/debian/control b/debian/control +index 6420b9d3e2..27359a297d 100644 +--- a/debian/control ++++ b/debian/control +@@ -9,6 +9,7 @@ Build-Depends: graphviz, + bzip2, + debhelper (>= 8), + dh-autoreconf, ++ dh-python, + libssl-dev, + libtool, + openssl, diff --git a/debian/openvswitch-switch.install b/debian/openvswitch-switch.install index 6a6e9a5435..5ac3df77b1 100644 --- a/debian/openvswitch-switch.install @@ -384,8829 +510,60873 @@ index 7fd7bc55da..088734b0dc 100644 debian/tmp/usr/share/man/man8/ovs-ctl.8 utilities/ovs-dpctl-top.8 utilities/ovs-dpctl.8 -diff --git a/dpdk/lib/vhost/vhost_user.c b/dpdk/lib/vhost/vhost_user.c -index a781346c4d..550b0ee8b5 100644 ---- a/dpdk/lib/vhost/vhost_user.c -+++ b/dpdk/lib/vhost/vhost_user.c -@@ -1603,6 +1603,9 @@ vhost_user_get_inflight_fd(struct virtio_net **pdev, - int numa_node = SOCKET_ID_ANY; - void *addr; +diff --git a/dpdk/MAINTAINERS b/dpdk/MAINTAINERS +index 18d9edaf88..84d8e261d5 100644 +--- a/dpdk/MAINTAINERS ++++ b/dpdk/MAINTAINERS +@@ -64,6 +64,8 @@ T: git://dpdk.org/next/dpdk-next-eventdev + Stable Branches + M: Luca Boccassi + M: Kevin Traynor ++M: Christian Ehrhardt ++M: Xueming Li + T: git://dpdk.org/dpdk-stable + + Security Issues +diff --git a/dpdk/VERSION b/dpdk/VERSION +index b570734337..63f795c0f8 100644 +--- a/dpdk/VERSION ++++ b/dpdk/VERSION +@@ -1 +1 @@ +-21.11.0 ++21.11.2 +diff --git a/dpdk/app/dumpcap/main.c b/dpdk/app/dumpcap/main.c +index c5fe440302..536ec64711 100644 +--- a/dpdk/app/dumpcap/main.c ++++ b/dpdk/app/dumpcap/main.c +@@ -679,8 +679,13 @@ static void enable_pdump(struct rte_ring *r, struct rte_mempool *mp) + flags |= RTE_PDUMP_FLAG_PCAPNG; + + TAILQ_FOREACH(intf, &interfaces, next) { +- if (promiscuous_mode) +- rte_eth_promiscuous_enable(intf->port); ++ if (promiscuous_mode) { ++ ret = rte_eth_promiscuous_enable(intf->port); ++ if (ret != 0) ++ fprintf(stderr, ++ "port %u set promiscuous enable failed: %d\n", ++ intf->port, ret); ++ } + + ret = rte_pdump_enable_bpf(intf->port, RTE_PDUMP_ALL_QUEUES, + flags, snaplen, +diff --git a/dpdk/app/pdump/main.c b/dpdk/app/pdump/main.c +index 46f9d25db0..101ac7db9a 100644 +--- a/dpdk/app/pdump/main.c ++++ b/dpdk/app/pdump/main.c +@@ -903,11 +903,21 @@ dump_packets_core(void *arg) + return 0; + } -+ if (validate_msg_fds(msg, 0) != 0) -+ return RTE_VHOST_MSG_RESULT_ERR; ++static unsigned int ++get_next_core(unsigned int lcore) ++{ ++ lcore = rte_get_next_lcore(lcore, 1, 0); ++ if (lcore == RTE_MAX_LCORE) ++ rte_exit(EXIT_FAILURE, ++ "Max core limit %u reached for packet capture", lcore); ++ return lcore; ++} + - if (msg->size != sizeof(msg->payload.inflight)) { - VHOST_LOG_CONFIG(ERR, - "invalid get_inflight_fd message size is %d\n", -@@ -1704,6 +1707,9 @@ vhost_user_set_inflight_fd(struct virtio_net **pdev, VhostUserMsg *msg, - int fd, i; - int numa_node = SOCKET_ID_ANY; + static inline void + dump_packets(void) + { + int i; +- uint32_t lcore_id = 0; ++ unsigned int lcore_id = 0; + + if (!multiple_core_capture) { + printf(" core (%u), capture for (%d) tuples\n", +@@ -933,12 +943,12 @@ dump_packets(void) + return; + } -+ if (validate_msg_fds(msg, 1) != 0) -+ return RTE_VHOST_MSG_RESULT_ERR; -+ - fd = msg->fds[0]; - if (msg->size != sizeof(msg->payload.inflight) || fd < 0) { - VHOST_LOG_CONFIG(ERR, -@@ -2873,6 +2879,9 @@ vhost_user_check_and_alloc_queue_pair(struct virtio_net *dev, - case VHOST_USER_SET_VRING_ADDR: - vring_idx = msg->payload.addr.index; - break; -+ case VHOST_USER_SET_INFLIGHT_FD: -+ vring_idx = msg->payload.inflight.num_queues - 1; -+ break; - default: - return 0; +- lcore_id = rte_get_next_lcore(lcore_id, 1, 0); ++ lcore_id = get_next_core(lcore_id); + + for (i = 0; i < num_tuples; i++) { + rte_eal_remote_launch(dump_packets_core, + &pdump_t[i], lcore_id); +- lcore_id = rte_get_next_lcore(lcore_id, 1, 0); ++ lcore_id = get_next_core(lcore_id); + + if (rte_eal_wait_lcore(lcore_id) < 0) + rte_exit(EXIT_FAILURE, "failed to wait\n"); +diff --git a/dpdk/app/proc-info/main.c b/dpdk/app/proc-info/main.c +index ce140aaf84..e1ccdbbaa5 100644 +--- a/dpdk/app/proc-info/main.c ++++ b/dpdk/app/proc-info/main.c +@@ -630,7 +630,7 @@ metrics_display(int port_id) + + names = rte_malloc(NULL, sizeof(struct rte_metric_name) * len, 0); + if (names == NULL) { +- printf("Cannot allocate memory for metrcis names\n"); ++ printf("Cannot allocate memory for metrics names\n"); + rte_free(metrics); + return; } -diff --git a/include/linux/automake.mk b/include/linux/automake.mk -index 8f063f482e..f857c7e088 100644 ---- a/include/linux/automake.mk -+++ b/include/linux/automake.mk -@@ -2,6 +2,7 @@ noinst_HEADERS += \ - include/linux/netlink.h \ - include/linux/netfilter/nf_conntrack_sctp.h \ - include/linux/pkt_cls.h \ -+ include/linux/gen_stats.h \ - include/linux/tc_act/tc_mpls.h \ - include/linux/tc_act/tc_pedit.h \ - include/linux/tc_act/tc_skbedit.h \ -diff --git a/include/linux/gen_stats.h b/include/linux/gen_stats.h -new file mode 100644 -index 0000000000..6fae6f727c ---- /dev/null -+++ b/include/linux/gen_stats.h -@@ -0,0 +1,81 @@ -+/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ -+#ifndef __LINUX_GEN_STATS_WRAPPER_H -+#define __LINUX_GEN_STATS_WRAPPER_H 1 +@@ -1109,7 +1109,7 @@ show_tm(void) + caplevel.n_nodes_max, + caplevel.n_nodes_nonleaf_max, + caplevel.n_nodes_leaf_max); +- printf("\t -- indetical: non leaf %u leaf %u\n", ++ printf("\t -- identical: non leaf %u leaf %u\n", + caplevel.non_leaf_nodes_identical, + caplevel.leaf_nodes_identical); + +@@ -1263,7 +1263,7 @@ show_ring(char *name) + printf(" - Name (%s) on socket (%d)\n" + " - flags:\n" + "\t -- Single Producer Enqueue (%u)\n" +- "\t -- Single Consmer Dequeue (%u)\n", ++ "\t -- Single Consumer Dequeue (%u)\n", + ptr->name, + ptr->memzone->socket_id, + ptr->flags & RING_F_SP_ENQ, +@@ -1504,10 +1504,10 @@ main(int argc, char **argv) + if (nb_ports == 0) + rte_exit(EXIT_FAILURE, "No Ethernet ports - bye\n"); + +- /* If no port mask was specified, then show non-owned ports */ ++ /* If no port mask was specified, then show all non-owned ports */ + if (enabled_port_mask == 0) { + RTE_ETH_FOREACH_DEV(i) +- enabled_port_mask = 1ul << i; ++ enabled_port_mask |= 1ul << i; + } + + for (i = 0; i < RTE_MAX_ETHPORTS; i++) { +diff --git a/dpdk/app/test-acl/main.c b/dpdk/app/test-acl/main.c +index c2de18770d..06e3847ab9 100644 +--- a/dpdk/app/test-acl/main.c ++++ b/dpdk/app/test-acl/main.c +@@ -386,8 +386,8 @@ parse_cb_ipv4_trace(char *str, struct ipv4_5tuple *v) + } + + /* +- * Parses IPV6 address, exepcts the following format: +- * XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX (where X - is a hexedecimal digit). ++ * Parse IPv6 address, expects the following format: ++ * XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX (where X is a hexadecimal digit). + */ + static int + parse_ipv6_addr(const char *in, const char **end, uint32_t v[IPV6_ADDR_U32], +@@ -994,7 +994,7 @@ print_usage(const char *prgname) + "should be either 1 or multiple of %zu, " + "but not greater then %u]\n" + "[--" OPT_MAX_SIZE +- "= " ++ "= " + "leave 0 for default behaviour]\n" + "[--" OPT_ITER_NUM "=]\n" + "[--" OPT_VERBOSE "=]\n" +diff --git a/dpdk/app/test-compress-perf/comp_perf_test_cyclecount.c b/dpdk/app/test-compress-perf/comp_perf_test_cyclecount.c +index da55b02b74..a3f6404eb2 100644 +--- a/dpdk/app/test-compress-perf/comp_perf_test_cyclecount.c ++++ b/dpdk/app/test-compress-perf/comp_perf_test_cyclecount.c +@@ -175,16 +175,17 @@ main_loop(struct cperf_cyclecount_ctx *ctx, enum rte_comp_xform_type type) + + /* one array for both enqueue and dequeue */ + ops = rte_zmalloc_socket(NULL, +- 2 * mem->total_bufs * sizeof(struct rte_comp_op *), ++ (test_data->burst_sz + mem->total_bufs) * ++ sizeof(struct rte_comp_op *), + 0, rte_socket_id()); + + if (ops == NULL) { + RTE_LOG(ERR, USER1, +- "Can't allocate memory for ops strucures\n"); ++ "Can't allocate memory for ops structures\n"); + return -1; + } + +- deq_ops = &ops[mem->total_bufs]; ++ deq_ops = &ops[test_data->burst_sz]; + + if (type == RTE_COMP_COMPRESS) { + xform = (struct rte_comp_xform) { +@@ -273,7 +274,7 @@ main_loop(struct cperf_cyclecount_ctx *ctx, enum rte_comp_xform_type type) + /* Allocate compression operations */ + if (ops_needed && rte_mempool_get_bulk( + mem->op_pool, +- (void **)ops, ++ (void **)&ops[ops_unused], + ops_needed) != 0) { + RTE_LOG(ERR, USER1, + "Could not allocate enough operations\n"); +diff --git a/dpdk/app/test-compress-perf/comp_perf_test_throughput.c b/dpdk/app/test-compress-perf/comp_perf_test_throughput.c +index d3dff070b0..4569599eb9 100644 +--- a/dpdk/app/test-compress-perf/comp_perf_test_throughput.c ++++ b/dpdk/app/test-compress-perf/comp_perf_test_throughput.c +@@ -72,7 +72,7 @@ main_loop(struct cperf_benchmark_ctx *ctx, enum rte_comp_xform_type type) + + if (ops == NULL) { + RTE_LOG(ERR, USER1, +- "Can't allocate memory for ops strucures\n"); ++ "Can't allocate memory for ops structures\n"); + return -1; + } + +diff --git a/dpdk/app/test-compress-perf/comp_perf_test_verify.c b/dpdk/app/test-compress-perf/comp_perf_test_verify.c +index f6e21368e8..7d06029488 100644 +--- a/dpdk/app/test-compress-perf/comp_perf_test_verify.c ++++ b/dpdk/app/test-compress-perf/comp_perf_test_verify.c +@@ -75,7 +75,7 @@ main_loop(struct cperf_verify_ctx *ctx, enum rte_comp_xform_type type) + + if (ops == NULL) { + RTE_LOG(ERR, USER1, +- "Can't allocate memory for ops strucures\n"); ++ "Can't allocate memory for ops structures\n"); + return -1; + } + +diff --git a/dpdk/app/test-compress-perf/main.c b/dpdk/app/test-compress-perf/main.c +index cc9951a9b1..ce9e80bedc 100644 +--- a/dpdk/app/test-compress-perf/main.c ++++ b/dpdk/app/test-compress-perf/main.c +@@ -67,7 +67,7 @@ comp_perf_check_capabilities(struct comp_test_data *test_data, uint8_t cdev_id) + + uint64_t comp_flags = cap->comp_feature_flags; + +- /* Huffman enconding */ ++ /* Huffman encoding */ + if (test_data->huffman_enc == RTE_COMP_HUFFMAN_FIXED && + (comp_flags & RTE_COMP_FF_HUFFMAN_FIXED) == 0) { + RTE_LOG(ERR, USER1, +@@ -168,7 +168,7 @@ comp_perf_initialize_compressdev(struct comp_test_data *test_data, + cdev_id = enabled_cdevs[i]; + + struct rte_compressdev_info cdev_info; +- uint8_t socket_id = rte_compressdev_socket_id(cdev_id); ++ int socket_id = rte_compressdev_socket_id(cdev_id); + + rte_compressdev_info_get(cdev_id, &cdev_info); + if (cdev_info.max_nb_queue_pairs && +@@ -194,6 +194,7 @@ comp_perf_initialize_compressdev(struct comp_test_data *test_data, + .max_nb_priv_xforms = NUM_MAX_XFORMS, + .max_nb_streams = 0 + }; ++ test_data->nb_qps = config.nb_queue_pairs; + + if (rte_compressdev_configure(cdev_id, &config) < 0) { + RTE_LOG(ERR, USER1, "Device configuration failed\n"); +diff --git a/dpdk/app/test-crypto-perf/cperf_test_pmd_cyclecount.c b/dpdk/app/test-crypto-perf/cperf_test_pmd_cyclecount.c +index ba1f104f72..5842f29d43 100644 +--- a/dpdk/app/test-crypto-perf/cperf_test_pmd_cyclecount.c ++++ b/dpdk/app/test-crypto-perf/cperf_test_pmd_cyclecount.c +@@ -334,7 +334,7 @@ pmd_cyclecount_bench_burst_sz( + * queue, so we never get any failed enqs unless the driver won't accept + * the exact number of descriptors we requested, or the driver won't + * wrap around the end of the TX ring. However, since we're only +- * dequeueing once we've filled up the queue, we have to benchmark it ++ * dequeuing once we've filled up the queue, we have to benchmark it + * piecemeal and then average out the results. + */ + cur_op = 0; +diff --git a/dpdk/app/test-eventdev/evt_options.c b/dpdk/app/test-eventdev/evt_options.c +index 753a7dbd7d..4ae44801da 100644 +--- a/dpdk/app/test-eventdev/evt_options.c ++++ b/dpdk/app/test-eventdev/evt_options.c +@@ -336,7 +336,7 @@ usage(char *program) + "\t--deq_tmo_nsec : global dequeue timeout\n" + "\t--prod_type_ethdev : use ethernet device as producer.\n" + "\t--prod_type_timerdev : use event timer device as producer.\n" +- "\t expity_nsec would be the timeout\n" ++ "\t expiry_nsec would be the timeout\n" + "\t in ns.\n" + "\t--prod_type_timerdev_burst : use timer device as producer\n" + "\t burst mode.\n" +diff --git a/dpdk/app/test-eventdev/test_order_common.c b/dpdk/app/test-eventdev/test_order_common.c +index ff7813f9c2..603e7c9178 100644 +--- a/dpdk/app/test-eventdev/test_order_common.c ++++ b/dpdk/app/test-eventdev/test_order_common.c +@@ -253,7 +253,7 @@ void + order_opt_dump(struct evt_options *opt) + { + evt_dump_producer_lcores(opt); +- evt_dump("nb_wrker_lcores", "%d", evt_nr_active_lcores(opt->wlcores)); ++ evt_dump("nb_worker_lcores", "%d", evt_nr_active_lcores(opt->wlcores)); + evt_dump_worker_lcores(opt); + evt_dump("nb_evdev_ports", "%d", order_nb_event_ports(opt)); + } +diff --git a/dpdk/app/test-fib/main.c b/dpdk/app/test-fib/main.c +index ecd420116a..830c32cc44 100644 +--- a/dpdk/app/test-fib/main.c ++++ b/dpdk/app/test-fib/main.c +@@ -624,7 +624,7 @@ print_usage(void) + "(if -f is not specified)>]\n" + "[-r ]\n" +- "[-c ]\n" ++ "[-c ]\n" + "[-6 ]\n" + "[-s ]\n" + "[-a ]\n" + "[-w ]\n" + "[-u ]\n" +- "[-v ]\n", +@@ -711,6 +711,10 @@ parse_opts(int argc, char **argv) + print_usage(); + rte_exit(-EINVAL, "Invalid option -n\n"); + } ++ ++ if (config.nb_routes < config.print_fract) ++ config.print_fract = config.nb_routes; ++ + break; + case 'd': + distrib_string = optarg; +@@ -1242,6 +1246,10 @@ main(int argc, char **argv) + config.nb_routes = 0; + while (fgets(line, sizeof(line), fr) != NULL) + config.nb_routes++; ++ ++ if (config.nb_routes < config.print_fract) ++ config.print_fract = config.nb_routes; ++ + rewind(fr); + } + +diff --git a/dpdk/app/test-flow-perf/config.h b/dpdk/app/test-flow-perf/config.h +index 0db2254bd1..29b63298e0 100644 +--- a/dpdk/app/test-flow-perf/config.h ++++ b/dpdk/app/test-flow-perf/config.h +@@ -28,7 +28,7 @@ + #define PORT_ID_DST 1 + #define TEID_VALUE 1 + +-/* Flow items/acctions max size */ ++/* Flow items/actions max size */ + #define MAX_ITEMS_NUM 32 + #define MAX_ACTIONS_NUM 32 + #define MAX_ATTRS_NUM 16 +diff --git a/dpdk/app/test-flow-perf/main.c b/dpdk/app/test-flow-perf/main.c +index 11f1ee0e1e..f375097028 100644 +--- a/dpdk/app/test-flow-perf/main.c ++++ b/dpdk/app/test-flow-perf/main.c +@@ -16,6 +16,7 @@ + * gives packet per second measurement. + */ + ++#include + #include + #include + #include +@@ -1519,7 +1520,7 @@ dump_used_cpu_time(const char *item, + * threads time. + * + * Throughput: total count of rte rules divided +- * over the average of the time cosumed by all ++ * over the average of the time consumed by all + * threads time. + */ + double insertion_latency_time; +@@ -1713,36 +1714,6 @@ do_tx(struct lcore_info *li, uint16_t cnt, uint16_t tx_port, + rte_pktmbuf_free(li->pkts[i]); + } + +-/* +- * Method to convert numbers into pretty numbers that easy +- * to read. The design here is to add comma after each three +- * digits and set all of this inside buffer. +- * +- * For example if n = 1799321, the output will be +- * 1,799,321 after this method which is easier to read. +- */ +-static char * +-pretty_number(uint64_t n, char *buf) +-{ +- char p[6][4]; +- int i = 0; +- int off = 0; +- +- while (n > 1000) { +- sprintf(p[i], "%03d", (int)(n % 1000)); +- n /= 1000; +- i += 1; +- } +- +- sprintf(p[i++], "%d", (int)n); +- +- while (i--) +- off += sprintf(buf + off, "%s,", p[i]); +- buf[strlen(buf) - 1] = '\0'; +- +- return buf; +-} +- + static void + packet_per_second_stats(void) + { +@@ -1764,7 +1735,6 @@ packet_per_second_stats(void) + uint64_t total_rx_pkts = 0; + uint64_t total_tx_drops = 0; + uint64_t tx_delta, rx_delta, drops_delta; +- char buf[3][32]; + int nr_valid_core = 0; + + sleep(1); +@@ -1789,10 +1759,8 @@ packet_per_second_stats(void) + tx_delta = li->tx_pkts - oli->tx_pkts; + rx_delta = li->rx_pkts - oli->rx_pkts; + drops_delta = li->tx_drops - oli->tx_drops; +- printf("%6d %16s %16s %16s\n", i, +- pretty_number(tx_delta, buf[0]), +- pretty_number(drops_delta, buf[1]), +- pretty_number(rx_delta, buf[2])); ++ printf("%6d %'16"PRId64" %'16"PRId64" %'16"PRId64"\n", ++ i, tx_delta, drops_delta, rx_delta); + + total_tx_pkts += tx_delta; + total_rx_pkts += rx_delta; +@@ -1803,10 +1771,9 @@ packet_per_second_stats(void) + } + + if (nr_valid_core > 1) { +- printf("%6s %16s %16s %16s\n", "total", +- pretty_number(total_tx_pkts, buf[0]), +- pretty_number(total_tx_drops, buf[1]), +- pretty_number(total_rx_pkts, buf[2])); ++ printf("%6s %'16"PRId64" %'16"PRId64" %'16"PRId64"\n", ++ "total", total_tx_pkts, total_tx_drops, ++ total_rx_pkts); + nr_lines += 1; + } + +@@ -2139,6 +2106,9 @@ main(int argc, char **argv) + if (argc > 1) + args_parse(argc, argv); + ++ /* For more fancy, localised integer formatting. */ ++ setlocale(LC_NUMERIC, ""); ++ + init_port(); + + nb_lcores = rte_lcore_count(); +diff --git a/dpdk/app/test-pmd/5tswap.c b/dpdk/app/test-pmd/5tswap.c +index 629d3e0d31..f041a5e1d5 100644 +--- a/dpdk/app/test-pmd/5tswap.c ++++ b/dpdk/app/test-pmd/5tswap.c +@@ -185,9 +185,22 @@ pkt_burst_5tuple_swap(struct fwd_stream *fs) + get_end_cycles(fs, start_tsc); + } + ++static void ++stream_init_5tuple_swap(struct fwd_stream *fs) ++{ ++ bool rx_stopped, tx_stopped; + -+#if defined(__KERNEL__) || defined(HAVE_TCA_STATS_PKT64) -+#include_next -+#else -+#include ++ rx_stopped = ports[fs->rx_port].rxq[fs->rx_queue].state == ++ RTE_ETH_QUEUE_STATE_STOPPED; ++ tx_stopped = ports[fs->tx_port].txq[fs->tx_queue].state == ++ RTE_ETH_QUEUE_STATE_STOPPED; ++ fs->disabled = rx_stopped || tx_stopped; ++} + -+enum { -+ TCA_STATS_UNSPEC, -+ TCA_STATS_BASIC, -+ TCA_STATS_RATE_EST, -+ TCA_STATS_QUEUE, -+ TCA_STATS_APP, -+ TCA_STATS_RATE_EST64, -+ TCA_STATS_PAD, -+ TCA_STATS_BASIC_HW, -+ TCA_STATS_PKT64, -+ __TCA_STATS_MAX, -+}; -+#define TCA_STATS_MAX (__TCA_STATS_MAX - 1) + struct fwd_engine five_tuple_swap_fwd_engine = { + .fwd_mode_name = "5tswap", + .port_fwd_begin = NULL, + .port_fwd_end = NULL, ++ .stream_init = stream_init_5tuple_swap, + .packet_fwd = pkt_burst_5tuple_swap, + }; +diff --git a/dpdk/app/test-pmd/cmd_flex_item.c b/dpdk/app/test-pmd/cmd_flex_item.c +index 908bcb3f47..3e54724237 100644 +--- a/dpdk/app/test-pmd/cmd_flex_item.c ++++ b/dpdk/app/test-pmd/cmd_flex_item.c +@@ -20,6 +20,8 @@ + struct flex_item *flex_items[RTE_MAX_ETHPORTS][FLEX_MAX_PARSERS_NUM]; + struct flex_pattern flex_patterns[FLEX_MAX_PATTERNS_NUM]; + ++#ifdef RTE_HAS_JANSSON ++ + static struct flex_item * + flex_parser_fetch(uint16_t port_id, uint16_t flex_id) + { +@@ -34,7 +36,6 @@ flex_parser_fetch(uint16_t port_id, uint16_t flex_id) + return flex_items[port_id][flex_id]; + } + +-#ifdef RTE_HAS_JANSSON + static __rte_always_inline bool + match_strkey(const char *key, const char *pattern) + { +@@ -133,7 +134,8 @@ flex_link_item_parse(const char *src, struct rte_flow_item *item) + struct rte_flow_item *pattern; + struct rte_flow_action *actions; + +- sprintf(flow_rule, "flow create 0 pattern %s / end", src); ++ sprintf(flow_rule, ++ "flow create 0 pattern %s / end actions drop / end", src); + src = flow_rule; + ret = flow_parse(src, (void *)data, sizeof(data), + &attr, &pattern, &actions); +@@ -368,23 +370,12 @@ flex_item_create(portid_t port_id, uint16_t flex_id, const char *filename) + free(fp); + } + +-#else /* RTE_HAS_JANSSON */ +-void flex_item_create(__rte_unused portid_t port_id, +- __rte_unused uint16_t flex_id, +- __rte_unused const char *filename) +-{ +- printf("cannot create flex item - no JSON library configured\n"); +-} +-#endif /* RTE_HAS_JANSSON */ +- + void + flex_item_destroy(portid_t port_id, uint16_t flex_id) + { + int ret; + struct rte_flow_error error; + struct flex_item *fp = flex_parser_fetch(port_id, flex_id); +- if (!flex_id) +- return; + if (fp == FLEX_PARSER_ERR) { + printf("Bad parameters: port_id=%u flex_id=%u\n", + port_id, flex_id); +@@ -405,6 +396,22 @@ flex_item_destroy(portid_t port_id, uint16_t flex_id) + } + } + ++#else /* RTE_HAS_JANSSON */ ++void flex_item_create(__rte_unused portid_t port_id, ++ __rte_unused uint16_t flex_id, ++ __rte_unused const char *filename) ++{ ++ printf("cannot create flex item - no JSON library configured\n"); ++} + -+/** -+ * struct gnet_stats_basic - byte/packet throughput statistics -+ * @bytes: number of seen bytes -+ * @packets: number of seen packets -+ */ -+struct gnet_stats_basic { -+ __u64 bytes; -+ __u32 packets; -+}; ++void ++flex_item_destroy(__rte_unused portid_t port_id, __rte_unused uint16_t flex_id) ++{ + -+/** -+ * struct gnet_stats_rate_est - rate estimator -+ * @bps: current byte rate -+ * @pps: current packet rate -+ */ -+struct gnet_stats_rate_est { -+ __u32 bps; -+ __u32 pps; -+}; ++} + -+/** -+ * struct gnet_stats_rate_est64 - rate estimator -+ * @bps: current byte rate -+ * @pps: current packet rate -+ */ -+struct gnet_stats_rate_est64 { -+ __u64 bps; -+ __u64 pps; -+}; ++#endif /* RTE_HAS_JANSSON */ + -+/** -+ * struct gnet_stats_queue - queuing statistics -+ * @qlen: queue length -+ * @backlog: backlog size of queue -+ * @drops: number of dropped packets -+ * @requeues: number of requeues -+ * @overlimits: number of enqueues over the limit -+ */ -+struct gnet_stats_queue { -+ __u32 qlen; -+ __u32 backlog; -+ __u32 drops; -+ __u32 requeues; -+ __u32 overlimits; -+}; + void + port_flex_item_flush(portid_t port_id) + { +diff --git a/dpdk/app/test-pmd/cmdline.c b/dpdk/app/test-pmd/cmdline.c +index 6e10afeedd..d9bf0eb3b3 100644 +--- a/dpdk/app/test-pmd/cmdline.c ++++ b/dpdk/app/test-pmd/cmdline.c +@@ -561,7 +561,7 @@ static void cmd_help_long_parsed(void *parsed_result, + " Set the option to enable display of RX and TX bursts.\n" + + "set port (port_id) vf (vf_id) rx|tx on|off\n" +- " Enable/Disable a VF receive/tranmit from a port\n\n" ++ " Enable/Disable a VF receive/transmit from a port\n\n" + + "set port (port_id) vf (vf_id) rxmode (AUPE|ROPE|BAM" + "|MPE) (on|off)\n" +@@ -2045,10 +2045,6 @@ cmd_config_mtu_parsed(void *parsed_result, + { + struct cmd_config_mtu_result *res = parsed_result; + +- if (res->value < RTE_ETHER_MIN_LEN) { +- fprintf(stderr, "mtu cannot be less than %d\n", RTE_ETHER_MIN_LEN); +- return; +- } + port_mtu_set(res->port_id, res->value); + } + +@@ -2651,8 +2647,10 @@ cmd_config_rxtx_queue_parsed(void *parsed_result, + __rte_unused void *data) + { + struct cmd_config_rxtx_queue *res = parsed_result; ++ struct rte_port *port; + uint8_t isrx; + uint8_t isstart; ++ uint8_t *state; + int ret = 0; + + if (test_done == 0) { +@@ -2700,8 +2698,15 @@ cmd_config_rxtx_queue_parsed(void *parsed_result, + else + ret = rte_eth_dev_tx_queue_stop(res->portid, res->qid); + +- if (ret == -ENOTSUP) ++ if (ret == -ENOTSUP) { + fprintf(stderr, "Function not supported in PMD\n"); ++ return; ++ } ++ ++ port = &ports[res->portid]; ++ state = isrx ? &port->rxq[res->qid].state : &port->txq[res->qid].state; ++ *state = isstart ? RTE_ETH_QUEUE_STATE_STARTED : ++ RTE_ETH_QUEUE_STATE_STOPPED; + } + + cmdline_parse_token_string_t cmd_config_rxtx_queue_port = +@@ -2770,11 +2775,11 @@ cmd_config_deferred_start_rxtx_queue_parsed(void *parsed_result, + + ison = !strcmp(res->state, "on"); + +- if (isrx && port->rx_conf[res->qid].rx_deferred_start != ison) { +- port->rx_conf[res->qid].rx_deferred_start = ison; ++ if (isrx && port->rxq[res->qid].conf.rx_deferred_start != ison) { ++ port->rxq[res->qid].conf.rx_deferred_start = ison; + needreconfig = 1; +- } else if (!isrx && port->tx_conf[res->qid].tx_deferred_start != ison) { +- port->tx_conf[res->qid].tx_deferred_start = ison; ++ } else if (!isrx && port->txq[res->qid].conf.tx_deferred_start != ison) { ++ port->txq[res->qid].conf.tx_deferred_start = ison; + needreconfig = 1; + } + +@@ -2892,7 +2897,7 @@ cmd_setup_rxtx_queue_parsed( + res->qid, + port->nb_rx_desc[res->qid], + socket_id, +- &port->rx_conf[res->qid], ++ &port->rxq[res->qid].conf, + mp); + if (ret) + fprintf(stderr, "Failed to setup RX queue\n"); +@@ -2910,7 +2915,7 @@ cmd_setup_rxtx_queue_parsed( + res->qid, + port->nb_tx_desc[res->qid], + socket_id, +- &port->tx_conf[res->qid]); ++ &port->txq[res->qid].conf); + if (ret) + fprintf(stderr, "Failed to setup TX queue\n"); + } +@@ -3120,7 +3125,7 @@ showport_parse_reta_config(struct rte_eth_rss_reta_entry64 *conf, + return -1; + } + for (i = 0; i < ret; i++) +- conf[i].mask = (uint64_t)strtoul(str_fld[i], &end, 0); ++ conf[i].mask = (uint64_t)strtoull(str_fld[i], &end, 0); + + return 0; + } +@@ -4686,7 +4691,7 @@ cmd_config_queue_tx_offloads(struct rte_port *port) + + /* Apply queue tx offloads configuration */ + for (k = 0; k < port->dev_info.max_tx_queues; k++) +- port->tx_conf[k].offloads = ++ port->txq[k].conf.offloads = + port->dev_conf.txmode.offloads; + } + +@@ -5915,6 +5920,19 @@ static void cmd_set_bonding_mode_parsed(void *parsed_result, + { + struct cmd_set_bonding_mode_result *res = parsed_result; + portid_t port_id = res->port_id; ++ struct rte_port *port = &ports[port_id]; ++ ++ /* ++ * Bonding mode changed means resources of device changed, like whether ++ * started rte timer or not. Device should be restarted when resources ++ * of device changed. ++ */ ++ if (port->port_status != RTE_PORT_STOPPED) { ++ fprintf(stderr, ++ "\t Error: Can't set bonding mode when port %d is not stopped\n", ++ port_id); ++ return; ++ } + + /* Set the bonding mode for the relevant port. */ + if (0 != rte_eth_bond_mode_set(port_id, res->value)) +@@ -6651,6 +6669,7 @@ static void cmd_create_bonded_device_parsed(void *parsed_result, + "Failed to enable promiscuous mode for port %u: %s - ignore\n", + port_id, rte_strerror(-ret)); + ++ ports[port_id].bond_flag = 1; + ports[port_id].need_setup = 0; + ports[port_id].port_status = RTE_PORT_STOPPED; + } +@@ -8754,6 +8773,7 @@ static void cmd_quit_parsed(__rte_unused void *parsed_result, + __rte_unused void *data) + { + cmdline_quit(cl); ++ cl_quit = 1; + } + + cmdline_parse_token_string_t cmd_quit_quit = +@@ -9273,6 +9293,7 @@ cmd_set_vf_rxmode_parsed(void *parsed_result, + } + + RTE_SET_USED(is_on); ++ RTE_SET_USED(vf_rxmode); + + #ifdef RTE_NET_IXGBE + if (ret == -ENOTSUP) +@@ -16068,7 +16089,7 @@ cmd_rx_offload_get_configuration_parsed( + + nb_rx_queues = dev_info.nb_rx_queues; + for (q = 0; q < nb_rx_queues; q++) { +- queue_offloads = port->rx_conf[q].offloads; ++ queue_offloads = port->rxq[q].conf.offloads; + printf(" Queue[%2d] :", q); + print_rx_offloads(queue_offloads); + printf("\n"); +@@ -16188,11 +16209,11 @@ cmd_config_per_port_rx_offload_parsed(void *parsed_result, + if (!strcmp(res->on_off, "on")) { + port->dev_conf.rxmode.offloads |= single_offload; + for (q = 0; q < nb_rx_queues; q++) +- port->rx_conf[q].offloads |= single_offload; ++ port->rxq[q].conf.offloads |= single_offload; + } else { + port->dev_conf.rxmode.offloads &= ~single_offload; + for (q = 0; q < nb_rx_queues; q++) +- port->rx_conf[q].offloads &= ~single_offload; ++ port->rxq[q].conf.offloads &= ~single_offload; + } + + cmd_reconfig_device_queue(port_id, 1, 1); +@@ -16298,9 +16319,9 @@ cmd_config_per_queue_rx_offload_parsed(void *parsed_result, + } + + if (!strcmp(res->on_off, "on")) +- port->rx_conf[queue_id].offloads |= single_offload; ++ port->rxq[queue_id].conf.offloads |= single_offload; + else +- port->rx_conf[queue_id].offloads &= ~single_offload; ++ port->rxq[queue_id].conf.offloads &= ~single_offload; + + cmd_reconfig_device_queue(port_id, 1, 1); + } +@@ -16487,7 +16508,7 @@ cmd_tx_offload_get_configuration_parsed( + + nb_tx_queues = dev_info.nb_tx_queues; + for (q = 0; q < nb_tx_queues; q++) { +- queue_offloads = port->tx_conf[q].offloads; ++ queue_offloads = port->txq[q].conf.offloads; + printf(" Queue[%2d] :", q); + print_tx_offloads(queue_offloads); + printf("\n"); +@@ -16611,11 +16632,11 @@ cmd_config_per_port_tx_offload_parsed(void *parsed_result, + if (!strcmp(res->on_off, "on")) { + port->dev_conf.txmode.offloads |= single_offload; + for (q = 0; q < nb_tx_queues; q++) +- port->tx_conf[q].offloads |= single_offload; ++ port->txq[q].conf.offloads |= single_offload; + } else { + port->dev_conf.txmode.offloads &= ~single_offload; + for (q = 0; q < nb_tx_queues; q++) +- port->tx_conf[q].offloads &= ~single_offload; ++ port->txq[q].conf.offloads &= ~single_offload; + } + + cmd_reconfig_device_queue(port_id, 1, 1); +@@ -16724,9 +16745,9 @@ cmd_config_per_queue_tx_offload_parsed(void *parsed_result, + } + + if (!strcmp(res->on_off, "on")) +- port->tx_conf[queue_id].offloads |= single_offload; ++ port->txq[queue_id].conf.offloads |= single_offload; + else +- port->tx_conf[queue_id].offloads &= ~single_offload; ++ port->txq[queue_id].conf.offloads &= ~single_offload; + + cmd_reconfig_device_queue(port_id, 1, 1); + } +@@ -17829,6 +17850,7 @@ cmdline_parse_ctx_t main_ctx[] = { + (cmdline_parse_inst_t *)&cmd_show_port_meter_cap, + (cmdline_parse_inst_t *)&cmd_add_port_meter_profile_srtcm, + (cmdline_parse_inst_t *)&cmd_add_port_meter_profile_trtcm, ++ (cmdline_parse_inst_t *)&cmd_add_port_meter_profile_trtcm_rfc4115, + (cmdline_parse_inst_t *)&cmd_del_port_meter_profile, + (cmdline_parse_inst_t *)&cmd_create_port_meter, + (cmdline_parse_inst_t *)&cmd_enable_port_meter, +diff --git a/dpdk/app/test-pmd/cmdline_flow.c b/dpdk/app/test-pmd/cmdline_flow.c +index bbe3dc0115..5c4544a753 100644 +--- a/dpdk/app/test-pmd/cmdline_flow.c ++++ b/dpdk/app/test-pmd/cmdline_flow.c +@@ -2162,7 +2162,7 @@ static const struct token token_list[] = { + }, + [COMMON_POLICY_ID] = { + .name = "{policy_id}", +- .type = "POLCIY_ID", ++ .type = "POLICY_ID", + .help = "policy id", + .call = parse_int, + .comp = comp_none, +@@ -2370,7 +2370,7 @@ static const struct token token_list[] = { + }, + [TUNNEL_DESTROY] = { + .name = "destroy", +- .help = "destroy tunel", ++ .help = "destroy tunnel", + .next = NEXT(NEXT_ENTRY(TUNNEL_DESTROY_ID), + NEXT_ENTRY(COMMON_PORT_ID)), + .args = ARGS(ARGS_ENTRY(struct buffer, port)), +@@ -2378,7 +2378,7 @@ static const struct token token_list[] = { + }, + [TUNNEL_DESTROY_ID] = { + .name = "id", +- .help = "tunnel identifier to testroy", ++ .help = "tunnel identifier to destroy", + .next = NEXT(NEXT_ENTRY(COMMON_UNSIGNED)), + .args = ARGS(ARGS_ENTRY(struct tunnel_ops, id)), + .call = parse_tunnel, +@@ -7702,16 +7702,14 @@ parse_string(struct context *ctx, const struct token *token, + static int + parse_hex_string(const char *src, uint8_t *dst, uint32_t *size) + { +- uint32_t left = *size; + const uint8_t *head = dst; ++ uint32_t left; + +- /* Check input parameters */ +- if ((src == NULL) || +- (dst == NULL) || +- (size == NULL) || +- (*size == 0)) ++ if (*size == 0) + return -1; + ++ left = *size; ++ + /* Convert chars to bytes */ + while (left) { + char tmp[3], *end = tmp; +@@ -9153,7 +9151,8 @@ cmd_set_raw_parsed(const struct buffer *in) + case RTE_FLOW_ITEM_TYPE_GENEVE_OPT: + opt = (const struct rte_flow_item_geneve_opt *) + item->spec; +- size = offsetof(struct rte_flow_item_geneve_opt, data); ++ size = offsetof(struct rte_flow_item_geneve_opt, ++ option_len) + sizeof(uint8_t); + if (opt->option_len && opt->data) { + *total_size += opt->option_len * + sizeof(uint32_t); +@@ -9210,19 +9209,15 @@ cmd_set_raw_parsed(const struct buffer *in) + } else { + const struct rte_flow_item_gtp_psc + *opt = item->spec; +- struct { +- uint8_t len; +- uint8_t pdu_type:4; +- uint8_t qfi:6; +- uint8_t next; +- } psc; +- psc.len = sizeof(psc) / 4; +- psc.pdu_type = opt->hdr.type; +- psc.qfi = opt->hdr.qfi; +- psc.next = 0; +- *total_size += sizeof(psc); +- rte_memcpy(data_tail - (*total_size), +- &psc, sizeof(psc)); ++ struct rte_gtp_psc_generic_hdr *hdr; ++ size_t hdr_size = RTE_ALIGN(sizeof(*hdr), ++ sizeof(int32_t)); ++ ++ *total_size += hdr_size; ++ hdr = (typeof(hdr))(data_tail - (*total_size)); ++ memset(hdr, 0, hdr_size); ++ *hdr = opt->hdr; ++ hdr->ext_hdr_len = 1; + gtp_psc = i; + size = 0; + } +diff --git a/dpdk/app/test-pmd/cmdline_mtr.c b/dpdk/app/test-pmd/cmdline_mtr.c +index ad7ef6ad98..a0f885f190 100644 +--- a/dpdk/app/test-pmd/cmdline_mtr.c ++++ b/dpdk/app/test-pmd/cmdline_mtr.c +@@ -817,8 +817,8 @@ static void cmd_create_port_meter_parsed(void *parsed_result, + cmdline_parse_inst_t cmd_create_port_meter = { + .f = cmd_create_port_meter_parsed, + .data = NULL, +- .help_str = "create port meter (yes|no) " +- " " ++ .help_str = "create port meter " ++ "(yes|no) " + "[ ...]", + .tokens = { + (void *)&cmd_create_port_meter_create, +diff --git a/dpdk/app/test-pmd/cmdline_tm.c b/dpdk/app/test-pmd/cmdline_tm.c +index bfbd43ca9b..c058b8946e 100644 +--- a/dpdk/app/test-pmd/cmdline_tm.c ++++ b/dpdk/app/test-pmd/cmdline_tm.c +@@ -69,7 +69,7 @@ print_err_msg(struct rte_tm_error *error) + [RTE_TM_ERROR_TYPE_NODE_PARAMS_N_SHARED_SHAPERS] + = "num shared shapers field (node params)", + [RTE_TM_ERROR_TYPE_NODE_PARAMS_WFQ_WEIGHT_MODE] +- = "wfq weght mode field (node params)", ++ = "wfq weight mode field (node params)", + [RTE_TM_ERROR_TYPE_NODE_PARAMS_N_SP_PRIORITIES] + = "num strict priorities field (node params)", + [RTE_TM_ERROR_TYPE_NODE_PARAMS_CMAN] +@@ -479,7 +479,7 @@ static void cmd_show_port_tm_level_cap_parsed(void *parsed_result, + cmdline_parse_inst_t cmd_show_port_tm_level_cap = { + .f = cmd_show_port_tm_level_cap_parsed, + .data = NULL, +- .help_str = "Show Port TM Hierarhical level Capabilities", ++ .help_str = "Show port TM hierarchical level capabilities", + .tokens = { + (void *)&cmd_show_port_tm_level_cap_show, + (void *)&cmd_show_port_tm_level_cap_port, +diff --git a/dpdk/app/test-pmd/config.c b/dpdk/app/test-pmd/config.c +index 1722d6c8f8..ad1b5f51d5 100644 +--- a/dpdk/app/test-pmd/config.c ++++ b/dpdk/app/test-pmd/config.c +@@ -66,8 +66,6 @@ + + #define NS_PER_SEC 1E9 + +-static char *flowtype_to_str(uint16_t flow_type); +- + static const struct { + enum tx_pkt_split split; + const char *name; +@@ -248,14 +246,20 @@ nic_stats_display(portid_t port_id) + diff_ns; + uint64_t mpps_rx, mpps_tx, mbps_rx, mbps_tx; + struct rte_eth_stats stats; +- + static const char *nic_stats_border = "########################"; ++ int ret; + + if (port_id_is_invalid(port_id, ENABLED_WARN)) { + print_valid_ports(); + return; + } +- rte_eth_stats_get(port_id, &stats); ++ ret = rte_eth_stats_get(port_id, &stats); ++ if (ret != 0) { ++ fprintf(stderr, ++ "%s: Error: failed to get stats (port %u): %d", ++ __func__, port_id, ret); ++ return; ++ } + printf("\n %s NIC statistics for port %-2d %s\n", + nic_stats_border, port_id, nic_stats_border); + +@@ -668,6 +672,19 @@ print_dev_capabilities(uint64_t capabilities) + } + } + ++const char * ++rsstypes_to_str(uint64_t rss_type) ++{ ++ uint16_t i; + -+/** -+ * struct gnet_estimator - rate estimator configuration -+ * @interval: sampling period -+ * @ewma_log: the log of measurement window weight -+ */ -+struct gnet_estimator { -+ signed char interval; -+ unsigned char ewma_log; -+}; ++ for (i = 0; rss_type_table[i].str != NULL; i++) { ++ if (rss_type_table[i].rss_type == rss_type) ++ return rss_type_table[i].str; ++ } + -+#endif /* __KERNEL__ || !HAVE_TCA_STATS_PKT64 */ -+#endif /* __LINUX_GEN_STATS_WRAPPER_H */ -diff --git a/include/openvswitch/flow.h b/include/openvswitch/flow.h -index 3054015d93..df10cf579e 100644 ---- a/include/openvswitch/flow.h -+++ b/include/openvswitch/flow.h -@@ -141,15 +141,14 @@ struct flow { - uint8_t nw_tos; /* IP ToS (including DSCP and ECN). */ - uint8_t nw_ttl; /* IP TTL/Hop Limit. */ - uint8_t nw_proto; /* IP protocol or low 8 bits of ARP opcode. */ -+ /* L4 (64-bit aligned) */ - struct in6_addr nd_target; /* IPv6 neighbor discovery (ND) target. */ - struct eth_addr arp_sha; /* ARP/ND source hardware address. */ - struct eth_addr arp_tha; /* ARP/ND target hardware address. */ -- ovs_be16 tcp_flags; /* TCP flags/ICMPv6 ND options type. -- * With L3 to avoid matching L4. */ -+ ovs_be16 tcp_flags; /* TCP flags/ICMPv6 ND options type. */ - ovs_be16 pad2; /* Pad to 64 bits. */ - struct ovs_key_nsh nsh; /* Network Service Header keys */ ++ return NULL; ++} ++ + void + port_infos_display(portid_t port_id) + { +@@ -772,19 +789,20 @@ port_infos_display(portid_t port_id) + if (!dev_info.flow_type_rss_offloads) + printf("No RSS offload flow type is supported.\n"); + else { ++ uint64_t rss_offload_types = dev_info.flow_type_rss_offloads; + uint16_t i; +- char *p; + + printf("Supported RSS offload flow types:\n"); +- for (i = RTE_ETH_FLOW_UNKNOWN + 1; +- i < sizeof(dev_info.flow_type_rss_offloads) * CHAR_BIT; i++) { +- if (!(dev_info.flow_type_rss_offloads & (1ULL << i))) +- continue; +- p = flowtype_to_str(i); +- if (p) +- printf(" %s\n", p); +- else +- printf(" user defined %d\n", i); ++ for (i = 0; i < sizeof(rss_offload_types) * CHAR_BIT; i++) { ++ uint64_t rss_offload = RTE_BIT64(i); ++ if ((rss_offload_types & rss_offload) != 0) { ++ const char *p = rsstypes_to_str(rss_offload); ++ if (p) ++ printf(" %s\n", p); ++ else ++ printf(" user defined %u\n", ++ i); ++ } + } + } -- /* L4 (64-bit aligned) */ - ovs_be16 tp_src; /* TCP/UDP/SCTP source port/ICMP type. */ - ovs_be16 tp_dst; /* TCP/UDP/SCTP destination port/ICMP code. */ - ovs_be16 ct_tp_src; /* CT original tuple source port/ICMP type. */ -@@ -179,7 +178,7 @@ BUILD_ASSERT_DECL(offsetof(struct flow, igmp_group_ip4) + sizeof(uint32_t) - enum { - FLOW_SEGMENT_1_ENDS_AT = offsetof(struct flow, dl_dst), - FLOW_SEGMENT_2_ENDS_AT = offsetof(struct flow, nw_src), -- FLOW_SEGMENT_3_ENDS_AT = offsetof(struct flow, tp_src), -+ FLOW_SEGMENT_3_ENDS_AT = offsetof(struct flow, nd_target), - }; - BUILD_ASSERT_DECL(FLOW_SEGMENT_1_ENDS_AT % sizeof(uint64_t) == 0); - BUILD_ASSERT_DECL(FLOW_SEGMENT_2_ENDS_AT % sizeof(uint64_t) == 0); -diff --git a/include/openvswitch/hmap.h b/include/openvswitch/hmap.h -index 4e001cc692..beb48295b9 100644 ---- a/include/openvswitch/hmap.h -+++ b/include/openvswitch/hmap.h -@@ -134,17 +134,17 @@ struct hmap_node *hmap_random_node(const struct hmap *); - * without using 'break', NODE will be NULL. This is true for all of the - * HMAP_FOR_EACH_*() macros. - */ --#define HMAP_FOR_EACH_WITH_HASH(NODE, MEMBER, HASH, HMAP) \ -- for (INIT_CONTAINER(NODE, hmap_first_with_hash(HMAP, HASH), MEMBER); \ -- (NODE != OBJECT_CONTAINING(NULL, NODE, MEMBER)) \ -- || ((NODE = NULL), false); \ -- ASSIGN_CONTAINER(NODE, hmap_next_with_hash(&(NODE)->MEMBER), \ -- MEMBER)) --#define HMAP_FOR_EACH_IN_BUCKET(NODE, MEMBER, HASH, HMAP) \ -- for (INIT_CONTAINER(NODE, hmap_first_in_bucket(HMAP, HASH), MEMBER); \ -- (NODE != OBJECT_CONTAINING(NULL, NODE, MEMBER)) \ -- || ((NODE = NULL), false); \ -- ASSIGN_CONTAINER(NODE, hmap_next_in_bucket(&(NODE)->MEMBER), MEMBER)) -+#define HMAP_FOR_EACH_WITH_HASH(NODE, MEMBER, HASH, HMAP) \ -+ for (INIT_MULTIVAR(NODE, MEMBER, hmap_first_with_hash(HMAP, HASH), \ -+ struct hmap_node); \ -+ CONDITION_MULTIVAR(NODE, MEMBER, ITER_VAR(NODE) != NULL); \ -+ UPDATE_MULTIVAR(NODE, hmap_next_with_hash(ITER_VAR(NODE)))) -+ -+#define HMAP_FOR_EACH_IN_BUCKET(NODE, MEMBER, HASH, HMAP) \ -+ for (INIT_MULTIVAR(NODE, MEMBER, hmap_first_in_bucket(HMAP, HASH), \ -+ struct hmap_node); \ -+ CONDITION_MULTIVAR(NODE, MEMBER, ITER_VAR(NODE) != NULL); \ -+ UPDATE_MULTIVAR(NODE, hmap_next_in_bucket(ITER_VAR(NODE)))) +@@ -912,10 +930,15 @@ port_eeprom_display(portid_t port_id) + return; + } - static inline struct hmap_node *hmap_first_with_hash(const struct hmap *, - size_t hash); -@@ -170,54 +170,80 @@ bool hmap_contains(const struct hmap *, const struct hmap_node *); - /* Iterates through every node in HMAP. */ - #define HMAP_FOR_EACH(NODE, MEMBER, HMAP) \ - HMAP_FOR_EACH_INIT(NODE, MEMBER, HMAP, (void) 0) --#define HMAP_FOR_EACH_INIT(NODE, MEMBER, HMAP, ...) \ -- for (INIT_CONTAINER(NODE, hmap_first(HMAP), MEMBER), __VA_ARGS__; \ -- (NODE != OBJECT_CONTAINING(NULL, NODE, MEMBER)) \ -- || ((NODE = NULL), false); \ -- ASSIGN_CONTAINER(NODE, hmap_next(HMAP, &(NODE)->MEMBER), MEMBER)) -+#define HMAP_FOR_EACH_INIT(NODE, MEMBER, HMAP, ...) \ -+ for (INIT_MULTIVAR_EXP(NODE, MEMBER, hmap_first(HMAP), struct hmap_node, \ -+ __VA_ARGS__); \ -+ CONDITION_MULTIVAR(NODE, MEMBER, ITER_VAR(NODE) != NULL); \ -+ UPDATE_MULTIVAR(NODE, hmap_next(HMAP, ITER_VAR(NODE)))) +- char buf[len_eeprom]; + einfo.offset = 0; + einfo.length = len_eeprom; +- einfo.data = buf; ++ einfo.data = calloc(1, len_eeprom); ++ if (!einfo.data) { ++ fprintf(stderr, ++ "Allocation of port %u eeprom data failed\n", ++ port_id); ++ return; ++ } + + ret = rte_eth_dev_get_eeprom(port_id, &einfo); + if (ret != 0) { +@@ -933,10 +956,12 @@ port_eeprom_display(portid_t port_id) + fprintf(stderr, "Unable to get EEPROM: %d\n", ret); + break; + } ++ free(einfo.data); + return; + } + rte_hexdump(stdout, "hexdump", einfo.data, einfo.length); + printf("Finish -- Port: %d EEPROM length: %d bytes\n", port_id, len_eeprom); ++ free(einfo.data); + } - /* Safe when NODE may be freed (not needed when NODE may be removed from the - * hash map but its members remain accessible and intact). */ --#define HMAP_FOR_EACH_SAFE(NODE, NEXT, MEMBER, HMAP) \ -- HMAP_FOR_EACH_SAFE_INIT(NODE, NEXT, MEMBER, HMAP, (void) 0) --#define HMAP_FOR_EACH_SAFE_INIT(NODE, NEXT, MEMBER, HMAP, ...) \ -- for (INIT_CONTAINER(NODE, hmap_first(HMAP), MEMBER), __VA_ARGS__; \ -- ((NODE != OBJECT_CONTAINING(NULL, NODE, MEMBER)) \ -- || ((NODE = NULL), false) \ -- ? INIT_CONTAINER(NEXT, hmap_next(HMAP, &(NODE)->MEMBER), MEMBER), 1 \ -- : 0); \ -- (NODE) = (NEXT)) -+#define HMAP_FOR_EACH_SAFE_LONG(NODE, NEXT, MEMBER, HMAP) \ -+ HMAP_FOR_EACH_SAFE_LONG_INIT (NODE, NEXT, MEMBER, HMAP, (void) NEXT) -+ -+#define HMAP_FOR_EACH_SAFE_LONG_INIT(NODE, NEXT, MEMBER, HMAP, ...) \ -+ for (INIT_MULTIVAR_SAFE_LONG_EXP(NODE, NEXT, MEMBER, hmap_first(HMAP), \ -+ struct hmap_node, __VA_ARGS__); \ -+ CONDITION_MULTIVAR_SAFE_LONG(NODE, NEXT, MEMBER, \ -+ ITER_VAR(NODE) != NULL, \ -+ ITER_VAR(NEXT) = hmap_next(HMAP, ITER_VAR(NODE)), \ -+ ITER_VAR(NEXT) != NULL); \ -+ UPDATE_MULTIVAR_SAFE_LONG(NODE, NEXT)) + void +@@ -972,10 +997,15 @@ port_module_eeprom_display(portid_t port_id) + return; + } + +- char buf[minfo.eeprom_len]; + einfo.offset = 0; + einfo.length = minfo.eeprom_len; +- einfo.data = buf; ++ einfo.data = calloc(1, minfo.eeprom_len); ++ if (!einfo.data) { ++ fprintf(stderr, ++ "Allocation of port %u eeprom data failed\n", ++ port_id); ++ return; ++ } + + ret = rte_eth_dev_get_module_eeprom(port_id, &einfo); + if (ret != 0) { +@@ -994,11 +1024,13 @@ port_module_eeprom_display(portid_t port_id) + ret); + break; + } ++ free(einfo.data); + return; + } + + rte_hexdump(stdout, "hexdump", einfo.data, einfo.length); + printf("Finish -- Port: %d MODULE EEPROM length: %d bytes\n", port_id, einfo.length); ++ free(einfo.data); + } + + int +@@ -1233,6 +1265,57 @@ port_reg_set(portid_t port_id, uint32_t reg_off, uint32_t reg_v) + display_port_reg_value(port_id, reg_off, reg_v); + } + ++static uint32_t ++eth_dev_get_overhead_len(uint32_t max_rx_pktlen, uint16_t max_mtu) ++{ ++ uint32_t overhead_len; + -+/* Short versions of HMAP_FOR_EACH_SAFE. */ -+#define HMAP_FOR_EACH_SAFE_SHORT(NODE, MEMBER, HMAP) \ -+ HMAP_FOR_EACH_SAFE_SHORT_INIT (NODE, MEMBER, HMAP, (void) 0) ++ if (max_mtu != UINT16_MAX && max_rx_pktlen > max_mtu) ++ overhead_len = max_rx_pktlen - max_mtu; ++ else ++ overhead_len = RTE_ETHER_HDR_LEN + RTE_ETHER_CRC_LEN; + -+#define HMAP_FOR_EACH_SAFE_SHORT_INIT(NODE, MEMBER, HMAP, ...) \ -+ for (INIT_MULTIVAR_SAFE_SHORT_EXP(NODE, MEMBER, hmap_first(HMAP), \ -+ struct hmap_node, __VA_ARGS__); \ -+ CONDITION_MULTIVAR_SAFE_SHORT(NODE, MEMBER, \ -+ ITER_VAR(NODE) != NULL, \ -+ ITER_NEXT_VAR(NODE) = hmap_next(HMAP, ITER_VAR(NODE))); \ -+ UPDATE_MULTIVAR_SAFE_SHORT(NODE)) ++ return overhead_len; ++} + -+#define HMAP_FOR_EACH_SAFE(...) \ -+ OVERLOAD_SAFE_MACRO(HMAP_FOR_EACH_SAFE_LONG, \ -+ HMAP_FOR_EACH_SAFE_SHORT, \ -+ 4, __VA_ARGS__) ++static int ++eth_dev_validate_mtu(uint16_t port_id, uint16_t mtu) ++{ ++ struct rte_eth_dev_info dev_info; ++ uint32_t overhead_len; ++ uint32_t frame_size; ++ int ret; ++ ++ ret = rte_eth_dev_info_get(port_id, &dev_info); ++ if (ret != 0) ++ return ret; ++ ++ if (mtu < dev_info.min_mtu) { ++ fprintf(stderr, ++ "MTU (%u) < device min MTU (%u) for port_id %u\n", ++ mtu, dev_info.min_mtu, port_id); ++ return -EINVAL; ++ } ++ if (mtu > dev_info.max_mtu) { ++ fprintf(stderr, ++ "MTU (%u) > device max MTU (%u) for port_id %u\n", ++ mtu, dev_info.max_mtu, port_id); ++ return -EINVAL; ++ } ++ ++ overhead_len = eth_dev_get_overhead_len(dev_info.max_rx_pktlen, ++ dev_info.max_mtu); ++ frame_size = mtu + overhead_len; ++ if (frame_size > dev_info.max_rx_pktlen) { ++ fprintf(stderr, ++ "Frame size (%u) > device max frame size (%u) for port_id %u\n", ++ frame_size, dev_info.max_rx_pktlen, port_id); ++ return -EINVAL; ++ } ++ ++ return 0; ++} + + void + port_mtu_set(portid_t port_id, uint16_t mtu) + { +@@ -1242,6 +1325,10 @@ port_mtu_set(portid_t port_id, uint16_t mtu) + if (port_id_is_invalid(port_id, ENABLED_WARN)) + return; + ++ diag = eth_dev_validate_mtu(port_id, mtu); ++ if (diag != 0) ++ return; ++ + if (port->need_reconfig == 0) { + diag = rte_eth_dev_set_mtu(port_id, mtu); + if (diag != 0) { +@@ -1682,6 +1769,37 @@ port_action_handle_destroy(portid_t port_id, + return ret; + } - /* Continues an iteration from just after NODE. */ - #define HMAP_FOR_EACH_CONTINUE(NODE, MEMBER, HMAP) \ - HMAP_FOR_EACH_CONTINUE_INIT(NODE, MEMBER, HMAP, (void) 0) --#define HMAP_FOR_EACH_CONTINUE_INIT(NODE, MEMBER, HMAP, ...) \ -- for (ASSIGN_CONTAINER(NODE, hmap_next(HMAP, &(NODE)->MEMBER), MEMBER), \ -- __VA_ARGS__; \ -- (NODE != OBJECT_CONTAINING(NULL, NODE, MEMBER)) \ -- || ((NODE = NULL), false); \ -- ASSIGN_CONTAINER(NODE, hmap_next(HMAP, &(NODE)->MEMBER), MEMBER)) -+#define HMAP_FOR_EACH_CONTINUE_INIT(NODE, MEMBER, HMAP, ...) \ -+ for (INIT_MULTIVAR_EXP(NODE, MEMBER, hmap_next(HMAP, &(NODE)->MEMBER), \ -+ struct hmap_node, __VA_ARGS__); \ -+ CONDITION_MULTIVAR(NODE, MEMBER, ITER_VAR(NODE) != NULL); \ -+ UPDATE_MULTIVAR(NODE, hmap_next(HMAP, ITER_VAR(NODE)))) -+ -+struct hmap_pop_helper_iter__ { -+ size_t bucket; -+ struct hmap_node *node; -+}; ++int ++port_action_handle_flush(portid_t port_id) ++{ ++ struct rte_port *port; ++ struct port_indirect_action **tmp; ++ int ret = 0; ++ ++ if (port_id_is_invalid(port_id, ENABLED_WARN) || ++ port_id == (portid_t)RTE_PORT_ALL) ++ return -EINVAL; ++ port = &ports[port_id]; ++ tmp = &port->actions_list; ++ while (*tmp != NULL) { ++ struct rte_flow_error error; ++ struct port_indirect_action *pia = *tmp; ++ ++ /* Poisoning to make sure PMDs update it in case of error. */ ++ memset(&error, 0x44, sizeof(error)); ++ if (pia->handle != NULL && ++ rte_flow_action_handle_destroy ++ (port_id, pia->handle, &error) != 0) { ++ printf("Indirect action #%u not destroyed\n", pia->id); ++ ret = port_flow_complain(&error); ++ tmp = &pia->next; ++ } else { ++ *tmp = pia->next; ++ free(pia); ++ } ++ } ++ return ret; ++} --static inline struct hmap_node * --hmap_pop_helper__(struct hmap *hmap, size_t *bucket) { -+static inline void -+hmap_pop_helper__(struct hmap *hmap, struct hmap_pop_helper_iter__ *iter) { + /** Get indirect action by port + id */ + struct rte_flow_action_handle * +@@ -2758,8 +2876,8 @@ rxtx_config_display(void) + nb_fwd_lcores, nb_fwd_ports); + + RTE_ETH_FOREACH_DEV(pid) { +- struct rte_eth_rxconf *rx_conf = &ports[pid].rx_conf[0]; +- struct rte_eth_txconf *tx_conf = &ports[pid].tx_conf[0]; ++ struct rte_eth_rxconf *rx_conf = &ports[pid].rxq[0].conf; ++ struct rte_eth_txconf *tx_conf = &ports[pid].txq[0].conf; + uint16_t *nb_rx_desc = &ports[pid].nb_rx_desc[0]; + uint16_t *nb_tx_desc = &ports[pid].nb_tx_desc[0]; + struct rte_eth_rxq_info rx_qinfo; +@@ -3017,7 +3135,7 @@ fwd_stream_on_other_lcores(uint16_t domain_id, lcoreid_t src_lc, + fs = fwd_streams[sm_id]; + port = &ports[fs->rx_port]; + dev_info = &port->dev_info; +- rxq_conf = &port->rx_conf[fs->rx_queue]; ++ rxq_conf = &port->rxq[fs->rx_queue].conf; + if ((dev_info->dev_capa & RTE_ETH_DEV_CAPA_RXQ_SHARE) + == 0 || rxq_conf->share_group == 0) + /* Not shared rxq. */ +@@ -3077,7 +3195,7 @@ pkt_fwd_shared_rxq_check(void) + fs->lcore = fwd_lcores[lc_id]; + port = &ports[fs->rx_port]; + dev_info = &port->dev_info; +- rxq_conf = &port->rx_conf[fs->rx_queue]; ++ rxq_conf = &port->rxq[fs->rx_queue].conf; + if ((dev_info->dev_capa & RTE_ETH_DEV_CAPA_RXQ_SHARE) + == 0 || rxq_conf->share_group == 0) + /* Not shared rxq. */ +@@ -4719,6 +4837,8 @@ set_record_burst_stats(uint8_t on_off) + record_burst_stats = on_off; + } -- for (; *bucket <= hmap->mask; (*bucket)++) { -- struct hmap_node *node = hmap->buckets[*bucket]; -+ for (; iter->bucket <= hmap->mask; (iter->bucket)++) { -+ struct hmap_node *node = hmap->buckets[iter->bucket]; ++#if defined(RTE_NET_I40E) || defined(RTE_NET_IXGBE) ++ + static char* + flowtype_to_str(uint16_t flow_type) + { +@@ -4762,8 +4882,6 @@ flowtype_to_str(uint16_t flow_type) + return NULL; + } - if (node) { - hmap_remove(hmap, node); -- return node; -+ iter->node = node; -+ return; - } - } +-#if defined(RTE_NET_I40E) || defined(RTE_NET_IXGBE) - -- return NULL; -+ iter->node = NULL; + static inline void + print_fdir_mask(struct rte_eth_fdir_masks *mask) + { +@@ -5185,6 +5303,25 @@ mcast_addr_pool_remove(struct rte_port *port, uint32_t addr_idx) + sizeof(struct rte_ether_addr) * (port->mc_addr_nb - addr_idx)); } --#define HMAP_FOR_EACH_POP(NODE, MEMBER, HMAP) \ -- for (size_t bucket__ = 0; \ -- INIT_CONTAINER(NODE, hmap_pop_helper__(HMAP, &bucket__), MEMBER), \ -- (NODE != OBJECT_CONTAINING(NULL, NODE, MEMBER)) \ -- || ((NODE = NULL), false);) -+#define HMAP_FOR_EACH_POP(NODE, MEMBER, HMAP) \ -+ for (struct hmap_pop_helper_iter__ ITER_VAR(NODE) = { 0, NULL }; \ -+ hmap_pop_helper__(HMAP, &ITER_VAR(NODE)), \ -+ (ITER_VAR(NODE).node != NULL) ? \ -+ (((NODE) = OBJECT_CONTAINING(ITER_VAR(NODE).node, \ -+ NODE, MEMBER)),1): \ -+ (((NODE) = NULL), 0);) - - static inline struct hmap_node *hmap_first(const struct hmap *); - static inline struct hmap_node *hmap_next(const struct hmap *, -diff --git a/include/openvswitch/list.h b/include/openvswitch/list.h -index 8ad5eeb327..6272d340cf 100644 ---- a/include/openvswitch/list.h -+++ b/include/openvswitch/list.h -@@ -72,37 +72,74 @@ static inline bool ovs_list_is_empty(const struct ovs_list *); - static inline bool ovs_list_is_singleton(const struct ovs_list *); - static inline bool ovs_list_is_short(const struct ovs_list *); ++int ++mcast_addr_pool_destroy(portid_t port_id) ++{ ++ struct rte_port *port; ++ ++ if (port_id_is_invalid(port_id, ENABLED_WARN) || ++ port_id == (portid_t)RTE_PORT_ALL) ++ return -EINVAL; ++ port = &ports[port_id]; ++ ++ if (port->mc_addr_nb != 0) { ++ /* free the pool of multicast addresses. */ ++ free(port->mc_addr_pool); ++ port->mc_addr_pool = NULL; ++ port->mc_addr_nb = 0; ++ } ++ return 0; ++} ++ + static int + eth_port_multicast_addr_list_set(portid_t port_id) + { +diff --git a/dpdk/app/test-pmd/csumonly.c b/dpdk/app/test-pmd/csumonly.c +index 2aeea243b6..d661e21e02 100644 +--- a/dpdk/app/test-pmd/csumonly.c ++++ b/dpdk/app/test-pmd/csumonly.c +@@ -222,15 +222,14 @@ parse_gtp(struct rte_udp_hdr *udp_hdr, + + gtp_hdr = (struct rte_gtp_hdr *)((char *)udp_hdr + + sizeof(struct rte_udp_hdr)); +- ++ if (gtp_hdr->e || gtp_hdr->s || gtp_hdr->pn) ++ gtp_len += sizeof(struct rte_gtp_hdr_ext_word); + /* + * Check message type. If message type is 0xff, it is + * a GTP data packet. If not, it is a GTP control packet + */ + if (gtp_hdr->msg_type == 0xff) { +- ip_ver = *(uint8_t *)((char *)udp_hdr + +- sizeof(struct rte_udp_hdr) + +- sizeof(struct rte_gtp_hdr)); ++ ip_ver = *(uint8_t *)((char *)gtp_hdr + gtp_len); + ip_ver = (ip_ver) & 0xf0; + + if (ip_ver == RTE_GTP_TYPE_IPV4) { +@@ -257,8 +256,7 @@ parse_gtp(struct rte_udp_hdr *udp_hdr, + /* Parse a vxlan header */ + static void + parse_vxlan(struct rte_udp_hdr *udp_hdr, +- struct testpmd_offload_info *info, +- uint32_t pkt_type) ++ struct testpmd_offload_info *info) + { + struct rte_ether_hdr *eth_hdr; + +@@ -266,8 +264,7 @@ parse_vxlan(struct rte_udp_hdr *udp_hdr, + * default vxlan port (rfc7348) or that the rx offload flag is set + * (i40e only currently) + */ +- if (udp_hdr->dst_port != _htons(RTE_VXLAN_DEFAULT_PORT) && +- RTE_ETH_IS_TUNNEL_PKT(pkt_type) == 0) ++ if (udp_hdr->dst_port != _htons(RTE_VXLAN_DEFAULT_PORT)) + return; + + update_tunnel_outer(info); +@@ -771,6 +768,28 @@ pkt_copy_split(const struct rte_mbuf *pkt) + return md[0]; + } --#define LIST_FOR_EACH(ITER, MEMBER, LIST) \ -- for (INIT_CONTAINER(ITER, (LIST)->next, MEMBER); \ -- &(ITER)->MEMBER != (LIST); \ -- ASSIGN_CONTAINER(ITER, (ITER)->MEMBER.next, MEMBER)) --#define LIST_FOR_EACH_CONTINUE(ITER, MEMBER, LIST) \ -- for (ASSIGN_CONTAINER(ITER, (ITER)->MEMBER.next, MEMBER); \ -- &(ITER)->MEMBER != (LIST); \ -- ASSIGN_CONTAINER(ITER, (ITER)->MEMBER.next, MEMBER)) --#define LIST_FOR_EACH_REVERSE(ITER, MEMBER, LIST) \ -- for (INIT_CONTAINER(ITER, (LIST)->prev, MEMBER); \ -- &(ITER)->MEMBER != (LIST); \ -- ASSIGN_CONTAINER(ITER, (ITER)->MEMBER.prev, MEMBER)) --#define LIST_FOR_EACH_REVERSE_SAFE(ITER, PREV, MEMBER, LIST) \ -- for (INIT_CONTAINER(ITER, (LIST)->prev, MEMBER); \ -- (&(ITER)->MEMBER != (LIST) \ -- ? INIT_CONTAINER(PREV, (ITER)->MEMBER.prev, MEMBER), 1 \ -- : 0); \ -- (ITER) = (PREV)) --#define LIST_FOR_EACH_REVERSE_CONTINUE(ITER, MEMBER, LIST) \ -- for (ASSIGN_CONTAINER(ITER, (ITER)->MEMBER.prev, MEMBER); \ -- &(ITER)->MEMBER != (LIST); \ -- ASSIGN_CONTAINER(ITER, (ITER)->MEMBER.prev, MEMBER)) --#define LIST_FOR_EACH_SAFE(ITER, NEXT, MEMBER, LIST) \ -- for (INIT_CONTAINER(ITER, (LIST)->next, MEMBER); \ -- (&(ITER)->MEMBER != (LIST) \ -- ? INIT_CONTAINER(NEXT, (ITER)->MEMBER.next, MEMBER), 1 \ -- : 0); \ -- (ITER) = (NEXT)) --#define LIST_FOR_EACH_POP(ITER, MEMBER, LIST) \ -- while (!ovs_list_is_empty(LIST) \ -- && (INIT_CONTAINER(ITER, ovs_list_pop_front(LIST), MEMBER), 1)) -+#define LIST_FOR_EACH(VAR, MEMBER, LIST) \ -+ for (INIT_MULTIVAR(VAR, MEMBER, (LIST)->next, struct ovs_list); \ -+ CONDITION_MULTIVAR(VAR, MEMBER, ITER_VAR(VAR) != (LIST)); \ -+ UPDATE_MULTIVAR(VAR, ITER_VAR(VAR)->next)) ++#if defined(RTE_LIB_GRO) || defined(RTE_LIB_GSO) ++/* ++ * Re-calculate IP checksum for merged/fragmented packets. ++ */ ++static void ++pkts_ip_csum_recalc(struct rte_mbuf **pkts_burst, const uint16_t nb_pkts, uint64_t tx_offloads) ++{ ++ int i; ++ struct rte_ipv4_hdr *ipv4_hdr; ++ for (i = 0; i < nb_pkts; i++) { ++ if ((pkts_burst[i]->ol_flags & RTE_MBUF_F_TX_IPV4) && ++ (tx_offloads & RTE_ETH_TX_OFFLOAD_IPV4_CKSUM) == 0) { ++ ipv4_hdr = rte_pktmbuf_mtod_offset(pkts_burst[i], ++ struct rte_ipv4_hdr *, ++ pkts_burst[i]->l2_len); ++ ipv4_hdr->hdr_checksum = 0; ++ ipv4_hdr->hdr_checksum = rte_ipv4_cksum(ipv4_hdr); ++ } ++ } ++} ++#endif + -+#define LIST_FOR_EACH_CONTINUE(VAR, MEMBER, LIST) \ -+ for (INIT_MULTIVAR(VAR, MEMBER, VAR->MEMBER.next, struct ovs_list); \ -+ CONDITION_MULTIVAR(VAR, MEMBER, ITER_VAR(VAR) != (LIST)); \ -+ UPDATE_MULTIVAR(VAR, ITER_VAR(VAR)->next)) + /* + * Receive a burst of packets, and for each packet: + * - parse packet, and try to recognize a supported packet type (1) +@@ -796,7 +815,7 @@ pkt_copy_split(const struct rte_mbuf *pkt) + * + * The testpmd command line for this forward engine sets the flags + * TESTPMD_TX_OFFLOAD_* in ports[tx_port].tx_ol_flags. They control +- * wether a checksum must be calculated in software or in hardware. The ++ * whether a checksum must be calculated in software or in hardware. The + * IP, UDP, TCP and SCTP flags always concern the inner layer. The + * OUTER_IP is only useful for tunnel packets. + */ +@@ -887,10 +906,6 @@ pkt_burst_checksum_forward(struct fwd_stream *fs) + * and inner headers */ + + eth_hdr = rte_pktmbuf_mtod(m, struct rte_ether_hdr *); +- rte_ether_addr_copy(&peer_eth_addrs[fs->peer_addr], +- ð_hdr->dst_addr); +- rte_ether_addr_copy(&ports[fs->tx_port].eth_addr, +- ð_hdr->src_addr); + parse_ethernet(eth_hdr, &info); + l3_hdr = (char *)eth_hdr + info.l2_len; + +@@ -912,8 +927,7 @@ pkt_burst_checksum_forward(struct fwd_stream *fs) + RTE_MBUF_F_TX_TUNNEL_VXLAN_GPE; + goto tunnel_update; + } +- parse_vxlan(udp_hdr, &info, +- m->packet_type); ++ parse_vxlan(udp_hdr, &info); + if (info.is_tunnel) { + tx_ol_flags |= + RTE_MBUF_F_TX_TUNNEL_VXLAN; +@@ -925,6 +939,12 @@ pkt_burst_checksum_forward(struct fwd_stream *fs) + RTE_MBUF_F_TX_TUNNEL_GENEVE; + goto tunnel_update; + } ++ /* Always keep last. */ ++ if (unlikely(RTE_ETH_IS_TUNNEL_PKT( ++ m->packet_type) != 0)) { ++ TESTPMD_LOG(DEBUG, "Unknown tunnel packet. UDP dst port: %hu", ++ udp_hdr->dst_port); ++ } + } else if (info.l4_proto == IPPROTO_GRE) { + struct simple_gre_hdr *gre_hdr; + +@@ -1089,6 +1109,8 @@ pkt_burst_checksum_forward(struct fwd_stream *fs) + fs->gro_times = 0; + } + } ++ ++ pkts_ip_csum_recalc(pkts_burst, nb_rx, tx_offloads); + } + #endif + +@@ -1122,6 +1144,8 @@ pkt_burst_checksum_forward(struct fwd_stream *fs) + + tx_pkts_burst = gso_segments; + nb_rx = nb_segments; + -+#define LIST_FOR_EACH_REVERSE(VAR, MEMBER, LIST) \ -+ for (INIT_MULTIVAR(VAR, MEMBER, (LIST)->prev, struct ovs_list); \ -+ CONDITION_MULTIVAR(VAR, MEMBER, ITER_VAR(VAR) != (LIST)); \ -+ UPDATE_MULTIVAR(VAR, ITER_VAR(VAR)->prev)) ++ pkts_ip_csum_recalc(tx_pkts_burst, nb_rx, tx_offloads); + } else + #endif + tx_pkts_burst = pkts_burst; +@@ -1164,9 +1188,22 @@ pkt_burst_checksum_forward(struct fwd_stream *fs) + get_end_cycles(fs, start_tsc); + } + ++static void ++stream_init_checksum_forward(struct fwd_stream *fs) ++{ ++ bool rx_stopped, tx_stopped; + -+#define LIST_FOR_EACH_REVERSE_CONTINUE(VAR, MEMBER, LIST) \ -+ for (INIT_MULTIVAR(VAR, MEMBER, VAR->MEMBER.prev, struct ovs_list); \ -+ CONDITION_MULTIVAR(VAR, MEMBER, ITER_VAR(VAR) != (LIST)); \ -+ UPDATE_MULTIVAR(VAR, ITER_VAR(VAR)->prev)) ++ rx_stopped = ports[fs->rx_port].rxq[fs->rx_queue].state == ++ RTE_ETH_QUEUE_STATE_STOPPED; ++ tx_stopped = ports[fs->tx_port].txq[fs->tx_queue].state == ++ RTE_ETH_QUEUE_STATE_STOPPED; ++ fs->disabled = rx_stopped || tx_stopped; ++} + -+/* LONG version of SAFE iterators. */ -+#define LIST_FOR_EACH_REVERSE_SAFE_LONG(VAR, PREV, MEMBER, LIST) \ -+ for (INIT_MULTIVAR_SAFE_LONG(VAR, PREV, MEMBER, (LIST)->prev, \ -+ struct ovs_list); \ -+ CONDITION_MULTIVAR_SAFE_LONG(VAR, PREV, MEMBER, \ -+ ITER_VAR(VAR) != (LIST), \ -+ ITER_VAR(PREV) = ITER_VAR(VAR)->prev, \ -+ ITER_VAR(PREV) != (LIST)); \ -+ UPDATE_MULTIVAR_SAFE_LONG(VAR, PREV)) + struct fwd_engine csum_fwd_engine = { + .fwd_mode_name = "csum", + .port_fwd_begin = NULL, + .port_fwd_end = NULL, ++ .stream_init = stream_init_checksum_forward, + .packet_fwd = pkt_burst_checksum_forward, + }; +diff --git a/dpdk/app/test-pmd/flowgen.c b/dpdk/app/test-pmd/flowgen.c +index 9ceef3b54a..1e01120ae9 100644 +--- a/dpdk/app/test-pmd/flowgen.c ++++ b/dpdk/app/test-pmd/flowgen.c +@@ -207,9 +207,22 @@ flowgen_begin(portid_t pi) + return 0; + } + ++static void ++flowgen_stream_init(struct fwd_stream *fs) ++{ ++ bool rx_stopped, tx_stopped; + -+#define LIST_FOR_EACH_SAFE_LONG(VAR, NEXT, MEMBER, LIST) \ -+ for (INIT_MULTIVAR_SAFE_LONG(VAR, NEXT, MEMBER, (LIST)->next, \ -+ struct ovs_list); \ -+ CONDITION_MULTIVAR_SAFE_LONG(VAR, NEXT, MEMBER, \ -+ ITER_VAR(VAR) != (LIST), \ -+ ITER_VAR(NEXT) = ITER_VAR(VAR)->next, \ -+ ITER_VAR(NEXT) != (LIST)); \ -+ UPDATE_MULTIVAR_SAFE_LONG(VAR, NEXT)) ++ rx_stopped = ports[fs->rx_port].rxq[fs->rx_queue].state == ++ RTE_ETH_QUEUE_STATE_STOPPED; ++ tx_stopped = ports[fs->tx_port].txq[fs->tx_queue].state == ++ RTE_ETH_QUEUE_STATE_STOPPED; ++ fs->disabled = rx_stopped || tx_stopped; ++} + -+/* SHORT version of SAFE iterators. */ -+#define LIST_FOR_EACH_REVERSE_SAFE_SHORT(VAR, MEMBER, LIST) \ -+ for (INIT_MULTIVAR_SAFE_SHORT(VAR, MEMBER, (LIST)->prev, struct ovs_list);\ -+ CONDITION_MULTIVAR_SAFE_SHORT(VAR, MEMBER, \ -+ ITER_VAR(VAR) != (LIST), \ -+ ITER_NEXT_VAR(VAR) = ITER_VAR(VAR)->prev); \ -+ UPDATE_MULTIVAR_SAFE_SHORT(VAR)) + struct fwd_engine flow_gen_engine = { + .fwd_mode_name = "flowgen", + .port_fwd_begin = flowgen_begin, + .port_fwd_end = NULL, ++ .stream_init = flowgen_stream_init, + .packet_fwd = pkt_burst_flow_gen, + }; +diff --git a/dpdk/app/test-pmd/icmpecho.c b/dpdk/app/test-pmd/icmpecho.c +index 99c94cb282..066f2a3ab7 100644 +--- a/dpdk/app/test-pmd/icmpecho.c ++++ b/dpdk/app/test-pmd/icmpecho.c +@@ -512,9 +512,22 @@ reply_to_icmp_echo_rqsts(struct fwd_stream *fs) + get_end_cycles(fs, start_tsc); + } + ++static void ++icmpecho_stream_init(struct fwd_stream *fs) ++{ ++ bool rx_stopped, tx_stopped; + -+#define LIST_FOR_EACH_SAFE_SHORT(VAR, MEMBER, LIST) \ -+ for (INIT_MULTIVAR_SAFE_SHORT(VAR, MEMBER, (LIST)->next, struct ovs_list);\ -+ CONDITION_MULTIVAR_SAFE_SHORT(VAR, MEMBER, \ -+ ITER_VAR(VAR) != (LIST), \ -+ ITER_NEXT_VAR(VAR) = ITER_VAR(VAR)->next); \ -+ UPDATE_MULTIVAR_SAFE_SHORT(VAR)) ++ rx_stopped = ports[fs->rx_port].rxq[fs->rx_queue].state == ++ RTE_ETH_QUEUE_STATE_STOPPED; ++ tx_stopped = ports[fs->tx_port].txq[fs->tx_queue].state == ++ RTE_ETH_QUEUE_STATE_STOPPED; ++ fs->disabled = rx_stopped || tx_stopped; ++} + -+#define LIST_FOR_EACH_SAFE(...) \ -+ OVERLOAD_SAFE_MACRO(LIST_FOR_EACH_SAFE_LONG, \ -+ LIST_FOR_EACH_SAFE_SHORT, \ -+ 4, __VA_ARGS__) + struct fwd_engine icmp_echo_engine = { + .fwd_mode_name = "icmpecho", + .port_fwd_begin = NULL, + .port_fwd_end = NULL, ++ .stream_init = icmpecho_stream_init, + .packet_fwd = reply_to_icmp_echo_rqsts, + }; +diff --git a/dpdk/app/test-pmd/ieee1588fwd.c b/dpdk/app/test-pmd/ieee1588fwd.c +index 9ff817aa68..fc4e2d014c 100644 +--- a/dpdk/app/test-pmd/ieee1588fwd.c ++++ b/dpdk/app/test-pmd/ieee1588fwd.c +@@ -211,9 +211,22 @@ port_ieee1588_fwd_end(portid_t pi) + rte_eth_timesync_disable(pi); + } + ++static void ++port_ieee1588_stream_init(struct fwd_stream *fs) ++{ ++ bool rx_stopped, tx_stopped; + -+#define LIST_FOR_EACH_REVERSE_SAFE(...) \ -+ OVERLOAD_SAFE_MACRO(LIST_FOR_EACH_REVERSE_SAFE_LONG, \ -+ LIST_FOR_EACH_REVERSE_SAFE_SHORT, \ -+ 4, __VA_ARGS__) ++ rx_stopped = ports[fs->rx_port].rxq[fs->rx_queue].state == ++ RTE_ETH_QUEUE_STATE_STOPPED; ++ tx_stopped = ports[fs->tx_port].txq[fs->tx_queue].state == ++ RTE_ETH_QUEUE_STATE_STOPPED; ++ fs->disabled = rx_stopped || tx_stopped; ++} + -+#define LIST_FOR_EACH_POP(ITER, MEMBER, LIST) \ -+ while (!ovs_list_is_empty(LIST) ? \ -+ (INIT_CONTAINER(ITER, ovs_list_pop_front(LIST), MEMBER), 1) : \ -+ (ITER = NULL, 0)) - - /* Inline implementations. */ + struct fwd_engine ieee1588_fwd_engine = { + .fwd_mode_name = "ieee1588", + .port_fwd_begin = port_ieee1588_fwd_begin, + .port_fwd_end = port_ieee1588_fwd_end, ++ .stream_init = port_ieee1588_stream_init, + .packet_fwd = ieee1588_packet_fwd, + }; +diff --git a/dpdk/app/test-pmd/iofwd.c b/dpdk/app/test-pmd/iofwd.c +index 19cd920f70..71849aaf96 100644 +--- a/dpdk/app/test-pmd/iofwd.c ++++ b/dpdk/app/test-pmd/iofwd.c +@@ -88,9 +88,22 @@ pkt_burst_io_forward(struct fwd_stream *fs) + get_end_cycles(fs, start_tsc); + } -diff --git a/include/openvswitch/ofp-actions.h b/include/openvswitch/ofp-actions.h -index 41bcb55d20..b7231c7bb3 100644 ---- a/include/openvswitch/ofp-actions.h -+++ b/include/openvswitch/ofp-actions.h -@@ -218,7 +218,9 @@ struct ofpact *ofpact_next_flattened(const struct ofpact *); - static inline struct ofpact * - ofpact_end(const struct ofpact *ofpacts, size_t ofpacts_len) - { -- return ALIGNED_CAST(struct ofpact *, (uint8_t *) ofpacts + ofpacts_len); -+ return ofpacts -+ ? ALIGNED_CAST(struct ofpact *, (uint8_t *) ofpacts + ofpacts_len) -+ : NULL; ++static void ++stream_init_forward(struct fwd_stream *fs) ++{ ++ bool rx_stopped, tx_stopped; ++ ++ rx_stopped = ports[fs->rx_port].rxq[fs->rx_queue].state == ++ RTE_ETH_QUEUE_STATE_STOPPED; ++ tx_stopped = ports[fs->tx_port].txq[fs->tx_queue].state == ++ RTE_ETH_QUEUE_STATE_STOPPED; ++ fs->disabled = rx_stopped || tx_stopped; ++} ++ + struct fwd_engine io_fwd_engine = { + .fwd_mode_name = "io", + .port_fwd_begin = NULL, + .port_fwd_end = NULL, ++ .stream_init = stream_init_forward, + .packet_fwd = pkt_burst_io_forward, + }; +diff --git a/dpdk/app/test-pmd/macfwd.c b/dpdk/app/test-pmd/macfwd.c +index 812a0c721f..79c9241d00 100644 +--- a/dpdk/app/test-pmd/macfwd.c ++++ b/dpdk/app/test-pmd/macfwd.c +@@ -119,9 +119,22 @@ pkt_burst_mac_forward(struct fwd_stream *fs) + get_end_cycles(fs, start_tsc); } - static inline bool -diff --git a/include/openvswitch/ofpbuf.h b/include/openvswitch/ofpbuf.h -index 1136ba04c8..32f03ea837 100644 ---- a/include/openvswitch/ofpbuf.h -+++ b/include/openvswitch/ofpbuf.h -@@ -179,7 +179,11 @@ static inline void ofpbuf_delete(struct ofpbuf *b) - static inline void *ofpbuf_at(const struct ofpbuf *b, size_t offset, - size_t size) - { -- return offset + size <= b->size ? (char *) b->data + offset : NULL; -+ if (offset + size <= b->size) { -+ ovs_assert(b->data); -+ return (char *) b->data + offset; -+ } -+ return NULL; ++static void ++stream_init_mac_forward(struct fwd_stream *fs) ++{ ++ bool rx_stopped, tx_stopped; ++ ++ rx_stopped = ports[fs->rx_port].rxq[fs->rx_queue].state == ++ RTE_ETH_QUEUE_STATE_STOPPED; ++ tx_stopped = ports[fs->tx_port].txq[fs->tx_queue].state == ++ RTE_ETH_QUEUE_STATE_STOPPED; ++ fs->disabled = rx_stopped || tx_stopped; ++} ++ + struct fwd_engine mac_fwd_engine = { + .fwd_mode_name = "mac", + .port_fwd_begin = NULL, + .port_fwd_end = NULL, ++ .stream_init = stream_init_mac_forward, + .packet_fwd = pkt_burst_mac_forward, + }; +diff --git a/dpdk/app/test-pmd/macswap.c b/dpdk/app/test-pmd/macswap.c +index 4627ff83e9..acb0fd7fb4 100644 +--- a/dpdk/app/test-pmd/macswap.c ++++ b/dpdk/app/test-pmd/macswap.c +@@ -97,9 +97,22 @@ pkt_burst_mac_swap(struct fwd_stream *fs) + get_end_cycles(fs, start_tsc); } - /* Returns a pointer to byte 'offset' in 'b', which must contain at least -@@ -188,20 +192,23 @@ static inline void *ofpbuf_at_assert(const struct ofpbuf *b, size_t offset, - size_t size) - { - ovs_assert(offset + size <= b->size); -- return ((char *) b->data) + offset; -+ ovs_assert(b->data); -+ return (char *) b->data + offset; ++static void ++stream_init_mac_swap(struct fwd_stream *fs) ++{ ++ bool rx_stopped, tx_stopped; ++ ++ rx_stopped = ports[fs->rx_port].rxq[fs->rx_queue].state == ++ RTE_ETH_QUEUE_STATE_STOPPED; ++ tx_stopped = ports[fs->tx_port].txq[fs->tx_queue].state == ++ RTE_ETH_QUEUE_STATE_STOPPED; ++ fs->disabled = rx_stopped || tx_stopped; ++} ++ + struct fwd_engine mac_swap_engine = { + .fwd_mode_name = "macswap", + .port_fwd_begin = NULL, + .port_fwd_end = NULL, ++ .stream_init = stream_init_mac_swap, + .packet_fwd = pkt_burst_mac_swap, + }; +diff --git a/dpdk/app/test-pmd/noisy_vnf.c b/dpdk/app/test-pmd/noisy_vnf.c +index e4434bea95..a92e810190 100644 +--- a/dpdk/app/test-pmd/noisy_vnf.c ++++ b/dpdk/app/test-pmd/noisy_vnf.c +@@ -277,9 +277,22 @@ noisy_fwd_begin(portid_t pi) + return 0; } - /* Returns a pointer to byte following the last byte of data in use in 'b'. */ - static inline void *ofpbuf_tail(const struct ofpbuf *b) ++static void ++stream_init_noisy_vnf(struct fwd_stream *fs) ++{ ++ bool rx_stopped, tx_stopped; ++ ++ rx_stopped = ports[fs->rx_port].rxq[fs->rx_queue].state == ++ RTE_ETH_QUEUE_STATE_STOPPED; ++ tx_stopped = ports[fs->tx_port].txq[fs->tx_queue].state == ++ RTE_ETH_QUEUE_STATE_STOPPED; ++ fs->disabled = rx_stopped || tx_stopped; ++} ++ + struct fwd_engine noisy_vnf_engine = { + .fwd_mode_name = "noisy", + .port_fwd_begin = noisy_fwd_begin, + .port_fwd_end = noisy_fwd_end, ++ .stream_init = stream_init_noisy_vnf, + .packet_fwd = pkt_burst_noisy_vnf, + }; +diff --git a/dpdk/app/test-pmd/parameters.c b/dpdk/app/test-pmd/parameters.c +index f9185065af..e3c9757f3f 100644 +--- a/dpdk/app/test-pmd/parameters.c ++++ b/dpdk/app/test-pmd/parameters.c +@@ -61,6 +61,9 @@ usage(char* progname) + "extended statistics to show. Used with --stats-period " + "specified or interactive commands that show Rx/Tx statistics " + "(i.e. 'show port stats').\n"); ++ printf(" --num-procs=N: set the total number of multi-process instances.\n"); ++ printf(" --proc-id=id: set the id of the current process from " ++ "multi-process instances (0 <= id < num-procs).\n"); + printf(" --nb-cores=N: set the number of forwarding cores " + "(1 <= N <= %d).\n", nb_lcores); + printf(" --nb-ports=N: set the number of forwarding ports " +@@ -110,7 +113,7 @@ usage(char* progname) + "If the drop-queue doesn't exist, the packet is dropped. " + "By default drop-queue=127.\n"); + #ifdef RTE_LIB_LATENCYSTATS +- printf(" --latencystats=N: enable latency and jitter statistcs " ++ printf(" --latencystats=N: enable latency and jitter statistics " + "monitoring on forwarding lcore id N.\n"); + #endif + printf(" --disable-crc-strip: disable CRC stripping by hardware.\n"); +@@ -940,11 +943,12 @@ launch_args_parse(int argc, char** argv) + } + if (!strcmp(lgopts[opt_idx].name, "total-num-mbufs")) { + n = atoi(optarg); +- if (n > 1024) ++ if (n > MIN_TOTAL_NUM_MBUFS) + param_total_num_mbufs = (unsigned)n; + else + rte_exit(EXIT_FAILURE, +- "total-num-mbufs should be > 1024\n"); ++ "total-num-mbufs should be > %d\n", ++ MIN_TOTAL_NUM_MBUFS); + } + if (!strcmp(lgopts[opt_idx].name, "max-pkt-len")) { + n = atoi(optarg); +diff --git a/dpdk/app/test-pmd/rxonly.c b/dpdk/app/test-pmd/rxonly.c +index d1a579d8d8..04457010f4 100644 +--- a/dpdk/app/test-pmd/rxonly.c ++++ b/dpdk/app/test-pmd/rxonly.c +@@ -68,9 +68,17 @@ pkt_burst_receive(struct fwd_stream *fs) + get_end_cycles(fs, start_tsc); + } + ++static void ++stream_init_receive(struct fwd_stream *fs) ++{ ++ fs->disabled = ports[fs->rx_port].rxq[fs->rx_queue].state == ++ RTE_ETH_QUEUE_STATE_STOPPED; ++} ++ + struct fwd_engine rx_only_engine = { + .fwd_mode_name = "rxonly", + .port_fwd_begin = NULL, + .port_fwd_end = NULL, ++ .stream_init = stream_init_receive, + .packet_fwd = pkt_burst_receive, + }; +diff --git a/dpdk/app/test-pmd/shared_rxq_fwd.c b/dpdk/app/test-pmd/shared_rxq_fwd.c +index da54a383fd..2e9047804b 100644 +--- a/dpdk/app/test-pmd/shared_rxq_fwd.c ++++ b/dpdk/app/test-pmd/shared_rxq_fwd.c +@@ -107,9 +107,17 @@ shared_rxq_fwd(struct fwd_stream *fs) + get_end_cycles(fs, start_tsc); + } + ++static void ++shared_rxq_stream_init(struct fwd_stream *fs) ++{ ++ fs->disabled = ports[fs->rx_port].rxq[fs->rx_queue].state == ++ RTE_ETH_QUEUE_STATE_STOPPED; ++} ++ + struct fwd_engine shared_rxq_engine = { + .fwd_mode_name = "shared_rxq", + .port_fwd_begin = NULL, + .port_fwd_end = NULL, ++ .stream_init = shared_rxq_stream_init, + .packet_fwd = shared_rxq_fwd, + }; +diff --git a/dpdk/app/test-pmd/testpmd.c b/dpdk/app/test-pmd/testpmd.c +index 55eb293cc0..3699c5fd64 100644 +--- a/dpdk/app/test-pmd/testpmd.c ++++ b/dpdk/app/test-pmd/testpmd.c +@@ -66,6 +66,9 @@ + #ifdef RTE_EXEC_ENV_WINDOWS + #include + #endif ++#ifdef RTE_NET_BOND ++#include ++#endif + + #include "testpmd.h" + +@@ -84,7 +87,13 @@ + #endif + + #define EXTMEM_HEAP_NAME "extmem" +-#define EXTBUF_ZONE_SIZE RTE_PGSIZE_2M ++/* ++ * Zone size with the malloc overhead (max of debug and release variants) ++ * must fit into the smallest supported hugepage size (2M), ++ * so that an IOVA-contiguous zone of this size can always be allocated ++ * if there are free 2M hugepages. ++ */ ++#define EXTBUF_ZONE_SIZE (RTE_PGSIZE_2M - 4 * RTE_CACHE_LINE_SIZE) + + uint16_t verbose_level = 0; /**< Silent by default. */ + int testpmd_logtype; /**< Log type for testpmd logs */ +@@ -220,6 +229,7 @@ unsigned int xstats_display_num; /**< Size of extended statistics to show */ + * option. Set flag to exit stats period loop after received SIGINT/SIGTERM. + */ + uint8_t f_quit; ++uint8_t cl_quit; /* Quit testpmd from cmdline. */ + + /* + * Max Rx frame size, set by '--max-pkt-len' parameter. +@@ -449,7 +459,7 @@ uint32_t bypass_timeout = RTE_PMD_IXGBE_BYPASS_TMT_OFF; + uint8_t latencystats_enabled; + + /* +- * Lcore ID to serive latency statistics. ++ * Lcore ID to service latency statistics. + */ + lcoreid_t latencystats_lcore_id = -1; + +@@ -591,11 +601,58 @@ eth_dev_configure_mp(uint16_t port_id, uint16_t nb_rx_q, uint16_t nb_tx_q, + return 0; + } + ++static int ++change_bonding_slave_port_status(portid_t bond_pid, bool is_stop) ++{ ++#ifdef RTE_NET_BOND ++ ++ portid_t slave_pids[RTE_MAX_ETHPORTS]; ++ struct rte_port *port; ++ int num_slaves; ++ portid_t slave_pid; ++ int i; ++ ++ num_slaves = rte_eth_bond_slaves_get(bond_pid, slave_pids, ++ RTE_MAX_ETHPORTS); ++ if (num_slaves < 0) { ++ fprintf(stderr, "Failed to get slave list for port = %u\n", ++ bond_pid); ++ return num_slaves; ++ } ++ ++ for (i = 0; i < num_slaves; i++) { ++ slave_pid = slave_pids[i]; ++ port = &ports[slave_pid]; ++ port->port_status = ++ is_stop ? RTE_PORT_STOPPED : RTE_PORT_STARTED; ++ } ++#else ++ RTE_SET_USED(bond_pid); ++ RTE_SET_USED(is_stop); ++#endif ++ return 0; ++} ++ + static int + eth_dev_start_mp(uint16_t port_id) { -- return (char *) b->data + b->size; -+ ovs_assert(b->data || !b->size); -+ return b->data ? (char *) b->data + b->size : NULL; +- if (is_proc_primary()) +- return rte_eth_dev_start(port_id); ++ int ret; ++ ++ if (is_proc_primary()) { ++ ret = rte_eth_dev_start(port_id); ++ if (ret != 0) ++ return ret; ++ ++ struct rte_port *port = &ports[port_id]; ++ ++ /* ++ * Starting a bonded port also starts all slaves under the bonded ++ * device. So if this port is bond device, we need to modify the ++ * port status of these slaves. ++ */ ++ if (port->bond_flag == 1) ++ return change_bonding_slave_port_status(port_id, false); ++ } + + return 0; + } +@@ -603,8 +660,23 @@ eth_dev_start_mp(uint16_t port_id) + static int + eth_dev_stop_mp(uint16_t port_id) + { +- if (is_proc_primary()) +- return rte_eth_dev_stop(port_id); ++ int ret; ++ ++ if (is_proc_primary()) { ++ ret = rte_eth_dev_stop(port_id); ++ if (ret != 0) ++ return ret; ++ ++ struct rte_port *port = &ports[port_id]; ++ ++ /* ++ * Stopping a bonded port also stops all slaves under the bonded ++ * device. So if this port is bond device, we need to modify the ++ * port status of these slaves. ++ */ ++ if (port->bond_flag == 1) ++ return change_bonding_slave_port_status(port_id, true); ++ } + + return 0; + } +@@ -1061,12 +1133,11 @@ setup_extbuf(uint32_t nb_mbufs, uint16_t mbuf_sz, unsigned int socket_id, + ext_num = 0; + break; + } +- mz = rte_memzone_reserve_aligned(mz_name, EXTBUF_ZONE_SIZE, +- socket_id, +- RTE_MEMZONE_IOVA_CONTIG | +- RTE_MEMZONE_1GB | +- RTE_MEMZONE_SIZE_HINT_ONLY, +- EXTBUF_ZONE_SIZE); ++ mz = rte_memzone_reserve(mz_name, EXTBUF_ZONE_SIZE, ++ socket_id, ++ RTE_MEMZONE_IOVA_CONTIG | ++ RTE_MEMZONE_1GB | ++ RTE_MEMZONE_SIZE_HINT_ONLY); + if (mz == NULL) { + /* + * The caller exits on external buffer creation +@@ -1569,10 +1640,10 @@ init_config_port_offloads(portid_t pid, uint32_t socket_id) + + /* Apply Rx offloads configuration */ + for (i = 0; i < port->dev_info.max_rx_queues; i++) +- port->rx_conf[i].offloads = port->dev_conf.rxmode.offloads; ++ port->rxq[i].conf.offloads = port->dev_conf.rxmode.offloads; + /* Apply Tx offloads configuration */ + for (i = 0; i < port->dev_info.max_tx_queues; i++) +- port->tx_conf[i].offloads = port->dev_conf.txmode.offloads; ++ port->txq[i].conf.offloads = port->dev_conf.txmode.offloads; + + if (eth_link_speed) + port->dev_conf.link_speeds = eth_link_speed; +@@ -1759,7 +1830,6 @@ reconfig(portid_t new_port_id, unsigned socket_id) + init_port_config(); } - /* Returns a pointer to byte following the last byte allocated for use (but - * not necessarily in use) in 'b'. */ - static inline void *ofpbuf_end(const struct ofpbuf *b) +- + int + init_fwd_streams(void) { -- return (char *) b->base + b->allocated; -+ ovs_assert(b->base || !b->allocated); -+ return b->base ? (char *) b->base + b->allocated : NULL; +@@ -1978,6 +2048,7 @@ fwd_stats_display(void) + struct rte_port *port; + streamid_t sm_id; + portid_t pt_id; ++ int ret; + int i; + + memset(ports_stats, 0, sizeof(ports_stats)); +@@ -2009,7 +2080,13 @@ fwd_stats_display(void) + pt_id = fwd_ports_ids[i]; + port = &ports[pt_id]; + +- rte_eth_stats_get(pt_id, &stats); ++ ret = rte_eth_stats_get(pt_id, &stats); ++ if (ret != 0) { ++ fprintf(stderr, ++ "%s: Error: failed to get stats (port %u): %d", ++ __func__, pt_id, ret); ++ continue; ++ } + stats.ipackets -= port->stats.ipackets; + stats.opackets -= port->stats.opackets; + stats.ibytes -= port->stats.ibytes; +@@ -2104,11 +2181,16 @@ fwd_stats_reset(void) + { + streamid_t sm_id; + portid_t pt_id; ++ int ret; + int i; + + for (i = 0; i < cur_fwd_config.nb_fwd_ports; i++) { + pt_id = fwd_ports_ids[i]; +- rte_eth_stats_get(pt_id, &ports[pt_id].stats); ++ ret = rte_eth_stats_get(pt_id, &ports[pt_id].stats); ++ if (ret != 0) ++ fprintf(stderr, ++ "%s: Error: failed to clear stats (port %u):%d", ++ __func__, pt_id, ret); + } + for (sm_id = 0; sm_id < cur_fwd_config.nb_fwd_streams; sm_id++) { + struct fwd_stream *fs = fwd_streams[sm_id]; +@@ -2152,6 +2234,12 @@ flush_fwd_rx_queues(void) + for (rxp = 0; rxp < cur_fwd_config.nb_fwd_ports; rxp++) { + for (rxq = 0; rxq < nb_rxq; rxq++) { + port_id = fwd_ports_ids[rxp]; ++ ++ /* Polling stopped queues is prohibited. */ ++ if (ports[port_id].rxq[rxq].state == ++ RTE_ETH_QUEUE_STATE_STOPPED) ++ continue; ++ + /** + * testpmd can stuck in the below do while loop + * if rte_eth_rx_burst() always returns nonzero +@@ -2197,7 +2285,8 @@ run_pkt_fwd_on_lcore(struct fwd_lcore *fc, packet_fwd_t pkt_fwd) + nb_fs = fc->stream_nb; + do { + for (sm_id = 0; sm_id < nb_fs; sm_id++) +- (*pkt_fwd)(fsm[sm_id]); ++ if (!fsm[sm_id]->disabled) ++ (*pkt_fwd)(fsm[sm_id]); + #ifdef RTE_LIB_BITRATESTATS + if (bitrate_enabled != 0 && + bitrate_lcore_id == rte_lcore_id()) { +@@ -2279,6 +2368,7 @@ start_packet_forwarding(int with_tx_first) + { + port_fwd_begin_t port_fwd_begin; + port_fwd_end_t port_fwd_end; ++ stream_init_t stream_init = cur_fwd_eng->stream_init; + unsigned int i; + + if (strcmp(cur_fwd_eng->fwd_mode_name, "rxonly") == 0 && !nb_rxq) +@@ -2309,6 +2399,10 @@ start_packet_forwarding(int with_tx_first) + if (!pkt_fwd_shared_rxq_check()) + return; + ++ if (stream_init != NULL) ++ for (i = 0; i < cur_fwd_config.nb_fwd_streams; i++) ++ stream_init(fwd_streams[i]); ++ + port_fwd_begin = cur_fwd_config.fwd_eng->port_fwd_begin; + if (port_fwd_begin != NULL) { + for (i = 0; i < cur_fwd_config.nb_fwd_ports; i++) { +@@ -2570,7 +2664,7 @@ rx_queue_setup(uint16_t port_id, uint16_t rx_queue_id, + ret = rte_eth_rx_queue_setup(port_id, rx_queue_id, + nb_rx_desc, socket_id, + rx_conf, mp); +- return ret; ++ goto exit; + } + for (i = 0; i < rx_pkt_nb_segs; i++) { + struct rte_eth_rxseg_split *rx_seg = &rx_useg[i].split; +@@ -2579,7 +2673,7 @@ rx_queue_setup(uint16_t port_id, uint16_t rx_queue_id, + * Use last valid pool for the segments with number + * exceeding the pool index. + */ +- mp_n = (i > mbuf_data_size_n) ? mbuf_data_size_n - 1 : i; ++ mp_n = (i >= mbuf_data_size_n) ? mbuf_data_size_n - 1 : i; + mpx = mbuf_pool_find(socket_id, mp_n); + /* Handle zero as mbuf data buffer size. */ + rx_seg->length = rx_pkt_seg_lengths[i] ? +@@ -2595,6 +2689,10 @@ rx_queue_setup(uint16_t port_id, uint16_t rx_queue_id, + socket_id, rx_conf, NULL); + rx_conf->rx_seg = NULL; + rx_conf->rx_nseg = 0; ++exit: ++ ports[port_id].rxq[rx_queue_id].state = rx_conf->rx_deferred_start ? ++ RTE_ETH_QUEUE_STATE_STOPPED : ++ RTE_ETH_QUEUE_STATE_STARTED; + return ret; } - /* Returns the number of bytes of headroom in 'b', that is, the number of bytes -@@ -249,6 +256,11 @@ static inline void *ofpbuf_pull(struct ofpbuf *b, size_t size) +@@ -2722,6 +2820,13 @@ start_port(portid_t pid) + if (pid != pi && pid != (portid_t)RTE_PORT_ALL) + continue; + ++ if (port_is_bonding_slave(pi)) { ++ fprintf(stderr, ++ "Please remove port %d from bonded device.\n", ++ pi); ++ continue; ++ } ++ + need_check_link_status = 0; + port = &ports[pi]; + if (port->port_status == RTE_PORT_STOPPED) +@@ -2790,7 +2895,7 @@ start_port(portid_t pid) + for (k = 0; + k < port->dev_info.max_rx_queues; + k++) +- port->rx_conf[k].offloads |= ++ port->rxq[k].conf.offloads |= + dev_conf.rxmode.offloads; + } + /* Apply Tx offloads configuration */ +@@ -2801,7 +2906,7 @@ start_port(portid_t pid) + for (k = 0; + k < port->dev_info.max_tx_queues; + k++) +- port->tx_conf[k].offloads |= ++ port->txq[k].conf.offloads |= + dev_conf.txmode.offloads; + } + } +@@ -2809,20 +2914,28 @@ start_port(portid_t pid) + port->need_reconfig_queues = 0; + /* setup tx queues */ + for (qi = 0; qi < nb_txq; qi++) { ++ struct rte_eth_txconf *conf = ++ &port->txq[qi].conf; ++ + if ((numa_support) && + (txring_numa[pi] != NUMA_NO_CONFIG)) + diag = rte_eth_tx_queue_setup(pi, qi, + port->nb_tx_desc[qi], + txring_numa[pi], +- &(port->tx_conf[qi])); ++ &(port->txq[qi].conf)); + else + diag = rte_eth_tx_queue_setup(pi, qi, + port->nb_tx_desc[qi], + port->socket_id, +- &(port->tx_conf[qi])); ++ &(port->txq[qi].conf)); + +- if (diag == 0) ++ if (diag == 0) { ++ port->txq[qi].state = ++ conf->tx_deferred_start ? ++ RTE_ETH_QUEUE_STATE_STOPPED : ++ RTE_ETH_QUEUE_STATE_STARTED; + continue; ++ } + + /* Fail to setup tx queue, return */ + if (port->port_status == RTE_PORT_HANDLING) +@@ -2855,7 +2968,7 @@ start_port(portid_t pid) + diag = rx_queue_setup(pi, qi, + port->nb_rx_desc[qi], + rxring_numa[pi], +- &(port->rx_conf[qi]), ++ &(port->rxq[qi].conf), + mp); + } else { + struct rte_mempool *mp = +@@ -2870,7 +2983,7 @@ start_port(portid_t pid) + diag = rx_queue_setup(pi, qi, + port->nb_rx_desc[qi], + port->socket_id, +- &(port->rx_conf[qi]), ++ &(port->rxq[qi].conf), + mp); + } + if (diag == 0) +@@ -3090,11 +3203,48 @@ remove_invalid_ports(void) + nb_cfg_ports = nb_fwd_ports; + } + ++static void ++flush_port_owned_resources(portid_t pi) ++{ ++ mcast_addr_pool_destroy(pi); ++ port_flow_flush(pi); ++ port_flex_item_flush(pi); ++ port_action_handle_flush(pi); ++} ++ ++static void ++clear_bonding_slave_device(portid_t *slave_pids, uint16_t num_slaves) ++{ ++ struct rte_port *port; ++ portid_t slave_pid; ++ uint16_t i; ++ ++ for (i = 0; i < num_slaves; i++) { ++ slave_pid = slave_pids[i]; ++ if (port_is_started(slave_pid) == 1) { ++ if (rte_eth_dev_stop(slave_pid) != 0) ++ fprintf(stderr, "rte_eth_dev_stop failed for port %u\n", ++ slave_pid); ++ ++ port = &ports[slave_pid]; ++ port->port_status = RTE_PORT_STOPPED; ++ } ++ ++ clear_port_slave_flag(slave_pid); ++ ++ /* Close slave device when testpmd quit or is killed. */ ++ if (cl_quit == 1 || f_quit == 1) ++ rte_eth_dev_close(slave_pid); ++ } ++} ++ + void + close_port(portid_t pid) { - ovs_assert(b->size >= size); - void *data = b->data; + portid_t pi; + struct rte_port *port; ++ portid_t slave_pids[RTE_MAX_ETHPORTS]; ++ int num_slaves = 0; + + if (port_id_is_invalid(pid, ENABLED_WARN)) + return; +@@ -3126,9 +3276,20 @@ close_port(portid_t pid) + } + + if (is_proc_primary()) { +- port_flow_flush(pi); +- port_flex_item_flush(pi); ++ flush_port_owned_resources(pi); ++#ifdef RTE_NET_BOND ++ if (port->bond_flag == 1) ++ num_slaves = rte_eth_bond_slaves_get(pi, ++ slave_pids, RTE_MAX_ETHPORTS); ++#endif + rte_eth_dev_close(pi); ++ /* ++ * If this port is bonded device, all slaves under the ++ * device need to be removed or closed. ++ */ ++ if (port->bond_flag == 1 && num_slaves > 0) ++ clear_bonding_slave_device(slave_pids, ++ num_slaves); + } + + free_xstats_display_info(pi); +@@ -3272,7 +3433,7 @@ detach_device(struct rte_device *dev) + sibling); + return; + } +- port_flow_flush(sibling); ++ flush_port_owned_resources(sibling); + } + } + +@@ -3339,7 +3500,7 @@ detach_devargs(char *identifier) + rte_devargs_reset(&da); + return; + } +- port_flow_flush(port_id); ++ flush_port_owned_resources(port_id); + } + } + +@@ -3645,59 +3806,59 @@ rxtx_port_config(portid_t pid) + struct rte_port *port = &ports[pid]; + + for (qid = 0; qid < nb_rxq; qid++) { +- offloads = port->rx_conf[qid].offloads; +- port->rx_conf[qid] = port->dev_info.default_rxconf; ++ offloads = port->rxq[qid].conf.offloads; ++ port->rxq[qid].conf = port->dev_info.default_rxconf; + + if (rxq_share > 0 && + (port->dev_info.dev_capa & RTE_ETH_DEV_CAPA_RXQ_SHARE)) { + /* Non-zero share group to enable RxQ share. */ +- port->rx_conf[qid].share_group = pid / rxq_share + 1; +- port->rx_conf[qid].share_qid = qid; /* Equal mapping. */ ++ port->rxq[qid].conf.share_group = pid / rxq_share + 1; ++ port->rxq[qid].conf.share_qid = qid; /* Equal mapping. */ + } + + if (offloads != 0) +- port->rx_conf[qid].offloads = offloads; ++ port->rxq[qid].conf.offloads = offloads; + + /* Check if any Rx parameters have been passed */ + if (rx_pthresh != RTE_PMD_PARAM_UNSET) +- port->rx_conf[qid].rx_thresh.pthresh = rx_pthresh; ++ port->rxq[qid].conf.rx_thresh.pthresh = rx_pthresh; + + if (rx_hthresh != RTE_PMD_PARAM_UNSET) +- port->rx_conf[qid].rx_thresh.hthresh = rx_hthresh; ++ port->rxq[qid].conf.rx_thresh.hthresh = rx_hthresh; + + if (rx_wthresh != RTE_PMD_PARAM_UNSET) +- port->rx_conf[qid].rx_thresh.wthresh = rx_wthresh; ++ port->rxq[qid].conf.rx_thresh.wthresh = rx_wthresh; + + if (rx_free_thresh != RTE_PMD_PARAM_UNSET) +- port->rx_conf[qid].rx_free_thresh = rx_free_thresh; ++ port->rxq[qid].conf.rx_free_thresh = rx_free_thresh; + + if (rx_drop_en != RTE_PMD_PARAM_UNSET) +- port->rx_conf[qid].rx_drop_en = rx_drop_en; ++ port->rxq[qid].conf.rx_drop_en = rx_drop_en; + + port->nb_rx_desc[qid] = nb_rxd; + } + + for (qid = 0; qid < nb_txq; qid++) { +- offloads = port->tx_conf[qid].offloads; +- port->tx_conf[qid] = port->dev_info.default_txconf; ++ offloads = port->txq[qid].conf.offloads; ++ port->txq[qid].conf = port->dev_info.default_txconf; + if (offloads != 0) +- port->tx_conf[qid].offloads = offloads; ++ port->txq[qid].conf.offloads = offloads; + + /* Check if any Tx parameters have been passed */ + if (tx_pthresh != RTE_PMD_PARAM_UNSET) +- port->tx_conf[qid].tx_thresh.pthresh = tx_pthresh; ++ port->txq[qid].conf.tx_thresh.pthresh = tx_pthresh; + + if (tx_hthresh != RTE_PMD_PARAM_UNSET) +- port->tx_conf[qid].tx_thresh.hthresh = tx_hthresh; ++ port->txq[qid].conf.tx_thresh.hthresh = tx_hthresh; + + if (tx_wthresh != RTE_PMD_PARAM_UNSET) +- port->tx_conf[qid].tx_thresh.wthresh = tx_wthresh; ++ port->txq[qid].conf.tx_thresh.wthresh = tx_wthresh; + + if (tx_rs_thresh != RTE_PMD_PARAM_UNSET) +- port->tx_conf[qid].tx_rs_thresh = tx_rs_thresh; ++ port->txq[qid].conf.tx_rs_thresh = tx_rs_thresh; + + if (tx_free_thresh != RTE_PMD_PARAM_UNSET) +- port->tx_conf[qid].tx_free_thresh = tx_free_thresh; ++ port->txq[qid].conf.tx_free_thresh = tx_free_thresh; + + port->nb_tx_desc[qid] = nb_txd; + } +@@ -3778,7 +3939,7 @@ init_port_config(void) + for (i = 0; + i < port->dev_info.nb_rx_queues; + i++) +- port->rx_conf[i].offloads &= ++ port->rxq[i].conf.offloads &= + ~RTE_ETH_RX_OFFLOAD_RSS_HASH; + } + } +@@ -3952,7 +4113,7 @@ init_port_dcb_config(portid_t pid, + if (port_conf.rxmode.mq_mode == RTE_ETH_MQ_RX_VMDQ_DCB) { + port_conf.rxmode.offloads &= ~RTE_ETH_RX_OFFLOAD_RSS_HASH; + for (i = 0; i < nb_rxq; i++) +- rte_port->rx_conf[i].offloads &= ++ rte_port->rxq[i].conf.offloads &= + ~RTE_ETH_RX_OFFLOAD_RSS_HASH; + } + +diff --git a/dpdk/app/test-pmd/testpmd.h b/dpdk/app/test-pmd/testpmd.h +index 2149ecd93a..18abee907c 100644 +--- a/dpdk/app/test-pmd/testpmd.h ++++ b/dpdk/app/test-pmd/testpmd.h +@@ -32,6 +32,8 @@ + #define RTE_PORT_CLOSED (uint16_t)2 + #define RTE_PORT_HANDLING (uint16_t)3 + ++extern uint8_t cl_quit; ++ + /* + * It is used to allocate the memory for hash key. + * The hash key size is NIC dependent. +@@ -72,6 +74,8 @@ + #define NUMA_NO_CONFIG 0xFF + #define UMA_NO_CONFIG 0xFF + ++#define MIN_TOTAL_NUM_MBUFS 1024 ++ + typedef uint8_t lcoreid_t; + typedef uint16_t portid_t; + typedef uint16_t queueid_t; +@@ -134,6 +138,7 @@ struct fwd_stream { + portid_t tx_port; /**< forwarding port of received packets */ + queueid_t tx_queue; /**< TX queue to send forwarded packets */ + streamid_t peer_addr; /**< index of peer ethernet address of packets */ ++ bool disabled; /**< the stream is disabled and should not run */ + + unsigned int retry_enabled; + +@@ -147,6 +152,7 @@ struct fwd_stream { + /**< received packets has bad outer l4 checksum */ + uint64_t rx_bad_outer_ip_csum; + /**< received packets having bad outer ip checksum */ ++ uint64_t ts_skew; /**< TX scheduling timestamp */ + #ifdef RTE_LIB_GRO + unsigned int gro_times; /**< GRO operation times */ + #endif +@@ -216,6 +222,18 @@ struct xstat_display_info { + bool allocated; + }; + ++/** RX queue configuration and state. */ ++struct port_rxqueue { ++ struct rte_eth_rxconf conf; ++ uint8_t state; /**< RTE_ETH_QUEUE_STATE_* value. */ ++}; + -+ if (!size) { -+ return data; -+ } ++/** TX queue configuration and state. */ ++struct port_txqueue { ++ struct rte_eth_txconf conf; ++ uint8_t state; /**< RTE_ETH_QUEUE_STATE_* value. */ ++}; + - b->data = (char*)b->data + size; - b->size = b->size - size; - return data; -@@ -270,7 +282,7 @@ static inline struct ofpbuf *ofpbuf_from_list(const struct ovs_list *list) - static inline bool ofpbuf_equal(const struct ofpbuf *a, const struct ofpbuf *b) + /** + * The data structure associated with each port. + */ +@@ -238,11 +256,12 @@ struct rte_port { + uint8_t dcb_flag; /**< enable dcb */ + uint16_t nb_rx_desc[RTE_MAX_QUEUES_PER_PORT+1]; /**< per queue rx desc number */ + uint16_t nb_tx_desc[RTE_MAX_QUEUES_PER_PORT+1]; /**< per queue tx desc number */ +- struct rte_eth_rxconf rx_conf[RTE_MAX_QUEUES_PER_PORT+1]; /**< per queue rx configuration */ +- struct rte_eth_txconf tx_conf[RTE_MAX_QUEUES_PER_PORT+1]; /**< per queue tx configuration */ ++ struct port_rxqueue rxq[RTE_MAX_QUEUES_PER_PORT+1]; /**< per queue Rx config and state */ ++ struct port_txqueue txq[RTE_MAX_QUEUES_PER_PORT+1]; /**< per queue Tx config and state */ + struct rte_ether_addr *mc_addr_pool; /**< pool of multicast addrs */ + uint32_t mc_addr_nb; /**< nb. of addr. in mc_addr_pool */ +- uint8_t slave_flag; /**< bonding slave port */ ++ uint8_t slave_flag : 1, /**< bonding slave port */ ++ bond_flag : 1; /**< port is bond device */ + struct port_flow *flow_list; /**< Associated flows. */ + struct port_indirect_action *actions_list; + /**< Associated indirect actions. */ +@@ -296,12 +315,14 @@ struct fwd_lcore { + */ + typedef int (*port_fwd_begin_t)(portid_t pi); + typedef void (*port_fwd_end_t)(portid_t pi); ++typedef void (*stream_init_t)(struct fwd_stream *fs); + typedef void (*packet_fwd_t)(struct fwd_stream *fs); + + struct fwd_engine { + const char *fwd_mode_name; /**< Forwarding mode name. */ + port_fwd_begin_t port_fwd_begin; /**< NULL if nothing special to do. */ + port_fwd_end_t port_fwd_end; /**< NULL if nothing special to do. */ ++ stream_init_t stream_init; /**< NULL if nothing special to do. */ + packet_fwd_t packet_fwd; /**< Mandatory. */ + }; + +@@ -880,6 +901,7 @@ int port_action_handle_create(portid_t port_id, uint32_t id, + const struct rte_flow_action *action); + int port_action_handle_destroy(portid_t port_id, + uint32_t n, const uint32_t *action); ++int port_action_handle_flush(portid_t port_id); + struct rte_flow_action_handle *port_action_handle_get_by_id(portid_t port_id, + uint32_t id); + int port_action_handle_update(portid_t port_id, uint32_t id, +@@ -897,6 +919,7 @@ int port_flow_create(portid_t port_id, + int port_action_handle_query(portid_t port_id, uint32_t id); + void update_age_action_context(const struct rte_flow_action *actions, + struct port_flow *pf); ++int mcast_addr_pool_destroy(portid_t port_id); + int port_flow_destroy(portid_t port_id, uint32_t n, const uint32_t *rule); + int port_flow_flush(portid_t port_id); + int port_flow_dump(portid_t port_id, bool dump_all, +@@ -1101,6 +1124,8 @@ extern int flow_parse(const char *src, void *result, unsigned int size, + struct rte_flow_item **pattern, + struct rte_flow_action **actions); + ++const char *rsstypes_to_str(uint64_t rss_type); ++ + /* + * Work-around of a compilation error with ICC on invocations of the + * rte_be_to_cpu_16() function. +diff --git a/dpdk/app/test-pmd/txonly.c b/dpdk/app/test-pmd/txonly.c +index b8497e733d..e1bc78b73d 100644 +--- a/dpdk/app/test-pmd/txonly.c ++++ b/dpdk/app/test-pmd/txonly.c +@@ -59,14 +59,10 @@ uint32_t tx_ip_dst_addr = (198U << 24) | (18 << 16) | (0 << 8) | 2; + static struct rte_ipv4_hdr pkt_ip_hdr; /**< IP header of transmitted packets. */ + RTE_DEFINE_PER_LCORE(uint8_t, _ip_var); /**< IP address variation */ + static struct rte_udp_hdr pkt_udp_hdr; /**< UDP header of tx packets. */ +-RTE_DEFINE_PER_LCORE(uint64_t, timestamp_qskew); +- /**< Timestamp offset per queue */ +-RTE_DEFINE_PER_LCORE(uint32_t, timestamp_idone); /**< Timestamp init done. */ + + static uint64_t timestamp_mask; /**< Timestamp dynamic flag mask */ + static int32_t timestamp_off; /**< Timestamp dynamic field offset */ + static bool timestamp_enable; /**< Timestamp enable */ +-static uint32_t timestamp_init_req; /**< Timestamp initialization request. */ + static uint64_t timestamp_initial[RTE_MAX_ETHPORTS]; + + static void +@@ -174,14 +170,14 @@ update_pkt_header(struct rte_mbuf *pkt, uint32_t total_pkt_len) + sizeof(struct rte_ether_hdr) + + sizeof(struct rte_ipv4_hdr) + + sizeof(struct rte_udp_hdr))); +- /* updata udp pkt length */ ++ /* update UDP packet length */ + udp_hdr = rte_pktmbuf_mtod_offset(pkt, struct rte_udp_hdr *, + sizeof(struct rte_ether_hdr) + + sizeof(struct rte_ipv4_hdr)); + pkt_len = (uint16_t) (pkt_data_len + sizeof(struct rte_udp_hdr)); + udp_hdr->dgram_len = RTE_CPU_TO_BE_16(pkt_len); + +- /* updata ip pkt length and csum */ ++ /* update IP packet length and checksum */ + ip_hdr = rte_pktmbuf_mtod_offset(pkt, struct rte_ipv4_hdr *, + sizeof(struct rte_ether_hdr)); + ip_hdr->hdr_checksum = 0; +@@ -194,7 +190,7 @@ static inline bool + pkt_burst_prepare(struct rte_mbuf *pkt, struct rte_mempool *mbp, + struct rte_ether_hdr *eth_hdr, const uint16_t vlan_tci, + const uint16_t vlan_tci_outer, const uint64_t ol_flags, +- const uint16_t idx, const struct fwd_stream *fs) ++ const uint16_t idx, struct fwd_stream *fs) { - return a->size == b->size && -- memcmp(a->data, b->data, a->size) == 0; -+ (a->size == 0 || memcmp(a->data, b->data, a->size) == 0); - } + struct rte_mbuf *pkt_segs[RTE_MAX_SEGS_PER_PKT]; + struct rte_mbuf *pkt_seg; +@@ -262,11 +258,10 @@ pkt_burst_prepare(struct rte_mbuf *pkt, struct rte_mempool *mbp, + update_pkt_header(pkt, pkt_len); + + if (unlikely(timestamp_enable)) { +- uint64_t skew = RTE_PER_LCORE(timestamp_qskew); ++ uint64_t skew = fs->ts_skew; + struct tx_timestamp timestamp_mark; + +- if (unlikely(timestamp_init_req != +- RTE_PER_LCORE(timestamp_idone))) { ++ if (unlikely(!skew)) { + struct rte_eth_dev_info dev_info; + unsigned int txqs_n; + uint64_t phase; +@@ -289,8 +284,7 @@ pkt_burst_prepare(struct rte_mbuf *pkt, struct rte_mempool *mbp, + */ + skew = timestamp_initial[fs->tx_port] + + tx_pkt_times_inter + phase; +- RTE_PER_LCORE(timestamp_qskew) = skew; +- RTE_PER_LCORE(timestamp_idone) = timestamp_init_req; ++ fs->ts_skew = skew; + } + timestamp_mark.pkt_idx = rte_cpu_to_be_16(idx); + timestamp_mark.queue_idx = rte_cpu_to_be_16(fs->tx_queue); +@@ -300,14 +294,14 @@ pkt_burst_prepare(struct rte_mbuf *pkt, struct rte_mempool *mbp, + pkt->ol_flags |= timestamp_mask; + *RTE_MBUF_DYNFIELD + (pkt, timestamp_off, uint64_t *) = skew; +- RTE_PER_LCORE(timestamp_qskew) = skew; ++ fs->ts_skew = skew; + timestamp_mark.ts = rte_cpu_to_be_64(skew); + } else if (tx_pkt_times_intra) { + skew += tx_pkt_times_intra; + pkt->ol_flags |= timestamp_mask; + *RTE_MBUF_DYNFIELD + (pkt, timestamp_off, uint64_t *) = skew; +- RTE_PER_LCORE(timestamp_qskew) = skew; ++ fs->ts_skew = skew; + timestamp_mark.ts = rte_cpu_to_be_64(skew); + } else { + timestamp_mark.ts = RTE_BE64(0); +@@ -461,7 +455,6 @@ tx_only_begin(portid_t pi) + timestamp_enable = false; + timestamp_mask = 0; + timestamp_off = -1; +- RTE_PER_LCORE(timestamp_qskew) = 0; + dynf = rte_mbuf_dynflag_lookup + (RTE_MBUF_DYNFLAG_TX_TIMESTAMP_NAME, NULL); + if (dynf >= 0) +@@ -504,7 +497,6 @@ tx_only_begin(portid_t pi) + return -EINVAL; + } + } +- timestamp_init_req++; + } - static inline bool ofpbuf_oversized(const struct ofpbuf *ofpacts) -diff --git a/include/openvswitch/shash.h b/include/openvswitch/shash.h -index c249e13e1f..4e7badd4dc 100644 ---- a/include/openvswitch/shash.h -+++ b/include/openvswitch/shash.h -@@ -41,13 +41,24 @@ struct shash { - BUILD_ASSERT_TYPE(SHASH_NODE, struct shash_node *), \ - BUILD_ASSERT_TYPE(SHASH, struct shash *)) + /* Make sure all settings are visible on forwarding cores.*/ +@@ -512,9 +504,17 @@ tx_only_begin(portid_t pi) + return 0; + } --#define SHASH_FOR_EACH_SAFE(SHASH_NODE, NEXT, SHASH) \ -- HMAP_FOR_EACH_SAFE_INIT ( \ -+#define SHASH_FOR_EACH_SAFE_SHORT(SHASH_NODE, SHASH) \ -+ HMAP_FOR_EACH_SAFE_SHORT_INIT ( \ -+ SHASH_NODE, node, &(SHASH)->map, \ -+ BUILD_ASSERT_TYPE(SHASH_NODE, struct shash_node *), \ -+ BUILD_ASSERT_TYPE(SHASH, struct shash *)) ++static void ++tx_only_stream_init(struct fwd_stream *fs) ++{ ++ fs->disabled = ports[fs->tx_port].txq[fs->tx_queue].state == ++ RTE_ETH_QUEUE_STATE_STOPPED; ++} + -+#define SHASH_FOR_EACH_SAFE_LONG(SHASH_NODE, NEXT, SHASH) \ -+ HMAP_FOR_EACH_SAFE_LONG_INIT ( \ - SHASH_NODE, NEXT, node, &(SHASH)->map, \ - BUILD_ASSERT_TYPE(SHASH_NODE, struct shash_node *), \ - BUILD_ASSERT_TYPE(NEXT, struct shash_node *), \ - BUILD_ASSERT_TYPE(SHASH, struct shash *)) + struct fwd_engine tx_only_engine = { + .fwd_mode_name = "txonly", + .port_fwd_begin = tx_only_begin, + .port_fwd_end = NULL, ++ .stream_init = tx_only_stream_init, + .packet_fwd = pkt_burst_transmit, + }; +diff --git a/dpdk/app/test-regex/main.c b/dpdk/app/test-regex/main.c +index 8e665df73c..ca0b0a5d6a 100644 +--- a/dpdk/app/test-regex/main.c ++++ b/dpdk/app/test-regex/main.c +@@ -385,10 +385,13 @@ run_regex(void *args) + char *data_buf = rgxc->data_buf; + long data_len = rgxc->data_len; + long job_len = rgxc->job_len; +- ++ long remainder; ++ long act_job_len = 0; ++ bool last_job = false; + char *buf = NULL; + uint32_t actual_jobs = 0; + uint32_t i; ++ uint32_t job_id; + uint16_t qp_id; + uint16_t dev_id = 0; + uint8_t nb_matches; +@@ -412,8 +415,8 @@ run_regex(void *args) + mbuf_mp = rte_pktmbuf_pool_create(mbuf_pool, + rte_align32pow2(nb_jobs * nb_qps * nb_segs), + 0, 0, (nb_segs == 1) ? MBUF_SIZE : +- (rte_align32pow2(job_len) / nb_segs + +- RTE_PKTMBUF_HEADROOM), ++ (rte_align32pow2(job_len + (data_len % nb_jobs)) / ++ nb_segs + RTE_PKTMBUF_HEADROOM), + rte_socket_id()); + if (mbuf_mp == NULL) { + printf("Error, can't create memory pool\n"); +@@ -459,9 +462,16 @@ run_regex(void *args) + /* Assign each mbuf with the data to handle. */ + actual_jobs = 0; + pos = 0; ++ remainder = data_len % nb_jobs; ++ + /* Allocate the jobs and assign each job with an mbuf. */ + for (i = 0; (pos < data_len) && (i < nb_jobs) ; i++) { +- long act_job_len = RTE_MIN(job_len, data_len - pos); ++ act_job_len = RTE_MIN(job_len, data_len - pos); ++ ++ if (i == (nb_jobs - 1)) { ++ last_job = true; ++ act_job_len += remainder; ++ } + + ops[i] = rte_malloc(NULL, sizeof(*ops[0]) + + nb_max_matches * +@@ -481,7 +491,12 @@ run_regex(void *args) + if (ops[i]->mbuf) { + rte_pktmbuf_attach_extbuf(ops[i]->mbuf, + &buf[pos], 0, act_job_len, &shinfo); +- ops[i]->mbuf->data_len = job_len; ++ ++ if (!last_job) ++ ops[i]->mbuf->data_len = job_len; ++ else ++ ops[i]->mbuf->data_len = act_job_len; ++ + ops[i]->mbuf->pkt_len = act_job_len; + } + } +@@ -509,6 +524,9 @@ run_regex(void *args) + qp = &qps[qp_id]; + qp->total_enqueue = 0; + qp->total_dequeue = 0; ++ /* Re-set user id after dequeue to match data in mbuf. */ ++ for (job_id = 0 ; job_id < nb_jobs; job_id++) ++ qp->ops[job_id]->user_id = job_id; + } + do { + update = false; +@@ -554,10 +572,10 @@ run_regex(void *args) + for (qp_id = 0; qp_id < nb_qps; qp_id++) { + qp = &qps[qp_id]; + time = (long double)qp->cycles / rte_get_timer_hz(); +- printf("Core=%u QP=%u Job=%ld Bytes Time=%Lf sec Perf=%Lf " ++ printf("Core=%u QP=%u Job=%ld Bytes Last Job=%ld Bytes Time=%Lf sec Perf=%Lf " + "Gbps\n", rte_lcore_id(), qp_id + qp_id_base, +- job_len, time, +- (((double)actual_jobs * job_len * nb_iterations * 8) ++ job_len, act_job_len, time, ++ (((double)data_len * nb_iterations * 8) + / time) / 1000000000.0); + } -+#define SHASH_FOR_EACH_SAFE(...) \ -+ OVERLOAD_SAFE_MACRO(SHASH_FOR_EACH_SAFE_LONG, \ -+ SHASH_FOR_EACH_SAFE_SHORT, \ -+ 3, __VA_ARGS__) +@@ -590,10 +608,10 @@ run_regex(void *args) + qp->total_matches += nb_matches; + match = qp->ops[d_ind % actual_jobs]->matches; + for (i = 0; i < nb_matches; i++) { +- printf("start = %ld, len = %d, rule = %d\n", +- match->start_offset + +- d_ind * job_len, +- match->len, match->rule_id); ++ printf("start = %d, len = %d, rule = %d\n", ++ match->start_offset + ++ (int)(qp->ops[d_ind % actual_jobs]->user_id * job_len), ++ match->len, match->rule_id); + match++; + } + } +@@ -714,6 +732,8 @@ main(int argc, char **argv) + rte_exit(EXIT_FAILURE, "Number of QPs must be greater than 0\n"); + if (nb_lcores == 0) + rte_exit(EXIT_FAILURE, "Number of lcores must be greater than 0\n"); ++ if (nb_jobs == 0) ++ rte_exit(EXIT_FAILURE, "Number of jobs must be greater than 0\n"); + if (distribute_qps_to_lcores(nb_lcores, nb_qps, &qps_per_lcore) < 0) + rte_exit(EXIT_FAILURE, "Failed to distribute queues to lcores!\n"); + ret = init_port(&nb_max_payload, rules_file, +diff --git a/dpdk/app/test/meson.build b/dpdk/app/test/meson.build +index 2b480adfba..c13776c0b9 100644 +--- a/dpdk/app/test/meson.build ++++ b/dpdk/app/test/meson.build +@@ -288,8 +288,6 @@ fast_tests = [ + # Tests known to have issues or which don't belong in other tests lists. + extra_test_names = [ + 'alarm_autotest', # ee00af60170b ("test: remove strict timing requirements some tests") +- 'cycles_autotest', # ee00af60170b ("test: remove strict timing requirements some tests") +- 'delay_us_sleep_autotest', # ee00af60170b ("test: remove strict timing requirements some tests") + 'red_autotest', # https://bugs.dpdk.org/show_bug.cgi?id=826 + ] + +@@ -492,7 +490,7 @@ dpdk_test = executable('dpdk-test', + driver_install_path), + install: true) + +-has_hugepage = run_command('has-hugepage.sh').stdout().strip() != '0' ++has_hugepage = run_command('has-hugepage.sh', check: true).stdout().strip() != '0' + message('hugepage availability: @0@'.format(has_hugepage)) + + # some perf tests (eg: memcpy perf autotest)take very long +diff --git a/dpdk/app/test/test_barrier.c b/dpdk/app/test/test_barrier.c +index 6d6d48749c..ec69af25bf 100644 +--- a/dpdk/app/test/test_barrier.c ++++ b/dpdk/app/test/test_barrier.c +@@ -11,7 +11,7 @@ + * (https://en.wikipedia.org/wiki/Peterson%27s_algorithm) + * for two execution units to make sure that rte_smp_mb() prevents + * store-load reordering to happen. +- * Also when executed on a single lcore could be used as a approxiamate ++ * Also when executed on a single lcore could be used as a approximate + * estimation of number of cycles particular implementation of rte_smp_mb() + * will take. + */ +diff --git a/dpdk/app/test/test_bpf.c b/dpdk/app/test/test_bpf.c +index 46bcb51f86..d70bb0fe85 100644 +--- a/dpdk/app/test/test_bpf.c ++++ b/dpdk/app/test/test_bpf.c +@@ -23,7 +23,7 @@ + /* + * Basic functional tests for librte_bpf. + * The main procedure - load eBPF program, execute it and +- * compare restuls with expected values. ++ * compare results with expected values. + */ + + struct dummy_offset { +@@ -2707,7 +2707,7 @@ test_ld_mbuf1_check(uint64_t rc, const void *arg) + } + + /* +- * same as ld_mbuf1, but then trancate the mbuf by 1B, ++ * same as ld_mbuf1, but then truncate the mbuf by 1B, + * so load of last 4B fail. + */ + static void +@@ -3250,7 +3250,16 @@ test_bpf(void) + + REGISTER_TEST_COMMAND(bpf_autotest, test_bpf); + +-#ifdef RTE_HAS_LIBPCAP ++#ifndef RTE_HAS_LIBPCAP + - void shash_init(struct shash *); - void shash_destroy(struct shash *); - void shash_destroy_free_data(struct shash *); -diff --git a/include/openvswitch/util.h b/include/openvswitch/util.h -index 228b185c3a..8e6c46a85f 100644 ---- a/include/openvswitch/util.h -+++ b/include/openvswitch/util.h -@@ -145,6 +145,150 @@ OVS_NO_RETURN void ovs_assert_failure(const char *, const char *, const char *); - #define INIT_CONTAINER(OBJECT, POINTER, MEMBER) \ - ((OBJECT) = NULL, ASSIGN_CONTAINER(OBJECT, POINTER, MEMBER)) ++static int ++test_bpf_convert(void) ++{ ++ printf("BPF convert RTE_HAS_LIBPCAP is undefined, skipping test\n"); ++ return TEST_SKIPPED; ++} ++ ++#else + #include + + static void +@@ -3259,8 +3268,10 @@ test_bpf_dump(struct bpf_program *cbf, const struct rte_bpf_prm *prm) + printf("cBPF program (%u insns)\n", cbf->bf_len); + bpf_dump(cbf, 1); + +- printf("\neBPF program (%u insns)\n", prm->nb_ins); +- rte_bpf_dump(stdout, prm->ins, prm->nb_ins); ++ if (prm != NULL) { ++ printf("\neBPF program (%u insns)\n", prm->nb_ins); ++ rte_bpf_dump(stdout, prm->ins, prm->nb_ins); ++ } + } + + static int +@@ -3446,5 +3457,6 @@ test_bpf_convert(void) + return rc; + } + +-REGISTER_TEST_COMMAND(bpf_convert_autotest, test_bpf_convert); + #endif /* RTE_HAS_LIBPCAP */ ++ ++REGISTER_TEST_COMMAND(bpf_convert_autotest, test_bpf_convert); +diff --git a/dpdk/app/test/test_compressdev.c b/dpdk/app/test/test_compressdev.c +index c63b5b6737..57c566aa92 100644 +--- a/dpdk/app/test/test_compressdev.c ++++ b/dpdk/app/test/test_compressdev.c +@@ -1256,7 +1256,7 @@ test_deflate_comp_run(const struct interim_data_params *int_data, + /* + * Store original operation index in private data, + * since ordering does not have to be maintained, +- * when dequeueing from compressdev, so a comparison ++ * when dequeuing from compressdev, so a comparison + * at the end of the test can be done. + */ + priv_data = (struct priv_op_data *) (ops[i] + 1); +diff --git a/dpdk/app/test/test_crc.c b/dpdk/app/test/test_crc.c +index bf1d344359..8231f81e4a 100644 +--- a/dpdk/app/test/test_crc.c ++++ b/dpdk/app/test/test_crc.c +@@ -80,6 +80,8 @@ test_crc_calc(void) + + /* 32-bit ethernet CRC: Test 2 */ + test_data = rte_zmalloc(NULL, CRC32_VEC_LEN1, 0); ++ if (test_data == NULL) ++ return -7; + + for (i = 0; i < CRC32_VEC_LEN1; i += 12) + rte_memcpy(&test_data[i], crc32_vec1, 12); +diff --git a/dpdk/app/test/test_cryptodev.c b/dpdk/app/test/test_cryptodev.c +index 10b48cdadb..b11be735d0 100644 +--- a/dpdk/app/test/test_cryptodev.c ++++ b/dpdk/app/test/test_cryptodev.c +@@ -209,6 +209,7 @@ process_sym_raw_dp_op(uint8_t dev_id, uint16_t qp_id, + int enqueue_status, dequeue_status; + struct crypto_unittest_params *ut_params = &unittest_params; + int is_sgl = sop->m_src->nb_segs > 1; ++ int is_oop = 0; + + ctx_service_size = rte_cryptodev_get_raw_dp_ctx_size(dev_id); + if (ctx_service_size < 0) { +@@ -247,6 +248,9 @@ process_sym_raw_dp_op(uint8_t dev_id, uint16_t qp_id, + + ofs.raw = 0; + ++ if ((sop->m_dst != NULL) && (sop->m_dst != sop->m_src)) ++ is_oop = 1; ++ + if (is_cipher && is_auth) { + cipher_offset = sop->cipher.data.offset; + cipher_len = sop->cipher.data.length; +@@ -277,6 +281,8 @@ process_sym_raw_dp_op(uint8_t dev_id, uint16_t qp_id, + if (is_sgl) { + uint32_t remaining_off = auth_offset + auth_len; + struct rte_mbuf *sgl_buf = sop->m_src; ++ if (is_oop) ++ sgl_buf = sop->m_dst; + + while (remaining_off >= rte_pktmbuf_data_len(sgl_buf) + && sgl_buf->next != NULL) { +@@ -293,7 +299,8 @@ process_sym_raw_dp_op(uint8_t dev_id, uint16_t qp_id, + /* Then check if digest-encrypted conditions are met */ + if ((auth_offset + auth_len < cipher_offset + cipher_len) && + (digest.iova == auth_end_iova) && is_sgl) +- max_len = RTE_MAX(max_len, auth_offset + auth_len + ++ max_len = RTE_MAX(max_len, ++ auth_offset + auth_len + + ut_params->auth_xform.auth.digest_length); + + } else if (is_cipher) { +@@ -356,7 +363,7 @@ process_sym_raw_dp_op(uint8_t dev_id, uint16_t qp_id, + + sgl.num = n; + /* Out of place */ +- if (sop->m_dst != NULL) { ++ if (is_oop) { + dest_sgl.vec = dest_data_vec; + vec.dest_sgl = &dest_sgl; + n = rte_crypto_mbuf_to_vec(sop->m_dst, 0, max_len, +@@ -6023,7 +6030,7 @@ test_zuc_encryption(const struct wireless_test_data *tdata) + retval = create_wireless_algo_cipher_operation(tdata->cipher_iv.data, + tdata->cipher_iv.len, + tdata->plaintext.len, +- 0); ++ tdata->validCipherOffsetInBits.len); + if (retval < 0) + return retval; + +@@ -6118,7 +6125,7 @@ test_zuc_encryption_sgl(const struct wireless_test_data *tdata) + /* Create ZUC operation */ + retval = create_wireless_algo_cipher_operation(tdata->cipher_iv.data, + tdata->cipher_iv.len, tdata->plaintext.len, +- 0); ++ tdata->validCipherOffsetInBits.len); + if (retval < 0) + return retval; + +@@ -6226,8 +6233,8 @@ test_zuc_authentication(const struct wireless_test_data *tdata) + else + ut_params->op = process_crypto_request(ts_params->valid_devs[0], + ut_params->op); +- ut_params->obuf = ut_params->op->sym->m_src; + TEST_ASSERT_NOT_NULL(ut_params->op, "failed to retrieve obuf"); ++ ut_params->obuf = ut_params->op->sym->m_src; + ut_params->digest = rte_pktmbuf_mtod(ut_params->obuf, uint8_t *) + + plaintext_pad_len; + +@@ -6553,7 +6560,7 @@ test_zuc_auth_cipher_sgl(const struct wireless_test_data *tdata, + retval = create_wireless_algo_auth_cipher_operation( + tdata->digest.data, tdata->digest.len, + tdata->cipher_iv.data, tdata->cipher_iv.len, +- NULL, 0, ++ tdata->auth_iv.data, tdata->auth_iv.len, + (tdata->digest.offset_bytes == 0 ? + (verify ? ciphertext_pad_len : plaintext_pad_len) + : tdata->digest.offset_bytes), +@@ -6870,7 +6877,7 @@ test_snow3g_decryption_with_digest_test_case_1(void) + } + + /* +- * Function prepare data for hash veryfication test case. ++ * Function prepare data for hash verification test case. + * Digest is allocated in 4 last bytes in plaintext, pattern. + */ + snow3g_hash_test_vector_setup(&snow3g_test_case_7, &snow3g_hash_data); +@@ -10540,9 +10547,11 @@ test_authenticated_encryption_oop(const struct aead_test_data *tdata) + rte_cryptodev_info_get(ts_params->valid_devs[0], &dev_info); + uint64_t feat_flags = dev_info.feature_flags; + +- if ((global_api_test_type == CRYPTODEV_RAW_API_TEST) && +- (!(feat_flags & RTE_CRYPTODEV_FF_SYM_RAW_DP))) ++ if ((global_api_test_type == CRYPTODEV_RAW_API_TEST) || ++ (!(feat_flags & RTE_CRYPTODEV_FF_SYM_RAW_DP))) { ++ printf("Device does not support RAW data-path APIs.\n"); + return TEST_SKIPPED; ++ } + + /* not supported with CPU crypto */ + if (gbl_action_type == RTE_SECURITY_ACTION_TYPE_CPU_CRYPTO) +@@ -15690,7 +15699,7 @@ test_cryptodev_dpaa2_sec_raw_api(void) + static int + test_cryptodev_dpaa_sec_raw_api(void) + { +- static const char *pmd_name = RTE_STR(CRYPTODEV_NAME_DPAA2_SEC_PMD); ++ static const char *pmd_name = RTE_STR(CRYPTODEV_NAME_DPAA_SEC_PMD); + int ret; + + ret = require_feature_flag(pmd_name, RTE_CRYPTODEV_FF_SYM_RAW_DP, +diff --git a/dpdk/app/test/test_cryptodev_asym.c b/dpdk/app/test/test_cryptodev_asym.c +index 9d19a6d6d9..1131290e88 100644 +--- a/dpdk/app/test/test_cryptodev_asym.c ++++ b/dpdk/app/test/test_cryptodev_asym.c +@@ -558,7 +558,7 @@ test_one_case(const void *test_case, int sessionless) + status = test_cryptodev_asym_op( + &testsuite_params, + &tc, test_msg, sessionless, i, +- RTE_RSA_KET_TYPE_QT); ++ RTE_RSA_KEY_TYPE_QT); + } + if (status) + break; +diff --git a/dpdk/app/test/test_cryptodev_rsa_test_vectors.h b/dpdk/app/test/test_cryptodev_rsa_test_vectors.h +index 48a72e1492..04539a1ecf 100644 +--- a/dpdk/app/test/test_cryptodev_rsa_test_vectors.h ++++ b/dpdk/app/test/test_cryptodev_rsa_test_vectors.h +@@ -378,7 +378,7 @@ struct rte_crypto_asym_xform rsa_xform_crt = { + .data = rsa_e, + .length = sizeof(rsa_e) + }, +- .key_type = RTE_RSA_KET_TYPE_QT, ++ .key_type = RTE_RSA_KEY_TYPE_QT, + .qt = { + .p = { + .data = rsa_p, +diff --git a/dpdk/app/test/test_cryptodev_snow3g_test_vectors.h b/dpdk/app/test/test_cryptodev_snow3g_test_vectors.h +index bbe05662be..b49a07bcf2 100644 +--- a/dpdk/app/test/test_cryptodev_snow3g_test_vectors.h ++++ b/dpdk/app/test/test_cryptodev_snow3g_test_vectors.h +@@ -138,11 +138,11 @@ struct snow3g_test_data snow3g_test_case_2 = { + .len = 16 + }, + .cipher_iv = { +- .data = { ++ .data = { + 0xE2, 0x8B, 0xCF, 0x7B, 0xC0, 0x00, 0x00, 0x00, + 0xE2, 0x8B, 0xCF, 0x7B, 0xC0, 0x00, 0x00, 0x00 + }, +- .len = 16 ++ .len = 16 + }, + .plaintext = { + .data = { +@@ -359,8 +359,8 @@ struct snow3g_test_data snow3g_auth_cipher_test_case_1 = { + }, + .cipher_iv = { + .data = { +- 0x14, 0x79, 0x3E, 0x41, 0x03, 0x97, 0xE8, 0xFD, +- 0x94, 0x79, 0x3E, 0x41, 0x03, 0x97, 0x68, 0xFD ++ 0x72, 0xA4, 0xF2, 0x0F, 0x48, 0x00, 0x00, 0x00, ++ 0x72, 0xA4, 0xF2, 0x0F, 0x48, 0x00, 0x00, 0x00 + }, + .len = 16 + }, +@@ -383,13 +383,13 @@ struct snow3g_test_data snow3g_auth_cipher_test_case_1 = { + .len = 384 + }, + .ciphertext = { +- .data = { +- 0x95, 0x2E, 0x5A, 0xE1, 0x50, 0xB8, 0x59, 0x2A, +- 0x9B, 0xA0, 0x38, 0xA9, 0x8E, 0x2F, 0xED, 0xAB, +- 0xFD, 0xC8, 0x3B, 0x47, 0x46, 0x0B, 0x50, 0x16, +- 0xEC, 0x88, 0x45, 0xB6, 0x05, 0xC7, 0x54, 0xF8, +- 0xBD, 0x91, 0xAA, 0xB6, 0xA4, 0xDC, 0x64, 0xB4, +- 0xCB, 0xEB, 0x97, 0x06, 0x4C, 0xF7, 0x02, 0x3D ++ .data = { ++ 0x86, 0x4F, 0x4D, 0xE8, 0x86, 0xE6, 0x3E, 0x66, ++ 0x52, 0x97, 0xC7, 0x62, 0xAE, 0x8E, 0xA2, 0xDB, ++ 0x01, 0xD6, 0x33, 0xA9, 0xA4, 0xCE, 0x02, 0xD5, ++ 0xC2, 0xC5, 0x5F, 0x90, 0xE0, 0x89, 0x48, 0xD4, ++ 0x92, 0xF4, 0xE5, 0x9A, 0xDA, 0x13, 0x76, 0xFF, ++ 0x6E, 0x76, 0x6B, 0x71, 0x62, 0x28, 0xB2, 0xEC + }, + .len = 384 + }, +@@ -428,15 +428,15 @@ struct snow3g_test_data snow3g_test_case_7 = { + }, + .cipher_iv = { + .data = { +- 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, +- 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08 ++ 0xE2, 0x8B, 0xCF, 0x7B, 0xC0, 0x00, 0x00, 0x00, ++ 0xE2, 0x8B, 0xCF, 0x7B, 0xC0, 0x00, 0x00, 0x00 + }, + .len = 16 + }, + .auth_iv = { + .data = { +- 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, +- 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 ++ 0x36, 0xAF, 0x61, 0x44, 0x98, 0x38, 0xF0, 0x3A, ++ 0x36, 0xAF, 0x61, 0x44, 0x98, 0x38, 0xF0, 0x3A + }, + .len = 16 + }, +@@ -457,28 +457,28 @@ struct snow3g_test_data snow3g_test_case_7 = { + 0x5A, 0x5A, 0x5A, 0x5A, 0x5A, 0x5A, 0x5A, 0x5A, + 0x5A, 0x5A, 0x5A, 0x5A, 0x5A, 0x5A, 0x5A, 0x5A, + 0x5A, 0x5A, 0x5A, 0x5A, 0x5A, 0x5A, 0x5A, 0x5A, +- 0x5A, 0x5A, 0x5A, 0x5A, 0xF1, 0x9E, 0x2B, 0x6F, ++ 0x5A, 0x5A, 0x5A, 0x5A, 0xBB, 0x2B, 0x8B, 0x15, + }, + .len = 128 << 3 + }, + .ciphertext = { + .data = { +- 0x5A, 0x5A, 0xE4, 0xAD, 0x29, 0xA2, 0x6A, 0xA6, +- 0x20, 0x1D, 0xCD, 0x08, 0x50, 0xD6, 0xE6, 0x47, +- 0xBC, 0x88, 0x08, 0x01, 0x17, 0xFA, 0x47, 0x5B, +- 0x90, 0x40, 0xBA, 0x0C, 0xB5, 0x58, 0xF3, 0x0C, +- 0xA0, 0xD4, 0x98, 0x83, 0x1B, 0xCE, 0x54, 0xE3, +- 0x29, 0x00, 0x3C, 0xA4, 0xAD, 0x74, 0xEE, 0x05, +- 0xA3, 0x6C, 0xD4, 0xAC, 0xC6, 0x30, 0x33, 0xC9, +- 0x37, 0x57, 0x41, 0x9B, 0xD4, 0x73, 0xB9, 0x77, +- 0x70, 0x8B, 0x63, 0xDD, 0x22, 0xB8, 0xE1, 0x85, +- 0xB2, 0x92, 0x7C, 0x37, 0xD3, 0x2E, 0xD9, 0xF4, +- 0x4A, 0x69, 0x25, 0x30, 0xE3, 0x5B, 0x8B, 0xF6, +- 0x0F, 0xDE, 0x0B, 0x92, 0xD5, 0x25, 0x52, 0x6D, +- 0x26, 0xEB, 0x2F, 0x8A, 0x3B, 0x8B, 0x38, 0xE2, +- 0x48, 0xD3, 0x4A, 0x98, 0xF7, 0x3A, 0xC2, 0x46, +- 0x69, 0x8D, 0x73, 0x3E, 0x57, 0x88, 0x2C, 0x80, +- 0xF0, 0xF2, 0x75, 0xB8, 0x7D, 0x27, 0xC6, 0xDA, ++ 0x5A, 0x5A, 0x8A, 0x35, 0xF7, 0x36, 0xDA, 0xD7, ++ 0xC4, 0x2C, 0x10, 0xEA, 0x92, 0x9C, 0x00, 0xF0, ++ 0xAE, 0x35, 0x5E, 0x8D, 0xB6, 0x88, 0x30, 0x66, ++ 0x74, 0x8B, 0xA2, 0x82, 0x5C, 0xA7, 0xF3, 0x54, ++ 0x75, 0x02, 0xA9, 0x90, 0x6B, 0x4B, 0x6A, 0x63, ++ 0xFF, 0x4B, 0x08, 0xFE, 0x11, 0x3C, 0x5A, 0x53, ++ 0xEE, 0x68, 0x14, 0x41, 0x17, 0xCD, 0x7B, 0x27, ++ 0x88, 0xAF, 0x99, 0xE2, 0x9C, 0x86, 0x42, 0x12, ++ 0x97, 0x93, 0xF0, 0xE6, 0xE2, 0xB2, 0x2D, 0xDA, ++ 0x2C, 0x59, 0xB0, 0xA7, 0x09, 0xF6, 0x32, 0xC0, ++ 0x35, 0x9A, 0xD3, 0xBA, 0xDC, 0x8F, 0x2E, 0x18, ++ 0x97, 0x87, 0x44, 0xD6, 0x43, 0xFA, 0x86, 0x5A, ++ 0xB0, 0xA2, 0x5A, 0xB8, 0x5F, 0x57, 0xE3, 0x2F, ++ 0x73, 0x9C, 0x01, 0x3A, 0x02, 0x08, 0x8C, 0xEB, ++ 0xA0, 0x5D, 0x74, 0x58, 0x5A, 0xA1, 0x58, 0x17, ++ 0x5E, 0x86, 0x96, 0xE6, 0x9C, 0xEE, 0x8C, 0xA8 + + }, + .len = 128 << 3 +@@ -493,7 +493,7 @@ struct snow3g_test_data snow3g_test_case_7 = { + }, + .digest = { + .data = { +- 0x7D, 0x27, 0xC6, 0xDA ++ 0x9C, 0xEE, 0x8C, 0xA8 + }, + .len = 4, + .offset_bytes = 124 +@@ -520,15 +520,15 @@ struct snow3g_test_data snow3g_auth_cipher_test_case_2 = { + }, + .cipher_iv = { + .data = { +- 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, +- 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08 ++ 0xFA, 0x55, 0x6B, 0x26, 0x1C, 0x00, 0x00, 0x00, ++ 0xFA, 0x55, 0x6B, 0x26, 0x1C, 0x00, 0x00, 0x00 + }, + .len = 16 + }, + .auth_iv = { + .data = { +- 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, +- 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 ++ 0x3E, 0xDC, 0x87, 0xE2, 0xA4, 0xF2, 0xD8, 0xE2, ++ 0x3E, 0xDC, 0x87, 0xE2, 0xA4, 0xF2, 0xD8, 0xE2 + }, + .len = 16 + }, +@@ -556,22 +556,22 @@ struct snow3g_test_data snow3g_auth_cipher_test_case_2 = { + }, + .ciphertext = { + .data = { +- 0x5A, 0x5A, 0xE4, 0xAD, 0x29, 0xA2, 0x6A, 0xA6, +- 0x20, 0x1D, 0xCD, 0x08, 0x50, 0xD6, 0xE6, 0x47, +- 0xBC, 0x88, 0x08, 0x01, 0x17, 0xFA, 0x47, 0x5B, +- 0x90, 0x40, 0xBA, 0x0C, 0xB5, 0x58, 0xF3, 0x0C, +- 0xA0, 0xD4, 0x98, 0x83, 0x1B, 0xCE, 0x54, 0xE3, +- 0x29, 0x00, 0x3C, 0xA4, 0xAD, 0x74, 0xEE, 0x05, +- 0xA3, 0x6C, 0xD4, 0xAC, 0xC6, 0x30, 0x33, 0xC9, +- 0x37, 0x57, 0x41, 0x9B, 0xD4, 0x73, 0xB9, 0x77, +- 0x70, 0x8B, 0x63, 0xDD, 0x22, 0xB8, 0xE1, 0x85, +- 0xB2, 0x92, 0x7C, 0x37, 0xD3, 0x2E, 0xD9, 0xF4, +- 0x4A, 0x69, 0x25, 0x30, 0xE3, 0x5B, 0x8B, 0xF6, +- 0x0F, 0xDE, 0x0B, 0x92, 0xD5, 0x25, 0x52, 0x6D, +- 0x26, 0xEB, 0x2F, 0x8A, 0x3B, 0x8B, 0x38, 0xE2, +- 0x48, 0xD3, 0x4A, 0x98, 0xF7, 0x3A, 0xC2, 0x46, +- 0x69, 0x8D, 0x73, 0x3E, 0x57, 0x88, 0x2C, 0x80, +- 0xF0, 0xF2, 0x75, 0xB8, 0x7D, 0x27, 0xC6, 0xDA, ++ 0x5A, 0x5A, 0xCF, 0xCF, 0x3D, 0x11, 0xBF, 0xD9, ++ 0xC3, 0x7F, 0x7C, 0xA8, 0x1A, 0x9F, 0x9F, 0x34, ++ 0xC5, 0x6E, 0x1B, 0x2C, 0xE0, 0x81, 0x4B, 0x66, ++ 0x87, 0xCB, 0xD5, 0x61, 0x04, 0xED, 0xBC, 0x69, ++ 0x79, 0x86, 0x73, 0x48, 0x69, 0x4A, 0xBA, 0x55, ++ 0x44, 0x6C, 0xEF, 0xD9, 0x34, 0x61, 0x59, 0x67, ++ 0x80, 0x4E, 0x03, 0x95, 0x0A, 0xA1, 0x6C, 0xBA, ++ 0x74, 0xBD, 0xAF, 0x11, 0x4B, 0xE6, 0x98, 0x61, ++ 0x4E, 0xD4, 0x3E, 0xE4, 0x99, 0x55, 0x5C, 0x3A, ++ 0x8C, 0x3E, 0xC0, 0x01, 0x6E, 0x15, 0xE1, 0x0E, ++ 0x71, 0x4C, 0x89, 0x43, 0x8A, 0x48, 0x69, 0x6D, ++ 0x02, 0x10, 0xC6, 0x54, 0x37, 0x18, 0xAA, 0x10, ++ 0x90, 0x80, 0x0B, 0x69, 0x08, 0xB4, 0xF9, 0x4D, ++ 0xD1, 0x2E, 0x43, 0xD9, 0x92, 0xAF, 0x06, 0x4A, ++ 0xAF, 0x26, 0x25, 0x77, 0x37, 0xD0, 0xFC, 0x3C, ++ 0xA0, 0xCB, 0xAF, 0x06, 0x95, 0x26, 0x30, 0x38, + + }, + .len = 128 << 3 +@@ -586,7 +586,7 @@ struct snow3g_test_data snow3g_auth_cipher_test_case_2 = { + }, + .digest = { + .data = { +- 0x7D, 0x27, 0xC6, 0xDA ++ 0x95, 0x26, 0x30, 0x38 + }, + .len = 4, + .offset_bytes = 124 +@@ -613,15 +613,15 @@ struct snow3g_test_data snow3g_auth_cipher_test_case_3 = { + }, + .cipher_iv = { + .data = { +- 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, +- 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08 ++ 0x39, 0x8A, 0x59, 0xB4, 0x2C, 0x00, 0x00, 0x00, ++ 0x39, 0x8A, 0x59, 0xB4, 0x2C, 0x00, 0x00, 0x00 + }, + .len = 16 + }, + .auth_iv = { + .data = { +- 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, +- 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 ++ 0x29, 0x6F, 0x39, 0x3C, 0x6B, 0x22, 0x77, 0x37, ++ 0x29, 0x6F, 0x39, 0x3C, 0x6B, 0x22, 0x77, 0x37 + }, + .len = 16 + }, +@@ -636,10 +636,10 @@ struct snow3g_test_data snow3g_auth_cipher_test_case_3 = { + }, + .ciphertext = { + .data = { +- 0x5A, 0x5A, 0xE4, 0xAD, 0x29, 0xA2, 0x6A, 0xA6, +- 0x20, 0x1D, 0xCD, 0x08, 0x50, 0xD6, 0xE6, 0x47, +- 0xBC, 0x88, 0x08, 0x01, 0x17, 0xFA, 0x47, 0x5B, +- 0x90, 0x40, 0xBA, 0x0C, 0xBA, 0x6D, 0x6A, 0x5E, ++ 0x5A, 0x5A, 0x93, 0xB0, 0x3F, 0xA4, 0xEB, 0xD4, ++ 0x51, 0x12, 0x3B, 0x95, 0x93, 0x12, 0xBF, 0xBE, ++ 0xF2, 0xFE, 0xA5, 0xAE, 0xE7, 0xF4, 0x80, 0x3E, ++ 0xB2, 0xD1, 0xFF, 0x5F, 0xD9, 0x32, 0x72, 0xFE, + }, + .len = 32 << 3 + }, +@@ -653,7 +653,7 @@ struct snow3g_test_data snow3g_auth_cipher_test_case_3 = { + }, + .digest = { + .data = { +- 0xBA, 0x6D, 0x6A, 0x5E ++ 0xD9, 0x32, 0x72, 0xFE + }, + .len = 4, + .offset_bytes = 28 +@@ -680,15 +680,15 @@ struct snow3g_test_data snow3g_auth_cipher_partial_digest_encryption = { + }, + .cipher_iv = { + .data = { +- 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, +- 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08 ++ 0x72, 0xA4, 0xF2, 0x0F, 0x48, 0x00, 0x00, 0x00, ++ 0x72, 0xA4, 0xF2, 0x0F, 0x48, 0x00, 0x00, 0x00 + }, + .len = 16 + }, + .auth_iv = { + .data = { +- 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, +- 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 ++ 0x14, 0x79, 0x3E, 0x41, 0x03, 0x97, 0xE8, 0xFD, ++ 0x14, 0x79, 0x3E, 0x41, 0x03, 0x97, 0xE8, 0xFD + }, + .len = 16 + }, +@@ -704,9 +704,9 @@ struct snow3g_test_data snow3g_auth_cipher_partial_digest_encryption = { + .ciphertext = { + .data = { + 0x5A, 0x5A, 0x5A, 0x5A, 0x5A, 0x5A, 0x5A, 0x5A, +- 0x5A, 0x5A, 0x5A, 0x5A, 0x5A, 0x5A, 0xE4, 0xAD, +- 0x29, 0xA2, 0x6A, 0xA6, 0x20, 0x1D, 0xCD, 0x08, +- 0x50, 0xD6, 0xE6, 0x47, 0xB3, 0xBD, 0xC3, 0x08 ++ 0x5A, 0x5A, 0x5A, 0x5A, 0x5A, 0x5A, 0xA2, 0xB7, ++ 0xDF, 0xA7, 0x98, 0xA1, 0xD8, 0xD4, 0x9B, 0x6E, ++ 0x2C, 0x7A, 0x66, 0x15, 0xCC, 0x4C, 0xE5, 0xE0 + }, + .len = 32 << 3 + }, +@@ -720,7 +720,7 @@ struct snow3g_test_data snow3g_auth_cipher_partial_digest_encryption = { + }, + .digest = { + .data = { +- 0xB3, 0xBD, 0xC3, 0x08 ++ 0xCC, 0x4C, 0xE5, 0xE0 + }, + .len = 4, + .offset_bytes = 28 +diff --git a/dpdk/app/test/test_cryptodev_zuc_test_vectors.h b/dpdk/app/test/test_cryptodev_zuc_test_vectors.h +index 5d1d264579..299d7649fe 100644 +--- a/dpdk/app/test/test_cryptodev_zuc_test_vectors.h ++++ b/dpdk/app/test/test_cryptodev_zuc_test_vectors.h +@@ -558,13 +558,13 @@ static struct wireless_test_data zuc_test_case_cipher_200b_auth_200b = { + }, + .auth_iv = { + .data = { +- 0xFA, 0x55, 0x6B, 0x26, 0x1C, 0x00, 0x00, 0x00, +- 0xFA, 0x55, 0x6B, 0x26, 0x1C, 0x00, 0x00, 0x00 ++ 0xFA, 0x55, 0x6B, 0x26, 0x18, 0x00, 0x00, 0x00, ++ 0xFA, 0x55, 0x6B, 0x26, 0x18, 0x00, 0x00, 0x00 + }, + .len = 16 + }, + .digest = { +- .data = {0x01, 0xFE, 0x5E, 0x38}, ++ .data = {0x2F, 0x45, 0x7D, 0x7B}, + .len = 4 + }, + .validAuthLenInBits = { +@@ -631,13 +631,13 @@ static struct wireless_test_data zuc_test_case_cipher_800b_auth_120b = { + }, + .auth_iv = { + .data = { +- 0xFA, 0x55, 0x6B, 0x26, 0x1C, 0x00, 0x00, 0x00, +- 0xFA, 0x55, 0x6B, 0x26, 0x1C, 0x00, 0x00, 0x00 ++ 0xFA, 0x55, 0x6B, 0x26, 0x18, 0x00, 0x00, 0x00, ++ 0xFA, 0x55, 0x6B, 0x26, 0x18, 0x00, 0x00, 0x00 + }, + .len = 16 + }, + .digest = { +- .data = {0x9D, 0x42, 0x1C, 0xEA}, ++ .data = {0xCA, 0xBB, 0x8D, 0x94}, + .len = 4 + }, + .validAuthLenInBits = { +@@ -1166,15 +1166,15 @@ struct wireless_test_data zuc_auth_cipher_test_case_1 = { + }, + .cipher_iv = { + .data = { +- 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, +- 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 ++ 0x66, 0x03, 0x54, 0x92, 0x78, 0x00, 0x00, 0x00, ++ 0x66, 0x03, 0x54, 0x92, 0x78, 0x00, 0x00, 0x00 + }, + .len = 16 + }, + .auth_iv = { + .data = { +- 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, +- 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 ++ 0xFA, 0x55, 0x6B, 0x26, 0x18, 0x00, 0x00, 0x00, ++ 0xFA, 0x55, 0x6B, 0x26, 0x18, 0x00, 0x00, 0x00 + }, + .len = 16 + }, +@@ -1201,22 +1201,22 @@ struct wireless_test_data zuc_auth_cipher_test_case_1 = { + }, + .ciphertext = { + .data = { +- 0x5A, 0x5A, 0xDB, 0x3D, 0xD5, 0xB7, 0xB9, 0x58, +- 0xA5, 0xD3, 0xE3, 0xF9, 0x18, 0x73, 0xB4, 0x74, +- 0x05, 0xF0, 0xE9, 0xB6, 0x5D, 0x9A, 0xE3, 0xFA, +- 0x5D, 0xFD, 0x24, 0x51, 0xAD, 0x73, 0xCA, 0x64, +- 0x91, 0xD5, 0xB3, 0x94, 0x10, 0x91, 0x89, 0xEA, +- 0x73, 0x6F, 0xB0, 0x2A, 0x0A, 0x63, 0x0F, 0x8D, +- 0x64, 0x87, 0xA3, 0x14, 0x6B, 0x93, 0x31, 0x0F, +- 0x14, 0xAD, 0xEA, 0x62, 0x80, 0x3F, 0x44, 0xDD, +- 0x4E, 0x30, 0xFA, 0xC8, 0x0E, 0x5F, 0x46, 0xE7, +- 0x60, 0xEC, 0xDF, 0x8B, 0x94, 0x7D, 0x2E, 0x63, +- 0x48, 0xD9, 0x69, 0x06, 0x13, 0xF2, 0x20, 0x49, +- 0x54, 0xA6, 0xD4, 0x98, 0xF4, 0xF6, 0x1D, 0x4A, +- 0xC9, 0xA5, 0xDA, 0x46, 0x3D, 0xD9, 0x02, 0x47, +- 0x1C, 0x20, 0x73, 0x35, 0x17, 0x1D, 0x81, 0x8D, +- 0x2E, 0xCD, 0x70, 0x37, 0x22, 0x55, 0x3C, 0xF3, +- 0xDA, 0x70, 0x42, 0x12, 0x0E, 0xAA, 0xC4, 0xAB ++ 0x5A, 0x5A, 0x94, 0xE7, 0xB8, 0xD7, 0x4E, 0xBB, ++ 0x4C, 0xC3, 0xD1, 0x16, 0xFC, 0x8C, 0xE4, 0x27, ++ 0x44, 0xEC, 0x04, 0x26, 0x60, 0x9C, 0xFF, 0x81, ++ 0xB6, 0x2B, 0x48, 0x1D, 0xEE, 0x26, 0xF7, 0x58, ++ 0x40, 0x38, 0x58, 0xEA, 0x22, 0x23, 0xE6, 0x34, ++ 0x9A, 0x69, 0x32, 0x68, 0xBD, 0xDD, 0x7D, 0xA3, ++ 0xC0, 0x04, 0x79, 0xF0, 0xF1, 0x58, 0x78, 0x5E, ++ 0xD0, 0xDF, 0x27, 0x9A, 0x53, 0x70, 0x5D, 0xFB, ++ 0x1B, 0xCA, 0xBA, 0x97, 0x12, 0x1F, 0x59, 0x6B, ++ 0x75, 0x7B, 0x94, 0xF6, 0xE7, 0xFA, 0x49, 0x6B, ++ 0x7D, 0x7F, 0x8F, 0x0F, 0x78, 0x56, 0x40, 0x52, ++ 0x84, 0x3E, 0xA9, 0xE8, 0x84, 0x6F, 0xEF, 0xFB, ++ 0x4A, 0x48, 0x3A, 0x4C, 0x81, 0x98, 0xDD, 0x17, ++ 0x89, 0x66, 0x3B, 0xC0, 0xEC, 0x71, 0xDB, 0xF6, ++ 0x44, 0xDF, 0xA7, 0x97, 0xB2, 0x9B, 0x84, 0xA7, ++ 0x2D, 0x2D, 0xC1, 0x93, 0x12, 0x37, 0xEA, 0xD2 + }, + .len = 128 << 3 + }, +@@ -1233,7 +1233,7 @@ struct wireless_test_data zuc_auth_cipher_test_case_1 = { + .len = 2 << 3 + }, + .digest = { +- .data = {0x0E, 0xAA, 0xC4, 0xAB}, ++ .data = {0x12, 0x37, 0xEA, 0xD2}, + .len = 4, + .offset_bytes = 124 + } +diff --git a/dpdk/app/test/test_dmadev.c b/dpdk/app/test/test_dmadev.c +index b206db27ae..e40c29c23b 100644 +--- a/dpdk/app/test/test_dmadev.c ++++ b/dpdk/app/test/test_dmadev.c +@@ -774,6 +774,9 @@ test_dmadev_instance(int16_t dev_id) + if (rte_dma_stats_get(dev_id, vchan, &stats) != 0) + ERR_RETURN("Error with rte_dma_stats_get()\n"); + ++ if (rte_dma_burst_capacity(dev_id, vchan) < 32) ++ ERR_RETURN("Error: Device does not have sufficient burst capacity to run tests"); ++ + if (stats.completed != 0 || stats.submitted != 0 || stats.errors != 0) + ERR_RETURN("Error device stats are not all zero: completed = %"PRIu64", " + "submitted = %"PRIu64", errors = %"PRIu64"\n", +@@ -795,7 +798,10 @@ test_dmadev_instance(int16_t dev_id) + goto err; + + /* run some burst capacity tests */ +- if (runtest("burst capacity", test_burst_capacity, 1, dev_id, vchan, CHECK_ERRS) < 0) ++ if (rte_dma_burst_capacity(dev_id, vchan) < 64) ++ printf("DMA Dev %u: insufficient burst capacity (64 required), skipping tests\n", ++ dev_id); ++ else if (runtest("burst capacity", test_burst_capacity, 1, dev_id, vchan, CHECK_ERRS) < 0) + goto err; + + /* to test error handling we can provide null pointers for source or dest in copies. This +diff --git a/dpdk/app/test/test_efd.c b/dpdk/app/test/test_efd.c +index 1b249e0447..c10c48cf37 100644 +--- a/dpdk/app/test/test_efd.c ++++ b/dpdk/app/test/test_efd.c +@@ -98,7 +98,7 @@ static inline uint64_t efd_get_all_sockets_bitmask(void) + unsigned int next_lcore = rte_get_main_lcore(); + const int val_true = 1, val_false = 0; + for (i = 0; i < rte_lcore_count(); i++) { +- all_cpu_sockets_bitmask |= 1 << rte_lcore_to_socket_id(next_lcore); ++ all_cpu_sockets_bitmask |= 1ULL << rte_lcore_to_socket_id(next_lcore); + next_lcore = rte_get_next_lcore(next_lcore, val_false, val_true); + } + +diff --git a/dpdk/app/test/test_fib_perf.c b/dpdk/app/test/test_fib_perf.c +index 86b2f832b8..7a25fe8df7 100644 +--- a/dpdk/app/test/test_fib_perf.c ++++ b/dpdk/app/test/test_fib_perf.c +@@ -346,7 +346,7 @@ test_fib_perf(void) + fib = rte_fib_create(__func__, SOCKET_ID_ANY, &config); + TEST_FIB_ASSERT(fib != NULL); + +- /* Measue add. */ ++ /* Measure add. */ + begin = rte_rdtsc(); + + for (i = 0; i < NUM_ROUTE_ENTRIES; i++) { +diff --git a/dpdk/app/test/test_hash_readwrite_lf_perf.c b/dpdk/app/test/test_hash_readwrite_lf_perf.c +index 8120cf43be..32f9ec9250 100644 +--- a/dpdk/app/test/test_hash_readwrite_lf_perf.c ++++ b/dpdk/app/test/test_hash_readwrite_lf_perf.c +@@ -59,7 +59,7 @@ struct rwc_perf { + uint32_t w_ks_r_hit_nsp[2][NUM_TEST]; + uint32_t w_ks_r_hit_sp[2][NUM_TEST]; + uint32_t w_ks_r_miss[2][NUM_TEST]; +- uint32_t multi_rw[NUM_TEST - 1][2][NUM_TEST]; ++ uint32_t multi_rw[NUM_TEST][2][NUM_TEST]; + uint32_t w_ks_r_hit_extbkt[2][NUM_TEST]; + uint32_t writer_add_del[NUM_TEST]; + }; +diff --git a/dpdk/app/test/test_ipsec.c b/dpdk/app/test/test_ipsec.c +index bc2a3dbc2e..3c6dcdc604 100644 +--- a/dpdk/app/test/test_ipsec.c ++++ b/dpdk/app/test/test_ipsec.c +@@ -543,12 +543,14 @@ struct rte_ipv4_hdr ipv4_outer = { + }; + + static struct rte_mbuf * +-setup_test_string(struct rte_mempool *mpool, +- const char *string, size_t len, uint8_t blocksize) ++setup_test_string(struct rte_mempool *mpool, const char *string, ++ size_t string_len, size_t len, uint8_t blocksize) + { + struct rte_mbuf *m = rte_pktmbuf_alloc(mpool); + size_t t_len = len - (blocksize ? (len % blocksize) : 0); + ++ RTE_VERIFY(len <= string_len); ++ + if (m) { + memset(m->buf_addr, 0, m->buf_len); + char *dst = rte_pktmbuf_append(m, t_len); +@@ -1354,7 +1356,8 @@ test_ipsec_crypto_outb_burst_null_null(int i) + /* Generate input mbuf data */ + for (j = 0; j < num_pkts && rc == 0; j++) { + ut_params->ibuf[j] = setup_test_string(ts_params->mbuf_pool, +- null_plain_data, test_cfg[i].pkt_sz, 0); ++ null_plain_data, sizeof(null_plain_data), ++ test_cfg[i].pkt_sz, 0); + if (ut_params->ibuf[j] == NULL) + rc = TEST_FAILED; + else { +@@ -1472,7 +1475,8 @@ test_ipsec_inline_crypto_inb_burst_null_null(int i) + /* Generate test mbuf data */ + ut_params->obuf[j] = setup_test_string( + ts_params->mbuf_pool, +- null_plain_data, test_cfg[i].pkt_sz, 0); ++ null_plain_data, sizeof(null_plain_data), ++ test_cfg[i].pkt_sz, 0); + if (ut_params->obuf[j] == NULL) + rc = TEST_FAILED; + } +@@ -1540,16 +1544,17 @@ test_ipsec_inline_proto_inb_burst_null_null(int i) + + /* Generate inbound mbuf data */ + for (j = 0; j < num_pkts && rc == 0; j++) { +- ut_params->ibuf[j] = setup_test_string( +- ts_params->mbuf_pool, +- null_plain_data, test_cfg[i].pkt_sz, 0); ++ ut_params->ibuf[j] = setup_test_string(ts_params->mbuf_pool, ++ null_plain_data, sizeof(null_plain_data), ++ test_cfg[i].pkt_sz, 0); + if (ut_params->ibuf[j] == NULL) + rc = TEST_FAILED; + else { + /* Generate test mbuf data */ + ut_params->obuf[j] = setup_test_string( + ts_params->mbuf_pool, +- null_plain_data, test_cfg[i].pkt_sz, 0); ++ null_plain_data, sizeof(null_plain_data), ++ test_cfg[i].pkt_sz, 0); + if (ut_params->obuf[j] == NULL) + rc = TEST_FAILED; + } +@@ -1649,7 +1654,8 @@ test_ipsec_inline_crypto_outb_burst_null_null(int i) + /* Generate test mbuf data */ + for (j = 0; j < num_pkts && rc == 0; j++) { + ut_params->ibuf[j] = setup_test_string(ts_params->mbuf_pool, +- null_plain_data, test_cfg[i].pkt_sz, 0); ++ null_plain_data, sizeof(null_plain_data), ++ test_cfg[i].pkt_sz, 0); + if (ut_params->ibuf[0] == NULL) + rc = TEST_FAILED; + +@@ -1727,15 +1733,17 @@ test_ipsec_inline_proto_outb_burst_null_null(int i) + /* Generate test mbuf data */ + for (j = 0; j < num_pkts && rc == 0; j++) { + ut_params->ibuf[j] = setup_test_string(ts_params->mbuf_pool, +- null_plain_data, test_cfg[i].pkt_sz, 0); ++ null_plain_data, sizeof(null_plain_data), ++ test_cfg[i].pkt_sz, 0); + if (ut_params->ibuf[0] == NULL) + rc = TEST_FAILED; + + if (rc == 0) { + /* Generate test tunneled mbuf data for comparison */ + ut_params->obuf[j] = setup_test_string( +- ts_params->mbuf_pool, +- null_plain_data, test_cfg[i].pkt_sz, 0); ++ ts_params->mbuf_pool, null_plain_data, ++ sizeof(null_plain_data), test_cfg[i].pkt_sz, ++ 0); + if (ut_params->obuf[j] == NULL) + rc = TEST_FAILED; + } +@@ -1804,7 +1812,8 @@ test_ipsec_lksd_proto_inb_burst_null_null(int i) + for (j = 0; j < num_pkts && rc == 0; j++) { + /* packet with sequence number 0 is invalid */ + ut_params->ibuf[j] = setup_test_string(ts_params->mbuf_pool, +- null_encrypted_data, test_cfg[i].pkt_sz, 0); ++ null_encrypted_data, sizeof(null_encrypted_data), ++ test_cfg[i].pkt_sz, 0); + if (ut_params->ibuf[j] == NULL) + rc = TEST_FAILED; + } +diff --git a/dpdk/app/test/test_ipsec_perf.c b/dpdk/app/test/test_ipsec_perf.c +index 92106bf374..7e07805ea3 100644 +--- a/dpdk/app/test/test_ipsec_perf.c ++++ b/dpdk/app/test/test_ipsec_perf.c +@@ -580,8 +580,8 @@ testsuite_teardown(void) + static int + test_libipsec_perf(void) + { +- struct ipsec_sa sa_out; +- struct ipsec_sa sa_in; ++ struct ipsec_sa sa_out = { .sa_prm = { 0 } }; ++ struct ipsec_sa sa_in = { .sa_prm = { 0 } }; + uint32_t i; + int ret; + +diff --git a/dpdk/app/test/test_kni.c b/dpdk/app/test/test_kni.c +index 40ab0d5c4c..2761de9b07 100644 +--- a/dpdk/app/test/test_kni.c ++++ b/dpdk/app/test/test_kni.c +@@ -326,7 +326,7 @@ test_kni_register_handler_mp(void) + + /* Check with the invalid parameters */ + if (rte_kni_register_handlers(kni, NULL) == 0) { +- printf("Unexpectedly register successuflly " ++ printf("Unexpectedly register successfully " + "with NULL ops pointer\n"); + exit(-1); + } +@@ -475,7 +475,7 @@ test_kni_processing(uint16_t port_id, struct rte_mempool *mp) + + /** + * Check multiple processes support on +- * registerring/unregisterring handlers. ++ * registering/unregistering handlers. + */ + if (test_kni_register_handler_mp() < 0) { + printf("fail to check multiple process support\n"); +diff --git a/dpdk/app/test/test_kvargs.c b/dpdk/app/test/test_kvargs.c +index a91ea8dc47..b7b97a0dd9 100644 +--- a/dpdk/app/test/test_kvargs.c ++++ b/dpdk/app/test/test_kvargs.c +@@ -11,7 +11,7 @@ + + #include "test.h" + +-/* incrementd in handler, to check it is properly called once per ++/* incremented in handler, to check it is properly called once per + * key/value association */ + static unsigned count; + +@@ -107,14 +107,14 @@ static int test_valid_kvargs(void) + goto fail; + } + count = 0; +- /* call check_handler() for all entries with key="unexistant_key" */ +- if (rte_kvargs_process(kvlist, "unexistant_key", check_handler, NULL) < 0) { ++ /* call check_handler() for all entries with key="nonexistent_key" */ ++ if (rte_kvargs_process(kvlist, "nonexistent_key", check_handler, NULL) < 0) { + printf("rte_kvargs_process() error\n"); + rte_kvargs_free(kvlist); + goto fail; + } + if (count != 0) { +- printf("invalid count value %d after rte_kvargs_process(unexistant_key)\n", ++ printf("invalid count value %d after rte_kvargs_process(nonexistent_key)\n", + count); + rte_kvargs_free(kvlist); + goto fail; +@@ -135,10 +135,10 @@ static int test_valid_kvargs(void) + rte_kvargs_free(kvlist); + goto fail; + } +- /* count all entries with key="unexistant_key" */ +- count = rte_kvargs_count(kvlist, "unexistant_key"); ++ /* count all entries with key="nonexistent_key" */ ++ count = rte_kvargs_count(kvlist, "nonexistent_key"); + if (count != 0) { +- printf("invalid count value %d after rte_kvargs_count(unexistant_key)\n", ++ printf("invalid count value %d after rte_kvargs_count(nonexistent_key)\n", + count); + rte_kvargs_free(kvlist); + goto fail; +@@ -156,7 +156,7 @@ static int test_valid_kvargs(void) + /* call check_handler() on all entries with key="check", it + * should fail as the value is not recognized by the handler */ + if (rte_kvargs_process(kvlist, "check", check_handler, NULL) == 0) { +- printf("rte_kvargs_process() is success bu should not\n"); ++ printf("rte_kvargs_process() is success but should not\n"); + rte_kvargs_free(kvlist); + goto fail; + } +diff --git a/dpdk/app/test/test_link_bonding.c b/dpdk/app/test/test_link_bonding.c +index dc6fc46b9c..12c50ef393 100644 +--- a/dpdk/app/test/test_link_bonding.c ++++ b/dpdk/app/test/test_link_bonding.c +@@ -181,6 +181,10 @@ configure_ethdev(uint16_t port_id, uint8_t start, uint8_t en_isr) + test_params->nb_tx_q, &default_pmd_conf), + "rte_eth_dev_configure for port %d failed", port_id); + ++ int ret = rte_eth_dev_set_mtu(port_id, 1550); ++ RTE_TEST_ASSERT(ret == 0 || ret == -ENOTSUP, ++ "rte_eth_dev_set_mtu for port %d failed", port_id); ++ + for (q_id = 0; q_id < test_params->nb_rx_q; q_id++) + TEST_ASSERT_SUCCESS(rte_eth_rx_queue_setup(port_id, q_id, RX_RING_SIZE, + rte_eth_dev_socket_id(port_id), &rx_conf_default, +diff --git a/dpdk/app/test/test_link_bonding_rssconf.c b/dpdk/app/test/test_link_bonding_rssconf.c +index f9eae93973..b3d71c6f3a 100644 +--- a/dpdk/app/test/test_link_bonding_rssconf.c ++++ b/dpdk/app/test/test_link_bonding_rssconf.c +@@ -128,6 +128,10 @@ configure_ethdev(uint16_t port_id, struct rte_eth_conf *eth_conf, + RXTX_QUEUE_COUNT, eth_conf) == 0, "Failed to configure device %u", + port_id); + ++ int ret = rte_eth_dev_set_mtu(port_id, 1550); ++ RTE_TEST_ASSERT(ret == 0 || ret == -ENOTSUP, ++ "rte_eth_dev_set_mtu for port %d failed", port_id); ++ + for (rxq = 0; rxq < RXTX_QUEUE_COUNT; rxq++) { + TEST_ASSERT(rte_eth_rx_queue_setup(port_id, rxq, RXTX_RING_SIZE, + rte_eth_dev_socket_id(port_id), NULL, +@@ -464,15 +468,85 @@ test_rss(void) + + TEST_ASSERT_SUCCESS(test_propagate(), "Propagation test failed"); + +- TEST_ASSERT(slave_remove_and_add() == 1, "New slave should be synced"); ++ TEST_ASSERT(slave_remove_and_add() == 1, "remove and add slaves success."); + + remove_slaves_and_stop_bonded_device(); + + return TEST_SUCCESS; + } -+/* Multi-variable container iterators. -+ * -+ * The following macros facilitate safe iteration over data structures -+ * contained in objects. It does so by using an internal iterator variable of -+ * the type of the member object pointer (i.e: pointer to the data structure). -+ */ + -+/* Multi-variable iterator variable name. -+ * Returns the name of the internal iterator variable. + /** +- * Test propagation logic, when RX_RSS mq_mode is turned off for bonding port ++ * Test RSS configuration over bonded and slaves. + */ -+#define ITER_VAR(NAME) NAME ## __iterator__ ++static int ++test_rss_config_lazy(void) ++{ ++ struct rte_eth_rss_conf bond_rss_conf = {0}; ++ struct slave_conf *port; ++ uint8_t rss_key[40]; ++ uint64_t rss_hf; ++ int retval; ++ uint16_t i; ++ uint8_t n; ++ ++ retval = rte_eth_dev_info_get(test_params.bond_port_id, ++ &test_params.bond_dev_info); ++ TEST_ASSERT((retval == 0), "Error during getting device (port %u) info: %s\n", ++ test_params.bond_port_id, strerror(-retval)); ++ ++ rss_hf = test_params.bond_dev_info.flow_type_rss_offloads; ++ if (rss_hf != 0) { ++ bond_rss_conf.rss_key = NULL; ++ bond_rss_conf.rss_hf = rss_hf; ++ retval = rte_eth_dev_rss_hash_update(test_params.bond_port_id, ++ &bond_rss_conf); ++ TEST_ASSERT(retval != 0, "Succeeded in setting bonded port hash function"); ++ } ++ ++ /* Set all keys to zero for all slaves */ ++ FOR_EACH_PORT(n, port) { ++ port = &test_params.slave_ports[n]; ++ retval = rte_eth_dev_rss_hash_conf_get(port->port_id, ++ &port->rss_conf); ++ TEST_ASSERT_SUCCESS(retval, "Cannot get slaves RSS configuration"); ++ memset(port->rss_key, 0, sizeof(port->rss_key)); ++ port->rss_conf.rss_key = port->rss_key; ++ port->rss_conf.rss_key_len = sizeof(port->rss_key); ++ retval = rte_eth_dev_rss_hash_update(port->port_id, ++ &port->rss_conf); ++ TEST_ASSERT(retval != 0, "Succeeded in setting slaves RSS keys"); ++ } ++ ++ /* Set RSS keys for bonded port */ ++ memset(rss_key, 1, sizeof(rss_key)); ++ bond_rss_conf.rss_hf = rss_hf; ++ bond_rss_conf.rss_key = rss_key; ++ bond_rss_conf.rss_key_len = sizeof(rss_key); ++ ++ retval = rte_eth_dev_rss_hash_update(test_params.bond_port_id, ++ &bond_rss_conf); ++ TEST_ASSERT(retval != 0, "Succeeded in setting bonded port RSS keys"); ++ ++ /* Test RETA propagation */ ++ for (i = 0; i < RXTX_QUEUE_COUNT; i++) { ++ FOR_EACH_PORT(n, port) { ++ port = &test_params.slave_ports[n]; ++ retval = reta_set(port->port_id, (i + 1) % RXTX_QUEUE_COUNT, ++ port->dev_info.reta_size); ++ TEST_ASSERT(retval != 0, "Succeeded in setting slaves RETA"); ++ } ++ ++ retval = reta_set(test_params.bond_port_id, i % RXTX_QUEUE_COUNT, ++ test_params.bond_dev_info.reta_size); ++ TEST_ASSERT(retval != 0, "Succeeded in setting bonded port RETA"); ++ } ++ ++ return TEST_SUCCESS; ++} + -+/* Multi-variable initialization. Creates an internal iterator variable that -+ * points to the provided pointer. The type of the iterator variable is -+ * ITER_TYPE*. It must be the same type as &VAR->MEMBER. -+ * -+ * The _EXP version evaluates the extra expressions once. ++/** ++ * Test RSS function logic, when RX_RSS mq_mode is turned off for bonding port + */ + static int + test_rss_lazy(void) +@@ -493,9 +567,7 @@ test_rss_lazy(void) + TEST_ASSERT_SUCCESS(rte_eth_dev_start(test_params.bond_port_id), + "Failed to start bonding port (%d).", test_params.bond_port_id); + +- TEST_ASSERT_SUCCESS(test_propagate(), "Propagation test failed"); +- +- TEST_ASSERT(slave_remove_and_add() == 0, "New slave shouldn't be synced"); ++ TEST_ASSERT_SUCCESS(test_rss_config_lazy(), "Succeeded in setting RSS hash when RX_RSS mq_mode is turned off"); + + remove_slaves_and_stop_bonded_device(); + +diff --git a/dpdk/app/test/test_lpm6_data.h b/dpdk/app/test/test_lpm6_data.h +index c3894f730e..da9b161f20 100644 +--- a/dpdk/app/test/test_lpm6_data.h ++++ b/dpdk/app/test/test_lpm6_data.h +@@ -22,7 +22,7 @@ struct ips_tbl_entry { + * in previous test_lpm6_routes.h . Because this table has only 1000 + * lines, keeping it doesn't make LPM6 test case so large and also + * make the algorithm to generate rule table unnecessary and the +- * algorithm to genertate test input IPv6 and associated expected ++ * algorithm to generate test input IPv6 and associated expected + * next_hop much simple. + */ + +diff --git a/dpdk/app/test/test_mbuf.c b/dpdk/app/test/test_mbuf.c +index f54d1d7c00..2a037a12da 100644 +--- a/dpdk/app/test/test_mbuf.c ++++ b/dpdk/app/test/test_mbuf.c +@@ -2031,8 +2031,6 @@ test_pktmbuf_read_from_offset(struct rte_mempool *pktmbuf_pool) + NULL); + if (data_copy == NULL) + GOTO_FAIL("%s: Error in reading packet data!\n", __func__); +- if (strlen(data_copy) != MBUF_TEST_DATA_LEN2 - 5) +- GOTO_FAIL("%s: Incorrect data length!\n", __func__); + for (off = 0; off < MBUF_TEST_DATA_LEN2 - 5; off++) { + if (data_copy[off] != (char)0xcc) + GOTO_FAIL("Data corrupted at offset %u", off); +@@ -2054,8 +2052,6 @@ test_pktmbuf_read_from_offset(struct rte_mempool *pktmbuf_pool) + data_copy = rte_pktmbuf_read(m, hdr_len, 0, NULL); + if (data_copy == NULL) + GOTO_FAIL("%s: Error in reading packet data!\n", __func__); +- if (strlen(data_copy) != MBUF_TEST_DATA_LEN2) +- GOTO_FAIL("%s: Corrupted data content!\n", __func__); + for (off = 0; off < MBUF_TEST_DATA_LEN2; off++) { + if (data_copy[off] != (char)0xcc) + GOTO_FAIL("Data corrupted at offset %u", off); +diff --git a/dpdk/app/test/test_member.c b/dpdk/app/test/test_member.c +index 40aa4c8627..af9d50915c 100644 +--- a/dpdk/app/test/test_member.c ++++ b/dpdk/app/test/test_member.c +@@ -459,7 +459,7 @@ static int test_member_multimatch(void) + MAX_MATCH, set_ids_cache); + /* + * For cache mode, keys overwrite when signature same. +- * the mutimatch should work like single match. ++ * the multimatch should work like single match. + */ + TEST_ASSERT(ret_ht == M_MATCH_CNT && ret_vbf == M_MATCH_CNT && + ret_cache == 1, +diff --git a/dpdk/app/test/test_memory.c b/dpdk/app/test/test_memory.c +index dbf6871e71..440e5ef838 100644 +--- a/dpdk/app/test/test_memory.c ++++ b/dpdk/app/test/test_memory.c +@@ -25,6 +25,11 @@ + * - Try to read all memory; it should not segfault. + */ + ++/* ++ * ASan complains about accessing unallocated memory. ++ * See: https://bugs.dpdk.org/show_bug.cgi?id=880 + */ -+#define INIT_MULTIVAR(VAR, MEMBER, POINTER, ITER_TYPE) \ -+ INIT_MULTIVAR_EXP(VAR, MEMBER, POINTER, ITER_TYPE, (void) 0) ++__rte_no_asan + static int + check_mem(const struct rte_memseg_list *msl __rte_unused, + const struct rte_memseg *ms, void *arg __rte_unused) +@@ -63,7 +68,7 @@ check_seg_fds(const struct rte_memseg_list *msl, const struct rte_memseg *ms, + /* we're able to get memseg fd - try getting its offset */ + ret = rte_memseg_get_fd_offset_thread_unsafe(ms, &offset); + if (ret < 0) { +- if (errno == ENOTSUP) ++ if (rte_errno == ENOTSUP) + return 1; + return -1; + } +diff --git a/dpdk/app/test/test_mempool.c b/dpdk/app/test/test_mempool.c +index f6c650d11f..8e493eda47 100644 +--- a/dpdk/app/test/test_mempool.c ++++ b/dpdk/app/test/test_mempool.c +@@ -304,7 +304,7 @@ static int test_mempool_single_consumer(void) + } + + /* +- * test function for mempool test based on singple consumer and single producer, ++ * test function for mempool test based on single consumer and single producer, + * can run on one lcore only + */ + static int +@@ -322,7 +322,7 @@ my_mp_init(struct rte_mempool *mp, __rte_unused void *arg) + } + + /* +- * it tests the mempool operations based on singple producer and single consumer ++ * it tests the mempool operations based on single producer and single consumer + */ + static int + test_mempool_sp_sc(void) +diff --git a/dpdk/app/test/test_memzone.c b/dpdk/app/test/test_memzone.c +index 6ddd0fbab5..c9255e5763 100644 +--- a/dpdk/app/test/test_memzone.c ++++ b/dpdk/app/test/test_memzone.c +@@ -543,7 +543,7 @@ test_memzone_reserve_max(void) + } + + if (mz->len != maxlen) { +- printf("Memzone reserve with 0 size did not return bigest block\n"); ++ printf("Memzone reserve with 0 size did not return biggest block\n"); + printf("Expected size = %zu, actual size = %zu\n", + maxlen, mz->len); + rte_dump_physmem_layout(stdout); +@@ -606,7 +606,7 @@ test_memzone_reserve_max_aligned(void) + + if (mz->len < minlen || mz->len > maxlen) { + printf("Memzone reserve with 0 size and alignment %u did not return" +- " bigest block\n", align); ++ " biggest block\n", align); + printf("Expected size = %zu-%zu, actual size = %zu\n", + minlen, maxlen, mz->len); + rte_dump_physmem_layout(stdout); +@@ -1054,7 +1054,7 @@ test_memzone_basic(void) + if (mz != memzone1) + return -1; + +- printf("test duplcate zone name\n"); ++ printf("test duplicate zone name\n"); + mz = rte_memzone_reserve(TEST_MEMZONE_NAME("testzone1"), 100, + SOCKET_ID_ANY, 0); + if (mz != NULL) +diff --git a/dpdk/app/test/test_metrics.c b/dpdk/app/test/test_metrics.c +index e736019ae4..11222133d0 100644 +--- a/dpdk/app/test/test_metrics.c ++++ b/dpdk/app/test/test_metrics.c +@@ -121,7 +121,7 @@ test_metrics_update_value(void) + err = rte_metrics_update_value(RTE_METRICS_GLOBAL, KEY, VALUE); + TEST_ASSERT(err >= 0, "%s, %d", __func__, __LINE__); + +- /* Successful Test: Valid port_id otherthan RTE_METRICS_GLOBAL, key ++ /* Successful Test: Valid port_id other than RTE_METRICS_GLOBAL, key + * and value + */ + err = rte_metrics_update_value(9, KEY, VALUE); +diff --git a/dpdk/app/test/test_pcapng.c b/dpdk/app/test/test_pcapng.c +index c2dbeaf603..34c5e12346 100644 +--- a/dpdk/app/test/test_pcapng.c ++++ b/dpdk/app/test/test_pcapng.c +@@ -109,7 +109,7 @@ test_setup(void) + return -1; + } + +- /* Make a pool for cloned packeets */ ++ /* Make a pool for cloned packets */ + mp = rte_pktmbuf_pool_create_by_ops("pcapng_test_pool", NUM_PACKETS, + 0, 0, + rte_pcapng_mbuf_size(pkt_len), +diff --git a/dpdk/app/test/test_pmd_perf.c b/dpdk/app/test/test_pmd_perf.c +index 0aa9dc1b1c..4094057b27 100644 +--- a/dpdk/app/test/test_pmd_perf.c ++++ b/dpdk/app/test/test_pmd_perf.c +@@ -454,6 +454,7 @@ main_loop(__rte_unused void *args) + #define PACKET_SIZE 64 + #define FRAME_GAP 12 + #define MAC_PREAMBLE 8 ++#define MAX_RETRY_COUNT 5 + struct rte_mbuf *pkts_burst[MAX_PKT_BURST]; + unsigned lcore_id; + unsigned i, portid, nb_rx = 0, nb_tx = 0; +@@ -461,6 +462,8 @@ main_loop(__rte_unused void *args) + int pkt_per_port; + uint64_t diff_tsc; + uint64_t packets_per_second, total_packets; ++ int retry_cnt = 0; ++ int free_pkt = 0; + + lcore_id = rte_lcore_id(); + conf = &lcore_conf[lcore_id]; +@@ -478,10 +481,19 @@ main_loop(__rte_unused void *args) + nb_tx = RTE_MIN(MAX_PKT_BURST, num); + nb_tx = rte_eth_tx_burst(portid, 0, + &tx_burst[idx], nb_tx); ++ if (nb_tx == 0) ++ retry_cnt++; + num -= nb_tx; + idx += nb_tx; ++ if (retry_cnt == MAX_RETRY_COUNT) { ++ retry_cnt = 0; ++ break; ++ } + } + } ++ for (free_pkt = idx; free_pkt < (MAX_TRAFFIC_BURST * conf->nb_ports); ++ free_pkt++) ++ rte_pktmbuf_free(tx_burst[free_pkt]); + printf("Total packets inject to prime ports = %u\n", idx); + + packets_per_second = (link_mbps * 1000 * 1000) / +diff --git a/dpdk/app/test/test_power_cpufreq.c b/dpdk/app/test/test_power_cpufreq.c +index 1a9549527e..4d013cd7bb 100644 +--- a/dpdk/app/test/test_power_cpufreq.c ++++ b/dpdk/app/test/test_power_cpufreq.c +@@ -659,7 +659,7 @@ test_power_cpufreq(void) + /* test of exit power management for an invalid lcore */ + ret = rte_power_exit(TEST_POWER_LCORE_INVALID); + if (ret == 0) { +- printf("Unpectedly exit power management successfully for " ++ printf("Unexpectedly exit power management successfully for " + "lcore %u\n", TEST_POWER_LCORE_INVALID); + rte_power_unset_env(); + return -1; +diff --git a/dpdk/app/test/test_rcu_qsbr.c b/dpdk/app/test/test_rcu_qsbr.c +index ab37a068cd..70404e89e6 100644 +--- a/dpdk/app/test/test_rcu_qsbr.c ++++ b/dpdk/app/test/test_rcu_qsbr.c +@@ -408,7 +408,7 @@ test_rcu_qsbr_synchronize_reader(void *arg) + + /* + * rte_rcu_qsbr_synchronize: Wait till all the reader threads have entered +- * the queiscent state. ++ * the quiescent state. + */ + static int + test_rcu_qsbr_synchronize(void) +@@ -443,7 +443,7 @@ test_rcu_qsbr_synchronize(void) + rte_rcu_qsbr_synchronize(t[0], RTE_MAX_LCORE - 1); + rte_rcu_qsbr_thread_offline(t[0], RTE_MAX_LCORE - 1); + +- /* Test if the API returns after unregisterng all the threads */ ++ /* Test if the API returns after unregistering all the threads */ + for (i = 0; i < RTE_MAX_LCORE; i++) + rte_rcu_qsbr_thread_unregister(t[0], i); + rte_rcu_qsbr_synchronize(t[0], RTE_QSBR_THRID_INVALID); +diff --git a/dpdk/app/test/test_red.c b/dpdk/app/test/test_red.c +index 05936cfee8..33a9f4ebb7 100644 +--- a/dpdk/app/test/test_red.c ++++ b/dpdk/app/test/test_red.c +@@ -1566,10 +1566,10 @@ static void ovfl_check_avg(uint32_t avg) + } + + static struct test_config ovfl_test1_config = { +- .ifname = "queue avergage overflow test interface", ++ .ifname = "queue average overflow test interface", + .msg = "overflow test 1 : use one RED configuration,\n" + " increase average queue size to target level,\n" +- " check maximum number of bits requirte_red to represent avg_s\n\n", ++ " check maximum number of bits required to represent avg_s\n\n", + .htxt = "avg queue size " + "wq_log2 " + "fraction bits " +@@ -1757,12 +1757,12 @@ test_invalid_parameters(void) + printf("%i: rte_red_config_init should have failed!\n", __LINE__); + return -1; + } +- /* min_treshold == max_treshold */ ++ /* min_threshold == max_threshold */ + if (rte_red_config_init(&config, 0, 1, 1, 0) == 0) { + printf("%i: rte_red_config_init should have failed!\n", __LINE__); + return -1; + } +- /* min_treshold > max_treshold */ ++ /* min_threshold > max_threshold */ + if (rte_red_config_init(&config, 0, 2, 1, 0) == 0) { + printf("%i: rte_red_config_init should have failed!\n", __LINE__); + return -1; +diff --git a/dpdk/app/test/test_security.c b/dpdk/app/test/test_security.c +index 060cf1ffa8..059731b65d 100644 +--- a/dpdk/app/test/test_security.c ++++ b/dpdk/app/test/test_security.c +@@ -237,7 +237,7 @@ + * increases .called counter. Function returns value stored in .ret field + * of the structure. + * In case of some parameters in some functions the expected value is unknown +- * and cannot be detrmined prior to call. Such parameters are stored ++ * and cannot be determined prior to call. Such parameters are stored + * in structure and can be compared or analyzed later in test case code. + * + * Below structures and functions follow the rules just described. +diff --git a/dpdk/app/test/test_table_pipeline.c b/dpdk/app/test/test_table_pipeline.c +index aabf4375db..915c451fed 100644 +--- a/dpdk/app/test/test_table_pipeline.c ++++ b/dpdk/app/test/test_table_pipeline.c +@@ -364,7 +364,7 @@ setup_pipeline(int test_type) + .action = RTE_PIPELINE_ACTION_PORT, + {.port_id = port_out_id[i^1]}, + }; +- printf("Setting secont table to output to port\n"); ++ printf("Setting second table to output to port\n"); + + /* Add the default action for the table. */ + ret = rte_pipeline_table_default_entry_add(p, +diff --git a/dpdk/app/test/test_table_tables.c b/dpdk/app/test/test_table_tables.c +index 4ff6ab16aa..494fb6ffaa 100644 +--- a/dpdk/app/test/test_table_tables.c ++++ b/dpdk/app/test/test_table_tables.c +@@ -290,10 +290,10 @@ test_table_lpm(void) + struct rte_mbuf *mbufs[RTE_PORT_IN_BURST_SIZE_MAX]; + void *table; + char *entries[RTE_PORT_IN_BURST_SIZE_MAX]; +- char entry; ++ uint64_t entry; + void *entry_ptr; + int key_found; +- uint32_t entry_size = 1; ++ uint32_t entry_size = sizeof(entry); + + /* Initialize params and create tables */ + struct rte_table_lpm_params lpm_params = { +@@ -355,7 +355,7 @@ test_table_lpm(void) + struct rte_table_lpm_key lpm_key; + lpm_key.ip = 0xadadadad; + +- table = rte_table_lpm_ops.f_create(&lpm_params, 0, 1); ++ table = rte_table_lpm_ops.f_create(&lpm_params, 0, entry_size); + if (table == NULL) + return -9; + +@@ -456,10 +456,10 @@ test_table_lpm_ipv6(void) + struct rte_mbuf *mbufs[RTE_PORT_IN_BURST_SIZE_MAX]; + void *table; + char *entries[RTE_PORT_IN_BURST_SIZE_MAX]; +- char entry; ++ uint64_t entry; + void *entry_ptr; + int key_found; +- uint32_t entry_size = 1; ++ uint32_t entry_size = sizeof(entry); + + /* Initialize params and create tables */ + struct rte_table_lpm_ipv6_params lpm_params = { +diff --git a/dpdk/app/test/test_thash.c b/dpdk/app/test/test_thash.c +index a62530673f..62ba4a9528 100644 +--- a/dpdk/app/test/test_thash.c ++++ b/dpdk/app/test/test_thash.c +@@ -684,7 +684,7 @@ test_predictable_rss_multirange(void) + + /* + * calculate hashes, complements, then adjust keys with +- * complements and recalsulate hashes ++ * complements and recalculate hashes + */ + for (i = 0; i < RTE_DIM(rng_arr); i++) { + for (k = 0; k < 100; k++) { +diff --git a/dpdk/buildtools/binutils-avx512-check.py b/dpdk/buildtools/binutils-avx512-check.py +index a4e14f3593..a0847a23d6 100644 +--- a/dpdk/buildtools/binutils-avx512-check.py ++++ b/dpdk/buildtools/binutils-avx512-check.py +@@ -1,5 +1,5 @@ + #! /usr/bin/env python3 +-# SPDX-License-Identitifer: BSD-3-Clause ++# SPDX-License-Identifier: BSD-3-Clause + # Copyright(c) 2020 Intel Corporation + + import subprocess +@@ -15,7 +15,7 @@ + src = '__asm__("vpgatherqq {}");'.format(gather_params).encode('utf-8') + subprocess.run(cc + ['-c', '-xc', '-o', obj.name, '-'], input=src, check=True) + asm = subprocess.run([objdump, '-d', '--no-show-raw-insn', obj.name], +- capture_output=True, check=True).stdout.decode('utf-8') ++ stdout=subprocess.PIPE, check=True).stdout.decode('utf-8') + if gather_params not in asm: + print('vpgatherqq displacement error with as') + sys.exit(1) +diff --git a/dpdk/buildtools/call-sphinx-build.py b/dpdk/buildtools/call-sphinx-build.py +index 26b199220a..39a60d09fa 100755 +--- a/dpdk/buildtools/call-sphinx-build.py ++++ b/dpdk/buildtools/call-sphinx-build.py +@@ -7,7 +7,7 @@ + import os + from os.path import join + from subprocess import run, PIPE, STDOUT +-from distutils.version import StrictVersion ++from packaging.version import Version + + # assign parameters to variables + (sphinx, version, src, dst, *extra_args) = sys.argv[1:] +@@ -19,7 +19,7 @@ + ver = run([sphinx, '--version'], stdout=PIPE, + stderr=STDOUT).stdout.decode().split()[-1] + sphinx_cmd = [sphinx] + extra_args +-if StrictVersion(ver) >= StrictVersion('1.7'): ++if Version(ver) >= Version('1.7'): + sphinx_cmd += ['-j', 'auto'] + + # find all the files sphinx will process so we can write them as dependencies +diff --git a/dpdk/buildtools/meson.build b/dpdk/buildtools/meson.build +index 22ea0ba375..e1c600e40f 100644 +--- a/dpdk/buildtools/meson.build ++++ b/dpdk/buildtools/meson.build +@@ -31,6 +31,9 @@ if host_machine.system() == 'windows' + pmdinfo += 'llvm-ar' + endif + pmdinfogen += 'coff' ++elif host_machine.system() == 'freebsd' ++ pmdinfo += 'llvm-ar' ++ pmdinfogen += 'elf' + else + pmdinfo += 'ar' + pmdinfogen += 'elf' +@@ -45,7 +48,7 @@ if host_machine.system() != 'windows' + endif + foreach module : python3_required_modules + script = 'import importlib.util; import sys; exit(importlib.util.find_spec("@0@") is None)' +- if run_command(py3, '-c', script.format(module)).returncode() != 0 ++ if run_command(py3, '-c', script.format(module), check: false).returncode() != 0 + error('missing python module: @0@'.format(module)) + endif + endforeach +diff --git a/dpdk/config/arm/arm32_armv8_linux_gcc b/dpdk/config/arm/arm32_armv8_linux_gcc +index 89f8a12881..0d4618ea4a 100644 +--- a/dpdk/config/arm/arm32_armv8_linux_gcc ++++ b/dpdk/config/arm/arm32_armv8_linux_gcc +@@ -1,6 +1,6 @@ + [binaries] + c = 'arm-linux-gnueabihf-gcc' +-cpp = 'arm-linux-gnueabihf-cpp' ++cpp = 'arm-linux-gnueabihf-g++' + ar = 'arm-linux-gnueabihf-gcc-ar' + strip = 'arm-linux-gnueabihf-strip' + pkgconfig = 'arm-linux-gnueabihf-pkg-config' +diff --git a/dpdk/config/arm/arm64_armada_linux_gcc b/dpdk/config/arm/arm64_armada_linux_gcc +index 301418949b..5043b82651 100644 +--- a/dpdk/config/arm/arm64_armada_linux_gcc ++++ b/dpdk/config/arm/arm64_armada_linux_gcc +@@ -1,6 +1,6 @@ + [binaries] + c = 'aarch64-linux-gnu-gcc' +-cpp = 'aarch64-linux-gnu-cpp' ++cpp = 'aarch64-linux-gnu-g++' + ar = 'aarch64-linux-gnu-ar' + as = 'aarch64-linux-gnu-as' + strip = 'aarch64-linux-gnu-strip' +diff --git a/dpdk/config/arm/arm64_armv8_linux_gcc b/dpdk/config/arm/arm64_armv8_linux_gcc +index 5391d35389..5c32f6b9ca 100644 +--- a/dpdk/config/arm/arm64_armv8_linux_gcc ++++ b/dpdk/config/arm/arm64_armv8_linux_gcc +@@ -1,6 +1,6 @@ + [binaries] + c = 'aarch64-linux-gnu-gcc' +-cpp = 'aarch64-linux-gnu-cpp' ++cpp = 'aarch64-linux-gnu-g++' + ar = 'aarch64-linux-gnu-gcc-ar' + strip = 'aarch64-linux-gnu-strip' + pkgconfig = 'aarch64-linux-gnu-pkg-config' +diff --git a/dpdk/config/arm/arm64_bluefield_linux_gcc b/dpdk/config/arm/arm64_bluefield_linux_gcc +index 248a9f031a..df6eccc046 100644 +--- a/dpdk/config/arm/arm64_bluefield_linux_gcc ++++ b/dpdk/config/arm/arm64_bluefield_linux_gcc +@@ -1,6 +1,6 @@ + [binaries] + c = 'aarch64-linux-gnu-gcc' +-cpp = 'aarch64-linux-gnu-cpp' ++cpp = 'aarch64-linux-gnu-g++' + ar = 'aarch64-linux-gnu-gcc-ar' + strip = 'aarch64-linux-gnu-strip' + pkgconfig = 'aarch64-linux-gnu-pkg-config' +diff --git a/dpdk/config/arm/arm64_centriq2400_linux_gcc b/dpdk/config/arm/arm64_centriq2400_linux_gcc +index dfe9110331..ddffc0503a 100644 +--- a/dpdk/config/arm/arm64_centriq2400_linux_gcc ++++ b/dpdk/config/arm/arm64_centriq2400_linux_gcc +@@ -1,6 +1,6 @@ + [binaries] + c = 'aarch64-linux-gnu-gcc' +-cpp = 'aarch64-linux-gnu-cpp' ++cpp = 'aarch64-linux-gnu-g++' + ar = 'aarch64-linux-gnu-gcc-ar' + strip = 'aarch64-linux-gnu-strip' + pkgconfig = 'aarch64-linux-gnu-pkg-config' +diff --git a/dpdk/config/arm/arm64_cn10k_linux_gcc b/dpdk/config/arm/arm64_cn10k_linux_gcc +index 88e5f10945..19068f0ec9 100644 +--- a/dpdk/config/arm/arm64_cn10k_linux_gcc ++++ b/dpdk/config/arm/arm64_cn10k_linux_gcc +@@ -1,6 +1,6 @@ + [binaries] + c = 'aarch64-linux-gnu-gcc' +-cpp = 'aarch64-linux-gnu-cpp' ++cpp = 'aarch64-linux-gnu-g++' + ar = 'aarch64-linux-gnu-gcc-ar' + strip = 'aarch64-linux-gnu-strip' + pkgconfig = 'aarch64-linux-gnu-pkg-config' +diff --git a/dpdk/config/arm/arm64_dpaa_linux_gcc b/dpdk/config/arm/arm64_dpaa_linux_gcc +index e9d5fd31fc..70df99fb02 100644 +--- a/dpdk/config/arm/arm64_dpaa_linux_gcc ++++ b/dpdk/config/arm/arm64_dpaa_linux_gcc +@@ -1,6 +1,6 @@ + [binaries] + c = 'aarch64-linux-gnu-gcc' +-cpp = 'aarch64-linux-gnu-cpp' ++cpp = 'aarch64-linux-gnu-g++' + ar = 'aarch64-linux-gnu-ar' + as = 'aarch64-linux-gnu-as' + strip = 'aarch64-linux-gnu-strip' +diff --git a/dpdk/config/arm/arm64_emag_linux_gcc b/dpdk/config/arm/arm64_emag_linux_gcc +index 9cdd931180..06f5eaecd0 100644 +--- a/dpdk/config/arm/arm64_emag_linux_gcc ++++ b/dpdk/config/arm/arm64_emag_linux_gcc +@@ -1,6 +1,6 @@ + [binaries] + c = 'aarch64-linux-gnu-gcc' +-cpp = 'aarch64-linux-gnu-cpp' ++cpp = 'aarch64-linux-gnu-g++' + ar = 'aarch64-linux-gnu-gcc-ar' + strip = 'aarch64-linux-gnu-strip' + pkgconfig = 'aarch64-linux-gnu-pkg-config' +diff --git a/dpdk/config/arm/arm64_graviton2_linux_gcc b/dpdk/config/arm/arm64_graviton2_linux_gcc +index 8016fd236c..24b2dbcca8 100644 +--- a/dpdk/config/arm/arm64_graviton2_linux_gcc ++++ b/dpdk/config/arm/arm64_graviton2_linux_gcc +@@ -1,6 +1,6 @@ + [binaries] + c = 'aarch64-linux-gnu-gcc' +-cpp = 'aarch64-linux-gnu-cpp' ++cpp = 'aarch64-linux-gnu-g++' + ar = 'aarch64-linux-gnu-gcc-ar' + strip = 'aarch64-linux-gnu-strip' + pkgconfig = 'aarch64-linux-gnu-pkg-config' +diff --git a/dpdk/config/arm/arm64_kunpeng920_linux_gcc b/dpdk/config/arm/arm64_kunpeng920_linux_gcc +index c4685b2458..4a71531e3e 100644 +--- a/dpdk/config/arm/arm64_kunpeng920_linux_gcc ++++ b/dpdk/config/arm/arm64_kunpeng920_linux_gcc +@@ -1,6 +1,6 @@ + [binaries] + c = 'aarch64-linux-gnu-gcc' +-cpp = 'aarch64-linux-gnu-cpp' ++cpp = 'aarch64-linux-gnu-g++' + ar = 'aarch64-linux-gnu-gcc-ar' + strip = 'aarch64-linux-gnu-strip' + pkgconfig = 'aarch64-linux-gnu-pkg-config' +diff --git a/dpdk/config/arm/arm64_kunpeng930_linux_gcc b/dpdk/config/arm/arm64_kunpeng930_linux_gcc +index fb85d2d710..383f0b0313 100644 +--- a/dpdk/config/arm/arm64_kunpeng930_linux_gcc ++++ b/dpdk/config/arm/arm64_kunpeng930_linux_gcc +@@ -1,6 +1,6 @@ + [binaries] + c = 'aarch64-linux-gnu-gcc' +-cpp = 'aarch64-linux-gnu-cpp' ++cpp = 'aarch64-linux-gnu-g++' + ar = 'aarch64-linux-gnu-gcc-ar' + strip = 'aarch64-linux-gnu-strip' + pkgconfig = 'aarch64-linux-gnu-pkg-config' +diff --git a/dpdk/config/arm/arm64_n1sdp_linux_gcc b/dpdk/config/arm/arm64_n1sdp_linux_gcc +index 0df283e2f4..5f6356caa2 100644 +--- a/dpdk/config/arm/arm64_n1sdp_linux_gcc ++++ b/dpdk/config/arm/arm64_n1sdp_linux_gcc +@@ -1,6 +1,6 @@ + [binaries] + c = 'aarch64-linux-gnu-gcc' +-cpp = 'aarch64-linux-gnu-cpp' ++cpp = 'aarch64-linux-gnu-g++' + ar = 'aarch64-linux-gnu-gcc-ar' + strip = 'aarch64-linux-gnu-strip' + pkgconfig = 'aarch64-linux-gnu-pkg-config' +diff --git a/dpdk/config/arm/arm64_n2_linux_gcc b/dpdk/config/arm/arm64_n2_linux_gcc +index 036aee2b0a..82806ba780 100644 +--- a/dpdk/config/arm/arm64_n2_linux_gcc ++++ b/dpdk/config/arm/arm64_n2_linux_gcc +@@ -1,6 +1,6 @@ + [binaries] + c = 'aarch64-linux-gnu-gcc' +-cpp = 'aarch64-linux-gnu-cpp' ++cpp = 'aarch64-linux-gnu-g++' + ar = 'aarch64-linux-gnu-gcc-ar' + strip = 'aarch64-linux-gnu-strip' + pkgconfig = 'aarch64-linux-gnu-pkg-config' +diff --git a/dpdk/config/arm/arm64_octeontx2_linux_gcc b/dpdk/config/arm/arm64_octeontx2_linux_gcc +index 8fbdd3868d..d23b6527ef 100644 +--- a/dpdk/config/arm/arm64_octeontx2_linux_gcc ++++ b/dpdk/config/arm/arm64_octeontx2_linux_gcc +@@ -1,6 +1,6 @@ + [binaries] + c = 'aarch64-linux-gnu-gcc' +-cpp = 'aarch64-linux-gnu-cpp' ++cpp = 'aarch64-linux-gnu-g++' + ar = 'aarch64-linux-gnu-gcc-ar' + strip = 'aarch64-linux-gnu-strip' + pkgconfig = 'aarch64-linux-gnu-pkg-config' +diff --git a/dpdk/config/arm/arm64_stingray_linux_gcc b/dpdk/config/arm/arm64_stingray_linux_gcc +index 319a4a151d..cf98337f0f 100644 +--- a/dpdk/config/arm/arm64_stingray_linux_gcc ++++ b/dpdk/config/arm/arm64_stingray_linux_gcc +@@ -1,6 +1,6 @@ + [binaries] + c = 'aarch64-linux-gnu-gcc' +-cpp = 'aarch64-linux-gnu-cpp' ++cpp = 'aarch64-linux-gnu-g++' + ar = 'aarch64-linux-gnu-gcc-ar' + strip = 'aarch64-linux-gnu-strip' + pkgconfig = 'aarch64-linux-gnu-pkg-config' +diff --git a/dpdk/config/arm/arm64_thunderx2_linux_gcc b/dpdk/config/arm/arm64_thunderx2_linux_gcc +index 69c71cbc82..616f6c263a 100644 +--- a/dpdk/config/arm/arm64_thunderx2_linux_gcc ++++ b/dpdk/config/arm/arm64_thunderx2_linux_gcc +@@ -1,6 +1,6 @@ + [binaries] + c = 'aarch64-linux-gnu-gcc' +-cpp = 'aarch64-linux-gnu-cpp' ++cpp = 'aarch64-linux-gnu-g++' + ar = 'aarch64-linux-gnu-gcc-ar' + strip = 'aarch64-linux-gnu-strip' + pkgconfig = 'aarch64-linux-gnu-pkg-config' +diff --git a/dpdk/config/arm/arm64_thunderxt88_linux_gcc b/dpdk/config/arm/arm64_thunderxt88_linux_gcc +index 372097ba01..131f56465a 100644 +--- a/dpdk/config/arm/arm64_thunderxt88_linux_gcc ++++ b/dpdk/config/arm/arm64_thunderxt88_linux_gcc +@@ -1,6 +1,6 @@ + [binaries] + c = 'aarch64-linux-gnu-gcc' +-cpp = 'aarch64-linux-gnu-cpp' ++cpp = 'aarch64-linux-gnu-g++' + ar = 'aarch64-linux-gnu-gcc-ar' + strip = 'aarch64-linux-gnu-strip' + pkgconfig = 'aarch64-linux-gnu-pkg-config' +diff --git a/dpdk/config/arm/meson.build b/dpdk/config/arm/meson.build +index 213324d262..89a3bf4213 100644 +--- a/dpdk/config/arm/meson.build ++++ b/dpdk/config/arm/meson.build +@@ -49,6 +49,7 @@ implementer_generic = { + ['RTE_ARCH_ARM_NEON_MEMCPY', false], + ['RTE_ARCH_STRICT_ALIGN', true], + ['RTE_ARCH_ARMv8_AARCH32', true], ++ ['RTE_ARCH', 'armv8_aarch32'], + ['RTE_CACHE_LINE_SIZE', 64] + ] + } +@@ -276,7 +277,8 @@ soc_cn10k = { + 'implementer' : '0x41', + 'flags': [ + ['RTE_MAX_LCORE', 24], +- ['RTE_MAX_NUMA_NODES', 1] ++ ['RTE_MAX_NUMA_NODES', 1], ++ ['RTE_MEMPOOL_ALIGN', 128] + ], + 'part_number': '0xd49', + 'extra_march_features': ['crypto'], +@@ -432,11 +434,15 @@ if dpdk_conf.get('RTE_ARCH_32') + else + # armv7 build + dpdk_conf.set('RTE_ARCH_ARMv7', true) ++ dpdk_conf.set('RTE_ARCH', 'armv7') ++ dpdk_conf.set('RTE_MAX_LCORE', 128) ++ dpdk_conf.set('RTE_MAX_NUMA_NODES', 1) + # the minimum architecture supported, armv7-a, needs the following, + machine_args += '-mfpu=neon' + endif + else + # armv8 build ++ dpdk_conf.set('RTE_ARCH', 'armv8') + update_flags = true + soc_config = {} + if not meson.is_cross_build() +@@ -460,7 +466,7 @@ else + # 'Primary Part number', 'Revision'] + detect_vendor = find_program(join_paths(meson.current_source_dir(), + 'armv8_machine.py')) +- cmd = run_command(detect_vendor.path()) ++ cmd = run_command(detect_vendor.path(), check: false) + if cmd.returncode() == 0 + cmd_output = cmd.stdout().to_lower().strip().split(' ') + implementer_id = cmd_output[0] +diff --git a/dpdk/config/meson.build b/dpdk/config/meson.build +index 805d5d51d0..ee12318d4f 100644 +--- a/dpdk/config/meson.build ++++ b/dpdk/config/meson.build +@@ -22,7 +22,8 @@ is_ms_linker = is_windows and (cc.get_id() == 'clang') + # depending on the configuration options + pver = meson.project_version().split('.') + major_version = '@0@.@1@'.format(pver.get(0), pver.get(1)) +-abi_version = run_command(find_program('cat', 'more'), abi_version_file).stdout().strip() ++abi_version = run_command(find_program('cat', 'more'), abi_version_file, ++ check: true).stdout().strip() + + # Libraries have the abi_version as the filename extension + # and have the soname be all but the final part of the abi_version. +@@ -334,7 +335,7 @@ if max_lcores == 'detect' + error('Discovery of max_lcores is not supported for cross-compilation.') + endif + # overwrite the default value with discovered values +- max_lcores = run_command(get_cpu_count_cmd).stdout().to_int() ++ max_lcores = run_command(get_cpu_count_cmd, check: true).stdout().to_int() + min_lcores = 2 + # DPDK must be built for at least 2 cores + if max_lcores < min_lcores +diff --git a/dpdk/config/ppc/ppc64le-power8-linux-gcc b/dpdk/config/ppc/ppc64le-power8-linux-gcc +index 51f7ceebf3..784c33df9e 100644 +--- a/dpdk/config/ppc/ppc64le-power8-linux-gcc ++++ b/dpdk/config/ppc/ppc64le-power8-linux-gcc +@@ -1,6 +1,6 @@ + [binaries] + c = 'powerpc64le-linux-gcc' +-cpp = 'powerpc64le-linux-cpp' ++cpp = 'powerpc64le-linux-g++' + ar = 'powerpc64le-linux-gcc-ar' + strip = 'powerpc64le-linux-strip' + +diff --git a/dpdk/config/x86/meson.build b/dpdk/config/x86/meson.build +index e25ed316f4..54345c4da3 100644 +--- a/dpdk/config/x86/meson.build ++++ b/dpdk/config/x86/meson.build +@@ -4,7 +4,7 @@ + # get binutils version for the workaround of Bug 97 + binutils_ok = true + if is_linux or cc.get_id() == 'gcc' +- binutils_ok = run_command(binutils_avx512_check).returncode() == 0 ++ binutils_ok = run_command(binutils_avx512_check, check: false).returncode() == 0 + if not binutils_ok and cc.has_argument('-mno-avx512f') + machine_args += '-mno-avx512f' + warning('Binutils error with AVX512 assembly, disabling AVX512 support') +diff --git a/dpdk/devtools/check-abi.sh b/dpdk/devtools/check-abi.sh +index ca523eb94c..9835e346da 100755 +--- a/dpdk/devtools/check-abi.sh ++++ b/dpdk/devtools/check-abi.sh +@@ -44,10 +44,6 @@ for dump in $(find $refdir -name "*.dump"); do + echo "Skipped glue library $name." + continue + fi +- if grep -qE "\&2 +diff --git a/dpdk/devtools/check-forbidden-tokens.awk b/dpdk/devtools/check-forbidden-tokens.awk +index 61ba707c9b..026844141c 100755 +--- a/dpdk/devtools/check-forbidden-tokens.awk ++++ b/dpdk/devtools/check-forbidden-tokens.awk +@@ -20,6 +20,9 @@ BEGIN { + # state machine assumes the comments structure is enforced by + # checkpatches.pl + (in_file) { ++ if ($0 ~ "^@@") { ++ in_comment = 0 ++ } + # comment start + if (index($0,comment_start) > 0) { + in_comment = 1 +diff --git a/dpdk/devtools/check-symbol-change.sh b/dpdk/devtools/check-symbol-change.sh +index 8fcd0ce1a1..8992214ac8 100755 +--- a/dpdk/devtools/check-symbol-change.sh ++++ b/dpdk/devtools/check-symbol-change.sh +@@ -25,7 +25,7 @@ build_map_changes() + + # Triggering this rule, which starts a line and ends it + # with a { identifies a versioned section. The section name is +- # the rest of the line with the + and { symbols remvoed. ++ # the rest of the line with the + and { symbols removed. + # Triggering this rule sets in_sec to 1, which actives the + # symbol rule below + /^.*{/ { +@@ -35,7 +35,7 @@ build_map_changes() + } + } + +- # This rule idenfies the end of a section, and disables the ++ # This rule identifies the end of a section, and disables the + # symbol rule + /.*}/ {in_sec=0} + +@@ -100,7 +100,7 @@ check_for_rule_violations() + # Just inform the user of this occurrence, but + # don't flag it as an error + echo -n "INFO: symbol $symname is added but " +- echo -n "patch has insuficient context " ++ echo -n "patch has insufficient context " + echo -n "to determine the section name " + echo -n "please ensure the version is " + echo "EXPERIMENTAL" +diff --git a/dpdk/devtools/check-symbol-maps.sh b/dpdk/devtools/check-symbol-maps.sh +index 5bd290ac97..32e1fa5c8f 100755 +--- a/dpdk/devtools/check-symbol-maps.sh ++++ b/dpdk/devtools/check-symbol-maps.sh +@@ -53,4 +53,11 @@ if [ -n "$duplicate_symbols" ] ; then + ret=1 + fi + ++local_miss_maps=$(grep -L 'local: \*;' $@ || true) ++if [ -n "$local_miss_maps" ] ; then ++ echo "Found maps without local catch-all:" ++ echo "$local_miss_maps" ++ ret=1 ++fi ++ + exit $ret +diff --git a/dpdk/devtools/libabigail.abignore b/dpdk/devtools/libabigail.abignore +index 4b676f317d..146a601ed3 100644 +--- a/dpdk/devtools/libabigail.abignore ++++ b/dpdk/devtools/libabigail.abignore +@@ -11,3 +11,23 @@ + ; Ignore generated PMD information strings + [suppress_variable] + name_regexp = _pmd_info$ ++ ++; Ignore changes to rte_crypto_asym_op, asymmetric crypto API is experimental ++[suppress_type] ++ name = rte_crypto_asym_op ++ ++; Ignore section attribute fixes in experimental regexdev library ++[suppress_file] ++ soname_regexp = ^librte_regexdev\. ++ ++; Ignore changes in common mlx5 driver, should be all internal ++[suppress_file] ++ soname_regexp = ^librte_common_mlx5\. ++ ++; Ignore visibility fix of local functions in experimental auxiliary driver ++[suppress_file] ++ soname_regexp = ^librte_bus_auxiliary\. ++ ++; Ignore visibility fix of local functions in experimental gpudev library ++[suppress_file] ++ soname_regexp = ^librte_gpudev\. +diff --git a/dpdk/devtools/test-null.sh b/dpdk/devtools/test-null.sh +index 4ba57a6829..6cd34f64f1 100755 +--- a/dpdk/devtools/test-null.sh ++++ b/dpdk/devtools/test-null.sh +@@ -27,6 +27,7 @@ else + fi + + (sleep 1 && echo stop) | +-$testpmd -c $coremask --no-huge -m 20 \ ++# testpmd only needs 20M, make it x2 (default number of cores) for NUMA systems ++$testpmd -c $coremask --no-huge -m 40 \ + $libs -a 0:0.0 --vdev net_null1 --vdev net_null2 $eal_options -- \ + --no-mlockall --total-num-mbufs=2048 $testpmd_options -ia +diff --git a/dpdk/doc/api/doxy-api-index.md b/dpdk/doc/api/doxy-api-index.md +index 4245b9635c..baecb2e52e 100644 +--- a/dpdk/doc/api/doxy-api-index.md ++++ b/dpdk/doc/api/doxy-api-index.md +@@ -9,222 +9,222 @@ API + The public API headers are grouped by topics: + + - **device**: +- [dev] (@ref rte_dev.h), +- [ethdev] (@ref rte_ethdev.h), +- [ethctrl] (@ref rte_eth_ctrl.h), +- [rte_flow] (@ref rte_flow.h), +- [rte_tm] (@ref rte_tm.h), +- [rte_mtr] (@ref rte_mtr.h), +- [bbdev] (@ref rte_bbdev.h), +- [cryptodev] (@ref rte_cryptodev.h), +- [security] (@ref rte_security.h), +- [compressdev] (@ref rte_compressdev.h), +- [compress] (@ref rte_comp.h), +- [regexdev] (@ref rte_regexdev.h), +- [dmadev] (@ref rte_dmadev.h), +- [eventdev] (@ref rte_eventdev.h), +- [event_eth_rx_adapter] (@ref rte_event_eth_rx_adapter.h), +- [event_eth_tx_adapter] (@ref rte_event_eth_tx_adapter.h), +- [event_timer_adapter] (@ref rte_event_timer_adapter.h), +- [event_crypto_adapter] (@ref rte_event_crypto_adapter.h), +- [rawdev] (@ref rte_rawdev.h), +- [metrics] (@ref rte_metrics.h), +- [bitrate] (@ref rte_bitrate.h), +- [latency] (@ref rte_latencystats.h), +- [devargs] (@ref rte_devargs.h), +- [PCI] (@ref rte_pci.h), +- [vdev] (@ref rte_bus_vdev.h), +- [vfio] (@ref rte_vfio.h) ++ [dev](@ref rte_dev.h), ++ [ethdev](@ref rte_ethdev.h), ++ [ethctrl](@ref rte_eth_ctrl.h), ++ [rte_flow](@ref rte_flow.h), ++ [rte_tm](@ref rte_tm.h), ++ [rte_mtr](@ref rte_mtr.h), ++ [bbdev](@ref rte_bbdev.h), ++ [cryptodev](@ref rte_cryptodev.h), ++ [security](@ref rte_security.h), ++ [compressdev](@ref rte_compressdev.h), ++ [compress](@ref rte_comp.h), ++ [regexdev](@ref rte_regexdev.h), ++ [dmadev](@ref rte_dmadev.h), ++ [eventdev](@ref rte_eventdev.h), ++ [event_eth_rx_adapter](@ref rte_event_eth_rx_adapter.h), ++ [event_eth_tx_adapter](@ref rte_event_eth_tx_adapter.h), ++ [event_timer_adapter](@ref rte_event_timer_adapter.h), ++ [event_crypto_adapter](@ref rte_event_crypto_adapter.h), ++ [rawdev](@ref rte_rawdev.h), ++ [metrics](@ref rte_metrics.h), ++ [bitrate](@ref rte_bitrate.h), ++ [latency](@ref rte_latencystats.h), ++ [devargs](@ref rte_devargs.h), ++ [PCI](@ref rte_pci.h), ++ [vdev](@ref rte_bus_vdev.h), ++ [vfio](@ref rte_vfio.h) + + - **device specific**: +- [softnic] (@ref rte_eth_softnic.h), +- [bond] (@ref rte_eth_bond.h), +- [vhost] (@ref rte_vhost.h), +- [vdpa] (@ref rte_vdpa.h), +- [KNI] (@ref rte_kni.h), +- [ixgbe] (@ref rte_pmd_ixgbe.h), +- [i40e] (@ref rte_pmd_i40e.h), +- [ice] (@ref rte_pmd_ice.h), +- [iavf] (@ref rte_pmd_iavf.h), +- [ioat] (@ref rte_ioat_rawdev.h), +- [bnxt] (@ref rte_pmd_bnxt.h), +- [dpaa] (@ref rte_pmd_dpaa.h), +- [dpaa2] (@ref rte_pmd_dpaa2.h), +- [mlx5] (@ref rte_pmd_mlx5.h), +- [dpaa2_mempool] (@ref rte_dpaa2_mempool.h), +- [dpaa2_cmdif] (@ref rte_pmd_dpaa2_cmdif.h), +- [dpaa2_qdma] (@ref rte_pmd_dpaa2_qdma.h), +- [crypto_scheduler] (@ref rte_cryptodev_scheduler.h), +- [dlb2] (@ref rte_pmd_dlb2.h), +- [ifpga] (@ref rte_pmd_ifpga.h) ++ [softnic](@ref rte_eth_softnic.h), ++ [bond](@ref rte_eth_bond.h), ++ [vhost](@ref rte_vhost.h), ++ [vdpa](@ref rte_vdpa.h), ++ [KNI](@ref rte_kni.h), ++ [ixgbe](@ref rte_pmd_ixgbe.h), ++ [i40e](@ref rte_pmd_i40e.h), ++ [ice](@ref rte_pmd_ice.h), ++ [iavf](@ref rte_pmd_iavf.h), ++ [ioat](@ref rte_ioat_rawdev.h), ++ [bnxt](@ref rte_pmd_bnxt.h), ++ [dpaa](@ref rte_pmd_dpaa.h), ++ [dpaa2](@ref rte_pmd_dpaa2.h), ++ [mlx5](@ref rte_pmd_mlx5.h), ++ [dpaa2_mempool](@ref rte_dpaa2_mempool.h), ++ [dpaa2_cmdif](@ref rte_pmd_dpaa2_cmdif.h), ++ [dpaa2_qdma](@ref rte_pmd_dpaa2_qdma.h), ++ [crypto_scheduler](@ref rte_cryptodev_scheduler.h), ++ [dlb2](@ref rte_pmd_dlb2.h), ++ [ifpga](@ref rte_pmd_ifpga.h) + + - **memory**: +- [memseg] (@ref rte_memory.h), +- [memzone] (@ref rte_memzone.h), +- [mempool] (@ref rte_mempool.h), +- [malloc] (@ref rte_malloc.h), +- [memcpy] (@ref rte_memcpy.h) ++ [memseg](@ref rte_memory.h), ++ [memzone](@ref rte_memzone.h), ++ [mempool](@ref rte_mempool.h), ++ [malloc](@ref rte_malloc.h), ++ [memcpy](@ref rte_memcpy.h) + + - **timers**: +- [cycles] (@ref rte_cycles.h), +- [timer] (@ref rte_timer.h), +- [alarm] (@ref rte_alarm.h) ++ [cycles](@ref rte_cycles.h), ++ [timer](@ref rte_timer.h), ++ [alarm](@ref rte_alarm.h) + + - **locks**: +- [atomic] (@ref rte_atomic.h), +- [mcslock] (@ref rte_mcslock.h), +- [pflock] (@ref rte_pflock.h), +- [rwlock] (@ref rte_rwlock.h), +- [spinlock] (@ref rte_spinlock.h), +- [ticketlock] (@ref rte_ticketlock.h), +- [RCU] (@ref rte_rcu_qsbr.h) ++ [atomic](@ref rte_atomic.h), ++ [mcslock](@ref rte_mcslock.h), ++ [pflock](@ref rte_pflock.h), ++ [rwlock](@ref rte_rwlock.h), ++ [spinlock](@ref rte_spinlock.h), ++ [ticketlock](@ref rte_ticketlock.h), ++ [RCU](@ref rte_rcu_qsbr.h) + + - **CPU arch**: +- [branch prediction] (@ref rte_branch_prediction.h), +- [cache prefetch] (@ref rte_prefetch.h), +- [SIMD] (@ref rte_vect.h), +- [byte order] (@ref rte_byteorder.h), +- [CPU flags] (@ref rte_cpuflags.h), +- [CPU pause] (@ref rte_pause.h), +- [I/O access] (@ref rte_io.h), +- [power management] (@ref rte_power_intrinsics.h) ++ [branch prediction](@ref rte_branch_prediction.h), ++ [cache prefetch](@ref rte_prefetch.h), ++ [SIMD](@ref rte_vect.h), ++ [byte order](@ref rte_byteorder.h), ++ [CPU flags](@ref rte_cpuflags.h), ++ [CPU pause](@ref rte_pause.h), ++ [I/O access](@ref rte_io.h), ++ [power management](@ref rte_power_intrinsics.h) + + - **CPU multicore**: +- [interrupts] (@ref rte_interrupts.h), +- [launch] (@ref rte_launch.h), +- [lcore] (@ref rte_lcore.h), +- [per-lcore] (@ref rte_per_lcore.h), +- [service cores] (@ref rte_service.h), +- [keepalive] (@ref rte_keepalive.h), +- [power/freq] (@ref rte_power.h), +- [PMD power] (@ref rte_power_pmd_mgmt.h) ++ [interrupts](@ref rte_interrupts.h), ++ [launch](@ref rte_launch.h), ++ [lcore](@ref rte_lcore.h), ++ [per-lcore](@ref rte_per_lcore.h), ++ [service cores](@ref rte_service.h), ++ [keepalive](@ref rte_keepalive.h), ++ [power/freq](@ref rte_power.h), ++ [PMD power](@ref rte_power_pmd_mgmt.h) + + - **layers**: +- [ethernet] (@ref rte_ether.h), +- [ARP] (@ref rte_arp.h), +- [HIGIG] (@ref rte_higig.h), +- [ICMP] (@ref rte_icmp.h), +- [ESP] (@ref rte_esp.h), +- [IPsec] (@ref rte_ipsec.h), +- [IPsec group] (@ref rte_ipsec_group.h), +- [IPsec SA] (@ref rte_ipsec_sa.h), +- [IPsec SAD] (@ref rte_ipsec_sad.h), +- [IP] (@ref rte_ip.h), +- [frag/reass] (@ref rte_ip_frag.h), +- [SCTP] (@ref rte_sctp.h), +- [TCP] (@ref rte_tcp.h), +- [UDP] (@ref rte_udp.h), +- [GTP] (@ref rte_gtp.h), +- [GRO] (@ref rte_gro.h), +- [GSO] (@ref rte_gso.h), +- [GRE] (@ref rte_gre.h), +- [MPLS] (@ref rte_mpls.h), +- [VXLAN] (@ref rte_vxlan.h), +- [Geneve] (@ref rte_geneve.h), +- [eCPRI] (@ref rte_ecpri.h), +- [L2TPv2] (@ref rte_l2tpv2.h), +- [PPP] (@ref rte_ppp.h) ++ [ethernet](@ref rte_ether.h), ++ [ARP](@ref rte_arp.h), ++ [HIGIG](@ref rte_higig.h), ++ [ICMP](@ref rte_icmp.h), ++ [ESP](@ref rte_esp.h), ++ [IPsec](@ref rte_ipsec.h), ++ [IPsec group](@ref rte_ipsec_group.h), ++ [IPsec SA](@ref rte_ipsec_sa.h), ++ [IPsec SAD](@ref rte_ipsec_sad.h), ++ [IP](@ref rte_ip.h), ++ [frag/reass](@ref rte_ip_frag.h), ++ [SCTP](@ref rte_sctp.h), ++ [TCP](@ref rte_tcp.h), ++ [UDP](@ref rte_udp.h), ++ [GTP](@ref rte_gtp.h), ++ [GRO](@ref rte_gro.h), ++ [GSO](@ref rte_gso.h), ++ [GRE](@ref rte_gre.h), ++ [MPLS](@ref rte_mpls.h), ++ [VXLAN](@ref rte_vxlan.h), ++ [Geneve](@ref rte_geneve.h), ++ [eCPRI](@ref rte_ecpri.h), ++ [L2TPv2](@ref rte_l2tpv2.h), ++ [PPP](@ref rte_ppp.h) + + - **QoS**: +- [metering] (@ref rte_meter.h), +- [scheduler] (@ref rte_sched.h), +- [RED congestion] (@ref rte_red.h) ++ [metering](@ref rte_meter.h), ++ [scheduler](@ref rte_sched.h), ++ [RED congestion](@ref rte_red.h) + + - **routing**: +- [LPM IPv4 route] (@ref rte_lpm.h), +- [LPM IPv6 route] (@ref rte_lpm6.h), +- [RIB IPv4] (@ref rte_rib.h), +- [RIB IPv6] (@ref rte_rib6.h), +- [FIB IPv4] (@ref rte_fib.h), +- [FIB IPv6] (@ref rte_fib6.h) ++ [LPM IPv4 route](@ref rte_lpm.h), ++ [LPM IPv6 route](@ref rte_lpm6.h), ++ [RIB IPv4](@ref rte_rib.h), ++ [RIB IPv6](@ref rte_rib6.h), ++ [FIB IPv4](@ref rte_fib.h), ++ [FIB IPv6](@ref rte_fib6.h) + + - **hashes**: +- [hash] (@ref rte_hash.h), +- [jhash] (@ref rte_jhash.h), +- [thash] (@ref rte_thash.h), +- [thash_gfni] (@ref rte_thash_gfni.h), +- [FBK hash] (@ref rte_fbk_hash.h), +- [CRC hash] (@ref rte_hash_crc.h) ++ [hash](@ref rte_hash.h), ++ [jhash](@ref rte_jhash.h), ++ [thash](@ref rte_thash.h), ++ [thash_gfni](@ref rte_thash_gfni.h), ++ [FBK hash](@ref rte_fbk_hash.h), ++ [CRC hash](@ref rte_hash_crc.h) + + - **classification** +- [reorder] (@ref rte_reorder.h), +- [distributor] (@ref rte_distributor.h), +- [EFD] (@ref rte_efd.h), +- [ACL] (@ref rte_acl.h), +- [member] (@ref rte_member.h), +- [flow classify] (@ref rte_flow_classify.h), +- [BPF] (@ref rte_bpf.h) ++ [reorder](@ref rte_reorder.h), ++ [distributor](@ref rte_distributor.h), ++ [EFD](@ref rte_efd.h), ++ [ACL](@ref rte_acl.h), ++ [member](@ref rte_member.h), ++ [flow classify](@ref rte_flow_classify.h), ++ [BPF](@ref rte_bpf.h) + + - **containers**: +- [mbuf] (@ref rte_mbuf.h), +- [mbuf pool ops] (@ref rte_mbuf_pool_ops.h), +- [ring] (@ref rte_ring.h), +- [stack] (@ref rte_stack.h), +- [tailq] (@ref rte_tailq.h), +- [bitmap] (@ref rte_bitmap.h) ++ [mbuf](@ref rte_mbuf.h), ++ [mbuf pool ops](@ref rte_mbuf_pool_ops.h), ++ [ring](@ref rte_ring.h), ++ [stack](@ref rte_stack.h), ++ [tailq](@ref rte_tailq.h), ++ [bitmap](@ref rte_bitmap.h) + + - **packet framework**: +- * [port] (@ref rte_port.h): +- [ethdev] (@ref rte_port_ethdev.h), +- [ring] (@ref rte_port_ring.h), +- [frag] (@ref rte_port_frag.h), +- [reass] (@ref rte_port_ras.h), +- [sched] (@ref rte_port_sched.h), +- [kni] (@ref rte_port_kni.h), +- [src/sink] (@ref rte_port_source_sink.h) +- * [table] (@ref rte_table.h): +- [lpm IPv4] (@ref rte_table_lpm.h), +- [lpm IPv6] (@ref rte_table_lpm_ipv6.h), +- [ACL] (@ref rte_table_acl.h), +- [hash] (@ref rte_table_hash.h), +- [array] (@ref rte_table_array.h), +- [stub] (@ref rte_table_stub.h) +- * [pipeline] (@ref rte_pipeline.h) +- [port_in_action] (@ref rte_port_in_action.h) +- [table_action] (@ref rte_table_action.h) ++ * [port](@ref rte_port.h): ++ [ethdev](@ref rte_port_ethdev.h), ++ [ring](@ref rte_port_ring.h), ++ [frag](@ref rte_port_frag.h), ++ [reass](@ref rte_port_ras.h), ++ [sched](@ref rte_port_sched.h), ++ [kni](@ref rte_port_kni.h), ++ [src/sink](@ref rte_port_source_sink.h) ++ * [table](@ref rte_table.h): ++ [lpm IPv4](@ref rte_table_lpm.h), ++ [lpm IPv6](@ref rte_table_lpm_ipv6.h), ++ [ACL](@ref rte_table_acl.h), ++ [hash](@ref rte_table_hash.h), ++ [array](@ref rte_table_array.h), ++ [stub](@ref rte_table_stub.h) ++ * [pipeline](@ref rte_pipeline.h) ++ [port_in_action](@ref rte_port_in_action.h) ++ [table_action](@ref rte_table_action.h) + * SWX pipeline: +- [control] (@ref rte_swx_ctl.h), +- [extern] (@ref rte_swx_extern.h), +- [pipeline] (@ref rte_swx_pipeline.h) ++ [control](@ref rte_swx_ctl.h), ++ [extern](@ref rte_swx_extern.h), ++ [pipeline](@ref rte_swx_pipeline.h) + * SWX port: +- [port] (@ref rte_swx_port.h), +- [ethdev] (@ref rte_swx_port_ethdev.h), +- [fd] (@ref rte_swx_port_fd.h), +- [ring] (@ref rte_swx_port_ring.h), +- [src/sink] (@ref rte_swx_port_source_sink.h) ++ [port](@ref rte_swx_port.h), ++ [ethdev](@ref rte_swx_port_ethdev.h), ++ [fd](@ref rte_swx_port_fd.h), ++ [ring](@ref rte_swx_port_ring.h), ++ [src/sink](@ref rte_swx_port_source_sink.h) + * SWX table: +- [table] (@ref rte_swx_table.h), +- [table_em] (@ref rte_swx_table_em.h) +- [table_wm] (@ref rte_swx_table_wm.h) +- * [graph] (@ref rte_graph.h): +- [graph_worker] (@ref rte_graph_worker.h) ++ [table](@ref rte_swx_table.h), ++ [table_em](@ref rte_swx_table_em.h) ++ [table_wm](@ref rte_swx_table_wm.h) ++ * [graph](@ref rte_graph.h): ++ [graph_worker](@ref rte_graph_worker.h) + * graph_nodes: +- [eth_node] (@ref rte_node_eth_api.h), +- [ip4_node] (@ref rte_node_ip4_api.h) ++ [eth_node](@ref rte_node_eth_api.h), ++ [ip4_node](@ref rte_node_ip4_api.h) + + - **basic**: +- [bitops] (@ref rte_bitops.h), +- [approx fraction] (@ref rte_approx.h), +- [random] (@ref rte_random.h), +- [config file] (@ref rte_cfgfile.h), +- [key/value args] (@ref rte_kvargs.h), +- [string] (@ref rte_string_fns.h) ++ [bitops](@ref rte_bitops.h), ++ [approx fraction](@ref rte_approx.h), ++ [random](@ref rte_random.h), ++ [config file](@ref rte_cfgfile.h), ++ [key/value args](@ref rte_kvargs.h), ++ [string](@ref rte_string_fns.h) + + - **debug**: +- [jobstats] (@ref rte_jobstats.h), +- [telemetry] (@ref rte_telemetry.h), +- [pcapng] (@ref rte_pcapng.h), +- [pdump] (@ref rte_pdump.h), +- [hexdump] (@ref rte_hexdump.h), +- [debug] (@ref rte_debug.h), +- [log] (@ref rte_log.h), +- [errno] (@ref rte_errno.h), +- [trace] (@ref rte_trace.h), +- [trace_point] (@ref rte_trace_point.h) ++ [jobstats](@ref rte_jobstats.h), ++ [telemetry](@ref rte_telemetry.h), ++ [pcapng](@ref rte_pcapng.h), ++ [pdump](@ref rte_pdump.h), ++ [hexdump](@ref rte_hexdump.h), ++ [debug](@ref rte_debug.h), ++ [log](@ref rte_log.h), ++ [errno](@ref rte_errno.h), ++ [trace](@ref rte_trace.h), ++ [trace_point](@ref rte_trace_point.h) + + - **misc**: +- [EAL config] (@ref rte_eal.h), +- [common] (@ref rte_common.h), +- [experimental APIs] (@ref rte_compat.h), +- [ABI versioning] (@ref rte_function_versioning.h), +- [version] (@ref rte_version.h) ++ [EAL config](@ref rte_eal.h), ++ [common](@ref rte_common.h), ++ [experimental APIs](@ref rte_compat.h), ++ [ABI versioning](@ref rte_function_versioning.h), ++ [version](@ref rte_version.h) +diff --git a/dpdk/doc/api/generate_examples.sh b/dpdk/doc/api/generate_examples.sh +index 3e08236596..48574563ca 100755 +--- a/dpdk/doc/api/generate_examples.sh ++++ b/dpdk/doc/api/generate_examples.sh +@@ -6,21 +6,15 @@ EXAMPLES_DIR=$1 + API_EXAMPLES=$2 + + FIND=find +-if [ "$(uname)" = "FreeBSD" ] ; then +-# on FreeBSD, we need GNU find for "-printf" flag +- FIND=gfind +- if ! which -s $FIND ; then +- echo "Error: need '$FIND' on FreeBSD. Install 'findutils' pkg" +- exit 1 +- fi +-fi + + # generate a .d file including both C files and also build files, so we can + # detect both file changes and file additions/deletions +-echo "$API_EXAMPLES: $($FIND ${EXAMPLES_DIR} -type f \( -name '*.c' -o -name 'meson.build' \) -printf '%p ' )" > ${API_EXAMPLES}.d ++echo "$API_EXAMPLES: $($FIND ${EXAMPLES_DIR} -type f \( -name '*.c' -o -name 'meson.build' \) | tr '\n' ' ' )" > ${API_EXAMPLES}.d + + exec > "${API_EXAMPLES}" + printf '/**\n' + printf '@page examples DPDK Example Programs\n\n' +-$FIND "${EXAMPLES_DIR}" -type f -name '*.c' -printf '@example examples/%P\n' | LC_ALL=C sort ++$FIND "${EXAMPLES_DIR}" -type f -name '*.c' | ++ sed "s|${EXAMPLES_DIR}|@example examples|" | ++ LC_ALL=C sort + printf '*/\n' +diff --git a/dpdk/doc/api/meson.build b/dpdk/doc/api/meson.build +index 7e2b429ac8..5c25b92092 100644 +--- a/dpdk/doc/api/meson.build ++++ b/dpdk/doc/api/meson.build +@@ -24,7 +24,7 @@ htmldir = join_paths(get_option('datadir'), 'doc', 'dpdk') + # So use a configure option for now. + example = custom_target('examples.dox', + output: 'examples.dox', +- command: [generate_examples, join_paths(meson.source_root(), 'examples'), '@OUTPUT@'], ++ command: [generate_examples, join_paths(dpdk_source_root, 'examples'), '@OUTPUT@'], + depfile: 'examples.dox.d', + install: get_option('enable_docs'), + install_dir: htmldir, +@@ -32,11 +32,11 @@ example = custom_target('examples.dox', + + cdata = configuration_data() + cdata.set('VERSION', meson.project_version()) +-cdata.set('API_EXAMPLES', join_paths(meson.build_root(), 'doc', 'api', 'examples.dox')) +-cdata.set('OUTPUT', join_paths(meson.build_root(), 'doc', 'api')) ++cdata.set('API_EXAMPLES', join_paths(dpdk_build_root, 'doc', 'api', 'examples.dox')) ++cdata.set('OUTPUT', join_paths(dpdk_build_root, 'doc', 'api')) + cdata.set('HTML_OUTPUT', 'html') +-cdata.set('TOPDIR', meson.source_root()) +-cdata.set('STRIP_FROM_PATH', ' '.join([meson.source_root(), join_paths(meson.build_root(), 'doc', 'api')])) ++cdata.set('TOPDIR', dpdk_source_root) ++cdata.set('STRIP_FROM_PATH', ' '.join([dpdk_source_root, join_paths(dpdk_build_root, 'doc', 'api')])) + cdata.set('WARN_AS_ERROR', 'NO') + if get_option('werror') + cdata.set('WARN_AS_ERROR', 'YES') +diff --git a/dpdk/doc/guides/compressdevs/mlx5.rst b/dpdk/doc/guides/compressdevs/mlx5.rst +index a4e17f65b3..7f2d6bdfff 100644 +--- a/dpdk/doc/guides/compressdevs/mlx5.rst ++++ b/dpdk/doc/guides/compressdevs/mlx5.rst +@@ -7,7 +7,7 @@ MLX5 compress driver + ==================== + + The MLX5 compress driver library +-(**librte_compress_mlx5**) provides support for **Mellanox BlueField 2** ++(**librte_compress_mlx5**) provides support for **Mellanox BlueField-2** + families of 25/50/100/200 Gb/s adapters. + + Design +@@ -21,7 +21,7 @@ These engines are part of the ARM complex of the BlueField chip, and as + such they do not use NIC related resources (e.g. RX/TX bandwidth). + They do share the same PCI and memory bandwidth. + +-So, using the BlueField device (starting from BlueField 2), the compress ++So, using the BlueField device (starting from BlueField-2), the compress + class operations can be supported in parallel to the net, vDPA and + RegEx class operations. + +@@ -95,7 +95,7 @@ Driver options + Supported NICs + -------------- + +-* Mellanox\ |reg| BlueField 2 SmartNIC ++* Mellanox\ |reg| BlueField-2 SmartNIC + + Prerequisites + ------------- +diff --git a/dpdk/doc/guides/conf.py b/dpdk/doc/guides/conf.py +index 1743ce301f..a55ce38800 100644 +--- a/dpdk/doc/guides/conf.py ++++ b/dpdk/doc/guides/conf.py +@@ -3,7 +3,7 @@ + # Copyright(c) 2010-2015 Intel Corporation + + from docutils import nodes +-from distutils.version import LooseVersion ++from packaging.version import Version + from sphinx import __version__ as sphinx_version + from os import listdir + from os import environ +@@ -28,7 +28,7 @@ + + project = 'Data Plane Development Kit' + html_logo = '../logo/DPDK_logo_vertical_rev_small.png' +-if LooseVersion(sphinx_version) >= LooseVersion('3.5'): ++if Version(sphinx_version) >= Version('3.5'): + html_permalinks = False + else: + html_add_permalinks = "" +@@ -427,7 +427,7 @@ def setup(app): + 'Features availability for Timer adapters', + 'Feature') + +- if LooseVersion(sphinx_version) < LooseVersion('1.3.1'): ++ if Version(sphinx_version) < Version('1.3.1'): + print('Upgrade sphinx to version >= 1.3.1 for ' + 'improved Figure/Table number handling.', + file=stderr) +diff --git a/dpdk/doc/guides/cryptodevs/mlx5.rst b/dpdk/doc/guides/cryptodevs/mlx5.rst +index e86a6205e8..9936556cc9 100644 +--- a/dpdk/doc/guides/cryptodevs/mlx5.rst ++++ b/dpdk/doc/guides/cryptodevs/mlx5.rst +@@ -88,7 +88,7 @@ The mlxreg dedicated tool should be used as follows: + should not be specified. + + All the device ports should set it in order to move to operational mode. +- For BlueField 2, the internal ports in the ARM system should also be set. ++ For BlueField-2, the internal ports in the ARM system should also be set. + + - Query CRYPTO_OPERATIONAL register to make sure the device is in Operational + mode. +@@ -142,7 +142,7 @@ Supported NICs + -------------- + + * Mellanox\ |reg| ConnectX\ |reg|-6 200G MCX654106A-HCAT (2x200G) +-* Mellanox\ |reg| BlueField 2 SmartNIC ++* Mellanox\ |reg| BlueField-2 SmartNIC + * Mellanox\ |reg| ConnectX\ |reg|-6 Dx + + +@@ -162,7 +162,7 @@ FW Prerequisites + ~~~~~~~~~~~~~~~~ + + - xx.31.0328 for ConnectX-6. +-- xx.32.0108 for ConnectX-6 Dx and BlueField 2. ++- xx.32.0108 for ConnectX-6 Dx and BlueField-2. + + Linux Prerequisites + ~~~~~~~~~~~~~~~~~~~ +diff --git a/dpdk/doc/guides/dmadevs/hisilicon.rst b/dpdk/doc/guides/dmadevs/hisilicon.rst +index 191e56f2f7..974bc49376 100644 +--- a/dpdk/doc/guides/dmadevs/hisilicon.rst ++++ b/dpdk/doc/guides/dmadevs/hisilicon.rst +@@ -29,8 +29,8 @@ which can be accessed using API from the ``rte_dmadev`` library. + + The name of the ``dmadev`` created is like "B:D.F-chX", e.g. DMA 0000:7b:00.0 + will create four ``dmadev``, +-the 1st ``dmadev`` name is "7b:00.0-ch0", +-and the 2nd ``dmadev`` name is "7b:00.0-ch1". ++the 1st ``dmadev`` name is "0000:7b:00.0-ch0", ++and the 2nd ``dmadev`` name is "0000:7b:00.0-ch1". + + Device Configuration + ~~~~~~~~~~~~~~~~~~~~~ +diff --git a/dpdk/doc/guides/dmadevs/idxd.rst b/dpdk/doc/guides/dmadevs/idxd.rst +index d4a210b854..81451afd4d 100644 +--- a/dpdk/doc/guides/dmadevs/idxd.rst ++++ b/dpdk/doc/guides/dmadevs/idxd.rst +@@ -55,7 +55,6 @@ such as priority or queue depth, need to be set for each queue. + To assign an engine to a group:: + + $ accel-config config-engine dsa0/engine0.0 --group-id=0 +- $ accel-config config-engine dsa0/engine0.1 --group-id=1 + + To assign work queues to groups for passing descriptors to the engines a similar accel-config command can be used. + However, the work queues also need to be configured depending on the use case. +@@ -71,7 +70,7 @@ Example configuration for a work queue:: + + $ accel-config config-wq dsa0/wq0.0 --group-id=0 \ + --mode=dedicated --priority=10 --wq-size=8 \ +- --type=user --name=dpdk_app1 ++ --max-batch-size=512 --type=user --name=dpdk_app1 + + Once the devices have been configured, they need to be enabled:: + +@@ -82,6 +81,32 @@ Check the device configuration:: + + $ accel-config list + ++Every Intel\ |reg| DSA instance supports multiple queues and each should be similarly configured. ++As a further example, the following set of commands will configure and enable 4 queues on instance 0, ++giving each an equal share of resources:: ++ ++ # configure 4 groups, each with one engine ++ accel-config config-engine dsa0/engine0.0 --group-id=0 ++ accel-config config-engine dsa0/engine0.1 --group-id=1 ++ accel-config config-engine dsa0/engine0.2 --group-id=2 ++ accel-config config-engine dsa0/engine0.3 --group-id=3 ++ ++ # configure 4 queues, putting each in a different group, so each ++ # is backed by a single engine ++ accel-config config-wq dsa0/wq0.0 --group-id=0 --type=user --wq-size=32 \ ++ --priority=10 --max-batch-size=1024 --mode=dedicated --name=dpdk_app1 ++ accel-config config-wq dsa0/wq0.1 --group-id=1 --type=user --wq-size=32 \ ++ --priority=10 --max-batch-size=1024 --mode=dedicated --name=dpdk_app1 ++ accel-config config-wq dsa0/wq0.2 --group-id=2 --type=user --wq-size=32 \ ++ --priority=10 --max-batch-size=1024 --mode=dedicated --name=dpdk_app1 ++ accel-config config-wq dsa0/wq0.3 --group-id=3 --type=user --wq-size=32 \ ++ --priority=10 --max-batch-size=1024 --mode=dedicated --name=dpdk_app1 ++ ++ # enable device and queues ++ accel-config enable-device dsa0 ++ accel-config enable-wq dsa0/wq0.0 dsa0/wq0.1 dsa0/wq0.2 dsa0/wq0.3 ++ ++ + Devices using VFIO/UIO drivers + ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +diff --git a/dpdk/doc/guides/eventdevs/dlb2.rst b/dpdk/doc/guides/eventdevs/dlb2.rst +index bce984ca08..bc53618b53 100644 +--- a/dpdk/doc/guides/eventdevs/dlb2.rst ++++ b/dpdk/doc/guides/eventdevs/dlb2.rst +@@ -151,7 +151,7 @@ load-balanced queues, and directed credits are used for directed queues. + These pools' sizes are controlled by the nb_events_limit field in struct + rte_event_dev_config. The load-balanced pool is sized to contain + nb_events_limit credits, and the directed pool is sized to contain +-nb_events_limit/4 credits. The directed pool size can be overridden with the ++nb_events_limit/2 credits. The directed pool size can be overridden with the + num_dir_credits devargs argument, like so: + + .. code-block:: console +@@ -239,8 +239,8 @@ queue A. + Due to this, workers should stop retrying after a time, release the events it + is attempting to enqueue, and dequeue more events. It is important that the + worker release the events and don't simply set them aside to retry the enqueue +-again later, because the port has limited history list size (by default, twice +-the port's dequeue_depth). ++again later, because the port has limited history list size (by default, same ++as port's dequeue_depth). + + Priority + ~~~~~~~~ +@@ -309,18 +309,11 @@ scheduled. The likelihood of this case depends on the eventdev configuration, + traffic behavior, event processing latency, potential for a worker to be + interrupted or otherwise delayed, etc. + +-By default, the PMD allocates 16 buffer entries for each load-balanced queue, +-which provides an even division across all 128 queues but potentially wastes ++By default, the PMD allocates 64 buffer entries for each load-balanced queue, ++which provides an even division across all 32 queues but potentially wastes + buffer space (e.g. if not all queues are used, or aren't used for atomic + scheduling). + +-The PMD provides a dev arg to override the default per-queue allocation. To +-increase per-queue atomic-inflight allocation to (for example) 64: +- +- .. code-block:: console +- +- --allow ea:00.0,atm_inflights=64 +- + QID Depth Threshold + ~~~~~~~~~~~~~~~~~~~ + +@@ -337,7 +330,7 @@ Per queue threshold metrics are tracked in the DLB xstats, and are also + returned in the impl_opaque field of each received event. + + The per qid threshold can be specified as part of the device args, and +-can be applied to all queue, a range of queues, or a single queue, as ++can be applied to all queues, a range of queues, or a single queue, as + shown below. + + .. code-block:: console +diff --git a/dpdk/doc/guides/eventdevs/features/dlb2.ini b/dpdk/doc/guides/eventdevs/features/dlb2.ini +index 29747b1c26..48a2a18aff 100644 +--- a/dpdk/doc/guides/eventdevs/features/dlb2.ini ++++ b/dpdk/doc/guides/eventdevs/features/dlb2.ini +@@ -4,12 +4,13 @@ + ; Refer to default.ini for the full list of available PMD features. + ; + [Scheduling Features] +-queue_qos = Y + event_qos = Y + distributed_sched = Y + queue_all_types = Y + burst_mode = Y + implicit_release_disable = Y ++runtime_port_link = Y ++multiple_queue_port = Y + maintenance_free = Y + + [Eth Rx adapter Features] +diff --git a/dpdk/doc/guides/gpus/features/cuda.ini b/dpdk/doc/guides/gpus/features/cuda.ini +new file mode 100644 +index 0000000000..9d587eed6e +--- /dev/null ++++ b/dpdk/doc/guides/gpus/features/cuda.ini +@@ -0,0 +1,10 @@ ++; ++; Supported features of the 'cuda' gpu driver. ++; ++; Refer to default.ini for the full list of available PMD features. ++; ++[Features] ++Get device info = Y ++Share CPU memory with device = Y ++Allocate device memory = Y ++Free memory = Y +diff --git a/dpdk/doc/guides/howto/img/virtio_user_for_container_networking.svg b/dpdk/doc/guides/howto/img/virtio_user_for_container_networking.svg +index de80806649..dc9b318e7e 100644 +--- a/dpdk/doc/guides/howto/img/virtio_user_for_container_networking.svg ++++ b/dpdk/doc/guides/howto/img/virtio_user_for_container_networking.svg +@@ -465,7 +465,7 @@ + v:mID="63" + id="shape63-63">Sheet.63Contanier/AppContainer/App ++ + .. _Enabling_Additional_Functionality: + + Enabling Additional Functionality +@@ -64,13 +66,62 @@ The application can then determine what action to take, if any, if the HPET is n + Running DPDK Applications Without Root Privileges + ------------------------------------------------- + +-In order to run DPDK as non-root, the following Linux filesystem objects' +-permissions should be adjusted to ensure that the Linux account being used to +-run the DPDK application has access to them: ++The following sections describe generic requirements and configuration ++for running DPDK applications as non-root. ++There may be additional requirements documented for some drivers. ++ ++Hugepages ++~~~~~~~~~ ++ ++Hugepages must be reserved as root before running the application as non-root, ++for example:: ++ ++ sudo dpdk-hugepages.py --reserve 1G ++ ++If multi-process is not required, running with ``--in-memory`` ++bypasses the need to access hugepage mount point and files within it. ++Otherwise, hugepage directory must be made accessible ++for writing to the unprivileged user. ++A good way for managing multiple applications using hugepages ++is to mount the filesystem with group permissions ++and add a supplementary group to each application or container. ++ ++One option is to mount manually:: ++ ++ mount -t hugetlbfs -o pagesize=1G,uid=`id -u`,gid=`id -g` nodev $HOME/huge-1G ++ ++In production environment, the OS can manage mount points ++(`systemd example `_). ++ ++The ``hugetlb`` filesystem has additional options to guarantee or limit ++the amount of memory that is possible to allocate using the mount point. ++Refer to the `documentation `_. ++ ++.. note:: ++ ++ Using ``vfio-pci`` kernel driver, if applicable, can eliminate the need ++ for physical addresses and therefore eliminate the permission requirements ++ described below. ++ ++If the driver requires using physical addresses (PA), ++the executable file must be granted additional capabilities: ++ ++* ``SYS_ADMIN`` to read ``/proc/self/pagemaps`` ++* ``IPC_LOCK`` to lock hugepages in memory + +-* All directories which serve as hugepage mount points, for example, ``/dev/hugepages`` ++.. code-block:: console ++ ++ setcap cap_ipc_lock,cap_sys_admin+ep ++ ++If physical addresses are not accessible, ++the following message will appear during EAL initialization:: ++ ++ EAL: rte_mem_virt2phy(): cannot open /proc/self/pagemap: Permission denied + +-* If the HPET is to be used, ``/dev/hpet`` ++It is harmless in case PA are not needed. ++ ++Resource Limits ++~~~~~~~~~~~~~~~ + + When running as non-root user, there may be some additional resource limits + that are imposed by the system. Specifically, the following resource limits may +@@ -85,8 +136,10 @@ need to be adjusted in order to ensure normal DPDK operation: + The above limits can usually be adjusted by editing + ``/etc/security/limits.conf`` file, and rebooting. + +-Additionally, depending on which kernel driver is in use, the relevant +-resources also should be accessible by the user running the DPDK application. ++Device Control ++~~~~~~~~~~~~~~ + -+#define INIT_MULTIVAR_EXP(VAR, MEMBER, POINTER, ITER_TYPE, ...) \ -+ ITER_TYPE *ITER_VAR(VAR) = ( __VA_ARGS__ , (ITER_TYPE *) POINTER) ++If the HPET is to be used, ``/dev/hpet`` permissions must be adjusted. + + For ``vfio-pci`` kernel driver, the following Linux file system objects' + permissions should be adjusted: +@@ -96,38 +149,18 @@ permissions should be adjusted: + * The directories under ``/dev/vfio`` that correspond to IOMMU group numbers of + devices intended to be used by DPDK, for example, ``/dev/vfio/50`` + +-.. note:: +- +- The instructions below will allow running DPDK with ``igb_uio`` or +- ``uio_pci_generic`` drivers as non-root with older Linux kernel versions. +- However, since version 4.0, the kernel does not allow unprivileged processes +- to read the physical address information from the pagemaps file, making it +- impossible for those processes to be used by non-privileged users. In such +- cases, using the VFIO driver is recommended. +- +-For ``igb_uio`` or ``uio_pci_generic`` kernel drivers, the following Linux file +-system objects' permissions should be adjusted: +- +-* The userspace-io device files in ``/dev``, for example, ``/dev/uio0``, ``/dev/uio1``, and so on +- +-* The userspace-io sysfs config and resource files, for example for ``uio0``:: +- +- /sys/class/uio/uio0/device/config +- /sys/class/uio/uio0/device/resource* +- +- + Power Management and Power Saving Functionality + ----------------------------------------------- + +-Enhanced Intel SpeedStep® Technology must be enabled in the platform BIOS if the power management feature of DPDK is to be used. ++Enhanced Intel SpeedStep\ |reg| Technology must be enabled in the platform BIOS if the power management feature of DPDK is to be used. + Otherwise, the sys file folder ``/sys/devices/system/cpu/cpu0/cpufreq`` will not exist, and the CPU frequency- based power management cannot be used. + Consult the relevant BIOS documentation to determine how these settings can be accessed. + +-For example, on some Intel reference platform BIOS variants, the path to Enhanced Intel SpeedStep® Technology is:: ++For example, on some Intel reference platform BIOS variants, the path to Enhanced Intel SpeedStep\ |reg| Technology is:: + + Advanced + -> Processor Configuration +- -> Enhanced Intel SpeedStep® Tech ++ -> Enhanced Intel SpeedStep\ |reg| Tech + + In addition, C3 and C6 should be enabled as well for power management. The path of C3 and C6 on the same platform BIOS is:: + +diff --git a/dpdk/doc/guides/linux_gsg/linux_drivers.rst b/dpdk/doc/guides/linux_gsg/linux_drivers.rst +index 2dd711bb37..75af2f01e1 100644 +--- a/dpdk/doc/guides/linux_gsg/linux_drivers.rst ++++ b/dpdk/doc/guides/linux_gsg/linux_drivers.rst +@@ -3,6 +3,8 @@ + Copyright 2017 Mellanox Technologies, Ltd + All rights reserved. + ++.. include:: ++ + .. _linux_gsg_linux_drivers: + + Linux Drivers +@@ -99,7 +101,7 @@ The token will be used for all PF and VF ports within the application. + + To make use of full VFIO functionality, + both kernel and BIOS must support and be configured +-to use IO virtualization (such as Intel® VT-d). ++to use IO virtualization (such as Intel\ |reg| VT-d). + + .. note:: + +@@ -172,6 +174,11 @@ It can be loaded as shown below: + sudo modprobe uio + sudo insmod igb_uio.ko + ++.. note:: ++ ++ For some devices which lack support for legacy interrupts, e.g. virtual function ++ (VF) devices, the ``igb_uio`` module may be needed in place of ``uio_pci_generic``. ++ + .. note:: + + If UEFI secure boot is enabled, +@@ -335,7 +342,7 @@ Please refer to earlier sections on how to configure kernel parameters + correctly for your system. + + If the kernel is configured correctly, one also has to make sure that +-the BIOS configuration has virtualization features (such as Intel® VT-d). ++the BIOS configuration has virtualization features (such as Intel\ |reg| VT-d). + There is no standard way to check if the platform is configured correctly, + so please check with your platform documentation to see if it has such features, + and how to enable them. +diff --git a/dpdk/doc/guides/linux_gsg/sys_reqs.rst b/dpdk/doc/guides/linux_gsg/sys_reqs.rst +index d95a78d156..cfaa2db301 100644 +--- a/dpdk/doc/guides/linux_gsg/sys_reqs.rst ++++ b/dpdk/doc/guides/linux_gsg/sys_reqs.rst +@@ -1,6 +1,8 @@ + .. SPDX-License-Identifier: BSD-3-Clause + Copyright(c) 2010-2014 Intel Corporation. + ++.. include:: ++ + System Requirements + =================== + +@@ -72,10 +74,10 @@ Compilation of the DPDK + + **Optional Tools:** + +-* Intel® C++ Compiler (icc). For installation, additional libraries may be required. ++* Intel\ |reg| C++ Compiler (icc). For installation, additional libraries may be required. + See the icc Installation Guide found in the Documentation directory under the compiler installation. + +-* IBM® Advance ToolChain for Powerlinux. This is a set of open source development tools and runtime libraries ++* IBM\ |reg| Advance ToolChain for Powerlinux. This is a set of open source development tools and runtime libraries + which allows users to take leading edge advantage of IBM's latest POWER hardware features on Linux. To install + it, see the IBM official installation document. + +diff --git a/dpdk/doc/guides/nics/af_packet.rst b/dpdk/doc/guides/nics/af_packet.rst +index 8292369141..66b977e1a2 100644 +--- a/dpdk/doc/guides/nics/af_packet.rst ++++ b/dpdk/doc/guides/nics/af_packet.rst +@@ -9,7 +9,7 @@ packets. This Linux-specific PMD binds to an AF_PACKET socket and allows + a DPDK application to send and receive raw packets through the Kernel. + + In order to improve Rx and Tx performance this implementation makes use of +-PACKET_MMAP, which provides a mmap'ed ring buffer, shared between user space ++PACKET_MMAP, which provides a mmapped ring buffer, shared between user space + and kernel, that's used to send and receive packets. This helps reducing system + calls and the copies needed between user space and Kernel. + +diff --git a/dpdk/doc/guides/nics/af_xdp.rst b/dpdk/doc/guides/nics/af_xdp.rst +index c9d0e1ad6c..db02ea1984 100644 +--- a/dpdk/doc/guides/nics/af_xdp.rst ++++ b/dpdk/doc/guides/nics/af_xdp.rst +@@ -43,9 +43,7 @@ Prerequisites + This is a Linux-specific PMD, thus the following prerequisites apply: + + * A Linux Kernel (version > v4.18) with XDP sockets configuration enabled; +-* libbpf (within kernel version > v5.1-rc4) with latest af_xdp support installed, +- User can install libbpf via `make install_lib` && `make install_headers` in +- /tools/lib/bpf; ++* Both libxdp >=v1.2.2 and libbpf libraries installed, or, libbpf <=v0.6.0 + * A Kernel bound interface to attach to; + * For need_wakeup feature, it requires kernel version later than v5.3-rc1; + * For PMD zero copy, it requires kernel version later than v5.4-rc1; +@@ -143,4 +141,4 @@ Limitations + NAPI context from a watchdog timer instead of from softirqs. More information + on this feature can be found at [1]. + +- [1] https://lwn.net/Articles/837010/ +\ No newline at end of file ++ [1] https://lwn.net/Articles/837010/ +diff --git a/dpdk/doc/guides/nics/features/ice_dcf.ini b/dpdk/doc/guides/nics/features/ice_dcf.ini +index 4d6fb6d849..54073f0b88 100644 +--- a/dpdk/doc/guides/nics/features/ice_dcf.ini ++++ b/dpdk/doc/guides/nics/features/ice_dcf.ini +@@ -3,6 +3,9 @@ + ; + ; Refer to default.ini for the full list of available PMD features. + ; ++; A feature with "P" indicates only be supported when non-vector path ++; is selected. ++; + [Features] + Queue start/stop = Y + Scattered Rx = Y +@@ -10,6 +13,8 @@ RSS hash = P + CRC offload = Y + L3 checksum offload = P + L4 checksum offload = P ++Inner L3 checksum = P ++Inner L4 checksum = P + Basic stats = Y + Linux = Y + x86-32 = Y +diff --git a/dpdk/doc/guides/nics/hns3.rst b/dpdk/doc/guides/nics/hns3.rst +index 5f68a10ecf..791c9cc2ed 100644 +--- a/dpdk/doc/guides/nics/hns3.rst ++++ b/dpdk/doc/guides/nics/hns3.rst +@@ -290,5 +290,10 @@ Currently, we only support VF device driven by DPDK driver when PF is driven + by kernel mode hns3 ethdev driver. VF is not supported when PF is driven by + DPDK driver. + ++For sake of Rx/Tx performance, IEEE 1588 is not supported when using vec or ++sve burst function. When enabling IEEE 1588, Rx/Tx burst mode should be ++simple or common. It is recommended that enable IEEE 1588 before ethdev ++start. In this way, the correct Rx/Tx burst function can be selected. ++ + Build with ICC is not supported yet. + X86-32, Power8, ARMv7 and BSD are not supported yet. +diff --git a/dpdk/doc/guides/nics/i40e.rst b/dpdk/doc/guides/nics/i40e.rst +index ef91b3a1ac..aedb1afc4b 100644 +--- a/dpdk/doc/guides/nics/i40e.rst ++++ b/dpdk/doc/guides/nics/i40e.rst +@@ -101,6 +101,14 @@ For X710/XL710/XXV710, + +--------------+-----------------------+------------------+ + | DPDK version | Kernel driver version | Firmware version | + +==============+=======================+==================+ ++ | 22.03 | 2.17.15 | 8.30 | ++ +--------------+-----------------------+------------------+ ++ | 21.11 | 2.17.4 | 8.30 | ++ +--------------+-----------------------+------------------+ ++ | 21.08 | 2.15.9 | 8.30 | ++ +--------------+-----------------------+------------------+ ++ | 21.05 | 2.15.9 | 8.30 | ++ +--------------+-----------------------+------------------+ + | 21.02 | 2.14.13 | 8.00 | + +--------------+-----------------------+------------------+ + | 20.11 | 2.14.13 | 8.00 | +@@ -148,6 +156,14 @@ For X722, + +--------------+-----------------------+------------------+ + | DPDK version | Kernel driver version | Firmware version | + +==============+=======================+==================+ ++ | 22.03 | 2.17.15 | 5.50 | ++ +--------------+-----------------------+------------------+ ++ | 21.11 | 2.17.4 | 5.30 | ++ +--------------+-----------------------+------------------+ ++ | 21.08 | 2.15.9 | 5.30 | ++ +--------------+-----------------------+------------------+ ++ | 21.05 | 2.15.9 | 5.30 | ++ +--------------+-----------------------+------------------+ + | 21.02 | 2.14.13 | 5.00 | + +--------------+-----------------------+------------------+ + | 20.11 | 2.13.10 | 5.00 | +@@ -771,6 +787,13 @@ it will fail and return the info "Conflict with the first rule's input set", + which means the current rule's input set conflicts with the first rule's. + Remove the first rule if want to change the input set of the PCTYPE. + ++PF reset fail after QinQ set with FW >= 8.4 ++~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ++ ++If upgrade FW to version 8.4 and higher, after set MAC VLAN filter and configure outer VLAN on PF, kill ++DPDK process will cause the card crash. ++ ++ + Example of getting best performance with l3fwd example + ------------------------------------------------------ + +diff --git a/dpdk/doc/guides/nics/ice.rst b/dpdk/doc/guides/nics/ice.rst +index f95fef8cf0..6b903b9bbc 100644 +--- a/dpdk/doc/guides/nics/ice.rst ++++ b/dpdk/doc/guides/nics/ice.rst +@@ -58,6 +58,12 @@ The detailed information can refer to chapter Tested Platforms/Tested NICs in re + +-----------+---------------+-----------------+-----------+--------------+-----------+ + | 21.05 | 1.6.5 | 1.3.26 | 1.3.30 | 1.3.6 | 3.0 | + +-----------+---------------+-----------------+-----------+--------------+-----------+ ++ | 21.08 | 1.7.16 | 1.3.27 | 1.3.31 | 1.3.7 | 3.1 | ++ +-----------+---------------+-----------------+-----------+--------------+-----------+ ++ | 21.11 | 1.7.16 | 1.3.27 | 1.3.31 | 1.3.7 | 3.1 | ++ +-----------+---------------+-----------------+-----------+--------------+-----------+ ++ | 22.03 | 1.8.3 | 1.3.28 | 1.3.35 | 1.3.8 | 3.2 | ++ +-----------+---------------+-----------------+-----------+--------------+-----------+ + + Pre-Installation Configuration + ------------------------------ +diff --git a/dpdk/doc/guides/nics/ixgbe.rst b/dpdk/doc/guides/nics/ixgbe.rst +index 82fa453fa2..ad1a3da610 100644 +--- a/dpdk/doc/guides/nics/ixgbe.rst ++++ b/dpdk/doc/guides/nics/ixgbe.rst +@@ -101,6 +101,23 @@ To guarantee the constraint, capabilities in dev_conf.rxmode.offloads will be ch + + fdir_conf->mode will also be checked. + ++Disable SDP3 TX_DISABLE for Fiber Links ++^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ++ ++The following ``devargs`` option can be enabled at runtime. It must ++be passed as part of EAL arguments. For example, ++ ++.. code-block:: console ++ ++ dpdk-testpmd -a fiber_sdp3_no_tx_disable=1 -- -i ++ ++- ``fiber_sdp3_no_tx_disable`` (default **0**) ++ ++ Not all IXGBE implementations with SFP cages use the SDP3 signal as ++ TX_DISABLE as a means to disable the laser on fiber SFP modules. ++ This option informs the driver that in this case, SDP3 is not to be ++ used as a check for link up by testing for laser on/off. ++ + VF Runtime Options + ^^^^^^^^^^^^^^^^^^ + +diff --git a/dpdk/doc/guides/nics/kni.rst b/dpdk/doc/guides/nics/kni.rst +index 37c5411a32..2a23bb3f3b 100644 +--- a/dpdk/doc/guides/nics/kni.rst ++++ b/dpdk/doc/guides/nics/kni.rst +@@ -33,7 +33,7 @@ Usage + + EAL ``--vdev`` argument can be used to create KNI device instance, like:: + +- dpdk-testpmd --vdev=net_kni0 --vdev=net_kn1 -- -i ++ dpdk-testpmd --vdev=net_kni0 --vdev=net_kni1 -- -i + + Above command will create ``kni0`` and ``kni1`` Linux network interfaces, + those interfaces can be controlled by standard Linux tools. +diff --git a/dpdk/doc/guides/nics/mlx4.rst b/dpdk/doc/guides/nics/mlx4.rst +index a25add7c47..66493a1157 100644 +--- a/dpdk/doc/guides/nics/mlx4.rst ++++ b/dpdk/doc/guides/nics/mlx4.rst +@@ -14,7 +14,7 @@ the `Mellanox website `_. Help is also provided by + the `Mellanox community `_. + + There is also a `section dedicated to this poll mode driver +-`_. ++`_. + + + Implementation details +@@ -178,7 +178,7 @@ DPDK and must be installed separately: + + - mlx4_core: hardware driver managing Mellanox ConnectX-3 devices. + - mlx4_en: Ethernet device driver that provides kernel network interfaces. +- - mlx4_ib: InifiniBand device driver. ++ - mlx4_ib: InfiniBand device driver. + - ib_uverbs: user space driver for verbs (entry point for libibverbs). + + - **Firmware update** +@@ -219,7 +219,7 @@ Mellanox OFED as a fallback + - `Mellanox OFED`_ version: **4.4, 4.5, 4.6**. + - firmware version: **2.42.5000** and above. + +-.. _`Mellanox OFED`: http://www.mellanox.com/page/products_dyn?product_family=26&mtag=linux_sw_drivers ++.. _`Mellanox OFED`: https://network.nvidia.com/products/infiniband-drivers/linux/mlnx_ofed/ + + .. note:: + +diff --git a/dpdk/doc/guides/nics/mlx5.rst b/dpdk/doc/guides/nics/mlx5.rst +index feb2e57cee..ce40d1cdac 100644 +--- a/dpdk/doc/guides/nics/mlx5.rst ++++ b/dpdk/doc/guides/nics/mlx5.rst +@@ -19,7 +19,7 @@ Information and documentation about these adapters can be found on the + `Mellanox community `__. + + There is also a `section dedicated to this poll mode driver +-`__. ++`_. + + + Design +@@ -250,7 +250,7 @@ Limitations + + - Flex item: + +- - Hardware support: BlueField 2. ++ - Hardware support: BlueField-2. + - Flex item is supported on PF only. + - Hardware limits ``header_length_mask_width`` up to 6 bits. + - Firmware supports 8 global sample fields. +@@ -360,6 +360,12 @@ Limitations + - can be applied to VF ports only. + - must specify PF port action (packet redirection from VF to PF). + ++- E-Switch Manager matching: + -+/* Multi-variable condition. -+ * Evaluates the condition expression (that must be based on the internal -+ * iterator variable). Only if the result of expression is true, the OBJECT is -+ * set to the object containing the current value of the iterator variable. -+ * -+ * It is up to the caller to make sure it is safe to run OBJECT_CONTAINING on -+ * the pointers that verify the condition. -+ */ -+#define CONDITION_MULTIVAR(VAR, MEMBER, EXPR) \ -+ ((EXPR) ? \ -+ (((VAR) = OBJECT_CONTAINING(ITER_VAR(VAR), VAR, MEMBER)), 1) : \ -+ (((VAR) = NULL), 0)) ++ - For Bluefield with old FW ++ which doesn't expose the E-Switch Manager vport ID in the capability, ++ matching E-Switch Manager should be used only in Bluefield embedded CPU mode. + -+/* Multi-variable update. -+ * Sets the iterator value to NEXT_ITER. -+ */ -+#define UPDATE_MULTIVAR(VAR, NEXT_ITER) \ -+ (ITER_VAR(VAR) = NEXT_ITER) + - Raw encapsulation: + + - The input buffer, used as outer header, is not validated. +@@ -420,6 +426,8 @@ Limitations + sample actions list. + - For E-Switch mirroring flow, supports ``RAW ENCAP``, ``Port ID``, + ``VXLAN ENCAP``, ``NVGRE ENCAP`` in the sample actions list. ++ - For ConnectX-5 trusted device, the application metadata with SET_TAG index 0 ++ is not supported before ``RTE_FLOW_ACTION_TYPE_SAMPLE`` action. + + - Modify Field flow: + +@@ -428,6 +436,12 @@ Limitations + - Modification of the 802.1Q Tag, VXLAN Network or GENEVE Network ID's is not supported. + - Encapsulation levels are not supported, can modify outermost header fields only. + - Offsets must be 32-bits aligned, cannot skip past the boundary of a field. ++ - If the field type is ``RTE_FLOW_FIELD_MAC_TYPE`` ++ and packet contains one or more VLAN headers, ++ the meaningful type field following the last VLAN header ++ is used as modify field operation argument. ++ The modify field action is not intended to modify VLAN headers type field, ++ dedicated VLAN push and pop actions should be used instead. + + - IPv6 header item 'proto' field, indicating the next header protocol, should + not be set as extension header. +@@ -462,17 +476,18 @@ Limitations + + - Integrity: + +- - Integrity offload is enabled for **ConnectX-6** family. ++ - Integrity offload is enabled starting from **ConnectX-6 Dx**. + - Verification bits provided by the hardware are ``l3_ok``, ``ipv4_csum_ok``, ``l4_ok``, ``l4_csum_ok``. + - ``level`` value 0 references outer headers. ++ - Negative integrity item verification is not supported. + - Multiple integrity items not supported in a single flow rule. + - Flow rule items supplied by application must explicitly specify network headers referred by integrity item. + For example, if integrity item mask sets ``l4_ok`` or ``l4_csum_ok`` bits, reference to L4 network header, + TCP or UDP, must be in the rule pattern as well:: + + flow create 0 ingress pattern integrity level is 0 value mask l3_ok value spec l3_ok / eth / ipv6 / end … +- or +- flow create 0 ingress pattern integrity level is 0 value mask l4_ok value spec 0 / eth / ipv4 proto is udp / end … + -+/* In the safe version of the multi-variable container iteration, the next -+ * value of the iterator is precalculated on the condition expression. -+ * This allows for the iterator to be freed inside the loop. -+ * -+ * Two versions of the macros are provided: -+ * -+ * * In the _SHORT version, the user does not have to provide a variable to -+ * store the next value of the iterator. Instead, a second iterator variable -+ * is declared in the INIT_ macro and its name is determined by -+ * ITER_NEXT_VAR(OBJECT). -+ * -+ * * In the _LONG version, the user provides another variable of the same type -+ * as the iterator object variable to store the next containing object. -+ * We still declare an iterator variable inside the loop but in this case it's -+ * name is derived from the name of the next containing variable. -+ * The value of the next containing object will only be set -+ * (via OBJECT_CONTAINING) if an additional condition is statisfied. This -+ * second condition must ensure it is safe to call OBJECT_CONTAINING on the -+ * next iterator variable. -+ * With respect to the value of the next containing object: -+ * - Inside of the loop: the variable is either NULL or safe to use. -+ * - Outside of the loop: the variable is NULL if the loop ends normally. -+ * If the loop ends with a "break;" statement, rules of Inside the loop -+ * apply. -+ */ -+#define ITER_NEXT_VAR(NAME) NAME ## __iterator__next__ ++ flow create 0 ingress pattern integrity level is 0 value mask l4_ok value spec l4_ok / eth / ipv4 proto is udp / end … + + - Connection tracking: + +@@ -508,6 +523,8 @@ Limitations + from the reference "Clock Queue" completions, + the scheduled send timestamps should not be specified with non-zero MSB. + ++- The NIC egress flow rules on representor port are not supported. ++ + Statistics + ---------- + +@@ -554,15 +571,6 @@ Environment variables + The register would be flushed to HW usually when the write-combining buffer + becomes full, but it depends on CPU design. + +- Except for vectorized Tx burst routines, a write memory barrier is enforced +- after updating the register so that the update can be immediately visible to +- HW. +- +- When vectorized Tx burst is called, the barrier is set only if the burst size +- is not aligned to MLX5_VPMD_TX_MAX_BURST. However, setting this environmental +- variable will bring better latency even though the maximum throughput can +- slightly decline. +- + Run-time configuration + ~~~~~~~~~~~~~~~~~~~~~~ + +@@ -649,7 +657,7 @@ Driver options + + A timeout value is set in the driver to control the waiting time before + dropping a packet. Once the timer is expired, the delay drop will be +- deactivated for all the Rx queues with this feature enable. To re-activeate ++ deactivated for all the Rx queues with this feature enable. To re-activate + it, a rearming is needed and it is part of the kernel driver starting from + OFED 5.5. + +@@ -1033,7 +1041,7 @@ Driver options + + For the MARK action the last 16 values in the full range are reserved for + internal PMD purposes (to emulate FLAG action). The valid range for the +- MARK action values is 0-0xFFEF for the 16-bit mode and 0-xFFFFEF ++ MARK action values is 0-0xFFEF for the 16-bit mode and 0-0xFFFFEF + for the 24-bit mode, the flows with the MARK action value outside + the specified range will be rejected. + +@@ -1317,7 +1325,7 @@ DPDK and must be installed separately: + - mlx5_core: hardware driver managing Mellanox + ConnectX-4/ConnectX-5/ConnectX-6/BlueField devices and related Ethernet kernel + network devices. +- - mlx5_ib: InifiniBand device driver. ++ - mlx5_ib: InfiniBand device driver. + - ib_uverbs: user space driver for Verbs (entry point for libibverbs). + + - **Firmware update** +@@ -1382,9 +1390,9 @@ managers on most distributions, this PMD requires Ethernet extensions that + may not be supported at the moment (this is a work in progress). + + `Mellanox OFED +-`__ and ++`__ and + `Mellanox EN +-`__ ++`__ + include the necessary support and should be used in the meantime. For DPDK, + only libibverbs, libmlx5, mlnx-ofed-kernel packages and firmware updates are + required from that distribution. +diff --git a/dpdk/doc/guides/prog_guide/bpf_lib.rst b/dpdk/doc/guides/prog_guide/bpf_lib.rst +index 1feb7734a3..1cf2d59429 100644 +--- a/dpdk/doc/guides/prog_guide/bpf_lib.rst ++++ b/dpdk/doc/guides/prog_guide/bpf_lib.rst +@@ -10,7 +10,7 @@ user-space dpdk application. + + It supports basic set of features from eBPF spec. + Please refer to the +-`eBPF spec ` ++`eBPF spec `_ + for more information. + Also it introduces basic framework to load/unload BPF-based filters + on eth devices (right now only via SW RX/TX callbacks). +@@ -48,9 +48,9 @@ For example, ``(BPF_IND | BPF_W | BPF_LD)`` means: + .. code-block:: c + + uint32_t tmp; +- R0 = rte_pktmbuf_read((const struct rte_mbuf *)R6, src_reg + imm32, +- sizeof(tmp), &tmp); +- if (R0 == NULL) return FAILED; ++ R0 = rte_pktmbuf_read((const struct rte_mbuf *)R6, src_reg + imm32, sizeof(tmp), &tmp); ++ if (R0 == NULL) ++ return FAILED; + R0 = ntohl(*(uint32_t *)R0); + + and ``R1-R5`` were scratched. +diff --git a/dpdk/doc/guides/prog_guide/compressdev.rst b/dpdk/doc/guides/prog_guide/compressdev.rst +index 07d1a62a63..2a59c434c1 100644 +--- a/dpdk/doc/guides/prog_guide/compressdev.rst ++++ b/dpdk/doc/guides/prog_guide/compressdev.rst +@@ -2,7 +2,7 @@ + Copyright(c) 2017-2018 Cavium Networks. + + Compression Device Library +-=========================== ++========================== + + The compression framework provides a generic set of APIs to perform compression services + as well as to query and configure compression devices both physical(hardware) and virtual(software) +@@ -32,10 +32,10 @@ From the command line using the --vdev EAL option + + .. Note:: + +- * If DPDK application requires multiple software compression PMD devices then required +- number of ``--vdev`` with appropriate libraries are to be added. ++ * If a DPDK application requires multiple software compression PMD devices then the ++ required number of ``--vdev`` args with appropriate libraries are to be added. + +- * An Application with multiple compression device instances exposed by the same PMD must ++ * An application with multiple compression device instances exposed by the same PMD must + specify a unique name for each device. + + Example: ``--vdev 'pmd0' --vdev 'pmd1'`` +@@ -53,7 +53,7 @@ All virtual compression devices support the following initialization parameters: + Device Identification + ~~~~~~~~~~~~~~~~~~~~~ + +-Each device, whether virtual or physical is uniquely designated by two ++Each device, whether virtual or physical, is uniquely designated by two + identifiers: + + - A unique device index used to designate the compression device in all functions +@@ -76,7 +76,7 @@ The ``rte_compressdev_configure`` API is used to configure a compression device. + The ``rte_compressdev_config`` structure is used to pass the configuration + parameters. + +-See *DPDK API Reference* for details. ++See the `DPDK API Reference `_ for details. + + Configuration of Queue Pairs + ~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +@@ -85,87 +85,88 @@ Each compression device queue pair is individually configured through the + ``rte_compressdev_queue_pair_setup`` API. + + The ``max_inflight_ops`` is used to pass maximum number of +-rte_comp_op that could be present in a queue at-a-time. +-PMD then can allocate resources accordingly on a specified socket. ++``rte_comp_op`` that could be present in a queue at a time. ++The PMD can then allocate resources accordingly on a specified socket. + +-See *DPDK API Reference* for details. ++See the `DPDK API Reference `_ for details. + +-Logical Cores, Memory and Queues Pair Relationships +-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ++Logical Cores, Memory and Queue Pair Relationships ++~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +-Library supports NUMA similarly as described in Cryptodev library section. ++The Compressdev library supports NUMA similarly as described in Cryptodev library section. + +-A queue pair cannot be shared and should be exclusively used by a single processing +-context for enqueuing operations or dequeuing operations on the same compression device ++A queue pair cannot be shared, and should be exclusively used by a single processing ++context for enqueuing operations or dequeuing operations on the same compression device, + since sharing would require global locks and hinder performance. It is however possible + to use a different logical core to dequeue an operation on a queue pair from the logical +-core on which it was enqueued. This means that a compression burst enqueue/dequeue ++core on which it was enqueued. This means that for a compression burst, enqueue/dequeue + APIs are a logical place to transition from one logical core to another in a + data processing pipeline. + + Device Features and Capabilities +---------------------------------- ++-------------------------------- + + Compression devices define their functionality through two mechanisms, global device +-features and algorithm features. Global devices features identify device +-wide level features which are applicable to the whole device such as supported hardware ++features and algorithm features. Global device features identify device ++wide level features which are applicable to the whole device, such as supported hardware + acceleration and CPU features. List of compression device features can be seen in the + RTE_COMPDEV_FF_XXX macros. + +-The algorithm features lists individual algo feature which device supports per-algorithm, +-such as a stateful compression/decompression, checksums operation etc. List of algorithm +-features can be seen in the RTE_COMP_FF_XXX macros. ++The algorithm features are features which the device supports per-algorithm, ++such as a stateful compression/decompression, checksums operation etc. ++The list of algorithm features can be seen in the RTE_COMP_FF_XXX macros. + + Capabilities + ~~~~~~~~~~~~ + Each PMD has a list of capabilities, including algorithms listed in +-enum ``rte_comp_algorithm`` and its associated feature flag and +-sliding window range in log base 2 value. Sliding window tells +-the minimum and maximum size of lookup window that algorithm uses ++the enum ``rte_comp_algorithm``, its associated feature flag, and ++sliding window range in log base 2 value. The sliding window range ++defines the minimum and maximum size of a lookup window that an algorithm uses + to find duplicates. + +-See *DPDK API Reference* for details. ++See the `DPDK API Reference `_ for details. + + Each Compression poll mode driver defines its array of capabilities +-for each algorithm it supports. See PMD implementation for capability ++for each algorithm it supports. See the PMD implementation for capability + initialization. + + Capabilities Discovery + ~~~~~~~~~~~~~~~~~~~~~~ + +-PMD capability and features are discovered via ``rte_compressdev_info_get`` function. ++PMD capability and features are discovered via the ``rte_compressdev_info_get`` function. + + The ``rte_compressdev_info`` structure contains all the relevant information for the device. + +-See *DPDK API Reference* for details. ++See the `DPDK API Reference `_ for details. + + Compression Operation +----------------------- ++--------------------- + + DPDK compression supports two types of compression methodologies: + +-- Stateless, data associated to a compression operation is compressed without any reference ++- Stateless - data associated with a compression operation is compressed without any reference + to another compression operation. + +-- Stateful, data in each compression operation is compressed with reference to previous compression ++- Stateful - data in each compression operation is compressed with reference to previous compression + operations in the same data stream i.e. history of data is maintained between the operations. + +-For more explanation, please refer RFC https://www.ietf.org/rfc/rfc1951.txt ++For more explanation, please refer to the RFC https://www.ietf.org/rfc/rfc1951.txt + + Operation Representation + ~~~~~~~~~~~~~~~~~~~~~~~~ + +-Compression operation is described via ``struct rte_comp_op``, which contains both input and ++A compression operation is described via ``struct rte_comp_op``, which contains both input and + output data. The operation structure includes the operation type (stateless or stateful), +-the operation status and the priv_xform/stream handle, source, destination and checksum buffer ++the operation status, the priv_xform/stream handle, source, destination and checksum buffer + pointers. It also contains the source mempool from which the operation is allocated. +-PMD updates consumed field with amount of data read from source buffer and produced +-field with amount of data of written into destination buffer along with status of +-operation. See section *Produced, Consumed And Operation Status* for more details. +- +-Compression operations mempool also has an ability to allocate private memory with the +-operation for application's purposes. Application software is responsible for specifying +-all the operation specific fields in the ``rte_comp_op`` structure which are then used ++The PMD updates the consumed field with the amount of data read from the source buffer, ++and the produced field with the amount of data written into the destination buffer, ++along with status of operation. ++See the section :ref:`compressdev_prod_cons_op_status`: for more details. ++ ++The compression operations mempool also has the ability to allocate private memory with the ++operation for the application's use. The application software is responsible for specifying ++all the operation specific fields in the ``rte_comp_op`` structure, which are then used + by the compression PMD to process the requested operation. + + +@@ -181,27 +182,27 @@ A ``rte_comp_op`` contains a field indicating the pool it originated from. + + ``rte_comp_op_alloc()`` and ``rte_comp_op_bulk_alloc()`` are used to allocate + compression operations from a given compression operation mempool. +-The operation gets reset before being returned to a user so that operation ++The operation gets reset before being returned to a user so that the operation + is always in a good known state before use by the application. + + ``rte_comp_op_free()`` is called by the application to return an operation to + its allocating pool. + +-See *DPDK API Reference* for details. ++See the `DPDK API Reference `_ for details. + + Passing source data as mbuf-chain +-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ++~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + If input data is scattered across several different buffers, then +-Application can either parse through all such buffers and make one ++the application can either parse through all such buffers and make one + mbuf-chain and enqueue it for processing or, alternatively, it can +-make multiple sequential enqueue_burst() calls for each of them +-processing them statefully. See *Compression API Stateful Operation* ++make multiple sequential enqueue_burst() calls for each of them, ++processing them statefully. See :ref:`compressdev_stateful_op`: + for stateful processing of ops. + + Operation Status + ~~~~~~~~~~~~~~~~ +-Each operation carries a status information updated by PMD after it is processed. +-Following are currently supported: ++Each operation carries status information updated by the PMD after it is processed. ++The following are currently supported: + + - RTE_COMP_OP_STATUS_SUCCESS, + Operation is successfully completed +@@ -225,22 +226,25 @@ Following are currently supported: + - RTE_COMP_OP_STATUS_OUT_OF_SPACE_RECOVERABLE, + Output buffer ran out of space before operation completed, but this + is not an error case. Output data up to op.produced can be used and +- next op in the stream should continue on from op.consumed+1. ++ the next op in the stream should continue on from op.consumed+1. + + Operation status after enqueue / dequeue + ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + Some of the above values may arise in the op after an +-``rte_compressdev_enqueue_burst()``. If number ops enqueued < number ops requested then +-the app should check the op.status of nb_enqd+1. If status is RTE_COMP_OP_STATUS_NOT_PROCESSED, +-it likely indicates a full-queue case for a hardware device and a retry after dequeuing some ops is likely +-to be successful. If the op holds any other status, e.g. RTE_COMP_OP_STATUS_INVALID_ARGS, a retry with ++``rte_compressdev_enqueue_burst()``. If the number of ops enqueued < the number of ops requested ++then the app should check the op.status of nb_enqd+1. ++If the status is RTE_COMP_OP_STATUS_NOT_PROCESSED, it likely indicates a full-queue case for a ++hardware device, and a retry after dequeuing some ops is likely to be successful. ++If the op holds any other status, e.g. RTE_COMP_OP_STATUS_INVALID_ARGS, a retry with + the same op is unlikely to be successful. + + ++.. _compressdev_prod_cons_op_status: ++ + Produced, Consumed And Operation Status + ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +-- If status is RTE_COMP_OP_STATUS_SUCCESS, ++- If the status is RTE_COMP_OP_STATUS_SUCCESS, + consumed = amount of data read from input buffer, and + produced = amount of data written in destination buffer + - If status is RTE_COMP_OP_STATUS_ERROR, +@@ -253,37 +257,37 @@ Produced, Consumed And Operation Status + - If status is RTE_COMP_OP_STATUS_OUT_OF_SPACE_RECOVERABLE, + consumed = amount of data read, and + produced = amount of data successfully produced until +- out of space condition hit. PMD has ability to recover +- from here, so application can submit next op from +- consumed+1 and a destination buffer with available space. ++ out of space condition hit. The PMD has ability to recover ++ from here, so an application can submit the next op from ++ consumed+1, and a destination buffer with available space. + + Transforms + ---------- + + Compression transforms (``rte_comp_xform``) are the mechanism + to specify the details of the compression operation such as algorithm, +-window size and checksum. ++window size, and checksum. + + Compression API Hash support + ---------------------------- + +-Compression API allows application to enable digest calculation ++The compression API allows an application to enable digest calculation + alongside compression and decompression of data. A PMD reflects its + support for hash algorithms via capability algo feature flags. +-If supported, PMD calculates digest always on plaintext i.e. ++If supported, the PMD always calculates the digest on plaintext i.e. + before compression and after decompression. + + Currently supported list of hash algos are SHA-1 and SHA2 family + SHA256. + +-See *DPDK API Reference* for details. ++See the `DPDK API Reference `_ for details. + +-If required, application should set valid hash algo in compress ++If required, the application should set the valid hash algo in compress + or decompress xforms during ``rte_compressdev_stream_create()`` +-or ``rte_compressdev_private_xform_create()`` and pass a valid ++or ``rte_compressdev_private_xform_create()``, and pass a valid + output buffer in ``rte_comp_op`` hash field struct to store the +-resulting digest. Buffer passed should be contiguous and large +-enough to store digest which is 20 bytes for SHA-1 and ++resulting digest. The buffer passed should be contiguous and large ++enough to store digest, which is 20 bytes for SHA-1 and + 32 bytes for SHA2-256. + + Compression API Stateless operation +@@ -295,20 +299,21 @@ An op is processed stateless if it has + (required only on compression side), + - All required input in source buffer + +-When all of the above conditions are met, PMD initiates stateless processing ++When all of the above conditions are met, the PMD initiates stateless processing + and releases acquired resources after processing of current operation is +-complete. Application can enqueue multiple stateless ops in a single burst ++complete. The application can enqueue multiple stateless ops in a single burst + and must attach priv_xform handle to such ops. + + priv_xform in Stateless operation + ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +-priv_xform is PMD internally managed private data that it maintains to do stateless processing. +-priv_xforms are initialized provided a generic xform structure by an application via making call +-to ``rte_compressdev_private_xform_create``, at an output PMD returns an opaque priv_xform reference. +-If PMD support SHAREABLE priv_xform indicated via algorithm feature flag, then application can +-attach same priv_xform with many stateless ops at-a-time. If not, then application needs to +-create as many priv_xforms as it expects to have stateless operations in-flight. ++A priv_xform is private data managed internally by the PMD to do stateless processing. ++A priv_xform is initialized by an application providing a generic xform structure ++to ``rte_compressdev_private_xform_create``, which returns an opaque priv_xform reference. ++If the PMD supports SHAREABLE priv_xform, indicated via algorithm feature flag, ++then the application can attach the same priv_xform with many stateless ops at a time. ++If not, then the application needs to create as many priv_xforms as it expects to have ++stateless operations in-flight. + + .. figure:: img/stateless-op.* + +@@ -320,8 +325,9 @@ create as many priv_xforms as it expects to have stateless operations in-flight. + Stateless Ops using Shareable priv_xform + + +-Application should call ``rte_compressdev_private_xform_create()`` and attach to stateless op before +-enqueuing them for processing and free via ``rte_compressdev_private_xform_free()`` during termination. ++The application should call ``rte_compressdev_private_xform_create()`` and attach it to a stateless ++op before enqueuing them for processing and free via ``rte_compressdev_private_xform_free()`` ++during termination. + + An example pseudocode to setup and process NUM_OPS stateless ops with each of length OP_LEN + using priv_xform would look like: +@@ -399,75 +405,80 @@ using priv_xform would look like: + + + Stateless and OUT_OF_SPACE +-~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ++~~~~~~~~~~~~~~~~~~~~~~~~~~ + +-OUT_OF_SPACE is a condition when output buffer runs out of space and where PMD +-still has more data to produce. If PMD runs into such condition, then PMD returns +-RTE_COMP_OP_OUT_OF_SPACE_TERMINATED error. In such case, PMD resets itself and can set ++OUT_OF_SPACE is a condition when the output buffer runs out of space and where the PMD ++still has more data to produce. If the PMD runs into such condition, then the PMD returns ++RTE_COMP_OP_OUT_OF_SPACE_TERMINATED error. In such case, the PMD resets itself and can set + consumed=0 and produced=amount of output it could produce before hitting out_of_space. +-Application would need to resubmit the whole input with a larger output buffer, if it ++The application would need to resubmit the whole input with a larger output buffer, if it + wants the operation to be completed. + + Hash in Stateless + ~~~~~~~~~~~~~~~~~ +-If hash is enabled, digest buffer will contain valid data after op is successfully ++If hash is enabled, the digest buffer will contain valid data after an op is successfully + processed i.e. dequeued with status = RTE_COMP_OP_STATUS_SUCCESS. + + Checksum in Stateless + ~~~~~~~~~~~~~~~~~~~~~ +-If checksum is enabled, checksum will only be available after op is successfully ++If checksum is enabled, checksum will only be available after an op is successfully + processed i.e. dequeued with status = RTE_COMP_OP_STATUS_SUCCESS. + ++.. _compressdev_stateful_op: ++ + Compression API Stateful operation + ----------------------------------- + +-Compression API provide RTE_COMP_FF_STATEFUL_COMPRESSION and +-RTE_COMP_FF_STATEFUL_DECOMPRESSION feature flag for PMD to reflect ++The compression API provides RTE_COMP_FF_STATEFUL_COMPRESSION and ++RTE_COMP_FF_STATEFUL_DECOMPRESSION feature flag for the PMD to reflect + its support for Stateful operations. + +-A Stateful operation in DPDK compression means application invokes enqueue +-burst() multiple times to process related chunk of data because +-application broke data into several ops. ++A Stateful operation in DPDK compression means the application invokes enqueue ++burst() multiple times to process a related chunk of data because the ++application broke the data into several ops. + +-In such case ++In such cases + - ops are setup with op_type RTE_COMP_OP_STATEFUL, +-- all ops except last set to flush value = RTE_COMP_FLUSH_NONE/SYNC +-and last set to flush value RTE_COMP_FLUSH_FULL/FINAL. ++- all ops except the last are set with flush value = RTE_COMP_FLUSH_NONE/SYNC ++and the last is set with flush value RTE_COMP_FLUSH_FULL/FINAL. + +-In case of either one or all of the above conditions, PMD initiates +-stateful processing and releases acquired resources after processing ++In case of either one or all of the above conditions, the PMD initiates ++stateful processing and releases acquired resources after processing the + operation with flush value = RTE_COMP_FLUSH_FULL/FINAL is complete. +-Unlike stateless, application can enqueue only one stateful op from +-a particular stream at a time and must attach stream handle ++Unlike stateless, the application can enqueue only one stateful op from ++a particular stream at a time and must attach a stream handle + to each op. + + Stream in Stateful operation + ~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +-`stream` in DPDK compression is a logical entity which identifies related set of ops, say, a one large +-file broken into multiple chunks then file is represented by a stream and each chunk of that file is +-represented by compression op `rte_comp_op`. Whenever application wants a stateful processing of such +-data, then it must get a stream handle via making call to ``rte_compressdev_stream_create()`` +-with xform, at an output the target PMD will return an opaque stream handle to application which +-it must attach to all of the ops carrying data of that stream. In stateful processing, every op +-requires previous op data for compression/decompression. A PMD allocates and set up resources such +-as history, states, etc. within a stream, which are maintained during the processing of the related ops. ++A stream in DPDK compression is a logical entity which identifies a related set of ops. ++For example, one large file broken into multiple chunks, then the file is represented by a stream, ++and each chunk of that file is represented by a compression op ``rte_comp_op``. ++Whenever an application wants stateful processing of such data, then it must get a stream handle ++via making call to ``rte_compressdev_stream_create()`` with an xform, which will return an opaque ++stream handle to attach to all of the ops carrying data of that stream. ++In stateful processing, every op requires previous op data for compression/decompression. ++A PMD allocates and sets up resources such as history, states, etc. within a stream, ++which are maintained during the processing of related ops. + +-Unlike priv_xforms, stream is always a NON_SHAREABLE entity. One stream handle must be attached to only +-one set of related ops and cannot be reused until all of them are processed with status Success or failure. ++Unlike priv_xforms, a stream is always a NON_SHAREABLE entity. One stream handle must be attached ++to only one set of related ops and cannot be reused until all of them are processed with a ++success/failure status. + + .. figure:: img/stateful-op.* + + Stateful Ops + + +-Application should call ``rte_compressdev_stream_create()`` and attach to op before ++An application should call ``rte_compressdev_stream_create()`` and attach it to the op before + enqueuing them for processing and free via ``rte_compressdev_stream_free()`` during +-termination. All ops that are to be processed statefully should carry *same* stream. ++termination. All ops that are to be processed statefully should carry the *same* stream. + +-See *DPDK API Reference* document for details. ++See the `DPDK API Reference `_ for details. + +-An example pseudocode to set up and process a stream having NUM_CHUNKS with each chunk size of CHUNK_LEN would look like: ++An example pseudocode to set up and process a stream having NUM_CHUNKS, ++with each chunk size of CHUNK_LEN, would look like: + + .. code-block:: c + +@@ -549,64 +560,65 @@ An example pseudocode to set up and process a stream having NUM_CHUNKS with each + + + Stateful and OUT_OF_SPACE +-~~~~~~~~~~~~~~~~~~~~~~~~~~~ ++~~~~~~~~~~~~~~~~~~~~~~~~~ + +-If PMD supports stateful operation, then OUT_OF_SPACE status is not an actual +-error for the PMD. In such case, PMD returns with status ++If a PMD supports stateful operation, then an OUT_OF_SPACE status is not an actual ++error for the PMD. In such a case, the PMD returns with status + RTE_COMP_OP_STATUS_OUT_OF_SPACE_RECOVERABLE with consumed = number of input bytes +-read and produced = length of complete output buffer. +-Application should enqueue next op with source starting at consumed+1 and an ++read, and produced = length of complete output buffer. ++The application should enqueue the next op with source starting at consumed+1, and an + output buffer with available space. + + Hash in Stateful + ~~~~~~~~~~~~~~~~ +-If enabled, digest buffer will contain valid digest after last op in stream ++If enabled, the digest buffer will contain valid digest after the last op in a stream + (having flush = RTE_COMP_FLUSH_FINAL) is successfully processed i.e. dequeued + with status = RTE_COMP_OP_STATUS_SUCCESS. + + Checksum in Stateful + ~~~~~~~~~~~~~~~~~~~~ +-If enabled, checksum will only be available after last op in stream ++If enabled, the checksum will only be available after the last op in a stream + (having flush = RTE_COMP_FLUSH_FINAL) is successfully processed i.e. dequeued + with status = RTE_COMP_OP_STATUS_SUCCESS. + + Burst in compression API +-------------------------- ++------------------------ + + Scheduling of compression operations on DPDK's application data path is + performed using a burst oriented asynchronous API set. A queue pair on a compression +-device accepts a burst of compression operations using enqueue burst API. On physical +-devices the enqueue burst API will place the operations to be processed ++device accepts a burst of compression operations using the enqueue burst API. ++On physical devices the enqueue burst API will place the operations to be processed + on the device's hardware input queue, for virtual devices the processing of the + operations is usually completed during the enqueue call to the compression + device. The dequeue burst API will retrieve any processed operations available + from the queue pair on the compression device, from physical devices this is usually +-directly from the devices processed queue, and for virtual device's from a ++directly from the devices processed queue, and for virtual device's from an + ``rte_ring`` where processed operations are placed after being processed on the + enqueue call. + +-A burst in DPDK compression can be a combination of stateless and stateful operations with a condition +-that for stateful ops only one op at-a-time should be enqueued from a particular stream i.e. no-two ops +-should belong to same stream in a single burst. However a burst may contain multiple stateful ops as long +-as each op is attached to a different stream i.e. a burst can look like: ++A burst in DPDK compression can be a combination of stateless and stateful operations with a ++condition that for stateful ops only one op at a time should be enqueued from a particular stream ++i.e. two ops should never belong to the same stream in a single burst. ++However, a burst may contain multiple stateful ops, as long as each op is attached to a different ++stream, i.e. a burst can look like: + + +---------------+--------------+--------------+-----------------+--------------+--------------+ + | enqueue_burst | op1.no_flush | op2.no_flush | op3.flush_final | op4.no_flush | op5.no_flush | + +---------------+--------------+--------------+-----------------+--------------+--------------+ + +-Where, op1 .. op5 all belong to different independent data units. op1, op2, op4, op5 must be stateful +-as stateless ops can only use flush full or final and op3 can be of type stateless or stateful. +-Every op with type set to RTE_COMP_OP_STATELESS must be attached to priv_xform and +-Every op with type set to RTE_COMP_OP_STATEFUL *must* be attached to stream. ++Where, op1 .. op5 all belong to different independent data units. op1, op2, op4, op5 must be ++stateful as stateless ops can only use flush full or final and op3 can be of type stateless or ++stateful. Every op with type set to RTE_COMP_OP_STATELESS must be attached to priv_xform and ++every op with type set to RTE_COMP_OP_STATEFUL *must* be attached to stream. + + Since each operation in a burst is independent and thus can be completed +-out-of-order, applications which need ordering, should setup per-op user data +-area with reordering information so that it can determine enqueue order at ++out of order, applications which need ordering should setup a per-op user data ++area, with reordering information so that it can determine enqueue order at + dequeue. + +-Also if multiple threads calls enqueue_burst() on same queue pair then it’s +-application onus to use proper locking mechanism to ensure exclusive enqueuing +-of operations. ++Also, if multiple threads calls enqueue_burst() on the same queue pair then it's ++the application's responsibility to use a proper locking mechanism to ensure ++exclusive enqueuing of operations. + + Enqueue / Dequeue Burst APIs + ~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +@@ -629,9 +641,10 @@ Sample code + ----------- + + There are unit test applications that show how to use the compressdev library inside +-app/test/test_compressdev.c ++``app/test/test_compressdev.c`` + + Compression Device API + ~~~~~~~~~~~~~~~~~~~~~~ + +-The compressdev Library API is described in the *DPDK API Reference* document. ++The compressdev Library API is described in the ++`DPDK API Reference `_. +diff --git a/dpdk/doc/guides/prog_guide/cryptodev_lib.rst b/dpdk/doc/guides/prog_guide/cryptodev_lib.rst +index 0af35f5e74..8e403cb949 100644 +--- a/dpdk/doc/guides/prog_guide/cryptodev_lib.rst ++++ b/dpdk/doc/guides/prog_guide/cryptodev_lib.rst +@@ -751,7 +751,7 @@ feature is useful when the user wants to abandon partially enqueued operations + for a failed enqueue burst operation and try enqueuing in a whole later. + + Similar as enqueue, there are two dequeue functions: +-``rte_cryptodev_raw_dequeue`` for dequeing single operation, and ++``rte_cryptodev_raw_dequeue`` for dequeuing single operation, and + ``rte_cryptodev_raw_dequeue_burst`` for dequeuing a burst of operations (e.g. + all operations in a ``struct rte_crypto_sym_vec`` descriptor). The + ``rte_cryptodev_raw_dequeue_burst`` function allows the user to provide callback +@@ -1309,6 +1309,7 @@ are shown below. + "enqueue_err_count": 0, "dequeue_err_count": 0}} + + #. Get the capabilities of a particular Crypto device:: ++ + --> /cryptodev/caps,0 + {"/cryptodev/caps": {"crypto_caps": [], "crypto_caps_n": }} +diff --git a/dpdk/doc/guides/prog_guide/env_abstraction_layer.rst b/dpdk/doc/guides/prog_guide/env_abstraction_layer.rst +index 29f6fefc48..c6accce701 100644 +--- a/dpdk/doc/guides/prog_guide/env_abstraction_layer.rst ++++ b/dpdk/doc/guides/prog_guide/env_abstraction_layer.rst +@@ -433,7 +433,7 @@ and decides on a preferred IOVA mode. + + - if all buses report RTE_IOVA_PA, then the preferred IOVA mode is RTE_IOVA_PA, + - if all buses report RTE_IOVA_VA, then the preferred IOVA mode is RTE_IOVA_VA, +-- if all buses report RTE_IOVA_DC, no bus expressed a preferrence, then the ++- if all buses report RTE_IOVA_DC, no bus expressed a preference, then the + preferred mode is RTE_IOVA_DC, + - if the buses disagree (at least one wants RTE_IOVA_PA and at least one wants + RTE_IOVA_VA), then the preferred IOVA mode is RTE_IOVA_DC (see below with the +@@ -658,7 +658,7 @@ Known Issues + + rte_ring + + rte_ring supports multi-producer enqueue and multi-consumer dequeue. +- However, it is non-preemptive, this has a knock on effect of making rte_mempool non-preemptable. ++ However, it is non-preemptive, this has a knock on effect of making rte_mempool non-preemptible. + + .. note:: + +diff --git a/dpdk/doc/guides/prog_guide/event_ethernet_rx_adapter.rst b/dpdk/doc/guides/prog_guide/event_ethernet_rx_adapter.rst +index 67b11e1563..3b4ef502b2 100644 +--- a/dpdk/doc/guides/prog_guide/event_ethernet_rx_adapter.rst ++++ b/dpdk/doc/guides/prog_guide/event_ethernet_rx_adapter.rst +@@ -257,8 +257,8 @@ A loop processing ``rte_event_vector`` containing mbufs is shown below. + /* Process each mbuf. */ + } + break; +- case ... +- ... ++ case default: ++ /* Handle other event_types. */ + } + + Rx event vectorization for SW Rx adapter +diff --git a/dpdk/doc/guides/prog_guide/img/flow_tru_droppper.png b/dpdk/doc/guides/prog_guide/img/flow_tru_dropper.png +similarity index 100% +rename from dpdk/doc/guides/prog_guide/img/flow_tru_droppper.png +rename to dpdk/doc/guides/prog_guide/img/flow_tru_dropper.png +diff --git a/dpdk/doc/guides/prog_guide/img/turbo_tb_decode.svg b/dpdk/doc/guides/prog_guide/img/turbo_tb_decode.svg +index a259f45866..95779c3642 100644 +--- a/dpdk/doc/guides/prog_guide/img/turbo_tb_decode.svg ++++ b/dpdk/doc/guides/prog_guide/img/turbo_tb_decode.svg +@@ -460,7 +460,7 @@ + height="14.642858" + x="39.285713" + y="287.16254" />offse offset offse offset = 4.2.0 +----------------------------------------------------------------- ++Vhost multi-queue reconnection failed with QEMU version 4.2.0 to 5.1.0 ++---------------------------------------------------------------------- + + **Description** + It's a QEMU regression bug (bad commit: c6beefd674ff). QEMU only saves + acked features for one vhost-net when vhost quits. When vhost reconnects + to virtio-net/virtio-pmd in multi-queue situations, the features been +- set multiple times are not consistent. ++ set multiple times are not consistent. QEMU-5.2.0 fixes this issue in commit ++ f66337bdbfda ("vhost-user: save features of multiqueues if chardev is closed"). + + **Implication** + Vhost cannot reconnect back to virtio-net/virtio-pmd normally. +diff --git a/dpdk/doc/guides/rel_notes/release_16_07.rst b/dpdk/doc/guides/rel_notes/release_16_07.rst +index 5be2d171f1..c4f2f71222 100644 +--- a/dpdk/doc/guides/rel_notes/release_16_07.rst ++++ b/dpdk/doc/guides/rel_notes/release_16_07.rst +@@ -192,7 +192,7 @@ EAL + + * **igb_uio: Fixed possible mmap failure for Linux >= 4.5.** + +- The mmaping of the iomem range of the PCI device fails for kernels that ++ The mmapping of the iomem range of the PCI device fails for kernels that + enabled the ``CONFIG_IO_STRICT_DEVMEM`` option. The error seen by the + user is as similar to the following:: + +diff --git a/dpdk/doc/guides/rel_notes/release_17_08.rst b/dpdk/doc/guides/rel_notes/release_17_08.rst +index 25439dad45..1fd1755858 100644 +--- a/dpdk/doc/guides/rel_notes/release_17_08.rst ++++ b/dpdk/doc/guides/rel_notes/release_17_08.rst +@@ -232,7 +232,7 @@ API Changes + * The ``rte_cryptodev_configure()`` function does not create the session + mempool for the device anymore. + * The ``rte_cryptodev_queue_pair_attach_sym_session()`` and +- ``rte_cryptodev_queue_pair_dettach_sym_session()`` functions require ++ ``rte_cryptodev_queue_pair_detach_sym_session()`` functions require + the new parameter ``device id``. + * Parameters of ``rte_cryptodev_sym_session_create()`` were modified to + accept ``mempool``, instead of ``device id`` and ``rte_crypto_sym_xform``. +diff --git a/dpdk/doc/guides/rel_notes/release_21_11.rst b/dpdk/doc/guides/rel_notes/release_21_11.rst +index db09ec01ea..69199a9583 100644 +--- a/dpdk/doc/guides/rel_notes/release_21_11.rst ++++ b/dpdk/doc/guides/rel_notes/release_21_11.rst +@@ -878,3 +878,975 @@ Tested Platforms + + * Kernel version: 5.10 + * Ubuntu 18.04 ++ ++21.11.1 Release Notes ++--------------------- + -+/* Safe initialization declares both iterators. */ -+#define INIT_MULTIVAR_SAFE_SHORT(VAR, MEMBER, POINTER, ITER_TYPE) \ -+ INIT_MULTIVAR_SAFE_SHORT_EXP(VAR, MEMBER, POINTER, ITER_TYPE, (void) 0) + -+#define INIT_MULTIVAR_SAFE_SHORT_EXP(VAR, MEMBER, POINTER, ITER_TYPE, ...) \ -+ ITER_TYPE *ITER_VAR(VAR) = ( __VA_ARGS__ , (ITER_TYPE *) POINTER), \ -+ *ITER_NEXT_VAR(VAR) = NULL ++21.11.1 Fixes ++~~~~~~~~~~~~~ ++ ++* acl: add missing C++ guards ++* app/compress-perf: fix cycle count operations allocation ++* app/compress-perf: fix number of queue pairs to setup ++* app/compress-perf: fix socket ID type during init ++* app/compress-perf: optimize operations pool allocation ++* app/dumpcap: check for failure to set promiscuous ++* app/fib: fix division by zero ++* app/pdump: abort on multi-core capture limit ++* app/regex: fix number of matches ++* app/testpmd: check starting port is not in bonding ++* app/testpmd: fix bonding mode set ++* app/testpmd: fix build without drivers ++* app/testpmd: fix dereference before null check ++* app/testpmd: fix external buffer allocation ++* app/testpmd: fix flow rule with flex input link ++* app/testpmd: fix GENEVE parsing in checksum mode ++* app/testpmd: fix GTP header parsing in checksum engine ++* app/testpmd: fix raw encap of GENEVE option ++* app/testpmd: fix show RSS RETA on Windows ++* app/testpmd: fix stack overflow for EEPROM display ++* app/testpmd: fix Tx scheduling interval ++* baseband/acc100: avoid out-of-bounds access ++* bpf: add missing C++ guards ++* bpf: fix build with some libpcap version on FreeBSD ++* build: fix build on FreeBSD with Meson 0.61.1 ++* build: fix warnings when running external commands ++* build: hide local symbols in shared libraries ++* build: remove deprecated Meson functions ++* build: suppress rte_crypto_asym_op abi check ++* buildtools: fix AVX512 check for Python 3.5 ++* bus/ifpga: remove useless check while browsing devices ++* bus/pci: assign driver pointer before mapping ++* common/cnxk: add missing checks of return values ++* common/cnxk: add workaround for vWQE flush ++* common/cnxk: always use single interrupt ID with NIX ++* common/cnxk: fix base rule merge ++* common/cnxk: fix bitmap usage for TM ++* common/cnxk: fix byte order of frag sizes and infos ++* common/cnxk: fix error checking ++* common/cnxk: fix flow deletion ++* common/cnxk: fix log level during MCAM allocation ++* common/cnxk: fix mbuf data offset for VF ++* common/cnxk: fix nibble parsing order when dumping MCAM ++* common/cnxk: fix NPC key extraction validation ++* common/cnxk: fix null pointer dereferences ++* common/cnxk: fix reset of fields ++* common/cnxk: fix shift offset for TL3 length disable ++* common/cnxk: fix uninitialized pointer read ++* common/cnxk: fix uninitialized variables ++* common/cnxk fix unintended sign extension ++* common/cnxk: reset stale values on error debug registers ++* common/mlx5: add minimum WQE size for striding RQ ++* common/mlx5: add Netlink event helpers ++* common/mlx5: consider local functions as internal ++* common/mlx5: fix error handling in multi-class probe ++* common/mlx5: fix missing validation in devargs parsing ++* common/mlx5: fix MR lookup for non-contiguous mempool ++* common/mlx5: fix probing failure code ++* common/mlx5: fix queue pair ack timeout configuration ++* common/sfc_efx/base: add missing handler for 1-byte fields ++* common/sfc_efx/base: fix recirculation ID set in outer rules ++* compressdev: add missing C++ guards ++* compressdev: fix missing space in log macro ++* compressdev: fix socket ID type ++* compress/mlx5: support out-of-space status ++* compress/octeontx: fix null pointer dereference ++* config: add arch define for Arm ++* config: align mempool elements to 128 bytes on CN10K ++* config/arm: add values for native armv7 ++* crypto/cnxk: enable allocated queues only ++* crypto/cnxk: fix extend tail calculation ++* crypto/cnxk: fix inflight count calculation ++* crypto/cnxk: fix update of number of descriptors ++* cryptodev: add missing C++ guards ++* cryptodev: fix clang C++ include ++* cryptodev: fix RSA key type name ++* crypto/dpaax_sec: fix auth/cipher xform chain checks ++* crypto/ipsec_mb: check missing operation types ++* crypto/ipsec_mb: fix buffer overrun ++* crypto/ipsec_mb: fix GCM requested digest length ++* crypto/ipsec_mb: fix GMAC parameters setting ++* crypto/ipsec_mb: fix length and offset settings ++* crypto/ipsec_mb: fix length and offset settings ++* crypto/ipsec_mb: fix premature dereference ++* crypto/ipsec_mb: fix queue cleanup null pointer dereference ++* crypto/ipsec_mb: fix queue setup null pointer dereference ++* crypto/ipsec_mb: fix tainted data for session ++* crypto/ipsec_mb: fix ZUC authentication verify ++* crypto/ipsec_mb: fix ZUC operation overwrite ++* crypto/ipsec_mb: remove useless check ++* crypto/qat: fix GEN4 AEAD job in raw data path ++* crypto/virtio: fix out-of-bounds access ++* devargs: fix crash with uninitialized parsing ++* devtools: fix comment detection in forbidden token check ++* devtools: fix symbols check ++* devtools: remove event/dlb exception in ABI check ++* distributor: fix potential overflow ++* dma/cnxk: fix installing internal headers ++* dmadev: add missing header include ++* dma/hisilicon: use common PCI device naming ++* dma/idxd: configure maximum batch size to high value ++* dma/idxd: fix burst capacity calculation ++* dma/idxd: fix paths to driver sysfs directory ++* dma/idxd: fix wrap-around in burst capacity calculation ++* doc: add CUDA driver features ++* doc: correct name of BlueField-2 in mlx5 guide ++* doc: fix dlb2 guide ++* doc: fix FIPS guide ++* doc: fix KNI PMD name typo ++* doc: fix missing note on UIO module in Linux guide ++* doc: fix modify field action description for mlx5 ++* doc: fix telemetry example in cryptodev guide ++* doc: fix typos and punctuation in flow API guide ++* doc: improve configuration examples in idxd guide ++* doc: remove dependency on findutils on FreeBSD ++* doc: remove obsolete vector Tx explanations from mlx5 guide ++* doc: replace broken links in mlx guides ++* doc: replace characters for (R) symbol in Linux guide ++* doc: replace deprecated distutils version parsing ++* doc: update matching versions in ice guide ++* eal: add missing C++ guards ++* eal: fix C++ include ++* eal/freebsd: add missing C++ include guards ++* eal/linux: fix device monitor stop return ++* eal/linux: fix illegal memory access in uevent handler ++* eal/linux: log hugepage create errors with filename ++* eal/windows: fix error code for not supported API ++* efd: fix uninitialized structure ++* ethdev: add internal function to device struct from name ++* ethdev: add missing C++ guards ++* ethdev: fix cast for C++ compatibility ++* ethdev: fix doxygen comments for device info struct ++* ethdev: fix MAC address in telemetry device info ++* ethdev: fix Rx queue telemetry memory leak on failure ++* ethdev: remove unnecessary null check ++* event/cnxk: fix QoS devargs parsing ++* event/cnxk: fix Rx adapter config check ++* event/cnxk: fix sub-event clearing mask length ++* event/cnxk: fix uninitialized local variables ++* event/cnxk: fix variables casting ++* eventdev: add missing C++ guards ++* eventdev/eth_rx: fix missing internal port checks ++* eventdev/eth_rx: fix parameters parsing memory leak ++* eventdev/eth_rx: fix queue config query ++* eventdev/eth_tx: fix queue add error code ++* eventdev: fix C++ include ++* eventdev: fix clang C++ include ++* event/dlb2: add shift value check in sparse dequeue ++* event/dlb2: poll HW CQ inflights before mapping queue ++* event/dlb2: update rolling mask used for dequeue ++* examples/distributor: reduce Tx queue number to 1 ++* examples/flow_classify: fix failure message ++* examples/ipsec-secgw: fix buffer freeing in vector mode ++* examples/ipsec-secgw: fix default flow rule creation ++* examples/ipsec-secgw: fix eventdev start sequence ++* examples/ipsec-secgw: fix offload flag used for TSO IPv6 ++* examples/kni: add missing trailing newline in log ++* examples/l2fwd-crypto: fix port mask overflow ++* examples/l3fwd: fix buffer overflow in Tx ++* examples/l3fwd: fix Rx burst size for event mode ++* examples/l3fwd: make Rx and Tx queue size configurable ++* examples/l3fwd: share queue size variables ++* examples/qos_sched: fix core mask overflow ++* examples/vhost: fix launch with physical port ++* fix spelling in comments and strings ++* gpu/cuda: fix dependency loading path ++* gpu/cuda: fix memory list cleanup ++* graph: fix C++ include ++* ipc: end multiprocess thread during cleanup ++* ipsec: fix C++ include ++* kni: add missing C++ guards ++* kni: fix freeing order in device release ++* maintainers: update for stable branches ++* mem: check allocation in dynamic hugepage init ++* mempool/cnxk: fix batch allocation failure path ++* metrics: add missing C++ guards ++* net/af_xdp: add missing trailing newline in logs ++* net/af_xdp: ensure socket is deleted on Rx queue setup error ++* net/af_xdp: fix build with -Wunused-function ++* net/af_xdp: fix custom program loading with multiple queues ++* net/axgbe: use PCI root complex device to distinguish device ++* net/bnxt: add null check for mark table ++* net/bnxt: cap maximum number of unicast MAC addresses ++* net/bnxt: check VF representor pointer before access ++* net/bnxt: fix check for autoneg enablement ++* net/bnxt: fix crash by validating pointer ++* net/bnxt: fix flow create when RSS is disabled ++* net/bnxt: fix handling of VF configuration change ++* net/bnxt: fix memzone allocation per VNIC ++* net/bnxt: fix multicast address set ++* net/bnxt: fix multicast MAC restore during reset recovery ++* net/bnxt: fix null dereference in session cleanup ++* net/bnxt: fix PAM4 mask setting ++* net/bnxt: fix queue stop operation ++* net/bnxt: fix restoring VLAN filtering after recovery ++* net/bnxt: fix ring calculation for representors ++* net/bnxt: fix ring teardown ++* net/bnxt: fix VF resource allocation strategy ++* net/bnxt: fix xstats names query overrun ++* net/bnxt: fix xstats query ++* net/bnxt: get maximum supported multicast filters count ++* net/bnxt: handle ring cleanup in case of error ++* net/bnxt: restore dependency on kernel modules ++* net/bnxt: restore RSS configuration after reset recovery ++* net/bnxt: set fast-path pointers only if recovery succeeds ++* net/bnxt: set HW coalescing parameters ++* net/bonding: fix mode type mismatch ++* net/bonding: fix MTU set for slaves ++* net/bonding: fix offloading configuration ++* net/bonding: fix promiscuous and allmulticast state ++* net/bonding: fix reference count on mbufs ++* net/bonding: fix RSS with early configure ++* net/bonding: fix slaves initializing on MTU setting ++* net/cnxk: fix build with GCC 12 ++* net/cnxk: fix build with optimization ++* net/cnxk: fix inline device RQ tag mask ++* net/cnxk: fix inline IPsec security error handling ++* net/cnxk: fix mbuf data length ++* net/cnxk: fix promiscuous mode in multicast enable flow ++* net/cnxk: fix RSS RETA table update ++* net/cnxk: fix Rx/Tx function update ++* net/cnxk: fix uninitialized local variable ++* net/cnxk: register callback early to handle initial packets ++* net/cxgbe: fix dangling pointer by mailbox access rework ++* net/dpaa2: fix null pointer dereference ++* net/dpaa2: fix timestamping for IEEE1588 ++* net/dpaa2: fix unregistering interrupt handler ++* net/ena: check memory BAR before initializing LLQ ++* net/ena: fix checksum flag for L4 ++* net/ena: fix meta descriptor DF flag setup ++* net/ena: fix reset reason being overwritten ++* net/ena: remove unused enumeration ++* net/ena: remove unused offload variables ++* net/ena: skip timer if reset is triggered ++* net/enic: fix dereference before null check ++* net: fix L2TPv2 common header ++* net/hns3: delete duplicated RSS type ++* net/hns3: fix double decrement of secondary count ++* net/hns3: fix insecure way to query MAC statistics ++* net/hns3: fix mailbox wait time ++* net/hns3: fix max packet size rollback in PF ++* net/hns3: fix operating queue when TCAM table is invalid ++* net/hns3: fix RSS key with null ++* net/hns3: fix RSS TC mode entry ++* net/hns3: fix Rx/Tx functions update ++* net/hns3: fix using enum as boolean ++* net/hns3: fix vector Rx/Tx when PTP enabled ++* net/hns3: fix VF RSS TC mode entry ++* net/hns3: increase time waiting for PF reset completion ++* net/hns3: remove duplicate macro definition ++* net/i40e: enable maximum frame size at port level ++* net/i40e: fix unintentional integer overflow ++* net/iavf: count continuous DD bits for Arm ++* net/iavf: count continuous DD bits for Arm in flex Rx ++* net/iavf: fix AES-GMAC IV size ++* net/iavf: fix function pointer in multi-process ++* net/iavf: fix null pointer dereference ++* net/iavf: fix potential out-of-bounds access ++* net/iavf: fix segmentation offload buffer size ++* net/iavf: fix segmentation offload condition ++* net/iavf: remove git residue symbol ++* net/iavf: reset security context pointer on stop ++* net/iavf: support NAT-T / UDP encapsulation ++* net/ice/base: add profile validation on switch filter ++* net/ice: fix build with 16-byte Rx descriptor ++* net/ice: fix link up when starting device ++* net/ice: fix mbuf offload flag for Rx timestamp ++* net/ice: fix overwriting of LSE bit by DCF ++* net/ice: fix pattern check for flow director parser ++* net/ice: fix pattern check in flow director ++* net/ice: fix Tx checksum offload ++* net/ice: fix Tx checksum offload capability ++* net/ice: fix Tx offload path choice ++* net/ice: track DCF state of PF ++* net/ixgbe: add vector Rx parameter check ++* net/ixgbe: check filter init failure ++* net/ixgbe: fix FSP check for X550EM devices ++* net/ixgbe: reset security context pointer on close ++* net/kni: fix config initialization ++* net/memif: remove pointer deference before null check ++* net/memif: remove unnecessary Rx interrupt stub ++* net/mlx5: fix ASO CT object release ++* net/mlx5: fix assertion on flags set in packet mbuf ++* net/mlx5: fix check in count action validation ++* net/mlx5: fix committed bucket size ++* net/mlx5: fix configuration without Rx queue ++* net/mlx5: fix CPU socket ID for Rx queue creation ++* net/mlx5: fix destroying empty matchers list ++* net/mlx5: fix entry in shared Rx queues list ++* net/mlx5: fix errno update in shared context creation ++* net/mlx5: fix E-Switch manager vport ID ++* net/mlx5: fix flex item availability ++* net/mlx5: fix flex item availability ++* net/mlx5: fix flex item header length translation ++* net/mlx5: fix GCC uninitialized variable warning ++* net/mlx5: fix GRE item translation in Verbs ++* net/mlx5: fix GRE protocol type translation for Verbs ++* net/mlx5: fix implicit tag insertion with sample action ++* net/mlx5: fix indexed pool fetch overlap ++* net/mlx5: fix ineffective metadata argument adjustment ++* net/mlx5: fix inet IPIP protocol type ++* net/mlx5: fix initial link status detection ++* net/mlx5: fix inline length for multi-segment TSO ++* net/mlx5: fix link status change detection ++* net/mlx5: fix mark enabling for Rx ++* net/mlx5: fix matcher priority with ICMP or ICMPv6 ++* net/mlx5: fix maximum packet headers size for TSO ++* net/mlx5: fix memory socket selection in ASO management ++* net/mlx5: fix metadata endianness in modify field action ++* net/mlx5: fix meter capabilities reporting ++* net/mlx5: fix meter creation default state ++* net/mlx5: fix meter policy creation assert ++* net/mlx5: fix meter sub-policy creation ++* net/mlx5: fix modify field MAC address offset ++* net/mlx5: fix modify port action validation ++* net/mlx5: fix MPLS/GRE Verbs spec ordering ++* net/mlx5: fix MPRQ stride devargs adjustment ++* net/mlx5: fix MPRQ WQE size assertion ++* net/mlx5: fix next protocol RSS expansion ++* net/mlx5: fix NIC egress flow mismatch in switchdev mode ++* net/mlx5: fix port matching in sample flow rule ++* net/mlx5: fix RSS expansion with explicit next protocol ++* net/mlx5: fix sample flow action on trusted device ++* net/mlx5: fix shared counter flag in flow validation ++* net/mlx5: fix shared RSS destroy ++* net/mlx5: fix sibling device config check ++* net/mlx5: fix VLAN push action validation ++* net/mlx5: forbid multiple ASO actions in a single rule ++* net/mlx5: improve stride parameter names ++* net/mlx5: reduce flex item flow handle size ++* net/mlx5: reject jump to root table ++* net/mlx5: relax headroom assertion ++* net/mlx5: remove unused function ++* net/mlx5: remove unused reference counter ++* net/mlx5: set flow error for hash list create ++* net/nfb: fix array indexes in deinit functions ++* net/nfb: fix multicast/promiscuous mode switching ++* net/nfp: free HW ring memzone on queue release ++* net/nfp: remove duplicated check when setting MAC address ++* net/nfp: remove useless range checks ++* net/ngbe: fix debug logs ++* net/ngbe: fix missed link interrupt ++* net/ngbe: fix packet statistics ++* net/ngbe: fix Rx by initializing packet buffer early ++* net/ngbe: fix Tx hang on queue disable ++* net/qede: fix maximum Rx packet length ++* net/qede: fix redundant condition in debug code ++* net/qede: fix Rx bulk ++* net/qede: fix Tx completion ++* net/sfc: demand Tx fast free offload on EF10 simple datapath ++* net/sfc: do not push fast free offload to default TxQ config ++* net/sfc: fix flow tunnel support detection ++* net/sfc: fix lock releases ++* net/sfc: fix memory allocation size for cache ++* net/sfc: reduce log level of tunnel restore info error ++* net/sfc: validate queue span when parsing flow action RSS ++* net/tap: fix to populate FDs in secondary process ++* net/txgbe: fix debug logs ++* net/txgbe: fix KR auto-negotiation ++* net/txgbe: fix link up and down ++* net/txgbe: fix queue statistics mapping ++* net/txgbe: reset security context pointer on close ++* net/virtio: fix slots number when indirect feature on ++* net/virtio: fix Tx queue 0 overriden by queue 128 ++* net/virtio: fix uninitialized RSS key ++* net/virtio-user: check FD flags getting failure ++* net/virtio-user: fix resource leak on probing failure ++* pcapng: handle failure of link status query ++* pflock: fix header file installation ++* pipeline: fix annotation checks ++* pipeline: fix table state memory allocation ++* raw/ifpga/base: fix port feature ID ++* raw/ifpga/base: fix SPI transaction ++* raw/ifpga: fix build with optimization ++* raw/ifpga: fix interrupt handle allocation ++* raw/ifpga: fix monitor thread ++* raw/ifpga: fix thread closing ++* raw/ifpga: fix variable initialization in probing ++* raw/ntb: clear all valid doorbell bits on init ++* regexdev: fix section attribute of symbols ++* regex/mlx5: fix memory allocation check ++* Revert "crypto/ipsec_mb: fix length and offset settings" ++* Revert "net/mlx5: fix flex item availability" ++* ring: fix error code when creating ring ++* ring: fix overflow in memory size calculation ++* sched: remove useless malloc in PIE data init ++* stack: fix stubs header export ++* table: fix C++ include ++* telemetry: add missing C++ guards ++* test/bpf: skip dump if conversion fails ++* test/crypto: fix out-of-place SGL in raw datapath ++* test/dma: fix missing checks for device capacity ++* test/efd: fix sockets mask size ++* test/mbuf: fix mbuf data content check ++* test/mem: fix error check ++* vdpa/ifc: fix log info mismatch ++* vdpa/mlx5: workaround queue stop with traffic ++* vdpa/sfc: fix null dereference during config ++* vdpa/sfc: fix null dereference during removal ++* version: 21.11.1-rc1 ++* vfio: cleanup the multiprocess sync handle ++* vhost: add missing C++ guards ++* vhost: fix C++ include ++* vhost: fix FD leak with inflight messages ++* vhost: fix field naming in guest page struct ++* vhost: fix guest to host physical address mapping ++* vhost: fix linker script syntax ++* vhost: fix physical address mapping ++* vhost: fix queue number check when setting inflight FD ++* vhost: fix unsafe vring addresses modifications ++ ++21.11.1 Validation ++~~~~~~~~~~~~~~~~~~ ++ ++* `Nvidia(R) Testing `_ ++ ++ * testpmd send and receive multiple types of traffic ++ * testpmd xstats counters ++ * testpmd timestamp ++ * Changing/checking link status through testpmd ++ * RTE flow ++ * Some RSS ++ * VLAN stripping and insertion ++ * checksum and TSO ++ * ptype ++ * ptype tests. ++ * link_status_interrupt example application ++ * l3fwd-power example application ++ * multi-process example applications ++ * Hardware LRO ++ * Regex application ++ * Buffer Split ++ * Tx scheduling ++ * Compilation tests ++ ++ * ConnectX-4 Lx ++ ++ * Ubuntu 20.04 ++ ++ * driver MLNX_OFED_LINUX-5.5-1.0.3.2 ++ * fw 14.32.1010 ++ ++ * ConnectX-5 ++ ++ * Ubuntu 20.04 ++ ++ * driver MLNX_OFED_LINUX-5.5-1.0.3.2 ++ * fw 16.32.2004 ++ ++ * ConnectX-6 Dx ++ ++ * Ubuntu 20.04 ++ ++ * driver MLNX_OFED_LINUX-5.5-1.0.3.2 ++ * fw 22.32.2004 ++ ++ * BlueField-2 ++ ++ * DOCA SW version: 1.2.1 ++ ++ ++* `Red Hat(R) Testing `_ ++ ++ * RHEL 8 ++ * Kernel 4.18 ++ * QEMU 6.2 ++ * Functionality ++ ++ * PF assignment ++ * VF assignment ++ * vhost single/multi queues and cross-NUMA ++ * vhostclient reconnect ++ * vhost live migration with single/multi queues and cross-NUMA ++ * OVS PVP ++ ++ * Tested NICs ++ ++ * X540-AT2 NIC(ixgbe, 10G) ++ ++ ++* `Intel(R) Testing `_ + -+/* Evaluate the condition expression and, if satisfied, update the _next_ -+ * iterator with the NEXT_EXPR. -+ * Both EXPR and NEXT_EXPR should only use ITER_VAR(VAR) and -+ * ITER_NEXT_VAR(VAR). -+ */ -+#define CONDITION_MULTIVAR_SAFE_SHORT(VAR, MEMBER, EXPR, NEXT_EXPR) \ -+ ((EXPR) ? \ -+ (((VAR) = OBJECT_CONTAINING(ITER_VAR(VAR), VAR, MEMBER)), \ -+ (NEXT_EXPR), 1) : \ -+ (((VAR) = NULL), 0)) ++ * Compilation tests + -+#define UPDATE_MULTIVAR_SAFE_SHORT(VAR) \ -+ UPDATE_MULTIVAR(VAR, ITER_NEXT_VAR(VAR)) ++ * Basic Intel(R) NIC(ixgbe, i40e, ice) + -+/* _LONG versions of the macros. */ ++ * PF (i40e, ixgbe, ice) ++ * VF (i40e, ixgbe, ice) ++ * Intel NIC single core/NIC performance ++ * IPsec test scenarios ++ * Power test scenarios + -+#define INIT_MULTIVAR_SAFE_LONG(VAR, NEXT_VAR, MEMBER, POINTER, ITER_TYPE) \ -+ INIT_MULTIVAR_SAFE_LONG_EXP(VAR, NEXT_VAR, MEMBER, POINTER, ITER_TYPE, \ -+ (void) 0) \ ++ * Basic cryptodev and virtio + -+#define INIT_MULTIVAR_SAFE_LONG_EXP(VAR, NEXT_VAR, MEMBER, POINTER, \ -+ ITER_TYPE, ...) \ -+ ITER_TYPE *ITER_VAR(VAR) = ( __VA_ARGS__ , (ITER_TYPE *) POINTER), \ -+ *ITER_VAR(NEXT_VAR) = NULL ++ * vhost/virtio basic loopback, PVP and performance ++ * cryptodev function ++ * cryptodev performance ++ * vhost_crypto unit test and function/performance test + -+/* Evaluate the condition expression and, if satisfied, update the _next_ -+ * iterator with the NEXT_EXPR. After, evaluate the NEXT_COND and, if -+ * satisfied, set the value to NEXT_VAR. NEXT_COND must use ITER_VAR(NEXT_VAR). -+ * -+ * Both EXPR and NEXT_EXPR should only use ITER_VAR(VAR) and -+ * ITER_VAR(NEXT_VAR). -+ */ -+#define CONDITION_MULTIVAR_SAFE_LONG(VAR, NEXT_VAR, MEMBER, EXPR, NEXT_EXPR, \ -+ NEXT_COND) \ -+ ((EXPR) ? \ -+ (((VAR) = OBJECT_CONTAINING(ITER_VAR(VAR), VAR, MEMBER)), \ -+ (NEXT_EXPR), ((NEXT_COND) ? \ -+ ((NEXT_VAR) = \ -+ OBJECT_CONTAINING(ITER_VAR(NEXT_VAR), NEXT_VAR, MEMBER)) : \ -+ ((NEXT_VAR) = NULL)), 1) : \ -+ (((VAR) = NULL), ((NEXT_VAR) = NULL), 0)) ++* `Canonical(R) Testing `_ + -+#define UPDATE_MULTIVAR_SAFE_LONG(VAR, NEXT_VAR) \ -+ UPDATE_MULTIVAR(VAR, ITER_VAR(NEXT_VAR)) ++ * Build tests of DPDK & OVS 2.13.3 on Ubuntu 20.04 (meson based) ++ * Functional and performance tests based on OVS-DPDK on x86_64 ++ * Autopkgtests for DPDK and OpenvSwitch + -+/* Helpers to allow overloading the *_SAFE iterator macros and select either -+ * the LONG or the SHORT version depending on the number of arguments. -+ */ -+#define GET_SAFE_MACRO2(_1, _2, NAME, ...) NAME -+#define GET_SAFE_MACRO3(_1, _2, _3, NAME, ...) NAME -+#define GET_SAFE_MACRO4(_1, _2, _3, _4, NAME, ...) NAME -+#define GET_SAFE_MACRO5(_1, _2, _3, _4, _5, NAME, ...) NAME -+#define GET_SAFE_MACRO6(_1, _2, _3, _4, _5, _6, NAME, ...) NAME -+#define GET_SAFE_MACRO(MAX_ARGS) GET_SAFE_MACRO ## MAX_ARGS ++21.11.1 Known Issues ++~~~~~~~~~~~~~~~~~~~~ + -+/* MSVC treats __VA_ARGS__ as a simple token in argument lists. Introduce -+ * a level of indirection to work around that. */ -+#define EXPAND_MACRO(name, args) name args ++* DPDK 21.11.1 contains fixes up to DPDK 22.03 ++* Issues identified/fixed in DPDK main branch after DPDK 22.03 may be present in DPDK 21.11.1 + -+/* Overload the LONG and the SHORT version of the macros. MAX_ARGS is the -+ * maximum number of arguments (i.e: the number of arguments of the LONG -+ * version). */ -+#define OVERLOAD_SAFE_MACRO(LONG, SHORT, MAX_ARGS, ...) \ -+ EXPAND_MACRO(GET_SAFE_MACRO(MAX_ARGS), \ -+ (__VA_ARGS__, LONG, SHORT))(__VA_ARGS__) ++21.11.2 Release Notes ++--------------------- + - /* Returns the number of elements in ARRAY. */ - #define ARRAY_SIZE(ARRAY) __ARRAY_SIZE(ARRAY) ++ ++21.11.2 Fixes ++~~~~~~~~~~~~~ ++ ++* acl: fix rules with 8-byte field size ++* app/flow-perf: fix build with GCC 12 ++* app/procinfo: show all non-owned ports ++* app/regex: avoid division by zero ++* app/regex: fix mbuf size for multi-segment buffer ++* app/testpmd: add help messages for multi-process ++* app/testpmd: check statistics query before printing ++* app/testpmd: cleanup port resources after implicit close ++* app/testpmd: do not poll stopped queues ++* app/testpmd: fix bonding slave devices not released ++* app/testpmd: fix flex parser destroy command ++* app/testpmd: fix GTP PSC raw processing ++* app/testpmd: fix GTP PSC raw processing ++* app/testpmd: fix help of create meter command ++* app/testpmd: fix metering and policing command for RFC4115 ++* app/testpmd: fix MTU verification ++* app/testpmd: fix multicast address pool leak ++* app/testpmd: fix packet segment allocation ++* app/testpmd: fix port status of bonding slave device ++* app/testpmd: fix supported RSS offload display ++* app/testpmd: fix use of indirect action after port close ++* app/testpmd: perform SW IP checksum for GRO/GSO packets ++* app/testpmd: remove useless pointer checks ++* app/testpmd: replace hardcoded min mbuf number with macro ++* app/testpmd: revert MAC update in checksum forwarding ++* avoid AltiVec keyword vector ++* baseband/acc100: add protection for some negative scenario ++* baseband/acc100: update companion PF configure function ++* bus/fslmc: fix VFIO setup ++* common/cnxk: allow changing PTP mode on CN10K ++* common/cnxk: fix decrypt packet count register update ++* common/cnxk: fix GRE tunnel parsing ++* common/cnxk: fix null pointer dereference ++* common/cnxk: fix SQ flush sequence ++* common/cnxk: fix unaligned access to device memory ++* common/cnxk: handle ROC model init failure ++* common/cnxk: swap zuc-256 key ++* common/cpt: fix build with GCC 12 ++* common/dpaax: fix short MAC-I IV calculation for ZUC ++* common/mlx5: fix memory region range calculation ++* common/mlx5: fix non-expandable global MR cache ++* common/mlx5: remove unused lcore check ++* common/sfc_efx/base: convert EFX PCIe INTF to MCDI value ++* config: fix C++ cross compiler for Arm and PPC ++* crypto/cnxk: fix build with GCC 12 ++* crypto/cnxk: swap zuc-256 iv ++* crypto/dpaa2_sec: fix buffer pool ID check ++* crypto/dpaa2_sec: fix chained FD length in raw datapath ++* crypto/dpaa2_sec: fix crypto operation pointer ++* crypto/dpaa2_sec: fix fle buffer leak ++* crypto/dpaa2_sec: fix operation status for simple FD ++* crypto/dpaa_sec: fix chained FD length in raw datapath ++* crypto/dpaa_sec: fix digest size ++* crypto/dpaa_sec: fix secondary process probing ++* crypto/ipsec_mb: fix build with GCC 12 ++* crypto/mlx5: fix login cleanup ++* crypto/qat: fix DOCSIS crash ++* crypto/scheduler: fix queue pair in scheduler failover ++* devargs: fix leak on hotplug failure ++* devtools: fix null test for NUMA systems ++* dma/hisilicon: enhance CQ scan robustness ++* dma/hisilicon: fix includes in header file ++* dma/hisilicon: fix index returned when no DMA completed ++* dma/idxd: fix AVX2 in non-datapath functions ++* dma/idxd: fix error code for PCI device commands ++* dma/idxd: fix memory leak in PCI close ++* dma/idxd: fix non-AVX builds with old compilers ++* dma/idxd: fix null dereference in PCI remove ++* dma/idxd: fix partial freeing in PCI close ++* dma/skeleton: fix index returned when no memcpy completed ++* doc: add missing auth algo for IPsec example ++* doc: add more instructions for running as non-root ++* doc: fix API index Markdown syntax ++* doc: fix build with sphinx 4.5 ++* doc: fix flow integrity hardware support in mlx5 guide ++* doc: fix formatting and link in BPF library guide ++* doc: fix grammar and formatting in compressdev guide ++* doc: fix grammar and parameters in l2fwd-crypto guide ++* doc: fix readability in vhost guide ++* doc: fix release note typo ++* doc: fix vhost multi-queue reconnection ++* doc: update matching versions in i40e guide ++* doc: update matching versions in ice guide ++* drivers/crypto: fix warnings for OpenSSL version ++* eal: fix C++ include for device event and DMA ++* eal/freebsd: fix use of newer cpuset macros ++* eal/ppc: fix compilation for musl ++* eal/windows: add missing C++ include guards ++* eal/windows: fix data race when creating threads ++* eal/x86: drop export of internal alignment macro ++* eal/x86: fix unaligned access for small memcpy ++* ethdev: fix build with vtune option ++* ethdev: fix memory leak in xstats telemetry ++* ethdev: fix port close in secondary process ++* ethdev: fix port state when stop ++* ethdev: fix possible null pointer access ++* ethdev: fix RSS update when RSS is disabled ++* ethdev: prohibit polling stopped queue ++* event/cnxk: fix out of bounds access in test ++* event/cnxk: fix QoS parameter handling ++* event/cnxk: fix Tx adapter enqueue return for CN10K ++* eventdev/eth_rx: fix telemetry Rx stats reset ++* eventdev/eth_tx: fix adapter creation ++* eventdev/eth_tx: fix queue delete ++* event/dlb2: fix advertized capabilities ++* event/dlb2: fix check of QID in-flight ++* event/dlb2: rework queue drain handling ++* event/octeontx: fix SSO fast path ++* examples/bond: fix invalid use of trylock ++* examples/distributor: fix distributor on Rx core ++* examples/dma: fix MTU configuration ++* examples/dma: fix Tx drop statistics ++* examples/fips_validation: handle empty payload ++* examples/ipsec-secgw: fix ESN setting ++* examples/ipsec-secgw: fix NAT-T header fields ++* examples/ipsec-secgw: fix promiscuous mode option ++* examples/ipsec-secgw: fix uninitialized memory access ++* examples/l2fwd-crypto: fix stats refresh rate ++* examples/link_status_interrupt: fix stats refresh rate ++* examples/performance-thread: fix build with GCC 12 ++* examples/vhost: fix crash when no VMDq ++* examples/vhost: fix retry logic on Rx path ++* gro: fix identifying fragmented packets ++* ipsec: fix NAT-T ports and length ++* kni: fix build ++* kni: fix build with Linux 5.18 ++* kni: use dedicated function to set MAC address ++* kni: use dedicated function to set random MAC address ++* malloc: fix allocation of almost hugepage size ++* malloc: fix ASan handling for unmapped memory ++* mbuf: dump outer VLAN ++* mem: skip attaching external memory in secondary process ++* net/af_xdp: make compatible with libbpf >= 0.7.0 ++* net/af_xdp: use libxdp if available ++* net/axgbe: fix xstats get return if xstats is null ++* net/bnxt: allow Tx only or Rx only ++* net/bnxt: avoid unnecessary endianness conversion ++* net/bnxt: check duplicate queue IDs ++* net/bnxt: cleanup MTU setting ++* net/bnxt: disallow MTU change when device is started ++* net/bnxt: fix check for autoneg enablement in the PHY FW ++* net/bnxt: fix compatibility with some old firmwares ++* net/bnxt: fix device capability reporting ++* net/bnxt: fix freeing VNIC filters ++* net/bnxt: fix link status when port is stopped ++* net/bnxt: fix reordering in NEON Rx ++* net/bnxt: fix ring group on Rx restart ++* net/bnxt: fix RSS action ++* net/bnxt: fix Rx configuration ++* net/bnxt: fix setting forced speed ++* net/bnxt: fix speed autonegotiation ++* net/bnxt: fix switch domain allocation ++* net/bnxt: fix tunnel stateless offloads ++* net/bnxt: fix ULP parser to ignore segment offset ++* net/bnxt: force PHY update on certain configurations ++* net/bnxt: handle queue stop during RSS flow create ++* net/bnxt: recheck FW readiness if in reset process ++* net/bnxt: remove unused macro ++* net/bonding: fix mbuf fast free usage ++* net/bonding: fix RSS inconsistency between ports ++* net/bonding: fix RSS key config with extended key length ++* net/bonding: fix slave stop and remove on port close ++* net/bonding: fix stopping non-active slaves ++* net/cnxk: add barrier after meta batch free in scalar ++* net/cnxk: add message on flow parsing failure ++* net/cnxk: fix possible null dereference in telemetry ++* net/cnxk: fix uninitialized variables ++* net/cxgbe: fix port ID in Rx mbuf ++* net/cxgbe: fix Tx queue stuck with mbuf chain coalescing ++* net/dpaa2: fix dpdmux default interface ++* net/dpaa: fix event queue detach ++* net/ena: fix build with GCC 12 ++* net/enetfec: fix build with GCC 12 ++* net/failsafe: fix device freeing ++* net: fix GTP PSC headers ++* net/hns3: delete unused code ++* net/hns3: fix an unreasonable memset ++* net/hns3: fix code check warning ++* net/hns3: fix crash from secondary process ++* net/hns3: fix descriptors check with SVE ++* net/hns3: fix link status capability query from VF ++* net/hns3: fix MAC and queues HW statistics overflow ++* net/hns3: fix mbuf free on Tx done cleanup ++* net/hns3: fix order of clearing imissed register in PF ++* net/hns3: fix pseudo-sharing between threads ++* net/hns3: fix PTP interrupt logging ++* net/hns3: fix return value for unsupported tuple ++* net/hns3: fix rollback on RSS hash update ++* net/hns3: fix RSS disable ++* net/hns3: fix statistics locking ++* net/hns3: fix TM capability ++* net/hns3: fix xstats get return if xstats is null ++* net/hns3: remove duplicate definition ++* net/hns3: remove redundant RSS tuple field ++* net/hns3: remove unnecessary RSS switch ++* net/hns3: support backplane media type ++* net/i40e: fix max frame size config at port level ++* net/i40e: populate error in flow director parser ++* net/iavf: fix data path selection ++* net/iavf: fix device initialization without inline crypto ++* net/iavf: fix device stop ++* net/iavf: fix GTP-U extension flow ++* net/iavf: fix mbuf release in multi-process ++* net/iavf: fix NAT-T payload length ++* net/iavf: fix queue start exception handling ++* net/iavf: fix Rx queue interrupt setting ++* net/iavf: fix segfaults when calling API after VF reset failed ++* net/iavf: fix VF reset ++* net/iavf: increase reset complete wait count ++* net/iavf: remove dead code ++* net/ice: add missing Tx burst mode name ++* net/ice/base: fix build with GCC 12 ++* net/ice/base: fix direction of flow that matches any ++* net/ice/base: fix getting sched node from ID type ++* net/ice: fix build with GCC 12 ++* net/ice: fix MTU info for DCF ++* net/ice: fix race condition in Rx timestamp ++* net/ice: fix raw flow input pattern parsing ++* net/ice: improve performance of Rx timestamp offload ++* net/ice: refactor parser usage ++* net/igc: support multi-process ++* net/ipn3ke: fix xstats get return if xstats is null ++* net/ixgbe: add option for link up check on pin SDP3 ++* net/memif: fix overwriting of head segment ++* net/mlx5: add limitation for E-Switch Manager match ++* net/mlx5: fix build with clang 14 ++* net/mlx5: fix counter in non-termination meter ++* net/mlx5: fix GTP handling in header modify action ++* net/mlx5: fix LRO configuration in drop Rx queue ++* net/mlx5: fix LRO validation in Rx setup ++* net/mlx5: fix metering on E-Switch Manager ++* net/mlx5: fix no-green metering with RSS ++* net/mlx5: fix probing with secondary bonding member ++* net/mlx5: fix RSS expansion for patterns with ICMP item ++* net/mlx5: fix RSS hash types adjustment ++* net/mlx5: fix Rx queue recovery mechanism ++* net/mlx5: fix Rx/Tx stats concurrency ++* net/mlx5: fix stack buffer overflow in drop action ++* net/mlx5: fix statistics read on Linux ++* net/mlx5: fix Tx recovery ++* net/mlx5: fix Tx when inlining is impossible ++* net/mlx5: reject negative integrity item configuration ++* net/mlx5: restrict Rx queue array access to boundary ++* net/mvpp2: fix xstats get return if xstats is null ++* net/netvsc: fix calculation of checksums based on mbuf flag ++* net/netvsc: fix hot adding multiple VF PCI devices ++* net/netvsc: fix vmbus device reference in multi-process ++* net/nfp: fix disabling VLAN stripping ++* net/nfp: fix initialization ++* net/nfp: make sure MTU is never larger than mbuf size ++* net/nfp: remove unneeded header inclusion ++* net/nfp: update how max MTU is read ++* net/ngbe: add more packet statistics ++* net/ngbe: fix link speed check ++* net/ngbe: fix PCIe related operations with bus API ++* net/ngbe: fix reading PHY ID ++* net/octeontx: fix port close ++* net/qede: fix build with GCC 12 ++* net/qede: fix build with GCC 13 ++* net/tap: fix device freeing ++* net/tap: fix interrupt handler freeing ++* net/txgbe: fix max number of queues for SR-IOV ++* net/txgbe: fix register polling ++* net/txgbe: fix SGMII mode to link up ++* net/vhost: fix access to freed memory ++* net/vhost: fix deadlock on vring state change ++* net/vhost: fix null pointer dereference ++* net/vhost: fix TSO feature default disablement ++* net/virtio: restore some optimisations with AVX512 ++* net/virtio: unmap PCI device in secondary process ++* net/virtio-user: fix Rx interrupts with multi-queue ++* net/virtio-user: fix socket non-blocking mode ++* net/vmxnet3: fix Rx data ring initialization ++* pcapng: fix timestamp wrapping in output files ++* pipeline: fix emit instruction for invalid headers ++* raw/ifpga: remove virtual devices on close ++* raw/ifpga: unregister interrupt on close ++* raw/ioat: fix build missing errno include ++* raw/ioat: fix build when ioat dmadev enabled ++* rib: fix references for IPv6 implementation ++* rib: fix traversal with /32 route ++* sched: remove unnecessary floating point ++* security: fix SA lifetime comments ++* service: fix lingering active status ++* test: avoid hang if queues are full and Tx fails ++* test/bonding: fix RSS test when disable RSS ++* test/bpf: skip test if libpcap is unavailable ++* test: check memory allocation for CRC ++* test/crypto: fix authentication IV for ZUC SGL ++* test/crypto: fix cipher offset for ZUC ++* test/crypto: fix driver name for DPAA raw API test ++* test/crypto: fix null check for ZUC authentication ++* test/crypto: fix SNOW3G vector IV format ++* test/crypto: fix ZUC vector IV format ++* test/crypto: skip oop test for raw api ++* test: drop reference to removed tests ++* test/hash: fix out of bound access ++* test/ipsec: fix build with GCC 12 ++* test/ipsec: fix performance test ++* test/mem: disable ASan when accessing unallocated memory ++* test/table: fix buffer overflow on lpm entry ++* trace: fix crash when exiting ++* trace: fix init with long file prefix ++* vdpa/ifc/base: fix null pointer dereference ++* vdpa/ifc: fix build with GCC 12 ++* vdpa/mlx5: fix dead loop when process interrupted ++* vdpa/mlx5: fix interrupt trash that leads to crash ++* vdpa/mlx5: fix leak on event thread creation ++* vdpa/mlx5: fix maximum number of virtqs ++* vdpa/mlx5: workaround var offset within page ++* vdpa/sfc: fix sync between QEMU and vhost-user ++* vdpa/sfc: resolve race between vhost lib and device conf ++* version: 21.11.2-rc1 ++* vhost: add some trailing newline in log messages ++* vhost/crypto: fix build with GCC 12 ++* vhost/crypto: fix descriptor processing ++* vhost: discard too small descriptor chains ++* vhost: fix async access ++* vhost: fix deadlock when message handling failed ++* vhost: fix header spanned across more than two descriptors ++* vhost: fix missing enqueue pseudo-header calculation ++* vhost: fix missing virtqueue lock protection ++* vhost: restore device information in log messages ++ ++21.11.2 Validation ++~~~~~~~~~~~~~~~~~~ ++ ++* `Red Hat(R) Testing `__ ++ ++ * Platform ++ ++ * RHEL 8 ++ * Kernel 4.18 ++ * Qemu 6.2 ++ * X540-AT2 NIC(ixgbe, 10G) ++ ++ * Functionality ++ ++ * Guest with device assignment(PF) throughput testing(1G hugepage size) ++ * Guest with device assignment(PF) throughput testing(2M hugepage size) ++ * Guest with device assignment(VF) throughput testing ++ * PVP (host dpdk testpmd as vswitch) 1Q: throughput testing ++ * PVP vhost-user 2Q throughput testing ++ * PVP vhost-user 1Q cross numa node throughput testing ++ * Guest with vhost-user 2 queues throughput testing ++ * vhost-user reconnect with dpdk-client, qemu-server: qemu reconnect ++ * vhost-user reconnect with dpdk-client, qemu-server: ovs reconnect ++ * PVP 1Q live migration testing ++ * PVP 1Q cross numa node live migration testing ++ * Guest with ovs+dpdk+vhost-user 1Q live migration testing ++ * Guest with ovs+dpdk+vhost-user 1Q live migration testing (2M) ++ * Guest with ovs+dpdk+vhost-user 2Q live migration testing ++ * Guest with ovs+dpdk+vhost-user 4Q live migration testing ++ * Host PF + DPDK testing ++ * Host VF + DPDK testing ++ ++ ++* `Intel(R) Testing `__ ++ ++ * Basic Intel(R) NIC(ixgbe, i40e and ice) testing ++ ++ * PF (i40e) ++ * PF (ixgbe) ++ * PF (ice) ++ * VF (i40e) ++ * VF (ixgbe) ++ * VF (ice) ++ * Compile Testing ++ * Intel NIC single core/NIC performance ++ * Power and IPsec ++ ++ * Basic cryptodev and virtio testing ++ ++ * vhost/virtio basic loopback, PVP and performance test ++ * cryptodev Function/Performance ++ ++ ++* `Nvidia(R) Testing `__ ++ ++ * Basic functionality with testpmd ++ ++ * Tx/Rx ++ * xstats ++ * Timestamps ++ * Link status ++ * RTE flow and flow_director ++ * RSS ++ * VLAN filtering, stripping and insertion ++ * Checksum/TSO ++ * ptype ++ * link_status_interrupt example application ++ * l3fwd-power example application ++ * Multi-process example applications ++ * Hardware LRO tests ++ * Regex application ++ * Buffer Split ++ * Tx scheduling ++ ++ * Build tests ++ ++ * Ubuntu 20.04.4 with MLNX_OFED_LINUX-5.7-1.0.2.0. ++ * Ubuntu 20.04.4 with rdma-core master (23a0021). ++ * Ubuntu 20.04.4 with rdma-core v28.0. ++ * Ubuntu 18.04.6 with rdma-core v17.1. ++ * Ubuntu 18.04.6 with rdma-core master (23a0021) (i386). ++ * Ubuntu 16.04.7 with rdma-core v22.7. ++ * Fedora 35 with rdma-core v39.0. ++ * Fedora 37 (Rawhide) with rdma-core v39.0 (with clang only). ++ * CentOS 7 7.9.2009 with rdma-core master (23a0021). ++ * CentOS 7 7.9.2009 with MLNX_OFED_LINUX-5.7-1.0.2.0. ++ * CentOS 8 8.4.2105 with rdma-core master (23a0021). ++ * OpenSUSE Leap 15.4 with rdma-core v38.1. ++ * Windows Server 2019 with Clang 11.0.0. ++ ++ * ConnectX-6 Dx ++ ++ * Ubuntu 20.04 ++ * Driver MLNX_OFED_LINUX-5.7-1.0.2.0 ++ * fw 22.34.1002 ++ ++ * ConnectX-5 ++ ++ * Ubuntu 20.04 ++ * Driver MLNX_OFED_LINUX-5.7-1.0.2.0 ++ * fw 16.34.1002 ++ ++ * ConnectX-4 Lx ++ ++ * Ubuntu 20.04 ++ * Driver MLNX_OFED_LINUX-5.7-1.0.2.0 ++ * fw 14.32.1010 ++ ++ * BlueField-2 ++ ++ * DOCA SW version: 1.4.0 ++ ++ ++* `Intel(R) Testing with Open vSwitch `__ ++ ++ * 21.11.2 validated by Intel for i40e, ICE, vhost and MTU for OVS with DPDK ++ ++21.11.2 Known Issues ++~~~~~~~~~~~~~~~~~~~~ ++ ++* DPDK 21.11.2 contains fixes up to DPDK 22.07 as well as fixes for CVE-2022-28199 and CVE-2022-2132 ++* Issues identified/fixed in DPDK main branch after DPDK 22.07 may be present in DPDK 21.11.2 +diff --git a/dpdk/doc/guides/rel_notes/release_2_1.rst b/dpdk/doc/guides/rel_notes/release_2_1.rst +index 35e6c88884..d0ad99ebce 100644 +--- a/dpdk/doc/guides/rel_notes/release_2_1.rst ++++ b/dpdk/doc/guides/rel_notes/release_2_1.rst +@@ -671,7 +671,7 @@ Resolved Issues + value 0. + + +- Fixes: 40b966a211ab ("ivshmem: library changes for mmaping using ivshmem") ++ Fixes: 40b966a211ab ("ivshmem: library changes for mmapping using ivshmem") + + + * **ixgbe/base: Fix SFP probing.** +diff --git a/dpdk/doc/guides/sample_app_ug/fips_validation.rst b/dpdk/doc/guides/sample_app_ug/fips_validation.rst +index 56df434215..39baea3346 100644 +--- a/dpdk/doc/guides/sample_app_ug/fips_validation.rst ++++ b/dpdk/doc/guides/sample_app_ug/fips_validation.rst +@@ -77,11 +77,12 @@ Compiling the Application + .. code-block:: console + + dos2unix AES/req/* +- dos2unix AES_GCM/req/* ++ dos2unix GCM/req/* + dos2unix CCM/req/* + dos2unix CMAC/req/* + dos2unix HMAC/req/* + dos2unix TDES/req/* ++ dos2unix SHA/req/* + + Running the Application + ----------------------- +diff --git a/dpdk/doc/guides/sample_app_ug/ip_reassembly.rst b/dpdk/doc/guides/sample_app_ug/ip_reassembly.rst +index 06289c2248..5280bf4ea0 100644 +--- a/dpdk/doc/guides/sample_app_ug/ip_reassembly.rst ++++ b/dpdk/doc/guides/sample_app_ug/ip_reassembly.rst +@@ -154,8 +154,8 @@ each RX queue uses its own mempool. + + .. literalinclude:: ../../../examples/ip_reassembly/main.c + :language: c +- :start-after: mbufs stored int the gragment table. 8< +- :end-before: >8 End of mbufs stored int the fragmentation table. ++ :start-after: mbufs stored in the fragment table. 8< ++ :end-before: >8 End of mbufs stored in the fragmentation table. + :dedent: 1 + + Packet Reassembly and Forwarding +diff --git a/dpdk/doc/guides/sample_app_ug/ipsec_secgw.rst b/dpdk/doc/guides/sample_app_ug/ipsec_secgw.rst +index c53ee7c386..468a977478 100644 +--- a/dpdk/doc/guides/sample_app_ug/ipsec_secgw.rst ++++ b/dpdk/doc/guides/sample_app_ug/ipsec_secgw.rst +@@ -116,7 +116,8 @@ Constraints + * No IPv6 options headers. + * No AH mode. + * Supported algorithms: AES-CBC, AES-CTR, AES-GCM, 3DES-CBC, HMAC-SHA1, +- AES-GMAC, AES_CTR, AES_XCBC_MAC, AES_CCM, CHACHA20_POLY1305 and NULL. ++ HMAC-SHA256, AES-GMAC, AES_CTR, AES_XCBC_MAC, AES_CCM, CHACHA20_POLY1305 ++ and NULL. + * Each SA must be handle by a unique lcore (*1 RX queue per port*). -@@ -285,6 +429,9 @@ is_pow2(uintmax_t x) - * segfault, so it is important to be aware of correct alignment. */ - #define ALIGNED_CAST(TYPE, ATTR) ((TYPE) (void *) (ATTR)) + Compiling the Application +@@ -586,6 +587,7 @@ where each options means: -+#define IS_PTR_ALIGNED(OBJ) \ -+ (!(OBJ) || (uintptr_t) (OBJ) % __alignof__(OVS_TYPEOF(OBJ)) == 0) -+ - #ifdef __cplusplus + * *null*: NULL algorithm + * *sha1-hmac*: HMAC SHA1 algorithm ++ * *sha256-hmac*: HMAC SHA256 algorithm + + ```` + +diff --git a/dpdk/doc/guides/sample_app_ug/l2_forward_cat.rst b/dpdk/doc/guides/sample_app_ug/l2_forward_cat.rst +index 440642ef7c..3ada3575ba 100644 +--- a/dpdk/doc/guides/sample_app_ug/l2_forward_cat.rst ++++ b/dpdk/doc/guides/sample_app_ug/l2_forward_cat.rst +@@ -176,7 +176,7 @@ function. The value returned is the number of parsed arguments: + .. literalinclude:: ../../../examples/l2fwd-cat/l2fwd-cat.c + :language: c + :start-after: Initialize the Environment Abstraction Layer (EAL). 8< +- :end-before: >8 End of initializion the Environment Abstraction Layer (EAL). ++ :end-before: >8 End of initialization the Environment Abstraction Layer (EAL). + :dedent: 1 + + The next task is to initialize the PQoS library and configure CAT. The +diff --git a/dpdk/doc/guides/sample_app_ug/l2_forward_crypto.rst b/dpdk/doc/guides/sample_app_ug/l2_forward_crypto.rst +index 1b4444b7d8..ce49eab96f 100644 +--- a/dpdk/doc/guides/sample_app_ug/l2_forward_crypto.rst ++++ b/dpdk/doc/guides/sample_app_ug/l2_forward_crypto.rst +@@ -15,7 +15,7 @@ Overview + The L2 Forwarding with Crypto sample application performs a crypto operation (cipher/hash) + specified by the user from command line (or using the default values), + with a crypto device capable of doing that operation, +-for each packet that is received on a RX_PORT and performs L2 forwarding. ++for each packet that is received on an RX_PORT and performs L2 forwarding. + The destination port is the adjacent port from the enabled portmask, that is, + if the first four ports are enabled (portmask 0xf), + ports 0 and 1 forward into each other, and ports 2 and 3 forward into each other. +@@ -54,37 +54,37 @@ The application requires a number of command line options: + + where, + +-* p PORTMASK: A hexadecimal bitmask of the ports to configure (default is all the ports) ++* p PORTMASK: A hexadecimal bitmask of the ports to configure. (Default is all the ports.) + +-* q NQ: A number of queues (=ports) per lcore (default is 1) ++* q NQ: A number of queues (=ports) per lcore. (Default is 1.) + +-* s: manage all ports from single core ++* s: manage all ports from a single core. + +-* T PERIOD: statistics will be refreshed each PERIOD seconds ++* T PERIOD: statistics will be refreshed each PERIOD seconds. + +- (0 to disable, 10 default, 86400 maximum) ++ (0 to disable, 10 default, 86400 maximum.) + +-* cdev_type: select preferred crypto device type: HW, SW or anything (ANY) ++* cdev_type: select preferred crypto device type: HW, SW or anything (ANY). + +- (default is ANY) ++ (Default is ANY.) + + * chain: select the operation chaining to perform: Cipher->Hash (CIPHER_HASH), + + Hash->Cipher (HASH_CIPHER), Cipher (CIPHER_ONLY), Hash (HASH_ONLY) + +- or AEAD (AEAD) ++ or AEAD (AEAD). + +- (default is Cipher->Hash) ++ (Default is Cipher->Hash.) + +-* cipher_algo: select the ciphering algorithm (default is aes-cbc) ++* cipher_algo: select the ciphering algorithm. (Default is aes-cbc.) + +-* cipher_op: select the ciphering operation to perform: ENCRYPT or DECRYPT ++* cipher_op: select the ciphering operation to perform: ENCRYPT or DECRYPT. + +- (default is ENCRYPT) ++ (Default is ENCRYPT.) + + * cipher_dataunit_len: set the length of the cipher data-unit. + +-* cipher_key: set the ciphering key to be used. Bytes has to be separated with ":" ++* cipher_key: set the ciphering key to be used. Bytes have to be separated with ":". + + * cipher_key_random_size: set the size of the ciphering key, + +@@ -92,19 +92,19 @@ where, + + Note that if --cipher_key is used, this will be ignored. + +-* cipher_iv: set the cipher IV to be used. Bytes has to be separated with ":" ++* cipher_iv: set the cipher IV to be used. Bytes have to be separated with ":". + + * cipher_iv_random_size: set the size of the cipher IV, which will be generated randomly. + + Note that if --cipher_iv is used, this will be ignored. + +-* auth_algo: select the authentication algorithm (default is sha1-hmac) ++* auth_algo: select the authentication algorithm. (Default is sha1-hmac.) + +-* auth_op: select the authentication operation to perform: GENERATE or VERIFY ++* auth_op: select the authentication operation to perform: GENERATE or VERIFY. + +- (default is GENERATE) ++ (Default is GENERATE.) + +-* auth_key: set the authentication key to be used. Bytes has to be separated with ":" ++* auth_key: set the authentication key to be used. Bytes have to be separated with ":". + + * auth_key_random_size: set the size of the authentication key, + +@@ -112,19 +112,19 @@ where, + + Note that if --auth_key is used, this will be ignored. + +-* auth_iv: set the auth IV to be used. Bytes has to be separated with ":" ++* auth_iv: set the auth IV to be used. Bytes have to be separated with ":". + + * auth_iv_random_size: set the size of the auth IV, which will be generated randomly. + + Note that if --auth_iv is used, this will be ignored. + +-* aead_algo: select the AEAD algorithm (default is aes-gcm) ++* aead_algo: select the AEAD algorithm. (Default is aes-gcm.) + +-* aead_op: select the AEAD operation to perform: ENCRYPT or DECRYPT ++* aead_op: select the AEAD operation to perform: ENCRYPT or DECRYPT. + +- (default is ENCRYPT) ++ (Default is ENCRYPT.) + +-* aead_key: set the AEAD key to be used. Bytes has to be separated with ":" ++* aead_key: set the AEAD key to be used. Bytes have to be separated with ":". + + * aead_key_random_size: set the size of the AEAD key, + +@@ -132,13 +132,13 @@ where, + + Note that if --aead_key is used, this will be ignored. + +-* aead_iv: set the AEAD IV to be used. Bytes has to be separated with ":" ++* aead_iv: set the AEAD IV to be used. Bytes have to be separated with ":". + + * aead_iv_random_size: set the size of the AEAD IV, which will be generated randomly. + + Note that if --aead_iv is used, this will be ignored. + +-* aad: set the AAD to be used. Bytes has to be separated with ":" ++* aad: set the AAD to be used. Bytes have to be separated with ":". + + * aad_random_size: set the size of the AAD, which will be generated randomly. + +@@ -151,9 +151,9 @@ where, + * cryptodev_mask: A hexadecimal bitmask of the cryptodevs to be used by the + application. + +- (default is all cryptodevs). ++ (Default is all cryptodevs.) + +-* [no-]mac-updating: Enable or disable MAC addresses updating (enabled by default). ++* [no-]mac-updating: Enable or disable MAC addresses updating. (Enabled by default.) + + + The application requires that crypto devices capable of performing +@@ -165,7 +165,7 @@ To run the application in linux environment with 2 lcores, 2 ports and 2 crypto + + .. code-block:: console + +- $ .//examples/dpdk-l2fwd-crypto -l 0-1 -n 4 --vdev "crypto_aesni_mb0" \ ++ $ .//examples/dpdk-l2fwd-crypto -l 0-1 --vdev "crypto_aesni_mb0" \ + --vdev "crypto_aesni_mb1" -- -p 0x3 --chain CIPHER_HASH \ + --cipher_op ENCRYPT --cipher_algo aes-cbc \ + --cipher_key 00:01:02:03:04:05:06:07:08:09:0a:0b:0c:0d:0e:0f \ +@@ -179,7 +179,7 @@ and the Environment Abstraction Layer (EAL) options. + + * The ``l2fwd-crypto`` sample application requires IPv4 packets for crypto operation. + +- * If multiple Ethernet ports is passed, then equal number of crypto devices are to be passed. ++ * If multiple Ethernet ports are passed, then equal number of crypto devices are to be passed. + + * All crypto devices shall use the same session. + +@@ -187,7 +187,7 @@ Explanation + ----------- + + The L2 forward with Crypto application demonstrates the performance of a crypto operation +-on a packet received on a RX PORT before forwarding it to a TX PORT. ++on a packet received on an RX PORT before forwarding it to a TX PORT. + + The following figure illustrates a sample flow of a packet in the application, + from reception until transmission. +@@ -196,7 +196,7 @@ from reception until transmission. + + .. figure:: img/l2_fwd_encrypt_flow.* + +- Encryption flow Through the L2 Forwarding with Crypto Application ++ Encryption flow through the L2 Forwarding with Crypto Application + + + The following sections provide some explanation of the application. +@@ -206,8 +206,8 @@ Crypto operation specification + + All the packets received in all the ports get transformed by the crypto device/s + (ciphering and/or authentication). +-The crypto operation to be performed on the packet is parsed from the command line +-(go to "Running the Application" section for all the options). ++The crypto operation to be performed on the packet is parsed from the command line. ++(Go to "Running the Application" section for all the options.) + + If no parameter is passed, the default crypto operation is: + +@@ -244,7 +244,7 @@ when running the application. + + The initialize_cryptodevs() function performs the device initialization. + It iterates through the list of the available crypto devices and +-check which ones are capable of performing the operation. ++checks which ones are capable of performing the operation. + Each device has a set of capabilities associated with it, + which are stored in the device info structure, so the function checks if the operation + is within the structure of each device. +@@ -291,7 +291,7 @@ This session is created and is later attached to the crypto operation: + Crypto operation creation + ~~~~~~~~~~~~~~~~~~~~~~~~~ + +-Given N packets received from a RX PORT, N crypto operations are allocated ++Given N packets received from an RX PORT, N crypto operations are allocated + and filled: + + .. literalinclude:: ../../../examples/l2fwd-crypto/main.c +diff --git a/dpdk/doc/guides/sample_app_ug/server_node_efd.rst b/dpdk/doc/guides/sample_app_ug/server_node_efd.rst +index 605eb09a61..c6cbc3def6 100644 +--- a/dpdk/doc/guides/sample_app_ug/server_node_efd.rst ++++ b/dpdk/doc/guides/sample_app_ug/server_node_efd.rst +@@ -191,7 +191,7 @@ flow is not handled by the node. + .. literalinclude:: ../../../examples/server_node_efd/node/node.c + :language: c + :start-after: Packets dequeued from the shared ring. 8< +- :end-before: >8 End of packets dequeueing. ++ :end-before: >8 End of packets dequeuing. + + Finally, note that both processes updates statistics, such as transmitted, received + and dropped packets, which are shown and refreshed by the server app. +diff --git a/dpdk/doc/guides/sample_app_ug/skeleton.rst b/dpdk/doc/guides/sample_app_ug/skeleton.rst +index 6d0de64401..08ddd7aa59 100644 +--- a/dpdk/doc/guides/sample_app_ug/skeleton.rst ++++ b/dpdk/doc/guides/sample_app_ug/skeleton.rst +@@ -54,7 +54,7 @@ function. The value returned is the number of parsed arguments: + .. literalinclude:: ../../../examples/skeleton/basicfwd.c + :language: c + :start-after: Initializion the Environment Abstraction Layer (EAL). 8< +- :end-before: >8 End of initializion the Environment Abstraction Layer (EAL). ++ :end-before: >8 End of initialization the Environment Abstraction Layer (EAL). + :dedent: 1 + + +diff --git a/dpdk/doc/guides/sample_app_ug/vm_power_management.rst b/dpdk/doc/guides/sample_app_ug/vm_power_management.rst +index 7160b6a63a..9ce87956c9 100644 +--- a/dpdk/doc/guides/sample_app_ug/vm_power_management.rst ++++ b/dpdk/doc/guides/sample_app_ug/vm_power_management.rst +@@ -681,7 +681,7 @@ The following is an example JSON string for a power management request. + "resource_id": 10 + }} + +-To query the available frequences of an lcore, use the query_cpu_freq command. ++To query the available frequencies of an lcore, use the query_cpu_freq command. + Where {core_num} is the lcore to query. + Before using this command, please enable responses via the set_query command on the host. + +diff --git a/dpdk/doc/guides/testpmd_app_ug/testpmd_funcs.rst b/dpdk/doc/guides/testpmd_app_ug/testpmd_funcs.rst +index 44228cd7d2..94792d88cc 100644 +--- a/dpdk/doc/guides/testpmd_app_ug/testpmd_funcs.rst ++++ b/dpdk/doc/guides/testpmd_app_ug/testpmd_funcs.rst +@@ -3510,7 +3510,7 @@ Tunnel offload + Indicate tunnel offload rule type + + - ``tunnel_set {tunnel_id}``: mark rule as tunnel offload decap_set type. +-- ``tunnel_match {tunnel_id}``: mark rule as tunel offload match type. ++- ``tunnel_match {tunnel_id}``: mark rule as tunnel offload match type. + + Matching pattern + ^^^^^^^^^^^^^^^^ +diff --git a/dpdk/drivers/baseband/acc100/acc100_pf_enum.h b/dpdk/drivers/baseband/acc100/acc100_pf_enum.h +index a1ee416d26..2fba667627 100644 +--- a/dpdk/drivers/baseband/acc100/acc100_pf_enum.h ++++ b/dpdk/drivers/baseband/acc100/acc100_pf_enum.h +@@ -238,6 +238,24 @@ enum { + HWPfPermonBTotalLatLowBusMon = 0x00BAC504, + HWPfPermonBTotalLatUpperBusMon = 0x00BAC508, + HWPfPermonBTotalReqCntBusMon = 0x00BAC50C, ++ HwPfFabI2MArbCntrlReg = 0x00BB0000, ++ HWPfFabricMode = 0x00BB1000, ++ HwPfFabI2MGrp0DebugReg = 0x00BBF000, ++ HwPfFabI2MGrp1DebugReg = 0x00BBF004, ++ HwPfFabI2MGrp2DebugReg = 0x00BBF008, ++ HwPfFabI2MGrp3DebugReg = 0x00BBF00C, ++ HwPfFabI2MBuf0DebugReg = 0x00BBF010, ++ HwPfFabI2MBuf1DebugReg = 0x00BBF014, ++ HwPfFabI2MBuf2DebugReg = 0x00BBF018, ++ HwPfFabI2MBuf3DebugReg = 0x00BBF01C, ++ HwPfFabM2IBuf0Grp0DebugReg = 0x00BBF020, ++ HwPfFabM2IBuf1Grp0DebugReg = 0x00BBF024, ++ HwPfFabM2IBuf0Grp1DebugReg = 0x00BBF028, ++ HwPfFabM2IBuf1Grp1DebugReg = 0x00BBF02C, ++ HwPfFabM2IBuf0Grp2DebugReg = 0x00BBF030, ++ HwPfFabM2IBuf1Grp2DebugReg = 0x00BBF034, ++ HwPfFabM2IBuf0Grp3DebugReg = 0x00BBF038, ++ HwPfFabM2IBuf1Grp3DebugReg = 0x00BBF03C, + HWPfFecUl5gCntrlReg = 0x00BC0000, + HWPfFecUl5gI2MThreshReg = 0x00BC0004, + HWPfFecUl5gVersionReg = 0x00BC0100, +diff --git a/dpdk/drivers/baseband/acc100/rte_acc100_pmd.c b/dpdk/drivers/baseband/acc100/rte_acc100_pmd.c +index 1c6080f2f8..6cdc6e65f7 100644 +--- a/dpdk/drivers/baseband/acc100/rte_acc100_pmd.c ++++ b/dpdk/drivers/baseband/acc100/rte_acc100_pmd.c +@@ -141,8 +141,8 @@ aqDepth(int qg_idx, struct rte_acc100_conf *acc100_conf) + int acc_enum = accFromQgid(qg_idx, acc100_conf); + qtopFromAcc(&q_top, acc_enum, acc100_conf); + if (unlikely(q_top == NULL)) +- return 0; +- return q_top->aq_depth_log2; ++ return 1; ++ return RTE_MAX(1, q_top->aq_depth_log2); } - #endif -diff --git a/ipsec/ovs-monitor-ipsec.in b/ipsec/ovs-monitor-ipsec.in -index a8b0705d9f..631a8fca80 100755 ---- a/ipsec/ovs-monitor-ipsec.in -+++ b/ipsec/ovs-monitor-ipsec.in -@@ -337,7 +337,14 @@ conn prevent_unencrypted_vxlan - Once strongSwan vici bindings will be distributed with major - Linux distributions this function could be simplified.""" - vlog.info("Refreshing StrongSwan configuration") -- subprocess.call([self.IPSEC, "update"]) -+ proc = subprocess.Popen([self.IPSEC, "update"], -+ stdout=subprocess.PIPE, -+ stderr=subprocess.PIPE) -+ outs, errs = proc.communicate() -+ if proc.returncode != 0: -+ vlog.err("StrongSwan failed to update configuration:\n" -+ "%s \n %s" % (str(outs), str(errs))) -+ - subprocess.call([self.IPSEC, "rereadsecrets"]) - # "ipsec update" command does not remove those tunnels that were - # updated or that disappeared from the ipsec.conf file. So, we have -@@ -708,6 +715,11 @@ conn prevent_unencrypted_vxlan - not re.match(r".*need --listen.*", pout): - break -+ if re.match(r".*[F|f]ailed to initiate connection.*", pout): -+ vlog.err('Failed to initiate connection through' -+ ' Interface %s.\n' % (conn.split('-')[0])) -+ vlog.err(pout) -+ - def _nss_clear_database(self): - """Remove all OVS IPsec related state from the NSS database""" - try: -diff --git a/lib/automake.mk b/lib/automake.mk -index a23cdc4ade..e9a5978e88 100644 ---- a/lib/automake.mk -+++ b/lib/automake.mk -@@ -38,8 +38,6 @@ lib_libopenvswitchavx512_la_CFLAGS = \ - -fPIC \ - $(AM_CFLAGS) - lib_libopenvswitchavx512_la_SOURCES = \ -- lib/cpu.c \ -- lib/cpu.h \ - lib/dpif-netdev-lookup-avx512-gather.c \ - lib/dpif-netdev-extract-avx512.c \ - lib/dpif-netdev-avx512.c -@@ -89,6 +87,8 @@ lib_libopenvswitch_la_SOURCES = \ - lib/conntrack.h \ - lib/coverage.c \ - lib/coverage.h \ -+ lib/cpu.c \ -+ lib/cpu.h \ - lib/crc32c.c \ - lib/crc32c.h \ - lib/csum.c \ -diff --git a/lib/cfm.c b/lib/cfm.c -index cc43e70e31..c3742f3de2 100644 ---- a/lib/cfm.c -+++ b/lib/cfm.c -@@ -416,7 +416,7 @@ cfm_run(struct cfm *cfm) OVS_EXCLUDED(mutex) - ovs_mutex_lock(&mutex); - if (timer_expired(&cfm->fault_timer)) { - long long int interval = cfm_fault_interval(cfm); -- struct remote_mp *rmp, *rmp_next; -+ struct remote_mp *rmp; - enum cfm_fault_reason old_cfm_fault = cfm->fault; - uint64_t old_flap_count = cfm->flap_count; - int old_health = cfm->health; -@@ -475,7 +475,7 @@ cfm_run(struct cfm *cfm) OVS_EXCLUDED(mutex) - cfm->rx_packets = rx_packets; - } + /* Return the AQ depth for a Queue Group Index */ +@@ -1236,6 +1236,8 @@ get_k0(uint16_t n_cb, uint16_t z_c, uint8_t bg, uint8_t rv_index) + return (bg == 1 ? ACC100_K0_3_1 : ACC100_K0_3_2) * z_c; + } + /* LBRM case - includes a division by N */ ++ if (unlikely(z_c == 0)) ++ return 0; + if (rv_index == 1) + return (((bg == 1 ? ACC100_K0_1_1 : ACC100_K0_1_2) * n_cb) + / n) * z_c; +@@ -1460,8 +1462,7 @@ acc100_dma_fill_blk_type_in(struct acc100_dma_req_desc *desc, + next_triplet++; + + while (cb_len > 0) { +- if (next_triplet < ACC100_DMA_MAX_NUM_POINTERS && +- m->next != NULL) { ++ if (next_triplet < ACC100_DMA_MAX_NUM_POINTERS_IN && m->next != NULL) { + + m = m->next; + *seg_total_left = rte_pktmbuf_data_len(m); +@@ -1765,6 +1766,10 @@ acc100_dma_desc_td_fill(struct rte_bbdev_dec_op *op, + + /* Soft output */ + if (check_bit(op->turbo_dec.op_flags, RTE_BBDEV_TURBO_SOFT_OUTPUT)) { ++ if (op->turbo_dec.soft_output.data == 0) { ++ rte_bbdev_log(ERR, "Soft output is not defined"); ++ return -1; ++ } + if (check_bit(op->turbo_dec.op_flags, + RTE_BBDEV_TURBO_EQUALIZER)) + *s_out_length = e; +@@ -4413,7 +4418,7 @@ rte_acc100_configure(const char *dev_name, struct rte_acc100_conf *conf) + { + rte_bbdev_log(INFO, "rte_acc100_configure"); + uint32_t value, address, status; +- int qg_idx, template_idx, vf_idx, acc, i; ++ int qg_idx, template_idx, vf_idx, acc, i, j; + struct rte_bbdev *bbdev = rte_bbdev_get_named_dev(dev_name); + + /* Compile time checks */ +@@ -4433,6 +4438,9 @@ rte_acc100_configure(const char *dev_name, struct rte_acc100_conf *conf) + /* Store configuration */ + rte_memcpy(&d->acc100_conf, conf, sizeof(d->acc100_conf)); + ++ value = acc100_reg_read(d, HwPfPcieGpexBridgeControl); ++ bool firstCfg = (value != ACC100_CFG_PCI_BRIDGE); ++ + /* PCIe Bridge configuration */ + acc100_reg_write(d, HwPfPcieGpexBridgeControl, ACC100_CFG_PCI_BRIDGE); + for (i = 1; i < ACC100_GPEX_AXIMAP_NUM; i++) +@@ -4453,20 +4461,9 @@ rte_acc100_configure(const char *dev_name, struct rte_acc100_conf *conf) + value = 1; + acc100_reg_write(d, address, value); + +- /* DDR Configuration */ +- address = HWPfDdrBcTim6; +- value = acc100_reg_read(d, address); +- value &= 0xFFFFFFFB; /* Bit 2 */ +-#ifdef ACC100_DDR_ECC_ENABLE +- value |= 0x4; +-#endif +- acc100_reg_write(d, address, value); +- address = HWPfDdrPhyDqsCountNum; +-#ifdef ACC100_DDR_ECC_ENABLE +- value = 9; +-#else +- value = 8; +-#endif ++ /* Enable granular dynamic clock gating */ ++ address = HWPfHiClkGateHystReg; ++ value = ACC100_CLOCK_GATING_EN; + acc100_reg_write(d, address, value); + + /* Set default descriptor signature */ +@@ -4484,6 +4481,17 @@ rte_acc100_configure(const char *dev_name, struct rte_acc100_conf *conf) + address = HWPfDmaAxcacheReg; + acc100_reg_write(d, address, value); + ++ /* Adjust PCIe Lane adaptation */ ++ for (i = 0; i < ACC100_QUAD_NUMS; i++) ++ for (j = 0; j < ACC100_LANES_PER_QUAD; j++) ++ acc100_reg_write(d, HwPfPcieLnAdaptctrl + i * ACC100_PCIE_QUAD_OFFSET ++ + j * ACC100_PCIE_LANE_OFFSET, ACC100_ADAPT); ++ ++ /* Enable PCIe live adaptation */ ++ for (i = 0; i < ACC100_QUAD_NUMS; i++) ++ acc100_reg_write(d, HwPfPciePcsEqControl + ++ i * ACC100_PCIE_QUAD_OFFSET, ACC100_PCS_EQ); ++ + /* Default DMA Configuration (Qmgr Enabled) */ + address = HWPfDmaConfig0Reg; + value = 0; +@@ -4502,6 +4510,11 @@ rte_acc100_configure(const char *dev_name, struct rte_acc100_conf *conf) + value = HWPfQmgrEgressQueuesTemplate; + acc100_reg_write(d, address, value); + ++ /* Default Fabric Mode */ ++ address = HWPfFabricMode; ++ value = ACC100_FABRIC_MODE; ++ acc100_reg_write(d, address, value); ++ + /* ===== Qmgr Configuration ===== */ + /* Configuration of the AQueue Depth QMGR_GRP_0_DEPTH_LOG2 for UL */ + int totalQgs = conf->q_ul_4g.num_qgroups + +@@ -4520,22 +4533,17 @@ rte_acc100_configure(const char *dev_name, struct rte_acc100_conf *conf) + } -- HMAP_FOR_EACH_SAFE (rmp, rmp_next, node, &cfm->remote_mps) { -+ HMAP_FOR_EACH_SAFE (rmp, node, &cfm->remote_mps) { - if (!rmp->recv) { - VLOG_INFO("%s: Received no CCM from RMP %"PRIu64" in the last" - " %lldms", cfm->name, rmp->mpid, -diff --git a/lib/classifier.c b/lib/classifier.c -index c4790ee6ba..0a89626cc3 100644 ---- a/lib/classifier.c -+++ b/lib/classifier.c -@@ -916,9 +916,9 @@ free_conjunctive_matches(struct hmap *matches, - struct conjunctive_match *cm_stubs, size_t n_cm_stubs) + /* Template Priority in incremental order */ +- for (template_idx = 0; template_idx < ACC100_NUM_TMPL; +- template_idx++) { +- address = HWPfQmgrGrpTmplateReg0Indx + +- ACC100_BYTES_IN_WORD * (template_idx % 8); ++ for (template_idx = 0; template_idx < ACC100_NUM_TMPL; template_idx++) { ++ address = HWPfQmgrGrpTmplateReg0Indx + ACC100_BYTES_IN_WORD * template_idx; + value = ACC100_TMPL_PRI_0; + acc100_reg_write(d, address, value); +- address = HWPfQmgrGrpTmplateReg1Indx + +- ACC100_BYTES_IN_WORD * (template_idx % 8); ++ address = HWPfQmgrGrpTmplateReg1Indx + ACC100_BYTES_IN_WORD * template_idx; + value = ACC100_TMPL_PRI_1; + acc100_reg_write(d, address, value); +- address = HWPfQmgrGrpTmplateReg2indx + +- ACC100_BYTES_IN_WORD * (template_idx % 8); ++ address = HWPfQmgrGrpTmplateReg2indx + ACC100_BYTES_IN_WORD * template_idx; + value = ACC100_TMPL_PRI_2; + acc100_reg_write(d, address, value); +- address = HWPfQmgrGrpTmplateReg3Indx + +- ACC100_BYTES_IN_WORD * (template_idx % 8); ++ address = HWPfQmgrGrpTmplateReg3Indx + ACC100_BYTES_IN_WORD * template_idx; + value = ACC100_TMPL_PRI_3; + acc100_reg_write(d, address, value); + } +@@ -4586,9 +4594,6 @@ rte_acc100_configure(const char *dev_name, struct rte_acc100_conf *conf) + numEngines++; + } else + acc100_reg_write(d, address, 0); +-#if RTE_ACC100_SINGLE_FEC == 1 +- value = 0; +-#endif + } + printf("Number of 5GUL engines %d\n", numEngines); + /* 4GDL */ +@@ -4603,9 +4608,6 @@ rte_acc100_configure(const char *dev_name, struct rte_acc100_conf *conf) + address = HWPfQmgrGrpTmplateReg4Indx + + ACC100_BYTES_IN_WORD * template_idx; + acc100_reg_write(d, address, value); +-#if RTE_ACC100_SINGLE_FEC == 1 +- value = 0; +-#endif + } + /* 5GDL */ + numQqsAcc += numQgs; +@@ -4619,13 +4621,10 @@ rte_acc100_configure(const char *dev_name, struct rte_acc100_conf *conf) + address = HWPfQmgrGrpTmplateReg4Indx + + ACC100_BYTES_IN_WORD * template_idx; + acc100_reg_write(d, address, value); +-#if RTE_ACC100_SINGLE_FEC == 1 +- value = 0; +-#endif + } + + /* Queue Group Function mapping */ +- int qman_func_id[5] = {0, 2, 1, 3, 4}; ++ int qman_func_id[8] = {0, 2, 1, 3, 4, 0, 0, 0}; + address = HWPfQmgrGrpFunction0; + value = 0; + for (qg_idx = 0; qg_idx < 8; qg_idx++) { +@@ -4656,7 +4655,7 @@ rte_acc100_configure(const char *dev_name, struct rte_acc100_conf *conf) + } + } + +- /* This pointer to ARAM (256kB) is shifted by 2 (4B per register) */ ++ /* This pointer to ARAM (128kB) is shifted by 2 (4B per register) */ + uint32_t aram_address = 0; + for (qg_idx = 0; qg_idx < totalQgs; qg_idx++) { + for (vf_idx = 0; vf_idx < conf->num_vf_bundles; vf_idx++) { +@@ -4681,6 +4680,11 @@ rte_acc100_configure(const char *dev_name, struct rte_acc100_conf *conf) + + /* ==== HI Configuration ==== */ + ++ /* No Info Ring/MSI by default */ ++ acc100_reg_write(d, HWPfHiInfoRingIntWrEnRegPf, 0); ++ acc100_reg_write(d, HWPfHiInfoRingVf2pfLoWrEnReg, 0); ++ acc100_reg_write(d, HWPfHiCfgMsiIntWrEnRegPf, 0xFFFFFFFF); ++ acc100_reg_write(d, HWPfHiCfgMsiVf2pfLoWrEnReg, 0xFFFFFFFF); + /* Prevent Block on Transmit Error */ + address = HWPfHiBlockTransmitOnErrorEn; + value = 0; +@@ -4693,10 +4697,6 @@ rte_acc100_configure(const char *dev_name, struct rte_acc100_conf *conf) + address = HWPfHiPfMode; + value = (conf->pf_mode_en) ? ACC100_PF_VAL : 0; + acc100_reg_write(d, address, value); +- /* Enable Error Detection in HW */ +- address = HWPfDmaErrorDetectionEn; +- value = 0x3D7; +- acc100_reg_write(d, address, value); + + /* QoS overflow init */ + value = 1; +@@ -4706,7 +4706,7 @@ rte_acc100_configure(const char *dev_name, struct rte_acc100_conf *conf) + acc100_reg_write(d, address, value); + + /* HARQ DDR Configuration */ +- unsigned int ddrSizeInMb = 512; /* Fixed to 512 MB per VF for now */ ++ unsigned int ddrSizeInMb = ACC100_HARQ_DDR; + for (vf_idx = 0; vf_idx < conf->num_vf_bundles; vf_idx++) { + address = HWPfDmaVfDdrBaseRw + vf_idx + * 0x10; +@@ -4720,6 +4720,88 @@ rte_acc100_configure(const char *dev_name, struct rte_acc100_conf *conf) + if (numEngines < (ACC100_SIG_UL_5G_LAST + 1)) + poweron_cleanup(bbdev, d, conf); + ++ uint32_t version = 0; ++ for (i = 0; i < 4; i++) ++ version += acc100_reg_read(d, ++ HWPfDdrPhyIdtmFwVersion + 4 * i) << (8 * i); ++ if (version != ACC100_PRQ_DDR_VER) { ++ printf("* Note: Not on DDR PRQ version %8x != %08x\n", ++ version, ACC100_PRQ_DDR_VER); ++ } else if (firstCfg) { ++ /* ---- DDR configuration at boot up --- */ ++ /* Read Clear Ddr training status */ ++ acc100_reg_read(d, HWPfChaDdrStDoneStatus); ++ /* Reset PHY/IDTM/UMMC */ ++ acc100_reg_write(d, HWPfChaDdrWbRstCfg, 3); ++ acc100_reg_write(d, HWPfChaDdrApbRstCfg, 2); ++ acc100_reg_write(d, HWPfChaDdrPhyRstCfg, 2); ++ acc100_reg_write(d, HWPfChaDdrCpuRstCfg, 3); ++ acc100_reg_write(d, HWPfChaDdrSifRstCfg, 2); ++ usleep(ACC100_MS_IN_US); ++ /* Reset WB and APB resets */ ++ acc100_reg_write(d, HWPfChaDdrWbRstCfg, 2); ++ acc100_reg_write(d, HWPfChaDdrApbRstCfg, 3); ++ /* Configure PHY-IDTM */ ++ acc100_reg_write(d, HWPfDdrPhyIdletimeout, 0x3e8); ++ /* IDTM timing registers */ ++ acc100_reg_write(d, HWPfDdrPhyRdLatency, 0x13); ++ acc100_reg_write(d, HWPfDdrPhyRdLatencyDbi, 0x15); ++ acc100_reg_write(d, HWPfDdrPhyWrLatency, 0x10011); ++ /* Configure SDRAM MRS registers */ ++ acc100_reg_write(d, HWPfDdrPhyMr01Dimm, 0x3030b70); ++ acc100_reg_write(d, HWPfDdrPhyMr01DimmDbi, 0x3030b50); ++ acc100_reg_write(d, HWPfDdrPhyMr23Dimm, 0x30); ++ acc100_reg_write(d, HWPfDdrPhyMr67Dimm, 0xc00); ++ acc100_reg_write(d, HWPfDdrPhyMr45Dimm, 0x4000000); ++ /* Configure active lanes */ ++ acc100_reg_write(d, HWPfDdrPhyDqsCountMax, 0x9); ++ acc100_reg_write(d, HWPfDdrPhyDqsCountNum, 0x9); ++ /* Configure WR/RD leveling timing registers */ ++ acc100_reg_write(d, HWPfDdrPhyWrlvlWwRdlvlRr, 0x101212); ++ /* Configure what trainings to execute */ ++ acc100_reg_write(d, HWPfDdrPhyTrngType, 0x2d3c); ++ /* Releasing PHY reset */ ++ acc100_reg_write(d, HWPfChaDdrPhyRstCfg, 3); ++ /* Configure Memory Controller registers */ ++ acc100_reg_write(d, HWPfDdrMemInitPhyTrng0, 0x3); ++ acc100_reg_write(d, HWPfDdrBcDram, 0x3c232003); ++ acc100_reg_write(d, HWPfDdrBcAddrMap, 0x31); ++ /* Configure UMMC BC timing registers */ ++ acc100_reg_write(d, HWPfDdrBcRef, 0xa22); ++ acc100_reg_write(d, HWPfDdrBcTim0, 0x4050501); ++ acc100_reg_write(d, HWPfDdrBcTim1, 0xf0b0476); ++ acc100_reg_write(d, HWPfDdrBcTim2, 0x103); ++ acc100_reg_write(d, HWPfDdrBcTim3, 0x144050a1); ++ acc100_reg_write(d, HWPfDdrBcTim4, 0x23300); ++ acc100_reg_write(d, HWPfDdrBcTim5, 0x4230276); ++ acc100_reg_write(d, HWPfDdrBcTim6, 0x857914); ++ acc100_reg_write(d, HWPfDdrBcTim7, 0x79100232); ++ acc100_reg_write(d, HWPfDdrBcTim8, 0x100007ce); ++ acc100_reg_write(d, HWPfDdrBcTim9, 0x50020); ++ acc100_reg_write(d, HWPfDdrBcTim10, 0x40ee); ++ /* Configure UMMC DFI timing registers */ ++ acc100_reg_write(d, HWPfDdrDfiInit, 0x5000); ++ acc100_reg_write(d, HWPfDdrDfiTim0, 0x15030006); ++ acc100_reg_write(d, HWPfDdrDfiTim1, 0x11305); ++ acc100_reg_write(d, HWPfDdrDfiPhyUpdEn, 0x1); ++ acc100_reg_write(d, HWPfDdrUmmcIntEn, 0x1f); ++ /* Release IDTM CPU out of reset */ ++ acc100_reg_write(d, HWPfChaDdrCpuRstCfg, 0x2); ++ /* Wait PHY-IDTM to finish static training */ ++ for (i = 0; i < ACC100_DDR_TRAINING_MAX; i++) { ++ usleep(ACC100_MS_IN_US); ++ value = acc100_reg_read(d, ++ HWPfChaDdrStDoneStatus); ++ if (value & 1) ++ break; ++ } ++ printf("DDR Training completed in %d ms", i); ++ /* Enable Memory Controller */ ++ acc100_reg_write(d, HWPfDdrUmmcCtrl, 0x401); ++ /* Release AXI interface reset */ ++ acc100_reg_write(d, HWPfChaDdrSifRstCfg, 3); ++ } ++ + rte_bbdev_log_debug("PF Tip configuration complete for %s", dev_name); + return 0; + } +diff --git a/dpdk/drivers/baseband/acc100/rte_acc100_pmd.h b/dpdk/drivers/baseband/acc100/rte_acc100_pmd.h +index 03ed0b3e1a..071b37cf9d 100644 +--- a/dpdk/drivers/baseband/acc100/rte_acc100_pmd.h ++++ b/dpdk/drivers/baseband/acc100/rte_acc100_pmd.h +@@ -31,11 +31,6 @@ + #define RTE_ACC100_PF_DEVICE_ID (0x0d5c) + #define RTE_ACC100_VF_DEVICE_ID (0x0d5d) + +-/* Define as 1 to use only a single FEC engine */ +-#ifndef RTE_ACC100_SINGLE_FEC +-#define RTE_ACC100_SINGLE_FEC 0 +-#endif +- + /* Values used in filling in descriptors */ + #define ACC100_DMA_DESC_TYPE 2 + #define ACC100_DMA_CODE_BLK_MODE 0 +@@ -113,6 +108,7 @@ + #define ACC100_SW_RING_MEM_ALLOC_ATTEMPTS 5 + #define ACC100_MAX_QUEUE_DEPTH 1024 + #define ACC100_DMA_MAX_NUM_POINTERS 14 ++#define ACC100_DMA_MAX_NUM_POINTERS_IN 7 + #define ACC100_DMA_DESC_PADDING 8 + #define ACC100_FCW_PADDING 12 + #define ACC100_DESC_FCW_OFFSET 192 +@@ -152,6 +148,12 @@ + #define ACC100_CFG_QMGR_HI_P 0x0F0F + #define ACC100_CFG_PCI_AXI 0xC003 + #define ACC100_CFG_PCI_BRIDGE 0x40006033 ++#define ACC100_QUAD_NUMS 4 ++#define ACC100_LANES_PER_QUAD 4 ++#define ACC100_PCIE_LANE_OFFSET 0x200 ++#define ACC100_PCIE_QUAD_OFFSET 0x2000 ++#define ACC100_PCS_EQ 0x6007 ++#define ACC100_ADAPT 0x8400 + #define ACC100_ENGINE_OFFSET 0x1000 + #define ACC100_RESET_HI 0x20100 + #define ACC100_RESET_LO 0x20000 +@@ -159,6 +161,15 @@ + #define ACC100_ENGINES_MAX 9 + #define ACC100_LONG_WAIT 1000 + #define ACC100_GPEX_AXIMAP_NUM 17 ++#define ACC100_CLOCK_GATING_EN 0x30000 ++#define ACC100_FABRIC_MODE 0xB ++/* DDR Size per VF - 512MB by default ++ * Can be increased up to 4 GB with single PF/VF ++ */ ++#define ACC100_HARQ_DDR (512 * 1) ++#define ACC100_PRQ_DDR_VER 0x10092020 ++#define ACC100_MS_IN_US (1000) ++#define ACC100_DDR_TRAINING_MAX (5000) + + /* ACC100 DMA Descriptor triplet */ + struct acc100_dma_triplet { +diff --git a/dpdk/drivers/baseband/fpga_lte_fec/fpga_lte_fec.c b/dpdk/drivers/baseband/fpga_lte_fec/fpga_lte_fec.c +index 92decc3e05..21d35292a3 100644 +--- a/dpdk/drivers/baseband/fpga_lte_fec/fpga_lte_fec.c ++++ b/dpdk/drivers/baseband/fpga_lte_fec/fpga_lte_fec.c +@@ -2097,7 +2097,7 @@ dequeue_enc_one_op_cb(struct fpga_queue *q, struct rte_bbdev_enc_op **op, + rte_bbdev_log_debug("DMA response desc %p", desc); + + *op = desc->enc_req.op_addr; +- /* Check the decriptor error field, return 1 on error */ ++ /* Check the descriptor error field, return 1 on error */ + desc_error = check_desc_error(desc->enc_req.error); + (*op)->status = desc_error << RTE_BBDEV_DATA_ERROR; + +@@ -2139,7 +2139,7 @@ dequeue_enc_one_op_tb(struct fpga_queue *q, struct rte_bbdev_enc_op **op, + for (cb_idx = 0; cb_idx < cbs_in_op; ++cb_idx) { + desc = q->ring_addr + ((q->head_free_desc + desc_offset + + cb_idx) & q->sw_ring_wrap_mask); +- /* Check the decriptor error field, return 1 on error */ ++ /* Check the descriptor error field, return 1 on error */ + desc_error = check_desc_error(desc->enc_req.error); + status |= desc_error << RTE_BBDEV_DATA_ERROR; + rte_bbdev_log_debug("DMA response desc %p", desc); +@@ -2177,7 +2177,7 @@ dequeue_dec_one_op_cb(struct fpga_queue *q, struct rte_bbdev_dec_op **op, + (*op)->turbo_dec.iter_count = (desc->dec_req.iter + 2) >> 1; + /* crc_pass = 0 when decoder fails */ + (*op)->status = !(desc->dec_req.crc_pass) << RTE_BBDEV_CRC_ERROR; +- /* Check the decriptor error field, return 1 on error */ ++ /* Check the descriptor error field, return 1 on error */ + desc_error = check_desc_error(desc->enc_req.error); + (*op)->status |= desc_error << RTE_BBDEV_DATA_ERROR; + return 1; +@@ -2221,7 +2221,7 @@ dequeue_dec_one_op_tb(struct fpga_queue *q, struct rte_bbdev_dec_op **op, + iter_count = RTE_MAX(iter_count, (uint8_t) desc->dec_req.iter); + /* crc_pass = 0 when decoder fails, one fails all */ + status |= !(desc->dec_req.crc_pass) << RTE_BBDEV_CRC_ERROR; +- /* Check the decriptor error field, return 1 on error */ ++ /* Check the descriptor error field, return 1 on error */ + desc_error = check_desc_error(desc->enc_req.error); + status |= desc_error << RTE_BBDEV_DATA_ERROR; + rte_bbdev_log_debug("DMA response desc %p", desc); +diff --git a/dpdk/drivers/baseband/null/bbdev_null.c b/dpdk/drivers/baseband/null/bbdev_null.c +index 753d920e18..08cff582b9 100644 +--- a/dpdk/drivers/baseband/null/bbdev_null.c ++++ b/dpdk/drivers/baseband/null/bbdev_null.c +@@ -31,7 +31,7 @@ struct bbdev_null_params { + uint16_t queues_num; /*< Null BBDEV queues number */ + }; + +-/* Accecptable params for null BBDEV devices */ ++/* Acceptable params for null BBDEV devices */ + #define BBDEV_NULL_MAX_NB_QUEUES_ARG "max_nb_queues" + #define BBDEV_NULL_SOCKET_ID_ARG "socket_id" + +diff --git a/dpdk/drivers/baseband/turbo_sw/bbdev_turbo_software.c b/dpdk/drivers/baseband/turbo_sw/bbdev_turbo_software.c +index b234bb751a..c6b1eb8679 100644 +--- a/dpdk/drivers/baseband/turbo_sw/bbdev_turbo_software.c ++++ b/dpdk/drivers/baseband/turbo_sw/bbdev_turbo_software.c +@@ -61,7 +61,7 @@ struct turbo_sw_params { + uint16_t queues_num; /*< Turbo SW device queues number */ + }; + +-/* Accecptable params for Turbo SW devices */ ++/* Acceptable params for Turbo SW devices */ + #define TURBO_SW_MAX_NB_QUEUES_ARG "max_nb_queues" + #define TURBO_SW_SOCKET_ID_ARG "socket_id" + +diff --git a/dpdk/drivers/bus/auxiliary/version.map b/dpdk/drivers/bus/auxiliary/version.map +index a52260657c..dc993e84ff 100644 +--- a/dpdk/drivers/bus/auxiliary/version.map ++++ b/dpdk/drivers/bus/auxiliary/version.map +@@ -4,4 +4,6 @@ EXPERIMENTAL { + # added in 21.08 + rte_auxiliary_register; + rte_auxiliary_unregister; ++ ++ local: *; + }; +diff --git a/dpdk/drivers/bus/dpaa/dpaa_bus.c b/dpdk/drivers/bus/dpaa/dpaa_bus.c +index 737ac8d8c5..5546a9cb8d 100644 +--- a/dpdk/drivers/bus/dpaa/dpaa_bus.c ++++ b/dpdk/drivers/bus/dpaa/dpaa_bus.c +@@ -70,7 +70,7 @@ compare_dpaa_devices(struct rte_dpaa_device *dev1, { - if (hmap_count(matches) > n_cm_stubs) { -- struct conjunctive_match *cm, *next; -+ struct conjunctive_match *cm; + int comp = 0; + +- /* Segragating ETH from SEC devices */ ++ /* Segregating ETH from SEC devices */ + if (dev1->device_type > dev2->device_type) + comp = 1; + else if (dev1->device_type < dev2->device_type) +diff --git a/dpdk/drivers/bus/dpaa/include/fsl_qman.h b/dpdk/drivers/bus/dpaa/include/fsl_qman.h +index 7ef2f3b2e3..9b63e559bc 100644 +--- a/dpdk/drivers/bus/dpaa/include/fsl_qman.h ++++ b/dpdk/drivers/bus/dpaa/include/fsl_qman.h +@@ -1353,7 +1353,7 @@ __rte_internal + int qman_irqsource_add(u32 bits); + + /** +- * qman_fq_portal_irqsource_add - samilar to qman_irqsource_add, but it ++ * qman_fq_portal_irqsource_add - similar to qman_irqsource_add, but it + * takes portal (fq specific) as input rather than using the thread affined + * portal. + */ +@@ -1416,7 +1416,7 @@ __rte_internal + struct qm_dqrr_entry *qman_dequeue(struct qman_fq *fq); + + /** +- * qman_dqrr_consume - Consume the DQRR entriy after volatile dequeue ++ * qman_dqrr_consume - Consume the DQRR entry after volatile dequeue + * @fq: Frame Queue on which the volatile dequeue command is issued + * @dq: DQRR entry to consume. This is the one which is provided by the + * 'qbman_dequeue' command. +@@ -2017,7 +2017,7 @@ int qman_create_cgr_to_dcp(struct qman_cgr *cgr, u32 flags, u16 dcp_portal, + * @cgr: the 'cgr' object to deregister + * + * "Unplugs" this CGR object from the portal affine to the cpu on which this API +- * is executed. This must be excuted on the same affine portal on which it was ++ * is executed. This must be executed on the same affine portal on which it was + * created. + */ + __rte_internal +diff --git a/dpdk/drivers/bus/dpaa/include/fsl_usd.h b/dpdk/drivers/bus/dpaa/include/fsl_usd.h +index dcf35e4adb..97279421ad 100644 +--- a/dpdk/drivers/bus/dpaa/include/fsl_usd.h ++++ b/dpdk/drivers/bus/dpaa/include/fsl_usd.h +@@ -40,7 +40,7 @@ struct dpaa_raw_portal { + /* Specifies the stash request queue this portal should use */ + uint8_t sdest; + +- /* Specifes a specific portal index to map or QBMAN_ANY_PORTAL_IDX ++ /* Specifies a specific portal index to map or QBMAN_ANY_PORTAL_IDX + * for don't care. The portal index will be populated by the + * driver when the ioctl() successfully completes. + */ +diff --git a/dpdk/drivers/bus/dpaa/include/process.h b/dpdk/drivers/bus/dpaa/include/process.h +index a922988607..48d6b5693f 100644 +--- a/dpdk/drivers/bus/dpaa/include/process.h ++++ b/dpdk/drivers/bus/dpaa/include/process.h +@@ -49,7 +49,7 @@ struct dpaa_portal_map { + struct dpaa_ioctl_portal_map { + /* Input parameter, is a qman or bman portal required. */ + enum dpaa_portal_type type; +- /* Specifes a specific portal index to map or 0xffffffff ++ /* Specifies a specific portal index to map or 0xffffffff + * for don't care. + */ + uint32_t index; +diff --git a/dpdk/drivers/bus/fslmc/fslmc_bus.c b/dpdk/drivers/bus/fslmc/fslmc_bus.c +index a0ef24cdc8..53fd75539e 100644 +--- a/dpdk/drivers/bus/fslmc/fslmc_bus.c ++++ b/dpdk/drivers/bus/fslmc/fslmc_bus.c +@@ -539,7 +539,7 @@ rte_fslmc_driver_unregister(struct rte_dpaa2_driver *driver) + + fslmc_bus = driver->fslmc_bus; + +- /* Cleanup the PA->VA Translation table; From whereever this function ++ /* Cleanup the PA->VA Translation table; From wherever this function + * is called from. + */ + if (rte_eal_iova_mode() == RTE_IOVA_PA) +diff --git a/dpdk/drivers/bus/fslmc/fslmc_vfio.c b/dpdk/drivers/bus/fslmc/fslmc_vfio.c +index b4704eeae4..abe1cab2ee 100644 +--- a/dpdk/drivers/bus/fslmc/fslmc_vfio.c ++++ b/dpdk/drivers/bus/fslmc/fslmc_vfio.c +@@ -979,6 +979,7 @@ fslmc_vfio_setup_group(void) + { + int groupid; + int ret; ++ int vfio_container_fd; + struct vfio_group_status status = { .argsz = sizeof(status) }; -- HMAP_FOR_EACH_SAFE (cm, next, hmap_node, matches) { -+ HMAP_FOR_EACH_SAFE (cm, hmap_node, matches) { - if (!(cm >= cm_stubs && cm < &cm_stubs[n_cm_stubs])) { - free(cm); - } -diff --git a/lib/cmap.h b/lib/cmap.h -index c502d23112..72e2ec5f71 100644 ---- a/lib/cmap.h -+++ b/lib/cmap.h -@@ -108,6 +108,8 @@ size_t cmap_replace(struct cmap *, struct cmap_node *old_node, + /* if already done once */ +@@ -997,8 +998,15 @@ fslmc_vfio_setup_group(void) + return 0; + } + ++ ret = rte_vfio_container_create(); ++ if (ret < 0) { ++ DPAA2_BUS_ERR("Failed to open VFIO container"); ++ return ret; ++ } ++ vfio_container_fd = ret; ++ + /* Get the actual group fd */ +- ret = rte_vfio_get_group_fd(groupid); ++ ret = rte_vfio_container_group_bind(vfio_container_fd, groupid); + if (ret < 0) + return ret; + vfio_group.fd = ret; +diff --git a/dpdk/drivers/bus/fslmc/portal/dpaa2_hw_dpio.c b/dpdk/drivers/bus/fslmc/portal/dpaa2_hw_dpio.c +index 2210a0fa4a..52605ea2c3 100644 +--- a/dpdk/drivers/bus/fslmc/portal/dpaa2_hw_dpio.c ++++ b/dpdk/drivers/bus/fslmc/portal/dpaa2_hw_dpio.c +@@ -178,7 +178,7 @@ static int dpaa2_dpio_intr_init(struct dpaa2_dpio_dev *dpio_dev) + dpio_epoll_fd = epoll_create(1); + ret = rte_dpaa2_intr_enable(dpio_dev->intr_handle, 0); + if (ret) { +- DPAA2_BUS_ERR("Interrupt registeration failed"); ++ DPAA2_BUS_ERR("Interrupt registration failed"); + return -1; + } + +diff --git a/dpdk/drivers/bus/fslmc/portal/dpaa2_hw_pvt.h b/dpdk/drivers/bus/fslmc/portal/dpaa2_hw_pvt.h +index b1bba1ac36..957fc62d4c 100644 +--- a/dpdk/drivers/bus/fslmc/portal/dpaa2_hw_pvt.h ++++ b/dpdk/drivers/bus/fslmc/portal/dpaa2_hw_pvt.h +@@ -156,7 +156,7 @@ struct dpaa2_queue { + struct rte_cryptodev_data *crypto_data; + }; + uint32_t fqid; /*!< Unique ID of this queue */ +- uint16_t flow_id; /*!< To be used by DPAA2 frmework */ ++ uint16_t flow_id; /*!< To be used by DPAA2 framework */ + uint8_t tc_index; /*!< traffic class identifier */ + uint8_t cgid; /*! < Congestion Group id for this queue */ + uint64_t rx_pkts; +diff --git a/dpdk/drivers/bus/fslmc/qbman/include/fsl_qbman_portal.h b/dpdk/drivers/bus/fslmc/qbman/include/fsl_qbman_portal.h +index eb68c9cab5..5375ea386d 100644 +--- a/dpdk/drivers/bus/fslmc/qbman/include/fsl_qbman_portal.h ++++ b/dpdk/drivers/bus/fslmc/qbman/include/fsl_qbman_portal.h +@@ -510,7 +510,7 @@ int qbman_result_has_new_result(struct qbman_swp *s, + struct qbman_result *dq); + + /** +- * qbman_check_command_complete() - Check if the previous issued dq commnd ++ * qbman_check_command_complete() - Check if the previous issued dq command + * is completed and results are available in memory. + * @s: the software portal object. + * @dq: the dequeue result read from the memory. +@@ -687,7 +687,7 @@ uint16_t qbman_result_DQ_seqnum(const struct qbman_result *dq); + + /** + * qbman_result_DQ_odpid() - Get the seqnum field in dequeue response +- * odpid is valid only if ODPVAILD flag is TRUE. ++ * odpid is valid only if ODPVALID flag is TRUE. + * @dq: the dequeue result. * - * CMAP and HASH are evaluated only once. NODE is evaluated many times. + * Return odpid. +@@ -743,7 +743,7 @@ const struct qbman_fd *qbman_result_DQ_fd(const struct qbman_result *dq); + * qbman_result_SCN_state() - Get the state field in State-change notification + * @scn: the state change notification. * -+ * After a normal exit of the loop (not through a "break;" statement) NODE is -+ * NULL. +- * Return the state in the notifiation. ++ * Return the state in the notification. + */ + __rte_internal + uint8_t qbman_result_SCN_state(const struct qbman_result *scn); +@@ -825,7 +825,7 @@ uint64_t qbman_result_bpscn_ctx(const struct qbman_result *scn); + + /* Parsing CGCU */ + /** +- * qbman_result_cgcu_cgid() - Check CGCU resouce id, i.e. cgid ++ * qbman_result_cgcu_cgid() - Check CGCU resource id, i.e. cgid + * @scn: the state change notification. * - * Thread-safety - * ============= -@@ -128,15 +130,15 @@ size_t cmap_replace(struct cmap *, struct cmap_node *old_node, - * CMAP_FOR_EACH_WITH_HASH_PROTECTED may only be used if CMAP is guaranteed not - * to change during iteration. It may be very slightly faster. + * Return the CGCU resource id. +@@ -903,14 +903,14 @@ void qbman_eq_desc_clear(struct qbman_eq_desc *d); + __rte_internal + void qbman_eq_desc_set_no_orp(struct qbman_eq_desc *d, int respond_success); + /** +- * qbman_eq_desc_set_orp() - Set order-resotration in the enqueue descriptor ++ * qbman_eq_desc_set_orp() - Set order-restoration in the enqueue descriptor + * @d: the enqueue descriptor. + * @response_success: 1 = enqueue with response always; 0 = enqueue with + * rejections returned on a FQ. + * @opr_id: the order point record id. + * @seqnum: the order restoration sequence number. +- * @incomplete: indiates whether this is the last fragments using the same +- * sequeue number. ++ * @incomplete: indicates whether this is the last fragments using the same ++ * sequence number. */ --#define CMAP_NODE_FOR_EACH(NODE, MEMBER, CMAP_NODE) \ -- for (INIT_CONTAINER(NODE, CMAP_NODE, MEMBER); \ -- (NODE) != OBJECT_CONTAINING(NULL, NODE, MEMBER); \ -- ASSIGN_CONTAINER(NODE, cmap_node_next(&(NODE)->MEMBER), MEMBER)) --#define CMAP_NODE_FOR_EACH_PROTECTED(NODE, MEMBER, CMAP_NODE) \ -- for (INIT_CONTAINER(NODE, CMAP_NODE, MEMBER); \ -- (NODE) != OBJECT_CONTAINING(NULL, NODE, MEMBER); \ -- ASSIGN_CONTAINER(NODE, cmap_node_next_protected(&(NODE)->MEMBER), \ -- MEMBER)) -+#define CMAP_NODE_FOR_EACH(NODE, MEMBER, CMAP_NODE) \ -+ for (INIT_MULTIVAR(NODE, MEMBER, CMAP_NODE, struct cmap_node); \ -+ CONDITION_MULTIVAR(NODE, MEMBER, ITER_VAR(NODE) != NULL); \ -+ UPDATE_MULTIVAR(NODE, cmap_node_next(ITER_VAR(NODE)))) -+#define CMAP_NODE_FOR_EACH_PROTECTED(NODE, MEMBER, CMAP_NODE) \ -+ for (INIT_MULTIVAR(NODE, MEMBER, CMAP_NODE, struct cmap_node); \ -+ CONDITION_MULTIVAR(NODE, MEMBER, ITER_VAR(NODE) != NULL); \ -+ UPDATE_MULTIVAR(NODE, cmap_node_next_protected(ITER_VAR(NODE)))) -+ - #define CMAP_FOR_EACH_WITH_HASH(NODE, MEMBER, HASH, CMAP) \ - CMAP_NODE_FOR_EACH(NODE, MEMBER, cmap_find(CMAP, HASH)) - #define CMAP_FOR_EACH_WITH_HASH_PROTECTED(NODE, MEMBER, HASH, CMAP) \ -@@ -223,7 +225,7 @@ unsigned long cmap_find_batch(const struct cmap *cmap, unsigned long map, - ? (INIT_CONTAINER(NODE, (CURSOR)->node, MEMBER), \ - cmap_cursor_advance(CURSOR), \ - true) \ -- : false) -+ : (NODE = NULL, false)) + __rte_internal + void qbman_eq_desc_set_orp(struct qbman_eq_desc *d, int respond_success, +@@ -1052,10 +1052,10 @@ __rte_internal + uint8_t qbman_result_eqresp_rspid(struct qbman_result *eqresp); + + /** +- * qbman_result_eqresp_rc() - determines if enqueue command is sucessful. ++ * qbman_result_eqresp_rc() - determines if enqueue command is successful. + * @eqresp: enqueue response. + * +- * Return 0 when command is sucessful. ++ * Return 0 when command is successful. + */ + __rte_internal + uint8_t qbman_result_eqresp_rc(struct qbman_result *eqresp); +@@ -1250,7 +1250,7 @@ int qbman_swp_fq_force(struct qbman_swp *s, uint32_t fqid); + /** + * These functions change the FQ flow-control stuff between XON/XOFF. (The + * default is XON.) This setting doesn't affect enqueues to the FQ, just +- * dequeues. XOFF FQs will remain in the tenatively-scheduled state, even when ++ * dequeues. XOFF FQs will remain in the tentatively-scheduled state, even when + * non-empty, meaning they won't be selected for scheduled dequeuing. If a FQ is + * changed to XOFF after it had already become truly-scheduled to a channel, and + * a pull dequeue of that channel occurs that selects that FQ for dequeuing, +diff --git a/dpdk/drivers/bus/ifpga/ifpga_bus.c b/dpdk/drivers/bus/ifpga/ifpga_bus.c +index cbc6809284..c5c8bbd572 100644 +--- a/dpdk/drivers/bus/ifpga/ifpga_bus.c ++++ b/dpdk/drivers/bus/ifpga/ifpga_bus.c +@@ -64,8 +64,7 @@ ifpga_find_afu_dev(const struct rte_rawdev *rdev, + struct rte_afu_device *afu_dev = NULL; + + TAILQ_FOREACH(afu_dev, &ifpga_afu_dev_list, next) { +- if (afu_dev && +- afu_dev->rawdev == rdev && ++ if (afu_dev->rawdev == rdev && + !ifpga_afu_id_cmp(&afu_dev->id, afu_id)) + return afu_dev; + } +@@ -78,8 +77,7 @@ rte_ifpga_find_afu_by_name(const char *name) + struct rte_afu_device *afu_dev = NULL; + + TAILQ_FOREACH(afu_dev, &ifpga_afu_dev_list, next) { +- if (afu_dev && +- !strcmp(afu_dev->device.name, name)) ++ if (!strcmp(afu_dev->device.name, name)) + return afu_dev; + } + return NULL; +diff --git a/dpdk/drivers/bus/pci/linux/pci_vfio.c b/dpdk/drivers/bus/pci/linux/pci_vfio.c +index 1a5e7c2d2a..cd0d0b1670 100644 +--- a/dpdk/drivers/bus/pci/linux/pci_vfio.c ++++ b/dpdk/drivers/bus/pci/linux/pci_vfio.c +@@ -815,7 +815,7 @@ pci_vfio_map_resource_primary(struct rte_pci_device *dev) + continue; + } + +- /* skip non-mmapable BARs */ ++ /* skip non-mmappable BARs */ + if ((reg->flags & VFIO_REGION_INFO_FLAG_MMAP) == 0) { + free(reg); + continue; +diff --git a/dpdk/drivers/bus/pci/pci_common.c b/dpdk/drivers/bus/pci/pci_common.c +index 4a3a87f24f..def372b67e 100644 +--- a/dpdk/drivers/bus/pci/pci_common.c ++++ b/dpdk/drivers/bus/pci/pci_common.c +@@ -247,9 +247,12 @@ rte_pci_probe_one_driver(struct rte_pci_driver *dr, + return -ENOMEM; + } + ++ dev->driver = dr; ++ + if (dr->drv_flags & RTE_PCI_DRV_NEED_MAPPING) { + ret = rte_pci_map_device(dev); + if (ret != 0) { ++ dev->driver = NULL; + rte_intr_instance_free(dev->vfio_req_intr_handle); + dev->vfio_req_intr_handle = NULL; + rte_intr_instance_free(dev->intr_handle); +@@ -257,8 +260,6 @@ rte_pci_probe_one_driver(struct rte_pci_driver *dr, + return ret; + } + } +- +- dev->driver = dr; + } - #define CMAP_CURSOR_FOR_EACH(NODE, MEMBER, CURSOR, CMAP) \ - for (*(CURSOR) = cmap_cursor_start(CMAP); \ -diff --git a/lib/conntrack.c b/lib/conntrack.c -index 33a1a92953..0103fb5396 100644 ---- a/lib/conntrack.c -+++ b/lib/conntrack.c -@@ -1526,14 +1526,14 @@ set_label(struct dp_packet *pkt, struct conn *conn, - static long long - ct_sweep(struct conntrack *ct, long long now, size_t limit) + RTE_LOG(INFO, EAL, "Probe PCI driver: %s (%x:%x) device: "PCI_PRI_FMT" (socket %i)\n", +diff --git a/dpdk/drivers/bus/vdev/rte_bus_vdev.h b/dpdk/drivers/bus/vdev/rte_bus_vdev.h +index 2856799953..5af6be009f 100644 +--- a/dpdk/drivers/bus/vdev/rte_bus_vdev.h ++++ b/dpdk/drivers/bus/vdev/rte_bus_vdev.h +@@ -197,7 +197,7 @@ rte_vdev_remove_custom_scan(rte_vdev_scan_callback callback, void *user_arg); + int rte_vdev_init(const char *name, const char *args); + + /** +- * Uninitalize a driver specified by name. ++ * Uninitialize a driver specified by name. + * + * @param name + * The pointer to a driver name to be uninitialized. +diff --git a/dpdk/drivers/bus/vmbus/private.h b/dpdk/drivers/bus/vmbus/private.h +index 1bca147e12..658303bc27 100644 +--- a/dpdk/drivers/bus/vmbus/private.h ++++ b/dpdk/drivers/bus/vmbus/private.h +@@ -74,6 +74,8 @@ struct vmbus_channel { + uint16_t relid; + uint16_t subchannel_id; + uint8_t monitor_id; ++ ++ struct vmbus_mon_page *monitor_page; + }; + + #define VMBUS_MAX_CHANNELS 64 +diff --git a/dpdk/drivers/bus/vmbus/vmbus_channel.c b/dpdk/drivers/bus/vmbus/vmbus_channel.c +index 119b9b367e..9bd01679c3 100644 +--- a/dpdk/drivers/bus/vmbus/vmbus_channel.c ++++ b/dpdk/drivers/bus/vmbus/vmbus_channel.c +@@ -27,7 +27,7 @@ vmbus_sync_set_bit(volatile uint32_t *addr, uint32_t mask) + } + + static inline void +-vmbus_set_monitor(const struct rte_vmbus_device *dev, uint32_t monitor_id) ++vmbus_set_monitor(const struct vmbus_channel *channel, uint32_t monitor_id) { -- struct conn *conn, *next; -+ struct conn *conn; - long long min_expiration = LLONG_MAX; - size_t count = 0; + uint32_t *monitor_addr, monitor_mask; + unsigned int trigger_index; +@@ -35,15 +35,14 @@ vmbus_set_monitor(const struct rte_vmbus_device *dev, uint32_t monitor_id) + trigger_index = monitor_id / HV_MON_TRIG_LEN; + monitor_mask = 1u << (monitor_id % HV_MON_TRIG_LEN); + +- monitor_addr = &dev->monitor_page->trigs[trigger_index].pending; ++ monitor_addr = &channel->monitor_page->trigs[trigger_index].pending; + vmbus_sync_set_bit(monitor_addr, monitor_mask); + } - ovs_mutex_lock(&ct->ct_lock); + static void +-vmbus_set_event(const struct rte_vmbus_device *dev, +- const struct vmbus_channel *chan) ++vmbus_set_event(const struct vmbus_channel *chan) + { +- vmbus_set_monitor(dev, chan->monitor_id); ++ vmbus_set_monitor(chan, chan->monitor_id); + } - for (unsigned i = 0; i < N_CT_TM; i++) { -- LIST_FOR_EACH_SAFE (conn, next, exp_node, &ct->exp_lists[i]) { -+ LIST_FOR_EACH_SAFE (conn, exp_node, &ct->exp_lists[i]) { - ovs_mutex_lock(&conn->lock); - if (now < conn->expiration || count >= limit) { - min_expiration = MIN(min_expiration, conn->expiration); -@@ -2242,7 +2242,7 @@ nat_range_hash(const struct conn *conn, uint32_t basis, - hash = ct_addr_hash_add(hash, &nat_info->min_addr); - hash = ct_addr_hash_add(hash, &nat_info->max_addr); - hash = hash_add(hash, -- (nat_info->max_port << 16) -+ ((uint32_t) nat_info->max_port << 16) - | nat_info->min_port); - hash = ct_endpoint_hash_add(hash, &conn->key.src); - hash = ct_endpoint_hash_add(hash, &conn->key.dst); -@@ -2265,8 +2265,16 @@ set_sport_range(const struct nat_action_info_t *ni, const struct conn_key *k, - if (((ni->nat_action & NAT_ACTION_SNAT_ALL) == NAT_ACTION_SRC) || - ((ni->nat_action & NAT_ACTION_DST))) { - *curr = ntohs(k->src.port); -- *min = MIN_NAT_EPHEMERAL_PORT; -- *max = MAX_NAT_EPHEMERAL_PORT; -+ if (*curr < 512) { -+ *min = 1; -+ *max = 511; -+ } else if (*curr < 1024) { -+ *min = 600; -+ *max = 1023; -+ } else { -+ *min = MIN_NAT_EPHEMERAL_PORT; -+ *max = MAX_NAT_EPHEMERAL_PORT; -+ } - } else { - *min = ni->min_port; - *max = ni->max_port; -@@ -2389,6 +2397,26 @@ next_addr_in_range_guarded(union ct_addr *curr, union ct_addr *min, - return exhausted; + /* +@@ -81,7 +80,6 @@ rte_vmbus_set_latency(const struct rte_vmbus_device *dev, + void + rte_vmbus_chan_signal_tx(const struct vmbus_channel *chan) + { +- const struct rte_vmbus_device *dev = chan->device; + const struct vmbus_br *tbr = &chan->txbr; + + /* Make sure all updates are done before signaling host */ +@@ -91,7 +89,7 @@ rte_vmbus_chan_signal_tx(const struct vmbus_channel *chan) + if (tbr->vbr->imask) + return; + +- vmbus_set_event(dev, chan); ++ vmbus_set_event(chan); } -+static bool -+nat_get_unique_l4(struct conntrack *ct, struct conn *nat_conn, -+ ovs_be16 *port, uint16_t curr, uint16_t min, -+ uint16_t max) -+{ -+ uint16_t orig = curr; -+ -+ FOR_EACH_PORT_IN_RANGE (curr, min, max) { -+ *port = htons(curr); -+ if (!conn_lookup(ct, &nat_conn->rev_key, -+ time_msec(), NULL, NULL)) { -+ return true; -+ } -+ } -+ -+ *port = htons(orig); -+ -+ return false; -+} -+ - /* This function tries to get a unique tuple. - * Every iteration checks that the reverse tuple doesn't - * collide with any existing one. -@@ -2403,9 +2431,11 @@ next_addr_in_range_guarded(union ct_addr *curr, union ct_addr *min, - * - * In case of DNAT: - * - For each dst IP address in the range (if any). -- * - For each dport in range (if any). -- * - Try to find a source port in the ephemeral range -- * (after testing the port used by the sender). -+ * - For each dport in range (if any) tries to find -+ * an unique tuple. -+ * - Eventually, if the previous attempt fails, -+ * tries to find a source port in the ephemeral -+ * range (after testing the port used by the sender). + +@@ -218,7 +216,7 @@ void rte_vmbus_chan_signal_read(struct vmbus_channel *chan, uint32_t bytes_read) + if (write_sz <= pending_sz) + return; + +- vmbus_set_event(chan->device, chan); ++ vmbus_set_event(chan); + } + + int rte_vmbus_chan_recv(struct vmbus_channel *chan, void *data, uint32_t *len, +@@ -325,6 +323,7 @@ int vmbus_chan_create(const struct rte_vmbus_device *device, + chan->subchannel_id = subid; + chan->relid = relid; + chan->monitor_id = monitor_id; ++ chan->monitor_page = device->monitor_page; + *new_chan = chan; + + err = vmbus_uio_map_rings(chan); +diff --git a/dpdk/drivers/bus/vmbus/vmbus_common.c b/dpdk/drivers/bus/vmbus/vmbus_common.c +index 519ca9c6fe..367727367e 100644 +--- a/dpdk/drivers/bus/vmbus/vmbus_common.c ++++ b/dpdk/drivers/bus/vmbus/vmbus_common.c +@@ -134,7 +134,7 @@ vmbus_probe_one_driver(struct rte_vmbus_driver *dr, + + /* + * If device class GUID matches, call the probe function of +- * registere drivers for the vmbus device. ++ * register drivers for the vmbus device. + * Return -1 if initialization failed, + * and 1 if no driver found for this device. + */ +diff --git a/dpdk/drivers/common/cnxk/cnxk_security.c b/dpdk/drivers/common/cnxk/cnxk_security.c +index 30562b46e3..787138b059 100644 +--- a/dpdk/drivers/common/cnxk/cnxk_security.c ++++ b/dpdk/drivers/common/cnxk/cnxk_security.c +@@ -444,10 +444,6 @@ cnxk_ot_ipsec_outb_sa_fill(struct roc_ot_ipsec_outb_sa *sa, + return -EINVAL; + } + +- /* Default options of DSCP and Flow label/DF */ +- sa->w2.s.dscp_src = ROC_IE_OT_SA_COPY_FROM_SA; +- sa->w2.s.ipv4_df_src_or_ipv6_flw_lbl_src = ROC_IE_OT_SA_COPY_FROM_SA; +- + skip_tunnel_info: + /* ESN */ + sa->w0.s.esn_en = !!ipsec_xfrm->options.esn; +diff --git a/dpdk/drivers/common/cnxk/cnxk_telemetry_nix.c b/dpdk/drivers/common/cnxk/cnxk_telemetry_nix.c +index df6458039d..4119e9ee4f 100644 +--- a/dpdk/drivers/common/cnxk/cnxk_telemetry_nix.c ++++ b/dpdk/drivers/common/cnxk/cnxk_telemetry_nix.c +@@ -765,6 +765,9 @@ cnxk_nix_tel_handle_info_x(const char *cmd, const char *params, + + plt_strlcpy(buf, params, PCI_PRI_STR_SIZE + 1); + name = strtok(buf, ","); ++ if (name == NULL) ++ goto exit; ++ + param = strtok(NULL, "\0"); + + node = nix_tel_node_get_by_pcidev_name(name); +diff --git a/dpdk/drivers/common/cnxk/hw/cpt.h b/dpdk/drivers/common/cnxk/hw/cpt.h +index 919f8420f0..3ade4dc0c9 100644 +--- a/dpdk/drivers/common/cnxk/hw/cpt.h ++++ b/dpdk/drivers/common/cnxk/hw/cpt.h +@@ -286,10 +286,11 @@ struct cpt_frag_info_s { + union { + uint64_t u64; + struct { +- union cpt_frag_info f3; +- union cpt_frag_info f2; +- union cpt_frag_info f1; ++ /* CPT HW swaps each 8B word implicitly */ + union cpt_frag_info f0; ++ union cpt_frag_info f1; ++ union cpt_frag_info f2; ++ union cpt_frag_info f3; + }; + } w0; + +@@ -297,10 +298,11 @@ struct cpt_frag_info_s { + union { + uint64_t u64; + struct { +- uint16_t frag_size3; +- uint16_t frag_size2; +- uint16_t frag_size1; ++ /* CPT HW swaps each 8B word implicitly */ + uint16_t frag_size0; ++ uint16_t frag_size1; ++ uint16_t frag_size2; ++ uint16_t frag_size3; + }; + } w1; + }; +diff --git a/dpdk/drivers/common/cnxk/roc_bphy_cgx.c b/dpdk/drivers/common/cnxk/roc_bphy_cgx.c +index 7449cbe77a..0cd7dff655 100644 +--- a/dpdk/drivers/common/cnxk/roc_bphy_cgx.c ++++ b/dpdk/drivers/common/cnxk/roc_bphy_cgx.c +@@ -14,7 +14,7 @@ + #define CGX_CMRX_INT_OVERFLW BIT_ULL(1) + /* + * CN10K stores number of lmacs in 4 bit filed +- * in contraty to CN9K which uses only 3 bits. ++ * in contrary to CN9K which uses only 3 bits. * - * If none can be found, return exhaustion to the caller. */ - static bool -@@ -2436,6 +2466,11 @@ nat_get_unique_tuple(struct conntrack *ct, const struct conn *conn, - set_dport_range(nat_info, &conn->key, hash, &curr_dport, - &min_dport, &max_dport); + * In theory masks should differ yet on CN9K + * bits beyond specified range contain zeros. +@@ -268,9 +268,6 @@ roc_bphy_cgx_ptp_rx_ena_dis(struct roc_bphy_cgx *roc_cgx, unsigned int lmac, + { + uint64_t scr1, scr0; -+ if (pat_proto) { -+ nat_conn->rev_key.src.port = htons(curr_dport); -+ nat_conn->rev_key.dst.port = htons(curr_sport); -+ } +- if (roc_model_is_cn10k()) +- return -ENOTSUP; +- + if (!roc_cgx) + return -EINVAL; + +diff --git a/dpdk/drivers/common/cnxk/roc_bphy_irq.c b/dpdk/drivers/common/cnxk/roc_bphy_irq.c +index f4e9b341af..f4954d2a28 100644 +--- a/dpdk/drivers/common/cnxk/roc_bphy_irq.c ++++ b/dpdk/drivers/common/cnxk/roc_bphy_irq.c +@@ -259,7 +259,7 @@ roc_bphy_irq_handler_set(struct roc_bphy_irq_chip *chip, int irq_num, + + CPU_ZERO(&intr_cpuset); + CPU_SET(curr_cpu, &intr_cpuset); +- retval = pthread_setaffinity_np(pthread_self(), sizeof(intr_cpuset), ++ rc = pthread_setaffinity_np(pthread_self(), sizeof(intr_cpuset), + &intr_cpuset); + if (rc < 0) { + plt_err("Failed to set affinity mask"); +diff --git a/dpdk/drivers/common/cnxk/roc_cpt.c b/dpdk/drivers/common/cnxk/roc_cpt.c +index 8f8e6d3821..6179df2f1f 100644 +--- a/dpdk/drivers/common/cnxk/roc_cpt.c ++++ b/dpdk/drivers/common/cnxk/roc_cpt.c +@@ -385,6 +385,9 @@ cpt_lfs_alloc(struct dev *dev, uint8_t eng_grpmsk, uint8_t blkaddr, + return -EINVAL; + + req = mbox_alloc_msg_cpt_lf_alloc(mbox); ++ if (!req) ++ return -ENOSPC; ++ + req->nix_pf_func = 0; + if (inl_dev_sso && nix_inl_dev_pffunc_get()) + req->sso_pf_func = nix_inl_dev_pffunc_get(); +@@ -568,9 +571,6 @@ cpt_lf_init(struct roc_cpt_lf *lf) + if (lf->nb_desc == 0 || lf->nb_desc > CPT_LF_MAX_NB_DESC) + lf->nb_desc = CPT_LF_DEFAULT_NB_DESC; + +- /* Update nb_desc to next power of 2 to aid in pending queue checks */ +- lf->nb_desc = plt_align32pow2(lf->nb_desc); +- + /* Allocate memory for instruction queue for CPT LF. */ + iq_mem = plt_zmalloc(cpt_lf_iq_mem_calc(lf->nb_desc), ROC_ALIGN); + if (iq_mem == NULL) +@@ -812,9 +812,9 @@ roc_cpt_eng_grp_add(struct roc_cpt *roc_cpt, enum cpt_eng_type eng_type) + void + roc_cpt_iq_disable(struct roc_cpt_lf *lf) + { ++ volatile union cpt_lf_q_grp_ptr grp_ptr = {.u = 0x0}; ++ volatile union cpt_lf_inprog lf_inprog = {.u = 0x0}; + union cpt_lf_ctl lf_ctl = {.u = 0x0}; +- union cpt_lf_q_grp_ptr grp_ptr; +- union cpt_lf_inprog lf_inprog; + int timeout = 20; + int cnt; + +diff --git a/dpdk/drivers/common/cnxk/roc_cpt_debug.c b/dpdk/drivers/common/cnxk/roc_cpt_debug.c +index 847d969268..be6ddb56aa 100644 +--- a/dpdk/drivers/common/cnxk/roc_cpt_debug.c ++++ b/dpdk/drivers/common/cnxk/roc_cpt_debug.c +@@ -200,7 +200,7 @@ cpt_lf_print(struct roc_cpt_lf *lf) + reg_val = plt_read64(lf->rbase + CPT_LF_CTX_DEC_BYTE_CNT); + plt_print(" Decrypted byte count:\t%" PRIu64, reg_val); + +- reg_val = plt_read64(lf->rbase + CPT_LF_CTX_ENC_PKT_CNT); ++ reg_val = plt_read64(lf->rbase + CPT_LF_CTX_DEC_PKT_CNT); + plt_print(" Decrypted packet count:\t%" PRIu64, reg_val); + } + +diff --git a/dpdk/drivers/common/cnxk/roc_dev.c b/dpdk/drivers/common/cnxk/roc_dev.c +index 926a916e44..9a869698c4 100644 +--- a/dpdk/drivers/common/cnxk/roc_dev.c ++++ b/dpdk/drivers/common/cnxk/roc_dev.c +@@ -57,7 +57,7 @@ pf_af_sync_msg(struct dev *dev, struct mbox_msghdr **rsp) + struct mbox *mbox = dev->mbox; + struct mbox_dev *mdev = &mbox->dev[0]; + +- volatile uint64_t int_status; ++ volatile uint64_t int_status = 0; + struct mbox_msghdr *msghdr; + uint64_t off; + int rc = 0; +@@ -152,6 +152,11 @@ af_pf_wait_msg(struct dev *dev, uint16_t vf, int num_msg) + /* Reserve PF/VF mbox message */ + size = PLT_ALIGN(size, MBOX_MSG_ALIGN); + rsp = mbox_alloc_msg(&dev->mbox_vfpf, vf, size); ++ if (!rsp) { ++ plt_err("Failed to reserve VF%d message", vf); ++ continue; ++ } ++ + mbox_rsp_init(msg->id, rsp); + + /* Copy message from AF<->PF mbox to PF<->VF mbox */ +@@ -236,6 +241,12 @@ vf_pf_process_msgs(struct dev *dev, uint16_t vf) + BIT_ULL(vf % max_bits); + rsp = (struct ready_msg_rsp *)mbox_alloc_msg( + mbox, vf, sizeof(*rsp)); ++ if (!rsp) { ++ plt_err("Failed to alloc VF%d READY message", ++ vf); ++ continue; ++ } ++ + mbox_rsp_init(msg->id, rsp); + + /* PF/VF function ID */ +@@ -988,6 +999,9 @@ dev_setup_shared_lmt_region(struct mbox *mbox, bool valid_iova, uint64_t iova) + struct lmtst_tbl_setup_req *req; + + req = mbox_alloc_msg_lmtst_tbl_setup(mbox); ++ if (!req) ++ return -ENOSPC; ++ + /* This pcifunc is defined with primary pcifunc whose LMT address + * will be shared. If call contains valid IOVA, following pcifunc + * field is of no use. +@@ -1061,6 +1075,11 @@ dev_lmt_setup(struct dev *dev) + */ + if (!dev->disable_shared_lmt) { + idev = idev_get_cfg(); ++ if (!idev) { ++ errno = EFAULT; ++ goto free; ++ } ++ + if (!__atomic_load_n(&idev->lmt_pf_func, __ATOMIC_ACQUIRE)) { + idev->lmt_base_addr = dev->lmt_base; + idev->lmt_pf_func = dev->pf_func; +diff --git a/dpdk/drivers/common/cnxk/roc_irq.c b/dpdk/drivers/common/cnxk/roc_irq.c +index 7a24297d72..010b121176 100644 +--- a/dpdk/drivers/common/cnxk/roc_irq.c ++++ b/dpdk/drivers/common/cnxk/roc_irq.c +@@ -160,7 +160,10 @@ dev_irq_register(struct plt_intr_handle *intr_handle, plt_intr_callback_fn cb, + return rc; + } + +- plt_intr_efds_index_set(intr_handle, vec, fd); ++ rc = plt_intr_efds_index_set(intr_handle, vec, fd); ++ if (rc) ++ return rc; ++ + nb_efd = (vec > (uint32_t)plt_intr_nb_efd_get(intr_handle)) ? + vec : (uint32_t)plt_intr_nb_efd_get(intr_handle); + plt_intr_nb_efd_set(intr_handle, nb_efd); +diff --git a/dpdk/drivers/common/cnxk/roc_mbox.h b/dpdk/drivers/common/cnxk/roc_mbox.h +index b63fe108c9..9a8ae6b216 100644 +--- a/dpdk/drivers/common/cnxk/roc_mbox.h ++++ b/dpdk/drivers/common/cnxk/roc_mbox.h +@@ -114,7 +114,7 @@ struct mbox_msghdr { + msg_rsp) \ + M(SSO_GRP_GET_PRIORITY, 0x606, sso_grp_get_priority, sso_info_req, \ + sso_grp_priority) \ +- M(SSO_WS_CACHE_INV, 0x607, sso_ws_cache_inv, msg_req, msg_rsp) \ ++ M(SSO_WS_CACHE_INV, 0x607, sso_ws_cache_inv, ssow_lf_inv_req, msg_rsp) \ + M(SSO_GRP_QOS_CONFIG, 0x608, sso_grp_qos_config, sso_grp_qos_cfg, \ + msg_rsp) \ + M(SSO_GRP_GET_STATS, 0x609, sso_grp_get_stats, sso_info_req, \ +@@ -123,6 +123,9 @@ struct mbox_msghdr { + sso_hws_stats) \ + M(SSO_HW_RELEASE_XAQ, 0x611, sso_hw_release_xaq_aura, \ + sso_hw_xaq_release, msg_rsp) \ ++ M(SSO_CONFIG_LSW, 0x612, ssow_config_lsw, ssow_config_lsw, msg_rsp) \ ++ M(SSO_HWS_CHNG_MSHIP, 0x613, ssow_chng_mship, ssow_chng_mship, \ ++ msg_rsp) \ + /* TIM mbox IDs (range 0x800 - 0x9FF) */ \ + M(TIM_LF_ALLOC, 0x800, tim_lf_alloc, tim_lf_alloc_req, \ + tim_lf_alloc_rsp) \ +@@ -247,7 +250,8 @@ struct mbox_msghdr { + M(NIX_CPT_BP_ENABLE, 0x8020, nix_cpt_bp_enable, nix_bp_cfg_req, \ + nix_bp_cfg_rsp) \ + M(NIX_CPT_BP_DISABLE, 0x8021, nix_cpt_bp_disable, nix_bp_cfg_req, \ +- msg_rsp) ++ msg_rsp) \ ++ M(NIX_RX_SW_SYNC, 0x8022, nix_rx_sw_sync, msg_req, msg_rsp) + + /* Messages initiated by AF (range 0xC00 - 0xDFF) */ + #define MBOX_UP_CGX_MESSAGES \ +@@ -1240,6 +1244,33 @@ struct ssow_lf_free_req { + uint16_t __io hws; + }; + ++#define SSOW_INVAL_SELECTIVE_VER 0x1000 ++struct ssow_lf_inv_req { ++ struct mbox_msghdr hdr; ++ uint16_t nb_hws; /* Number of HWS to invalidate*/ ++ uint16_t hws[MAX_RVU_BLKLF_CNT]; /* Array of HWS */ ++}; + - another_round: - store_addr_to_key(&curr_addr, &nat_conn->rev_key, - nat_info->nat_action); -@@ -2449,15 +2484,19 @@ another_round: - goto next_addr; - } ++struct ssow_config_lsw { ++ struct mbox_msghdr hdr; ++#define SSOW_LSW_DIS 0 ++#define SSOW_LSW_GW_WAIT 1 ++#define SSOW_LSW_GW_IMM 2 ++ uint8_t __io lsw_mode; ++#define SSOW_WQE_REL_LSW_WAIT 0 ++#define SSOW_WQE_REL_IMM 1 ++ uint8_t __io wqe_release; ++}; ++ ++struct ssow_chng_mship { ++ struct mbox_msghdr hdr; ++ uint8_t __io set; /* Membership set to modify. */ ++ uint8_t __io enable; /* Enable/Disable the hwgrps. */ ++ uint8_t __io hws; /* HWS to modify. */ ++ uint16_t __io nb_hwgrps; /* Number of hwgrps in the array */ ++ uint16_t __io hwgrps[MAX_RVU_BLKLF_CNT]; /* Array of hwgrps. */ ++}; ++ + struct sso_hw_setconfig { + struct mbox_msghdr hdr; + uint32_t __io npa_aura_id; +diff --git a/dpdk/drivers/common/cnxk/roc_nix.h b/dpdk/drivers/common/cnxk/roc_nix.h +index 69a5e8e7b4..986aac9e57 100644 +--- a/dpdk/drivers/common/cnxk/roc_nix.h ++++ b/dpdk/drivers/common/cnxk/roc_nix.h +@@ -808,6 +808,7 @@ int __roc_api roc_nix_ptp_sync_time_adjust(struct roc_nix *roc_nix, + int __roc_api roc_nix_ptp_info_cb_register(struct roc_nix *roc_nix, + ptp_info_update_t ptp_update); + void __roc_api roc_nix_ptp_info_cb_unregister(struct roc_nix *roc_nix); ++bool __roc_api roc_nix_ptp_is_enable(struct roc_nix *roc_nix); + + /* VLAN */ + int __roc_api +diff --git a/dpdk/drivers/common/cnxk/roc_nix_bpf.c b/dpdk/drivers/common/cnxk/roc_nix_bpf.c +index 6996a54be0..4941f62995 100644 +--- a/dpdk/drivers/common/cnxk/roc_nix_bpf.c ++++ b/dpdk/drivers/common/cnxk/roc_nix_bpf.c +@@ -138,7 +138,7 @@ nix_lf_bpf_dump(__io struct nix_band_prof_s *bpf) + { + plt_dump("W0: cir_mantissa \t\t\t%d\nW0: pebs_mantissa \t\t\t0x%03x", + bpf->cir_mantissa, bpf->pebs_mantissa); +- plt_dump("W0: peir_matissa \t\t\t\t%d\nW0: cbs_exponent \t\t\t%d", ++ plt_dump("W0: peir_mantissa \t\t\t\t%d\nW0: cbs_exponent \t\t\t%d", + bpf->peir_mantissa, bpf->cbs_exponent); + plt_dump("W0: cir_exponent \t\t\t%d\nW0: pebs_exponent \t\t\t%d", + bpf->cir_exponent, bpf->pebs_exponent); +diff --git a/dpdk/drivers/common/cnxk/roc_nix_debug.c b/dpdk/drivers/common/cnxk/roc_nix_debug.c +index 266935a6c5..9829c13b69 100644 +--- a/dpdk/drivers/common/cnxk/roc_nix_debug.c ++++ b/dpdk/drivers/common/cnxk/roc_nix_debug.c +@@ -52,7 +52,9 @@ nix_bitmap_dump(struct plt_bitmap *bmp) + int i; + + plt_bitmap_scan_init(bmp); +- plt_bitmap_scan(bmp, &pos, &slab); ++ if (!plt_bitmap_scan(bmp, &pos, &slab)) ++ return; ++ + start_pos = pos; + + nix_dump_no_nl(" \t\t["); +@@ -323,6 +325,9 @@ nix_q_ctx_get(struct dev *dev, uint8_t ctype, uint16_t qid, __io void **ctx_p) + int rc; + + aq = mbox_alloc_msg_nix_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = qid; + aq->ctype = ctype; + aq->op = NIX_AQ_INSTOP_READ; +@@ -341,6 +346,9 @@ nix_q_ctx_get(struct dev *dev, uint8_t ctype, uint16_t qid, __io void **ctx_p) + struct nix_cn10k_aq_enq_req *aq; + + aq = mbox_alloc_msg_nix_cn10k_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = qid; + aq->ctype = ctype; + aq->op = NIX_AQ_INSTOP_READ; +diff --git a/dpdk/drivers/common/cnxk/roc_nix_fc.c b/dpdk/drivers/common/cnxk/roc_nix_fc.c +index ca29cd2bf9..d31137188e 100644 +--- a/dpdk/drivers/common/cnxk/roc_nix_fc.c ++++ b/dpdk/drivers/common/cnxk/roc_nix_fc.c +@@ -113,6 +113,9 @@ nix_fc_cq_config_get(struct roc_nix *roc_nix, struct roc_nix_fc_cfg *fc_cfg) + struct nix_aq_enq_req *aq; + + aq = mbox_alloc_msg_nix_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = fc_cfg->cq_cfg.rq; + aq->ctype = NIX_AQ_CTYPE_CQ; + aq->op = NIX_AQ_INSTOP_READ; +@@ -120,6 +123,9 @@ nix_fc_cq_config_get(struct roc_nix *roc_nix, struct roc_nix_fc_cfg *fc_cfg) + struct nix_cn10k_aq_enq_req *aq; + + aq = mbox_alloc_msg_nix_cn10k_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = fc_cfg->cq_cfg.rq; + aq->ctype = NIX_AQ_CTYPE_CQ; + aq->op = NIX_AQ_INSTOP_READ; +@@ -147,6 +153,9 @@ nix_fc_cq_config_set(struct roc_nix *roc_nix, struct roc_nix_fc_cfg *fc_cfg) + struct nix_aq_enq_req *aq; + + aq = mbox_alloc_msg_nix_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = fc_cfg->cq_cfg.rq; + aq->ctype = NIX_AQ_CTYPE_CQ; + aq->op = NIX_AQ_INSTOP_WRITE; +@@ -164,6 +173,9 @@ nix_fc_cq_config_set(struct roc_nix *roc_nix, struct roc_nix_fc_cfg *fc_cfg) + struct nix_cn10k_aq_enq_req *aq; + + aq = mbox_alloc_msg_nix_cn10k_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = fc_cfg->cq_cfg.rq; + aq->ctype = NIX_AQ_CTYPE_CQ; + aq->op = NIX_AQ_INSTOP_WRITE; +diff --git a/dpdk/drivers/common/cnxk/roc_nix_inl.c b/dpdk/drivers/common/cnxk/roc_nix_inl.c +index f0fc690417..e71774cf4a 100644 +--- a/dpdk/drivers/common/cnxk/roc_nix_inl.c ++++ b/dpdk/drivers/common/cnxk/roc_nix_inl.c +@@ -533,7 +533,7 @@ roc_nix_inl_dev_rq_get(struct roc_nix_rq *rq) + + inl_rq->flow_tag_width = 20; + /* Special tag mask */ +- inl_rq->tag_mask = 0xFFF00000; ++ inl_rq->tag_mask = rq->tag_mask; + inl_rq->tt = SSO_TT_ORDERED; + inl_rq->hwgrp = 0; + inl_rq->wqe_skip = 1; +@@ -595,8 +595,7 @@ roc_nix_inl_dev_rq_put(struct roc_nix_rq *rq) + plt_err("Failed to disable inline device rq, rc=%d", rc); + + /* Flush NIX LF for CN10K */ +- if (roc_model_is_cn10k()) +- plt_write64(0, inl_dev->nix_base + NIX_LF_OP_VWQE_FLUSH); ++ nix_rq_vwqe_flush(rq, inl_dev->vwqe_interval); + + return rc; + } +diff --git a/dpdk/drivers/common/cnxk/roc_nix_inl_dev.c b/dpdk/drivers/common/cnxk/roc_nix_inl_dev.c +index a0fe6ecd82..10912a6c93 100644 +--- a/dpdk/drivers/common/cnxk/roc_nix_inl_dev.c ++++ b/dpdk/drivers/common/cnxk/roc_nix_inl_dev.c +@@ -346,6 +346,7 @@ nix_inl_nix_setup(struct nix_inl_dev *inl_dev) + struct mbox *mbox = dev->mbox; + struct nix_lf_alloc_rsp *rsp; + struct nix_lf_alloc_req *req; ++ struct nix_hw_info *hw_info; + size_t inb_sa_sz; + int i, rc = -ENOSPC; + void *sa; +@@ -382,6 +383,17 @@ nix_inl_nix_setup(struct nix_inl_dev *inl_dev) + inl_dev->qints = rsp->qints; + inl_dev->cints = rsp->cints; + ++ /* Get VWQE info if supported */ ++ if (roc_model_is_cn10k()) { ++ mbox_alloc_msg_nix_get_hw_info(mbox); ++ rc = mbox_process_msg(mbox, (void *)&hw_info); ++ if (rc) { ++ plt_err("Failed to get HW info, rc=%d", rc); ++ goto lf_free; ++ } ++ inl_dev->vwqe_interval = hw_info->vwqe_delay; ++ } ++ + /* Register nix interrupts */ + rc = nix_inl_nix_register_irqs(inl_dev); + if (rc) { +diff --git a/dpdk/drivers/common/cnxk/roc_nix_inl_priv.h b/dpdk/drivers/common/cnxk/roc_nix_inl_priv.h +index 3dc526f929..be53a3fa81 100644 +--- a/dpdk/drivers/common/cnxk/roc_nix_inl_priv.h ++++ b/dpdk/drivers/common/cnxk/roc_nix_inl_priv.h +@@ -35,6 +35,7 @@ struct nix_inl_dev { + /* NIX data */ + uint8_t lf_tx_stats; + uint8_t lf_rx_stats; ++ uint16_t vwqe_interval; + uint16_t cints; + uint16_t qints; + struct roc_nix_rq rq; +diff --git a/dpdk/drivers/common/cnxk/roc_nix_irq.c b/dpdk/drivers/common/cnxk/roc_nix_irq.c +index a5cd9d4b02..7dcd533ea9 100644 +--- a/dpdk/drivers/common/cnxk/roc_nix_irq.c ++++ b/dpdk/drivers/common/cnxk/roc_nix_irq.c +@@ -202,9 +202,12 @@ nix_lf_sq_debug_reg(struct nix *nix, uint32_t off) + uint64_t reg; + + reg = plt_read64(nix->base + off); +- if (reg & BIT_ULL(44)) ++ if (reg & BIT_ULL(44)) { + plt_err("SQ=%d err_code=0x%x", (int)((reg >> 8) & 0xfffff), + (uint8_t)(reg & 0xff)); ++ /* Clear valid bit */ ++ plt_write64(BIT_ULL(44), nix->base + off); ++ } + } -- FOR_EACH_PORT_IN_RANGE(curr_dport, min_dport, max_dport) { -- nat_conn->rev_key.src.port = htons(curr_dport); -- FOR_EACH_PORT_IN_RANGE(curr_sport, min_sport, max_sport) { -- nat_conn->rev_key.dst.port = htons(curr_sport); -- if (!conn_lookup(ct, &nat_conn->rev_key, -- time_msec(), NULL, NULL)) { -- return true; -- } -- } -+ bool found = false; -+ if (nat_info->nat_action & NAT_ACTION_DST_PORT) { -+ found = nat_get_unique_l4(ct, nat_conn, &nat_conn->rev_key.src.port, -+ curr_dport, min_dport, max_dport); -+ } + static void +diff --git a/dpdk/drivers/common/cnxk/roc_nix_npc.c b/dpdk/drivers/common/cnxk/roc_nix_npc.c +index c0666c87d5..ad8839dde8 100644 +--- a/dpdk/drivers/common/cnxk/roc_nix_npc.c ++++ b/dpdk/drivers/common/cnxk/roc_nix_npc.c +@@ -96,7 +96,7 @@ roc_nix_npc_mcast_config(struct roc_nix *roc_nix, bool mcast_enable, + + if (mcast_enable) + req->mode = NIX_RX_MODE_ALLMULTI; +- else if (prom_enable) ++ if (prom_enable) + req->mode = NIX_RX_MODE_PROMISC; + + return mbox_process(mbox); +diff --git a/dpdk/drivers/common/cnxk/roc_nix_priv.h b/dpdk/drivers/common/cnxk/roc_nix_priv.h +index 04575af295..deb2a6ba11 100644 +--- a/dpdk/drivers/common/cnxk/roc_nix_priv.h ++++ b/dpdk/drivers/common/cnxk/roc_nix_priv.h +@@ -377,6 +377,7 @@ int nix_rq_cfg(struct dev *dev, struct roc_nix_rq *rq, uint16_t qints, bool cfg, + int nix_rq_ena_dis(struct dev *dev, struct roc_nix_rq *rq, bool enable); + int nix_tm_bp_config_get(struct roc_nix *roc_nix, bool *is_enabled); + int nix_tm_bp_config_set(struct roc_nix *roc_nix, bool enable); ++void nix_rq_vwqe_flush(struct roc_nix_rq *rq, uint16_t vwqe_interval); + + /* + * TM priv utils. +diff --git a/dpdk/drivers/common/cnxk/roc_nix_ptp.c b/dpdk/drivers/common/cnxk/roc_nix_ptp.c +index 03c4c6e5fd..05e4211de9 100644 +--- a/dpdk/drivers/common/cnxk/roc_nix_ptp.c ++++ b/dpdk/drivers/common/cnxk/roc_nix_ptp.c +@@ -120,3 +120,11 @@ roc_nix_ptp_info_cb_unregister(struct roc_nix *roc_nix) + + dev->ops->ptp_info_update = NULL; + } + -+ if (!found) { -+ found = nat_get_unique_l4(ct, nat_conn, &nat_conn->rev_key.dst.port, -+ curr_sport, min_sport, max_sport); -+ } ++bool ++roc_nix_ptp_is_enable(struct roc_nix *roc_nix) ++{ ++ struct nix *nix = roc_nix_to_nix_priv(roc_nix); + -+ if (found) { -+ return true; - } ++ return nix->ptp_en; ++} +diff --git a/dpdk/drivers/common/cnxk/roc_nix_queue.c b/dpdk/drivers/common/cnxk/roc_nix_queue.c +index c8c8401d81..e79a2d63e2 100644 +--- a/dpdk/drivers/common/cnxk/roc_nix_queue.c ++++ b/dpdk/drivers/common/cnxk/roc_nix_queue.c +@@ -28,6 +28,22 @@ nix_qsize_clampup(uint32_t val) + return i; + } - /* Check if next IP is in range and respin. Otherwise, notify -@@ -2857,8 +2896,8 @@ expectation_clean(struct conntrack *ct, const struct conn_key *parent_key) ++void ++nix_rq_vwqe_flush(struct roc_nix_rq *rq, uint16_t vwqe_interval) ++{ ++ uint64_t wait_ns; ++ ++ if (!roc_model_is_cn10k()) ++ return; ++ /* Due to HW errata writes to VWQE_FLUSH might hang, so instead ++ * wait for max vwqe timeout interval. ++ */ ++ if (rq->vwqe_ena) { ++ wait_ns = rq->vwqe_wait_tmo * (vwqe_interval + 1) * 100; ++ plt_delay_us((wait_ns / 1E3) + 1); ++ } ++} ++ + int + nix_rq_ena_dis(struct dev *dev, struct roc_nix_rq *rq, bool enable) { - ovs_rwlock_wrlock(&ct->resources_lock); +@@ -38,6 +54,9 @@ nix_rq_ena_dis(struct dev *dev, struct roc_nix_rq *rq, bool enable) + struct nix_aq_enq_req *aq; + + aq = mbox_alloc_msg_nix_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = rq->qid; + aq->ctype = NIX_AQ_CTYPE_RQ; + aq->op = NIX_AQ_INSTOP_WRITE; +@@ -48,6 +67,9 @@ nix_rq_ena_dis(struct dev *dev, struct roc_nix_rq *rq, bool enable) + struct nix_cn10k_aq_enq_req *aq; + + aq = mbox_alloc_msg_nix_cn10k_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = rq->qid; + aq->ctype = NIX_AQ_CTYPE_RQ; + aq->op = NIX_AQ_INSTOP_WRITE; +@@ -66,9 +88,8 @@ roc_nix_rq_ena_dis(struct roc_nix_rq *rq, bool enable) + int rc; + + rc = nix_rq_ena_dis(&nix->dev, rq, enable); ++ nix_rq_vwqe_flush(rq, nix->vwqe_interval); + +- if (roc_model_is_cn10k()) +- plt_write64(rq->qid, nix->base + NIX_LF_OP_VWQE_FLUSH); + return rc; + } -- struct alg_exp_node *node, *next; -- HINDEX_FOR_EACH_WITH_HASH_SAFE (node, next, node_ref, -+ struct alg_exp_node *node; -+ HINDEX_FOR_EACH_WITH_HASH_SAFE (node, node_ref, - conn_key_hash(parent_key, ct->hash_basis), - &ct->alg_expectation_refs) { - if (!conn_key_cmp(&node->parent_key, parent_key)) { -diff --git a/lib/dns-resolve.c b/lib/dns-resolve.c -index d344514343..1afcc65adb 100644 ---- a/lib/dns-resolve.c -+++ b/lib/dns-resolve.c -@@ -189,8 +189,8 @@ dns_resolve_destroy(void) - ub_ctx_delete(ub_ctx__); - ub_ctx__ = NULL; +@@ -80,6 +101,9 @@ nix_rq_cn9k_cfg(struct dev *dev, struct roc_nix_rq *rq, uint16_t qints, + struct nix_aq_enq_req *aq; + + aq = mbox_alloc_msg_nix_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = rq->qid; + aq->ctype = NIX_AQ_CTYPE_RQ; + aq->op = cfg ? NIX_AQ_INSTOP_WRITE : NIX_AQ_INSTOP_INIT; +@@ -195,6 +219,9 @@ nix_rq_cfg(struct dev *dev, struct roc_nix_rq *rq, uint16_t qints, bool cfg, + struct mbox *mbox = dev->mbox; + + aq = mbox_alloc_msg_nix_cn10k_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = rq->qid; + aq->ctype = NIX_AQ_CTYPE_RQ; + aq->op = cfg ? NIX_AQ_INSTOP_WRITE : NIX_AQ_INSTOP_INIT; +@@ -463,6 +490,9 @@ roc_nix_cq_init(struct roc_nix *roc_nix, struct roc_nix_cq *cq) + struct nix_aq_enq_req *aq; + + aq = mbox_alloc_msg_nix_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = cq->qid; + aq->ctype = NIX_AQ_CTYPE_CQ; + aq->op = NIX_AQ_INSTOP_INIT; +@@ -471,6 +501,9 @@ roc_nix_cq_init(struct roc_nix *roc_nix, struct roc_nix_cq *cq) + struct nix_cn10k_aq_enq_req *aq; + + aq = mbox_alloc_msg_nix_cn10k_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = cq->qid; + aq->ctype = NIX_AQ_CTYPE_CQ; + aq->op = NIX_AQ_INSTOP_INIT; +@@ -547,6 +580,9 @@ roc_nix_cq_fini(struct roc_nix_cq *cq) + struct nix_aq_enq_req *aq; + + aq = mbox_alloc_msg_nix_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = cq->qid; + aq->ctype = NIX_AQ_CTYPE_CQ; + aq->op = NIX_AQ_INSTOP_WRITE; +@@ -558,6 +594,9 @@ roc_nix_cq_fini(struct roc_nix_cq *cq) + struct nix_cn10k_aq_enq_req *aq; + + aq = mbox_alloc_msg_nix_cn10k_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = cq->qid; + aq->ctype = NIX_AQ_CTYPE_CQ; + aq->op = NIX_AQ_INSTOP_WRITE; +@@ -649,7 +688,7 @@ sqb_pool_populate(struct roc_nix *roc_nix, struct roc_nix_sq *sq) + return rc; + } -- struct resolve_request *req, *next; -- HMAP_FOR_EACH_SAFE (req, next, hmap_node, &all_reqs__) { -+ struct resolve_request *req; -+ HMAP_FOR_EACH_SAFE (req, hmap_node, &all_reqs__) { - ub_resolve_free(req->ub_result); - free(req->addr); - free(req->name); -@@ -265,7 +265,7 @@ resolve_callback__(void *req_, int err, struct ub_result *result) - if (err != 0 || (result->qtype == ns_t_aaaa && !result->havedata)) { - ub_resolve_free(result); - req->state = RESOLVE_ERROR; -- VLOG_ERR_RL(&rl, "%s: failed to resolve", req->name); -+ VLOG_WARN_RL(&rl, "%s: failed to resolve", req->name); - return; - } +-static void ++static int + sq_cn9k_init(struct nix *nix, struct roc_nix_sq *sq, uint32_t rr_quantum, + uint16_t smq) + { +@@ -657,6 +696,9 @@ sq_cn9k_init(struct nix *nix, struct roc_nix_sq *sq, uint32_t rr_quantum, + struct nix_aq_enq_req *aq; + + aq = mbox_alloc_msg_nix_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = sq->qid; + aq->ctype = NIX_AQ_CTYPE_SQ; + aq->op = NIX_AQ_INSTOP_INIT; +@@ -680,7 +722,12 @@ sq_cn9k_init(struct nix *nix, struct roc_nix_sq *sq, uint32_t rr_quantum, + aq->sq.sq_int_ena |= BIT(NIX_SQINT_MNQ_ERR); + + /* Many to one reduction */ +- aq->sq.qint_idx = sq->qid % nix->qints; ++ /* Assigning QINT 0 to all the SQs, an errata exists where NIXTX can ++ * send incorrect QINT_IDX when reporting queue interrupt (QINT). This ++ * might result in software missing the interrupt. ++ */ ++ aq->sq.qint_idx = 0; ++ return 0; + } -diff --git a/lib/dpif-netdev-avx512.c b/lib/dpif-netdev-avx512.c -index b7131ba3f1..82a4138184 100644 ---- a/lib/dpif-netdev-avx512.c -+++ b/lib/dpif-netdev-avx512.c -@@ -20,7 +20,6 @@ + static int +@@ -694,6 +741,9 @@ sq_cn9k_fini(struct nix *nix, struct roc_nix_sq *sq) + int rc, count; + + aq = mbox_alloc_msg_nix_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = sq->qid; + aq->ctype = NIX_AQ_CTYPE_SQ; + aq->op = NIX_AQ_INSTOP_READ; +@@ -707,6 +757,9 @@ sq_cn9k_fini(struct nix *nix, struct roc_nix_sq *sq) + + /* Disable sq */ + aq = mbox_alloc_msg_nix_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = sq->qid; + aq->ctype = NIX_AQ_CTYPE_SQ; + aq->op = NIX_AQ_INSTOP_WRITE; +@@ -718,6 +771,9 @@ sq_cn9k_fini(struct nix *nix, struct roc_nix_sq *sq) + + /* Read SQ and free sqb's */ + aq = mbox_alloc_msg_nix_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = sq->qid; + aq->ctype = NIX_AQ_CTYPE_SQ; + aq->op = NIX_AQ_INSTOP_READ; +@@ -749,7 +805,7 @@ sq_cn9k_fini(struct nix *nix, struct roc_nix_sq *sq) + return 0; + } - #include +-static void ++static int + sq_init(struct nix *nix, struct roc_nix_sq *sq, uint32_t rr_quantum, + uint16_t smq) + { +@@ -757,6 +813,9 @@ sq_init(struct nix *nix, struct roc_nix_sq *sq, uint32_t rr_quantum, + struct nix_cn10k_aq_enq_req *aq; + + aq = mbox_alloc_msg_nix_cn10k_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = sq->qid; + aq->ctype = NIX_AQ_CTYPE_SQ; + aq->op = NIX_AQ_INSTOP_INIT; +@@ -779,8 +838,12 @@ sq_init(struct nix *nix, struct roc_nix_sq *sq, uint32_t rr_quantum, + aq->sq.sq_int_ena |= BIT(NIX_SQINT_SEND_ERR); + aq->sq.sq_int_ena |= BIT(NIX_SQINT_MNQ_ERR); + +- /* Many to one reduction */ +- aq->sq.qint_idx = sq->qid % nix->qints; ++ /* Assigning QINT 0 to all the SQs, an errata exists where NIXTX can ++ * send incorrect QINT_IDX when reporting queue interrupt (QINT). This ++ * might result in software missing the interrupt. ++ */ ++ aq->sq.qint_idx = 0; ++ return 0; + } --#include "cpu.h" - #include "dpif-netdev.h" - #include "dpif-netdev-perf.h" - #include "dpif-netdev-private.h" -@@ -59,19 +58,6 @@ struct dpif_userdata { - struct pkt_flow_meta pkt_meta[NETDEV_MAX_BURST]; - }; + static int +@@ -794,6 +857,9 @@ sq_fini(struct nix *nix, struct roc_nix_sq *sq) + int rc, count; + + aq = mbox_alloc_msg_nix_cn10k_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = sq->qid; + aq->ctype = NIX_AQ_CTYPE_SQ; + aq->op = NIX_AQ_INSTOP_READ; +@@ -807,6 +873,9 @@ sq_fini(struct nix *nix, struct roc_nix_sq *sq) + + /* Disable sq */ + aq = mbox_alloc_msg_nix_cn10k_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = sq->qid; + aq->ctype = NIX_AQ_CTYPE_SQ; + aq->op = NIX_AQ_INSTOP_WRITE; +@@ -818,6 +887,9 @@ sq_fini(struct nix *nix, struct roc_nix_sq *sq) + + /* Read SQ and free sqb's */ + aq = mbox_alloc_msg_nix_cn10k_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = sq->qid; + aq->ctype = NIX_AQ_CTYPE_SQ; + aq->op = NIX_AQ_INSTOP_READ; +@@ -888,9 +960,12 @@ roc_nix_sq_init(struct roc_nix *roc_nix, struct roc_nix_sq *sq) + + /* Init SQ context */ + if (roc_model_is_cn9k()) +- sq_cn9k_init(nix, sq, rr_quantum, smq); ++ rc = sq_cn9k_init(nix, sq, rr_quantum, smq); + else +- sq_init(nix, sq, rr_quantum, smq); ++ rc = sq_init(nix, sq, rr_quantum, smq); ++ ++ if (rc) ++ goto nomem; + + rc = mbox_process(mbox); + if (rc) +diff --git a/dpdk/drivers/common/cnxk/roc_nix_stats.c b/dpdk/drivers/common/cnxk/roc_nix_stats.c +index c50c8fa629..756111fb1c 100644 +--- a/dpdk/drivers/common/cnxk/roc_nix_stats.c ++++ b/dpdk/drivers/common/cnxk/roc_nix_stats.c +@@ -124,6 +124,9 @@ nix_stat_rx_queue_reset(struct nix *nix, uint16_t qid) + struct nix_aq_enq_req *aq; + + aq = mbox_alloc_msg_nix_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = qid; + aq->ctype = NIX_AQ_CTYPE_RQ; + aq->op = NIX_AQ_INSTOP_WRITE; +@@ -143,6 +146,9 @@ nix_stat_rx_queue_reset(struct nix *nix, uint16_t qid) + struct nix_cn10k_aq_enq_req *aq; + + aq = mbox_alloc_msg_nix_cn10k_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = qid; + aq->ctype = NIX_AQ_CTYPE_RQ; + aq->op = NIX_AQ_INSTOP_WRITE; +@@ -174,6 +180,9 @@ nix_stat_tx_queue_reset(struct nix *nix, uint16_t qid) + struct nix_aq_enq_req *aq; + + aq = mbox_alloc_msg_nix_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = qid; + aq->ctype = NIX_AQ_CTYPE_SQ; + aq->op = NIX_AQ_INSTOP_WRITE; +@@ -190,6 +199,9 @@ nix_stat_tx_queue_reset(struct nix *nix, uint16_t qid) + struct nix_cn10k_aq_enq_req *aq; + + aq = mbox_alloc_msg_nix_cn10k_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = qid; + aq->ctype = NIX_AQ_CTYPE_SQ; + aq->op = NIX_AQ_INSTOP_WRITE; +@@ -295,6 +307,9 @@ roc_nix_xstats_get(struct roc_nix *roc_nix, struct roc_nix_xstat *xstats, + + if (roc_model_is_cn9k()) { + req = mbox_alloc_msg_cgx_stats(mbox); ++ if (!req) ++ return -ENOSPC; ++ + req->hdr.pcifunc = roc_nix_get_pf_func(roc_nix); + + rc = mbox_process_msg(mbox, (void *)&cgx_resp); +@@ -316,6 +331,9 @@ roc_nix_xstats_get(struct roc_nix *roc_nix, struct roc_nix_xstat *xstats, + } + } else { + req = mbox_alloc_msg_rpm_stats(mbox); ++ if (!req) ++ return -ENOSPC; ++ + req->hdr.pcifunc = roc_nix_get_pf_func(roc_nix); + + rc = mbox_process_msg(mbox, (void *)&rpm_resp); +diff --git a/dpdk/drivers/common/cnxk/roc_nix_tm.c b/dpdk/drivers/common/cnxk/roc_nix_tm.c +index b3d8ebd3c2..506cb066ce 100644 +--- a/dpdk/drivers/common/cnxk/roc_nix_tm.c ++++ b/dpdk/drivers/common/cnxk/roc_nix_tm.c +@@ -424,7 +424,7 @@ nix_tm_bp_config_get(struct roc_nix *roc_nix, bool *is_enabled) + + if (req) { + req->num_regs = k; +- rc = mbox_process(mbox); ++ rc = mbox_process_msg(mbox, (void **)&rsp); + if (rc) + goto err; + /* Report it as enabled only if enabled or all */ +@@ -564,6 +564,7 @@ nix_tm_sq_flush_pre(struct roc_nix_sq *sq) + struct nix_tm_node *node, *sibling; + struct nix_tm_node_list *list; + enum roc_nix_tm_tree tree; ++ struct msg_req *req; + struct mbox *mbox; + struct nix *nix; + uint16_t qid; +@@ -653,6 +654,12 @@ nix_tm_sq_flush_pre(struct roc_nix_sq *sq) + rc); + goto cleanup; + } ++ ++ req = mbox_alloc_msg_nix_rx_sw_sync(mbox); ++ if (!req) ++ return -ENOSPC; ++ ++ rc = mbox_process(mbox); + cleanup: + /* Restore cgx state */ + if (!roc_nix->io_enabled) { +@@ -766,6 +773,9 @@ nix_tm_sq_sched_conf(struct nix *nix, struct nix_tm_node *node, + struct nix_aq_enq_req *aq; + + aq = mbox_alloc_msg_nix_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = qid; + aq->ctype = NIX_AQ_CTYPE_SQ; + aq->op = NIX_AQ_INSTOP_WRITE; +@@ -781,6 +791,9 @@ nix_tm_sq_sched_conf(struct nix *nix, struct nix_tm_node *node, + struct nix_cn10k_aq_enq_req *aq; + + aq = mbox_alloc_msg_nix_cn10k_aq_enq(mbox); ++ if (!aq) ++ return -ENOSPC; ++ + aq->qidx = qid; + aq->ctype = NIX_AQ_CTYPE_SQ; + aq->op = NIX_AQ_INSTOP_WRITE; +diff --git a/dpdk/drivers/common/cnxk/roc_nix_tm_ops.c b/dpdk/drivers/common/cnxk/roc_nix_tm_ops.c +index 3257fa67c7..3d81247a12 100644 +--- a/dpdk/drivers/common/cnxk/roc_nix_tm_ops.c ++++ b/dpdk/drivers/common/cnxk/roc_nix_tm_ops.c +@@ -107,7 +107,7 @@ nix_tm_adjust_shaper_pps_rate(struct nix_tm_shaper_profile *profile) + if (profile->peak.rate && min_rate > profile->peak.rate) + min_rate = profile->peak.rate; + +- /* Each packet accomulate single count, whereas HW ++ /* Each packet accumulate single count, whereas HW + * considers each unit as Byte, so we need convert + * user pps to bps + */ +diff --git a/dpdk/drivers/common/cnxk/roc_nix_tm_utils.c b/dpdk/drivers/common/cnxk/roc_nix_tm_utils.c +index 543adf9e56..9e80c2a5fe 100644 +--- a/dpdk/drivers/common/cnxk/roc_nix_tm_utils.c ++++ b/dpdk/drivers/common/cnxk/roc_nix_tm_utils.c +@@ -642,6 +642,7 @@ nix_tm_shaper_reg_prep(struct nix_tm_node *node, + else if (profile) + adjust = profile->pkt_len_adj; + ++ adjust &= 0x1FF; + plt_tm_dbg("Shaper config node %s(%u) lvl %u id %u, " + "pir %" PRIu64 "(%" PRIu64 "B)," + " cir %" PRIu64 "(%" PRIu64 "B)" +@@ -708,7 +709,7 @@ nix_tm_shaper_reg_prep(struct nix_tm_node *node, + /* Configure RED algo */ + reg[k] = NIX_AF_TL3X_SHAPE(schq); + regval[k] = (adjust | (uint64_t)node->red_algo << 9 | +- (uint64_t)node->pkt_mode); ++ (uint64_t)node->pkt_mode << 24); + k++; --int32_t --dp_netdev_input_outer_avx512_probe(void) --{ -- bool avx512f_available = cpu_has_isa(OVS_CPU_ISA_X86_AVX512F); -- bool bmi2_available = cpu_has_isa(OVS_CPU_ISA_X86_BMI2); -- -- if (!avx512f_available || !bmi2_available) { -- return -ENOTSUP; -- } -- -- return 0; --} -- - int32_t - dp_netdev_input_outer_avx512(struct dp_netdev_pmd_thread *pmd, - struct dp_packet_batch *packets, -@@ -159,7 +145,7 @@ dp_netdev_input_outer_avx512(struct dp_netdev_pmd_thread *pmd, - mf_mask = mfex_func(packets, keys, batch_size, in_port, pmd); - } + break; +diff --git a/dpdk/drivers/common/cnxk/roc_npc.c b/dpdk/drivers/common/cnxk/roc_npc.c +index 503c74748f..5ee7ff5e41 100644 +--- a/dpdk/drivers/common/cnxk/roc_npc.c ++++ b/dpdk/drivers/common/cnxk/roc_npc.c +@@ -969,14 +969,14 @@ npc_vtag_insert_action_configure(struct mbox *mbox, struct roc_npc_flow *flow, + vtag_cfg->cfg_type = VTAG_TX; + vtag_cfg->vtag_size = NIX_VTAGSIZE_T4; + vtag_cfg->tx.vtag0 = +- ((vlan_info[0].vlan_ethtype << 16) | ++ (((uint32_t)vlan_info[0].vlan_ethtype << 16) | + (vlan_info[0].vlan_pcp << 13) | vlan_info[0].vlan_id); + + vtag_cfg->tx.cfg_vtag0 = 1; + + if (flow->vtag_insert_count == 2) { + vtag_cfg->tx.vtag1 = +- ((vlan_info[1].vlan_ethtype << 16) | ++ (((uint32_t)vlan_info[1].vlan_ethtype << 16) | + (vlan_info[1].vlan_pcp << 13) | vlan_info[1].vlan_id); + + vtag_cfg->tx.cfg_vtag1 = 1; +@@ -1246,6 +1246,16 @@ roc_npc_flow_destroy(struct roc_npc *roc_npc, struct roc_npc_flow *flow) + return rc; + } -- uint32_t lookup_pkts_bitmask = (1ULL << batch_size) - 1; -+ uint32_t lookup_pkts_bitmask = (UINT64_C(1) << batch_size) - 1; - uint32_t iter = lookup_pkts_bitmask; - while (iter) { - uint32_t i = raw_ctz(iter); -@@ -183,7 +169,7 @@ dp_netdev_input_outer_avx512(struct dp_netdev_pmd_thread *pmd, - * classifed by vector mfex else do a scalar miniflow extract - * for that packet. - */ -- bool mfex_hit = !!(mf_mask & (1 << i)); -+ bool mfex_hit = !!(mf_mask & (UINT32_C(1) << i)); ++ if (flow->ctr_id != NPC_COUNTER_NONE) { ++ rc = roc_npc_mcam_clear_counter(roc_npc, flow->ctr_id); ++ if (rc != 0) ++ return rc; ++ ++ rc = npc_mcam_free_counter(npc, flow->ctr_id); ++ if (rc != 0) ++ return rc; ++ } ++ + rc = npc_mcam_free_entry(npc, flow->mcam_id); + if (rc != 0) + return rc; +diff --git a/dpdk/drivers/common/cnxk/roc_npc_mcam.c b/dpdk/drivers/common/cnxk/roc_npc_mcam.c +index ba7f89b45b..a16ba3f7be 100644 +--- a/dpdk/drivers/common/cnxk/roc_npc_mcam.c ++++ b/dpdk/drivers/common/cnxk/roc_npc_mcam.c +@@ -234,7 +234,7 @@ npc_get_kex_capability(struct npc *npc) + /* Ethtype: Offset 12B, len 2B */ + kex_cap.bit.ethtype_0 = npc_is_kex_enabled( + npc, NPC_LID_LA, NPC_LT_LA_ETHER, 12 * 8, 2 * 8); +- /* QINQ VLAN Ethtype: ofset 8B, len 2B */ ++ /* QINQ VLAN Ethtype: offset 8B, len 2B */ + kex_cap.bit.ethtype_x = npc_is_kex_enabled( + npc, NPC_LID_LB, NPC_LT_LB_STAG_QINQ, 8 * 8, 2 * 8); + /* VLAN ID0 : Outer VLAN: Offset 2B, len 2B */ +@@ -670,7 +670,7 @@ npc_program_mcam(struct npc *npc, struct npc_parse_state *pst, bool mcam_alloc) + memcpy(pst->flow->mcam_data, key_data, key_len); + memcpy(pst->flow->mcam_mask, key_mask, key_len); + +- if (pst->is_vf) { ++ if (pst->is_vf && pst->flow->nix_intf == NIX_INTF_RX) { + (void)mbox_alloc_msg_npc_read_base_steer_rule(npc->mbox); + rc = mbox_process_msg(npc->mbox, (void *)&base_rule_rsp); + if (rc) { +diff --git a/dpdk/drivers/common/cnxk/roc_npc_mcam_dump.c b/dpdk/drivers/common/cnxk/roc_npc_mcam_dump.c +index 19b4901a52..278056591e 100644 +--- a/dpdk/drivers/common/cnxk/roc_npc_mcam_dump.c ++++ b/dpdk/drivers/common/cnxk/roc_npc_mcam_dump.c +@@ -159,6 +159,12 @@ npc_flow_print_parse_nibbles(FILE *file, struct roc_npc_flow *flow, + offset += 4; + } - /* Check for a partial hardware offload match. */ - if (hwol_enabled) { -@@ -204,7 +190,7 @@ dp_netdev_input_outer_avx512(struct dp_netdev_pmd_thread *pmd, ++ if (rx_parse->laflags) { ++ data = npc_get_nibbles(flow, 2, offset); ++ fprintf(file, "\tNPC_PARSE_NIBBLE_LA_FLAGS:%#02X\n", data); ++ offset += 8; ++ } ++ + if (rx_parse->latype) { + data = npc_get_nibbles(flow, 1, offset); + fprintf(file, "\tNPC_PARSE_NIBBLE_LA_LTYPE:%s\n", +@@ -166,9 +172,9 @@ npc_flow_print_parse_nibbles(FILE *file, struct roc_npc_flow *flow, + offset += 4; + } - pkt_meta[i].bytes = dp_packet_size(packet); - phwol_hits++; -- hwol_emc_smc_hitmask |= (1 << i); -+ hwol_emc_smc_hitmask |= (UINT32_C(1) << i); - continue; - } - } -@@ -227,7 +213,7 @@ dp_netdev_input_outer_avx512(struct dp_netdev_pmd_thread *pmd, - if (f) { - rules[i] = &f->cr; - emc_hits++; -- hwol_emc_smc_hitmask |= (1 << i); -+ hwol_emc_smc_hitmask |= (UINT32_C(1) << i); - continue; - } - } -@@ -237,7 +223,7 @@ dp_netdev_input_outer_avx512(struct dp_netdev_pmd_thread *pmd, - if (f) { - rules[i] = &f->cr; - smc_hits++; -- smc_hitmask |= (1 << i); -+ smc_hitmask |= (UINT32_C(1) << i); - continue; - } - } -diff --git a/lib/dpif-netdev-extract-avx512.c b/lib/dpif-netdev-extract-avx512.c -index c1c1fefb6a..92980ca1b9 100644 ---- a/lib/dpif-netdev-extract-avx512.c -+++ b/lib/dpif-netdev-extract-avx512.c -@@ -42,7 +42,6 @@ - #include - #include +- if (rx_parse->laflags) { ++ if (rx_parse->lbflags) { + data = npc_get_nibbles(flow, 2, offset); +- fprintf(file, "\tNPC_PARSE_NIBBLE_LA_FLAGS:%#02X\n", data); ++ fprintf(file, "\tNPC_PARSE_NIBBLE_LB_FLAGS:%#02X\n", data); + offset += 8; + } --#include "cpu.h" - #include "flow.h" +@@ -179,9 +185,9 @@ npc_flow_print_parse_nibbles(FILE *file, struct roc_npc_flow *flow, + offset += 4; + } - #include "dpif-netdev-private-dpcls.h" -@@ -544,7 +543,11 @@ mfex_avx512_process(struct dp_packet_batch *packets, - */ - __m512i v512_zeros = _mm512_setzero_si512(); - __m512i v_blk0; -+#if __GNUC__ >= 4 - if (__builtin_constant_p(use_vbmi) && use_vbmi) { -+#else -+ if (use_vbmi) { -+#endif - v_blk0 = _mm512_maskz_permutexvar_epi8_wrap(k_shuf, v_shuf, - v_pkt0); - } else { -@@ -619,7 +622,7 @@ mfex_avx512_process(struct dp_packet_batch *packets, - }; +- if (rx_parse->lbflags) { ++ if (rx_parse->lcflags) { + data = npc_get_nibbles(flow, 2, offset); +- fprintf(file, "\tNPC_PARSE_NIBBLE_LB_FLAGS:%#02X\n", data); ++ fprintf(file, "\tNPC_PARSE_NIBBLE_LC_FLAGS:%#02X\n", data); + offset += 8; + } - /* This packet has its miniflow created, add to hitmask. */ -- hitmask |= 1 << i; -+ hitmask |= UINT32_C(1) << i; - } +@@ -192,9 +198,9 @@ npc_flow_print_parse_nibbles(FILE *file, struct roc_npc_flow *flow, + offset += 4; + } - return hitmask; -@@ -659,47 +662,5 @@ DECLARE_MFEX_FUNC(ip_udp, PROFILE_ETH_IPV4_UDP) - DECLARE_MFEX_FUNC(ip_tcp, PROFILE_ETH_IPV4_TCP) - DECLARE_MFEX_FUNC(dot1q_ip_udp, PROFILE_ETH_VLAN_IPV4_UDP) - DECLARE_MFEX_FUNC(dot1q_ip_tcp, PROFILE_ETH_VLAN_IPV4_TCP) -- -- --static int32_t --avx512_isa_probe(uint32_t needs_vbmi) --{ -- static enum ovs_cpu_isa isa_required[] = { -- OVS_CPU_ISA_X86_AVX512F, -- OVS_CPU_ISA_X86_AVX512BW, -- OVS_CPU_ISA_X86_BMI2, -- }; -- -- int32_t ret = 0; -- for (uint32_t i = 0; i < ARRAY_SIZE(isa_required); i++) { -- if (!cpu_has_isa(isa_required[i])) { -- ret = -ENOTSUP; -- } -- } -- -- if (needs_vbmi) { -- if (!cpu_has_isa(OVS_CPU_ISA_X86_AVX512VBMI)) { -- ret = -ENOTSUP; -- } -- } -- -- return ret; --} -- --/* Probe functions to check ISA requirements. */ --int32_t --mfex_avx512_probe(void) --{ -- const uint32_t needs_vbmi = 0; -- return avx512_isa_probe(needs_vbmi); --} -- --int32_t --mfex_avx512_vbmi_probe(void) --{ -- const uint32_t needs_vbmi = 1; -- return avx512_isa_probe(needs_vbmi); --} -- - #endif /* __CHECKER__ */ - #endif /* __x86_64__ */ -diff --git a/lib/dpif-netdev-lookup-avx512-gather.c b/lib/dpif-netdev-lookup-avx512-gather.c -index 7bc1e9e9a5..fb2084392a 100644 ---- a/lib/dpif-netdev-lookup-avx512-gather.c -+++ b/lib/dpif-netdev-lookup-avx512-gather.c -@@ -23,7 +23,6 @@ - #include "dpif-netdev-lookup.h" +- if (rx_parse->lcflags) { ++ if (rx_parse->ldflags) { + data = npc_get_nibbles(flow, 2, offset); +- fprintf(file, "\tNPC_PARSE_NIBBLE_LC_FLAGS:%#02X\n", data); ++ fprintf(file, "\tNPC_PARSE_NIBBLE_LD_FLAGS:%#02X\n", data); + offset += 8; + } - #include "cmap.h" --#include "cpu.h" - #include "flow.h" - #include "pvector.h" - #include "openvswitch/vlog.h" -@@ -396,18 +395,11 @@ dpcls_avx512_gather_mf_any(struct dpcls_subtable *subtable, uint32_t keys_map, - } +@@ -205,9 +211,9 @@ npc_flow_print_parse_nibbles(FILE *file, struct roc_npc_flow *flow, + offset += 4; + } - dpcls_subtable_lookup_func --dpcls_subtable_avx512_gather_probe(uint32_t u0_bits, uint32_t u1_bits) -+dpcls_subtable_avx512_gather_probe__(uint32_t u0_bits, uint32_t u1_bits, -+ bool use_vpop) - { - dpcls_subtable_lookup_func f = NULL; +- if (rx_parse->ldflags) { ++ if (rx_parse->leflags) { + data = npc_get_nibbles(flow, 2, offset); +- fprintf(file, "\tNPC_PARSE_NIBBLE_LD_FLAGS:%#02X\n", data); ++ fprintf(file, "\tNPC_PARSE_NIBBLE_LE_FLAGS:%#02X\n", data); + offset += 8; + } -- int avx512f_available = cpu_has_isa(OVS_CPU_ISA_X86_AVX512F); -- int bmi2_available = cpu_has_isa(OVS_CPU_ISA_X86_BMI2); -- if (!avx512f_available || !bmi2_available) { -- return NULL; -- } -- -- int use_vpop = cpu_has_isa(OVS_CPU_ISA_X86_VPOPCNTDQ); +@@ -218,9 +224,9 @@ npc_flow_print_parse_nibbles(FILE *file, struct roc_npc_flow *flow, + offset += 4; + } + +- if (rx_parse->leflags) { ++ if (rx_parse->lfflags) { + data = npc_get_nibbles(flow, 2, offset); +- fprintf(file, "\tNPC_PARSE_NIBBLE_LE_FLAGS:%#02X\n", data); ++ fprintf(file, "\tNPC_PARSE_NIBBLE_LF_FLAGS:%#02X\n", data); + offset += 8; + } + +@@ -231,9 +237,9 @@ npc_flow_print_parse_nibbles(FILE *file, struct roc_npc_flow *flow, + offset += 4; + } + +- if (rx_parse->lfflags) { ++ if (rx_parse->lgflags) { + data = npc_get_nibbles(flow, 2, offset); +- fprintf(file, "\tNPC_PARSE_NIBBLE_LF_FLAGS:%#02X\n", data); ++ fprintf(file, "\tNPC_PARSE_NIBBLE_LG_FLAGS:%#02X\n", data); + offset += 8; + } + +@@ -244,10 +250,9 @@ npc_flow_print_parse_nibbles(FILE *file, struct roc_npc_flow *flow, + offset += 4; + } + +- if (rx_parse->lgflags) { ++ if (rx_parse->lhflags) { + data = npc_get_nibbles(flow, 2, offset); +- fprintf(file, "\tNPC_PARSE_NIBBLE_LG_FLAGS:%#02X\n", data); +- offset += 8; ++ fprintf(file, "\tNPC_PARSE_NIBBLE_LH_FLAGS:%#02X\n", data); + } + + if (rx_parse->lhtype) { +@@ -256,11 +261,6 @@ npc_flow_print_parse_nibbles(FILE *file, struct roc_npc_flow *flow, + ltype_str[NPC_LID_LH][data]); + offset += 4; + } - - CHECK_LOOKUP_FUNCTION(9, 4, use_vpop); - CHECK_LOOKUP_FUNCTION(9, 1, use_vpop); - CHECK_LOOKUP_FUNCTION(5, 3, use_vpop); -diff --git a/lib/dpif-netdev-lookup.c b/lib/dpif-netdev-lookup.c -index bd0a99abe7..b1d2801575 100644 ---- a/lib/dpif-netdev-lookup.c -+++ b/lib/dpif-netdev-lookup.c -@@ -18,10 +18,25 @@ - #include - #include "dpif-netdev-lookup.h" +- if (rx_parse->lhflags) { +- data = npc_get_nibbles(flow, 2, offset); +- fprintf(file, "\tNPC_PARSE_NIBBLE_LH_FLAGS:%#02X\n", data); +- } + } -+#include "cpu.h" - #include "openvswitch/vlog.h" + static void +diff --git a/dpdk/drivers/common/cnxk/roc_npc_parse.c b/dpdk/drivers/common/cnxk/roc_npc_parse.c +index 8125035dd8..9742ac90f7 100644 +--- a/dpdk/drivers/common/cnxk/roc_npc_parse.c ++++ b/dpdk/drivers/common/cnxk/roc_npc_parse.c +@@ -38,6 +38,7 @@ npc_parse_cpt_hdr(struct npc_parse_state *pst) + info.hw_hdr_len = 0; + + /* Prepare for parsing the item */ ++ info.def_mask = NULL; + info.hw_mask = &hw_mask; + info.len = pst->pattern->size; + npc_get_hw_supp_mask(pst, &info, lid, lt); +@@ -75,6 +76,7 @@ npc_parse_higig2_hdr(struct npc_parse_state *pst) + } - VLOG_DEFINE_THIS_MODULE(dpif_netdev_lookup); + /* Prepare for parsing the item */ ++ info.def_mask = NULL; + info.hw_mask = &hw_mask; + info.len = pst->pattern->size; + npc_get_hw_supp_mask(pst, &info, lid, lt); +@@ -121,6 +123,7 @@ npc_parse_la(struct npc_parse_state *pst) + } -+#if (__x86_64__ && HAVE_AVX512F && HAVE_LD_AVX512_GOOD && __SSE4_2__) -+static dpcls_subtable_lookup_func -+dpcls_subtable_avx512_gather_probe(uint32_t u0_bits, uint32_t u1_bits) -+{ -+ if (!cpu_has_isa(OVS_CPU_ISA_X86_AVX512F) -+ || !cpu_has_isa(OVS_CPU_ISA_X86_BMI2)) { -+ return NULL; -+ } -+ -+ return dpcls_subtable_avx512_gather_probe__(u0_bits, u1_bits, -+ cpu_has_isa(OVS_CPU_ISA_X86_VPOPCNTDQ)); -+} -+#endif + /* Prepare for parsing the item */ ++ info.def_mask = NULL; + info.hw_mask = &hw_mask; + info.len = pst->pattern->size; + npc_get_hw_supp_mask(pst, &info, lid, lt); +@@ -179,6 +182,7 @@ npc_parse_lb(struct npc_parse_state *pst) + int nr_vlans = 0; + int rc; + ++ info.def_mask = NULL; + info.spec = NULL; + info.mask = NULL; + info.def_mask = NULL; +@@ -307,12 +311,12 @@ npc_parse_mpls_label_stack(struct npc_parse_state *pst, int *flag) + * pst->pattern points to first MPLS label. We only check + * that subsequent labels do not have anything to match. + */ ++ info.def_mask = NULL; + info.hw_mask = NULL; + info.len = pattern->size; + info.spec = NULL; + info.mask = NULL; + info.hw_hdr_len = 0; +- info.def_mask = NULL; + + while (pattern->type == ROC_NPC_ITEM_TYPE_MPLS) { + nr_labels++; +@@ -358,6 +362,7 @@ npc_parse_mpls(struct npc_parse_state *pst, int lid) + info.len = pst->pattern->size; + info.spec = NULL; + info.mask = NULL; ++ info.def_mask = NULL; + info.hw_hdr_len = 0; + + npc_get_hw_supp_mask(pst, &info, lid, lt); +@@ -405,6 +410,7 @@ npc_parse_lc(struct npc_parse_state *pst) + if (pst->pattern->type == ROC_NPC_ITEM_TYPE_MPLS) + return npc_parse_mpls(pst, NPC_LID_LC); + ++ info.def_mask = NULL; + info.hw_mask = &hw_mask; + info.spec = NULL; + info.mask = NULL; +@@ -492,10 +498,10 @@ npc_parse_ld(struct npc_parse_state *pst) + return npc_parse_mpls(pst, NPC_LID_LD); + return 0; + } ++ info.def_mask = NULL; + info.hw_mask = &hw_mask; + info.spec = NULL; + info.mask = NULL; +- info.def_mask = NULL; + info.len = 0; + info.hw_hdr_len = 0; + +@@ -529,11 +535,13 @@ npc_parse_ld(struct npc_parse_state *pst) + case ROC_NPC_ITEM_TYPE_GRE: + lt = NPC_LT_LD_GRE; + info.len = pst->pattern->size; ++ pst->tunnel = 1; + break; + case ROC_NPC_ITEM_TYPE_GRE_KEY: + lt = NPC_LT_LD_GRE; + info.len = pst->pattern->size; + info.hw_hdr_len = 4; ++ pst->tunnel = 1; + break; + case ROC_NPC_ITEM_TYPE_NVGRE: + lt = NPC_LT_LD_NVGRE; +@@ -651,6 +659,7 @@ npc_parse_lf(struct npc_parse_state *pst) + lflags = 0; + + /* No match support for vlan tags */ ++ info.def_mask = NULL; + info.hw_mask = NULL; + info.len = pst->pattern->size; + info.spec = NULL; +@@ -709,6 +718,7 @@ npc_parse_lg(struct npc_parse_state *pst) + if (!pst->tunnel) + return 0; + ++ info.def_mask = NULL; + info.hw_mask = &hw_mask; + info.spec = NULL; + info.mask = NULL; +@@ -745,6 +755,7 @@ npc_parse_lh(struct npc_parse_state *pst) + if (!pst->tunnel) + return 0; + ++ info.def_mask = NULL; + info.hw_mask = &hw_mask; + info.spec = NULL; + info.mask = NULL; +diff --git a/dpdk/drivers/common/cnxk/roc_npc_priv.h b/dpdk/drivers/common/cnxk/roc_npc_priv.h +index 712302bc5c..74e0fb2ece 100644 +--- a/dpdk/drivers/common/cnxk/roc_npc_priv.h ++++ b/dpdk/drivers/common/cnxk/roc_npc_priv.h +@@ -363,7 +363,7 @@ struct npc { + uint32_t rss_grps; /* rss groups supported */ + uint16_t flow_prealloc_size; /* Pre allocated mcam size */ + uint16_t flow_max_priority; /* Max priority for flow */ +- uint16_t switch_header_type; /* Suppprted switch header type */ ++ uint16_t switch_header_type; /* Supported switch header type */ + uint32_t mark_actions; /* Number of mark actions */ + uint32_t vtag_strip_actions; /* vtag insert/strip actions */ + uint16_t pf_func; /* pf_func of device */ +diff --git a/dpdk/drivers/common/cnxk/roc_npc_utils.c b/dpdk/drivers/common/cnxk/roc_npc_utils.c +index ed0ef5c462..e36a312576 100644 +--- a/dpdk/drivers/common/cnxk/roc_npc_utils.c ++++ b/dpdk/drivers/common/cnxk/roc_npc_utils.c +@@ -145,6 +145,9 @@ npc_parse_item_basic(const struct roc_npc_item_info *item, + info->mask = item->mask; + } + ++ if (info->mask == NULL) ++ return NPC_ERR_INVALID_MASK; + - /* Actual list of implementations goes here */ - static struct dpcls_subtable_lookup_info_t subtable_lookups[] = { - /* The autovalidator implementation will not be used by default, it must -diff --git a/lib/dpif-netdev-lookup.h b/lib/dpif-netdev-lookup.h -index 59f51faa0e..5d2d845945 100644 ---- a/lib/dpif-netdev-lookup.h -+++ b/lib/dpif-netdev-lookup.h -@@ -44,7 +44,8 @@ dpcls_subtable_generic_probe(uint32_t u0_bit_count, uint32_t u1_bit_count); + /* mask specified must be subset of hw supported mask + * mask | hw_mask == hw_mask + */ +@@ -166,6 +169,9 @@ npc_update_extraction_data(struct npc_parse_state *pst, + int len = 0; - /* Probe function for AVX-512 gather implementation */ - dpcls_subtable_lookup_func --dpcls_subtable_avx512_gather_probe(uint32_t u0_bit_cnt, uint32_t u1_bit_cnt); -+dpcls_subtable_avx512_gather_probe__(uint32_t u0_bit_cnt, uint32_t u1_bit_cnt, -+ bool use_vpop); + x = xinfo; ++ if (x->len > NPC_MAX_EXTRACT_DATA_LEN) ++ return NPC_ERR_INVALID_SIZE; ++ + len = x->len; + hdr_off = x->hdr_off; +@@ -579,7 +585,7 @@ npc_allocate_mcam_entry(struct mbox *mbox, int prio, + if (!rsp_cmd->count) + return -ENOSPC; - /* Subtable registration and iteration helpers */ -diff --git a/lib/dpif-netdev-private-dpif.c b/lib/dpif-netdev-private-dpif.c -index 84d4ec156e..5ae119a308 100644 ---- a/lib/dpif-netdev-private-dpif.c -+++ b/lib/dpif-netdev-private-dpif.c -@@ -22,6 +22,7 @@ - #include - #include +- memcpy(rsp_local, rsp_cmd, sizeof(*rsp)); ++ mbox_memcpy(rsp_local, rsp_cmd, sizeof(*rsp)); -+#include "cpu.h" - #include "openvswitch/dynamic-string.h" - #include "openvswitch/vlog.h" - #include "util.h" -@@ -33,6 +34,19 @@ enum dpif_netdev_impl_info_idx { - DPIF_NETDEV_IMPL_AVX512 + return 0; + } +@@ -664,14 +670,14 @@ npc_get_free_mcam_entry(struct mbox *mbox, struct roc_npc_flow *flow, + + new_entry->flow = flow; + +- plt_info("npc: kernel allocated MCAM entry %d", rsp_local.entry); ++ plt_npc_dbg("kernel allocated MCAM entry %d", rsp_local.entry); + + rc = npc_sort_mcams_by_user_prio_level(mbox, new_entry, npc, + &rsp_local); + if (rc) + goto err; + +- plt_info("npc: allocated MCAM entry after sorting %d", rsp_local.entry); ++ plt_npc_dbg("allocated MCAM entry after sorting %d", rsp_local.entry); + flow->mcam_id = rsp_local.entry; + npc_insert_into_flow_list(npc, new_entry); + +diff --git a/dpdk/drivers/common/cnxk/roc_platform.c b/dpdk/drivers/common/cnxk/roc_platform.c +index 74dbdeceb9..6cf0c4113e 100644 +--- a/dpdk/drivers/common/cnxk/roc_platform.c ++++ b/dpdk/drivers/common/cnxk/roc_platform.c +@@ -37,7 +37,11 @@ roc_plt_init(void) + plt_err("Failed to reserve mem for roc_model"); + return -ENOMEM; + } +- roc_model_init(mz->addr); ++ if (roc_model_init(mz->addr)) { ++ plt_err("Failed to init roc_model"); ++ rte_memzone_free(mz); ++ return -EINVAL; ++ } + } + } else { + if (mz == NULL) { +diff --git a/dpdk/drivers/common/cnxk/roc_se.c b/dpdk/drivers/common/cnxk/roc_se.c +index ffe537af30..3f0821e400 100644 +--- a/dpdk/drivers/common/cnxk/roc_se.c ++++ b/dpdk/drivers/common/cnxk/roc_se.c +@@ -283,6 +283,8 @@ roc_se_auth_key_set(struct roc_se_ctx *se_ctx, roc_se_auth_type type, + return ret; + se_ctx->pdcp_alg_type = ROC_SE_PDCP_ALG_TYPE_ZUC; + memcpy(ci_key, key, key_len); ++ if (key_len == 32) ++ roc_se_zuc_bytes_swap(ci_key, key_len); + cpt_pdcp_update_zuc_const(zuc_const, key_len, mac_len); + se_ctx->fc_type = ROC_SE_PDCP; + se_ctx->zsk_flags = 0x1; +@@ -459,9 +461,10 @@ roc_se_ciph_key_set(struct roc_se_ctx *se_ctx, roc_se_cipher_type type, + zs_ctx->zuc.otk_ctx.w0.s.alg_type = ROC_SE_PDCP_ALG_TYPE_ZUC; + se_ctx->pdcp_alg_type = ROC_SE_PDCP_ALG_TYPE_ZUC; + memcpy(ci_key, key, key_len); +- if (key_len == 32) ++ if (key_len == 32) { ++ roc_se_zuc_bytes_swap(ci_key, key_len); + memcpy(zuc_const, zuc_key256, 16); +- else ++ } else + memcpy(zuc_const, zuc_key128, 32); + + se_ctx->zsk_flags = 0; +diff --git a/dpdk/drivers/common/cnxk/roc_se.h b/dpdk/drivers/common/cnxk/roc_se.h +index 5be832fa75..500f94ac11 100644 +--- a/dpdk/drivers/common/cnxk/roc_se.h ++++ b/dpdk/drivers/common/cnxk/roc_se.h +@@ -297,6 +297,27 @@ struct roc_se_ctx { + uint8_t *auth_key; }; -+#if (__x86_64__ && HAVE_AVX512F && HAVE_LD_AVX512_GOOD && __SSE4_2__) -+static int32_t -+dp_netdev_input_outer_avx512_probe(void) ++static inline void ++roc_se_zuc_bytes_swap(uint8_t *arr, int len) +{ -+ if (!cpu_has_isa(OVS_CPU_ISA_X86_AVX512F) -+ || !cpu_has_isa(OVS_CPU_ISA_X86_BMI2)) { -+ return -ENOTSUP; -+ } -+ -+ return 0; ++ int start, end; ++ uint8_t tmp; ++ ++ if (len <= 0) ++ return; ++ ++ start = 0; ++ end = len - 1; ++ ++ while (start < end) { ++ tmp = arr[start]; ++ arr[start] = arr[end]; ++ arr[end] = tmp; ++ start++; ++ end--; ++ } +} -+#endif + - /* Actual list of implementations goes here. */ - static struct dpif_netdev_impl_info_t dpif_impls[] = { - /* The default scalar C code implementation. */ -diff --git a/lib/dpif-netdev-private-dpif.h b/lib/dpif-netdev-private-dpif.h -index 0da639c55a..3e38630f53 100644 ---- a/lib/dpif-netdev-private-dpif.h -+++ b/lib/dpif-netdev-private-dpif.h -@@ -67,10 +67,7 @@ dp_netdev_input(struct dp_netdev_pmd_thread *pmd, - struct dp_packet_batch *packets, - odp_port_t in_port); + int __roc_api roc_se_auth_key_set(struct roc_se_ctx *se_ctx, + roc_se_auth_type type, const uint8_t *key, + uint16_t key_len, uint16_t mac_len); +@@ -306,4 +327,5 @@ int __roc_api roc_se_ciph_key_set(struct roc_se_ctx *se_ctx, + uint16_t key_len, uint8_t *salt); + + void __roc_api roc_se_ctx_swap(struct roc_se_ctx *se_ctx); ++ + #endif /* __ROC_SE_H__ */ +diff --git a/dpdk/drivers/common/cnxk/roc_sso.c b/dpdk/drivers/common/cnxk/roc_sso.c +index 45ff16ca0e..ede0a8a3bb 100644 +--- a/dpdk/drivers/common/cnxk/roc_sso.c ++++ b/dpdk/drivers/common/cnxk/roc_sso.c +@@ -378,10 +378,10 @@ roc_sso_hwgrp_qos_config(struct roc_sso *roc_sso, struct roc_sso_hwgrp_qos *qos, + } + req->grp = qos[i].hwgrp; + req->xaq_limit = (nb_xaq * (xaq_prcnt ? xaq_prcnt : 100)) / 100; +- req->taq_thr = (SSO_HWGRP_IAQ_MAX_THR_MASK * ++ req->iaq_thr = (SSO_HWGRP_IAQ_MAX_THR_MASK * + (iaq_prcnt ? iaq_prcnt : 100)) / + 100; +- req->iaq_thr = (SSO_HWGRP_TAQ_MAX_THR_MASK * ++ req->taq_thr = (SSO_HWGRP_TAQ_MAX_THR_MASK * + (taq_prcnt ? taq_prcnt : 100)) / + 100; + } +diff --git a/dpdk/drivers/common/cnxk/version.map b/dpdk/drivers/common/cnxk/version.map +index 07c6720f0c..556369bc02 100644 +--- a/dpdk/drivers/common/cnxk/version.map ++++ b/dpdk/drivers/common/cnxk/version.map +@@ -198,6 +198,7 @@ INTERNAL { + roc_nix_ptp_clock_read; + roc_nix_ptp_info_cb_register; + roc_nix_ptp_info_cb_unregister; ++ roc_nix_ptp_is_enable; + roc_nix_ptp_rx_ena_dis; + roc_nix_ptp_sync_time_adjust; + roc_nix_ptp_tx_ena_dis; +diff --git a/dpdk/drivers/common/cpt/cpt_mcode_defines.h b/dpdk/drivers/common/cpt/cpt_mcode_defines.h +index f16ee44297..e6dcb7674c 100644 +--- a/dpdk/drivers/common/cpt/cpt_mcode_defines.h ++++ b/dpdk/drivers/common/cpt/cpt_mcode_defines.h +@@ -387,7 +387,7 @@ typedef struct buf_ptr { + /* IOV Pointer */ + typedef struct{ + int buf_cnt; +- buf_ptr_t bufs[0]; ++ buf_ptr_t bufs[]; + } iov_ptr_t; + + typedef struct fc_params { +diff --git a/dpdk/drivers/common/cpt/cpt_ucode.h b/dpdk/drivers/common/cpt/cpt_ucode.h +index e015cf66a1..22aabab6ac 100644 +--- a/dpdk/drivers/common/cpt/cpt_ucode.h ++++ b/dpdk/drivers/common/cpt/cpt_ucode.h +@@ -246,7 +246,7 @@ cpt_fc_ciph_set_key(struct cpt_ctx *cpt_ctx, cipher_type_t type, + if (cpt_ctx->fc_type == FC_GEN) { + /* + * We need to always say IV is from DPTR as user can +- * sometimes iverride IV per operation. ++ * sometimes override IV per operation. + */ + fctx->enc.iv_source = CPT_FROM_DPTR; + +@@ -394,27 +394,26 @@ fill_sg_comp_from_iov(sg_comp_t *list, + int32_t j; + uint32_t extra_len = extra_buf ? extra_buf->size : 0; + uint32_t size = *psize; +- buf_ptr_t *bufs; + +- bufs = from->bufs; + for (j = 0; (j < from->buf_cnt) && size; j++) { ++ phys_addr_t dma_addr = from->bufs[j].dma_addr; ++ uint32_t buf_sz = from->bufs[j].size; ++ sg_comp_t *to = &list[i >> 2]; + phys_addr_t e_dma_addr; + uint32_t e_len; +- sg_comp_t *to = &list[i >> 2]; + + if (unlikely(from_offset)) { +- if (from_offset >= bufs[j].size) { +- from_offset -= bufs[j].size; ++ if (from_offset >= buf_sz) { ++ from_offset -= buf_sz; + continue; + } +- e_dma_addr = bufs[j].dma_addr + from_offset; +- e_len = (size > (bufs[j].size - from_offset)) ? +- (bufs[j].size - from_offset) : size; ++ e_dma_addr = dma_addr + from_offset; ++ e_len = (size > (buf_sz - from_offset)) ? ++ (buf_sz - from_offset) : size; + from_offset = 0; + } else { +- e_dma_addr = bufs[j].dma_addr; +- e_len = (size > bufs[j].size) ? +- bufs[j].size : size; ++ e_dma_addr = dma_addr; ++ e_len = (size > buf_sz) ? buf_sz : size; + } + + to->u.s.len[i % 4] = rte_cpu_to_be_16(e_len); +@@ -3035,7 +3034,7 @@ prepare_iov_from_pkt_inplace(struct rte_mbuf *pkt, + tailroom = rte_pktmbuf_tailroom(pkt); + if (likely((headroom >= 24) && + (tailroom >= 8))) { +- /* In 83XX this is prerequivisit for Direct mode */ ++ /* In 83XX this is prerequisite for Direct mode */ + *flags |= SINGLE_BUF_HEADTAILROOM; + } + param->bufs[0].vaddr = seg_data; +diff --git a/dpdk/drivers/common/cpt/cpt_ucode_asym.h b/dpdk/drivers/common/cpt/cpt_ucode_asym.h +index a67ded642a..f0b5dddd8c 100644 +--- a/dpdk/drivers/common/cpt/cpt_ucode_asym.h ++++ b/dpdk/drivers/common/cpt/cpt_ucode_asym.h +@@ -779,7 +779,7 @@ cpt_ecdsa_verify_prep(struct rte_crypto_ecdsa_op_param *ecdsa, + * Set dlen = sum(sizeof(fpm address), ROUNDUP8(message len), + * ROUNDUP8(sign len(r and s), public key len(x and y coordinates), + * prime len, order len)). +- * Please note sign, public key and order can not excede prime length ++ * Please note sign, public key and order can not exceed prime length + * i.e. 6 * p_align + */ + dlen = sizeof(fpm_table_iova) + m_align + (8 * p_align); +diff --git a/dpdk/drivers/common/dpaax/caamflib/desc/algo.h b/dpdk/drivers/common/dpaax/caamflib/desc/algo.h +index 6bb915054a..e0848f0940 100644 +--- a/dpdk/drivers/common/dpaax/caamflib/desc/algo.h ++++ b/dpdk/drivers/common/dpaax/caamflib/desc/algo.h +@@ -67,7 +67,7 @@ cnstr_shdsc_zuce(uint32_t *descbuf, bool ps, bool swap, + * @authlen: size of digest + * + * The IV prepended before hmac payload must be 8 bytes consisting +- * of COUNT||BEAERER||DIR. The COUNT is of 32-bits, bearer is of 5 bits and ++ * of COUNT||BEARER||DIR. The COUNT is of 32-bits, bearer is of 5 bits and + * direction is of 1 bit - totalling to 38 bits. + * + * Return: size of descriptor written in words or negative number on error +diff --git a/dpdk/drivers/common/dpaax/caamflib/desc/pdcp.h b/dpdk/drivers/common/dpaax/caamflib/desc/pdcp.h +index 8e8daf5ba8..2c9c631cfd 100644 +--- a/dpdk/drivers/common/dpaax/caamflib/desc/pdcp.h ++++ b/dpdk/drivers/common/dpaax/caamflib/desc/pdcp.h +@@ -3795,7 +3795,7 @@ cnstr_shdsc_pdcp_short_mac(uint32_t *descbuf, + return -ENOTSUP; + } + iv[0] = 0xFFFFFFFF; +- iv[1] = swap ? swab32(0xFC000000) : 0xFC000000; ++ iv[1] = swab32(0xFC000000); + iv[2] = 0x00000000; /* unused */ + + KEY(p, KEY2, authdata->key_enc_flags, authdata->key, +diff --git a/dpdk/drivers/common/dpaax/caamflib/desc/sdap.h b/dpdk/drivers/common/dpaax/caamflib/desc/sdap.h +index b2497a5424..07f55b5b40 100644 +--- a/dpdk/drivers/common/dpaax/caamflib/desc/sdap.h ++++ b/dpdk/drivers/common/dpaax/caamflib/desc/sdap.h +@@ -492,10 +492,10 @@ pdcp_sdap_insert_snoop_op(struct program *p, bool swap __maybe_unused, + + /* Set the variable size of data the register will write */ + if (dir == OP_TYPE_ENCAP_PROTOCOL) { +- /* We will add the interity data so add its length */ ++ /* We will add the integrity data so add its length */ + MATHI(p, SEQINSZ, ADD, PDCP_MAC_I_LEN, VSEQOUTSZ, 4, IMMED2); + } else { +- /* We will check the interity data so remove its length */ ++ /* We will check the integrity data so remove its length */ + MATHI(p, SEQINSZ, SUB, PDCP_MAC_I_LEN, VSEQOUTSZ, 4, IMMED2); + /* Do not take the ICV in the out-snooping configuration */ + MATHI(p, SEQINSZ, SUB, PDCP_MAC_I_LEN, VSEQINSZ, 4, IMMED2); +@@ -803,7 +803,7 @@ static inline int pdcp_sdap_insert_no_snoop_op( + CLRW_CLR_C1MODE, + CLRW, 0, 4, IMMED); + +- /* Load the key for authentcation */ ++ /* Load the key for authentication */ + KEY(p, KEY1, authdata->key_enc_flags, authdata->key, + authdata->keylen, INLINE_KEY(authdata)); + +diff --git a/dpdk/drivers/common/dpaax/dpaax_iova_table.c b/dpdk/drivers/common/dpaax/dpaax_iova_table.c +index 3d661102cc..9daac4bc03 100644 +--- a/dpdk/drivers/common/dpaax/dpaax_iova_table.c ++++ b/dpdk/drivers/common/dpaax/dpaax_iova_table.c +@@ -261,7 +261,7 @@ dpaax_iova_table_depopulate(void) + rte_free(dpaax_iova_table_p->entries); + dpaax_iova_table_p = NULL; + +- DPAAX_DEBUG("IOVA Table cleanedup"); ++ DPAAX_DEBUG("IOVA Table cleaned"); + } --/* AVX512 enabled DPIF implementation and probe functions. */ --int32_t --dp_netdev_input_outer_avx512_probe(void); + int +diff --git a/dpdk/drivers/common/iavf/iavf_type.h b/dpdk/drivers/common/iavf/iavf_type.h +index 51267ca3b3..1cd87587d6 100644 +--- a/dpdk/drivers/common/iavf/iavf_type.h ++++ b/dpdk/drivers/common/iavf/iavf_type.h +@@ -1006,7 +1006,7 @@ struct iavf_profile_tlv_section_record { + u8 data[12]; + }; + +-/* Generic AQ section in proflie */ ++/* Generic AQ section in profile */ + struct iavf_profile_aq_section { + u16 opcode; + u16 flags; +diff --git a/dpdk/drivers/common/iavf/virtchnl.h b/dpdk/drivers/common/iavf/virtchnl.h +index 269578f7c0..80e754a1b2 100644 +--- a/dpdk/drivers/common/iavf/virtchnl.h ++++ b/dpdk/drivers/common/iavf/virtchnl.h +@@ -233,7 +233,7 @@ static inline const char *virtchnl_op_str(enum virtchnl_ops v_opcode) + case VIRTCHNL_OP_DCF_CMD_DESC: + return "VIRTCHNL_OP_DCF_CMD_DESC"; + case VIRTCHNL_OP_DCF_CMD_BUFF: +- return "VIRTCHHNL_OP_DCF_CMD_BUFF"; ++ return "VIRTCHNL_OP_DCF_CMD_BUFF"; + case VIRTCHNL_OP_DCF_DISABLE: + return "VIRTCHNL_OP_DCF_DISABLE"; + case VIRTCHNL_OP_DCF_GET_VSI_MAP: +diff --git a/dpdk/drivers/common/iavf/virtchnl_inline_ipsec.h b/dpdk/drivers/common/iavf/virtchnl_inline_ipsec.h +index 1e9134501e..2f4bf15725 100644 +--- a/dpdk/drivers/common/iavf/virtchnl_inline_ipsec.h ++++ b/dpdk/drivers/common/iavf/virtchnl_inline_ipsec.h +@@ -446,6 +446,15 @@ struct virtchnl_ipsec_sp_cfg { + + /* Set TC (congestion domain) if true. For future use. */ + u8 set_tc; ++ ++ /* 0 for NAT-T unsupported, 1 for NAT-T supported */ ++ u8 is_udp; ++ ++ /* reserved */ ++ u8 reserved; ++ ++ /* NAT-T UDP port number. Only valid in case NAT-T supported */ ++ u16 udp_port; + } __rte_packed; + + +diff --git a/dpdk/drivers/common/mlx5/linux/meson.build b/dpdk/drivers/common/mlx5/linux/meson.build +index 7909f23e21..4c7b53b9bd 100644 +--- a/dpdk/drivers/common/mlx5/linux/meson.build ++++ b/dpdk/drivers/common/mlx5/linux/meson.build +@@ -36,7 +36,7 @@ foreach libname:libnames + endforeach + if static_ibverbs or dlopen_ibverbs + # Build without adding shared libs to Requires.private +- ibv_cflags = run_command(pkgconf, '--cflags', 'libibverbs').stdout() ++ ibv_cflags = run_command(pkgconf, '--cflags', 'libibverbs', check: true).stdout() + ext_deps += declare_dependency(compile_args: ibv_cflags.split()) + endif + if static_ibverbs +diff --git a/dpdk/drivers/common/mlx5/linux/mlx5_common_os.c b/dpdk/drivers/common/mlx5/linux/mlx5_common_os.c +index 0d3e24e04e..eeb583a553 100644 +--- a/dpdk/drivers/common/mlx5/linux/mlx5_common_os.c ++++ b/dpdk/drivers/common/mlx5/linux/mlx5_common_os.c +@@ -456,21 +456,33 @@ mlx5_os_get_ibv_device(const struct rte_pci_addr *addr) + int n; + struct ibv_device **ibv_list = mlx5_glue->get_device_list(&n); + struct ibv_device *ibv_match = NULL; ++ uint8_t guid1[32] = {0}; ++ uint8_t guid2[32] = {0}; ++ int ret1, ret2 = -1; ++ struct rte_pci_addr paddr; + +- if (ibv_list == NULL) { ++ if (ibv_list == NULL || !n) { + rte_errno = ENOSYS; ++ if (ibv_list) ++ mlx5_glue->free_device_list(ibv_list); + return NULL; + } ++ ret1 = mlx5_get_device_guid(addr, guid1, sizeof(guid1)); + while (n-- > 0) { +- struct rte_pci_addr paddr; - -+/* AVX512 enabled DPIF implementation function. */ - int32_t - dp_netdev_input_outer_avx512(struct dp_netdev_pmd_thread *pmd, - struct dp_packet_batch *packets, -diff --git a/lib/dpif-netdev-private-extract.c b/lib/dpif-netdev-private-extract.c -index a29bdcfa78..fe04ea80ff 100644 ---- a/lib/dpif-netdev-private-extract.c -+++ b/lib/dpif-netdev-private-extract.c -@@ -19,6 +19,7 @@ - #include - #include + DRV_LOG(DEBUG, "Checking device \"%s\"..", ibv_list[n]->name); + if (mlx5_get_pci_addr(ibv_list[n]->ibdev_path, &paddr) != 0) + continue; +- if (rte_pci_addr_cmp(addr, &paddr) != 0) +- continue; +- ibv_match = ibv_list[n]; +- break; ++ if (ret1 > 0) ++ ret2 = mlx5_get_device_guid(&paddr, guid2, sizeof(guid2)); ++ /* Bond device can bond secondary PCIe */ ++ if ((strstr(ibv_list[n]->name, "bond") && ++ ((ret1 > 0 && ret2 > 0 && !memcmp(guid1, guid2, sizeof(guid1))) || ++ (addr->domain == paddr.domain && addr->bus == paddr.bus && ++ addr->devid == paddr.devid))) || ++ !rte_pci_addr_cmp(addr, &paddr)) { ++ ibv_match = ibv_list[n]; ++ break; ++ } + } + if (ibv_match == NULL) { + DRV_LOG(WARNING, +@@ -487,7 +499,7 @@ mlx5_os_get_ibv_device(const struct rte_pci_addr *addr) + static int + mlx5_nl_roce_disable(const char *addr) + { +- int nlsk_fd = mlx5_nl_init(NETLINK_GENERIC); ++ int nlsk_fd = mlx5_nl_init(NETLINK_GENERIC, 0); + int devlink_id; + int enable; + int ret; +diff --git a/dpdk/drivers/common/mlx5/linux/mlx5_common_os.h b/dpdk/drivers/common/mlx5/linux/mlx5_common_os.h +index 83066e752d..a6190a34e6 100644 +--- a/dpdk/drivers/common/mlx5/linux/mlx5_common_os.h ++++ b/dpdk/drivers/common/mlx5/linux/mlx5_common_os.h +@@ -9,6 +9,7 @@ + #include + + #include ++#include + #include + #include + #include +@@ -300,6 +301,7 @@ mlx5_set_context_attr(struct rte_device *dev, struct ibv_context *ctx); + * 0 if OFED doesn't support. + * >0 if success. + */ ++__rte_internal + int + mlx5_get_device_guid(const struct rte_pci_addr *dev, uint8_t *guid, size_t len); -+#include "cpu.h" - #include "dp-packet.h" - #include "dpif-netdev-private-dpcls.h" - #include "dpif-netdev-private-extract.h" -@@ -33,6 +34,43 @@ VLOG_DEFINE_THIS_MODULE(dpif_netdev_extract); - /* Variable to hold the default MFEX implementation. */ - static ATOMIC(miniflow_extract_func) default_mfex_func; - -+#if (__x86_64__ && HAVE_AVX512F && HAVE_LD_AVX512_GOOD && __SSE4_2__) -+static int32_t -+avx512_isa_probe(bool needs_vbmi) -+{ -+ static enum ovs_cpu_isa isa_required[] = { -+ OVS_CPU_ISA_X86_AVX512F, -+ OVS_CPU_ISA_X86_AVX512BW, -+ OVS_CPU_ISA_X86_BMI2, -+ }; -+ -+ for (uint32_t i = 0; i < ARRAY_SIZE(isa_required); i++) { -+ if (!cpu_has_isa(isa_required[i])) { -+ return -ENOTSUP; -+ } -+ } -+ -+ if (needs_vbmi && !cpu_has_isa(OVS_CPU_ISA_X86_AVX512VBMI)) { -+ return -ENOTSUP; -+ } +diff --git a/dpdk/drivers/common/mlx5/linux/mlx5_nl.c b/dpdk/drivers/common/mlx5/linux/mlx5_nl.c +index fd4c2d2625..5d04857b38 100644 +--- a/dpdk/drivers/common/mlx5/linux/mlx5_nl.c ++++ b/dpdk/drivers/common/mlx5/linux/mlx5_nl.c +@@ -185,19 +185,22 @@ uint32_t atomic_sn; + * + * @param protocol + * Netlink protocol (e.g. NETLINK_ROUTE, NETLINK_RDMA). ++ * @param groups ++ * Groups to listen (e.g. RTMGRP_LINK), can be 0. + * + * @return + * A file descriptor on success, a negative errno value otherwise and + * rte_errno is set. + */ + int +-mlx5_nl_init(int protocol) ++mlx5_nl_init(int protocol, int groups) + { + int fd; + int buf_size; + socklen_t opt_size; + struct sockaddr_nl local = { + .nl_family = AF_NETLINK, ++ .nl_groups = groups, + }; + int ret; + +@@ -1862,3 +1865,100 @@ mlx5_nl_enable_roce_set(int nlsk_fd, int family_id, const char *pci_addr, + /* Now, need to reload the driver. */ + return mlx5_nl_driver_reload(nlsk_fd, family_id, pci_addr); + } + -+ return 0; ++/** ++ * Try to parse a Netlink message as a link status update. ++ * ++ * @param hdr ++ * Netlink message header. ++ * @param[out] ifindex ++ * Index of the updated interface. ++ * ++ * @return ++ * 0 on success, negative on failure. ++ */ ++int ++mlx5_nl_parse_link_status_update(struct nlmsghdr *hdr, uint32_t *ifindex) ++{ ++ struct ifinfomsg *info; ++ ++ switch (hdr->nlmsg_type) { ++ case RTM_NEWLINK: ++ case RTM_DELLINK: ++ case RTM_GETLINK: ++ case RTM_SETLINK: ++ info = NLMSG_DATA(hdr); ++ *ifindex = info->ifi_index; ++ return 0; ++ } ++ return -1; +} + -+/* Probe functions to check ISA requirements. */ -+static int32_t -+mfex_avx512_probe(void) ++/** ++ * Read pending events from a Netlink socket. ++ * ++ * @param nlsk_fd ++ * Netlink socket. ++ * @param cb ++ * Callback invoked for each of the events. ++ * @param cb_arg ++ * User data for the callback. ++ * ++ * @return ++ * 0 on success, including the case when there are no events. ++ * Negative on failure and rte_errno is set. ++ */ ++int ++mlx5_nl_read_events(int nlsk_fd, mlx5_nl_event_cb *cb, void *cb_arg) +{ -+ return avx512_isa_probe(false); ++ char buf[8192]; ++ struct sockaddr_nl addr; ++ struct iovec iov = { ++ .iov_base = buf, ++ .iov_len = sizeof(buf), ++ }; ++ struct msghdr msg = { ++ .msg_name = &addr, ++ .msg_namelen = sizeof(addr), ++ .msg_iov = &iov, ++ .msg_iovlen = 1, ++ }; ++ struct nlmsghdr *hdr; ++ ssize_t size; ++ ++ while (1) { ++ size = recvmsg(nlsk_fd, &msg, MSG_DONTWAIT); ++ if (size < 0) { ++ if (errno == EAGAIN) ++ return 0; ++ if (errno == EINTR) ++ continue; ++ DRV_LOG(DEBUG, "Failed to receive netlink message: %s", ++ strerror(errno)); ++ rte_errno = errno; ++ return -rte_errno; ++ } ++ hdr = (struct nlmsghdr *)buf; ++ while (size >= (ssize_t)sizeof(*hdr)) { ++ ssize_t msg_len = hdr->nlmsg_len; ++ ssize_t data_len = msg_len - sizeof(*hdr); ++ ssize_t aligned_len; ++ ++ if (data_len < 0) { ++ DRV_LOG(DEBUG, "Netlink message too short"); ++ rte_errno = EINVAL; ++ return -rte_errno; ++ } ++ aligned_len = NLMSG_ALIGN(msg_len); ++ if (aligned_len > size) { ++ DRV_LOG(DEBUG, "Netlink message too long"); ++ rte_errno = EINVAL; ++ return -rte_errno; ++ } ++ cb(hdr, cb_arg); ++ hdr = RTE_PTR_ADD(hdr, aligned_len); ++ size -= aligned_len; ++ } ++ } ++ return 0; ++} +diff --git a/dpdk/drivers/common/mlx5/linux/mlx5_nl.h b/dpdk/drivers/common/mlx5/linux/mlx5_nl.h +index 2063c0deeb..0b7552338a 100644 +--- a/dpdk/drivers/common/mlx5/linux/mlx5_nl.h ++++ b/dpdk/drivers/common/mlx5/linux/mlx5_nl.h +@@ -11,6 +11,7 @@ + + #include "mlx5_common.h" + ++typedef void (mlx5_nl_event_cb)(struct nlmsghdr *hdr, void *user_data); + + /* VLAN netdev for VLAN workaround. */ + struct mlx5_nl_vlan_dev { +@@ -30,7 +31,7 @@ struct mlx5_nl_vlan_vmwa_context { + }; + + __rte_internal +-int mlx5_nl_init(int protocol); ++int mlx5_nl_init(int protocol, int groups); + __rte_internal + int mlx5_nl_mac_addr_add(int nlsk_fd, unsigned int iface_idx, uint64_t *mac_own, + struct rte_ether_addr *mac, uint32_t index); +@@ -75,4 +76,9 @@ int mlx5_nl_enable_roce_get(int nlsk_fd, int family_id, const char *pci_addr, + int mlx5_nl_enable_roce_set(int nlsk_fd, int family_id, const char *pci_addr, + int enable); + ++__rte_internal ++int mlx5_nl_read_events(int nlsk_fd, mlx5_nl_event_cb *cb, void *cb_arg); ++__rte_internal ++int mlx5_nl_parse_link_status_update(struct nlmsghdr *hdr, uint32_t *ifindex); ++ + #endif /* RTE_PMD_MLX5_NL_H_ */ +diff --git a/dpdk/drivers/common/mlx5/mlx5_common.c b/dpdk/drivers/common/mlx5/mlx5_common.c +index f1650f94c6..4faae6c86d 100644 +--- a/dpdk/drivers/common/mlx5/mlx5_common.c ++++ b/dpdk/drivers/common/mlx5/mlx5_common.c +@@ -111,6 +111,11 @@ mlx5_common_args_check_handler(const char *key, const char *val, void *opaque) + struct mlx5_common_dev_config *config = opaque; + signed long tmp; + ++ if (val == NULL || *val == '\0') { ++ DRV_LOG(ERR, "Key %s is missing value.", key); ++ rte_errno = EINVAL; ++ return -rte_errno; ++ } + errno = 0; + tmp = strtol(val, NULL, 0); + if (errno) { +@@ -616,7 +621,6 @@ drivers_remove(struct mlx5_common_device *cdev, uint32_t enabled_classes) + unsigned int i = 0; + int ret = 0; + +- enabled_classes &= cdev->classes_loaded; + while (enabled_classes) { + driver = driver_get(RTE_BIT64(i)); + if (driver != NULL) { +@@ -640,7 +644,7 @@ drivers_probe(struct mlx5_common_device *cdev, uint32_t user_classes) + struct mlx5_class_driver *driver; + uint32_t enabled_classes = 0; + bool already_loaded; +- int ret; ++ int ret = -EINVAL; + + TAILQ_FOREACH(driver, &drivers_list, next) { + if ((driver->drv_class & user_classes) == 0) +@@ -662,12 +666,16 @@ drivers_probe(struct mlx5_common_device *cdev, uint32_t user_classes) + } + enabled_classes |= driver->drv_class; + } +- cdev->classes_loaded |= enabled_classes; +- return 0; ++ if (!ret) { ++ cdev->classes_loaded |= enabled_classes; ++ return 0; ++ } + probe_err: +- /* Only unload drivers which are enabled which were enabled +- * in this probe instance. ++ /* ++ * Need to remove only drivers which were not probed before this probe ++ * instance, but have already been probed before this failure. + */ ++ enabled_classes &= ~cdev->classes_loaded; + drivers_remove(cdev, enabled_classes); + return ret; + } +@@ -754,6 +762,7 @@ mlx5_common_dev_dma_map(struct rte_device *rte_dev, void *addr, + uint64_t iova __rte_unused, size_t len) + { + struct mlx5_common_device *dev; ++ struct mlx5_mr_btree *bt; + struct mlx5_mr *mr; + + dev = to_mlx5_device(rte_dev); +@@ -771,7 +780,36 @@ mlx5_common_dev_dma_map(struct rte_device *rte_dev, void *addr, + rte_errno = EINVAL; + return -1; + } ++try_insert: + rte_rwlock_write_lock(&dev->mr_scache.rwlock); ++ bt = &dev->mr_scache.cache; ++ if (bt->len == bt->size) { ++ uint32_t size; ++ int ret; ++ ++ size = bt->size + 1; ++ MLX5_ASSERT(size > bt->size); ++ /* ++ * Avoid deadlock (numbers show the sequence of events): ++ * mlx5_mr_create_primary(): ++ * 1) take EAL memory lock ++ * 3) take MR lock ++ * this function: ++ * 2) take MR lock ++ * 4) take EAL memory lock while allocating the new cache ++ * Releasing the MR lock before step 4 ++ * allows another thread to execute step 3. ++ */ ++ rte_rwlock_write_unlock(&dev->mr_scache.rwlock); ++ ret = mlx5_mr_expand_cache(&dev->mr_scache, size, ++ rte_dev->numa_node); ++ if (ret < 0) { ++ mlx5_mr_free(mr, dev->mr_scache.dereg_mr_cb); ++ rte_errno = ret; ++ return -1; ++ } ++ goto try_insert; ++ } + LIST_INSERT_HEAD(&dev->mr_scache.mr_list, mr, mr); + /* Insert to the global cache table. */ + mlx5_mr_insert_cache(&dev->mr_scache, mr); +@@ -854,7 +892,7 @@ static void mlx5_common_driver_init(void) + static bool mlx5_common_initialized; + + /** +- * One time innitialization routine for run-time dependency on glue library ++ * One time initialization routine for run-time dependency on glue library + * for multiple PMDs. Each mlx5 PMD that depends on mlx5_common module, + * must invoke in its constructor. + */ +diff --git a/dpdk/drivers/common/mlx5/mlx5_common_mr.c b/dpdk/drivers/common/mlx5/mlx5_common_mr.c +index c694aaf28c..26fdf22386 100644 +--- a/dpdk/drivers/common/mlx5/mlx5_common_mr.c ++++ b/dpdk/drivers/common/mlx5/mlx5_common_mr.c +@@ -78,7 +78,7 @@ mlx5_mprq_buf_free_cb(void *addr __rte_unused, void *opaque) + * 0 on success, -1 on failure. + */ + static int +-mr_btree_expand(struct mlx5_mr_btree *bt, int n) ++mr_btree_expand(struct mlx5_mr_btree *bt, uint32_t n) + { + void *mem; + int ret = 0; +@@ -123,11 +123,11 @@ mr_btree_expand(struct mlx5_mr_btree *bt, int n) + * Searched LKey on success, UINT32_MAX on no match. + */ + static uint32_t +-mr_btree_lookup(struct mlx5_mr_btree *bt, uint16_t *idx, uintptr_t addr) ++mr_btree_lookup(struct mlx5_mr_btree *bt, uint32_t *idx, uintptr_t addr) + { + struct mr_cache_entry *lkp_tbl; +- uint16_t n; +- uint16_t base = 0; ++ uint32_t n; ++ uint32_t base = 0; + + MLX5_ASSERT(bt != NULL); + lkp_tbl = *bt->table; +@@ -137,7 +137,7 @@ mr_btree_lookup(struct mlx5_mr_btree *bt, uint16_t *idx, uintptr_t addr) + lkp_tbl[0].lkey == UINT32_MAX)); + /* Binary search. */ + do { +- register uint16_t delta = n >> 1; ++ register uint32_t delta = n >> 1; + + if (addr < lkp_tbl[base + delta].start) { + n = delta; +@@ -169,7 +169,7 @@ static int + mr_btree_insert(struct mlx5_mr_btree *bt, struct mr_cache_entry *entry) + { + struct mr_cache_entry *lkp_tbl; +- uint16_t idx = 0; ++ uint32_t idx = 0; + size_t shift; + + MLX5_ASSERT(bt != NULL); +@@ -185,11 +185,8 @@ mr_btree_insert(struct mlx5_mr_btree *bt, struct mr_cache_entry *entry) + /* Already exist, return. */ + return 0; + } +- /* If table is full, return error. */ +- if (unlikely(bt->len == bt->size)) { +- bt->overflow = 1; +- return -1; +- } ++ /* Caller must ensure that there is enough place for a new entry. */ ++ MLX5_ASSERT(bt->len < bt->size); + /* Insert entry. */ + ++idx; + shift = (bt->len - idx) * sizeof(struct mr_cache_entry); +@@ -273,7 +270,7 @@ void + mlx5_mr_btree_dump(struct mlx5_mr_btree *bt __rte_unused) + { + #ifdef RTE_LIBRTE_MLX5_DEBUG +- int idx; ++ uint32_t idx; + struct mr_cache_entry *lkp_tbl; + + if (bt == NULL) +@@ -409,13 +406,8 @@ mlx5_mr_insert_cache(struct mlx5_mr_share_cache *share_cache, + n = mr_find_next_chunk(mr, &entry, n); + if (!entry.end) + break; +- if (mr_btree_insert(&share_cache->cache, &entry) < 0) { +- /* +- * Overflowed, but the global table cannot be expanded +- * because of deadlock. +- */ ++ if (mr_btree_insert(&share_cache->cache, &entry) < 0) + return -1; +- } + } + return 0; + } +@@ -477,26 +469,12 @@ static uint32_t + mlx5_mr_lookup_cache(struct mlx5_mr_share_cache *share_cache, + struct mr_cache_entry *entry, uintptr_t addr) + { +- uint16_t idx; +- uint32_t lkey = UINT32_MAX; +- struct mlx5_mr *mr; ++ uint32_t idx; ++ uint32_t lkey; + +- /* +- * If the global cache has overflowed since it failed to expand the +- * B-tree table, it can't have all the existing MRs. Then, the address +- * has to be searched by traversing the original MR list instead, which +- * is very slow path. Otherwise, the global cache is all inclusive. +- */ +- if (!unlikely(share_cache->cache.overflow)) { +- lkey = mr_btree_lookup(&share_cache->cache, &idx, addr); +- if (lkey != UINT32_MAX) +- *entry = (*share_cache->cache.table)[idx]; +- } else { +- /* Falling back to the slowest path. */ +- mr = mlx5_mr_lookup_list(share_cache, entry, addr); +- if (mr != NULL) +- lkey = entry->lkey; +- } ++ lkey = mr_btree_lookup(&share_cache->cache, &idx, addr); ++ if (lkey != UINT32_MAX) ++ *entry = (*share_cache->cache.table)[idx]; + MLX5_ASSERT(lkey == UINT32_MAX || (addr >= entry->start && + addr < entry->end)); + return lkey; +@@ -529,7 +507,6 @@ mlx5_mr_rebuild_cache(struct mlx5_mr_share_cache *share_cache) + DRV_LOG(DEBUG, "Rebuild dev cache[] %p", (void *)share_cache); + /* Flush cache to rebuild. */ + share_cache->cache.len = 1; +- share_cache->cache.overflow = 0; + /* Iterate all the existing MRs. */ + LIST_FOREACH(mr, &share_cache->mr_list, mr) + if (mlx5_mr_insert_cache(share_cache, mr) < 0) +@@ -585,6 +562,74 @@ mr_find_contig_memsegs_cb(const struct rte_memseg_list *msl, + return 1; + } + ++/** ++ * Get the number of virtually-contiguous chunks in the MR. ++ * HW MR does not need to be already created to use this function. ++ * ++ * @param mr ++ * Pointer to the MR. ++ * ++ * @return ++ * Number of chunks. ++ */ ++static uint32_t ++mr_get_chunk_count(const struct mlx5_mr *mr) ++{ ++ uint32_t i, count = 0; ++ bool was_in_chunk = false; ++ bool is_in_chunk; ++ ++ /* There is only one chunk in case of external memory. */ ++ if (mr->msl == NULL) ++ return 1; ++ for (i = 0; i < mr->ms_bmp_n; i++) { ++ is_in_chunk = rte_bitmap_get(mr->ms_bmp, i); ++ if (!was_in_chunk && is_in_chunk) ++ count++; ++ was_in_chunk = is_in_chunk; ++ } ++ return count; +} + -+static int32_t -+mfex_avx512_vbmi_probe(void) ++/** ++ * Thread-safely expand the global MR cache to at least @p new_size slots. ++ * ++ * @param share_cache ++ * Shared MR cache for locking. ++ * @param new_size ++ * Desired cache size. ++ * @param socket ++ * NUMA node. ++ * ++ * @return ++ * 0 in success, negative on failure and rte_errno is set. ++ */ ++int ++mlx5_mr_expand_cache(struct mlx5_mr_share_cache *share_cache, ++ uint32_t size, int socket) +{ -+ return avx512_isa_probe(true); ++ struct mlx5_mr_btree cache = {0}; ++ struct mlx5_mr_btree *bt; ++ struct mr_cache_entry *lkp_tbl; ++ int ret; ++ ++ size = rte_align32pow2(size); ++ ret = mlx5_mr_btree_init(&cache, size, socket); ++ if (ret < 0) ++ return ret; ++ rte_rwlock_write_lock(&share_cache->rwlock); ++ bt = &share_cache->cache; ++ lkp_tbl = *bt->table; ++ if (cache.size > bt->size) { ++ rte_memcpy(cache.table, lkp_tbl, bt->len * sizeof(lkp_tbl[0])); ++ RTE_SWAP(*bt, cache); ++ DRV_LOG(DEBUG, "Global MR cache expanded to %u slots", size); ++ } ++ rte_rwlock_write_unlock(&share_cache->rwlock); ++ mlx5_mr_btree_free(&cache); ++ return 0; +} -+#endif + - /* Implementations of available extract options and - * the implementations are always in order of preference. + /** + * Create a new global Memory Region (MR) for a missing virtual address. + * This API should be called on a secondary process, then a request is sent to +@@ -660,12 +705,14 @@ mlx5_mr_create_primary(void *pd, + struct mr_find_contig_memsegs_data data_re; + const struct rte_memseg_list *msl; + const struct rte_memseg *ms; ++ struct mlx5_mr_btree *bt; + struct mlx5_mr *mr = NULL; + int ms_idx_shift = -1; + uint32_t bmp_size; + void *bmp_mem; + uint32_t ms_n; + uint32_t n; ++ uint32_t chunks_n; + size_t len; + + DRV_LOG(DEBUG, "Creating a MR using address (%p)", (void *)addr); +@@ -677,6 +724,7 @@ mlx5_mr_create_primary(void *pd, + * is quite opportunistic. + */ + mlx5_mr_garbage_collect(share_cache); ++find_range: + /* + * If enabled, find out a contiguous virtual address chunk in use, to + * which the given address belongs, in order to register maximum range. +@@ -828,6 +876,33 @@ mlx5_mr_create_primary(void *pd, + len = data.end - data.start; + mr->ms_bmp_n = len / msl->page_sz; + MLX5_ASSERT(ms_idx_shift + mr->ms_bmp_n <= ms_n); ++ /* ++ * It is now known how many entries will be used in the global cache. ++ * If there is not enough, expand the cache. ++ * This cannot be done while holding the memory hotplug lock. ++ * While it is released, memory layout may change, ++ * so the process must be repeated from the beginning. ++ */ ++ bt = &share_cache->cache; ++ chunks_n = mr_get_chunk_count(mr); ++ if (bt->len + chunks_n > bt->size) { ++ struct mlx5_common_device *cdev; ++ uint32_t size; ++ ++ size = bt->size + chunks_n; ++ MLX5_ASSERT(size > bt->size); ++ cdev = container_of(share_cache, struct mlx5_common_device, ++ mr_scache); ++ rte_rwlock_write_unlock(&share_cache->rwlock); ++ rte_mcfg_mem_read_unlock(); ++ if (mlx5_mr_expand_cache(share_cache, size, ++ cdev->dev->numa_node) < 0) { ++ DRV_LOG(ERR, "Failed to expand global MR cache to %u slots", ++ size); ++ goto err_nolock; ++ } ++ goto find_range; ++ } + /* + * Finally create an MR for the memory chunk. Verbs: ibv_reg_mr() can + * be called with holding the memory lock because it doesn't use +@@ -938,7 +1013,7 @@ mr_lookup_caches(struct mlx5_mr_ctrl *mr_ctrl, + container_of(share_cache, struct mlx5_common_device, mr_scache); + struct mlx5_mr_btree *bt = &mr_ctrl->cache_bh; + uint32_t lkey; +- uint16_t idx; ++ uint32_t idx; + + /* If local cache table is full, try to double it. */ + if (unlikely(bt->len == bt->size)) +@@ -989,7 +1064,7 @@ static uint32_t + mlx5_mr_addr2mr_bh(struct mlx5_mr_ctrl *mr_ctrl, uintptr_t addr) + { + uint32_t lkey; +- uint16_t bh_idx = 0; ++ uint32_t bh_idx = 0; + /* Victim in top-half cache to replace with new entry. */ + struct mr_cache_entry *repl = &mr_ctrl->cache[mr_ctrl->head]; + +@@ -1086,7 +1161,6 @@ mlx5_mr_flush_local_cache(struct mlx5_mr_ctrl *mr_ctrl) + memset(mr_ctrl->cache, 0, sizeof(mr_ctrl->cache)); + /* Reset the B-tree table. */ + mr_ctrl->cache_bh.len = 1; +- mr_ctrl->cache_bh.overflow = 0; + /* Update the generation number. */ + mr_ctrl->cur_gen = *mr_ctrl->dev_gen_ptr; + DRV_LOG(DEBUG, "mr_ctrl(%p): flushed, cur_gen=%d", +@@ -1290,11 +1364,12 @@ mlx5_range_from_mempool_chunk(struct rte_mempool *mp, void *opaque, + unsigned int idx) + { + struct mlx5_range *ranges = opaque, *range = &ranges[idx]; ++ uintptr_t start = (uintptr_t)memhdr->addr; + uint64_t page_size = rte_mem_page_size(); + + RTE_SET_USED(mp); +- range->start = RTE_ALIGN_FLOOR((uintptr_t)memhdr->addr, page_size); +- range->end = RTE_ALIGN_CEIL(range->start + memhdr->len, page_size); ++ range->start = RTE_ALIGN_FLOOR(start, page_size); ++ range->end = RTE_ALIGN_CEIL(start + memhdr->len, page_size); + } + + /** +@@ -1541,7 +1616,7 @@ mlx5_mempool_reg_create(struct rte_mempool *mp, unsigned int mrs_n, + * Destroy a mempool registration object. + * + * @param standalone +- * Whether @p mpr owns its MRs excludively, i.e. they are not shared. ++ * Whether @p mpr owns its MRs exclusively, i.e. they are not shared. + */ + static void + mlx5_mempool_reg_destroy(struct mlx5_mr_share_cache *share_cache, +@@ -1834,12 +1909,13 @@ mlx5_mempool_reg_addr2mr(struct mlx5_mempool_reg *mpr, uintptr_t addr, + + for (i = 0; i < mpr->mrs_n; i++) { + const struct mlx5_pmd_mr *mr = &mpr->mrs[i].pmd_mr; +- uintptr_t mr_addr = (uintptr_t)mr->addr; ++ uintptr_t mr_start = (uintptr_t)mr->addr; ++ uintptr_t mr_end = mr_start + mr->len; + +- if (mr_addr <= addr) { ++ if (mr_start <= addr && addr < mr_end) { + lkey = rte_cpu_to_be_32(mr->lkey); +- entry->start = mr_addr; +- entry->end = mr_addr + mr->len; ++ entry->start = mr_start; ++ entry->end = mr_end; + entry->lkey = lkey; + break; + } +@@ -1932,7 +2008,7 @@ mlx5_mr_mempool_populate_cache(struct mlx5_mr_ctrl *mr_ctrl, + struct mlx5_mempool_mr *mr = &mpr->mrs[i]; + struct mr_cache_entry entry; + uint32_t lkey; +- uint16_t idx; ++ uint32_t idx; + + lkey = mr_btree_lookup(bt, &idx, (uintptr_t)mr->pmd_mr.addr); + if (lkey != UINT32_MAX) +@@ -1970,7 +2046,7 @@ mlx5_mr_mempool2mr_bh(struct mlx5_mr_ctrl *mr_ctrl, + { + struct mr_cache_entry *repl = &mr_ctrl->cache[mr_ctrl->head]; + uint32_t lkey; +- uint16_t bh_idx = 0; ++ uint32_t bh_idx = 0; + + /* Binary-search MR translation table. */ + lkey = mr_btree_lookup(&mr_ctrl->cache_bh, &bh_idx, addr); +diff --git a/dpdk/drivers/common/mlx5/mlx5_common_mr.h b/dpdk/drivers/common/mlx5/mlx5_common_mr.h +index cf384b6748..213f5427cb 100644 +--- a/dpdk/drivers/common/mlx5/mlx5_common_mr.h ++++ b/dpdk/drivers/common/mlx5/mlx5_common_mr.h +@@ -56,9 +56,8 @@ struct mr_cache_entry { + + /* MR Cache table for Binary search. */ + struct mlx5_mr_btree { +- uint16_t len; /* Number of entries. */ +- uint16_t size; /* Total number of entries. */ +- int overflow; /* Mark failure of table expansion. */ ++ uint32_t len; /* Number of entries. */ ++ uint32_t size; /* Total number of entries. */ + struct mr_cache_entry (*table)[]; + } __rte_packed; + +@@ -218,6 +217,8 @@ void mlx5_mr_btree_dump(struct mlx5_mr_btree *bt __rte_unused); + __rte_internal + uint32_t mlx5_mr_mempool2mr_bh(struct mlx5_mr_ctrl *mr_ctrl, + struct rte_mempool *mp, uintptr_t addr); ++int mlx5_mr_expand_cache(struct mlx5_mr_share_cache *share_cache, ++ uint32_t new_size, int socket); + void mlx5_mr_release_cache(struct mlx5_mr_share_cache *mr_cache); + int mlx5_mr_create_cache(struct mlx5_mr_share_cache *share_cache, int socket); + void mlx5_mr_dump_cache(struct mlx5_mr_share_cache *share_cache __rte_unused); +diff --git a/dpdk/drivers/common/mlx5/mlx5_common_utils.c b/dpdk/drivers/common/mlx5/mlx5_common_utils.c +index 775fabd478..58d744b4d4 100644 +--- a/dpdk/drivers/common/mlx5/mlx5_common_utils.c ++++ b/dpdk/drivers/common/mlx5/mlx5_common_utils.c +@@ -293,11 +293,9 @@ _mlx5_list_unregister(struct mlx5_list_inconst *l_inconst, + l_const->cb_clone_free(l_const->ctx, entry); + else + l_const->cb_remove(l_const->ctx, entry); +- } else if (likely(lcore_idx != -1)) { ++ } else { + __atomic_add_fetch(&l_inconst->cache[entry->lcore_idx]->inv_cnt, + 1, __ATOMIC_RELAXED); +- } else { +- return 0; + } + if (!l_const->lcores_share) { + __atomic_sub_fetch(&l_inconst->count, 1, __ATOMIC_RELAXED); +diff --git a/dpdk/drivers/common/mlx5/mlx5_devx_cmds.c b/dpdk/drivers/common/mlx5/mlx5_devx_cmds.c +index e52b995ee3..70a430f134 100644 +--- a/dpdk/drivers/common/mlx5/mlx5_devx_cmds.c ++++ b/dpdk/drivers/common/mlx5/mlx5_devx_cmds.c +@@ -823,6 +823,7 @@ mlx5_devx_cmd_query_hca_attr(void *ctx, + { + uint32_t in[MLX5_ST_SZ_DW(query_hca_cap_in)] = {0}; + uint32_t out[MLX5_ST_SZ_DW(query_hca_cap_out)] = {0}; ++ bool hca_cap_2_sup; + uint64_t general_obj_types_supported = 0; + void *hcattr; + int rc, i; +@@ -832,6 +833,7 @@ mlx5_devx_cmd_query_hca_attr(void *ctx, + MLX5_HCA_CAP_OPMOD_GET_CUR); + if (!hcattr) + return rc; ++ hca_cap_2_sup = MLX5_GET(cmd_hca_cap, hcattr, hca_cap_2); + attr->max_wqe_sz_sq = MLX5_GET(cmd_hca_cap, hcattr, max_wqe_sz_sq); + attr->flow_counter_bulk_alloc_bitmap = + MLX5_GET(cmd_hca_cap, hcattr, flow_counter_bulk_alloc); +@@ -967,6 +969,20 @@ mlx5_devx_cmd_query_hca_attr(void *ctx, + general_obj_types) & + MLX5_GENERAL_OBJ_TYPES_CAP_CONN_TRACK_OFFLOAD); + attr->rq_delay_drop = MLX5_GET(cmd_hca_cap, hcattr, rq_delay_drop); ++ if (hca_cap_2_sup) { ++ hcattr = mlx5_devx_get_hca_cap(ctx, in, out, &rc, ++ MLX5_GET_HCA_CAP_OP_MOD_GENERAL_DEVICE_2 | ++ MLX5_HCA_CAP_OPMOD_GET_CUR); ++ if (!hcattr) { ++ DRV_LOG(DEBUG, ++ "Failed to query DevX HCA capabilities 2."); ++ return rc; ++ } ++ attr->log_min_stride_wqe_sz = MLX5_GET(cmd_hca_cap_2, hcattr, ++ log_min_stride_wqe_sz); ++ } ++ if (attr->log_min_stride_wqe_sz == 0) ++ attr->log_min_stride_wqe_sz = MLX5_MPRQ_LOG_MIN_STRIDE_WQE_SIZE; + if (attr->qos.sup) { + hcattr = mlx5_devx_get_hca_cap(ctx, in, out, &rc, + MLX5_GET_HCA_CAP_OP_MOD_QOS_CAP | +@@ -1114,6 +1130,18 @@ mlx5_devx_cmd_query_hca_attr(void *ctx, + goto error; + } + } ++ if (attr->eswitch_manager) { ++ hcattr = mlx5_devx_get_hca_cap(ctx, in, out, &rc, ++ MLX5_SET_HCA_CAP_OP_MOD_ESW | ++ MLX5_HCA_CAP_OPMOD_GET_CUR); ++ if (!hcattr) ++ return rc; ++ attr->esw_mgr_vport_id_valid = ++ MLX5_GET(esw_cap, hcattr, ++ esw_manager_vport_number_valid); ++ attr->esw_mgr_vport_id = ++ MLX5_GET(esw_cap, hcattr, esw_manager_vport_number); ++ } + return 0; + error: + rc = (rc > 0) ? -rc : rc; +@@ -1822,7 +1850,7 @@ mlx5_devx_cmd_create_td(void *ctx) + * Pointer to file stream. + * + * @return +- * 0 on success, a nagative value otherwise. ++ * 0 on success, a negative value otherwise. */ -diff --git a/lib/dpif-netdev-private-extract.h b/lib/dpif-netdev-private-extract.h -index f9a757ba41..3e06148c5a 100644 ---- a/lib/dpif-netdev-private-extract.h -+++ b/lib/dpif-netdev-private-extract.h -@@ -176,10 +176,8 @@ mfex_study_traffic(struct dp_packet_batch *packets, int - mfex_set_study_pkt_cnt(uint32_t pkt_cmp_count, const char *name); + mlx5_devx_cmd_flow_dump(void *fdb_domain __rte_unused, +@@ -2263,7 +2291,7 @@ mlx5_devx_cmd_modify_qp_state(struct mlx5_devx_obj *qp, uint32_t qp_st_mod_op, + case MLX5_CMD_OP_RTR2RTS_QP: + qpc = MLX5_ADDR_OF(rtr2rts_qp_in, &in, qpc); + MLX5_SET(rtr2rts_qp_in, &in, qpn, qp->id); +- MLX5_SET(qpc, qpc, primary_address_path.ack_timeout, 14); ++ MLX5_SET(qpc, qpc, primary_address_path.ack_timeout, 16); + MLX5_SET(qpc, qpc, log_ack_req_freq, 0); + MLX5_SET(qpc, qpc, retry_count, 7); + MLX5_SET(qpc, qpc, rnr_retry, 7); +diff --git a/dpdk/drivers/common/mlx5/mlx5_devx_cmds.h b/dpdk/drivers/common/mlx5/mlx5_devx_cmds.h +index d7f71646a3..4373761c29 100644 +--- a/dpdk/drivers/common/mlx5/mlx5_devx_cmds.h ++++ b/dpdk/drivers/common/mlx5/mlx5_devx_cmds.h +@@ -251,6 +251,9 @@ struct mlx5_hca_attr { + uint32_t log_max_mmo_decompress:5; + uint32_t umr_modify_entity_size_disabled:1; + uint32_t umr_indirect_mkey_disabled:1; ++ uint32_t log_min_stride_wqe_sz:5; ++ uint32_t esw_mgr_vport_id_valid:1; /* E-Switch Mgr vport ID is valid. */ ++ uint16_t esw_mgr_vport_id; /* E-Switch Mgr vport ID . */ + uint16_t max_wqe_sz_sq; + }; --/* AVX512 MFEX Probe and Implementations functions. */ -+/* AVX512 MFEX Implementation functions. */ - #ifdef __x86_64__ --int32_t mfex_avx512_probe(void); --int32_t mfex_avx512_vbmi_probe(void); +diff --git a/dpdk/drivers/common/mlx5/mlx5_malloc.c b/dpdk/drivers/common/mlx5/mlx5_malloc.c +index b19501e1bc..cef3b88e11 100644 +--- a/dpdk/drivers/common/mlx5/mlx5_malloc.c ++++ b/dpdk/drivers/common/mlx5/mlx5_malloc.c +@@ -58,7 +58,7 @@ static struct mlx5_sys_mem mlx5_sys_mem = { + * Check if the address belongs to memory seg list. + * + * @param addr +- * Memory address to be ckeced. ++ * Memory address to be checked. + * @param msl + * Memory seg list. + * +@@ -109,7 +109,7 @@ mlx5_mem_update_msl(void *addr) + * Check if the address belongs to rte memory. + * + * @param addr +- * Memory address to be ckeced. ++ * Memory address to be checked. + * + * @return + * True if it belongs, false otherwise. +diff --git a/dpdk/drivers/common/mlx5/mlx5_malloc.h b/dpdk/drivers/common/mlx5/mlx5_malloc.h +index 74b7eeb26e..92149f7b92 100644 +--- a/dpdk/drivers/common/mlx5/mlx5_malloc.h ++++ b/dpdk/drivers/common/mlx5/mlx5_malloc.h +@@ -19,7 +19,7 @@ extern "C" { + + enum mlx5_mem_flags { + MLX5_MEM_ANY = 0, +- /* Memory will be allocated dpends on sys_mem_en. */ ++ /* Memory will be allocated depends on sys_mem_en. */ + MLX5_MEM_SYS = 1 << 0, + /* Memory should be allocated from system. */ + MLX5_MEM_RTE = 1 << 1, +diff --git a/dpdk/drivers/common/mlx5/mlx5_prm.h b/dpdk/drivers/common/mlx5/mlx5_prm.h +index 2ded67e85e..58aa72df64 100644 +--- a/dpdk/drivers/common/mlx5/mlx5_prm.h ++++ b/dpdk/drivers/common/mlx5/mlx5_prm.h +@@ -249,6 +249,9 @@ + /* Maximum number of DS in WQE. Limited by 6-bit field. */ + #define MLX5_DSEG_MAX 63 + ++/* The 32 bit syndrome offset in struct mlx5_err_cqe. */ ++#define MLX5_ERROR_CQE_SYNDROME_OFFSET 52 ++ + /* The completion mode offset in the WQE control segment line 2. */ + #define MLX5_COMP_MODE_OFFSET 2 + +@@ -264,6 +267,9 @@ + /* The maximum log value of segments per RQ WQE. */ + #define MLX5_MAX_LOG_RQ_SEGS 5u + ++/* Log 2 of the default size of a WQE for Multi-Packet RQ. */ ++#define MLX5_MPRQ_LOG_MIN_STRIDE_WQE_SIZE 14U ++ + /* The alignment needed for WQ buffer. */ + #define MLX5_WQE_BUF_ALIGNMENT rte_mem_page_size() + +@@ -556,6 +562,8 @@ struct mlx5_rdma_write_wqe { + #define MLX5_GGA_COMP_LOG_BLOCK_SIZE_MAX 15u + #define MLX5_GGA_COMP_LOG_DYNAMIC_SIZE_MAX 15u + #define MLX5_GGA_COMP_LOG_DYNAMIC_SIZE_MIN 0u ++#define MLX5_GGA_COMP_OUT_OF_SPACE_SYNDROME_BE 0x29D0084 ++#define MLX5_GGA_COMP_MISSING_BFINAL_SYNDROME_BE 0x29D0011 + + struct mlx5_wqe_metadata_seg { + uint32_t mmo_control_31_0; /* mmo_control_63_32 is in ctrl_seg.imm */ +@@ -1261,6 +1269,7 @@ enum { + MLX5_GET_HCA_CAP_OP_MOD_QOS_CAP = 0xc << 1, + MLX5_GET_HCA_CAP_OP_MOD_ROCE = 0x4 << 1, + MLX5_GET_HCA_CAP_OP_MOD_NIC_FLOW_TABLE = 0x7 << 1, ++ MLX5_SET_HCA_CAP_OP_MOD_ESW = 0x9 << 1, + MLX5_GET_HCA_CAP_OP_MOD_VDPA_EMULATION = 0x13 << 1, + MLX5_GET_HCA_CAP_OP_MOD_PARSE_GRAPH_NODE_CAP = 0x1C << 1, + MLX5_GET_HCA_CAP_OP_MOD_GENERAL_DEVICE_2 = 0x20 << 1, +@@ -1342,7 +1351,9 @@ enum { + #define MLX5_STEERING_LOGIC_FORMAT_CONNECTX_6DX 0x1 + + struct mlx5_ifc_cmd_hca_cap_bits { +- u8 reserved_at_0[0x30]; ++ u8 reserved_at_0[0x20]; ++ u8 hca_cap_2[0x1]; ++ u8 reserved_at_21[0xf]; + u8 vhca_id[0x10]; + u8 reserved_at_40[0x20]; + u8 reserved_at_60[0x3]; +@@ -1909,7 +1920,8 @@ struct mlx5_ifc_cmd_hca_cap_2_bits { + u8 max_reformat_insert_offset[0x8]; + u8 max_reformat_remove_size[0x8]; + u8 max_reformat_remove_offset[0x8]; /* End of DW6. */ +- u8 aso_conntrack_reg_id[0x8]; ++ u8 reserved_at_c0[0x3]; ++ u8 log_min_stride_wqe_sz[0x5]; + u8 reserved_at_c8[0x3]; + u8 log_conn_track_granularity[0x5]; + u8 reserved_at_d0[0x3]; +@@ -1920,13 +1932,25 @@ struct mlx5_ifc_cmd_hca_cap_2_bits { + u8 reserved_at_100[0x700]; + }; - #define DECLARE_AVX512_MFEX_PROTOTYPE(name) \ - uint32_t \ -diff --git a/lib/dpif-netdev-private-flow.h b/lib/dpif-netdev-private-flow.h -index 66016eb099..7425dd44e7 100644 ---- a/lib/dpif-netdev-private-flow.h -+++ b/lib/dpif-netdev-private-flow.h -@@ -104,6 +104,7 @@ struct dp_netdev_flow { - bool dead; - uint32_t mark; /* Unique flow mark for netdev offloading. */ - uint64_t simple_match_mark; /* Unique flow mark for the simple match. */ -+ odp_port_t orig_in_port; ++struct mlx5_ifc_esw_cap_bits { ++ u8 reserved_at_0[0x60]; ++ ++ u8 esw_manager_vport_number_valid[0x1]; ++ u8 reserved_at_61[0xf]; ++ u8 esw_manager_vport_number[0x10]; ++ ++ u8 reserved_at_80[0x780]; ++}; ++ + union mlx5_ifc_hca_cap_union_bits { + struct mlx5_ifc_cmd_hca_cap_bits cmd_hca_cap; ++ struct mlx5_ifc_cmd_hca_cap_2_bits cmd_hca_cap_2; + struct mlx5_ifc_per_protocol_networking_offload_caps_bits + per_protocol_networking_offload_caps; + struct mlx5_ifc_qos_cap_bits qos_cap; + struct mlx5_ifc_virtio_emulation_cap_bits vdpa_caps; + struct mlx5_ifc_flow_table_nic_cap_bits flow_table_nic_cap; ++ struct mlx5_ifc_esw_cap_bits esw_cap; + struct mlx5_ifc_roce_caps_bits roce_caps; + u8 reserved_at_0[0x8000]; + }; +@@ -4172,7 +4196,7 @@ mlx5_flow_mark_get(uint32_t val) + * timestamp format supported by the queue. + * + * @return +- * Converted timstamp format settings. ++ * Converted timestamp format settings. + */ + static inline uint32_t + mlx5_ts_format_conv(uint32_t ts_format) +diff --git a/dpdk/drivers/common/mlx5/version.map b/dpdk/drivers/common/mlx5/version.map +index 34e86004a0..130d47ad8c 100644 +--- a/dpdk/drivers/common/mlx5/version.map ++++ b/dpdk/drivers/common/mlx5/version.map +@@ -79,6 +79,7 @@ INTERNAL { + + mlx5_free; + ++ mlx5_get_device_guid; # WINDOWS_NO_EXPORT + mlx5_get_ifname_sysfs; # WINDOWS_NO_EXPORT + mlx5_get_pci_addr; # WINDOWS_NO_EXPORT + +@@ -123,9 +124,11 @@ INTERNAL { + mlx5_nl_mac_addr_flush; # WINDOWS_NO_EXPORT + mlx5_nl_mac_addr_remove; # WINDOWS_NO_EXPORT + mlx5_nl_mac_addr_sync; # WINDOWS_NO_EXPORT ++ mlx5_nl_parse_link_status_update; # WINDOWS_NO_EXPORT + mlx5_nl_port_state; # WINDOWS_NO_EXPORT + mlx5_nl_portnum; # WINDOWS_NO_EXPORT + mlx5_nl_promisc; # WINDOWS_NO_EXPORT ++ mlx5_nl_read_events; # WINDOWS_NO_EXPORT + mlx5_nl_switch_info; # WINDOWS_NO_EXPORT + mlx5_nl_vf_mac_addr_modify; # WINDOWS_NO_EXPORT + mlx5_nl_vlan_vmwa_create; # WINDOWS_NO_EXPORT +@@ -146,4 +149,6 @@ INTERNAL { + mlx5_mp_req_mempool_reg; + mlx5_mr_mempool2mr_bh; + mlx5_mr_mempool_populate_cache; ++ ++ local: *; + }; +diff --git a/dpdk/drivers/common/mlx5/windows/meson.build b/dpdk/drivers/common/mlx5/windows/meson.build +index 980f76b11c..edbbaa9ae1 100644 +--- a/dpdk/drivers/common/mlx5/windows/meson.build ++++ b/dpdk/drivers/common/mlx5/windows/meson.build +@@ -8,8 +8,8 @@ sources += files( + 'mlx5_common_os.c', + ) - /* Statistics. */ - struct dp_netdev_flow_stats stats; -diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c -index 9f35713ef5..3d9d8929f7 100644 ---- a/lib/dpif-netdev.c -+++ b/lib/dpif-netdev.c -@@ -93,7 +93,8 @@ VLOG_DEFINE_THIS_MODULE(dpif_netdev); - /* Auto Load Balancing Defaults */ - #define ALB_IMPROVEMENT_THRESHOLD 25 - #define ALB_LOAD_THRESHOLD 95 --#define ALB_REBALANCE_INTERVAL 1 /* 1 Min */ -+#define ALB_REBALANCE_INTERVAL 1 /* 1 Min */ -+#define MAX_ALB_REBALANCE_INTERVAL 20000 /* 20000 Min */ - #define MIN_TO_MSEC 60000 +-res_lib = run_command(python3, '-c', 'import os; print(os.environ["DEVX_LIB_PATH"])') +-res_inc = run_command(python3, '-c', 'import os; print(os.environ["DEVX_INC_PATH"])') ++res_lib = run_command(python3, '-c', 'import os; print(os.environ["DEVX_LIB_PATH"])', check: false) ++res_inc = run_command(python3, '-c', 'import os; print(os.environ["DEVX_INC_PATH"])', check: false) + + if (res_lib.returncode() != 0 or res_inc.returncode() != 0) + build = false +diff --git a/dpdk/drivers/common/mlx5/windows/mlx5_common_os.c b/dpdk/drivers/common/mlx5/windows/mlx5_common_os.c +index 162c7476cc..c3cfc315f2 100644 +--- a/dpdk/drivers/common/mlx5/windows/mlx5_common_os.c ++++ b/dpdk/drivers/common/mlx5/windows/mlx5_common_os.c +@@ -302,7 +302,7 @@ mlx5_os_umem_dereg(void *pumem) + } - #define FLOW_DUMP_MAX_BATCH 50 -@@ -1932,13 +1933,13 @@ static void - dp_netdev_free(struct dp_netdev *dp) - OVS_REQUIRES(dp_netdev_mutex) - { -- struct dp_netdev_port *port, *next; -+ struct dp_netdev_port *port; - struct tx_bond *bond; + /** +- * Register mr. Given protection doamin pointer, pointer to addr and length ++ * Register mr. Given protection domain pointer, pointer to addr and length + * register the memory region. + * + * @param[in] pd +@@ -310,7 +310,7 @@ mlx5_os_umem_dereg(void *pumem) + * @param[in] addr + * Pointer to memory start address (type devx_device_ctx). + * @param[in] length +- * Lengtoh of the memory to register. ++ * Length of the memory to register. + * @param[out] pmd_mr + * pmd_mr struct set with lkey, address, length, pointer to mr object, mkey + * +diff --git a/dpdk/drivers/common/mlx5/windows/mlx5_common_os.h b/dpdk/drivers/common/mlx5/windows/mlx5_common_os.h +index 3afce56cd9..61fc8dd761 100644 +--- a/dpdk/drivers/common/mlx5/windows/mlx5_common_os.h ++++ b/dpdk/drivers/common/mlx5/windows/mlx5_common_os.h +@@ -21,7 +21,7 @@ + /** + * This API allocates aligned or non-aligned memory. The free can be on either + * aligned or nonaligned memory. To be protected - even though there may be no +- * alignment - in Windows this API will unconditioanlly call _aligned_malloc() ++ * alignment - in Windows this API will unconditionally call _aligned_malloc() + * with at least a minimal alignment size. + * + * @param[in] align +diff --git a/dpdk/drivers/common/qat/qat_adf/adf_transport_access_macros.h b/dpdk/drivers/common/qat/qat_adf/adf_transport_access_macros.h +index a6d403fac3..12a7258c60 100644 +--- a/dpdk/drivers/common/qat/qat_adf/adf_transport_access_macros.h ++++ b/dpdk/drivers/common/qat/qat_adf/adf_transport_access_macros.h +@@ -72,7 +72,7 @@ + #define ADF_SIZE_TO_RING_SIZE_IN_BYTES(SIZE) ((1 << (SIZE - 1)) << 7) + #define ADF_RING_SIZE_IN_BYTES_TO_SIZE(SIZE) ((1 << (SIZE - 1)) >> 7) + +-/* Minimum ring bufer size for memory allocation */ ++/* Minimum ring buffer size for memory allocation */ + #define ADF_RING_SIZE_BYTES_MIN(SIZE) ((SIZE < ADF_RING_SIZE_4K) ? \ + ADF_RING_SIZE_4K : SIZE) + #define ADF_RING_SIZE_MODULO(SIZE) (SIZE + 0x6) +diff --git a/dpdk/drivers/common/sfc_efx/base/efx_impl.h b/dpdk/drivers/common/sfc_efx/base/efx_impl.h +index e2802e6672..ba00eeeb47 100644 +--- a/dpdk/drivers/common/sfc_efx/base/efx_impl.h ++++ b/dpdk/drivers/common/sfc_efx/base/efx_impl.h +@@ -1555,6 +1555,12 @@ efx_mcdi_intf_from_pcie( + __in uint32_t pcie_intf, + __out efx_pcie_interface_t *efx_intf); + ++LIBEFX_INTERNAL ++extern __checkReturn efx_rc_t ++efx_mcdi_intf_to_pcie( ++ __in efx_pcie_interface_t efx_intf, ++ __out uint32_t *pcie_intf); ++ + LIBEFX_INTERNAL + extern __checkReturn efx_rc_t + efx_mcdi_init_evq( +diff --git a/dpdk/drivers/common/sfc_efx/base/efx_mae.c b/dpdk/drivers/common/sfc_efx/base/efx_mae.c +index 7b24e3fee4..31f51b5548 100644 +--- a/dpdk/drivers/common/sfc_efx/base/efx_mae.c ++++ b/dpdk/drivers/common/sfc_efx/base/efx_mae.c +@@ -1027,6 +1027,10 @@ efx_mae_match_spec_field_set( + memcpy(mvp + descp->emmd_value_offset, + &dword, sizeof (dword)); + break; ++ case 1: ++ memcpy(mvp + descp->emmd_value_offset, ++ value, 1); ++ break; + default: + EFSYS_ASSERT(B_FALSE); + } +@@ -1039,6 +1043,10 @@ efx_mae_match_spec_field_set( + memcpy(mvp + descp->emmd_mask_offset, + &dword, sizeof (dword)); + break; ++ case 1: ++ memcpy(mvp + descp->emmd_mask_offset, ++ mask, 1); ++ break; + default: + EFSYS_ASSERT(B_FALSE); + } +@@ -2242,7 +2250,8 @@ efx_mae_outer_rule_insert( + memcpy(payload + offset, spec->emms_mask_value_pairs.outer, + MAE_ENC_FIELD_PAIRS_LEN); + +- MCDI_IN_SET_BYTE(req, MAE_OUTER_RULE_INSERT_IN_RECIRC_ID, ++ MCDI_IN_SET_DWORD_FIELD(req, MAE_OUTER_RULE_INSERT_IN_LOOKUP_CONTROL, ++ MAE_OUTER_RULE_INSERT_IN_RECIRC_ID, + spec->emms_outer_rule_recirc_id); + + efx_mcdi_execute(enp, &req); +diff --git a/dpdk/drivers/common/sfc_efx/base/efx_mcdi.c b/dpdk/drivers/common/sfc_efx/base/efx_mcdi.c +index 9189a7a8b3..404ca23d58 100644 +--- a/dpdk/drivers/common/sfc_efx/base/efx_mcdi.c ++++ b/dpdk/drivers/common/sfc_efx/base/efx_mcdi.c +@@ -659,6 +659,7 @@ efx_mcdi_get_client_handle( + EFX_MCDI_DECLARE_BUF(payload, + MC_CMD_GET_CLIENT_HANDLE_IN_LEN, + MC_CMD_GET_CLIENT_HANDLE_OUT_LEN); ++ uint32_t pcie_intf; + efx_rc_t rc; + + if (handle == NULL) { +@@ -666,6 +667,10 @@ efx_mcdi_get_client_handle( + goto fail1; + } - shash_find_and_delete(&dp_netdevs, dp->name); ++ rc = efx_mcdi_intf_to_pcie(intf, &pcie_intf); ++ if (rc != 0) ++ goto fail2; ++ + req.emr_cmd = MC_CMD_GET_CLIENT_HANDLE; + req.emr_in_buf = payload; + req.emr_in_length = MC_CMD_GET_CLIENT_HANDLE_IN_LEN; +@@ -676,23 +681,25 @@ efx_mcdi_get_client_handle( + MC_CMD_GET_CLIENT_HANDLE_IN_TYPE_FUNC); + MCDI_IN_SET_WORD(req, GET_CLIENT_HANDLE_IN_FUNC_PF, pf); + MCDI_IN_SET_WORD(req, GET_CLIENT_HANDLE_IN_FUNC_VF, vf); +- MCDI_IN_SET_DWORD(req, GET_CLIENT_HANDLE_IN_FUNC_INTF, intf); ++ MCDI_IN_SET_DWORD(req, GET_CLIENT_HANDLE_IN_FUNC_INTF, pcie_intf); + + efx_mcdi_execute(enp, &req); + + if (req.emr_rc != 0) { + rc = req.emr_rc; +- goto fail2; ++ goto fail3; + } - ovs_rwlock_wrlock(&dp->port_rwlock); -- HMAP_FOR_EACH_SAFE (port, next, node, &dp->ports) { -+ HMAP_FOR_EACH_SAFE (port, node, &dp->ports) { - do_del_port(dp, port); - } - ovs_rwlock_unlock(&dp->port_rwlock); -@@ -3006,7 +3007,7 @@ static void - queue_netdev_flow_put(struct dp_netdev_pmd_thread *pmd, - struct dp_netdev_flow *flow, struct match *match, - const struct nlattr *actions, size_t actions_len, -- odp_port_t orig_in_port, int op) -+ int op) + if (req.emr_out_length_used < MC_CMD_GET_CLIENT_HANDLE_OUT_LEN) { + rc = EMSGSIZE; +- goto fail3; ++ goto fail4; + } + + *handle = MCDI_OUT_DWORD(req, GET_CLIENT_HANDLE_OUT_HANDLE); + + return 0; ++fail4: ++ EFSYS_PROBE(fail4); + fail3: + EFSYS_PROBE(fail3); + fail2: +@@ -709,7 +716,7 @@ efx_mcdi_get_own_client_handle( { - struct dp_offload_thread_item *item; - struct dp_offload_flow_item *flow_offload; -@@ -3021,7 +3022,7 @@ queue_netdev_flow_put(struct dp_netdev_pmd_thread *pmd, - flow_offload->actions = xmalloc(actions_len); - memcpy(flow_offload->actions, actions, actions_len); - flow_offload->actions_len = actions_len; -- flow_offload->orig_in_port = orig_in_port; -+ flow_offload->orig_in_port = flow->orig_in_port; + efx_rc_t rc; + +- rc = efx_mcdi_get_client_handle(enp, PCIE_INTERFACE_CALLER, ++ rc = efx_mcdi_get_client_handle(enp, EFX_PCIE_INTERFACE_CALLER, + PCIE_FUNCTION_PF_NULL, PCIE_FUNCTION_VF_NULL, handle); + if (rc != 0) + goto fail1; +@@ -2233,6 +2240,35 @@ efx_mcdi_intf_from_pcie( + return (rc); + } - item->timestamp = pmd->ctx.now; - dp_netdev_offload_flow_enqueue(item); -@@ -4095,6 +4096,7 @@ dp_netdev_flow_add(struct dp_netdev_pmd_thread *pmd, - flow->dead = false; - flow->batch = NULL; - flow->mark = INVALID_FLOW_MARK; -+ flow->orig_in_port = orig_in_port; - *CONST_CAST(unsigned *, &flow->pmd_id) = pmd->core_id; - *CONST_CAST(struct flow *, &flow->flow) = match->flow; - *CONST_CAST(ovs_u128 *, &flow->ufid) = *ufid; -@@ -4129,7 +4131,7 @@ dp_netdev_flow_add(struct dp_netdev_pmd_thread *pmd, - } ++ __checkReturn efx_rc_t ++efx_mcdi_intf_to_pcie( ++ __in efx_pcie_interface_t efx_intf, ++ __out uint32_t *pcie_intf) ++{ ++ efx_rc_t rc; ++ ++ switch (efx_intf) { ++ case EFX_PCIE_INTERFACE_CALLER: ++ *pcie_intf = PCIE_INTERFACE_CALLER; ++ break; ++ case EFX_PCIE_INTERFACE_HOST_PRIMARY: ++ *pcie_intf = PCIE_INTERFACE_HOST_PRIMARY; ++ break; ++ case EFX_PCIE_INTERFACE_NIC_EMBEDDED: ++ *pcie_intf = PCIE_INTERFACE_NIC_EMBEDDED; ++ break; ++ default: ++ rc = EINVAL; ++ goto fail1; ++ } ++ ++ return (0); ++ ++fail1: ++ EFSYS_PROBE1(fail1, efx_rc_t, rc); ++ return (rc); ++} ++ + /* + * This function returns the pf and vf number of a function. If it is a pf the + * vf number is 0xffff. The vf number is the index of the vf on that +diff --git a/dpdk/drivers/common/sfc_efx/efsys.h b/dpdk/drivers/common/sfc_efx/efsys.h +index 3860c2835a..224254bee7 100644 +--- a/dpdk/drivers/common/sfc_efx/efsys.h ++++ b/dpdk/drivers/common/sfc_efx/efsys.h +@@ -616,7 +616,7 @@ typedef struct efsys_bar_s { + + #define EFSYS_DMA_SYNC_FOR_KERNEL(_esmp, _offset, _size) ((void)0) + +-/* Just avoid store and compiler (impliciltly) reordering */ ++/* Just avoid store and compiler (implicitly) reordering */ + #define EFSYS_DMA_SYNC_FOR_DEVICE(_esmp, _offset, _size) rte_wmb() + + /* TIMESTAMP */ +diff --git a/dpdk/drivers/compress/mlx5/mlx5_compress.c b/dpdk/drivers/compress/mlx5/mlx5_compress.c +index 82b871bd86..a18ec8a6cf 100644 +--- a/dpdk/drivers/compress/mlx5/mlx5_compress.c ++++ b/dpdk/drivers/compress/mlx5/mlx5_compress.c +@@ -563,7 +563,18 @@ mlx5_compress_cqe_err_handle(struct mlx5_compress_qp *qp, + qp->qp.wqes; + volatile struct mlx5_gga_compress_opaque *opaq = qp->opaque_mr.addr; + +- op->status = RTE_COMP_OP_STATUS_ERROR; ++ volatile uint32_t *synd_word = RTE_PTR_ADD(cqe, MLX5_ERROR_CQE_SYNDROME_OFFSET); ++ switch (*synd_word) { ++ case MLX5_GGA_COMP_OUT_OF_SPACE_SYNDROME_BE: ++ op->status = RTE_COMP_OP_STATUS_OUT_OF_SPACE_TERMINATED; ++ DRV_LOG(DEBUG, "OUT OF SPACE error, output is bigger than dst buffer."); ++ break; ++ case MLX5_GGA_COMP_MISSING_BFINAL_SYNDROME_BE: ++ DRV_LOG(DEBUG, "The last compressed block missed the B-final flag; maybe the compressed data is not complete or garbaged?"); ++ /* fallthrough */ ++ default: ++ op->status = RTE_COMP_OP_STATUS_ERROR; ++ } + op->consumed = 0; + op->produced = 0; + op->output_chksum = 0; +diff --git a/dpdk/drivers/compress/octeontx/include/zip_regs.h b/dpdk/drivers/compress/octeontx/include/zip_regs.h +index 96e538bb75..94a48cde66 100644 +--- a/dpdk/drivers/compress/octeontx/include/zip_regs.h ++++ b/dpdk/drivers/compress/octeontx/include/zip_regs.h +@@ -195,7 +195,7 @@ union zip_inst_s { + uint64_t bf : 1; + /** Comp/decomp operation */ + uint64_t op : 2; +- /** Data sactter */ ++ /** Data scatter */ + uint64_t ds : 1; + /** Data gather */ + uint64_t dg : 1; +@@ -376,7 +376,7 @@ union zip_inst_s { + uint64_t bf : 1; + /** Comp/decomp operation */ + uint64_t op : 2; +- /** Data sactter */ ++ /** Data scatter */ + uint64_t ds : 1; + /** Data gather */ + uint64_t dg : 1; +diff --git a/dpdk/drivers/compress/octeontx/otx_zip.h b/dpdk/drivers/compress/octeontx/otx_zip.h +index e43f7f5c3e..118a95d738 100644 +--- a/dpdk/drivers/compress/octeontx/otx_zip.h ++++ b/dpdk/drivers/compress/octeontx/otx_zip.h +@@ -31,7 +31,7 @@ extern int octtx_zip_logtype_driver; + /**< PCI device id of ZIP VF */ + #define PCI_DEVICE_ID_OCTEONTX_ZIPVF 0xA037 + +-/* maxmum number of zip vf devices */ ++/* maximum number of zip vf devices */ + #define ZIP_MAX_VFS 8 + + /* max size of one chunk */ +diff --git a/dpdk/drivers/compress/octeontx/otx_zip_pmd.c b/dpdk/drivers/compress/octeontx/otx_zip_pmd.c +index dd62285b86..1b6178f661 100644 +--- a/dpdk/drivers/compress/octeontx/otx_zip_pmd.c ++++ b/dpdk/drivers/compress/octeontx/otx_zip_pmd.c +@@ -392,6 +392,8 @@ zip_pmd_qp_setup(struct rte_compressdev *dev, uint16_t qp_id, + } - queue_netdev_flow_put(pmd, flow, match, actions, actions_len, -- orig_in_port, DP_NETDEV_FLOW_OFFLOAD_OP_ADD); -+ DP_NETDEV_FLOW_OFFLOAD_OP_ADD); - log_netdev_flow_change(flow, match, NULL, actions, actions_len); + name = rte_malloc(NULL, RTE_COMPRESSDEV_NAME_MAX_LEN, 0); ++ if (name == NULL) ++ return (-ENOMEM); + snprintf(name, RTE_COMPRESSDEV_NAME_MAX_LEN, + "zip_pmd_%u_qp_%u", + dev->data->dev_id, qp_id); +@@ -399,8 +401,10 @@ zip_pmd_qp_setup(struct rte_compressdev *dev, uint16_t qp_id, + /* Allocate the queue pair data structure. */ + qp = rte_zmalloc_socket(name, sizeof(*qp), + RTE_CACHE_LINE_SIZE, socket_id); +- if (qp == NULL) ++ if (qp == NULL) { ++ rte_free(name); + return (-ENOMEM); ++ } + + qp->name = name; + +diff --git a/dpdk/drivers/compress/qat/qat_comp_pmd.c b/dpdk/drivers/compress/qat/qat_comp_pmd.c +index 9b24d46e97..da6404c017 100644 +--- a/dpdk/drivers/compress/qat/qat_comp_pmd.c ++++ b/dpdk/drivers/compress/qat/qat_comp_pmd.c +@@ -463,7 +463,7 @@ qat_comp_create_stream_pool(struct qat_comp_dev_private *comp_dev, + } else if (info.error) { + rte_mempool_obj_iter(mp, qat_comp_stream_destroy, NULL); + QAT_LOG(ERR, +- "Destoying mempool %s as at least one element failed initialisation", ++ "Destroying mempool %s as at least one element failed initialisation", + stream_pool_name); + rte_mempool_free(mp); + mp = NULL; +diff --git a/dpdk/drivers/crypto/bcmfs/bcmfs_device.h b/dpdk/drivers/crypto/bcmfs/bcmfs_device.h +index e5ca866977..4901a6cfd9 100644 +--- a/dpdk/drivers/crypto/bcmfs/bcmfs_device.h ++++ b/dpdk/drivers/crypto/bcmfs/bcmfs_device.h +@@ -32,7 +32,7 @@ enum bcmfs_device_type { + BCMFS_UNKNOWN + }; - return flow; -@@ -4171,7 +4173,7 @@ flow_put_on_pmd(struct dp_netdev_pmd_thread *pmd, - ovsrcu_set(&netdev_flow->actions, new_actions); +-/* A table to store registered queue pair opertations */ ++/* A table to store registered queue pair operations */ + struct bcmfs_hw_queue_pair_ops_table { + rte_spinlock_t tl; + /* Number of used ops structs in the table. */ +diff --git a/dpdk/drivers/crypto/bcmfs/bcmfs_qp.c b/dpdk/drivers/crypto/bcmfs/bcmfs_qp.c +index cb5ff6c61b..61d457f4e0 100644 +--- a/dpdk/drivers/crypto/bcmfs/bcmfs_qp.c ++++ b/dpdk/drivers/crypto/bcmfs/bcmfs_qp.c +@@ -212,7 +212,7 @@ bcmfs_qp_setup(struct bcmfs_qp **qp_addr, + nb_descriptors = FS_RM_MAX_REQS; + + if (qp_conf->iobase == NULL) { +- BCMFS_LOG(ERR, "IO onfig space null"); ++ BCMFS_LOG(ERR, "IO config space null"); + return -EINVAL; + } - queue_netdev_flow_put(pmd, netdev_flow, match, -- put->actions, put->actions_len, ODPP_NONE, -+ put->actions, put->actions_len, - DP_NETDEV_FLOW_OFFLOAD_OP_MOD); - log_netdev_flow_change(netdev_flow, match, old_actions, - put->actions, put->actions_len); -@@ -4778,8 +4780,8 @@ dpif_netdev_set_config(struct dpif *dpif, const struct smap *other_config) - uint32_t insert_min, cur_min; - uint32_t tx_flush_interval, cur_tx_flush_interval; - uint64_t rebalance_intvl; -- uint8_t rebalance_load, cur_rebalance_load; -- uint8_t rebalance_improve; -+ uint8_t cur_rebalance_load; -+ uint32_t rebalance_load, rebalance_improve; - bool log_autolb = false; - enum sched_assignment_type pmd_rxq_assign_type; +diff --git a/dpdk/drivers/crypto/bcmfs/bcmfs_sym_defs.h b/dpdk/drivers/crypto/bcmfs/bcmfs_sym_defs.h +index eaefe97e26..9bb8a695a0 100644 +--- a/dpdk/drivers/crypto/bcmfs/bcmfs_sym_defs.h ++++ b/dpdk/drivers/crypto/bcmfs/bcmfs_sym_defs.h +@@ -20,11 +20,11 @@ struct bcmfs_sym_request; + + /** Crypto Request processing successful. */ + #define BCMFS_SYM_RESPONSE_SUCCESS (0) +-/** Crypot Request processing protocol failure. */ ++/** Crypto Request processing protocol failure. */ + #define BCMFS_SYM_RESPONSE_PROTO_FAILURE (1) +-/** Crypot Request processing completion failure. */ ++/** Crypto Request processing completion failure. */ + #define BCMFS_SYM_RESPONSE_COMPL_ERROR (2) +-/** Crypot Request processing hash tag check error. */ ++/** Crypto Request processing hash tag check error. */ + #define BCMFS_SYM_RESPONSE_HASH_TAG_ERROR (3) + + /** Maximum threshold length to adjust AAD in continuation +diff --git a/dpdk/drivers/crypto/bcmfs/bcmfs_sym_engine.h b/dpdk/drivers/crypto/bcmfs/bcmfs_sym_engine.h +index d9594246b5..51ff9f75ed 100644 +--- a/dpdk/drivers/crypto/bcmfs/bcmfs_sym_engine.h ++++ b/dpdk/drivers/crypto/bcmfs/bcmfs_sym_engine.h +@@ -12,7 +12,7 @@ + #include "bcmfs_sym_defs.h" + #include "bcmfs_sym_req.h" + +-/* structure to hold element's arrtibutes */ ++/* structure to hold element's attributes */ + struct fsattr { + void *va; + uint64_t pa; +diff --git a/dpdk/drivers/crypto/bcmfs/hw/bcmfs5_rm.c b/dpdk/drivers/crypto/bcmfs/hw/bcmfs5_rm.c +index 86e53051dd..c677c0cd9b 100644 +--- a/dpdk/drivers/crypto/bcmfs/hw/bcmfs5_rm.c ++++ b/dpdk/drivers/crypto/bcmfs/hw/bcmfs5_rm.c +@@ -441,7 +441,7 @@ static void bcmfs5_write_doorbell(struct bcmfs_qp *qp) + { + struct bcmfs_queue *txq = &qp->tx_q; + +- /* sync in bfeore ringing the door-bell */ ++ /* sync in before ringing the door-bell */ + rte_wmb(); + + FS_MMIO_WRITE32(txq->descs_inflight, +diff --git a/dpdk/drivers/crypto/caam_jr/caam_jr_hw_specific.h b/dpdk/drivers/crypto/caam_jr/caam_jr_hw_specific.h +index bbe8bc3f90..6ee7f7cef3 100644 +--- a/dpdk/drivers/crypto/caam_jr/caam_jr_hw_specific.h ++++ b/dpdk/drivers/crypto/caam_jr/caam_jr_hw_specific.h +@@ -376,7 +376,7 @@ struct sec_job_ring_t { + void *register_base_addr; /* Base address for SEC's + * register memory for this job ring. + */ +- uint8_t coalescing_en; /* notifies if coelescing is ++ uint8_t coalescing_en; /* notifies if coalescing is + * enabled for the job ring + */ + sec_job_ring_state_t jr_state; /* The state of this job ring */ +@@ -479,7 +479,7 @@ void hw_job_ring_error_print(struct sec_job_ring_t *job_ring, int code); + + /* @brief Set interrupt coalescing parameters on the Job Ring. + * @param [in] job_ring The job ring +- * @param [in] irq_coalesing_timer Interrupt coalescing timer threshold. ++ * @param [in] irq_coalescing_timer Interrupt coalescing timer threshold. + * This value determines the maximum + * amount of time after processing a + * descriptor before raising an interrupt. +diff --git a/dpdk/drivers/crypto/caam_jr/caam_jr_pvt.h b/dpdk/drivers/crypto/caam_jr/caam_jr_pvt.h +index 552d6b9b1b..52f872bcd0 100644 +--- a/dpdk/drivers/crypto/caam_jr/caam_jr_pvt.h ++++ b/dpdk/drivers/crypto/caam_jr/caam_jr_pvt.h +@@ -169,7 +169,7 @@ struct sec4_sg_entry { + + /* Structure encompassing a job descriptor which is to be processed + * by SEC. User should also initialise this structure with the callback +- * function pointer which will be called by driver after recieving proccessed ++ * function pointer which will be called by driver after receiving processed + * descriptor from SEC. User data is also passed in this data structure which + * will be sent as an argument to the user callback function. + */ +@@ -288,7 +288,7 @@ int caam_jr_enable_irqs(int uio_fd); + * value that indicates an IRQ disable action into UIO file descriptor + * of this job ring. + * +- * @param [in] uio_fd UIO File descripto ++ * @param [in] uio_fd UIO File descriptor + * @retval 0 for success + * @retval -1 value for error + * +diff --git a/dpdk/drivers/crypto/caam_jr/caam_jr_uio.c b/dpdk/drivers/crypto/caam_jr/caam_jr_uio.c +index e4ee102344..583ba3b523 100644 +--- a/dpdk/drivers/crypto/caam_jr/caam_jr_uio.c ++++ b/dpdk/drivers/crypto/caam_jr/caam_jr_uio.c +@@ -227,7 +227,7 @@ caam_jr_enable_irqs(int uio_fd) + * value that indicates an IRQ disable action into UIO file descriptor + * of this job ring. + * +- * @param [in] uio_fd UIO File descripto ++ * @param [in] uio_fd UIO File descriptor + * @retval 0 for success + * @retval -1 value for error + * +diff --git a/dpdk/drivers/crypto/ccp/ccp_crypto.c b/dpdk/drivers/crypto/ccp/ccp_crypto.c +index 70daed791e..4bab18323b 100644 +--- a/dpdk/drivers/crypto/ccp/ccp_crypto.c ++++ b/dpdk/drivers/crypto/ccp/ccp_crypto.c +@@ -2,6 +2,8 @@ + * Copyright(c) 2018 Advanced Micro Devices, Inc. All rights reserved. + */ -@@ -4880,8 +4882,12 @@ dpif_netdev_set_config(struct dpif *dpif, const struct smap *other_config) ++#define OPENSSL_API_COMPAT 0x10100000L ++ + #include + #include + #include +@@ -1299,7 +1301,7 @@ ccp_auth_slot(struct ccp_session *session) + case CCP_AUTH_ALGO_SHA512_HMAC: + /** + * 1. Load PHash1 = H(k ^ ipad); to LSB +- * 2. generate IHash = H(hash on meassage with PHash1 ++ * 2. generate IHash = H(hash on message with PHash1 + * as init values); + * 3. Retrieve IHash 2 slots for 384/512 + * 4. Load Phash2 = H(k ^ opad); to LSB +diff --git a/dpdk/drivers/crypto/ccp/ccp_crypto.h b/dpdk/drivers/crypto/ccp/ccp_crypto.h +index 8e6d03efc8..d307f73ee4 100644 +--- a/dpdk/drivers/crypto/ccp/ccp_crypto.h ++++ b/dpdk/drivers/crypto/ccp/ccp_crypto.h +@@ -70,7 +70,7 @@ + /* Maximum length for digest */ + #define DIGEST_LENGTH_MAX 64 + +-/* SHA LSB intialiazation values */ ++/* SHA LSB initialization values */ + + #define SHA1_H0 0x67452301UL + #define SHA1_H1 0xefcdab89UL +diff --git a/dpdk/drivers/crypto/ccp/ccp_dev.h b/dpdk/drivers/crypto/ccp/ccp_dev.h +index 85c8fc47a2..2a205cd446 100644 +--- a/dpdk/drivers/crypto/ccp/ccp_dev.h ++++ b/dpdk/drivers/crypto/ccp/ccp_dev.h +@@ -19,7 +19,7 @@ + #include + #include + +-/**< CCP sspecific */ ++/**< CCP specific */ + #define MAX_HW_QUEUES 5 + #define CCP_MAX_TRNG_RETRIES 10 + #define CCP_ALIGN(x, y) ((((x) + (y - 1)) / y) * y) +diff --git a/dpdk/drivers/crypto/cnxk/cn9k_ipsec_la_ops.h b/dpdk/drivers/crypto/cnxk/cn9k_ipsec_la_ops.h +index 2dc8913feb..2b0261e057 100644 +--- a/dpdk/drivers/crypto/cnxk/cn9k_ipsec_la_ops.h ++++ b/dpdk/drivers/crypto/cnxk/cn9k_ipsec_la_ops.h +@@ -77,9 +77,10 @@ process_outb_sa(struct rte_crypto_op *cop, struct cn9k_ipsec_sa *sa, + const unsigned int hdr_len = sizeof(struct roc_ie_on_outb_hdr); + struct rte_crypto_sym_op *sym_op = cop->sym; + struct rte_mbuf *m_src = sym_op->m_src; +- uint32_t dlen, rlen, extend_tail; + struct roc_ie_on_outb_sa *out_sa; + struct roc_ie_on_outb_hdr *hdr; ++ uint32_t dlen, rlen; ++ int32_t extend_tail; + + out_sa = &sa->out_sa; + +@@ -88,7 +89,8 @@ process_outb_sa(struct rte_crypto_op *cop, struct cn9k_ipsec_sa *sa, + + extend_tail = rlen - dlen; + if (unlikely(extend_tail > rte_pktmbuf_tailroom(m_src))) { +- plt_dp_err("Not enough tail room"); ++ plt_dp_err("Not enough tail room (required: %d, available: %d", ++ extend_tail, rte_pktmbuf_tailroom(m_src)); + return -ENOMEM; + } - struct pmd_auto_lb *pmd_alb = &dp->pmd_alb; +diff --git a/dpdk/drivers/crypto/cnxk/cnxk_cryptodev_ops.c b/dpdk/drivers/crypto/cnxk/cnxk_cryptodev_ops.c +index a2281fb8de..0d99c891d9 100644 +--- a/dpdk/drivers/crypto/cnxk/cnxk_cryptodev_ops.c ++++ b/dpdk/drivers/crypto/cnxk/cnxk_cryptodev_ops.c +@@ -100,8 +100,13 @@ cnxk_cpt_dev_start(struct rte_cryptodev *dev) + uint16_t nb_lf = roc_cpt->nb_lf; + uint16_t qp_id; -- rebalance_intvl = smap_get_int(other_config, "pmd-auto-lb-rebal-interval", -- ALB_REBALANCE_INTERVAL); -+ rebalance_intvl = smap_get_ullong(other_config, -+ "pmd-auto-lb-rebal-interval", -+ ALB_REBALANCE_INTERVAL); -+ if (rebalance_intvl > MAX_ALB_REBALANCE_INTERVAL) { -+ rebalance_intvl = ALB_REBALANCE_INTERVAL; -+ } +- for (qp_id = 0; qp_id < nb_lf; qp_id++) ++ for (qp_id = 0; qp_id < nb_lf; qp_id++) { ++ /* Application may not setup all queue pair */ ++ if (roc_cpt->lf[qp_id] == NULL) ++ continue; ++ + roc_cpt_iq_enable(roc_cpt->lf[qp_id]); ++ } - /* Input is in min, convert it to msec. */ - rebalance_intvl = -@@ -4894,21 +4900,21 @@ dpif_netdev_set_config(struct dpif *dpif, const struct smap *other_config) - log_autolb = true; - } + return 0; + } +@@ -114,8 +119,12 @@ cnxk_cpt_dev_stop(struct rte_cryptodev *dev) + uint16_t nb_lf = roc_cpt->nb_lf; + uint16_t qp_id; -- rebalance_improve = smap_get_int(other_config, -- "pmd-auto-lb-improvement-threshold", -- ALB_IMPROVEMENT_THRESHOLD); -+ rebalance_improve = smap_get_uint(other_config, -+ "pmd-auto-lb-improvement-threshold", -+ ALB_IMPROVEMENT_THRESHOLD); - if (rebalance_improve > 100) { - rebalance_improve = ALB_IMPROVEMENT_THRESHOLD; - } - if (rebalance_improve != pmd_alb->rebalance_improve_thresh) { - pmd_alb->rebalance_improve_thresh = rebalance_improve; - VLOG_INFO("PMD auto load balance improvement threshold set to " -- "%"PRIu8"%%", rebalance_improve); -+ "%"PRIu32"%%", rebalance_improve); - log_autolb = true; - } +- for (qp_id = 0; qp_id < nb_lf; qp_id++) ++ for (qp_id = 0; qp_id < nb_lf; qp_id++) { ++ if (roc_cpt->lf[qp_id] == NULL) ++ continue; ++ + roc_cpt_iq_disable(roc_cpt->lf[qp_id]); ++ } + } -- rebalance_load = smap_get_int(other_config, "pmd-auto-lb-load-threshold", -- ALB_LOAD_THRESHOLD); -+ rebalance_load = smap_get_uint(other_config, "pmd-auto-lb-load-threshold", -+ ALB_LOAD_THRESHOLD); - if (rebalance_load > 100) { - rebalance_load = ALB_LOAD_THRESHOLD; - } -@@ -4916,7 +4922,7 @@ dpif_netdev_set_config(struct dpif *dpif, const struct smap *other_config) - if (rebalance_load != cur_rebalance_load) { - atomic_store_relaxed(&pmd_alb->rebalance_load_thresh, - rebalance_load); -- VLOG_INFO("PMD auto load balance load threshold set to %"PRIu8"%%", -+ VLOG_INFO("PMD auto load balance load threshold set to %"PRIu32"%%", - rebalance_load); - log_autolb = true; - } -@@ -5684,23 +5690,28 @@ sched_numa_list_put_in_place(struct sched_numa_list *numa_list) - } + int +@@ -352,6 +361,7 @@ cnxk_cpt_queue_pair_setup(struct rte_cryptodev *dev, uint16_t qp_id, + struct roc_cpt *roc_cpt = &vf->cpt; + struct rte_pci_device *pci_dev; + struct cnxk_cpt_qp *qp; ++ uint32_t nb_desc; + int ret; + + if (dev->data->queue_pairs[qp_id] != NULL) +@@ -364,14 +374,17 @@ cnxk_cpt_queue_pair_setup(struct rte_cryptodev *dev, uint16_t qp_id, + return -EIO; + } + +- qp = cnxk_cpt_qp_create(dev, qp_id, conf->nb_descriptors); ++ /* Update nb_desc to next power of 2 to aid in pending queue checks */ ++ nb_desc = plt_align32pow2(conf->nb_descriptors); ++ ++ qp = cnxk_cpt_qp_create(dev, qp_id, nb_desc); + if (qp == NULL) { + plt_err("Could not create queue pair %d", qp_id); + return -ENOMEM; + } + + qp->lf.lf_id = qp_id; +- qp->lf.nb_desc = conf->nb_descriptors; ++ qp->lf.nb_desc = nb_desc; + + ret = roc_cpt_lf_init(roc_cpt, &qp->lf); + if (ret < 0) { +diff --git a/dpdk/drivers/crypto/cnxk/cnxk_cryptodev_ops.h b/dpdk/drivers/crypto/cnxk/cnxk_cryptodev_ops.h +index 0d363651ff..1d1b1fc94b 100644 +--- a/dpdk/drivers/crypto/cnxk/cnxk_cryptodev_ops.h ++++ b/dpdk/drivers/crypto/cnxk/cnxk_cryptodev_ops.h +@@ -166,7 +166,11 @@ pending_queue_retreat(uint64_t *index, const uint64_t mask, uint64_t nb_entry) + static __rte_always_inline uint64_t + pending_queue_infl_cnt(uint64_t head, uint64_t tail, const uint64_t mask) + { +- return (head - tail) & mask; ++ /* ++ * Mask is nb_desc - 1. Add nb_desc to head and mask to account for ++ * cases when tail > head, which happens during wrap around. ++ */ ++ return ((head + mask + 1) - tail) & mask; } -+/* Returns 'true' if OVS rxq scheduling algorithm assigned any unpinned rxq to -+ * a PMD thread core on a non-local numa node. */ - static bool - sched_numa_list_cross_numa_polling(struct sched_numa_list *numa_list) + static __rte_always_inline uint64_t +diff --git a/dpdk/drivers/crypto/cnxk/cnxk_se.h b/dpdk/drivers/crypto/cnxk/cnxk_se.h +index 37237de21a..af86ef18d8 100644 +--- a/dpdk/drivers/crypto/cnxk/cnxk_se.h ++++ b/dpdk/drivers/crypto/cnxk/cnxk_se.h +@@ -39,17 +39,16 @@ struct cnxk_se_sess { + static inline void + cpt_pack_iv(uint8_t *iv_src, uint8_t *iv_dst) { - struct sched_numa *numa; +- iv_dst[16] = iv_src[16]; +- /* pack the last 8 bytes of IV to 6 bytes. ++ /* pack the first 8 bytes of IV to 6 bytes. + * discard the 2 MSB bits of each byte + */ +- iv_dst[17] = (((iv_src[17] & 0x3f) << 2) | ((iv_src[18] >> 4) & 0x3)); +- iv_dst[18] = (((iv_src[18] & 0xf) << 4) | ((iv_src[19] >> 2) & 0xf)); +- iv_dst[19] = (((iv_src[19] & 0x3) << 6) | (iv_src[20] & 0x3f)); ++ iv_dst[0] = (((iv_src[0] & 0x3f) << 2) | ((iv_src[1] >> 4) & 0x3)); ++ iv_dst[1] = (((iv_src[1] & 0xf) << 4) | ((iv_src[2] >> 2) & 0xf)); ++ iv_dst[2] = (((iv_src[2] & 0x3) << 6) | (iv_src[3] & 0x3f)); + +- iv_dst[20] = (((iv_src[21] & 0x3f) << 2) | ((iv_src[22] >> 4) & 0x3)); +- iv_dst[21] = (((iv_src[22] & 0xf) << 4) | ((iv_src[23] >> 2) & 0xf)); +- iv_dst[22] = (((iv_src[23] & 0x3) << 6) | (iv_src[24] & 0x3f)); ++ iv_dst[3] = (((iv_src[4] & 0x3f) << 2) | ((iv_src[5] >> 4) & 0x3)); ++ iv_dst[4] = (((iv_src[5] & 0xf) << 4) | ((iv_src[6] >> 2) & 0xf)); ++ iv_dst[5] = (((iv_src[6] & 0x3) << 6) | (iv_src[7] & 0x3f)); + } -- /* For each numa */ - HMAP_FOR_EACH (numa, node, &numa_list->numas) { -- /* For each pmd */ - for (int i = 0; i < numa->n_pmds; i++) { - struct sched_pmd *sched_pmd; + static inline void +@@ -71,10 +70,11 @@ pdcp_iv_copy(uint8_t *iv_d, uint8_t *iv_s, const uint8_t pdcp_alg_type, + iv_temp[j] = iv_s_temp[3 - j]; + memcpy(iv_d, iv_temp, 16); + } else { +- /* ZUC doesn't need a swap */ +- memcpy(iv_d, iv_s, 16); +- if (pack_iv) ++ if (pack_iv) { + cpt_pack_iv(iv_s, iv_d); ++ memcpy(iv_d + 6, iv_s + 8, 17); ++ } else ++ memcpy(iv_d, iv_s, 16); + } + } - sched_pmd = &numa->pmds[i]; -- /* For each rxq. */ -+ if (sched_pmd->isolated) { -+ /* All rxqs on this PMD thread core are pinned. */ -+ continue; -+ } - for (unsigned k = 0; k < sched_pmd->n_rxq; k++) { - struct dp_netdev_rxq *rxq = sched_pmd->rxqs[k]; -- -- if (!sched_pmd->isolated && -+ /* Check if the rxq is not pinned to a specific PMD thread core -+ * by the user AND the PMD thread core that OVS assigned is -+ * non-local to the rxq port. */ -+ if (rxq->core_id == OVS_CORE_UNSPEC && - rxq->pmd->numa_id != - netdev_get_numa_id(rxq->port->netdev)) { - return true; -@@ -6000,10 +6011,10 @@ sched_numa_list_schedule(struct sched_numa_list *numa_list, - /* Find any numa with available PMDs. */ - for (int j = 0; j < n_numa; j++) { - numa = sched_numa_list_next(numa_list, last_cross_numa); -+ last_cross_numa = numa; - if (sched_numa_noniso_pmd_count(numa)) { - break; - } -- last_cross_numa = numa; - numa = NULL; - } - } -@@ -6111,7 +6122,7 @@ sched_numa_list_variance(struct sched_numa_list *numa_list) - * pmd_rebalance_dry_run() can be avoided when it is not needed. +@@ -179,27 +179,27 @@ fill_sg_comp_from_iov(struct roc_se_sglist_comp *list, uint32_t i, + int32_t j; + uint32_t extra_len = extra_buf ? extra_buf->size : 0; + uint32_t size = *psize; +- struct roc_se_buf_ptr *bufs; + +- bufs = from->bufs; + for (j = 0; (j < from->buf_cnt) && size; j++) { ++ struct roc_se_sglist_comp *to = &list[i >> 2]; ++ uint32_t buf_sz = from->bufs[j].size; ++ void *vaddr = from->bufs[j].vaddr; + uint64_t e_vaddr; + uint32_t e_len; +- struct roc_se_sglist_comp *to = &list[i >> 2]; + + if (unlikely(from_offset)) { +- if (from_offset >= bufs[j].size) { +- from_offset -= bufs[j].size; ++ if (from_offset >= buf_sz) { ++ from_offset -= buf_sz; + continue; + } +- e_vaddr = (uint64_t)bufs[j].vaddr + from_offset; +- e_len = (size > (bufs[j].size - from_offset)) ? +- (bufs[j].size - from_offset) : ++ e_vaddr = (uint64_t)vaddr + from_offset; ++ e_len = (size > (buf_sz - from_offset)) ? ++ (buf_sz - from_offset) : + size; + from_offset = 0; + } else { +- e_vaddr = (uint64_t)bufs[j].vaddr; +- e_len = (size > bufs[j].size) ? bufs[j].size : size; ++ e_vaddr = (uint64_t)vaddr; ++ e_len = (size > buf_sz) ? buf_sz : size; + } + + to->u.s.len[i % 4] = rte_cpu_to_be_16(e_len); +@@ -1020,6 +1020,7 @@ cpt_zuc_snow3g_prep(uint32_t req_flags, uint64_t d_offs, uint64_t d_lens, + iv_len = params->auth_iv_len; + + if (iv_len == 25) { ++ roc_se_zuc_bytes_swap(iv_s, iv_len); + iv_len -= 2; + pack_iv = 1; + } +@@ -1049,6 +1050,7 @@ cpt_zuc_snow3g_prep(uint32_t req_flags, uint64_t d_offs, uint64_t d_lens, + iv_len = params->cipher_iv_len; + + if (iv_len == 25) { ++ roc_se_zuc_bytes_swap(iv_s, iv_len); + iv_len -= 2; + pack_iv = 1; + } +diff --git a/dpdk/drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c b/dpdk/drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c +index a5b052375d..c9745f1db0 100644 +--- a/dpdk/drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c ++++ b/dpdk/drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c +@@ -1,7 +1,7 @@ + /* SPDX-License-Identifier: BSD-3-Clause + * + * Copyright (c) 2016 Freescale Semiconductor, Inc. All rights reserved. +- * Copyright 2016-2021 NXP ++ * Copyright 2016-2022 NXP + * */ - static bool --pmd_reblance_dry_run_needed(struct dp_netdev *dp) -+pmd_rebalance_dry_run_needed(struct dp_netdev *dp) - OVS_REQ_RDLOCK(dp->port_rwlock) - { - struct dp_netdev_pmd_thread *pmd; -@@ -6342,11 +6353,11 @@ pmd_remove_stale_ports(struct dp_netdev *dp, - OVS_EXCLUDED(pmd->port_mutex) - OVS_REQ_RDLOCK(dp->port_rwlock) - { -- struct rxq_poll *poll, *poll_next; -- struct tx_port *tx, *tx_next; -+ struct rxq_poll *poll; -+ struct tx_port *tx; - ovs_mutex_lock(&pmd->port_mutex); -- HMAP_FOR_EACH_SAFE (poll, poll_next, node, &pmd->poll_list) { -+ HMAP_FOR_EACH_SAFE (poll, node, &pmd->poll_list) { - struct dp_netdev_port *port = poll->rxq->port; +@@ -52,6 +52,27 @@ - if (port->need_reconfigure -@@ -6354,7 +6365,7 @@ pmd_remove_stale_ports(struct dp_netdev *dp, - dp_netdev_del_rxq_from_pmd(pmd, poll); - } - } -- HMAP_FOR_EACH_SAFE (tx, tx_next, node, &pmd->tx_ports) { -+ HMAP_FOR_EACH_SAFE (tx, node, &pmd->tx_ports) { - struct dp_netdev_port *port = tx->port; + uint8_t cryptodev_driver_id; - if (port->need_reconfigure -@@ -6430,8 +6441,7 @@ reconfigure_datapath(struct dp_netdev *dp) - /* We only reconfigure the ports that we determined above, because they're - * not being used by any pmd thread at the moment. If a port fails to - * reconfigure we remove it from the datapath. */ -- struct dp_netdev_port *next_port; -- HMAP_FOR_EACH_SAFE (port, next_port, node, &dp->ports) { -+ HMAP_FOR_EACH_SAFE (port, node, &dp->ports) { - int err; ++static inline void ++free_fle(const struct qbman_fd *fd) ++{ ++ struct qbman_fle *fle; ++ struct rte_crypto_op *op; ++ struct ctxt_priv *priv; ++ ++#ifdef RTE_LIB_SECURITY ++ if (DPAA2_FD_GET_FORMAT(fd) == qbman_fd_single) ++ return; ++#endif ++ fle = (struct qbman_fle *)DPAA2_IOVA_TO_VADDR(DPAA2_GET_FD_ADDR(fd)); ++ op = (struct rte_crypto_op *)DPAA2_GET_FLE_ADDR((fle - 1)); ++ /* free the fle memory */ ++ if (likely(rte_pktmbuf_is_contiguous(op->sym->m_src))) { ++ priv = (struct ctxt_priv *)(size_t)DPAA2_GET_FLE_CTXT(fle - 1); ++ rte_mempool_put(priv->fle_pool, (void *)(fle-1)); ++ } else ++ rte_free((void *)(fle-1)); ++} ++ + #ifdef RTE_LIB_SECURITY + static inline int + build_proto_compound_sg_fd(dpaa2_sec_session *sess, +@@ -1501,6 +1522,12 @@ dpaa2_sec_enqueue_burst(void *qp, struct rte_crypto_op **ops, + if (retry_count > DPAA2_MAX_TX_RETRY_COUNT) { + num_tx += loop; + nb_ops -= loop; ++ DPAA2_SEC_DP_DEBUG("Enqueue fail\n"); ++ /* freeing the fle buffers */ ++ while (loop < frames_to_send) { ++ free_fle(&fd_arr[loop]); ++ loop++; ++ } + goto skip_tx; + } + } else { +@@ -1527,6 +1554,10 @@ sec_simple_fd_to_mbuf(const struct qbman_fd *fd) + int16_t diff = 0; + dpaa2_sec_session *sess_priv __rte_unused; + ++ if (unlikely(DPAA2_GET_FD_IVP(fd))) { ++ DPAA2_SEC_ERR("error: non inline buffer"); ++ return NULL; ++ } + struct rte_mbuf *mbuf = DPAA2_INLINE_MBUF_FROM_BUF( + DPAA2_IOVA_TO_VADDR(DPAA2_GET_FD_ADDR(fd)), + rte_dpaa2_bpid_info[DPAA2_GET_FD_BPID(fd)].meta_data_size); +@@ -1545,6 +1576,14 @@ sec_simple_fd_to_mbuf(const struct qbman_fd *fd) + else + mbuf->data_off += SEC_FLC_DHR_INBOUND; + ++ if (unlikely(fd->simple.frc)) { ++ DPAA2_SEC_ERR("SEC returned Error - %x", ++ fd->simple.frc); ++ op->status = RTE_CRYPTO_OP_STATUS_ERROR; ++ } else { ++ op->status = RTE_CRYPTO_OP_STATUS_SUCCESS; ++ } ++ + return op; + } + #endif +@@ -1573,11 +1612,6 @@ sec_fd_to_mbuf(const struct qbman_fd *fd) + * We can have a better approach to use the inline Mbuf + */ + +- if (unlikely(DPAA2_GET_FD_IVP(fd))) { +- /* TODO complete it. */ +- DPAA2_SEC_ERR("error: non inline buffer"); +- return NULL; +- } + op = (struct rte_crypto_op *)DPAA2_GET_FLE_ADDR((fle - 1)); + + /* Prefeth op */ +@@ -3138,13 +3172,15 @@ dpaa2_sec_set_pdcp_session(struct rte_cryptodev *dev, + /* find xfrm types */ + if (xform->type == RTE_CRYPTO_SYM_XFORM_CIPHER) { + cipher_xform = &xform->cipher; +- if (xform->next != NULL) { ++ if (xform->next != NULL && ++ xform->next->type == RTE_CRYPTO_SYM_XFORM_AUTH) { + session->ext_params.aead_ctxt.auth_cipher_text = true; + auth_xform = &xform->next->auth; + } + } else if (xform->type == RTE_CRYPTO_SYM_XFORM_AUTH) { + auth_xform = &xform->auth; +- if (xform->next != NULL) { ++ if (xform->next != NULL && ++ xform->next->type == RTE_CRYPTO_SYM_XFORM_CIPHER) { + session->ext_params.aead_ctxt.auth_cipher_text = false; + cipher_xform = &xform->next->cipher; + } +@@ -3723,7 +3759,7 @@ dpaa2_sec_process_atomic_event(struct qbman_swp *swp __rte_unused, + struct rte_event *ev) + { + uint8_t dqrr_index; +- struct rte_crypto_op *crypto_op = (struct rte_crypto_op *)ev->event_ptr; ++ struct rte_crypto_op *crypto_op; + /* Prefetching mbuf */ + rte_prefetch0((void *)(size_t)(DPAA2_GET_FD_ADDR(fd)- + rte_dpaa2_bpid_info[DPAA2_GET_FD_BPID(fd)].meta_data_size)); +@@ -3739,12 +3775,13 @@ dpaa2_sec_process_atomic_event(struct qbman_swp *swp __rte_unused, + ev->queue_id = rxq->ev.queue_id; + ev->priority = rxq->ev.priority; + +- ev->event_ptr = sec_fd_to_mbuf(fd); ++ crypto_op = sec_fd_to_mbuf(fd); + dqrr_index = qbman_get_dqrr_idx(dq); + *dpaa2_seqn(crypto_op->sym->m_src) = dqrr_index + 1; + DPAA2_PER_LCORE_DQRR_SIZE++; + DPAA2_PER_LCORE_DQRR_HELD |= 1 << dqrr_index; + DPAA2_PER_LCORE_DQRR_MBUF(dqrr_index) = crypto_op->sym->m_src; ++ ev->event_ptr = crypto_op; + } - if (!port->need_reconfigure) { -@@ -6487,10 +6497,10 @@ reconfigure_datapath(struct dp_netdev *dp) - } + int +diff --git a/dpdk/drivers/crypto/dpaa2_sec/dpaa2_sec_raw_dp.c b/dpdk/drivers/crypto/dpaa2_sec/dpaa2_sec_raw_dp.c +index 74f2045637..e68a4875dd 100644 +--- a/dpdk/drivers/crypto/dpaa2_sec/dpaa2_sec_raw_dp.c ++++ b/dpdk/drivers/crypto/dpaa2_sec/dpaa2_sec_raw_dp.c +@@ -1,5 +1,5 @@ + /* SPDX-License-Identifier: BSD-3-Clause +- * Copyright 2021 NXP ++ * Copyright 2021-2022 NXP + */ - CMAP_FOR_EACH (pmd, node, &dp->poll_threads) { -- struct rxq_poll *poll, *poll_next; -+ struct rxq_poll *poll; + #include +@@ -44,8 +44,8 @@ build_raw_dp_chain_fd(uint8_t *drv_ctx, + uint16_t auth_hdr_len = ofs.ofs.cipher.head - + ofs.ofs.auth.head; + +- uint16_t auth_tail_len = ofs.ofs.auth.tail; +- uint32_t auth_only_len = (auth_tail_len << 16) | auth_hdr_len; ++ uint16_t auth_tail_len; ++ uint32_t auth_only_len; + int icv_len = sess->digest_length; + uint8_t *old_icv; + uint8_t *iv_ptr = iv->va; +@@ -55,6 +55,8 @@ build_raw_dp_chain_fd(uint8_t *drv_ctx, + + cipher_len = data_len - ofs.ofs.cipher.head - ofs.ofs.cipher.tail; + auth_len = data_len - ofs.ofs.auth.head - ofs.ofs.auth.tail; ++ auth_tail_len = auth_len - cipher_len - auth_hdr_len; ++ auth_only_len = (auth_tail_len << 16) | auth_hdr_len; + /* first FLE entry used to store session ctxt */ + fle = (struct qbman_fle *)rte_malloc(NULL, + FLE_SG_MEM_SIZE(2 * sgl->num), +@@ -104,6 +106,7 @@ build_raw_dp_chain_fd(uint8_t *drv_ctx, + DPAA2_SET_FLE_OFFSET(sge, 0); + sge->length = dest_sgl->vec[i].len; + } ++ sge->length -= ofs.ofs.cipher.tail; + } else { + /* Configure Output SGE for Encap/Decap */ + DPAA2_SET_FLE_ADDR(sge, sgl->vec[0].iova); +@@ -117,6 +120,7 @@ build_raw_dp_chain_fd(uint8_t *drv_ctx, + DPAA2_SET_FLE_OFFSET(sge, 0); + sge->length = sgl->vec[i].len; + } ++ sge->length -= ofs.ofs.cipher.tail; + } - ovs_mutex_lock(&pmd->port_mutex); -- HMAP_FOR_EACH_SAFE (poll, poll_next, node, &pmd->poll_list) { -+ HMAP_FOR_EACH_SAFE (poll, node, &pmd->poll_list) { - if (poll->rxq->pmd != pmd) { - dp_netdev_del_rxq_from_pmd(pmd, poll); + if (sess->dir == DIR_ENC) { +diff --git a/dpdk/drivers/crypto/dpaa_sec/dpaa_sec.c b/dpdk/drivers/crypto/dpaa_sec/dpaa_sec.c +index a552e64506..0a6126ad97 100644 +--- a/dpdk/drivers/crypto/dpaa_sec/dpaa_sec.c ++++ b/dpdk/drivers/crypto/dpaa_sec/dpaa_sec.c +@@ -723,7 +723,7 @@ dpaa_sec_deq(struct dpaa_sec_qp *qp, struct rte_crypto_op **ops, int nb_ops) + } + ops[pkts++] = op; + +- /* report op status to sym->op and then free the ctx memeory */ ++ /* report op status to sym->op and then free the ctx memory */ + rte_mempool_put(ctx->ctx_pool, (void *)ctx); + + qman_dqrr_consume(fq, dq); +@@ -2986,11 +2986,13 @@ dpaa_sec_set_pdcp_session(struct rte_cryptodev *dev, + /* find xfrm types */ + if (xform->type == RTE_CRYPTO_SYM_XFORM_CIPHER) { + cipher_xform = &xform->cipher; +- if (xform->next != NULL) ++ if (xform->next != NULL && ++ xform->next->type == RTE_CRYPTO_SYM_XFORM_AUTH) + auth_xform = &xform->next->auth; + } else if (xform->type == RTE_CRYPTO_SYM_XFORM_AUTH) { + auth_xform = &xform->auth; +- if (xform->next != NULL) ++ if (xform->next != NULL && ++ xform->next->type == RTE_CRYPTO_SYM_XFORM_CIPHER) + cipher_xform = &xform->next->cipher; + } else { + DPAA_SEC_ERR("Invalid crypto type"); +@@ -3552,23 +3554,24 @@ cryptodev_dpaa_sec_probe(struct rte_dpaa_driver *dpaa_drv __rte_unused, + + int retval; + ++ if (rte_eal_process_type() != RTE_PROC_PRIMARY) ++ return 0; ++ + snprintf(cryptodev_name, sizeof(cryptodev_name), "%s", dpaa_dev->name); + + cryptodev = rte_cryptodev_pmd_allocate(cryptodev_name, rte_socket_id()); + if (cryptodev == NULL) + return -ENOMEM; + +- if (rte_eal_process_type() == RTE_PROC_PRIMARY) { +- cryptodev->data->dev_private = rte_zmalloc_socket( +- "cryptodev private structure", +- sizeof(struct dpaa_sec_dev_private), +- RTE_CACHE_LINE_SIZE, +- rte_socket_id()); ++ cryptodev->data->dev_private = rte_zmalloc_socket( ++ "cryptodev private structure", ++ sizeof(struct dpaa_sec_dev_private), ++ RTE_CACHE_LINE_SIZE, ++ rte_socket_id()); + +- if (cryptodev->data->dev_private == NULL) +- rte_panic("Cannot allocate memzone for private " +- "device data"); +- } ++ if (cryptodev->data->dev_private == NULL) ++ rte_panic("Cannot allocate memzone for private " ++ "device data"); + + dpaa_dev->crypto_dev = cryptodev; + cryptodev->device = &dpaa_dev->device; +@@ -3610,8 +3613,7 @@ cryptodev_dpaa_sec_probe(struct rte_dpaa_driver *dpaa_drv __rte_unused, + retval = -ENXIO; + out: + /* In case of error, cleanup is done */ +- if (rte_eal_process_type() == RTE_PROC_PRIMARY) +- rte_free(cryptodev->data->dev_private); ++ rte_free(cryptodev->data->dev_private); + + rte_cryptodev_pmd_release_device(cryptodev); + +diff --git a/dpdk/drivers/crypto/dpaa_sec/dpaa_sec.h b/dpdk/drivers/crypto/dpaa_sec/dpaa_sec.h +index 7890687828..b3f2258ead 100644 +--- a/dpdk/drivers/crypto/dpaa_sec/dpaa_sec.h ++++ b/dpdk/drivers/crypto/dpaa_sec/dpaa_sec.h +@@ -1,6 +1,6 @@ + /* SPDX-License-Identifier: BSD-3-Clause + * +- * Copyright 2016-2021 NXP ++ * Copyright 2016-2022 NXP + * + */ -@@ -6682,7 +6692,7 @@ dpif_netdev_run(struct dpif *dpif) - if (pmd_rebalance && - !dp_netdev_is_reconf_required(dp) && - !ports_require_restart(dp) && -- pmd_reblance_dry_run_needed(dp) && -+ pmd_rebalance_dry_run_needed(dp) && - pmd_rebalance_dry_run(dp)) { - VLOG_INFO("PMD auto load balance dry run. " - "Requesting datapath reconfigure."); -@@ -7364,15 +7374,15 @@ static struct dp_netdev_pmd_thread * - dp_netdev_get_pmd(struct dp_netdev *dp, unsigned core_id) - { - struct dp_netdev_pmd_thread *pmd; -- const struct cmap_node *pnode; +@@ -231,7 +231,7 @@ struct dpaa_sec_job { + struct qm_sg_entry sg[MAX_JOB_SG_ENTRIES]; + }; -- pnode = cmap_find(&dp->poll_threads, hash_int(core_id, 0)); -- if (!pnode) { -- return NULL; -+ CMAP_FOR_EACH_WITH_HASH (pmd, node, hash_int(core_id, 0), -+ &dp->poll_threads) { -+ if (pmd->core_id == core_id) { -+ return dp_netdev_pmd_try_ref(pmd) ? pmd : NULL; -+ } - } -- pmd = CONTAINER_OF(pnode, struct dp_netdev_pmd_thread, node); +-#define DPAA_MAX_NB_MAX_DIGEST 32 ++#define DPAA_MAX_NB_MAX_DIGEST 64 + struct dpaa_sec_op_ctx { + struct dpaa_sec_job job; + union { +diff --git a/dpdk/drivers/crypto/dpaa_sec/dpaa_sec_raw_dp.c b/dpdk/drivers/crypto/dpaa_sec/dpaa_sec_raw_dp.c +index 522685f8cf..29f4e6d40b 100644 +--- a/dpdk/drivers/crypto/dpaa_sec/dpaa_sec_raw_dp.c ++++ b/dpdk/drivers/crypto/dpaa_sec/dpaa_sec_raw_dp.c +@@ -1,5 +1,5 @@ + /* SPDX-License-Identifier: BSD-3-Clause +- * Copyright 2021 NXP ++ * Copyright 2021-2022 NXP + */ -- return dp_netdev_pmd_try_ref(pmd) ? pmd : NULL; -+ return NULL; - } + #include +@@ -397,8 +397,8 @@ build_dpaa_raw_dp_chain_fd(uint8_t *drv_ctx, + unsigned int i; + uint16_t auth_hdr_len = ofs.ofs.cipher.head - + ofs.ofs.auth.head; +- uint16_t auth_tail_len = ofs.ofs.auth.tail; +- uint32_t auth_only_len = (auth_tail_len << 16) | auth_hdr_len; ++ uint16_t auth_tail_len; ++ uint32_t auth_only_len; + int data_len = 0, auth_len = 0, cipher_len = 0; + + for (i = 0; i < sgl->num; i++) +@@ -406,6 +406,8 @@ build_dpaa_raw_dp_chain_fd(uint8_t *drv_ctx, + + cipher_len = data_len - ofs.ofs.cipher.head - ofs.ofs.cipher.tail; + auth_len = data_len - ofs.ofs.auth.head - ofs.ofs.auth.tail; ++ auth_tail_len = auth_len - cipher_len - auth_hdr_len; ++ auth_only_len = (auth_tail_len << 16) | auth_hdr_len; + + if (sgl->num > MAX_SG_ENTRIES) { + DPAA_SEC_DP_ERR("Cipher-Auth: Max sec segs supported is %d", +@@ -448,6 +450,7 @@ build_dpaa_raw_dp_chain_fd(uint8_t *drv_ctx, + qm_sg_entry_set64(sg, dest_sgl->vec[i].iova); + sg->length = dest_sgl->vec[i].len; + } ++ sg->length -= ofs.ofs.cipher.tail; + } else { + qm_sg_entry_set64(sg, sgl->vec[0].iova); + sg->length = sgl->vec[0].len - ofs.ofs.cipher.head; +@@ -460,6 +463,7 @@ build_dpaa_raw_dp_chain_fd(uint8_t *drv_ctx, + qm_sg_entry_set64(sg, sgl->vec[i].iova); + sg->length = sgl->vec[i].len; + } ++ sg->length -= ofs.ofs.cipher.tail; + } - /* Sets the 'struct dp_netdev_pmd_thread' for non-pmd threads. */ -@@ -7505,6 +7515,7 @@ dp_netdev_destroy_pmd(struct dp_netdev_pmd_thread *pmd) - seq_destroy(pmd->reload_seq); - ovs_mutex_destroy(&pmd->port_mutex); - ovs_mutex_destroy(&pmd->bond_mutex); -+ free(pmd->netdev_input_func_userdata); - free(pmd); + if (is_encode(ses)) { +diff --git a/dpdk/drivers/crypto/ipsec_mb/ipsec_mb_ops.c b/dpdk/drivers/crypto/ipsec_mb/ipsec_mb_ops.c +index 189262c4ad..58ea4ee476 100644 +--- a/dpdk/drivers/crypto/ipsec_mb/ipsec_mb_ops.c ++++ b/dpdk/drivers/crypto/ipsec_mb/ipsec_mb_ops.c +@@ -221,8 +221,11 @@ ipsec_mb_qp_setup(struct rte_cryptodev *dev, uint16_t qp_id, + IMB_VERSION_STR, IMB_MP_REQ_VER_STR); + return -EINVAL; + #endif +- if (dev->data->queue_pairs[qp_id] != NULL) +- qp = dev->data->queue_pairs[qp_id]; ++ qp = dev->data->queue_pairs[qp_id]; ++ if (qp == NULL) { ++ IPSEC_MB_LOG(ERR, "Primary process hasn't configured device qp."); ++ return -EINVAL; ++ } + } else { + /* Free memory prior to re-allocation if needed. */ + if (dev->data->queue_pairs[qp_id] != NULL) +@@ -291,8 +294,7 @@ ipsec_mb_qp_setup(struct rte_cryptodev *dev, uint16_t qp_id, + if (qp->mb_mgr_mz) + rte_memzone_free(qp->mb_mgr_mz); + #endif +- if (qp) +- rte_free(qp); ++ rte_free(qp); + return ret; } -diff --git a/lib/dpif-netlink.c b/lib/dpif-netlink.c -index 71e35ccdda..06e1e8ca02 100644 ---- a/lib/dpif-netlink.c -+++ b/lib/dpif-netlink.c -@@ -85,7 +85,7 @@ enum { MAX_PORTS = USHRT_MAX }; - #define EPOLLEXCLUSIVE (1u << 28) - #endif +diff --git a/dpdk/drivers/crypto/ipsec_mb/ipsec_mb_private.h b/dpdk/drivers/crypto/ipsec_mb/ipsec_mb_private.h +index 866722d6f4..e53101acf1 100644 +--- a/dpdk/drivers/crypto/ipsec_mb/ipsec_mb_private.h ++++ b/dpdk/drivers/crypto/ipsec_mb/ipsec_mb_private.h +@@ -191,13 +191,13 @@ ipsec_mb_parse_xform(const struct rte_crypto_sym_xform *xform, + const struct rte_crypto_sym_xform **cipher_xform, + const struct rte_crypto_sym_xform **aead_xform) + { +- const struct rte_crypto_sym_xform *next = xform->next; +- + if (xform == NULL) { + *mode = IPSEC_MB_OP_NOT_SUPPORTED; + return -ENOTSUP; + } --#define OVS_DP_F_UNSUPPORTED (1 << 31); -+#define OVS_DP_F_UNSUPPORTED (1u << 31); ++ const struct rte_crypto_sym_xform *next = xform->next; ++ + if (xform->type == RTE_CRYPTO_SYM_XFORM_CIPHER) { + if (next == NULL) { + if (xform->cipher.op == RTE_CRYPTO_CIPHER_OP_ENCRYPT) { +diff --git a/dpdk/drivers/crypto/ipsec_mb/pmd_aesni_gcm.c b/dpdk/drivers/crypto/ipsec_mb/pmd_aesni_gcm.c +index 2c203795ab..2c033c6f28 100644 +--- a/dpdk/drivers/crypto/ipsec_mb/pmd_aesni_gcm.c ++++ b/dpdk/drivers/crypto/ipsec_mb/pmd_aesni_gcm.c +@@ -96,7 +96,9 @@ aesni_gcm_session_configure(IMB_MGR *mb_mgr, void *session, + sess->iv.length = auth_xform->auth.iv.length; + key_length = auth_xform->auth.key.length; + key = auth_xform->auth.key.data; +- sess->req_digest_length = auth_xform->auth.digest_length; ++ sess->req_digest_length = ++ RTE_MIN(auth_xform->auth.digest_length, ++ DIGEST_LENGTH_MAX); + break; + case IPSEC_MB_OP_AEAD_AUTHENTICATED_ENCRYPT: + case IPSEC_MB_OP_AEAD_AUTHENTICATED_DECRYPT: +@@ -116,7 +118,9 @@ aesni_gcm_session_configure(IMB_MGR *mb_mgr, void *session, + key_length = aead_xform->aead.key.length; + key = aead_xform->aead.key.data; + sess->aad_length = aead_xform->aead.aad_length; +- sess->req_digest_length = aead_xform->aead.digest_length; ++ sess->req_digest_length = ++ RTE_MIN(aead_xform->aead.digest_length, ++ DIGEST_LENGTH_MAX); + break; + default: + IPSEC_MB_LOG( +@@ -713,19 +717,17 @@ aesni_gcm_process_bulk(struct rte_cryptodev *dev, + __rte_unused union rte_crypto_sym_ofs ofs, + struct rte_crypto_sym_vec *vec) + { +- void *sess_priv; + struct aesni_gcm_session *s; + struct gcm_context_data gdata_ctx; + IMB_MGR *mb_mgr; + +- sess_priv = get_sym_session_private_data(sess, dev->driver_id); +- if (unlikely(sess_priv == NULL)) { ++ s = (struct aesni_gcm_session *) get_sym_session_private_data(sess, ++ dev->driver_id); ++ if (unlikely(s == NULL)) { + aesni_gcm_fill_error_code(vec, EINVAL); + return 0; + } - /* This PID is not used by the kernel datapath when using dispatch per CPU, - * but it is required to be set (not zero). */ -diff --git a/lib/dynamic-string.c b/lib/dynamic-string.c -index fd0127ed17..3b4520f87c 100644 ---- a/lib/dynamic-string.c -+++ b/lib/dynamic-string.c -@@ -152,7 +152,10 @@ ds_put_format_valist(struct ds *ds, const char *format, va_list args_) +- s = sess_priv; +- + /* get per-thread MB MGR, create one if needed */ + mb_mgr = get_per_thread_mb_mgr(); + if (unlikely(mb_mgr == NULL)) +diff --git a/dpdk/drivers/crypto/ipsec_mb/pmd_aesni_mb.c b/dpdk/drivers/crypto/ipsec_mb/pmd_aesni_mb.c +index a308d42ffa..536a586e98 100644 +--- a/dpdk/drivers/crypto/ipsec_mb/pmd_aesni_mb.c ++++ b/dpdk/drivers/crypto/ipsec_mb/pmd_aesni_mb.c +@@ -918,7 +918,9 @@ aesni_mb_set_docsis_sec_session_parameters( + + static inline uint64_t + auth_start_offset(struct rte_crypto_op *op, struct aesni_mb_session *session, +- uint32_t oop) ++ uint32_t oop, const uint32_t auth_offset, ++ const uint32_t cipher_offset, const uint32_t auth_length, ++ const uint32_t cipher_length) + { + struct rte_mbuf *m_src, *m_dst; + uint8_t *p_src, *p_dst; +@@ -927,7 +929,7 @@ auth_start_offset(struct rte_crypto_op *op, struct aesni_mb_session *session, + + /* Only cipher then hash needs special calculation. */ + if (!oop || session->chain_order != IMB_ORDER_CIPHER_HASH) +- return op->sym->auth.data.offset; ++ return auth_offset; + + m_src = op->sym->m_src; + m_dst = op->sym->m_dst; +@@ -935,24 +937,24 @@ auth_start_offset(struct rte_crypto_op *op, struct aesni_mb_session *session, + p_src = rte_pktmbuf_mtod(m_src, uint8_t *); + p_dst = rte_pktmbuf_mtod(m_dst, uint8_t *); + u_src = (uintptr_t)p_src; +- u_dst = (uintptr_t)p_dst + op->sym->auth.data.offset; ++ u_dst = (uintptr_t)p_dst + auth_offset; + + /** + * Copy the content between cipher offset and auth offset for generating + * correct digest. + */ +- if (op->sym->cipher.data.offset > op->sym->auth.data.offset) +- memcpy(p_dst + op->sym->auth.data.offset, +- p_src + op->sym->auth.data.offset, +- op->sym->cipher.data.offset - +- op->sym->auth.data.offset); ++ if (cipher_offset > auth_offset) ++ memcpy(p_dst + auth_offset, ++ p_src + auth_offset, ++ cipher_offset - ++ auth_offset); + + /** + * Copy the content between (cipher offset + length) and (auth offset + + * length) for generating correct digest + */ +- cipher_end = op->sym->cipher.data.offset + op->sym->cipher.data.length; +- auth_end = op->sym->auth.data.offset + op->sym->auth.data.length; ++ cipher_end = cipher_offset + cipher_length; ++ auth_end = auth_offset + auth_length; + if (cipher_end < auth_end) + memcpy(p_dst + cipher_end, p_src + cipher_end, + auth_end - cipher_end); +@@ -1099,6 +1101,10 @@ set_mb_job_params(IMB_JOB *job, struct ipsec_mb_qp *qp, + struct aesni_mb_qp_data *qp_data = ipsec_mb_get_qp_private_data(qp); + struct aesni_mb_session *session; + uint32_t m_offset, oop; ++ uint32_t auth_off_in_bytes; ++ uint32_t ciph_off_in_bytes; ++ uint32_t auth_len_in_bytes; ++ uint32_t ciph_len_in_bytes; + + session = ipsec_mb_get_session_private(qp, op); + if (session == NULL) { +@@ -1207,6 +1213,7 @@ set_mb_job_params(IMB_JOB *job, struct ipsec_mb_qp *qp, + if (job->cipher_mode == IMB_CIPHER_ZUC_EEA3) { + job->enc_keys = session->cipher.zuc_cipher_key; + job->dec_keys = session->cipher.zuc_cipher_key; ++ m_offset >>= 3; + } else if (job->cipher_mode == IMB_CIPHER_SNOW3G_UEA2_BITLEN) { + job->enc_keys = &session->cipher.pKeySched_snow3g_cipher; + m_offset = 0; +@@ -1264,9 +1271,6 @@ set_mb_job_params(IMB_JOB *job, struct ipsec_mb_qp *qp, + + switch (job->hash_alg) { + case IMB_AUTH_AES_CCM: +- job->cipher_start_src_offset_in_bytes = +- op->sym->aead.data.offset; +- job->msg_len_to_cipher_in_bytes = op->sym->aead.data.length; + job->hash_start_src_offset_in_bytes = op->sym->aead.data.offset; + job->msg_len_to_hash_in_bytes = op->sym->aead.data.length; + +@@ -1276,21 +1280,13 @@ set_mb_job_params(IMB_JOB *job, struct ipsec_mb_qp *qp, + + case IMB_AUTH_AES_GMAC: + if (session->cipher.mode == IMB_CIPHER_GCM) { +- job->cipher_start_src_offset_in_bytes = +- op->sym->aead.data.offset; + job->hash_start_src_offset_in_bytes = + op->sym->aead.data.offset; +- job->msg_len_to_cipher_in_bytes = +- op->sym->aead.data.length; + job->msg_len_to_hash_in_bytes = + op->sym->aead.data.length; +- } else { +- job->cipher_start_src_offset_in_bytes = +- op->sym->auth.data.offset; +- job->hash_start_src_offset_in_bytes = +- op->sym->auth.data.offset; +- job->msg_len_to_cipher_in_bytes = 0; ++ } else { /* AES-GMAC only, only AAD used */ + job->msg_len_to_hash_in_bytes = 0; ++ job->hash_start_src_offset_in_bytes = 0; + } + + job->iv = rte_crypto_op_ctod_offset(op, uint8_t *, +@@ -1298,36 +1294,100 @@ set_mb_job_params(IMB_JOB *job, struct ipsec_mb_qp *qp, + break; - va_copy(args, args_); - available = ds->string ? ds->allocated - ds->length + 1 : 0; -- needed = vsnprintf(&ds->string[ds->length], available, format, args); -+ needed = vsnprintf(ds->string -+ ? &ds->string[ds->length] -+ : NULL, -+ available, format, args); - va_end(args); + case IMB_AUTH_CHACHA20_POLY1305: +- job->cipher_start_src_offset_in_bytes = +- op->sym->aead.data.offset; + job->hash_start_src_offset_in_bytes = + op->sym->aead.data.offset; +- job->msg_len_to_cipher_in_bytes = +- op->sym->aead.data.length; + job->msg_len_to_hash_in_bytes = + op->sym->aead.data.length; + + job->iv = rte_crypto_op_ctod_offset(op, uint8_t *, + session->iv.offset); + break; +- default: +- /* For SNOW3G, length and offsets are already in bits */ +- job->cipher_start_src_offset_in_bytes = +- op->sym->cipher.data.offset; +- job->msg_len_to_cipher_in_bytes = op->sym->cipher.data.length; ++ /* ZUC and SNOW3G require length in bits and offset in bytes */ ++ case IMB_AUTH_ZUC_EIA3_BITLEN: ++ case IMB_AUTH_ZUC256_EIA3_BITLEN: ++ case IMB_AUTH_SNOW3G_UIA2_BITLEN: ++ auth_off_in_bytes = op->sym->auth.data.offset >> 3; ++ ciph_off_in_bytes = op->sym->cipher.data.offset >> 3; ++ auth_len_in_bytes = op->sym->auth.data.length >> 3; ++ ciph_len_in_bytes = op->sym->cipher.data.length >> 3; ++ ++ job->hash_start_src_offset_in_bytes = auth_start_offset(op, ++ session, oop, auth_off_in_bytes, ++ ciph_off_in_bytes, auth_len_in_bytes, ++ ciph_len_in_bytes); ++ job->msg_len_to_hash_in_bits = op->sym->auth.data.length; ++ ++ job->iv = rte_crypto_op_ctod_offset(op, uint8_t *, ++ session->iv.offset); ++ break; ++ ++ /* KASUMI requires lengths and offset in bytes */ ++ case IMB_AUTH_KASUMI_UIA1: ++ auth_off_in_bytes = op->sym->auth.data.offset >> 3; ++ ciph_off_in_bytes = op->sym->cipher.data.offset >> 3; ++ auth_len_in_bytes = op->sym->auth.data.length >> 3; ++ ciph_len_in_bytes = op->sym->cipher.data.length >> 3; + + job->hash_start_src_offset_in_bytes = auth_start_offset(op, +- session, oop); ++ session, oop, auth_off_in_bytes, ++ ciph_off_in_bytes, auth_len_in_bytes, ++ ciph_len_in_bytes); ++ job->msg_len_to_hash_in_bytes = auth_len_in_bytes; ++ ++ job->iv = rte_crypto_op_ctod_offset(op, uint8_t *, ++ session->iv.offset); ++ break; ++ ++ default: ++ job->hash_start_src_offset_in_bytes = auth_start_offset(op, ++ session, oop, op->sym->auth.data.offset, ++ op->sym->cipher.data.offset, ++ op->sym->auth.data.length, ++ op->sym->cipher.data.length); + job->msg_len_to_hash_in_bytes = op->sym->auth.data.length; - if (needed < available) { -@@ -162,7 +165,8 @@ ds_put_format_valist(struct ds *ds, const char *format, va_list args_) + job->iv = rte_crypto_op_ctod_offset(op, uint8_t *, + session->iv.offset); + } - va_copy(args, args_); - available = ds->allocated - ds->length + 1; -- needed = vsnprintf(&ds->string[ds->length], available, format, args); -+ needed = vsnprintf(&ds->string[ds->length], -+ available, format, args); - va_end(args); +- if (job->cipher_mode == IMB_CIPHER_ZUC_EEA3) +- job->msg_len_to_cipher_in_bytes >>= 3; +- else if (job->hash_alg == IMB_AUTH_KASUMI_UIA1) +- job->msg_len_to_hash_in_bytes >>= 3; ++ switch (job->cipher_mode) { ++ /* ZUC requires length and offset in bytes */ ++ case IMB_CIPHER_ZUC_EEA3: ++ job->cipher_start_src_offset_in_bytes = ++ op->sym->cipher.data.offset >> 3; ++ job->msg_len_to_cipher_in_bytes = ++ op->sym->cipher.data.length >> 3; ++ break; ++ /* ZUC and SNOW3G require length and offset in bits */ ++ case IMB_CIPHER_SNOW3G_UEA2_BITLEN: ++ case IMB_CIPHER_KASUMI_UEA1_BITLEN: ++ job->cipher_start_src_offset_in_bits = ++ op->sym->cipher.data.offset; ++ job->msg_len_to_cipher_in_bits = ++ op->sym->cipher.data.length; ++ break; ++ case IMB_CIPHER_GCM: ++ if (session->cipher.mode == IMB_CIPHER_NULL) { ++ /* AES-GMAC only (only AAD used) */ ++ job->msg_len_to_cipher_in_bytes = 0; ++ job->cipher_start_src_offset_in_bytes = 0; ++ } else { ++ job->cipher_start_src_offset_in_bytes = ++ op->sym->aead.data.offset; ++ job->msg_len_to_cipher_in_bytes = op->sym->aead.data.length; ++ } ++ break; ++ case IMB_CIPHER_CCM: ++ case IMB_CIPHER_CHACHA20_POLY1305: ++ job->cipher_start_src_offset_in_bytes = ++ op->sym->aead.data.offset; ++ job->msg_len_to_cipher_in_bytes = op->sym->aead.data.length; ++ break; ++ default: ++ job->cipher_start_src_offset_in_bytes = ++ op->sym->cipher.data.offset; ++ job->msg_len_to_cipher_in_bytes = op->sym->cipher.data.length; ++ } + + /* Set user data to be crypto operation data struct */ + job->user_data = op; +diff --git a/dpdk/drivers/crypto/ipsec_mb/pmd_aesni_mb_priv.h b/dpdk/drivers/crypto/ipsec_mb/pmd_aesni_mb_priv.h +index d37cc787a0..d177961ea5 100644 +--- a/dpdk/drivers/crypto/ipsec_mb/pmd_aesni_mb_priv.h ++++ b/dpdk/drivers/crypto/ipsec_mb/pmd_aesni_mb_priv.h +@@ -848,7 +848,7 @@ struct aesni_mb_session { + + struct gcm_key_data gcm_key; + /* *< Expanded GCM key */ +- uint8_t zuc_cipher_key[16]; ++ uint8_t zuc_cipher_key[32]; + /* *< ZUC cipher key */ + snow3g_key_schedule_t pKeySched_snow3g_cipher; + /* *< SNOW3G scheduled cipher key */ +@@ -893,7 +893,7 @@ struct aesni_mb_session { + /* *< k3. */ + } cmac; + /* *< Expanded XCBC authentication keys */ +- uint8_t zuc_auth_key[16]; ++ uint8_t zuc_auth_key[32]; + /* *< ZUC authentication key */ + snow3g_key_schedule_t pKeySched_snow3g_auth; + /* *< SNOW3G scheduled authentication key */ +diff --git a/dpdk/drivers/crypto/ipsec_mb/pmd_snow3g.c b/dpdk/drivers/crypto/ipsec_mb/pmd_snow3g.c +index ebc9a0b562..9a85f46721 100644 +--- a/dpdk/drivers/crypto/ipsec_mb/pmd_snow3g.c ++++ b/dpdk/drivers/crypto/ipsec_mb/pmd_snow3g.c +@@ -422,12 +422,13 @@ process_op_bit(struct rte_crypto_op *op, struct snow3g_session *session, + op->sym->session = NULL; + } - ovs_assert(needed < available); -@@ -198,10 +202,11 @@ ds_put_strftime_msec(struct ds *ds, const char *template, long long int when, - localtime_msec(when, &tm); - } +- enqueued_op = rte_ring_enqueue_burst(qp->ingress_queue, +- (void **)&op, processed_op, NULL); ++ if (unlikely(processed_op != 1)) ++ return 0; ++ enqueued_op = rte_ring_enqueue(qp->ingress_queue, op); + qp->stats.enqueued_count += enqueued_op; + *accumulated_enqueued_ops += enqueued_op; -+ ds_reserve(ds, 64); - for (;;) { -- size_t avail = ds->string ? ds->allocated - ds->length + 1 : 0; -- size_t used = strftime_msec(&ds->string[ds->length], avail, template, -- &tm); -+ size_t avail = ds->allocated - ds->length + 1; -+ char *dest = &ds->string[ds->length]; -+ size_t used = strftime_msec(dest, avail, template, &tm); - if (used) { - ds->length += used; - return; -diff --git a/lib/fat-rwlock.c b/lib/fat-rwlock.c -index d913b2088f..771ccc9737 100644 ---- a/lib/fat-rwlock.c -+++ b/lib/fat-rwlock.c -@@ -97,14 +97,14 @@ fat_rwlock_init(struct fat_rwlock *rwlock) - void - fat_rwlock_destroy(struct fat_rwlock *rwlock) +- return enqueued_op; ++ return 1; + } + + static uint16_t +diff --git a/dpdk/drivers/crypto/ipsec_mb/pmd_zuc.c b/dpdk/drivers/crypto/ipsec_mb/pmd_zuc.c +index 2eae1d1ec7..e36c7092d6 100644 +--- a/dpdk/drivers/crypto/ipsec_mb/pmd_zuc.c ++++ b/dpdk/drivers/crypto/ipsec_mb/pmd_zuc.c +@@ -166,7 +166,7 @@ process_zuc_hash_op(struct ipsec_mb_qp *qp, struct rte_crypto_op **ops, + + hash_keys[i] = sess->pKey_hash; + if (sess->auth_op == RTE_CRYPTO_AUTH_OP_VERIFY) +- dst[i] = (uint32_t *)qp_data->temp_digest; ++ dst[i] = (uint32_t *)qp_data->temp_digest[i]; + else + dst[i] = (uint32_t *)ops[i]->sym->auth.digest.data; + +@@ -198,7 +198,7 @@ process_ops(struct rte_crypto_op **ops, enum ipsec_mb_operation op_type, + struct ipsec_mb_qp *qp, uint8_t num_ops) { -- struct fat_rwlock_slot *slot, *next; -+ struct fat_rwlock_slot *slot; + unsigned int i; +- unsigned int processed_ops; ++ unsigned int processed_ops = 0; + + switch (op_type) { + case IPSEC_MB_OP_ENCRYPT_ONLY: +@@ -212,18 +212,21 @@ process_ops(struct rte_crypto_op **ops, enum ipsec_mb_operation op_type, + num_ops); + break; + case IPSEC_MB_OP_ENCRYPT_THEN_HASH_GEN: ++ case IPSEC_MB_OP_DECRYPT_THEN_HASH_VERIFY: + processed_ops = process_zuc_cipher_op(qp, ops, sessions, + num_ops); + process_zuc_hash_op(qp, ops, sessions, processed_ops); + break; + case IPSEC_MB_OP_HASH_VERIFY_THEN_DECRYPT: ++ case IPSEC_MB_OP_HASH_GEN_THEN_ENCRYPT: + processed_ops = process_zuc_hash_op(qp, ops, sessions, + num_ops); + process_zuc_cipher_op(qp, ops, sessions, processed_ops); + break; + default: + /* Operation not supported. */ +- processed_ops = 0; ++ for (i = 0; i < num_ops; i++) ++ ops[i]->status = RTE_CRYPTO_OP_STATUS_INVALID_SESSION; + } - /* Order is important here. By destroying the thread-specific data first, - * before we destroy the slots, we ensure that the thread-specific - * data destructor can't race with our loop below. */ - ovsthread_key_delete(rwlock->key); + for (i = 0; i < num_ops; i++) { +@@ -256,6 +259,7 @@ zuc_pmd_dequeue_burst(void *queue_pair, + + struct zuc_session *curr_sess; + struct zuc_session *sessions[ZUC_MAX_BURST]; ++ struct rte_crypto_op *int_c_ops[ZUC_MAX_BURST]; + enum ipsec_mb_operation prev_zuc_op = IPSEC_MB_OP_NOT_SUPPORTED; + enum ipsec_mb_operation curr_zuc_op; + struct ipsec_mb_qp *qp = queue_pair; +@@ -287,11 +291,11 @@ zuc_pmd_dequeue_burst(void *queue_pair, + */ + if (burst_size == 0) { + prev_zuc_op = curr_zuc_op; +- c_ops[0] = curr_c_op; ++ int_c_ops[0] = curr_c_op; + sessions[0] = curr_sess; + burst_size++; + } else if (curr_zuc_op == prev_zuc_op) { +- c_ops[burst_size] = curr_c_op; ++ int_c_ops[burst_size] = curr_c_op; + sessions[burst_size] = curr_sess; + burst_size++; + /* +@@ -299,7 +303,7 @@ zuc_pmd_dequeue_burst(void *queue_pair, + * process them, and start a new batch. + */ + if (burst_size == ZUC_MAX_BURST) { +- processed_ops = process_ops(c_ops, curr_zuc_op, ++ processed_ops = process_ops(int_c_ops, curr_zuc_op, + sessions, qp, burst_size); + if (processed_ops < burst_size) { + burst_size = 0; +@@ -313,7 +317,7 @@ zuc_pmd_dequeue_burst(void *queue_pair, + * Different operation type, process the ops + * of the previous type. + */ +- processed_ops = process_ops(c_ops, prev_zuc_op, ++ processed_ops = process_ops(int_c_ops, prev_zuc_op, + sessions, qp, burst_size); + if (processed_ops < burst_size) { + burst_size = 0; +@@ -323,7 +327,7 @@ zuc_pmd_dequeue_burst(void *queue_pair, + burst_size = 0; + prev_zuc_op = curr_zuc_op; + +- c_ops[0] = curr_c_op; ++ int_c_ops[0] = curr_c_op; + sessions[0] = curr_sess; + burst_size++; + } +@@ -331,7 +335,7 @@ zuc_pmd_dequeue_burst(void *queue_pair, + + if (burst_size != 0) { + /* Process the crypto ops of the last operation type. */ +- processed_ops = process_ops(c_ops, prev_zuc_op, ++ processed_ops = process_ops(int_c_ops, prev_zuc_op, + sessions, qp, burst_size); + } -- LIST_FOR_EACH_SAFE (slot, next, list_node, &rwlock->threads) { -+ LIST_FOR_EACH_SAFE (slot, list_node, &rwlock->threads) { - free_slot(slot); - } - ovs_mutex_destroy(&rwlock->mutex); -diff --git a/lib/hindex.h b/lib/hindex.h -index 876c5a9e39..ea7402587e 100644 ---- a/lib/hindex.h -+++ b/lib/hindex.h -@@ -128,18 +128,38 @@ void hindex_remove(struct hindex *, struct hindex_node *); - * Evaluates HASH only once. +diff --git a/dpdk/drivers/crypto/ipsec_mb/pmd_zuc_priv.h b/dpdk/drivers/crypto/ipsec_mb/pmd_zuc_priv.h +index 46d5bfae37..76fd6758c2 100644 +--- a/dpdk/drivers/crypto/ipsec_mb/pmd_zuc_priv.h ++++ b/dpdk/drivers/crypto/ipsec_mb/pmd_zuc_priv.h +@@ -75,7 +75,7 @@ struct zuc_session { + + struct zuc_qp_data { + +- uint8_t temp_digest[ZUC_DIGEST_LENGTH]; ++ uint8_t temp_digest[ZUC_MAX_BURST][ZUC_DIGEST_LENGTH]; + /* *< Buffers used to store the digest generated + * by the driver when verifying a digest provided + * by the user (using authentication verify operation) +diff --git a/dpdk/drivers/crypto/mlx5/mlx5_crypto.c b/dpdk/drivers/crypto/mlx5/mlx5_crypto.c +index 421c23748a..36db31aae5 100644 +--- a/dpdk/drivers/crypto/mlx5/mlx5_crypto.c ++++ b/dpdk/drivers/crypto/mlx5/mlx5_crypto.c +@@ -952,6 +952,7 @@ mlx5_crypto_dev_probe(struct mlx5_common_device *cdev) + ret = mlx5_crypto_configure_wqe_size(priv, + cdev->config.hca_attr.max_wqe_sz_sq, devarg_prms.max_segs_num); + if (ret) { ++ claim_zero(mlx5_devx_cmd_destroy(priv->login_obj)); + mlx5_devx_uar_release(&priv->uar); + rte_cryptodev_pmd_destroy(priv->crypto_dev); + return -1; +diff --git a/dpdk/drivers/crypto/octeontx/otx_cryptodev_hw_access.c b/dpdk/drivers/crypto/octeontx/otx_cryptodev_hw_access.c +index 20b288334a..27604459e4 100644 +--- a/dpdk/drivers/crypto/octeontx/otx_cryptodev_hw_access.c ++++ b/dpdk/drivers/crypto/octeontx/otx_cryptodev_hw_access.c +@@ -296,7 +296,7 @@ cpt_vq_init(struct cpt_vf *cptvf, uint8_t group) + /* CPT VF device initialization */ + otx_cpt_vfvq_init(cptvf); + +- /* Send msg to PF to assign currnet Q to required group */ ++ /* Send msg to PF to assign current Q to required group */ + cptvf->vfgrp = group; + err = otx_cpt_send_vf_grp_msg(cptvf, group); + if (err) { +diff --git a/dpdk/drivers/crypto/octeontx/otx_cryptodev_mbox.h b/dpdk/drivers/crypto/octeontx/otx_cryptodev_mbox.h +index 508f3afd47..c1eedc1b9e 100644 +--- a/dpdk/drivers/crypto/octeontx/otx_cryptodev_mbox.h ++++ b/dpdk/drivers/crypto/octeontx/otx_cryptodev_mbox.h +@@ -70,7 +70,7 @@ void + otx_cpt_handle_mbox_intr(struct cpt_vf *cptvf); + + /* +- * Checks if VF is able to comminicate with PF ++ * Checks if VF is able to communicate with PF + * and also gets the CPT number this VF is associated to. + */ + int +diff --git a/dpdk/drivers/crypto/octeontx/otx_cryptodev_ops.c b/dpdk/drivers/crypto/octeontx/otx_cryptodev_ops.c +index 9e8fd495cf..f7ca8a8a8e 100644 +--- a/dpdk/drivers/crypto/octeontx/otx_cryptodev_ops.c ++++ b/dpdk/drivers/crypto/octeontx/otx_cryptodev_ops.c +@@ -558,7 +558,7 @@ otx_cpt_enq_single_sym(struct cpt_instance *instance, + &mdata, (void **)&prep_req); + + if (unlikely(ret)) { +- CPT_LOG_DP_ERR("prep cryto req : op %p, cpt_op 0x%x " ++ CPT_LOG_DP_ERR("prep crypto req : op %p, cpt_op 0x%x " + "ret 0x%x", op, (unsigned int)cpt_op, ret); + return NULL; + } +diff --git a/dpdk/drivers/crypto/openssl/rte_openssl_pmd.c b/dpdk/drivers/crypto/openssl/rte_openssl_pmd.c +index 5794ed8159..5977bc746c 100644 +--- a/dpdk/drivers/crypto/openssl/rte_openssl_pmd.c ++++ b/dpdk/drivers/crypto/openssl/rte_openssl_pmd.c +@@ -2,6 +2,8 @@ + * Copyright(c) 2016-2017 Intel Corporation */ - #define HINDEX_FOR_EACH_WITH_HASH(NODE, MEMBER, HASH, HINDEX) \ -- for (INIT_CONTAINER(NODE, hindex_node_with_hash(HINDEX, HASH), MEMBER); \ -- NODE != OBJECT_CONTAINING(NULL, NODE, MEMBER); \ -- ASSIGN_CONTAINER(NODE, (NODE)->MEMBER.s, MEMBER)) -+ for (INIT_MULTIVAR(NODE, MEMBER, hindex_node_with_hash(HINDEX, HASH), \ -+ struct hindex_node); \ -+ CONDITION_MULTIVAR(NODE, MEMBER, ITER_VAR(NODE) != NULL); \ -+ UPDATE_MULTIVAR(NODE, ITER_VAR(NODE)->s)) - /* Safe when NODE may be freed (not needed when NODE may be removed from the - * hash map but its members remain accessible and intact). */ --#define HINDEX_FOR_EACH_WITH_HASH_SAFE(NODE, NEXT, MEMBER, HASH, HINDEX) \ -- for (INIT_CONTAINER(NODE, hindex_node_with_hash(HINDEX, HASH), MEMBER); \ -- (NODE != OBJECT_CONTAINING(NULL, NODE, MEMBER) \ -- ? INIT_CONTAINER(NEXT, (NODE)->MEMBER.s, MEMBER), 1 \ -- : 0); \ -- (NODE) = (NEXT)) -+#define HINDEX_FOR_EACH_WITH_HASH_SAFE_LONG(NODE, NEXT, MEMBER, HASH, HINDEX) \ -+ for (INIT_MULTIVAR_SAFE_LONG(NODE, NEXT, MEMBER, \ -+ hindex_node_with_hash(HINDEX, HASH), \ -+ struct hindex_node); \ -+ CONDITION_MULTIVAR_SAFE_LONG(NODE, NEXT, MEMBER, \ -+ ITER_VAR(NODE) != NULL, \ -+ ITER_VAR(NEXT) = ITER_VAR(NODE)->s, \ -+ ITER_VAR(NEXT) != NULL); \ -+ UPDATE_MULTIVAR_SAFE_LONG(NODE, NEXT)) -+ -+/* Short version of HINDEX_FOR_EACH_WITH_HASH_SAFE. */ -+#define HINDEX_FOR_EACH_WITH_HASH_SAFE_SHORT(NODE, MEMBER, HASH, HINDEX) \ -+ for (INIT_MULTIVAR_SAFE_SHORT(NODE, MEMBER, \ -+ hindex_node_with_hash(HINDEX, HASH), \ -+ struct hindex_node); \ -+ CONDITION_MULTIVAR_SAFE_SHORT(NODE, MEMBER, \ -+ ITER_VAR(NODE) != NULL, \ -+ ITER_NEXT_VAR(NODE) = ITER_VAR(NODE)->s); \ -+ UPDATE_MULTIVAR_SAFE_SHORT(NODE)) -+ -+#define HINDEX_FOR_EACH_WITH_HASH_SAFE(...) \ -+ OVERLOAD_SAFE_MACRO(HINDEX_FOR_EACH_WITH_HASH_SAFE_LONG, \ -+ HINDEX_FOR_EACH_WITH_HASH_SAFE_SHORT, \ -+ 5, __VA_ARGS__) ++#define OPENSSL_API_COMPAT 0x10100000L ++ + #include + #include + #include +diff --git a/dpdk/drivers/crypto/openssl/rte_openssl_pmd_ops.c b/dpdk/drivers/crypto/openssl/rte_openssl_pmd_ops.c +index 52715f86f8..35c4ad13ba 100644 +--- a/dpdk/drivers/crypto/openssl/rte_openssl_pmd_ops.c ++++ b/dpdk/drivers/crypto/openssl/rte_openssl_pmd_ops.c +@@ -2,6 +2,8 @@ + * Copyright(c) 2016-2017 Intel Corporation + */ + ++#define OPENSSL_API_COMPAT 0x10100000L + + #include - /* Returns the head node in 'hindex' with the given 'hash', or a null pointer - * if no nodes have that hash value. */ -@@ -157,19 +177,36 @@ hindex_node_with_hash(const struct hindex *hindex, size_t hash) - /* Iteration. */ + #include +diff --git a/dpdk/drivers/crypto/qat/qat_asym.c b/dpdk/drivers/crypto/qat/qat_asym.c +index f893508030..bd0bf5f0cb 100644 +--- a/dpdk/drivers/crypto/qat/qat_asym.c ++++ b/dpdk/drivers/crypto/qat/qat_asym.c +@@ -97,7 +97,7 @@ static void qat_clear_arrays_by_alg(struct qat_asym_op_cookie *cookie, + qat_clear_arrays(cookie, QAT_ASYM_MODINV_NUM_IN_PARAMS, + QAT_ASYM_MODINV_NUM_OUT_PARAMS, alg_size); + else if (xform->xform_type == RTE_CRYPTO_ASYM_XFORM_RSA) { +- if (xform->rsa.key_type == RTE_RSA_KET_TYPE_QT) ++ if (xform->rsa.key_type == RTE_RSA_KEY_TYPE_QT) + qat_clear_arrays_crt(cookie, alg_size); + else { + qat_clear_arrays(cookie, QAT_ASYM_RSA_NUM_IN_PARAMS, +@@ -109,7 +109,7 @@ static void qat_clear_arrays_by_alg(struct qat_asym_op_cookie *cookie, + static int qat_asym_check_nonzero(rte_crypto_param n) + { + if (n.length < 8) { +- /* Not a case for any cryptograpic function except for DH ++ /* Not a case for any cryptographic function except for DH + * generator which very often can be of one byte length + */ + size_t i; +@@ -370,7 +370,7 @@ qat_asym_fill_arrays(struct rte_crypto_asym_op *asym_op, + return -(EINVAL); + } + } +- if (xform->rsa.key_type == RTE_RSA_KET_TYPE_QT) { ++ if (xform->rsa.key_type == RTE_RSA_KEY_TYPE_QT) { + + qat_req->input_param_count = + QAT_ASYM_RSA_QT_NUM_IN_PARAMS; +diff --git a/dpdk/drivers/crypto/qat/qat_sym.c b/dpdk/drivers/crypto/qat/qat_sym.c +index 93b257522b..0dd83ee2ee 100644 +--- a/dpdk/drivers/crypto/qat/qat_sym.c ++++ b/dpdk/drivers/crypto/qat/qat_sym.c +@@ -2,6 +2,8 @@ + * Copyright(c) 2015-2019 Intel Corporation + */ - /* Iterates through every node in HINDEX. */ --#define HINDEX_FOR_EACH(NODE, MEMBER, HINDEX) \ -- for (INIT_CONTAINER(NODE, hindex_first(HINDEX), MEMBER); \ -- NODE != OBJECT_CONTAINING(NULL, NODE, MEMBER); \ -- ASSIGN_CONTAINER(NODE, hindex_next(HINDEX, &(NODE)->MEMBER), MEMBER)) -+#define HINDEX_FOR_EACH(NODE, MEMBER, HINDEX) \ -+ for (INIT_MULTIVAR(NODE, MEMBER, hindex_first(HINDEX), \ -+ struct hindex_node); \ -+ CONDITION_MULTIVAR(NODE, MEMBER, ITER_VAR(NODE) != NULL); \ -+ UPDATE_MULTIVAR(NODE, hindex_next(HINDEX, ITER_VAR(NODE)))) ++#define OPENSSL_API_COMPAT 0x10100000L ++ + #include + + #include +@@ -419,7 +421,7 @@ qat_sym_build_request(void *in_op, uint8_t *out_msg, + ICP_QAT_HW_AUTH_ALGO_AES_CBC_MAC) { + + /* In case of AES-CCM this may point to user selected +- * memory or iv offset in cypto_op ++ * memory or iv offset in crypto_op + */ + uint8_t *aad_data = op->sym->aead.aad.data; + /* This is true AAD length, it not includes 18 bytes of +diff --git a/dpdk/drivers/crypto/qat/qat_sym_hw_dp.c b/dpdk/drivers/crypto/qat/qat_sym_hw_dp.c +index 12825e448b..792ad2b213 100644 +--- a/dpdk/drivers/crypto/qat/qat_sym_hw_dp.c ++++ b/dpdk/drivers/crypto/qat/qat_sym_hw_dp.c +@@ -533,8 +533,20 @@ enqueue_one_aead_job(struct qat_sym_session *ctx, + /* CPM 1.7 uses single pass to treat AEAD as cipher operation */ + if (ctx->is_single_pass) { + enqueue_one_cipher_job(ctx, req, iv, ofs, data_len); +- cipher_param->spc_aad_addr = aad->iova; +- cipher_param->spc_auth_res_addr = digest->iova; ++ ++ if (ctx->is_ucs) { ++ /* QAT GEN4 uses single pass to treat AEAD as cipher ++ * operation ++ */ ++ struct icp_qat_fw_la_cipher_20_req_params *cipher_param_20 = ++ (void *)&req->serv_specif_rqpars; ++ cipher_param_20->spc_aad_addr = aad->iova; ++ cipher_param_20->spc_auth_res_addr = digest->iova; ++ } else { ++ cipher_param->spc_aad_addr = aad->iova; ++ cipher_param->spc_auth_res_addr = digest->iova; ++ } ++ + return; + } - /* Safe when NODE may be freed (not needed when NODE may be removed from the - * hash index but its members remain accessible and intact). */ --#define HINDEX_FOR_EACH_SAFE(NODE, NEXT, MEMBER, HINDEX) \ -- for (INIT_CONTAINER(NODE, hindex_first(HINDEX), MEMBER); \ -- (NODE != OBJECT_CONTAINING(NULL, NODE, MEMBER) \ -- ? INIT_CONTAINER(NEXT, hindex_next(HINDEX, &(NODE)->MEMBER), MEMBER), 1 \ -- : 0); \ -- (NODE) = (NEXT)) -+#define HINDEX_FOR_EACH_SAFE_LONG(NODE, NEXT, MEMBER, HINDEX) \ -+ for (INIT_MULTIVAR_SAFE_LONG(NODE, NEXT, MEMBER, hindex_first(HINDEX), \ -+ struct hindex_node); \ -+ CONDITION_MULTIVAR_SAFE_LONG(NODE, NEXT, MEMBER, \ -+ ITER_VAR(NODE) != NULL, \ -+ ITER_VAR(NEXT) = hindex_next(HINDEX, ITER_VAR(NODE)), \ -+ ITER_VAR(NEXT) != NULL); \ -+ UPDATE_MULTIVAR_SAFE_LONG(NODE, NEXT)) -+ -+/* Short version of HINDEX_FOR_EACH_SAFE. */ -+#define HINDEX_FOR_EACH_SAFE_SHORT(NODE, MEMBER, HINDEX) \ -+ for (INIT_MULTIVAR_SAFE_SHORT(NODE, MEMBER, hindex_first(HINDEX), \ -+ struct hindex_node); \ -+ CONDITION_MULTIVAR_SAFE_SHORT(NODE, MEMBER, \ -+ ITER_VAR(NODE) != NULL, \ -+ ITER_NEXT_VAR(NODE) = hindex_next(HINDEX, ITER_VAR(NODE))); \ -+ UPDATE_MULTIVAR_SAFE_SHORT(NODE)) -+ -+#define HINDEX_FOR_EACH_SAFE(...) \ -+ OVERLOAD_SAFE_MACRO(HINDEX_FOR_EACH_SAFE_LONG, \ -+ HINDEX_FOR_EACH_SAFE_SHORT, \ -+ 4, __VA_ARGS__) +diff --git a/dpdk/drivers/crypto/qat/qat_sym_session.c b/dpdk/drivers/crypto/qat/qat_sym_session.c +index 8ca475ca8b..80d6fbfa46 100644 +--- a/dpdk/drivers/crypto/qat/qat_sym_session.c ++++ b/dpdk/drivers/crypto/qat/qat_sym_session.c +@@ -2,6 +2,8 @@ + * Copyright(c) 2015-2019 Intel Corporation + */ - struct hindex_node *hindex_first(const struct hindex *); - struct hindex_node *hindex_next(const struct hindex *, -diff --git a/lib/hmapx.c b/lib/hmapx.c -index eadfe640ac..68192fc2c5 100644 ---- a/lib/hmapx.c -+++ b/lib/hmapx.c -@@ -123,9 +123,9 @@ hmapx_add_assert(struct hmapx *map, void *data) - void - hmapx_clear(struct hmapx *map) ++#define OPENSSL_API_COMPAT 0x10100000L ++ + #include /* Needed to calculate pre-compute values */ + #include /* Needed to calculate pre-compute values */ + #include /* Needed to calculate pre-compute values */ +@@ -124,8 +126,10 @@ bpi_cipher_ctx_init(enum rte_crypto_cipher_algorithm cryptodev_algo, + return 0; + + ctx_init_err: +- if (*ctx != NULL) ++ if (*ctx != NULL) { + EVP_CIPHER_CTX_free(*ctx); ++ *ctx = NULL; ++ } + return ret; + } + +diff --git a/dpdk/drivers/crypto/scheduler/scheduler_failover.c b/dpdk/drivers/crypto/scheduler/scheduler_failover.c +index 5023577ef8..2a0e29fa72 100644 +--- a/dpdk/drivers/crypto/scheduler/scheduler_failover.c ++++ b/dpdk/drivers/crypto/scheduler/scheduler_failover.c +@@ -157,6 +157,9 @@ scheduler_start(struct rte_cryptodev *dev) + ((struct scheduler_qp_ctx *) + dev->data->queue_pairs[i])->private_qp_ctx; + ++ sched_ctx->workers[PRIMARY_WORKER_IDX].qp_id = i; ++ sched_ctx->workers[SECONDARY_WORKER_IDX].qp_id = i; ++ + rte_memcpy(&qp_ctx->primary_worker, + &sched_ctx->workers[PRIMARY_WORKER_IDX], + sizeof(struct scheduler_worker)); +diff --git a/dpdk/drivers/crypto/virtio/virtio_rxtx.c b/dpdk/drivers/crypto/virtio/virtio_rxtx.c +index a65524a306..08359b3a39 100644 +--- a/dpdk/drivers/crypto/virtio/virtio_rxtx.c ++++ b/dpdk/drivers/crypto/virtio/virtio_rxtx.c +@@ -264,6 +264,9 @@ virtqueue_crypto_sym_enqueue_xmit( + if (cop->phys_addr) + desc[idx].addr = cop->phys_addr + session->iv.offset; + else { ++ if (session->iv.length > VIRTIO_CRYPTO_MAX_IV_SIZE) ++ return -ENOMEM; ++ + rte_memcpy(crypto_op_cookie->iv, + rte_crypto_op_ctod_offset(cop, + uint8_t *, session->iv.offset), +diff --git a/dpdk/drivers/crypto/virtio/virtqueue.h b/dpdk/drivers/crypto/virtio/virtqueue.h +index bf10c6579b..c96ca62992 100644 +--- a/dpdk/drivers/crypto/virtio/virtqueue.h ++++ b/dpdk/drivers/crypto/virtio/virtqueue.h +@@ -145,7 +145,7 @@ virtqueue_notify(struct virtqueue *vq) { -- struct hmapx_node *node, *next; -+ struct hmapx_node *node; + /* + * Ensure updated avail->idx is visible to host. +- * For virtio on IA, the notificaiton is through io port operation ++ * For virtio on IA, the notification is through io port operation + * which is a serialization instruction itself. + */ + VTPCI_OPS(vq->hw)->notify_queue(vq->hw, vq); +diff --git a/dpdk/drivers/dma/cnxk/meson.build b/dpdk/drivers/dma/cnxk/meson.build +index 633e92a20d..d4be4ee860 100644 +--- a/dpdk/drivers/dma/cnxk/meson.build ++++ b/dpdk/drivers/dma/cnxk/meson.build +@@ -3,4 +3,3 @@ + + deps += ['bus_pci', 'common_cnxk', 'dmadev'] + sources = files('cnxk_dmadev.c') +-headers = files('cnxk_dmadev.h') +diff --git a/dpdk/drivers/dma/hisilicon/hisi_dmadev.c b/dpdk/drivers/dma/hisilicon/hisi_dmadev.c +index 05066b4d0e..ec687ef67e 100644 +--- a/dpdk/drivers/dma/hisilicon/hisi_dmadev.c ++++ b/dpdk/drivers/dma/hisilicon/hisi_dmadev.c +@@ -581,7 +581,7 @@ hisi_dma_scan_cq(struct hisi_dma_dev *hw) + uint16_t count = 0; + uint64_t misc; + +- while (true) { ++ while (count < hw->cq_depth) { + cqe = &hw->cqe[cq_head]; + misc = cqe->misc; + misc = rte_le_to_cpu_64(misc); +@@ -589,6 +589,16 @@ hisi_dma_scan_cq(struct hisi_dma_dev *hw) + break; + + csq_head = FIELD_GET(CQE_SQ_HEAD_MASK, misc); ++ if (unlikely(csq_head > hw->sq_depth_mask)) { ++ /** ++ * Defensive programming to prevent overflow of the ++ * status array indexed by csq_head. Only error logs ++ * are used for prompting. ++ */ ++ HISI_DMA_ERR(hw, "invalid csq_head:%u!\n", csq_head); ++ count = 0; ++ break; ++ } + if (unlikely(misc & CQE_STATUS_MASK)) + hw->status[csq_head] = FIELD_GET(CQE_STATUS_MASK, + misc); +@@ -649,12 +659,12 @@ hisi_dma_completed(void *dev_private, + } + sq_head = (sq_head + 1) & hw->sq_depth_mask; + } ++ *last_idx = hw->cridx + i - 1; + if (i > 0) { + hw->cridx += i; +- *last_idx = hw->cridx - 1; + hw->sq_head = sq_head; ++ hw->completed += i; + } +- hw->completed += i; -- HMAPX_FOR_EACH_SAFE (node, next, map) { -+ HMAPX_FOR_EACH_SAFE (node, map) { - hmapx_delete(map, node); - } + return i; } -diff --git a/lib/hmapx.h b/lib/hmapx.h -index 06a6bbe67d..0b03411491 100644 ---- a/lib/hmapx.h -+++ b/lib/hmapx.h -@@ -67,10 +67,20 @@ bool hmapx_equals(const struct hmapx *, const struct hmapx *); +@@ -708,12 +718,12 @@ hisi_dma_completed_status(void *dev_private, + hw->status[sq_head] = HISI_DMA_STATUS_SUCCESS; + sq_head = (sq_head + 1) & hw->sq_depth_mask; + } ++ *last_idx = hw->cridx + cpl_num - 1; + if (likely(cpl_num > 0)) { + hw->cridx += cpl_num; +- *last_idx = hw->cridx - 1; + hw->sq_head = sq_head; ++ hw->completed += cpl_num; + } +- hw->completed += cpl_num; - /* Safe when NODE may be freed (not needed when NODE may be removed from the - * hash map but its members remain accessible and intact). */ --#define HMAPX_FOR_EACH_SAFE(NODE, NEXT, HMAPX) \ -- HMAP_FOR_EACH_SAFE_INIT(NODE, NEXT, hmap_node, &(HMAPX)->map, \ -+#define HMAPX_FOR_EACH_SAFE_SHORT(NODE, HMAPX) \ -+ HMAP_FOR_EACH_SAFE_SHORT_INIT (NODE, hmap_node, &(HMAPX)->map, \ -+ BUILD_ASSERT_TYPE(NODE, struct hmapx_node *), \ -+ BUILD_ASSERT_TYPE(HMAPX, struct hmapx *)) -+ -+#define HMAPX_FOR_EACH_SAFE_LONG(NODE, NEXT, HMAPX) \ -+ HMAP_FOR_EACH_SAFE_LONG_INIT (NODE, NEXT, hmap_node, &(HMAPX)->map, \ - BUILD_ASSERT_TYPE(NODE, struct hmapx_node *), \ - BUILD_ASSERT_TYPE(NEXT, struct hmapx_node *), \ - BUILD_ASSERT_TYPE(HMAPX, struct hmapx *)) + return cpl_num; + } +@@ -731,24 +741,15 @@ hisi_dma_burst_capacity(const void *dev_private, uint16_t vchan) + sq_head - 1 - sq_tail; + } -+#define HMAPX_FOR_EACH_SAFE(...) \ -+ OVERLOAD_SAFE_MACRO(HMAPX_FOR_EACH_SAFE_LONG, \ -+ HMAPX_FOR_EACH_SAFE_SHORT, \ -+ 3, __VA_ARGS__) +-static void +-hisi_dma_gen_pci_device_name(const struct rte_pci_device *pci_dev, +- char *name, size_t size) +-{ +- memset(name, 0, size); +- (void)snprintf(name, size, "%x:%x.%x", +- pci_dev->addr.bus, pci_dev->addr.devid, +- pci_dev->addr.function); +-} +- + static void + hisi_dma_gen_dev_name(const struct rte_pci_device *pci_dev, +- uint8_t queue_id, char *name, size_t size) ++ uint8_t queue_id, char *dev_name, size_t size) + { +- memset(name, 0, size); +- (void)snprintf(name, size, "%x:%x.%x-ch%u", +- pci_dev->addr.bus, pci_dev->addr.devid, +- pci_dev->addr.function, queue_id); ++ char name[RTE_DEV_NAME_MAX_LEN] = { 0 }; ++ ++ memset(dev_name, 0, size); ++ rte_pci_device_name(&pci_dev->addr, name, sizeof(name)); ++ (void)snprintf(dev_name, size, "%s-ch%u", name, queue_id); + } + + /** +@@ -864,7 +865,7 @@ hisi_dma_probe(struct rte_pci_driver *pci_drv __rte_unused, + uint8_t i; + int ret; + +- hisi_dma_gen_pci_device_name(pci_dev, name, sizeof(name)); ++ rte_pci_device_name(&pci_dev->addr, name, sizeof(name)); + + if (pci_dev->mem_resource[2].addr == NULL) { + HISI_DMA_LOG(ERR, "%s BAR2 is NULL!\n", name); +diff --git a/dpdk/drivers/dma/hisilicon/hisi_dmadev.h b/dpdk/drivers/dma/hisilicon/hisi_dmadev.h +index 12e209c86e..f06c851825 100644 +--- a/dpdk/drivers/dma/hisilicon/hisi_dmadev.h ++++ b/dpdk/drivers/dma/hisilicon/hisi_dmadev.h +@@ -7,6 +7,8 @@ + + #include + #include ++#include ++#include + + #define BIT(x) (1ul << (x)) + #define BITS_PER_LONG (__SIZEOF_LONG__ * 8) +diff --git a/dpdk/drivers/dma/idxd/dpdk_idxd_cfg.py b/dpdk/drivers/dma/idxd/dpdk_idxd_cfg.py +index fcc27822ef..3f5d5ee752 100755 +--- a/dpdk/drivers/dma/idxd/dpdk_idxd_cfg.py ++++ b/dpdk/drivers/dma/idxd/dpdk_idxd_cfg.py +@@ -29,9 +29,17 @@ def write_values(self, values): + f.write(str(contents)) + + ++def get_drv_dir(dtype): ++ "Get the sysfs path for the driver, either 'idxd' or 'user'" ++ drv_dir = "/sys/bus/dsa/drivers/" + dtype ++ if not os.path.exists(drv_dir): ++ return "/sys/bus/dsa/drivers/dsa" ++ return drv_dir ++ ++ + def reset_device(dsa_id): + "Reset the DSA device and all its queues" +- drv_dir = SysfsDir("/sys/bus/dsa/drivers/dsa") ++ drv_dir = SysfsDir(get_drv_dir("idxd")) + drv_dir.write_values({"unbind": f"dsa{dsa_id}"}) + + +@@ -58,7 +66,6 @@ def get_dsa_id(pci): + def configure_dsa(dsa_id, queues, prefix): + "Configure the DSA instance with appropriate number of queues" + dsa_dir = SysfsDir(f"/sys/bus/dsa/devices/dsa{dsa_id}") +- drv_dir = SysfsDir("/sys/bus/dsa/drivers/dsa") + + max_groups = dsa_dir.read_int("max_groups") + max_engines = dsa_dir.read_int("max_engines") +@@ -82,12 +89,16 @@ def configure_dsa(dsa_id, queues, prefix): + "mode": "dedicated", + "name": f"{prefix}_wq{dsa_id}.{q}", + "priority": 1, ++ "max_batch_size": 1024, + "size": int(max_work_queues_size / nb_queues)}) + + # enable device and then queues +- drv_dir.write_values({"bind": f"dsa{dsa_id}"}) ++ idxd_dir = SysfsDir(get_drv_dir("idxd")) ++ idxd_dir.write_values({"bind": f"dsa{dsa_id}"}) ++ ++ user_dir = SysfsDir(get_drv_dir("user")) + for q in range(nb_queues): +- drv_dir.write_values({"bind": f"wq{dsa_id}.{q}"}) ++ user_dir.write_values({"bind": f"wq{dsa_id}.{q}"}) + + + def main(args): +diff --git a/dpdk/drivers/dma/idxd/idxd_common.c b/dpdk/drivers/dma/idxd/idxd_common.c +index fc11b11337..c77200a457 100644 +--- a/dpdk/drivers/dma/idxd/idxd_common.c ++++ b/dpdk/drivers/dma/idxd/idxd_common.c +@@ -13,12 +13,23 @@ + + #define IDXD_PMD_NAME_STR "dmadev_idxd" + ++/* systems with DSA all support AVX2 so allow our data-path functions to ++ * always use at least that instruction set ++ */ ++#ifndef __AVX2__ ++#define __use_avx2 __attribute__((target("avx2"))) ++#else ++#define __use_avx2 ++#endif + - #endif /* hmapx.h */ -diff --git a/lib/id-fpool.c b/lib/id-fpool.c -index 15cef5d003..7108c104a4 100644 ---- a/lib/id-fpool.c -+++ b/lib/id-fpool.c -@@ -166,11 +166,10 @@ void - id_fpool_destroy(struct id_fpool *pool) ++__use_avx2 + static __rte_always_inline rte_iova_t + __desc_idx_to_iova(struct idxd_dmadev *idxd, uint16_t n) { - struct id_slab *slab; -- struct id_slab *next; - size_t i; + return idxd->desc_iova + (n * sizeof(struct idxd_hw_desc)); + } - id_fpool_lock(&pool->pool_lock); -- LIST_FOR_EACH_SAFE (slab, next, node, &pool->free_slabs) { -+ LIST_FOR_EACH_SAFE (slab, node, &pool->free_slabs) { - free(slab); - } - ovs_list_poison(&pool->free_slabs); -diff --git a/lib/ipf.c b/lib/ipf.c -index 507db2aea2..d452663743 100644 ---- a/lib/ipf.c -+++ b/lib/ipf.c -@@ -1058,9 +1058,9 @@ ipf_send_completed_frags(struct ipf *ipf, struct dp_packet_batch *pb, - } ++__use_avx2 + static __rte_always_inline void + __idxd_movdir64b(volatile void *dst, const struct idxd_hw_desc *src) + { +@@ -28,6 +39,7 @@ __idxd_movdir64b(volatile void *dst, const struct idxd_hw_desc *src) + : "memory"); + } - ovs_mutex_lock(&ipf->ipf_lock); -- struct ipf_list *ipf_list, *next; -+ struct ipf_list *ipf_list; ++__use_avx2 + static __rte_always_inline void + __submit(struct idxd_dmadev *idxd) + { +@@ -74,6 +86,7 @@ __submit(struct idxd_dmadev *idxd) + _mm256_setzero_si256()); + } -- LIST_FOR_EACH_SAFE (ipf_list, next, list_node, &ipf->frag_complete_list) { -+ LIST_FOR_EACH_SAFE (ipf_list, list_node, &ipf->frag_complete_list) { - if (ipf_send_frags_in_list(ipf, ipf_list, pb, IPF_FRAG_COMPLETED_LIST, - v6, now)) { - ipf_completed_list_clean(&ipf->frag_lists, ipf_list); -@@ -1090,10 +1090,10 @@ ipf_send_expired_frags(struct ipf *ipf, struct dp_packet_batch *pb, - } ++__use_avx2 + static __rte_always_inline int + __idxd_write_desc(struct idxd_dmadev *idxd, + const uint32_t op_flags, +@@ -112,6 +125,7 @@ __idxd_write_desc(struct idxd_dmadev *idxd, + return job_id; + } - ovs_mutex_lock(&ipf->ipf_lock); -- struct ipf_list *ipf_list, *next; -+ struct ipf_list *ipf_list; - size_t lists_removed = 0; ++__use_avx2 + int + idxd_enqueue_copy(void *dev_private, uint16_t qid __rte_unused, rte_iova_t src, + rte_iova_t dst, unsigned int length, uint64_t flags) +@@ -126,6 +140,7 @@ idxd_enqueue_copy(void *dev_private, uint16_t qid __rte_unused, rte_iova_t src, + flags); + } -- LIST_FOR_EACH_SAFE (ipf_list, next, list_node, &ipf->frag_exp_list) { -+ LIST_FOR_EACH_SAFE (ipf_list, list_node, &ipf->frag_exp_list) { - if (now <= ipf_list->expiration || - lists_removed >= IPF_FRAG_LIST_MAX_EXPIRED) { - break; -@@ -1121,9 +1121,9 @@ ipf_execute_reass_pkts(struct ipf *ipf, struct dp_packet_batch *pb) - } ++__use_avx2 + int + idxd_enqueue_fill(void *dev_private, uint16_t qid __rte_unused, uint64_t pattern, + rte_iova_t dst, unsigned int length, uint64_t flags) +@@ -136,6 +151,7 @@ idxd_enqueue_fill(void *dev_private, uint16_t qid __rte_unused, uint64_t pattern + flags); + } - ovs_mutex_lock(&ipf->ipf_lock); -- struct reassembled_pkt *rp, *next; -+ struct reassembled_pkt *rp; ++__use_avx2 + int + idxd_submit(void *dev_private, uint16_t qid __rte_unused) + { +@@ -143,6 +159,7 @@ idxd_submit(void *dev_private, uint16_t qid __rte_unused) + return 0; + } -- LIST_FOR_EACH_SAFE (rp, next, rp_list_node, &ipf->reassembled_pkt_list) { -+ LIST_FOR_EACH_SAFE (rp, rp_list_node, &ipf->reassembled_pkt_list) { - if (!rp->list->reass_execute_ctx && - ipf_dp_packet_batch_add(pb, rp->pkt, false)) { - rp->list->reass_execute_ctx = rp->pkt; -@@ -1144,9 +1144,9 @@ ipf_post_execute_reass_pkts(struct ipf *ipf, - } - - ovs_mutex_lock(&ipf->ipf_lock); -- struct reassembled_pkt *rp, *next; -+ struct reassembled_pkt *rp; ++__use_avx2 + static enum rte_dma_status_code + get_comp_status(struct idxd_completion *c) + { +@@ -163,6 +180,7 @@ get_comp_status(struct idxd_completion *c) + } + } -- LIST_FOR_EACH_SAFE (rp, next, rp_list_node, &ipf->reassembled_pkt_list) { -+ LIST_FOR_EACH_SAFE (rp, rp_list_node, &ipf->reassembled_pkt_list) { - const size_t pb_cnt = dp_packet_batch_size(pb); - int pb_idx; - struct dp_packet *pkt; -@@ -1271,15 +1271,15 @@ ipf_clean_thread_main(void *f) ++__use_avx2 + int + idxd_vchan_status(const struct rte_dma_dev *dev, uint16_t vchan __rte_unused, + enum rte_dma_vchan_status *status) +@@ -180,6 +198,7 @@ idxd_vchan_status(const struct rte_dma_dev *dev, uint16_t vchan __rte_unused, + return 0; + } - ovs_mutex_lock(&ipf->ipf_lock); ++__use_avx2 + static __rte_always_inline int + batch_ok(struct idxd_dmadev *idxd, uint16_t max_ops, enum rte_dma_status_code *status) + { +@@ -224,6 +243,7 @@ batch_ok(struct idxd_dmadev *idxd, uint16_t max_ops, enum rte_dma_status_code *s + return -1; /* error case */ + } -- struct ipf_list *ipf_list, *next; -- LIST_FOR_EACH_SAFE (ipf_list, next, list_node, -+ struct ipf_list *ipf_list; -+ LIST_FOR_EACH_SAFE (ipf_list, list_node, - &ipf->frag_exp_list) { - if (ipf_purge_list_check(ipf, ipf_list, now)) { - ipf_expiry_list_clean(&ipf->frag_lists, ipf_list); - } - } ++__use_avx2 + static inline uint16_t + batch_completed(struct idxd_dmadev *idxd, uint16_t max_ops, bool *has_error) + { +@@ -275,6 +295,7 @@ batch_completed(struct idxd_dmadev *idxd, uint16_t max_ops, bool *has_error) + return ret; + } -- LIST_FOR_EACH_SAFE (ipf_list, next, list_node, -+ LIST_FOR_EACH_SAFE (ipf_list, list_node, - &ipf->frag_complete_list) { - if (ipf_purge_list_check(ipf, ipf_list, now)) { - ipf_completed_list_clean(&ipf->frag_lists, ipf_list); -diff --git a/lib/json.c b/lib/json.c -index 720c73d940..042aab83b3 100644 ---- a/lib/json.c -+++ b/lib/json.c -@@ -397,9 +397,9 @@ json_destroy__(struct json *json) - static void - json_destroy_object(struct shash *object) ++__use_avx2 + static uint16_t + batch_completed_status(struct idxd_dmadev *idxd, uint16_t max_ops, enum rte_dma_status_code *status) { -- struct shash_node *node, *next; -+ struct shash_node *node; +@@ -366,6 +387,7 @@ batch_completed_status(struct idxd_dmadev *idxd, uint16_t max_ops, enum rte_dma_ + return ret; + } -- SHASH_FOR_EACH_SAFE (node, next, object) { -+ SHASH_FOR_EACH_SAFE (node, object) { - struct json *value = node->data; ++__use_avx2 + uint16_t + idxd_completed(void *dev_private, uint16_t qid __rte_unused, uint16_t max_ops, + uint16_t *last_idx, bool *has_error) +@@ -383,6 +405,7 @@ idxd_completed(void *dev_private, uint16_t qid __rte_unused, uint16_t max_ops, + return ret; + } - json_destroy(value); -diff --git a/lib/lacp.c b/lib/lacp.c -index 89d711225f..3252f17ebf 100644 ---- a/lib/lacp.c -+++ b/lib/lacp.c -@@ -280,10 +280,10 @@ void - lacp_unref(struct lacp *lacp) OVS_EXCLUDED(mutex) - { - if (lacp && ovs_refcount_unref_relaxed(&lacp->ref_cnt) == 1) { -- struct member *member, *next; -+ struct member *member; ++__use_avx2 + uint16_t + idxd_completed_status(void *dev_private, uint16_t qid __rte_unused, uint16_t max_ops, + uint16_t *last_idx, enum rte_dma_status_code *status) +@@ -480,12 +503,12 @@ idxd_burst_capacity(const void *dev_private, uint16_t vchan __rte_unused) + idxd->batch_idx_write + 1 == idxd->batch_idx_read) + return 0; - lacp_lock(); -- HMAP_FOR_EACH_SAFE (member, next, node, &lacp->members) { -+ HMAP_FOR_EACH_SAFE (member, node, &lacp->members) { - member_destroy(member); - } +- /* For descriptors, check for wrap-around on write but not read */ +- if (idxd->ids_returned > write_idx) +- write_idx += idxd->desc_ring_mask + 1; +- used_space = write_idx - idxd->ids_returned; ++ /* Subtract and mask to get in correct range */ ++ used_space = (write_idx - idxd->ids_returned) & idxd->desc_ring_mask; + +- return RTE_MIN((idxd->desc_ring_mask - used_space), idxd->max_batch_size); ++ const int ret = RTE_MIN((idxd->desc_ring_mask - used_space), ++ (idxd->max_batch_size - idxd->batch_size)); ++ return ret < 0 ? 0 : (uint16_t)ret; + } -diff --git a/lib/lldp/lldpd-structs.c b/lib/lldp/lldpd-structs.c -index 499b441746..a8c7fad098 100644 ---- a/lib/lldp/lldpd-structs.c -+++ b/lib/lldp/lldpd-structs.c -@@ -64,11 +64,11 @@ lldpd_remote_cleanup(struct lldpd_hardware *hw, - struct lldpd_port *), - bool all) - { -- struct lldpd_port *port, *port_next; -+ struct lldpd_port *port; - time_t now = time_now(); + int +diff --git a/dpdk/drivers/dma/idxd/idxd_internal.h b/dpdk/drivers/dma/idxd/idxd_internal.h +index 3375600217..180a8587c6 100644 +--- a/dpdk/drivers/dma/idxd/idxd_internal.h ++++ b/dpdk/drivers/dma/idxd/idxd_internal.h +@@ -7,6 +7,7 @@ + + #include + #include ++#include + + #include "idxd_hw_defs.h" + +@@ -33,6 +34,7 @@ struct idxd_pci_common { + rte_spinlock_t lk; + + uint8_t wq_cfg_sz; ++ rte_atomic16_t ref_count; + volatile struct rte_idxd_bar0 *regs; + volatile uint32_t *wq_regs_base; + volatile struct rte_idxd_grpcfg *grp_regs; +diff --git a/dpdk/drivers/dma/idxd/idxd_pci.c b/dpdk/drivers/dma/idxd/idxd_pci.c +index 9ca1ec64e9..2f8ec06d9e 100644 +--- a/dpdk/drivers/dma/idxd/idxd_pci.c ++++ b/dpdk/drivers/dma/idxd/idxd_pci.c +@@ -6,6 +6,7 @@ + #include + #include + #include ++#include + + #include "idxd_internal.h" + +@@ -38,13 +39,13 @@ idxd_pci_dev_command(struct idxd_dmadev *idxd, enum rte_idxd_cmds command) + IDXD_PMD_ERR("Timeout waiting for command response from HW"); + rte_spinlock_unlock(&idxd->u.pci->lk); + err_code &= CMDSTATUS_ERR_MASK; +- return -err_code; ++ return err_code; + } + } while (err_code & CMDSTATUS_ACTIVE_MASK); + rte_spinlock_unlock(&idxd->u.pci->lk); + + err_code &= CMDSTATUS_ERR_MASK; +- return -err_code; ++ return err_code; + } - VLOG_DBG("cleanup remote port on %s", hw->h_ifname); -- LIST_FOR_EACH_SAFE (port, port_next, p_entries, &hw->h_rports) { -+ LIST_FOR_EACH_SAFE (port, p_entries, &hw->h_rports) { - bool del = all; - if (!all && expire && - (now >= port->p_lastupdate + port->p_chassis->c_ttl)) { -@@ -99,11 +99,10 @@ static void - lldpd_aa_maps_cleanup(struct lldpd_port *port) + static uint32_t * +@@ -115,20 +116,38 @@ idxd_pci_dev_close(struct rte_dma_dev *dev) { - struct lldpd_aa_isid_vlan_maps_tlv *isid_vlan_map = NULL; -- struct lldpd_aa_isid_vlan_maps_tlv *isid_vlan_map_next = NULL; - - if (!ovs_list_is_empty(&port->p_isid_vlan_maps)) { + struct idxd_dmadev *idxd = dev->fp_obj->dev_private; + uint8_t err_code; ++ int is_last_wq; + +- /* disable the device */ +- err_code = idxd_pci_dev_command(idxd, idxd_disable_dev); +- if (err_code) { +- IDXD_PMD_ERR("Error disabling device: code %#x", err_code); +- return err_code; ++ if (idxd_is_wq_enabled(idxd)) { ++ /* disable the wq */ ++ err_code = idxd_pci_dev_command(idxd, idxd_disable_wq); ++ if (err_code) { ++ IDXD_PMD_ERR("Error disabling wq: code %#x", err_code); ++ return err_code; ++ } ++ IDXD_PMD_DEBUG("IDXD WQ disabled OK"); + } +- IDXD_PMD_DEBUG("IDXD Device disabled OK"); + + /* free device memory */ + IDXD_PMD_DEBUG("Freeing device driver memory"); +- rte_free(idxd->batch_idx_ring); ++ rte_free(idxd->batch_comp_ring); + rte_free(idxd->desc_ring); + ++ /* if this is the last WQ on the device, disable the device and free ++ * the PCI struct ++ */ ++ is_last_wq = rte_atomic16_dec_and_test(&idxd->u.pci->ref_count); ++ if (is_last_wq) { ++ /* disable the device */ ++ err_code = idxd_pci_dev_command(idxd, idxd_disable_dev); ++ if (err_code) { ++ IDXD_PMD_ERR("Error disabling device: code %#x", err_code); ++ return err_code; ++ } ++ IDXD_PMD_DEBUG("IDXD device disabled OK"); ++ rte_free(idxd->u.pci); ++ } ++ + return 0; + } -- LIST_FOR_EACH_SAFE (isid_vlan_map, isid_vlan_map_next, m_entries, -+ LIST_FOR_EACH_SAFE (isid_vlan_map, m_entries, - &port->p_isid_vlan_maps) { +@@ -159,12 +178,13 @@ init_pci_device(struct rte_pci_device *dev, struct idxd_dmadev *idxd, + uint8_t lg2_max_batch, lg2_max_copy_size; + unsigned int i, err_code; - ovs_list_remove(&isid_vlan_map->m_entries); -diff --git a/lib/lldp/lldpd.c b/lib/lldp/lldpd.c -index a024dc5e58..4bff7b017f 100644 ---- a/lib/lldp/lldpd.c -+++ b/lib/lldp/lldpd.c -@@ -134,24 +134,20 @@ lldpd_hardware_cleanup(struct lldpd *cfg, struct lldpd_hardware *hardware) - void - lldpd_cleanup(struct lldpd *cfg) - { -- struct lldpd_hardware *hw, *hw_next; -- struct lldpd_chassis *chassis, *chassis_next; -+ struct lldpd_hardware *hw; -+ struct lldpd_chassis *chassis; +- pci = malloc(sizeof(*pci)); ++ pci = rte_malloc(NULL, sizeof(*pci), 0); + if (pci == NULL) { + IDXD_PMD_ERR("%s: Can't allocate memory", __func__); + err_code = -1; + goto err; + } ++ memset(pci, 0, sizeof(*pci)); + rte_spinlock_init(&pci->lk); + + /* assign the bar registers, and then configure device */ +@@ -330,6 +350,7 @@ idxd_dmadev_probe_pci(struct rte_pci_driver *drv, struct rte_pci_device *dev) + free(idxd.u.pci); + return ret; + } ++ rte_atomic16_inc(&idxd.u.pci->ref_count); + } - VLOG_DBG("cleanup all ports"); + return 0; +@@ -359,10 +380,10 @@ idxd_dmadev_remove_pci(struct rte_pci_device *dev) + IDXD_PMD_INFO("Closing %s on NUMA node %d", name, dev->device.numa_node); + + RTE_DMA_FOREACH_DEV(i) { +- struct rte_dma_info *info = {0}; +- rte_dma_info_get(i, info); +- if (strncmp(name, info->dev_name, strlen(name)) == 0) +- idxd_dmadev_destroy(info->dev_name); ++ struct rte_dma_info info; ++ rte_dma_info_get(i, &info); ++ if (strncmp(name, info.dev_name, strlen(name)) == 0) ++ idxd_dmadev_destroy(info.dev_name); + } -- LIST_FOR_EACH_SAFE (hw, hw_next, h_entries, &cfg->g_hardware) { -- if (!hw->h_flags) { -- ovs_list_remove(&hw->h_entries); -- lldpd_remote_cleanup(hw, NULL, true); -- lldpd_hardware_cleanup(cfg, hw); -- } else { -- lldpd_remote_cleanup(hw, NULL, false); -- } -+ LIST_FOR_EACH_SAFE (hw, h_entries, &cfg->g_hardware) { -+ ovs_list_remove(&hw->h_entries); -+ lldpd_remote_cleanup(hw, NULL, true); -+ lldpd_hardware_cleanup(cfg, hw); - } + return 0; +diff --git a/dpdk/drivers/dma/idxd/meson.build b/dpdk/drivers/dma/idxd/meson.build +index f1396be945..c5403b431c 100644 +--- a/dpdk/drivers/dma/idxd/meson.build ++++ b/dpdk/drivers/dma/idxd/meson.build +@@ -4,8 +4,18 @@ + build = dpdk_conf.has('RTE_ARCH_X86') + reason = 'only supported on x86' + ++test_avx2_code = ''' ++#include ++__attribute__((target("avx2"))) ++__m256i fn(void *x) { return _mm256_loadu_si256(x); } ++''' ++if build and not cc.compiles(test_avx2_code, args:machine_args) ++ build = false ++ reason = 'missing support for AVX2 function attribute' ++ subdir_done() ++endif ++ + deps += ['bus_pci'] +-cflags += '-mavx2' # all platforms with idxd HW support AVX + sources = files( + 'idxd_common.c', + 'idxd_pci.c', +diff --git a/dpdk/drivers/dma/skeleton/skeleton_dmadev.c b/dpdk/drivers/dma/skeleton/skeleton_dmadev.c +index d9e4f731d7..6b0bb14e2c 100644 +--- a/dpdk/drivers/dma/skeleton/skeleton_dmadev.c ++++ b/dpdk/drivers/dma/skeleton/skeleton_dmadev.c +@@ -118,6 +118,7 @@ skeldma_start(struct rte_dma_dev *dev) + fflush_ring(hw, hw->desc_running); + fflush_ring(hw, hw->desc_completed); + hw->ridx = 0; ++ hw->last_ridx = hw->ridx - 1; + hw->submitted_count = 0; + hw->zero_req_count = 0; + hw->completed_count = 0; +@@ -169,7 +170,7 @@ vchan_setup(struct skeldma_hw *hw, uint16_t nb_desc) + struct rte_ring *completed; + uint16_t i; + +- desc = rte_zmalloc_socket("dma_skelteon_desc", ++ desc = rte_zmalloc_socket("dma_skeleton_desc", + nb_desc * sizeof(struct skeldma_desc), + RTE_CACHE_LINE_SIZE, hw->socket_id); + if (desc == NULL) { +@@ -322,9 +323,11 @@ skeldma_dump(const struct rte_dma_dev *dev, FILE *f) + GET_RING_COUNT(hw->desc_completed)); + (void)fprintf(f, + " next_ring_idx: %u\n" ++ " last_ring_idx: %u\n" + " submitted_count: %" PRIu64 "\n" + " completed_count: %" PRIu64 "\n", +- hw->ridx, hw->submitted_count, hw->completed_count); ++ hw->ridx, hw->last_ridx, ++ hw->submitted_count, hw->completed_count); + + return 0; + } +@@ -398,11 +401,15 @@ skeldma_completed(void *dev_private, + count = RTE_MIN(nb_cpls, rte_ring_count(hw->desc_completed)); + while (index < count) { + (void)rte_ring_dequeue(hw->desc_completed, (void **)&desc); +- if (index == count - 1) ++ if (index == count - 1) { ++ hw->last_ridx = desc->ridx; + *last_idx = desc->ridx; ++ } + index++; + (void)rte_ring_enqueue(hw->desc_empty, (void *)desc); + } ++ if (unlikely(count == 0)) ++ *last_idx = hw->last_ridx; - VLOG_DBG("cleanup all chassis"); + return count; + } +@@ -422,11 +429,15 @@ skeldma_completed_status(void *dev_private, + count = RTE_MIN(nb_cpls, rte_ring_count(hw->desc_completed)); + while (index < count) { + (void)rte_ring_dequeue(hw->desc_completed, (void **)&desc); +- if (index == count - 1) ++ if (index == count - 1) { ++ hw->last_ridx = desc->ridx; + *last_idx = desc->ridx; ++ } + status[index++] = RTE_DMA_STATUS_SUCCESSFUL; + (void)rte_ring_enqueue(hw->desc_empty, (void *)desc); + } ++ if (unlikely(count == 0)) ++ *last_idx = hw->last_ridx; -- LIST_FOR_EACH_SAFE (chassis, chassis_next, list, &cfg->g_chassis) { -+ LIST_FOR_EACH_SAFE (chassis, list, &cfg->g_chassis) { - if (chassis->c_refcount == 0) { - ovs_list_remove(&chassis->list); - lldpd_chassis_cleanup(chassis, 1); -diff --git a/lib/mac-learning.c b/lib/mac-learning.c -index 3fcd7d9b77..a60794fb26 100644 ---- a/lib/mac-learning.c -+++ b/lib/mac-learning.c -@@ -244,10 +244,10 @@ void - mac_learning_unref(struct mac_learning *ml) + return count; + } +diff --git a/dpdk/drivers/dma/skeleton/skeleton_dmadev.h b/dpdk/drivers/dma/skeleton/skeleton_dmadev.h +index 91eb5460fc..6f89400480 100644 +--- a/dpdk/drivers/dma/skeleton/skeleton_dmadev.h ++++ b/dpdk/drivers/dma/skeleton/skeleton_dmadev.h +@@ -50,6 +50,7 @@ struct skeldma_hw { + /* Cache delimiter for dataplane API's operation data */ + char cache1 __rte_cache_aligned; + uint16_t ridx; /* ring idx */ ++ uint16_t last_ridx; + uint64_t submitted_count; + + /* Cache delimiter for cpucopy thread's operation data */ +diff --git a/dpdk/drivers/event/cnxk/cn10k_eventdev.c b/dpdk/drivers/event/cnxk/cn10k_eventdev.c +index c5a8c1ae8f..4d878fc2b7 100644 +--- a/dpdk/drivers/event/cnxk/cn10k_eventdev.c ++++ b/dpdk/drivers/event/cnxk/cn10k_eventdev.c +@@ -111,10 +111,10 @@ cn10k_sso_hws_release(void *arg, void *hws) { - if (ml && ovs_refcount_unref(&ml->ref_cnt) == 1) { -- struct mac_entry *e, *next; -+ struct mac_entry *e; + struct cnxk_sso_evdev *dev = arg; + struct cn10k_sso_hws *ws = hws; +- int i; ++ uint16_t i; + + for (i = 0; i < dev->nb_event_queues; i++) +- roc_sso_hws_unlink(&dev->sso, ws->hws_id, (uint16_t *)&i, 1); ++ roc_sso_hws_unlink(&dev->sso, ws->hws_id, &i, 1); + memset(ws, 0, sizeof(*ws)); + } - ovs_rwlock_wrlock(&ml->rwlock); -- HMAP_FOR_EACH_SAFE (e, next, hmap_node, &ml->table) { -+ HMAP_FOR_EACH_SAFE (e, hmap_node, &ml->table) { - mac_learning_expire(ml, e); - } - hmap_destroy(&ml->table); -diff --git a/lib/mcast-snooping.c b/lib/mcast-snooping.c -index 6730301b67..029ca28558 100644 ---- a/lib/mcast-snooping.c -+++ b/lib/mcast-snooping.c -@@ -356,11 +356,11 @@ mcast_snooping_prune_expired(struct mcast_snooping *ms, - OVS_REQ_WRLOCK(ms->rwlock) +diff --git a/dpdk/drivers/event/cnxk/cn10k_worker.h b/dpdk/drivers/event/cnxk/cn10k_worker.h +index f8331e88d7..f67c36f888 100644 +--- a/dpdk/drivers/event/cnxk/cn10k_worker.h ++++ b/dpdk/drivers/event/cnxk/cn10k_worker.h +@@ -169,7 +169,6 @@ cn10k_process_vwqe(uintptr_t vwqe, uint16_t port_id, const uint32_t flags, + CNXK_SSO_WQE_SG_PTR); + cnxk_nix_mbuf_to_tstamp((struct rte_mbuf *)mbuf, tstamp, + flags & NIX_RX_OFFLOAD_TSTAMP_F, +- flags & NIX_RX_MULTI_SEG_F, + (uint64_t *)tstamp_ptr); + wqe[0] = (uint64_t *)mbuf; + non_vec--; +@@ -261,7 +260,6 @@ cn10k_sso_hws_get_work(struct cn10k_sso_hws *ws, struct rte_event *ev, + cnxk_nix_mbuf_to_tstamp((struct rte_mbuf *)mbuf, + ws->tstamp, + flags & NIX_RX_OFFLOAD_TSTAMP_F, +- flags & NIX_RX_MULTI_SEG_F, + (uint64_t *)tstamp_ptr); + gw.u64[1] = mbuf; + } else if (CNXK_EVENT_TYPE_FROM_TAG(gw.u64[0]) == +@@ -526,7 +524,7 @@ cn10k_sso_hws_event_tx(struct cn10k_sso_hws *ws, struct rte_event *ev, + ev->sched_type, ws->tx_base, txq_data, flags); + } + rte_mempool_put(rte_mempool_from_obj(ev->vec), ev->vec); +- return (meta & 0xFFFF); ++ return 1; + } + + m = ev->mbuf; +diff --git a/dpdk/drivers/event/cnxk/cn9k_eventdev.c b/dpdk/drivers/event/cnxk/cn9k_eventdev.c +index b68ce6c0a4..d2ec4aedd7 100644 +--- a/dpdk/drivers/event/cnxk/cn9k_eventdev.c ++++ b/dpdk/drivers/event/cnxk/cn9k_eventdev.c +@@ -109,24 +109,21 @@ cn9k_sso_hws_release(void *arg, void *hws) + struct cnxk_sso_evdev *dev = arg; + struct cn9k_sso_hws_dual *dws; + struct cn9k_sso_hws *ws; +- int i; ++ uint16_t i; + + if (dev->dual_ws) { + dws = hws; + for (i = 0; i < dev->nb_event_queues; i++) { + roc_sso_hws_unlink(&dev->sso, +- CN9K_DUAL_WS_PAIR_ID(dws->hws_id, 0), +- (uint16_t *)&i, 1); ++ CN9K_DUAL_WS_PAIR_ID(dws->hws_id, 0), &i, 1); + roc_sso_hws_unlink(&dev->sso, +- CN9K_DUAL_WS_PAIR_ID(dws->hws_id, 1), +- (uint16_t *)&i, 1); ++ CN9K_DUAL_WS_PAIR_ID(dws->hws_id, 1), &i, 1); + } + memset(dws, 0, sizeof(*dws)); + } else { + ws = hws; + for (i = 0; i < dev->nb_event_queues; i++) +- roc_sso_hws_unlink(&dev->sso, ws->hws_id, +- (uint16_t *)&i, 1); ++ roc_sso_hws_unlink(&dev->sso, ws->hws_id, &i, 1); + memset(ws, 0, sizeof(*ws)); + } + } +diff --git a/dpdk/drivers/event/cnxk/cn9k_worker.h b/dpdk/drivers/event/cnxk/cn9k_worker.h +index 9377fa50e7..8abdd13b66 100644 +--- a/dpdk/drivers/event/cnxk/cn9k_worker.h ++++ b/dpdk/drivers/event/cnxk/cn9k_worker.h +@@ -209,7 +209,6 @@ cn9k_sso_hws_dual_get_work(uint64_t base, uint64_t pair_base, + CNXK_SSO_WQE_SG_PTR); + cnxk_nix_mbuf_to_tstamp((struct rte_mbuf *)mbuf, tstamp, + flags & NIX_RX_OFFLOAD_TSTAMP_F, +- flags & NIX_RX_MULTI_SEG_F, + (uint64_t *)tstamp_ptr); + gw.u64[1] = mbuf; + } +@@ -288,7 +287,6 @@ cn9k_sso_hws_get_work(struct cn9k_sso_hws *ws, struct rte_event *ev, + cnxk_nix_mbuf_to_tstamp((struct rte_mbuf *)mbuf, + ws->tstamp, + flags & NIX_RX_OFFLOAD_TSTAMP_F, +- flags & NIX_RX_MULTI_SEG_F, + (uint64_t *)tstamp_ptr); + gw.u64[1] = mbuf; + } +diff --git a/dpdk/drivers/event/cnxk/cnxk_eventdev.c b/dpdk/drivers/event/cnxk/cnxk_eventdev.c +index f7a5026250..46a788ef4e 100644 +--- a/dpdk/drivers/event/cnxk/cnxk_eventdev.c ++++ b/dpdk/drivers/event/cnxk/cnxk_eventdev.c +@@ -417,10 +417,10 @@ cnxk_sso_start(struct rte_eventdev *event_dev, cnxk_sso_hws_reset_t reset_fn, + + plt_sso_dbg(); + for (i = 0; i < dev->qos_queue_cnt; i++) { +- qos->hwgrp = dev->qos_parse_data[i].queue; +- qos->iaq_prcnt = dev->qos_parse_data[i].iaq_prcnt; +- qos->taq_prcnt = dev->qos_parse_data[i].taq_prcnt; +- qos->xaq_prcnt = dev->qos_parse_data[i].xaq_prcnt; ++ qos[i].hwgrp = dev->qos_parse_data[i].queue; ++ qos[i].iaq_prcnt = dev->qos_parse_data[i].iaq_prcnt; ++ qos[i].taq_prcnt = dev->qos_parse_data[i].taq_prcnt; ++ qos[i].xaq_prcnt = dev->qos_parse_data[i].xaq_prcnt; + } + rc = roc_sso_hwgrp_qos_config(&dev->sso, qos, dev->qos_queue_cnt, + dev->xae_cnt); +@@ -482,7 +482,7 @@ static void + parse_queue_param(char *value, void *opaque) { - int expired; -- struct mcast_group_bundle *b, *next_b; -+ struct mcast_group_bundle *b; - time_t timenow = time_now(); + struct cnxk_sso_qos queue_qos = {0}; +- uint8_t *val = (uint8_t *)&queue_qos; ++ uint16_t *val = (uint16_t *)&queue_qos; + struct cnxk_sso_evdev *dev = opaque; + char *tok = strtok(value, "-"); + struct cnxk_sso_qos *old_ptr; +diff --git a/dpdk/drivers/event/cnxk/cnxk_eventdev.h b/dpdk/drivers/event/cnxk/cnxk_eventdev.h +index 305c6a3b9e..39c13b02fc 100644 +--- a/dpdk/drivers/event/cnxk/cnxk_eventdev.h ++++ b/dpdk/drivers/event/cnxk/cnxk_eventdev.h +@@ -44,7 +44,7 @@ + #define CNXK_TT_FROM_EVENT(x) (((x) >> 38) & SSO_TT_EMPTY) + #define CNXK_EVENT_TYPE_FROM_TAG(x) (((x) >> 28) & 0xf) + #define CNXK_SUB_EVENT_FROM_TAG(x) (((x) >> 20) & 0xff) +-#define CNXK_CLR_SUB_EVENT(x) (~(0xffu << 20) & x) ++#define CNXK_CLR_SUB_EVENT(x) (~(0xffull << 20) & x) + #define CNXK_GRP_FROM_TAG(x) (((x) >> 36) & 0x3ff) + #define CNXK_SWTAG_PEND(x) (BIT_ULL(62) & x) + +@@ -74,9 +74,9 @@ typedef void (*cnxk_sso_hws_flush_t)(void *ws, uint8_t queue_id, uintptr_t base, + + struct cnxk_sso_qos { + uint16_t queue; +- uint8_t xaq_prcnt; +- uint8_t taq_prcnt; +- uint8_t iaq_prcnt; ++ uint16_t xaq_prcnt; ++ uint16_t taq_prcnt; ++ uint16_t iaq_prcnt; + }; - expired = 0; -- LIST_FOR_EACH_SAFE (b, next_b, bundle_node, &grp->bundle_lru) { -+ LIST_FOR_EACH_SAFE (b, bundle_node, &grp->bundle_lru) { - /* This list is sorted on expiration time. */ - if (b->expires > timenow) { - break; -@@ -946,15 +946,15 @@ mcast_snooping_wait(struct mcast_snooping *ms) - void - mcast_snooping_flush_bundle(struct mcast_snooping *ms, void *port) + struct cnxk_sso_evdev { +diff --git a/dpdk/drivers/event/cnxk/cnxk_eventdev_adptr.c b/dpdk/drivers/event/cnxk/cnxk_eventdev_adptr.c +index fdcd68ca63..54c3d6a3cb 100644 +--- a/dpdk/drivers/event/cnxk/cnxk_eventdev_adptr.c ++++ b/dpdk/drivers/event/cnxk/cnxk_eventdev_adptr.c +@@ -230,7 +230,7 @@ cnxk_sso_rx_adapter_queue_add( + cnxk_eth_dev, (uint16_t)rx_queue_id, port, + &queue_conf->ev, + !!(queue_conf->rx_queue_flags & +- RTE_EVENT_ETH_RX_ADAPTER_CAP_OVERRIDE_FLOW_ID)); ++ RTE_EVENT_ETH_RX_ADAPTER_QUEUE_FLOW_ID_VALID)); + if (queue_conf->rx_queue_flags & + RTE_EVENT_ETH_RX_ADAPTER_QUEUE_EVENT_VECTOR) { + cnxk_sso_updt_xae_cnt(dev, queue_conf->vector_mp, +diff --git a/dpdk/drivers/event/cnxk/cnxk_eventdev_selftest.c b/dpdk/drivers/event/cnxk/cnxk_eventdev_selftest.c +index 69c15b1d0a..3aa6f081a7 100644 +--- a/dpdk/drivers/event/cnxk/cnxk_eventdev_selftest.c ++++ b/dpdk/drivers/event/cnxk/cnxk_eventdev_selftest.c +@@ -140,7 +140,7 @@ _eventdev_setup(int mode) + struct rte_event_dev_info info; + int i, ret; + +- /* Create and destrory pool for each test case to make it standalone */ ++ /* Create and destroy pool for each test case to make it standalone */ + eventdev_test_mempool = rte_pktmbuf_pool_create( + pool_name, MAX_EVENTS, 0, 0, 512, rte_socket_id()); + if (!eventdev_test_mempool) { +@@ -626,6 +626,12 @@ launch_workers_and_wait(int (*main_thread)(void *), + /* start core */ -1, + /* skip main */ 1, + /* wrap */ 0); ++ if (w_lcore == RTE_MAX_LCORE) { ++ plt_err("Failed to get next available lcore"); ++ free(param); ++ return -1; ++ } ++ + rte_eal_remote_launch(main_thread, ¶m[0], w_lcore); + + for (port = 1; port < nb_workers; port++) { +@@ -635,6 +641,12 @@ launch_workers_and_wait(int (*main_thread)(void *), + param[port].dequeue_tmo_ticks = dequeue_tmo_ticks; + rte_atomic_thread_fence(__ATOMIC_RELEASE); + w_lcore = rte_get_next_lcore(w_lcore, 1, 0); ++ if (w_lcore == RTE_MAX_LCORE) { ++ plt_err("Failed to get next available lcore"); ++ free(param); ++ return -1; ++ } ++ + rte_eal_remote_launch(worker_thread, ¶m[port], w_lcore); + } + +@@ -1543,7 +1555,7 @@ cnxk_sso_selftest(const char *dev_name) + cn9k_sso_set_rsrc(dev); + if (cnxk_sso_testsuite_run(dev_name)) + return rc; +- /* Verift dual ws mode. */ ++ /* Verify dual ws mode. */ + printf("Verifying CN9K Dual workslot mode\n"); + dev->dual_ws = 1; + cn9k_sso_set_rsrc(dev); +diff --git a/dpdk/drivers/event/cnxk/cnxk_tim_worker.c b/dpdk/drivers/event/cnxk/cnxk_tim_worker.c +index 3ce99864a6..dfcfbdc797 100644 +--- a/dpdk/drivers/event/cnxk/cnxk_tim_worker.c ++++ b/dpdk/drivers/event/cnxk/cnxk_tim_worker.c +@@ -63,7 +63,7 @@ cnxk_tim_timer_arm_burst(const struct rte_event_timer_adapter *adptr, + struct cnxk_tim_ring *tim_ring = adptr->data->adapter_priv; + struct cnxk_tim_ent entry; + uint16_t index; +- int ret; ++ int ret = 0; + + cnxk_tim_sync_start_cyc(tim_ring); + for (index = 0; index < nb_timers; index++) { +diff --git a/dpdk/drivers/event/cnxk/cnxk_tim_worker.h b/dpdk/drivers/event/cnxk/cnxk_tim_worker.h +index 78e36ffafe..0c9f29cfbe 100644 +--- a/dpdk/drivers/event/cnxk/cnxk_tim_worker.h ++++ b/dpdk/drivers/event/cnxk/cnxk_tim_worker.h +@@ -233,8 +233,8 @@ cnxk_tim_add_entry_sp(struct cnxk_tim_ring *const tim_ring, + const struct cnxk_tim_ent *const pent, + const uint8_t flags) { -- struct mcast_group *g, *next_g; -- struct mcast_mrouter_bundle *m, *next_m; -+ struct mcast_group *g; -+ struct mcast_mrouter_bundle *m; ++ struct cnxk_tim_ent *chunk = NULL; + struct cnxk_tim_bkt *mirr_bkt; +- struct cnxk_tim_ent *chunk; + struct cnxk_tim_bkt *bkt; + uint64_t lock_sema; + int16_t rem; +@@ -316,8 +316,8 @@ cnxk_tim_add_entry_mp(struct cnxk_tim_ring *const tim_ring, + const struct cnxk_tim_ent *const pent, + const uint8_t flags) + { ++ struct cnxk_tim_ent *chunk = NULL; + struct cnxk_tim_bkt *mirr_bkt; +- struct cnxk_tim_ent *chunk; + struct cnxk_tim_bkt *bkt; + uint64_t lock_sema; + int64_t rem; +diff --git a/dpdk/drivers/event/dlb2/dlb2.c b/dpdk/drivers/event/dlb2/dlb2.c +index 16e9764dbf..543f793ed1 100644 +--- a/dpdk/drivers/event/dlb2/dlb2.c ++++ b/dpdk/drivers/event/dlb2/dlb2.c +@@ -61,12 +61,13 @@ static struct rte_event_dev_info evdev_dlb2_default_info = { + .max_num_events = DLB2_MAX_NUM_LDB_CREDITS, + .max_single_link_event_port_queue_pairs = + DLB2_MAX_NUM_DIR_PORTS(DLB2_HW_V2), +- .event_dev_cap = (RTE_EVENT_DEV_CAP_QUEUE_QOS | +- RTE_EVENT_DEV_CAP_EVENT_QOS | +- RTE_EVENT_DEV_CAP_BURST_MODE | ++ .event_dev_cap = (RTE_EVENT_DEV_CAP_EVENT_QOS | + RTE_EVENT_DEV_CAP_DISTRIBUTED_SCHED | +- RTE_EVENT_DEV_CAP_IMPLICIT_RELEASE_DISABLE | + RTE_EVENT_DEV_CAP_QUEUE_ALL_TYPES | ++ RTE_EVENT_DEV_CAP_BURST_MODE | ++ RTE_EVENT_DEV_CAP_IMPLICIT_RELEASE_DISABLE | ++ RTE_EVENT_DEV_CAP_RUNTIME_PORT_LINK | ++ RTE_EVENT_DEV_CAP_MULTIPLE_QUEUE_PORT | + RTE_EVENT_DEV_CAP_MAINTENANCE_FREE), + }; - if (!mcast_snooping_enabled(ms)) { - return; - } +@@ -2145,7 +2146,7 @@ dlb2_event_queue_detach_ldb(struct dlb2_eventdev *dlb2, + } - ovs_rwlock_wrlock(&ms->rwlock); -- LIST_FOR_EACH_SAFE (g, next_g, group_node, &ms->group_lru) { -+ LIST_FOR_EACH_SAFE (g, group_node, &ms->group_lru) { - if (mcast_group_delete_bundle(ms, g, port)) { - ms->need_revalidate = true; + /* This is expected with eventdev API! +- * It blindly attemmpts to unmap all queues. ++ * It blindly attempts to unmap all queues. + */ + if (i == DLB2_MAX_NUM_QIDS_PER_LDB_CQ) { + DLB2_LOG_DBG("dlb2: ignoring LB QID %d not mapped for qm_port %d.\n", +@@ -3897,31 +3898,47 @@ dlb2_hw_dequeue_sparse(struct dlb2_eventdev *dlb2, + while (num < max_num) { + struct dlb2_dequeue_qe qes[DLB2_NUM_QES_PER_CACHE_LINE]; + int num_avail; ++ + if (use_scalar) { ++ int n_iter = 0; ++ uint64_t m_rshift, m_lshift, m2_rshift, m2_lshift; ++ + num_avail = dlb2_recv_qe_sparse(qm_port, qes); + num_avail = RTE_MIN(num_avail, max_num - num); + dlb2_inc_cq_idx(qm_port, num_avail << 2); + if (num_avail == DLB2_NUM_QES_PER_CACHE_LINE) +- num += dlb2_process_dequeue_four_qes(ev_port, +- qm_port, +- &events[num], +- &qes[0]); ++ n_iter = dlb2_process_dequeue_four_qes(ev_port, ++ qm_port, ++ &events[num], ++ &qes[0]); + else if (num_avail) +- num += dlb2_process_dequeue_qes(ev_port, ++ n_iter = dlb2_process_dequeue_qes(ev_port, + qm_port, + &events[num], + &qes[0], + num_avail); ++ if (n_iter != 0) { ++ num += n_iter; ++ /* update rolling_mask for vector code support */ ++ m_rshift = qm_port->cq_rolling_mask >> n_iter; ++ m_lshift = qm_port->cq_rolling_mask << (64 - n_iter); ++ m2_rshift = qm_port->cq_rolling_mask_2 >> n_iter; ++ m2_lshift = qm_port->cq_rolling_mask_2 << ++ (64 - n_iter); ++ qm_port->cq_rolling_mask = (m_rshift | m2_lshift); ++ qm_port->cq_rolling_mask_2 = (m2_rshift | m_lshift); ++ } + } else { /* !use_scalar */ + num_avail = dlb2_recv_qe_sparse_vec(qm_port, + &events[num], + max_num - num); +- num += num_avail; + dlb2_inc_cq_idx(qm_port, num_avail << 2); ++ num += num_avail; + DLB2_INC_STAT(ev_port->stats.traffic.rx_ok, num_avail); + } + if (!num_avail) { +- if (num > 0) ++ if ((timeout == 0) || (num > 0)) ++ /* Not waiting in any form or 1+ events recd */ + break; + else if (dlb2_dequeue_wait(dlb2, ev_port, qm_port, + timeout, start_ticks)) +diff --git a/dpdk/drivers/event/dlb2/dlb2_priv.h b/dpdk/drivers/event/dlb2/dlb2_priv.h +index a5e2f8e46b..7837ae8733 100644 +--- a/dpdk/drivers/event/dlb2/dlb2_priv.h ++++ b/dpdk/drivers/event/dlb2/dlb2_priv.h +@@ -519,7 +519,7 @@ struct dlb2_eventdev_port { + bool setup_done; + /* enq_configured is set when the qm port is created */ + bool enq_configured; +- uint8_t implicit_release; /* release events before dequeueing */ ++ uint8_t implicit_release; /* release events before dequeuing */ + } __rte_cache_aligned; + + struct dlb2_queue { +diff --git a/dpdk/drivers/event/dlb2/dlb2_selftest.c b/dpdk/drivers/event/dlb2/dlb2_selftest.c +index 2113bc2c99..1863ffe049 100644 +--- a/dpdk/drivers/event/dlb2/dlb2_selftest.c ++++ b/dpdk/drivers/event/dlb2/dlb2_selftest.c +@@ -223,7 +223,7 @@ test_stop_flush(struct test *t) /* test to check we can properly flush events */ + 0, + RTE_EVENT_PORT_ATTR_DEQ_DEPTH, + &dequeue_depth)) { +- printf("%d: Error retrieveing dequeue depth\n", __LINE__); ++ printf("%d: Error retrieving dequeue depth\n", __LINE__); + goto err; + } -@@ -964,7 +964,7 @@ mcast_snooping_flush_bundle(struct mcast_snooping *ms, void *port) - } - } +diff --git a/dpdk/drivers/event/dlb2/pf/base/dlb2_hw_types.h b/dpdk/drivers/event/dlb2/pf/base/dlb2_hw_types.h +index 6b8fee3416..9511521e67 100644 +--- a/dpdk/drivers/event/dlb2/pf/base/dlb2_hw_types.h ++++ b/dpdk/drivers/event/dlb2/pf/base/dlb2_hw_types.h +@@ -27,7 +27,7 @@ + #define DLB2_MAX_NUM_SEQUENCE_NUMBER_GROUPS 2 + #define DLB2_MAX_NUM_SEQUENCE_NUMBER_MODES 5 + #define DLB2_MAX_CQ_COMP_CHECK_LOOPS 409600 +-#define DLB2_MAX_QID_EMPTY_CHECK_LOOPS (32 * 64 * 1024 * (800 / 30)) ++#define DLB2_MAX_QID_EMPTY_CHECK_LOOPS (4 * DLB2_MAX_NUM_LDB_CREDITS) + + #define DLB2_FUNC_BAR 0 + #define DLB2_CSR_BAR 2 +diff --git a/dpdk/drivers/event/dlb2/pf/base/dlb2_resource.c b/dpdk/drivers/event/dlb2/pf/base/dlb2_resource.c +index 3661b940c3..4011c24aef 100644 +--- a/dpdk/drivers/event/dlb2/pf/base/dlb2_resource.c ++++ b/dpdk/drivers/event/dlb2/pf/base/dlb2_resource.c +@@ -1057,7 +1057,7 @@ static u32 dlb2_dir_cq_token_count(struct dlb2_hw *hw, + port->init_tkn_cnt; + } -- LIST_FOR_EACH_SAFE (m, next_m, mrouter_node, &ms->mrouter_lru) { -+ LIST_FOR_EACH_SAFE (m, mrouter_node, &ms->mrouter_lru) { - if (m->port == port) { - mcast_snooping_flush_mrouter(m); - ms->need_revalidate = true; -diff --git a/lib/meta-flow.c b/lib/meta-flow.c -index e03cd8d0c5..c576ae6202 100644 ---- a/lib/meta-flow.c -+++ b/lib/meta-flow.c -@@ -3442,7 +3442,9 @@ mf_get_vl_mff(const struct mf_field *mff, - const struct vl_mff_map *vl_mff_map) +-static void dlb2_drain_dir_cq(struct dlb2_hw *hw, ++static int dlb2_drain_dir_cq(struct dlb2_hw *hw, + struct dlb2_dir_pq_pair *port) { - if (mff && mff->variable_len && vl_mff_map) { -- return &mf_get_vl_mff__(mff->id, vl_mff_map)->mf; -+ struct vl_mf_field *vl_mff = mf_get_vl_mff__(mff->id, vl_mff_map); + unsigned int port_id = port->id.phys_id; +@@ -1089,6 +1089,8 @@ static void dlb2_drain_dir_cq(struct dlb2_hw *hw, + + os_unmap_producer_port(hw, pp_addr); + } + -+ return vl_mff ? &vl_mff->mf : NULL; - } ++ return cnt; + } - return NULL; -diff --git a/lib/namemap.c b/lib/namemap.c -index 785cda4c27..dd317ea52e 100644 ---- a/lib/namemap.c -+++ b/lib/namemap.c -@@ -90,9 +90,9 @@ void - namemap_destroy(struct namemap *map) + static void dlb2_dir_port_cq_enable(struct dlb2_hw *hw, +@@ -1107,6 +1109,7 @@ static int dlb2_domain_drain_dir_cqs(struct dlb2_hw *hw, { - if (map) { -- struct namemap_node *node, *next; -+ struct namemap_node *node; + struct dlb2_list_entry *iter; + struct dlb2_dir_pq_pair *port; ++ int drain_cnt = 0; + RTE_SET_USED(iter); -- HMAP_FOR_EACH_SAFE (node, next, name_node, &map->by_name) { -+ HMAP_FOR_EACH_SAFE (node, name_node, &map->by_name) { - hmap_remove(&map->by_name, &node->name_node); - hmap_remove(&map->by_number, &node->number_node); - free(node->name); -diff --git a/lib/netdev-afxdp.c b/lib/netdev-afxdp.c -index 482400d8d1..ca3f2431ea 100644 ---- a/lib/netdev-afxdp.c -+++ b/lib/netdev-afxdp.c -@@ -235,11 +235,11 @@ netdev_afxdp_cleanup_unused_pool(struct unused_pool *pool) - static void - netdev_afxdp_sweep_unused_pools(void *aux OVS_UNUSED) - { -- struct unused_pool *pool, *next; -+ struct unused_pool *pool; - unsigned int count; + DLB2_DOM_LIST_FOR(domain->used_dir_pq_pairs, port, iter) { +@@ -1120,13 +1123,13 @@ static int dlb2_domain_drain_dir_cqs(struct dlb2_hw *hw, + if (toggle_port) + dlb2_dir_port_cq_disable(hw, port); - ovs_mutex_lock(&unused_pools_mutex); -- LIST_FOR_EACH_SAFE (pool, next, list_node, &unused_pools) { -+ LIST_FOR_EACH_SAFE (pool, list_node, &unused_pools) { +- dlb2_drain_dir_cq(hw, port); ++ drain_cnt = dlb2_drain_dir_cq(hw, port); - count = umem_pool_count(&pool->umem_info->mpool); - ovs_assert(count + pool->lost_in_rings <= NUM_FRAMES); -diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c -index b6b29c75e3..e28e397d7e 100644 ---- a/lib/netdev-dpdk.c -+++ b/lib/netdev-dpdk.c -@@ -622,9 +622,9 @@ dpdk_mp_full(const struct rte_mempool *mp) OVS_REQUIRES(dpdk_mp_mutex) + if (toggle_port) + dlb2_dir_port_cq_enable(hw, port); + } + +- return 0; ++ return drain_cnt; + } + + static u32 dlb2_dir_queue_depth(struct dlb2_hw *hw, +@@ -1170,10 +1173,20 @@ static int dlb2_domain_drain_dir_queues(struct dlb2_hw *hw, + return 0; + + for (i = 0; i < DLB2_MAX_QID_EMPTY_CHECK_LOOPS; i++) { +- dlb2_domain_drain_dir_cqs(hw, domain, true); ++ int drain_cnt; ++ ++ drain_cnt = dlb2_domain_drain_dir_cqs(hw, domain, false); + + if (dlb2_domain_dir_queues_empty(hw, domain)) + break; ++ ++ /* ++ * Allow time for DLB to schedule QEs before draining ++ * the CQs again. ++ */ ++ if (!drain_cnt) ++ rte_delay_us(1); ++ + } + + if (i == DLB2_MAX_QID_EMPTY_CHECK_LOOPS) { +@@ -1249,7 +1262,7 @@ static u32 dlb2_ldb_cq_token_count(struct dlb2_hw *hw, + port->init_tkn_cnt; + } + +-static void dlb2_drain_ldb_cq(struct dlb2_hw *hw, struct dlb2_ldb_port *port) ++static int dlb2_drain_ldb_cq(struct dlb2_hw *hw, struct dlb2_ldb_port *port) + { + u32 infl_cnt, tkn_cnt; + unsigned int i; +@@ -1289,32 +1302,37 @@ static void dlb2_drain_ldb_cq(struct dlb2_hw *hw, struct dlb2_ldb_port *port) + + os_unmap_producer_port(hw, pp_addr); + } ++ ++ return tkn_cnt; + } + +-static void dlb2_domain_drain_ldb_cqs(struct dlb2_hw *hw, ++static int dlb2_domain_drain_ldb_cqs(struct dlb2_hw *hw, + struct dlb2_hw_domain *domain, + bool toggle_port) + { + struct dlb2_list_entry *iter; + struct dlb2_ldb_port *port; ++ int drain_cnt = 0; + int i; + RTE_SET_USED(iter); + + /* If the domain hasn't been started, there's no traffic to drain */ + if (!domain->started) +- return; ++ return 0; + + for (i = 0; i < DLB2_NUM_COS_DOMAINS; i++) { + DLB2_DOM_LIST_FOR(domain->used_ldb_ports[i], port, iter) { + if (toggle_port) + dlb2_ldb_port_cq_disable(hw, port); + +- dlb2_drain_ldb_cq(hw, port); ++ drain_cnt = dlb2_drain_ldb_cq(hw, port); + + if (toggle_port) + dlb2_ldb_port_cq_enable(hw, port); + } + } ++ ++ return drain_cnt; + } + + static u32 dlb2_ldb_queue_depth(struct dlb2_hw *hw, +@@ -1375,10 +1393,19 @@ static int dlb2_domain_drain_mapped_queues(struct dlb2_hw *hw, + } + + for (i = 0; i < DLB2_MAX_QID_EMPTY_CHECK_LOOPS; i++) { +- dlb2_domain_drain_ldb_cqs(hw, domain, true); ++ int drain_cnt; ++ ++ drain_cnt = dlb2_domain_drain_ldb_cqs(hw, domain, false); + + if (dlb2_domain_mapped_queues_empty(hw, domain)) + break; ++ ++ /* ++ * Allow time for DLB to schedule QEs before draining ++ * the CQs again. ++ */ ++ if (!drain_cnt) ++ rte_delay_us(1); + } + + if (i == DLB2_MAX_QID_EMPTY_CHECK_LOOPS) { +@@ -2356,16 +2383,26 @@ static bool dlb2_domain_finish_unmap_port(struct dlb2_hw *hw, + { + u32 infl_cnt; + int i; ++ const int max_iters = 1000; ++ const int iter_poll_us = 100; + + if (port->num_pending_removals == 0) + return false; + + /* + * The unmap requires all the CQ's outstanding inflights to be +- * completed. ++ * completed. Poll up to 100ms. + */ +- infl_cnt = DLB2_CSR_RD(hw, DLB2_LSP_CQ_LDB_INFL_CNT(hw->ver, ++ for (i = 0; i < max_iters; i++) { ++ infl_cnt = DLB2_CSR_RD(hw, DLB2_LSP_CQ_LDB_INFL_CNT(hw->ver, + port->id.phys_id)); ++ ++ if (DLB2_BITS_GET(infl_cnt, ++ DLB2_LSP_CQ_LDB_INFL_CNT_COUNT) == 0) ++ break; ++ rte_delay_us_sleep(iter_poll_us); ++ } ++ + if (DLB2_BITS_GET(infl_cnt, DLB2_LSP_CQ_LDB_INFL_CNT_COUNT) > 0) + return false; + +@@ -3691,7 +3728,7 @@ dlb2_verify_create_ldb_queue_args(struct dlb2_hw *hw, + } + } + +- if (args->num_qid_inflights > 4096) { ++ if (args->num_qid_inflights < 1 || args->num_qid_inflights > 2048) { + resp->status = DLB2_ST_INVALID_QID_INFLIGHT_ALLOCATION; + return -EINVAL; + } +@@ -5316,6 +5353,7 @@ static void dlb2_log_map_qid(struct dlb2_hw *hw, + * EINVAL - A requested resource is unavailable, invalid port or queue ID, or + * the domain is not configured. + * EFAULT - Internal error (resp->status not set). ++ * EBUSY - The requested port has outstanding detach operations. + */ + int dlb2_hw_map_qid(struct dlb2_hw *hw, + u32 domain_id, +@@ -5356,8 +5394,12 @@ int dlb2_hw_map_qid(struct dlb2_hw *hw, + * attempt to complete them. This may be necessary to free up a QID + * slot for this requested mapping. + */ +- if (port->num_pending_removals) +- dlb2_domain_finish_unmap_port(hw, domain, port); ++ if (port->num_pending_removals) { ++ bool bool_ret; ++ bool_ret = dlb2_domain_finish_unmap_port(hw, domain, port); ++ if (!bool_ret) ++ return -EBUSY; ++ } + + ret = dlb2_verify_map_qid_slot_available(port, queue, resp); + if (ret) +diff --git a/dpdk/drivers/event/dlb2/rte_pmd_dlb2.h b/dpdk/drivers/event/dlb2/rte_pmd_dlb2.h +index 74399db018..1dbd885a16 100644 +--- a/dpdk/drivers/event/dlb2/rte_pmd_dlb2.h ++++ b/dpdk/drivers/event/dlb2/rte_pmd_dlb2.h +@@ -24,7 +24,7 @@ extern "C" { + * Selects the token pop mode for a DLB2 port. + */ + enum dlb2_token_pop_mode { +- /* Pop the CQ tokens immediately after dequeueing. */ ++ /* Pop the CQ tokens immediately after dequeuing. */ + AUTO_POP, + /* Pop CQ tokens after (dequeue_depth - 1) events are released. + * Supported on load-balanced ports only. +diff --git a/dpdk/drivers/event/dpaa2/dpaa2_eventdev_selftest.c b/dpdk/drivers/event/dpaa2/dpaa2_eventdev_selftest.c +index bbbd20951f..b549bdfcbb 100644 +--- a/dpdk/drivers/event/dpaa2/dpaa2_eventdev_selftest.c ++++ b/dpdk/drivers/event/dpaa2/dpaa2_eventdev_selftest.c +@@ -118,7 +118,7 @@ _eventdev_setup(int mode) + struct rte_event_dev_info info; + const char *pool_name = "evdev_dpaa2_test_pool"; + +- /* Create and destrory pool for each test case to make it standalone */ ++ /* Create and destroy pool for each test case to make it standalone */ + eventdev_test_mempool = rte_pktmbuf_pool_create(pool_name, + MAX_EVENTS, + 0 /*MBUF_CACHE_SIZE*/, +diff --git a/dpdk/drivers/event/dsw/dsw_evdev.h b/dpdk/drivers/event/dsw/dsw_evdev.h +index e64ae26f6e..c907c00c78 100644 +--- a/dpdk/drivers/event/dsw/dsw_evdev.h ++++ b/dpdk/drivers/event/dsw/dsw_evdev.h +@@ -24,7 +24,7 @@ + /* Multiple 24-bit flow ids will map to the same DSW-level flow. The + * number of DSW flows should be high enough make it unlikely that + * flow ids of several large flows hash to the same DSW-level flow. +- * Such collisions will limit parallism and thus the number of cores ++ * Such collisions will limit parallelism and thus the number of cores + * that may be utilized. However, configuring a large number of DSW + * flows might potentially, depending on traffic and actual + * application flow id value range, result in each such DSW-level flow +@@ -104,7 +104,7 @@ + /* Only one outstanding migration per port is allowed */ + #define DSW_MAX_PAUSED_FLOWS (DSW_MAX_PORTS*DSW_MAX_FLOWS_PER_MIGRATION) + +-/* Enough room for paus request/confirm and unpaus request/confirm for ++/* Enough room for pause request/confirm and unpaus request/confirm for + * all possible senders. + */ + #define DSW_CTL_IN_RING_SIZE ((DSW_MAX_PORTS-1)*4) +diff --git a/dpdk/drivers/event/dsw/dsw_event.c b/dpdk/drivers/event/dsw/dsw_event.c +index c6ed470286..e209cd5b00 100644 +--- a/dpdk/drivers/event/dsw/dsw_event.c ++++ b/dpdk/drivers/event/dsw/dsw_event.c +@@ -1096,7 +1096,7 @@ dsw_port_ctl_process(struct dsw_evdev *dsw, struct dsw_port *port) static void - dpdk_mp_sweep(void) OVS_REQUIRES(dpdk_mp_mutex) + dsw_port_note_op(struct dsw_port *port, uint16_t num_events) { -- struct dpdk_mp *dmp, *next; -+ struct dpdk_mp *dmp; +- /* To pull the control ring reasonbly often on busy ports, ++ /* To pull the control ring reasonably often on busy ports, + * each dequeued/enqueued event is considered an 'op' too. + */ + port->ops_since_bg_task += (num_events+1); +@@ -1180,7 +1180,7 @@ dsw_event_enqueue_burst_generic(struct dsw_port *source_port, + * addition, a port cannot be left "unattended" (e.g. unused) + * for long periods of time, since that would stall + * migration. Eventdev API extensions to provide a cleaner way +- * to archieve both of these functions should be ++ * to archive both of these functions should be + * considered. + */ + if (unlikely(events_len == 0)) { +diff --git a/dpdk/drivers/event/octeontx/ssovf_evdev.h b/dpdk/drivers/event/octeontx/ssovf_evdev.h +index bb1056a955..e46dc055eb 100644 +--- a/dpdk/drivers/event/octeontx/ssovf_evdev.h ++++ b/dpdk/drivers/event/octeontx/ssovf_evdev.h +@@ -88,7 +88,7 @@ + + /* + * In Cavium OCTEON TX SoC, all accesses to the device registers are +- * implictly strongly ordered. So, The relaxed version of IO operation is ++ * implicitly strongly ordered. So, The relaxed version of IO operation is + * safe to use with out any IO memory barriers. + */ + #define ssovf_read64 rte_read64_relaxed +diff --git a/dpdk/drivers/event/octeontx/ssovf_evdev_selftest.c b/dpdk/drivers/event/octeontx/ssovf_evdev_selftest.c +index d7b0d22111..b55523632a 100644 +--- a/dpdk/drivers/event/octeontx/ssovf_evdev_selftest.c ++++ b/dpdk/drivers/event/octeontx/ssovf_evdev_selftest.c +@@ -151,7 +151,7 @@ _eventdev_setup(int mode) + struct rte_event_dev_info info; + const char *pool_name = "evdev_octeontx_test_pool"; + +- /* Create and destrory pool for each test case to make it standalone */ ++ /* Create and destroy pool for each test case to make it standalone */ + eventdev_test_mempool = rte_pktmbuf_pool_create(pool_name, + MAX_EVENTS, + 0 /*MBUF_CACHE_SIZE*/, +diff --git a/dpdk/drivers/event/octeontx/ssovf_worker.h b/dpdk/drivers/event/octeontx/ssovf_worker.h +index e6ee292688..57be476394 100644 +--- a/dpdk/drivers/event/octeontx/ssovf_worker.h ++++ b/dpdk/drivers/event/octeontx/ssovf_worker.h +@@ -179,16 +179,22 @@ ssows_get_work(struct ssows *ws, struct rte_event *ev, const uint16_t flag) + ev->event = sched_type_queue | (get_work0 & 0xffffffff); + + if (get_work1) { +- if (ev->event_type == RTE_EVENT_TYPE_ETHDEV) +- get_work1 = (uintptr_t)ssovf_octeontx_wqe_to_pkt( +- get_work1, (ev->event >> 20) & 0x7F, flag, +- ws->lookup_mem); +- else if (ev->event_type == RTE_EVENT_TYPE_CRYPTODEV) ++ if (ev->event_type == RTE_EVENT_TYPE_ETHDEV) { ++ uint16_t port = (ev->event >> 20) & 0x7F; ++ ++ ev->sub_event_type = 0; ++ ev->mbuf = ssovf_octeontx_wqe_to_pkt( ++ get_work1, port, flag, ws->lookup_mem); ++ } else if (ev->event_type == RTE_EVENT_TYPE_CRYPTODEV) { + get_work1 = otx_crypto_adapter_dequeue(get_work1); +- ev->u64 = get_work1; +- } else if (unlikely((get_work0 & 0xFFFFFFFF) == 0xFFFFFFFF)) { +- ssovf_octeontx_wqe_free(get_work1); +- return 0; ++ ev->u64 = get_work1; ++ } else { ++ if (unlikely((get_work0 & 0xFFFFFFFF) == 0xFFFFFFFF)) { ++ ssovf_octeontx_wqe_free(get_work1); ++ return 0; ++ } ++ ev->u64 = get_work1; ++ } + } -- LIST_FOR_EACH_SAFE (dmp, next, list_node, &dpdk_mp_list) { -+ LIST_FOR_EACH_SAFE (dmp, list_node, &dpdk_mp_list) { - if (!dmp->refcount && dpdk_mp_full(dmp->mp)) { - VLOG_DBG("Freeing mempool \"%s\"", dmp->mp->name); - ovs_list_remove(&dmp->list_node); -@@ -2561,90 +2561,6 @@ netdev_dpdk_vhost_update_tx_counters(struct netdev_dpdk *dev, - } + return !!get_work1; +diff --git a/dpdk/drivers/event/octeontx2/otx2_evdev_selftest.c b/dpdk/drivers/event/octeontx2/otx2_evdev_selftest.c +index 48bfaf893d..a89637d60f 100644 +--- a/dpdk/drivers/event/octeontx2/otx2_evdev_selftest.c ++++ b/dpdk/drivers/event/octeontx2/otx2_evdev_selftest.c +@@ -139,7 +139,7 @@ _eventdev_setup(int mode) + struct rte_event_dev_info info; + int i, ret; + +- /* Create and destrory pool for each test case to make it standalone */ ++ /* Create and destroy pool for each test case to make it standalone */ + eventdev_test_mempool = rte_pktmbuf_pool_create(pool_name, MAX_EVENTS, + 0, 0, 512, + rte_socket_id()); +diff --git a/dpdk/drivers/event/octeontx2/otx2_worker_dual.h b/dpdk/drivers/event/octeontx2/otx2_worker_dual.h +index 36ae4dd88f..ca06d51c8a 100644 +--- a/dpdk/drivers/event/octeontx2/otx2_worker_dual.h ++++ b/dpdk/drivers/event/octeontx2/otx2_worker_dual.h +@@ -74,7 +74,7 @@ otx2_ssogws_dual_get_work(struct otx2_ssogws_state *ws, + event.flow_id, flags, lookup_mem); + /* Extracting tstamp, if PTP enabled. CGX will prepend + * the timestamp at starting of packet data and it can +- * be derieved from WQE 9 dword which corresponds to SG ++ * be derived from WQE 9 dword which corresponds to SG + * iova. + * rte_pktmbuf_mtod_offset can be used for this purpose + * but it brings down the performance as it reads +diff --git a/dpdk/drivers/event/opdl/opdl_evdev.c b/dpdk/drivers/event/opdl/opdl_evdev.c +index 15c10240b0..8b6890b220 100644 +--- a/dpdk/drivers/event/opdl/opdl_evdev.c ++++ b/dpdk/drivers/event/opdl/opdl_evdev.c +@@ -703,7 +703,7 @@ opdl_probe(struct rte_vdev_device *vdev) + } + + PMD_DRV_LOG(INFO, "DEV_ID:[%02d] : " +- "Success - creating eventdev device %s, numa_node:[%d], do_valdation:[%s]" ++ "Success - creating eventdev device %s, numa_node:[%d], do_validation:[%s]" + " , self_test:[%s]\n", + dev->data->dev_id, + name, +diff --git a/dpdk/drivers/event/opdl/opdl_test.c b/dpdk/drivers/event/opdl/opdl_test.c +index e4fc70a440..24b92df476 100644 +--- a/dpdk/drivers/event/opdl/opdl_test.c ++++ b/dpdk/drivers/event/opdl/opdl_test.c +@@ -864,7 +864,7 @@ qid_basic(struct test *t) + } + + +- /* Start the devicea */ ++ /* Start the device */ + if (!err) { + if (rte_event_dev_start(evdev) < 0) { + PMD_DRV_LOG(ERR, "%s:%d: Error with start call\n", +diff --git a/dpdk/drivers/event/sw/sw_evdev.h b/dpdk/drivers/event/sw/sw_evdev.h +index 33645bd1df..4fd1054470 100644 +--- a/dpdk/drivers/event/sw/sw_evdev.h ++++ b/dpdk/drivers/event/sw/sw_evdev.h +@@ -180,7 +180,7 @@ struct sw_port { + uint16_t outstanding_releases __rte_cache_aligned; + uint16_t inflight_max; /* app requested max inflights for this port */ + uint16_t inflight_credits; /* num credits this port has right now */ +- uint8_t implicit_release; /* release events before dequeueing */ ++ uint8_t implicit_release; /* release events before dequeuing */ + + uint16_t last_dequeue_burst_sz; /* how big the burst was */ + uint64_t last_dequeue_ticks; /* used to track burst processing time */ +diff --git a/dpdk/drivers/event/sw/sw_evdev_selftest.c b/dpdk/drivers/event/sw/sw_evdev_selftest.c +index 9768d3a0c7..cb97a4d615 100644 +--- a/dpdk/drivers/event/sw/sw_evdev_selftest.c ++++ b/dpdk/drivers/event/sw/sw_evdev_selftest.c +@@ -1109,7 +1109,7 @@ xstats_tests(struct test *t) + NULL, + 0); + +- /* Verify that the resetable stats are reset, and others are not */ ++ /* Verify that the resettable stats are reset, and others are not */ + static const uint64_t queue_expected_zero[] = { + 0 /* rx */, + 0 /* tx */, +diff --git a/dpdk/drivers/gpu/cuda/cuda.c b/dpdk/drivers/gpu/cuda/cuda.c +index 882df08e56..fd577f7167 100644 +--- a/dpdk/drivers/gpu/cuda/cuda.c ++++ b/dpdk/drivers/gpu/cuda/cuda.c +@@ -163,7 +163,7 @@ cuda_loader(void) + if (getenv("CUDA_PATH_L") == NULL) + snprintf(cuda_path, 1024, "%s", "libcuda.so"); + else +- snprintf(cuda_path, 1024, "%s%s", getenv("CUDA_PATH_L"), "libcuda.so"); ++ snprintf(cuda_path, 1024, "%s/%s", getenv("CUDA_PATH_L"), "libcuda.so"); + + cudalib = dlopen(cuda_path, RTLD_LAZY); + if (cudalib == NULL) { +@@ -437,9 +437,11 @@ mem_list_del_item(cuda_ptr_key pk) + return -EINVAL; + + /* if key is in head */ +- if (mem_alloc_list_cur->prev == NULL) ++ if (mem_alloc_list_cur->prev == NULL) { + mem_alloc_list_head = mem_alloc_list_cur->next; +- else { ++ if (mem_alloc_list_head != NULL) ++ mem_alloc_list_head->prev = NULL; ++ } else { + mem_alloc_list_cur->prev->next = mem_alloc_list_cur->next; + if (mem_alloc_list_cur->next != NULL) + mem_alloc_list_cur->next->prev = mem_alloc_list_cur->prev; +diff --git a/dpdk/drivers/mempool/cnxk/cn10k_mempool_ops.c b/dpdk/drivers/mempool/cnxk/cn10k_mempool_ops.c +index 4c669b878f..6ebbf91de5 100644 +--- a/dpdk/drivers/mempool/cnxk/cn10k_mempool_ops.c ++++ b/dpdk/drivers/mempool/cnxk/cn10k_mempool_ops.c +@@ -202,7 +202,7 @@ cn10k_mempool_deq(struct rte_mempool *mp, void **obj_table, unsigned int n) + BATCH_ALLOC_SZ, 0, 1); + /* If issue fails, try falling back to default alloc */ + if (unlikely(rc)) +- return cn10k_mempool_enq(mp, obj_table, n); ++ return cnxk_mempool_deq(mp, obj_table, n); + mem->status = BATCH_ALLOC_OP_ISSUED; + } + +diff --git a/dpdk/drivers/mempool/dpaa/dpaa_mempool.c b/dpdk/drivers/mempool/dpaa/dpaa_mempool.c +index f17aff9655..32639a3bfd 100644 +--- a/dpdk/drivers/mempool/dpaa/dpaa_mempool.c ++++ b/dpdk/drivers/mempool/dpaa/dpaa_mempool.c +@@ -258,7 +258,7 @@ dpaa_mbuf_alloc_bulk(struct rte_mempool *pool, + } + /* assigning mbuf from the acquired objects */ + for (i = 0; (i < ret) && bufs[i].addr; i++) { +- /* TODO-errata - objerved that bufs may be null ++ /* TODO-errata - observed that bufs may be null + * i.e. first buffer is valid, remaining 6 buffers + * may be null. + */ +diff --git a/dpdk/drivers/mempool/octeontx/octeontx_fpavf.c b/dpdk/drivers/mempool/octeontx/octeontx_fpavf.c +index 94dc5cd815..8fd9edced2 100644 +--- a/dpdk/drivers/mempool/octeontx/octeontx_fpavf.c ++++ b/dpdk/drivers/mempool/octeontx/octeontx_fpavf.c +@@ -669,7 +669,7 @@ octeontx_fpa_bufpool_destroy(uintptr_t handle, int node_id) + break; + } + +- /* Imsert it into an ordered linked list */ ++ /* Insert it into an ordered linked list */ + for (curr = &head; curr[0] != NULL; curr = curr[0]) { + if ((uintptr_t)node <= (uintptr_t)curr[0]) + break; +@@ -705,7 +705,7 @@ octeontx_fpa_bufpool_destroy(uintptr_t handle, int node_id) + + ret = octeontx_fpapf_aura_detach(gpool); + if (ret) { +- fpavf_log_err("Failed to dettach gaura %u. error code=%d\n", ++ fpavf_log_err("Failed to detach gaura %u. error code=%d\n", + gpool, ret); + } + +diff --git a/dpdk/drivers/net/af_xdp/compat.h b/dpdk/drivers/net/af_xdp/compat.h +index 3880dc7dd7..28ea64aeaa 100644 +--- a/dpdk/drivers/net/af_xdp/compat.h ++++ b/dpdk/drivers/net/af_xdp/compat.h +@@ -2,12 +2,17 @@ + * Copyright(c) 2020 Intel Corporation. + */ + ++#ifdef RTE_NET_AF_XDP_LIBXDP ++#include ++#else + #include ++#endif ++#include + #include + #include + + #if KERNEL_VERSION(5, 10, 0) <= LINUX_VERSION_CODE && \ +- defined(RTE_LIBRTE_AF_XDP_PMD_SHARED_UMEM) ++ defined(RTE_NET_AF_XDP_SHARED_UMEM) + #define ETH_AF_XDP_SHARED_UMEM 1 + #endif + +@@ -54,3 +59,41 @@ tx_syscall_needed(struct xsk_ring_prod *q __rte_unused) + return 1; + } + #endif ++ ++#ifdef RTE_NET_AF_XDP_LIBBPF_OBJ_OPEN ++static int load_program(const char *prog_path, struct bpf_object **obj) ++{ ++ struct bpf_program *prog; ++ int err; ++ ++ *obj = bpf_object__open_file(prog_path, NULL); ++ err = libbpf_get_error(*obj); ++ if (err) ++ return -1; ++ ++ err = bpf_object__load(*obj); ++ if (err) ++ goto out; ++ ++ prog = bpf_object__next_program(*obj, NULL); ++ if (!prog) ++ goto out; ++ ++ return bpf_program__fd(prog); ++ ++out: ++ bpf_object__close(*obj); ++ return -1; ++} ++#else ++static int load_program(const char *prog_path, struct bpf_object **obj) ++{ ++ int ret, prog_fd; ++ ++ ret = bpf_prog_load(prog_path, BPF_PROG_TYPE_XDP, obj, &prog_fd); ++ if (ret) ++ return -1; ++ ++ return prog_fd; ++} ++#endif +diff --git a/dpdk/drivers/net/af_xdp/meson.build b/dpdk/drivers/net/af_xdp/meson.build +index 3ed2b29784..1e0de23705 100644 +--- a/dpdk/drivers/net/af_xdp/meson.build ++++ b/dpdk/drivers/net/af_xdp/meson.build +@@ -9,19 +9,49 @@ endif + + sources = files('rte_eth_af_xdp.c') + ++xdp_dep = dependency('libxdp', version : '>=1.2.2', required: false, method: 'pkg-config') + bpf_dep = dependency('libbpf', required: false, method: 'pkg-config') + if not bpf_dep.found() + bpf_dep = cc.find_library('bpf', required: false) + endif + +-if bpf_dep.found() and cc.has_header('bpf/xsk.h') and cc.has_header('linux/if_xdp.h') +- ext_deps += bpf_dep +- bpf_ver_dep = dependency('libbpf', version : '>=0.2.0', +- required: false, method: 'pkg-config') +- if bpf_ver_dep.found() +- dpdk_conf.set('RTE_LIBRTE_AF_XDP_PMD_SHARED_UMEM', 1) ++if cc.has_header('linux/if_xdp.h') ++ if xdp_dep.found() and cc.has_header('xdp/xsk.h') ++ if bpf_dep.found() and cc.has_header('bpf/bpf.h') ++ cflags += ['-DRTE_NET_AF_XDP_LIBXDP'] ++ cflags += ['-DRTE_NET_AF_XDP_SHARED_UMEM'] ++ ext_deps += xdp_dep ++ ext_deps += bpf_dep ++ bpf_ver_dep = dependency('libbpf', version : '>=0.7.0', ++ required: false, method: 'pkg-config') ++ if bpf_ver_dep.found() ++ cflags += ['-DRTE_NET_AF_XDP_LIBBPF_OBJ_OPEN'] ++ endif ++ else ++ build = false ++ reason = 'missing dependency, libbpf' ++ endif ++ elif bpf_dep.found() and cc.has_header('bpf/xsk.h') and cc.has_header('bpf/bpf.h') ++ # libxdp not found. Rely solely on libbpf for xsk functionality ++ # which is only available in versions <= v0.6.0. ++ bpf_ver_dep = dependency('libbpf', version : '<=0.6.0', ++ required: false, method: 'pkg-config') ++ if bpf_ver_dep.found() ++ ext_deps += bpf_dep ++ bpf_shumem_ver_dep = dependency('libbpf', version : '>=0.2.0', ++ required: false, method: 'pkg-config') ++ if bpf_shumem_ver_dep.found() ++ cflags += ['-DRTE_NET_AF_XDP_SHARED_UMEM'] ++ endif ++ else ++ build = false ++ reason = 'missing dependency, "libxdp" or "libbpf <= v0.6.0"' ++ endif ++ else ++ build = false ++ reason = 'missing dependency, "libxdp" and "libbpf"' + endif + else + build = false +- reason = 'missing dependency, "libbpf"' ++ reason = 'missing header, "linux/if_xdp.h"' + endif +diff --git a/dpdk/drivers/net/af_xdp/rte_eth_af_xdp.c b/dpdk/drivers/net/af_xdp/rte_eth_af_xdp.c +index 96c2c9d939..9db76d4562 100644 +--- a/dpdk/drivers/net/af_xdp/rte_eth_af_xdp.c ++++ b/dpdk/drivers/net/af_xdp/rte_eth_af_xdp.c +@@ -15,8 +15,6 @@ + #include + #include + #include "af_xdp_deps.h" +-#include +-#include + + #include + #include +@@ -697,67 +695,6 @@ find_internal_resource(struct pmd_internals *port_int) + return list; } --static void --__netdev_dpdk_vhost_send(struct netdev *netdev, int qid, -- struct dp_packet **pkts, int cnt) +-/* Check if the netdev,qid context already exists */ +-static inline bool +-ctx_exists(struct pkt_rx_queue *rxq, const char *ifname, +- struct pkt_rx_queue *list_rxq, const char *list_ifname) -{ -- struct netdev_dpdk *dev = netdev_dpdk_cast(netdev); -- struct rte_mbuf **cur_pkts = (struct rte_mbuf **) pkts; -- struct netdev_dpdk_sw_stats sw_stats_add; -- unsigned int n_packets_to_free = cnt; -- unsigned int total_packets = cnt; -- int i, retries = 0; -- int max_retries = VHOST_ENQ_RETRY_MIN; -- int vid = netdev_dpdk_get_vid(dev); -- -- qid = dev->tx_q[qid % netdev->n_txq].map; +- bool exists = false; - -- if (OVS_UNLIKELY(vid < 0 || !dev->vhost_reconfigured || qid < 0 -- || !(dev->flags & NETDEV_UP))) { -- rte_spinlock_lock(&dev->stats_lock); -- dev->stats.tx_dropped+= cnt; -- rte_spinlock_unlock(&dev->stats_lock); -- goto out; -- } -- -- if (OVS_UNLIKELY(!rte_spinlock_trylock(&dev->tx_q[qid].tx_lock))) { -- COVERAGE_INC(vhost_tx_contention); -- rte_spinlock_lock(&dev->tx_q[qid].tx_lock); -- } -- -- sw_stats_add.tx_invalid_hwol_drops = cnt; -- if (userspace_tso_enabled()) { -- cnt = netdev_dpdk_prep_hwol_batch(dev, cur_pkts, cnt); -- } -- -- sw_stats_add.tx_invalid_hwol_drops -= cnt; -- sw_stats_add.tx_mtu_exceeded_drops = cnt; -- cnt = netdev_dpdk_filter_packet_len(dev, cur_pkts, cnt); -- sw_stats_add.tx_mtu_exceeded_drops -= cnt; -- -- /* Check has QoS has been configured for the netdev */ -- sw_stats_add.tx_qos_drops = cnt; -- cnt = netdev_dpdk_qos_run(dev, cur_pkts, cnt, true); -- sw_stats_add.tx_qos_drops -= cnt; -- -- n_packets_to_free = cnt; +- if (rxq->xsk_queue_idx == list_rxq->xsk_queue_idx && +- !strncmp(ifname, list_ifname, IFNAMSIZ)) { +- AF_XDP_LOG(ERR, "ctx %s,%i already exists, cannot share umem\n", +- ifname, rxq->xsk_queue_idx); +- exists = true; +- } - -- do { -- int vhost_qid = qid * VIRTIO_QNUM + VIRTIO_RXQ; -- unsigned int tx_pkts; +- return exists; +-} - -- tx_pkts = rte_vhost_enqueue_burst(vid, vhost_qid, cur_pkts, cnt); -- if (OVS_LIKELY(tx_pkts)) { -- /* Packets have been sent.*/ -- cnt -= tx_pkts; -- /* Prepare for possible retry.*/ -- cur_pkts = &cur_pkts[tx_pkts]; -- if (OVS_UNLIKELY(cnt && !retries)) { -- /* -- * Read max retries as there are packets not sent -- * and no retries have already occurred. -- */ -- atomic_read_relaxed(&dev->vhost_tx_retries_max, &max_retries); -- } -- } else { -- /* No packets sent - do not retry.*/ -- break; -- } -- } while (cnt && (retries++ < max_retries)); +-/* Get a pointer to an existing UMEM which overlays the rxq's mb_pool */ +-static inline int +-get_shared_umem(struct pkt_rx_queue *rxq, const char *ifname, +- struct xsk_umem_info **umem) +-{ +- struct internal_list *list; +- struct pmd_internals *internals; +- int i = 0, ret = 0; +- struct rte_mempool *mb_pool = rxq->mb_pool; - -- rte_spinlock_unlock(&dev->tx_q[qid].tx_lock); +- if (mb_pool == NULL) +- return ret; - -- sw_stats_add.tx_failure_drops = cnt; -- sw_stats_add.tx_retries = MIN(retries, max_retries); +- pthread_mutex_lock(&internal_list_lock); - -- rte_spinlock_lock(&dev->stats_lock); -- netdev_dpdk_vhost_update_tx_counters(dev, pkts, total_packets, -- &sw_stats_add); -- rte_spinlock_unlock(&dev->stats_lock); +- TAILQ_FOREACH(list, &internal_list, next) { +- internals = list->eth_dev->data->dev_private; +- for (i = 0; i < internals->queue_cnt; i++) { +- struct pkt_rx_queue *list_rxq = +- &internals->rx_queues[i]; +- if (rxq == list_rxq) +- continue; +- if (mb_pool == internals->rx_queues[i].mb_pool) { +- if (ctx_exists(rxq, ifname, list_rxq, +- internals->if_name)) { +- ret = -1; +- goto out; +- } +- if (__atomic_load_n( +- &internals->rx_queues[i].umem->refcnt, +- __ATOMIC_ACQUIRE)) { +- *umem = internals->rx_queues[i].umem; +- goto out; +- } +- } +- } +- } - -out: -- for (i = 0; i < n_packets_to_free; i++) { -- dp_packet_delete(pkts[i]); -- } +- pthread_mutex_unlock(&internal_list_lock); +- +- return ret; -} - - static void - netdev_dpdk_extbuf_free(void *addr OVS_UNUSED, void *opaque) + static int + eth_dev_configure(struct rte_eth_dev *dev) { -@@ -2749,76 +2665,69 @@ dpdk_copy_dp_packet_to_mbuf(struct rte_mempool *mp, struct dp_packet *pkt_orig) - return pkt_dest; +@@ -1013,6 +950,66 @@ static inline uintptr_t get_base_addr(struct rte_mempool *mp, uint64_t *align) + return aligned_addr; } --/* Tx function. Transmit packets indefinitely */ --static void --dpdk_do_tx_copy(struct netdev *netdev, int qid, struct dp_packet_batch *batch) -- OVS_NO_THREAD_SAFETY_ANALYSIS --{ -- const size_t batch_cnt = dp_packet_batch_size(batch); --#if !defined(__CHECKER__) && !defined(_WIN32) -- const size_t PKT_ARRAY_SIZE = batch_cnt; --#else -- /* Sparse or MSVC doesn't like variable length array. */ -- enum { PKT_ARRAY_SIZE = NETDEV_MAX_BURST }; --#endif -+/* Replace packets in a 'batch' with their corresponding copies using -+ * DPDK memory. -+ * -+ * Returns the number of good packets in the batch. */ -+static size_t -+dpdk_copy_batch_to_mbuf(struct netdev *netdev, struct dp_packet_batch *batch) ++/* Check if the netdev,qid context already exists */ ++static inline bool ++ctx_exists(struct pkt_rx_queue *rxq, const char *ifname, ++ struct pkt_rx_queue *list_rxq, const char *list_ifname) +{ - struct netdev_dpdk *dev = netdev_dpdk_cast(netdev); -- struct dp_packet *pkts[PKT_ARRAY_SIZE]; -- struct netdev_dpdk_sw_stats *sw_stats = dev->sw_stats; -- uint32_t cnt = batch_cnt; -- uint32_t dropped = 0; -- uint32_t tx_failure = 0; -- uint32_t mtu_drops = 0; -- uint32_t qos_drops = 0; -- -- if (dev->type != DPDK_DEV_VHOST) { -- /* Check if QoS has been configured for this netdev. */ -- cnt = netdev_dpdk_qos_run(dev, (struct rte_mbuf **) batch->packets, -- batch_cnt, false); -- qos_drops = batch_cnt - cnt; -- } -- -- uint32_t txcnt = 0; -- -- for (uint32_t i = 0; i < cnt; i++) { -- struct dp_packet *packet = batch->packets[i]; -- uint32_t size = dp_packet_size(packet); -- -- if (size > dev->max_packet_len -- && !(packet->mbuf.ol_flags & RTE_MBUF_F_TX_TCP_SEG)) { -- VLOG_WARN_RL(&rl, "Too big size %u max_packet_len %d", size, -- dev->max_packet_len); -- mtu_drops++; -- continue; -- } -+ size_t i, size = dp_packet_batch_size(batch); -+ struct dp_packet *packet; ++ bool exists = false; ++ ++ if (rxq->xsk_queue_idx == list_rxq->xsk_queue_idx && ++ !strncmp(ifname, list_ifname, IFNAMSIZ)) { ++ AF_XDP_LOG(ERR, "ctx %s,%i already exists, cannot share umem\n", ++ ifname, rxq->xsk_queue_idx); ++ exists = true; ++ } ++ ++ return exists; ++} ++ ++/* Get a pointer to an existing UMEM which overlays the rxq's mb_pool */ ++static inline int ++get_shared_umem(struct pkt_rx_queue *rxq, const char *ifname, ++ struct xsk_umem_info **umem) ++{ ++ struct internal_list *list; ++ struct pmd_internals *internals; ++ int i = 0, ret = 0; ++ struct rte_mempool *mb_pool = rxq->mb_pool; ++ ++ if (mb_pool == NULL) ++ return ret; ++ ++ pthread_mutex_lock(&internal_list_lock); ++ ++ TAILQ_FOREACH(list, &internal_list, next) { ++ internals = list->eth_dev->data->dev_private; ++ for (i = 0; i < internals->queue_cnt; i++) { ++ struct pkt_rx_queue *list_rxq = ++ &internals->rx_queues[i]; ++ if (rxq == list_rxq) ++ continue; ++ if (mb_pool == internals->rx_queues[i].mb_pool) { ++ if (ctx_exists(rxq, ifname, list_rxq, ++ internals->if_name)) { ++ ret = -1; ++ goto out; ++ } ++ if (__atomic_load_n(&internals->rx_queues[i].umem->refcnt, ++ __ATOMIC_ACQUIRE)) { ++ *umem = internals->rx_queues[i].umem; ++ goto out; ++ } ++ } ++ } ++ } ++ ++out: ++ pthread_mutex_unlock(&internal_list_lock); ++ ++ return ret; ++} ++ + static struct + xsk_umem_info *xdp_umem_configure(struct pmd_internals *internals, + struct pkt_rx_queue *rxq) +@@ -1052,7 +1049,7 @@ xsk_umem_info *xdp_umem_configure(struct pmd_internals *internals, + umem = rte_zmalloc_socket("umem", sizeof(*umem), 0, + rte_socket_id()); + if (umem == NULL) { +- AF_XDP_LOG(ERR, "Failed to allocate umem info"); ++ AF_XDP_LOG(ERR, "Failed to allocate umem info\n"); + return NULL; + } + +@@ -1065,7 +1062,7 @@ xsk_umem_info *xdp_umem_configure(struct pmd_internals *internals, + ret = xsk_umem__create(&umem->umem, base_addr, umem_size, + &rxq->fq, &rxq->cq, &usr_config); + if (ret) { +- AF_XDP_LOG(ERR, "Failed to create umem"); ++ AF_XDP_LOG(ERR, "Failed to create umem\n"); + goto err; + } + umem->buffer = base_addr; +@@ -1099,7 +1096,7 @@ xsk_umem_info *xdp_umem_configure(struct pmd_internals *internals, + + umem = rte_zmalloc_socket("umem", sizeof(*umem), 0, rte_socket_id()); + if (umem == NULL) { +- AF_XDP_LOG(ERR, "Failed to allocate umem info"); ++ AF_XDP_LOG(ERR, "Failed to allocate umem info\n"); + return NULL; + } -- pkts[txcnt] = dpdk_copy_dp_packet_to_mbuf(dev->dpdk_mp->mp, packet); -- if (OVS_UNLIKELY(!pkts[txcnt])) { -- dropped = cnt - i; -- break; -- } -+ DP_PACKET_BATCH_REFILL_FOR_EACH (i, size, packet, batch) { -+ if (OVS_UNLIKELY(packet->source == DPBUF_DPDK)) { -+ dp_packet_batch_refill(batch, packet, i); -+ } else { -+ struct dp_packet *pktcopy; +@@ -1135,7 +1132,7 @@ xsk_umem_info *xdp_umem_configure(struct pmd_internals *internals, + &usr_config); -- txcnt++; -- } -+ pktcopy = dpdk_copy_dp_packet_to_mbuf(dev->dpdk_mp->mp, packet); -+ if (pktcopy) { -+ dp_packet_batch_refill(batch, pktcopy, i); -+ } + if (ret) { +- AF_XDP_LOG(ERR, "Failed to create umem"); ++ AF_XDP_LOG(ERR, "Failed to create umem\n"); + goto err; + } + umem->mz = mz; +@@ -1151,13 +1148,13 @@ xsk_umem_info *xdp_umem_configure(struct pmd_internals *internals, + static int + load_custom_xdp_prog(const char *prog_path, int if_index, struct bpf_map **map) + { +- int ret, prog_fd = -1; ++ int ret, prog_fd; + struct bpf_object *obj; + +- ret = bpf_prog_load(prog_path, BPF_PROG_TYPE_XDP, &obj, &prog_fd); +- if (ret) { ++ prog_fd = load_program(prog_path, &obj); ++ if (prog_fd < 0) { + AF_XDP_LOG(ERR, "Failed to load program %s\n", prog_path); +- return ret; ++ return -1; + } -- if (OVS_LIKELY(txcnt)) { -- if (dev->type == DPDK_DEV_VHOST) { -- __netdev_dpdk_vhost_send(netdev, qid, pkts, txcnt); -- } else { -- tx_failure += netdev_dpdk_eth_tx_burst(dev, qid, -- (struct rte_mbuf **)pkts, -- txcnt); -+ dp_packet_delete(packet); - } - } + /* +@@ -1269,18 +1266,19 @@ xsk_configure(struct pmd_internals *internals, struct pkt_rx_queue *rxq, + cfg.bind_flags |= XDP_USE_NEED_WAKEUP; + #endif -- dropped += qos_drops + mtu_drops + tx_failure; -- if (OVS_UNLIKELY(dropped)) { -- rte_spinlock_lock(&dev->stats_lock); -- dev->stats.tx_dropped += dropped; -- sw_stats->tx_failure_drops += tx_failure; -- sw_stats->tx_mtu_exceeded_drops += mtu_drops; -- sw_stats->tx_qos_drops += qos_drops; -- rte_spinlock_unlock(&dev->stats_lock); -+ return dp_packet_batch_size(batch); -+} -+ -+static size_t -+netdev_dpdk_common_send(struct netdev *netdev, struct dp_packet_batch *batch, -+ struct netdev_dpdk_sw_stats *stats) -+{ -+ struct rte_mbuf **pkts = (struct rte_mbuf **) batch->packets; -+ struct netdev_dpdk *dev = netdev_dpdk_cast(netdev); -+ size_t cnt, pkt_cnt = dp_packet_batch_size(batch); -+ -+ memset(stats, 0, sizeof *stats); -+ -+ /* Copy dp-packets to mbufs. */ -+ if (OVS_UNLIKELY(batch->packets[0]->source != DPBUF_DPDK)) { -+ cnt = dpdk_copy_batch_to_mbuf(netdev, batch); -+ stats->tx_failure_drops += pkt_cnt - cnt; -+ pkt_cnt = cnt; - } -+ -+ /* Drop oversized packets. */ -+ cnt = netdev_dpdk_filter_packet_len(dev, pkts, pkt_cnt); -+ stats->tx_mtu_exceeded_drops += pkt_cnt - cnt; -+ pkt_cnt = cnt; -+ -+ /* Prepare each mbuf for hardware offloading. */ -+ if (userspace_tso_enabled()) { -+ cnt = netdev_dpdk_prep_hwol_batch(dev, pkts, pkt_cnt); -+ stats->tx_invalid_hwol_drops += pkt_cnt - cnt; -+ pkt_cnt = cnt; -+ } -+ -+ /* Apply Quality of Service policy. */ -+ cnt = netdev_dpdk_qos_run(dev, pkts, pkt_cnt, true); -+ stats->tx_qos_drops += pkt_cnt - cnt; -+ -+ return cnt; +- if (strnlen(internals->prog_path, PATH_MAX) && +- !internals->custom_prog_configured) { +- ret = load_custom_xdp_prog(internals->prog_path, +- internals->if_index, +- &internals->map); +- if (ret) { +- AF_XDP_LOG(ERR, "Failed to load custom XDP program %s\n", +- internals->prog_path); +- goto err; ++ if (strnlen(internals->prog_path, PATH_MAX)) { ++ if (!internals->custom_prog_configured) { ++ ret = load_custom_xdp_prog(internals->prog_path, ++ internals->if_index, ++ &internals->map); ++ if (ret) { ++ AF_XDP_LOG(ERR, "Failed to load custom XDP program %s\n", ++ internals->prog_path); ++ goto out_umem; ++ } ++ internals->custom_prog_configured = 1; + } +- internals->custom_prog_configured = 1; +- cfg.libbpf_flags = XSK_LIBBPF_FLAGS__INHIBIT_PROG_LOAD; ++ cfg.libbpf_flags |= XSK_LIBBPF_FLAGS__INHIBIT_PROG_LOAD; + } + + if (internals->shared_umem) +@@ -1294,7 +1292,7 @@ xsk_configure(struct pmd_internals *internals, struct pkt_rx_queue *rxq, + + if (ret) { + AF_XDP_LOG(ERR, "Failed to create xsk socket.\n"); +- goto err; ++ goto out_umem; + } + + /* insert the xsk into the xsks_map */ +@@ -1306,7 +1304,7 @@ xsk_configure(struct pmd_internals *internals, struct pkt_rx_queue *rxq, + &rxq->xsk_queue_idx, &fd, 0); + if (err) { + AF_XDP_LOG(ERR, "Failed to insert xsk in map.\n"); +- goto err; ++ goto out_xsk; + } + } + +@@ -1314,7 +1312,7 @@ xsk_configure(struct pmd_internals *internals, struct pkt_rx_queue *rxq, + ret = rte_pktmbuf_alloc_bulk(rxq->umem->mb_pool, fq_bufs, reserve_size); + if (ret) { + AF_XDP_LOG(DEBUG, "Failed to get enough buffers for fq.\n"); +- goto err; ++ goto out_xsk; + } + #endif + +@@ -1322,20 +1320,21 @@ xsk_configure(struct pmd_internals *internals, struct pkt_rx_queue *rxq, + ret = configure_preferred_busy_poll(rxq); + if (ret) { + AF_XDP_LOG(ERR, "Failed configure busy polling.\n"); +- goto err; ++ goto out_xsk; + } + } + + ret = reserve_fill_queue(rxq->umem, reserve_size, fq_bufs, &rxq->fq); + if (ret) { +- xsk_socket__delete(rxq->xsk); + AF_XDP_LOG(ERR, "Failed to reserve fill queue.\n"); +- goto err; ++ goto out_xsk; + } + + return 0; + +-err: ++out_xsk: ++ xsk_socket__delete(rxq->xsk); ++out_umem: + if (__atomic_sub_fetch(&rxq->umem->refcnt, 1, __ATOMIC_ACQUIRE) == 0) + xdp_umem_destroy(rxq->umem); + +diff --git a/dpdk/drivers/net/ark/ark_global.h b/dpdk/drivers/net/ark/ark_global.h +index 6f9b3013d8..49193ac5b3 100644 +--- a/dpdk/drivers/net/ark/ark_global.h ++++ b/dpdk/drivers/net/ark/ark_global.h +@@ -67,7 +67,7 @@ + typedef void (*rx_user_meta_hook_fn)(struct rte_mbuf *mbuf, + const uint32_t *meta, + void *ext_user_data); +-/* TX hook poplulate *meta, with up to 20 bytes. meta_cnt ++/* TX hook populate *meta, with up to 20 bytes. meta_cnt + * returns the number of uint32_t words populated, 0 to 5 + */ + typedef void (*tx_user_meta_hook_fn)(const struct rte_mbuf *mbuf, +diff --git a/dpdk/drivers/net/atlantic/atl_ethdev.c b/dpdk/drivers/net/atlantic/atl_ethdev.c +index 1c03e8bfa1..3a028f4290 100644 +--- a/dpdk/drivers/net/atlantic/atl_ethdev.c ++++ b/dpdk/drivers/net/atlantic/atl_ethdev.c +@@ -1423,7 +1423,7 @@ atl_dev_interrupt_action(struct rte_eth_dev *dev, + * @param handle + * Pointer to interrupt handle. + * @param param +- * The address of parameter (struct rte_eth_dev *) regsitered before. ++ * The address of parameter (struct rte_eth_dev *) registered before. + * + * @return + * void +diff --git a/dpdk/drivers/net/atlantic/atl_rxtx.c b/dpdk/drivers/net/atlantic/atl_rxtx.c +index e3f57ded73..aeb79bf5a2 100644 +--- a/dpdk/drivers/net/atlantic/atl_rxtx.c ++++ b/dpdk/drivers/net/atlantic/atl_rxtx.c +@@ -1094,7 +1094,7 @@ atl_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t nb_pkts) + * register. + * Update the RDT with the value of the last processed RX descriptor + * minus 1, to guarantee that the RDT register is never equal to the +- * RDH register, which creates a "full" ring situtation from the ++ * RDH register, which creates a "full" ring situation from the + * hardware point of view... + */ + nb_hold = (uint16_t)(nb_hold + rxq->nb_rx_hold); +diff --git a/dpdk/drivers/net/atlantic/hw_atl/hw_atl_b0.c b/dpdk/drivers/net/atlantic/hw_atl/hw_atl_b0.c +index 7d0e724019..d0eb4af928 100644 +--- a/dpdk/drivers/net/atlantic/hw_atl/hw_atl_b0.c ++++ b/dpdk/drivers/net/atlantic/hw_atl/hw_atl_b0.c +@@ -281,7 +281,7 @@ int hw_atl_b0_hw_init_rx_path(struct aq_hw_s *self) + hw_atl_rpf_vlan_outer_etht_set(self, 0x88A8U); + hw_atl_rpf_vlan_inner_etht_set(self, 0x8100U); + +- /* VLAN proimisc bu defauld */ ++ /* VLAN promisc by default */ + hw_atl_rpf_vlan_prom_mode_en_set(self, 1); + + /* Rx Interrupts */ +diff --git a/dpdk/drivers/net/axgbe/axgbe_dev.c b/dpdk/drivers/net/axgbe/axgbe_dev.c +index daeb3308f4..6a7fddffca 100644 +--- a/dpdk/drivers/net/axgbe/axgbe_dev.c ++++ b/dpdk/drivers/net/axgbe/axgbe_dev.c +@@ -1046,7 +1046,7 @@ static int axgbe_config_rx_threshold(struct axgbe_port *pdata, + return 0; } - static int -@@ -2826,25 +2735,92 @@ netdev_dpdk_vhost_send(struct netdev *netdev, int qid, - struct dp_packet_batch *batch, - bool concurrent_txq OVS_UNUSED) +-/*Distrubting fifo size */ ++/* Distributing FIFO size */ + static void axgbe_config_rx_fifo_size(struct axgbe_port *pdata) { -+ struct netdev_dpdk *dev = netdev_dpdk_cast(netdev); -+ int max_retries = VHOST_ENQ_RETRY_MIN; -+ int cnt, batch_cnt, vhost_batch_cnt; -+ int vid = netdev_dpdk_get_vid(dev); -+ struct netdev_dpdk_sw_stats stats; -+ struct rte_mbuf **pkts; -+ int retries; + unsigned int fifo_size; +diff --git a/dpdk/drivers/net/axgbe/axgbe_ethdev.c b/dpdk/drivers/net/axgbe/axgbe_ethdev.c +index 7d40c18a86..5add403235 100644 +--- a/dpdk/drivers/net/axgbe/axgbe_ethdev.c ++++ b/dpdk/drivers/net/axgbe/axgbe_ethdev.c +@@ -10,6 +10,8 @@ + #include "axgbe_regs.h" + #include "rte_time.h" + ++#include "eal_filesystem.h" ++ + static int eth_axgbe_dev_init(struct rte_eth_dev *eth_dev); + static int axgbe_dev_configure(struct rte_eth_dev *dev); + static int axgbe_dev_start(struct rte_eth_dev *dev); +@@ -284,7 +286,7 @@ static int axgbe_phy_reset(struct axgbe_port *pdata) + * @param handle + * Pointer to interrupt handle. + * @param param +- * The address of parameter (struct rte_eth_dev *) regsitered before. ++ * The address of parameter (struct rte_eth_dev *) registered before. + * + * @return + * void +@@ -1009,18 +1011,18 @@ axgbe_dev_xstats_get(struct rte_eth_dev *dev, struct rte_eth_xstat *stats, + struct axgbe_port *pdata = dev->data->dev_private; + unsigned int i; + +- if (!stats) +- return 0; ++ if (n < AXGBE_XSTATS_COUNT) ++ return AXGBE_XSTATS_COUNT; + + axgbe_read_mmc_stats(pdata); + +- for (i = 0; i < n && i < AXGBE_XSTATS_COUNT; i++) { ++ for (i = 0; i < AXGBE_XSTATS_COUNT; i++) { + stats[i].id = i; + stats[i].value = *(u64 *)((uint8_t *)&pdata->mmc_stats + + axgbe_xstats_strings[i].offset); + } -- if (OVS_UNLIKELY(batch->packets[0]->source != DPBUF_DPDK)) { -- dpdk_do_tx_copy(netdev, qid, batch); -+ batch_cnt = cnt = dp_packet_batch_size(batch); -+ qid = dev->tx_q[qid % netdev->n_txq].map; -+ if (OVS_UNLIKELY(vid < 0 || !dev->vhost_reconfigured || qid < 0 -+ || !(dev->flags & NETDEV_UP))) { -+ rte_spinlock_lock(&dev->stats_lock); -+ dev->stats.tx_dropped += cnt; -+ rte_spinlock_unlock(&dev->stats_lock); - dp_packet_delete_batch(batch, true); -- } else { -- __netdev_dpdk_vhost_send(netdev, qid, batch->packets, -- dp_packet_batch_size(batch)); -+ return 0; -+ } -+ -+ if (OVS_UNLIKELY(!rte_spinlock_trylock(&dev->tx_q[qid].tx_lock))) { -+ COVERAGE_INC(vhost_tx_contention); -+ rte_spinlock_lock(&dev->tx_q[qid].tx_lock); -+ } -+ -+ cnt = netdev_dpdk_common_send(netdev, batch, &stats); -+ -+ pkts = (struct rte_mbuf **) batch->packets; -+ vhost_batch_cnt = cnt; -+ retries = 0; -+ do { -+ int vhost_qid = qid * VIRTIO_QNUM + VIRTIO_RXQ; -+ int tx_pkts; -+ -+ tx_pkts = rte_vhost_enqueue_burst(vid, vhost_qid, pkts, cnt); -+ if (OVS_LIKELY(tx_pkts)) { -+ /* Packets have been sent.*/ -+ cnt -= tx_pkts; -+ /* Prepare for possible retry.*/ -+ pkts = &pkts[tx_pkts]; -+ if (OVS_UNLIKELY(cnt && !retries)) { -+ /* -+ * Read max retries as there are packets not sent -+ * and no retries have already occurred. -+ */ -+ atomic_read_relaxed(&dev->vhost_tx_retries_max, &max_retries); -+ } -+ } else { -+ /* No packets sent - do not retry.*/ -+ break; -+ } -+ } while (cnt && (retries++ < max_retries)); -+ -+ rte_spinlock_unlock(&dev->tx_q[qid].tx_lock); -+ -+ stats.tx_failure_drops += cnt; -+ stats.tx_retries = MIN(retries, max_retries); -+ -+ rte_spinlock_lock(&dev->stats_lock); -+ netdev_dpdk_vhost_update_tx_counters(dev, batch->packets, batch_cnt, -+ &stats); -+ rte_spinlock_unlock(&dev->stats_lock); -+ -+ pkts = (struct rte_mbuf **) batch->packets; -+ for (int i = 0; i < vhost_batch_cnt; i++) { -+ rte_pktmbuf_free(pkts[i]); - } -+ - return 0; +- return i; ++ return AXGBE_XSTATS_COUNT; } --static inline void --netdev_dpdk_send__(struct netdev_dpdk *dev, int qid, -- struct dp_packet_batch *batch, -- bool concurrent_txq) -+static int -+netdev_dpdk_eth_send(struct netdev *netdev, int qid, -+ struct dp_packet_batch *batch, bool concurrent_txq) + static int +@@ -2117,28 +2119,27 @@ static void axgbe_default_config(struct axgbe_port *pdata) + pdata->power_down = 0; + } + +-static int +-pci_device_cmp(const struct rte_device *dev, const void *_pci_id) ++/* ++ * Return PCI root complex device id on success else 0 ++ */ ++static uint16_t ++get_pci_rc_devid(void) { -+ struct rte_mbuf **pkts = (struct rte_mbuf **) batch->packets; -+ struct netdev_dpdk *dev = netdev_dpdk_cast(netdev); -+ int batch_cnt = dp_packet_batch_size(batch); -+ struct netdev_dpdk_sw_stats stats; -+ int cnt, dropped; -+ - if (OVS_UNLIKELY(!(dev->flags & NETDEV_UP))) { -+ rte_spinlock_lock(&dev->stats_lock); -+ dev->stats.tx_dropped += dp_packet_batch_size(batch); -+ rte_spinlock_unlock(&dev->stats_lock); - dp_packet_delete_batch(batch, true); -- return; -+ return 0; - } +- const struct rte_pci_device *pdev = RTE_DEV_TO_PCI_CONST(dev); +- const struct rte_pci_id *pcid = _pci_id; ++ char pci_sysfs[PATH_MAX]; ++ const struct rte_pci_addr pci_rc_addr = {0, 0, 0, 0}; ++ unsigned long device_id; + +- if (pdev->id.vendor_id == AMD_PCI_VENDOR_ID && +- pdev->id.device_id == pcid->device_id) +- return 0; +- return 1; +-} ++ snprintf(pci_sysfs, sizeof(pci_sysfs), "%s/" PCI_PRI_FMT "/device", ++ rte_pci_get_sysfs_path(), pci_rc_addr.domain, ++ pci_rc_addr.bus, pci_rc_addr.devid, pci_rc_addr.function); - if (OVS_UNLIKELY(concurrent_txq)) { -@@ -2852,56 +2828,27 @@ netdev_dpdk_send__(struct netdev_dpdk *dev, int qid, - rte_spinlock_lock(&dev->tx_q[qid].tx_lock); - } +-static bool +-pci_search_device(int device_id) +-{ +- struct rte_bus *pci_bus; +- struct rte_pci_id dev_id; ++ /* get device id */ ++ if (eal_parse_sysfs_value(pci_sysfs, &device_id) < 0) { ++ PMD_INIT_LOG(ERR, "Error in reading PCI sysfs\n"); ++ return 0; ++ } + +- dev_id.device_id = device_id; +- pci_bus = rte_bus_find_by_name("pci"); +- return (pci_bus != NULL) && +- (pci_bus->find_device(NULL, pci_device_cmp, &dev_id) != NULL); ++ return (uint16_t)device_id; + } -- if (OVS_UNLIKELY(batch->packets[0]->source != DPBUF_DPDK)) { -- struct netdev *netdev = &dev->up; -+ cnt = netdev_dpdk_common_send(netdev, batch, &stats); + /* +@@ -2180,7 +2181,7 @@ eth_axgbe_dev_init(struct rte_eth_dev *eth_dev) + /* + * Use root complex device ID to differentiate RV AXGBE vs SNOWY AXGBE + */ +- if (pci_search_device(AMD_PCI_RV_ROOT_COMPLEX_ID)) { ++ if ((get_pci_rc_devid()) == AMD_PCI_RV_ROOT_COMPLEX_ID) { + pdata->xpcs_window_def_reg = PCS_V2_RV_WINDOW_DEF; + pdata->xpcs_window_sel_reg = PCS_V2_RV_WINDOW_SELECT; + } else { +diff --git a/dpdk/drivers/net/axgbe/axgbe_ethdev.h b/dpdk/drivers/net/axgbe/axgbe_ethdev.h +index a207f2ae1b..e06d40f9eb 100644 +--- a/dpdk/drivers/net/axgbe/axgbe_ethdev.h ++++ b/dpdk/drivers/net/axgbe/axgbe_ethdev.h +@@ -641,7 +641,7 @@ struct axgbe_port { + + unsigned int kr_redrv; + +- /* Auto-negotiation atate machine support */ ++ /* Auto-negotiation state machine support */ + unsigned int an_int; + unsigned int an_status; + enum axgbe_an an_result; +diff --git a/dpdk/drivers/net/axgbe/axgbe_phy_impl.c b/dpdk/drivers/net/axgbe/axgbe_phy_impl.c +index 02236ec192..72104f8a3f 100644 +--- a/dpdk/drivers/net/axgbe/axgbe_phy_impl.c ++++ b/dpdk/drivers/net/axgbe/axgbe_phy_impl.c +@@ -347,7 +347,7 @@ static int axgbe_phy_i2c_read(struct axgbe_port *pdata, unsigned int target, + + retry = 1; + again2: +- /* Read the specfied register */ ++ /* Read the specified register */ + i2c_op.cmd = AXGBE_I2C_CMD_READ; + i2c_op.target = target; + i2c_op.len = val_len; +@@ -1093,7 +1093,7 @@ static int axgbe_phy_an_config(struct axgbe_port *pdata __rte_unused) + { + return 0; + /* Dummy API since there is no case to support +- * external phy devices registred through kerenl apis ++ * external phy devices registered through kernel APIs + */ + } -- dpdk_do_tx_copy(netdev, qid, batch); -- dp_packet_delete_batch(batch, true); -- } else { -+ dropped = batch_cnt - cnt; -+ -+ dropped += netdev_dpdk_eth_tx_burst(dev, qid, pkts, cnt); -+ if (OVS_UNLIKELY(dropped)) { - struct netdev_dpdk_sw_stats *sw_stats = dev->sw_stats; -- int dropped; -- int tx_failure, mtu_drops, qos_drops, hwol_drops; -- int batch_cnt = dp_packet_batch_size(batch); -- struct rte_mbuf **pkts = (struct rte_mbuf **) batch->packets; +diff --git a/dpdk/drivers/net/axgbe/axgbe_rxtx_vec_sse.c b/dpdk/drivers/net/axgbe/axgbe_rxtx_vec_sse.c +index 816371cd79..d95a446bef 100644 +--- a/dpdk/drivers/net/axgbe/axgbe_rxtx_vec_sse.c ++++ b/dpdk/drivers/net/axgbe/axgbe_rxtx_vec_sse.c +@@ -11,7 +11,7 @@ + #include + #include + +-/* Useful to avoid shifting for every descriptor prepration*/ ++/* Useful to avoid shifting for every descriptor preparation */ + #define TX_DESC_CTRL_FLAGS 0xb000000000000000 + #define TX_DESC_CTRL_FLAG_TMST 0x40000000 + #define TX_FREE_BULK 8 +diff --git a/dpdk/drivers/net/bnx2x/bnx2x.c b/dpdk/drivers/net/bnx2x/bnx2x.c +index f67db015b5..74e3018eab 100644 +--- a/dpdk/drivers/net/bnx2x/bnx2x.c ++++ b/dpdk/drivers/net/bnx2x/bnx2x.c +@@ -926,7 +926,7 @@ storm_memset_eq_prod(struct bnx2x_softc *sc, uint16_t eq_prod, uint16_t pfid) + * block. + * + * RAMROD_CMD_ID_ETH_UPDATE +- * Used to update the state of the leading connection, usually to udpate ++ * Used to update the state of the leading connection, usually to update + * the RSS indirection table. Completes on the RCQ of the leading + * connection. (Not currently used under FreeBSD until OS support becomes + * available.) +@@ -941,7 +941,7 @@ storm_memset_eq_prod(struct bnx2x_softc *sc, uint16_t eq_prod, uint16_t pfid) + * the RCQ of the leading connection. + * + * RAMROD_CMD_ID_ETH_CFC_DEL +- * Used when tearing down a conneciton prior to driver unload. Completes ++ * Used when tearing down a connection prior to driver unload. Completes + * on the RCQ of the leading connection (since the current connection + * has been completely removed from controller memory). + * +@@ -1072,7 +1072,7 @@ bnx2x_sp_post(struct bnx2x_softc *sc, int command, int cid, uint32_t data_hi, + + /* + * It's ok if the actual decrement is issued towards the memory +- * somewhere between the lock and unlock. Thus no more explict ++ * somewhere between the lock and unlock. Thus no more explicit + * memory barrier is needed. + */ + if (common) { +@@ -1190,7 +1190,7 @@ bnx2x_sp_event(struct bnx2x_softc *sc, struct bnx2x_fastpath *fp, + break; -- hwol_drops = batch_cnt; -- if (userspace_tso_enabled()) { -- batch_cnt = netdev_dpdk_prep_hwol_batch(dev, pkts, batch_cnt); -- } -- hwol_drops -= batch_cnt; -- mtu_drops = batch_cnt; -- batch_cnt = netdev_dpdk_filter_packet_len(dev, pkts, batch_cnt); -- mtu_drops -= batch_cnt; -- qos_drops = batch_cnt; -- batch_cnt = netdev_dpdk_qos_run(dev, pkts, batch_cnt, true); -- qos_drops -= batch_cnt; -- -- tx_failure = netdev_dpdk_eth_tx_burst(dev, qid, pkts, batch_cnt); -- -- dropped = tx_failure + mtu_drops + qos_drops + hwol_drops; -- if (OVS_UNLIKELY(dropped)) { -- rte_spinlock_lock(&dev->stats_lock); -- dev->stats.tx_dropped += dropped; -- sw_stats->tx_failure_drops += tx_failure; -- sw_stats->tx_mtu_exceeded_drops += mtu_drops; -- sw_stats->tx_qos_drops += qos_drops; -- sw_stats->tx_invalid_hwol_drops += hwol_drops; -- rte_spinlock_unlock(&dev->stats_lock); -- } -+ rte_spinlock_lock(&dev->stats_lock); -+ dev->stats.tx_dropped += dropped; -+ sw_stats->tx_failure_drops += stats.tx_failure_drops; -+ sw_stats->tx_mtu_exceeded_drops += stats.tx_mtu_exceeded_drops; -+ sw_stats->tx_qos_drops += stats.tx_qos_drops; -+ sw_stats->tx_invalid_hwol_drops += stats.tx_invalid_hwol_drops; -+ rte_spinlock_unlock(&dev->stats_lock); - } + case (RAMROD_CMD_ID_ETH_TERMINATE): +- PMD_DRV_LOG(DEBUG, sc, "got MULTI[%d] teminate ramrod", cid); ++ PMD_DRV_LOG(DEBUG, sc, "got MULTI[%d] terminate ramrod", cid); + drv_cmd = ECORE_Q_CMD_TERMINATE; + break; - if (OVS_UNLIKELY(concurrent_txq)) { - rte_spinlock_unlock(&dev->tx_q[qid].tx_lock); - } --} -- --static int --netdev_dpdk_eth_send(struct netdev *netdev, int qid, -- struct dp_packet_batch *batch, bool concurrent_txq) --{ -- struct netdev_dpdk *dev = netdev_dpdk_cast(netdev); +@@ -1476,7 +1476,7 @@ bnx2x_fill_accept_flags(struct bnx2x_softc *sc, uint32_t rx_mode, + case BNX2X_RX_MODE_ALLMULTI_PROMISC: + case BNX2X_RX_MODE_PROMISC: + /* +- * According to deffinition of SI mode, iface in promisc mode ++ * According to definition of SI mode, iface in promisc mode + * should receive matched and unmatched (in resolution of port) + * unicast packets. + */ +@@ -1944,7 +1944,7 @@ static void bnx2x_disable_close_the_gate(struct bnx2x_softc *sc) + + /* + * Cleans the object that have internal lists without sending +- * ramrods. Should be run when interrutps are disabled. ++ * ramrods. Should be run when interrupts are disabled. + */ + static void bnx2x_squeeze_objects(struct bnx2x_softc *sc) + { +@@ -2043,7 +2043,7 @@ bnx2x_nic_unload(struct bnx2x_softc *sc, uint32_t unload_mode, uint8_t keep_link + + /* + * Nothing to do during unload if previous bnx2x_nic_load() +- * did not completed successfully - all resourses are released. ++ * did not complete successfully - all resources are released. + */ + if ((sc->state == BNX2X_STATE_CLOSED) || (sc->state == BNX2X_STATE_ERROR)) { + return 0; +@@ -2084,7 +2084,7 @@ bnx2x_nic_unload(struct bnx2x_softc *sc, uint32_t unload_mode, uint8_t keep_link + /* + * Prevent transactions to host from the functions on the + * engine that doesn't reset global blocks in case of global +- * attention once gloabl blocks are reset and gates are opened ++ * attention once global blocks are reset and gates are opened + * (the engine which leader will perform the recovery + * last). + */ +@@ -2101,7 +2101,7 @@ bnx2x_nic_unload(struct bnx2x_softc *sc, uint32_t unload_mode, uint8_t keep_link + + /* + * At this stage no more interrupts will arrive so we may safely clean +- * the queue'able objects here in case they failed to get cleaned so far. ++ * the queueable objects here in case they failed to get cleaned so far. + */ + if (IS_PF(sc)) { + bnx2x_squeeze_objects(sc); +@@ -2151,7 +2151,7 @@ bnx2x_nic_unload(struct bnx2x_softc *sc, uint32_t unload_mode, uint8_t keep_link + } -- netdev_dpdk_send__(dev, qid, batch, concurrent_txq); - return 0; + /* +- * Encapsulte an mbuf cluster into the tx bd chain and makes the memory ++ * Encapsulate an mbuf cluster into the Tx BD chain and makes the memory + * visible to the controller. + * + * If an mbuf is submitted to this routine and cannot be given to the +@@ -2719,7 +2719,7 @@ static uint8_t bnx2x_clear_pf_load(struct bnx2x_softc *sc) + return val1 != 0; } -@@ -4686,11 +4633,11 @@ trtcm_policer_qos_construct(const struct smap *details, - static void - trtcm_policer_qos_destruct(struct qos_conf *conf) +-/* send load requrest to mcp and analyze response */ ++/* send load request to MCP and analyze response */ + static int bnx2x_nic_load_request(struct bnx2x_softc *sc, uint32_t * load_code) { -- struct trtcm_policer_queue *queue, *next_queue; -+ struct trtcm_policer_queue *queue; - struct trtcm_policer *policer = CONTAINER_OF(conf, struct trtcm_policer, - qos_conf); + PMD_INIT_FUNC_TRACE(sc); +@@ -5325,7 +5325,7 @@ static void bnx2x_func_init(struct bnx2x_softc *sc, struct bnx2x_func_init_param + * sum of vn_min_rates. + * or + * 0 - if all the min_rates are 0. +- * In the later case fainess algorithm should be deactivated. ++ * In the later case fairness algorithm should be deactivated. + * If all min rates are not zero then those that are zeroes will be set to 1. + */ + static void bnx2x_calc_vn_min(struct bnx2x_softc *sc, struct cmng_init_input *input) +@@ -6564,7 +6564,7 @@ bnx2x_pf_tx_q_prep(struct bnx2x_softc *sc, struct bnx2x_fastpath *fp, + txq_init->fw_sb_id = fp->fw_sb_id; + + /* +- * set the TSS leading client id for TX classfication to the ++ * set the TSS leading client id for Tx classification to the + * leading RSS client id + */ + txq_init->tss_leading_cl_id = BNX2X_FP(sc, 0, cl_id); +@@ -7634,8 +7634,8 @@ static uint8_t bnx2x_is_pcie_pending(struct bnx2x_softc *sc) + } -- HMAP_FOR_EACH_SAFE (queue, next_queue, hmap_node, &policer->queues) { -+ HMAP_FOR_EACH_SAFE (queue, hmap_node, &policer->queues) { - hmap_remove(&policer->queues, &queue->hmap_node); - free(queue); - } -diff --git a/lib/netdev-linux.c b/lib/netdev-linux.c -index 620a451dec..2766b3f2bf 100644 ---- a/lib/netdev-linux.c -+++ b/lib/netdev-linux.c -@@ -5331,11 +5331,11 @@ static void - hfsc_tc_destroy(struct tc *tc) - { - struct hfsc *hfsc; -- struct hfsc_class *hc, *next; -+ struct hfsc_class *hc; + /* +-* Walk the PCI capabiites list for the device to find what features are +-* supported. These capabilites may be enabled/disabled by firmware so it's ++* Walk the PCI capabilities list for the device to find what features are ++* supported. These capabilities may be enabled/disabled by firmware so it's + * best to walk the list rather than make assumptions. + */ + static void bnx2x_probe_pci_caps(struct bnx2x_softc *sc) +@@ -8425,7 +8425,7 @@ static int bnx2x_get_device_info(struct bnx2x_softc *sc) + } else { + sc->devinfo.int_block = INT_BLOCK_IGU; + +-/* do not allow device reset during IGU info preocessing */ ++/* do not allow device reset during IGU info processing */ + bnx2x_acquire_hw_lock(sc, HW_LOCK_RESOURCE_RESET); + + val = REG_RD(sc, IGU_REG_BLOCK_CONFIGURATION); +@@ -9765,7 +9765,7 @@ int bnx2x_attach(struct bnx2x_softc *sc) + + sc->igu_base_addr = IS_VF(sc) ? PXP_VF_ADDR_IGU_START : BAR_IGU_INTMEM; + +- /* get PCI capabilites */ ++ /* get PCI capabilities */ + bnx2x_probe_pci_caps(sc); + + if (sc->devinfo.pcie_msix_cap_reg != 0) { +@@ -10284,7 +10284,7 @@ static int bnx2x_init_hw_common(struct bnx2x_softc *sc) + * stay set) + * f. If this is VNIC 3 of a port then also init + * first_timers_ilt_entry to zero and last_timers_ilt_entry +- * to the last enrty in the ILT. ++ * to the last entry in the ILT. + * + * Notes: + * Currently the PF error in the PGLC is non recoverable. +@@ -11090,7 +11090,7 @@ static void bnx2x_hw_enable_status(struct bnx2x_softc *sc) + /** + * bnx2x_pf_flr_clnup + * a. re-enable target read on the PF +- * b. poll cfc per function usgae counter ++ * b. poll cfc per function usage counter + * c. poll the qm perfunction usage counter + * d. poll the tm per function usage counter + * e. poll the tm per function scan-done indication +diff --git a/dpdk/drivers/net/bnx2x/bnx2x.h b/dpdk/drivers/net/bnx2x/bnx2x.h +index 80d19cbfd6..d7e1729e68 100644 +--- a/dpdk/drivers/net/bnx2x/bnx2x.h ++++ b/dpdk/drivers/net/bnx2x/bnx2x.h +@@ -681,13 +681,13 @@ struct bnx2x_slowpath { + }; /* struct bnx2x_slowpath */ + + /* +- * Port specifc data structure. ++ * Port specific data structure. + */ + struct bnx2x_port { + /* + * Port Management Function (for 57711E only). + * When this field is set the driver instance is +- * responsible for managing port specifc ++ * responsible for managing port specific + * configurations such as handling link attentions. + */ + uint32_t pmf; +@@ -732,7 +732,7 @@ struct bnx2x_port { + + /* + * MCP scratchpad address for port specific statistics. +- * The device is responsible for writing statistcss ++ * The device is responsible for writing statistics + * back to the MCP for use with management firmware such + * as UMP/NC-SI. + */ +@@ -937,8 +937,8 @@ struct bnx2x_devinfo { + * already registered for this port (which means that the user wants storage + * services). + * 2. During cnic-related load, to know if offload mode is already configured +- * in the HW or needs to be configrued. Since the transition from nic-mode to +- * offload-mode in HW causes traffic coruption, nic-mode is configured only ++ * in the HW or needs to be configured. Since the transition from nic-mode to ++ * offload-mode in HW causes traffic corruption, nic-mode is configured only + * in ports on which storage services where never requested. + */ + #define CONFIGURE_NIC_MODE(sc) (!CHIP_IS_E1x(sc) && !CNIC_ENABLED(sc)) +diff --git a/dpdk/drivers/net/bnx2x/bnx2x_stats.c b/dpdk/drivers/net/bnx2x/bnx2x_stats.c +index 1cd972591a..c07b01510a 100644 +--- a/dpdk/drivers/net/bnx2x/bnx2x_stats.c ++++ b/dpdk/drivers/net/bnx2x/bnx2x_stats.c +@@ -1358,7 +1358,7 @@ bnx2x_prep_fw_stats_req(struct bnx2x_softc *sc) + + /* + * Prepare the first stats ramrod (will be completed with +- * the counters equal to zero) - init counters to somethig different. ++ * the counters equal to zero) - init counters to something different. + */ + memset(&sc->fw_stats_data->storm_counters, 0xff, + sizeof(struct stats_counter)); +diff --git a/dpdk/drivers/net/bnx2x/bnx2x_stats.h b/dpdk/drivers/net/bnx2x/bnx2x_stats.h +index 635412bdd3..11ddab5039 100644 +--- a/dpdk/drivers/net/bnx2x/bnx2x_stats.h ++++ b/dpdk/drivers/net/bnx2x/bnx2x_stats.h +@@ -314,7 +314,7 @@ struct bnx2x_eth_stats_old { + }; - hfsc = CONTAINER_OF(tc, struct hfsc, tc); + struct bnx2x_eth_q_stats_old { +- /* Fields to perserve over fw reset*/ ++ /* Fields to preserve over FW reset */ + uint32_t total_unicast_bytes_received_hi; + uint32_t total_unicast_bytes_received_lo; + uint32_t total_broadcast_bytes_received_hi; +@@ -328,7 +328,7 @@ struct bnx2x_eth_q_stats_old { + uint32_t total_multicast_bytes_transmitted_hi; + uint32_t total_multicast_bytes_transmitted_lo; + +- /* Fields to perserve last of */ ++ /* Fields to preserve last of */ + uint32_t total_bytes_received_hi; + uint32_t total_bytes_received_lo; + uint32_t total_bytes_transmitted_hi; +diff --git a/dpdk/drivers/net/bnx2x/bnx2x_vfpf.c b/dpdk/drivers/net/bnx2x/bnx2x_vfpf.c +index 945e3df84f..63953c2979 100644 +--- a/dpdk/drivers/net/bnx2x/bnx2x_vfpf.c ++++ b/dpdk/drivers/net/bnx2x/bnx2x_vfpf.c +@@ -73,7 +73,7 @@ bnx2x_add_tlv(__rte_unused struct bnx2x_softc *sc, void *tlvs_list, + tl->length = length; + } -- HMAP_FOR_EACH_SAFE (hc, next, tc_queue.hmap_node, &hfsc->tc.queues) { -+ HMAP_FOR_EACH_SAFE (hc, tc_queue.hmap_node, &hfsc->tc.queues) { - hmap_remove(&hfsc->tc.queues, &hc->tc_queue.hmap_node); - free(hc); - } -@@ -6295,7 +6295,14 @@ get_stats_via_netlink(const struct netdev *netdev_, struct netdev_stats *stats) - if (ofpbuf_try_pull(reply, NLMSG_HDRLEN + sizeof(struct ifinfomsg))) { - const struct nlattr *a = nl_attr_find(reply, 0, IFLA_STATS64); - if (a && nl_attr_get_size(a) >= sizeof(struct rtnl_link_stats64)) { -- netdev_stats_from_rtnl_link_stats64(stats, nl_attr_get(a)); -+ const struct rtnl_link_stats64 *lstats = nl_attr_get(a); -+ struct rtnl_link_stats64 aligned_lstats; -+ -+ if (!IS_PTR_ALIGNED(lstats)) { -+ memcpy(&aligned_lstats, lstats, sizeof aligned_lstats); -+ lstats = &aligned_lstats; -+ } -+ netdev_stats_from_rtnl_link_stats64(stats, lstats); - error = 0; - } else { - a = nl_attr_find(reply, 0, IFLA_STATS); -diff --git a/lib/netdev-offload-dpdk.c b/lib/netdev-offload-dpdk.c -index 94dc6a9b74..12d299603a 100644 ---- a/lib/netdev-offload-dpdk.c -+++ b/lib/netdev-offload-dpdk.c -@@ -363,6 +363,8 @@ dump_flow_pattern(struct ds *s, +-/* Initiliaze header of the first tlv and clear mailbox*/ ++/* Initialize header of the first TLV and clear mailbox */ + static void + bnx2x_vf_prep(struct bnx2x_softc *sc, struct vf_first_tlv *first_tlv, + uint16_t type, uint16_t length) +diff --git a/dpdk/drivers/net/bnx2x/bnx2x_vfpf.h b/dpdk/drivers/net/bnx2x/bnx2x_vfpf.h +index 9577341266..d71e81c005 100644 +--- a/dpdk/drivers/net/bnx2x/bnx2x_vfpf.h ++++ b/dpdk/drivers/net/bnx2x/bnx2x_vfpf.h +@@ -241,7 +241,7 @@ struct vf_close_tlv { + uint8_t pad[2]; + }; - ds_put_cstr(s, "eth "); - if (eth_spec) { -+ uint32_t has_vlan_mask; -+ - if (!eth_mask) { - eth_mask = &rte_flow_item_eth_mask; - } -@@ -377,6 +379,9 @@ dump_flow_pattern(struct ds *s, - DUMP_PATTERN_ITEM(eth_mask->type, false, "type", "0x%04"PRIx16, - ntohs(eth_spec->type), - ntohs(eth_mask->type), 0); -+ has_vlan_mask = eth_mask->has_vlan ? UINT32_MAX : 0; -+ DUMP_PATTERN_ITEM(has_vlan_mask, false, "has_vlan", "%d", -+ eth_spec->has_vlan, eth_mask->has_vlan, 0); - } - ds_put_cstr(s, "/ "); - } else if (item->type == RTE_FLOW_ITEM_TYPE_VLAN) { -@@ -1369,6 +1374,7 @@ parse_flow_match(struct netdev *netdev, - struct flow_patterns *patterns, - struct match *match) +-/* rlease the VF's acquired resources */ ++/* release the VF's acquired resources */ + struct vf_release_tlv { + struct vf_first_tlv first_tlv; + uint16_t vf_id; /* for debug */ +diff --git a/dpdk/drivers/net/bnx2x/ecore_fw_defs.h b/dpdk/drivers/net/bnx2x/ecore_fw_defs.h +index 93bca8ad33..6fc1fce7e2 100644 +--- a/dpdk/drivers/net/bnx2x/ecore_fw_defs.h ++++ b/dpdk/drivers/net/bnx2x/ecore_fw_defs.h +@@ -379,7 +379,7 @@ + /* temporarily used for RTT */ + #define XSEMI_CLK1_RESUL_CHIP (1e-3) + +-/* used for Host Coallescing */ ++/* used for Host Coalescing */ + #define SDM_TIMER_TICK_RESUL_CHIP (4 * (1e-6)) + #define TSDM_TIMER_TICK_RESUL_CHIP (1 * (1e-6)) + +diff --git a/dpdk/drivers/net/bnx2x/ecore_hsi.h b/dpdk/drivers/net/bnx2x/ecore_hsi.h +index 5508c53639..eda79408e9 100644 +--- a/dpdk/drivers/net/bnx2x/ecore_hsi.h ++++ b/dpdk/drivers/net/bnx2x/ecore_hsi.h +@@ -1062,7 +1062,7 @@ struct port_feat_cfg { /* port 0: 0x454 port 1: 0x4c8 */ + #define PORT_FEATURE_MBA_LINK_SPEED_20G 0x20000000 + + /* Secondary MBA configuration, +- * see mba_config for the fileds defination. ++ * see mba_config for the fields definition. + */ + uint32_t mba_config2; + +@@ -1075,7 +1075,7 @@ struct port_feat_cfg { /* port 0: 0x454 port 1: 0x4c8 */ + #define PORT_FEATURE_BOFM_CFGD_VEN 0x00080000 + + /* Secondary MBA configuration, +- * see mba_vlan_cfg for the fileds defination. ++ * see mba_vlan_cfg for the fields definition. + */ + uint32_t mba_vlan_cfg2; + +@@ -1429,7 +1429,7 @@ struct extended_dev_info_shared_cfg { /* NVRAM OFFSET */ + #define EXTENDED_DEV_INFO_SHARED_CFG_DBG_GEN3_COMPLI_ENA 0x00080000 + + /* Override Rx signal detect threshold when enabled the threshold +- * will be set staticaly ++ * will be set statically + */ + #define EXTENDED_DEV_INFO_SHARED_CFG_OVERRIDE_RX_SIG_MASK 0x00100000 + #define EXTENDED_DEV_INFO_SHARED_CFG_OVERRIDE_RX_SIG_SHIFT 20 +@@ -2189,9 +2189,9 @@ struct eee_remote_vals { + * elements on a per byte or word boundary. + * + * example: an array with 8 entries each 4 bit wide. This array will fit into +- * a single dword. The diagrmas below show the array order of the nibbles. ++ * a single dword. The diagrams below show the array order of the nibbles. + * +- * SHMEM_ARRAY_BITPOS(i, 4, 4) defines the stadard ordering: ++ * SHMEM_ARRAY_BITPOS(i, 4, 4) defines the standard ordering: + * + * | | | | + * 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | +@@ -2519,17 +2519,17 @@ struct shmem_lfa { + }; + + /* +- * Used to suppoert NSCI get OS driver version ++ * Used to support NSCI get OS driver version + * On driver load the version value will be set + * On driver unload driver value of 0x0 will be set + */ + struct os_drv_ver { + #define DRV_VER_NOT_LOADED 0 +- /*personalites orrder is importent */ ++ /* personalities order is important */ + #define DRV_PERS_ETHERNET 0 + #define DRV_PERS_ISCSI 1 + #define DRV_PERS_FCOE 2 +- /*shmem2 struct is constatnt can't add more personalites here*/ ++ /* shmem2 struct is constant can't add more personalities here */ + #define MAX_DRV_PERS 3 + uint32_t versions[MAX_DRV_PERS]; + }; +@@ -2821,7 +2821,7 @@ struct shmem2_region { + /* Flag to the driver that PF's drv_info_host_addr buffer was read */ + uint32_t mfw_drv_indication; /* Offset 0x19c */ + +- /* We use inidcation for each PF (0..3) */ ++ /* We use indication for each PF (0..3) */ + #define MFW_DRV_IND_READ_DONE_OFFSET(_pf_) (1 << (_pf_)) + + union { /* For various OEMs */ /* Offset 0x1a0 */ +@@ -6195,7 +6195,7 @@ struct hc_sb_data { + + + /* +- * Segment types for host coaslescing ++ * Segment types for host coalescing + */ + enum hc_segment { + HC_REGULAR_SEGMENT, +@@ -6242,7 +6242,7 @@ struct hc_status_block_data_e2 { + + + /* +- * IGU block operartion modes (in Everest2) ++ * IGU block operation modes (in Everest2) + */ + enum igu_mode { + HC_IGU_BC_MODE, +@@ -6508,7 +6508,7 @@ struct stats_query_header { + + + /* +- * Types of statistcis query entry ++ * Types of statistics query entry + */ + enum stats_query_type { + STATS_TYPE_QUEUE, +@@ -6542,7 +6542,7 @@ enum storm_id { + + + /* +- * Taffic types used in ETS and flow control algorithms ++ * Traffic types used in ETS and flow control algorithms + */ + enum traffic_type { + LLFC_TRAFFIC_TYPE_NW, +diff --git a/dpdk/drivers/net/bnx2x/ecore_init_ops.h b/dpdk/drivers/net/bnx2x/ecore_init_ops.h +index 0945e79993..4ed811fdd4 100644 +--- a/dpdk/drivers/net/bnx2x/ecore_init_ops.h ++++ b/dpdk/drivers/net/bnx2x/ecore_init_ops.h +@@ -534,7 +534,7 @@ static void ecore_init_pxp_arb(struct bnx2x_softc *sc, int r_order, + REG_WR(sc, PXP2_REG_WR_CDU_MPS, val); + } + +- /* Validate number of tags suppoted by device */ ++ /* Validate number of tags supported by device */ + #define PCIE_REG_PCIER_TL_HDR_FC_ST 0x2980 + val = REG_RD(sc, PCIE_REG_PCIER_TL_HDR_FC_ST); + val &= 0xFF; +@@ -714,7 +714,7 @@ static void ecore_ilt_client_init_op_ilt(struct bnx2x_softc *sc, + for (i = ilt_cli->start; i <= ilt_cli->end; i++) + ecore_ilt_line_init_op(sc, ilt, i, initop); + +- /* init/clear the ILT boundries */ ++ /* init/clear the ILT boundaries */ + ecore_ilt_boundary_init_op(sc, ilt_cli, ilt->start_line, initop); + } + +@@ -765,7 +765,7 @@ static void ecore_ilt_init_client_psz(struct bnx2x_softc *sc, int cli_num, + + /* + * called during init common stage, ilt clients should be initialized +- * prioir to calling this function ++ * prior to calling this function + */ + static void ecore_ilt_init_page_size(struct bnx2x_softc *sc, uint8_t initop) { -+ struct rte_flow_item_eth *eth_spec = NULL, *eth_mask = NULL; - struct flow *consumed_masks; - uint8_t proto = 0; +diff --git a/dpdk/drivers/net/bnx2x/ecore_reg.h b/dpdk/drivers/net/bnx2x/ecore_reg.h +index bb92d131f8..6f7b0522f2 100644 +--- a/dpdk/drivers/net/bnx2x/ecore_reg.h ++++ b/dpdk/drivers/net/bnx2x/ecore_reg.h +@@ -19,7 +19,7 @@ + #define ATC_ATC_INT_STS_REG_ATC_RCPL_TO_EMPTY_CNT (0x1 << 3) + #define ATC_ATC_INT_STS_REG_ATC_TCPL_ERROR (0x1 << 4) + #define ATC_ATC_INT_STS_REG_ATC_TCPL_TO_NOT_PEND (0x1 << 1) +-/* [R 1] ATC initalization done */ ++/* [R 1] ATC initialization done */ + #define ATC_REG_ATC_INIT_DONE 0x1100bc + /* [RW 6] Interrupt mask register #0 read/write */ + #define ATC_REG_ATC_INT_MASK 0x1101c8 +@@ -56,7 +56,7 @@ + #define BRB1_REG_PAUSE_HIGH_THRESHOLD_0 0x60078 + /* [RW 10] Write client 0: Assert pause threshold. Not Functional */ + #define BRB1_REG_PAUSE_LOW_THRESHOLD_0 0x60068 +-/* [R 24] The number of full blocks occpied by port. */ ++/* [R 24] The number of full blocks occupied by port. */ + #define BRB1_REG_PORT_NUM_OCC_BLOCKS_0 0x60094 + /* [R 5] Used to read the value of the XX protection CAM occupancy counter. */ + #define CCM_REG_CAM_OCCUP 0xd0188 +@@ -456,7 +456,7 @@ + #define IGU_REG_PCI_PF_MSIX_FUNC_MASK 0x130148 + #define IGU_REG_PCI_PF_MSI_EN 0x130140 + /* [WB_R 32] Each bit represent the pending bits status for that SB. 0 = no +- * pending; 1 = pending. Pendings means interrupt was asserted; and write ++ * pending; 1 = pending. Pending means interrupt was asserted; and write + * done was not received. Data valid only in addresses 0-4. all the rest are + * zero. + */ +@@ -1059,14 +1059,14 @@ + /* [R 28] this field hold the last information that caused reserved + * attention. bits [19:0] - address; [22:20] function; [23] reserved; + * [27:24] the master that caused the attention - according to the following +- * encodeing:1 = pxp; 2 = mcp; 3 = usdm; 4 = tsdm; 5 = xsdm; 6 = csdm; 7 = ++ * encoding:1 = pxp; 2 = mcp; 3 = usdm; 4 = tsdm; 5 = xsdm; 6 = csdm; 7 = + * dbu; 8 = dmae + */ + #define MISC_REG_GRC_RSV_ATTN 0xa3c0 + /* [R 28] this field hold the last information that caused timeout + * attention. bits [19:0] - address; [22:20] function; [23] reserved; + * [27:24] the master that caused the attention - according to the following +- * encodeing:1 = pxp; 2 = mcp; 3 = usdm; 4 = tsdm; 5 = xsdm; 6 = csdm; 7 = ++ * encoding:1 = pxp; 2 = mcp; 3 = usdm; 4 = tsdm; 5 = xsdm; 6 = csdm; 7 = + * dbu; 8 = dmae + */ + #define MISC_REG_GRC_TIMEOUT_ATTN 0xa3c4 +@@ -1567,7 +1567,7 @@ + * MAC DA 2. The reset default is set to mask out all parameters. + */ + #define NIG_REG_P0_LLH_PTP_PARAM_MASK 0x187a0 +-/* [RW 14] Mask regiser for the rules used in detecting PTP packets. Set ++/* [RW 14] Mask register for the rules used in detecting PTP packets. Set + * each bit to 1 to mask out that particular rule. 0-{IPv4 DA 0; UDP DP 0} . + * 1-{IPv4 DA 0; UDP DP 1} . 2-{IPv4 DA 1; UDP DP 0} . 3-{IPv4 DA 1; UDP DP + * 1} . 4-{IPv6 DA 0; UDP DP 0} . 5-{IPv6 DA 0; UDP DP 1} . 6-{IPv6 DA 1; +@@ -1672,7 +1672,7 @@ + * MAC DA 2. The reset default is set to mask out all parameters. + */ + #define NIG_REG_P0_TLLH_PTP_PARAM_MASK 0x187f0 +-/* [RW 14] Mask regiser for the rules used in detecting PTP packets. Set ++/* [RW 14] Mask register for the rules used in detecting PTP packets. Set + * each bit to 1 to mask out that particular rule. 0-{IPv4 DA 0; UDP DP 0} . + * 1-{IPv4 DA 0; UDP DP 1} . 2-{IPv4 DA 1; UDP DP 0} . 3-{IPv4 DA 1; UDP DP + * 1} . 4-{IPv6 DA 0; UDP DP 0} . 5-{IPv6 DA 0; UDP DP 1} . 6-{IPv6 DA 1; +@@ -1839,7 +1839,7 @@ + * MAC DA 2. The reset default is set to mask out all parameters. + */ + #define NIG_REG_P1_LLH_PTP_PARAM_MASK 0x187c8 +-/* [RW 14] Mask regiser for the rules used in detecting PTP packets. Set ++/* [RW 14] Mask register for the rules used in detecting PTP packets. Set + * each bit to 1 to mask out that particular rule. 0-{IPv4 DA 0; UDP DP 0} . + * 1-{IPv4 DA 0; UDP DP 1} . 2-{IPv4 DA 1; UDP DP 0} . 3-{IPv4 DA 1; UDP DP + * 1} . 4-{IPv6 DA 0; UDP DP 0} . 5-{IPv6 DA 0; UDP DP 1} . 6-{IPv6 DA 1; +@@ -1926,7 +1926,7 @@ + * MAC DA 2. The reset default is set to mask out all parameters. + */ + #define NIG_REG_P1_TLLH_PTP_PARAM_MASK 0x187f8 +-/* [RW 14] Mask regiser for the rules used in detecting PTP packets. Set ++/* [RW 14] Mask register for the rules used in detecting PTP packets. Set + * each bit to 1 to mask out that particular rule. 0-{IPv4 DA 0; UDP DP 0} . + * 1-{IPv4 DA 0; UDP DP 1} . 2-{IPv4 DA 1; UDP DP 0} . 3-{IPv4 DA 1; UDP DP + * 1} . 4-{IPv6 DA 0; UDP DP 0} . 5-{IPv6 DA 0; UDP DP 1} . 6-{IPv6 DA 1; +@@ -2306,7 +2306,7 @@ + #define PBF_REG_HDRS_AFTER_BASIC 0x15c0a8 + /* [RW 6] Bit-map indicating which L2 hdrs may appear after L2 tag 0 */ + #define PBF_REG_HDRS_AFTER_TAG_0 0x15c0b8 +-/* [R 1] Removed for E3 B0 - Indicates which COS is conncted to the highest ++/* [R 1] Removed for E3 B0 - Indicates which COS is connected to the highest + * priority in the command arbiter. + */ + #define PBF_REG_HIGH_PRIORITY_COS_NUM 0x15c04c +@@ -2366,7 +2366,7 @@ + */ + #define PBF_REG_NUM_STRICT_ARB_SLOTS 0x15c064 + /* [R 11] Removed for E3 B0 - Port 0 threshold used by arbiter in 16 byte +- * lines used when pause not suppoterd. ++ * lines used when pause not supported. + */ + #define PBF_REG_P0_ARB_THRSH 0x1400e4 + /* [R 11] Removed for E3 B0 - Current credit for port 0 in the tx port +@@ -3503,7 +3503,7 @@ + * queues. + */ + #define QM_REG_OVFERROR 0x16805c +-/* [RC 6] the Q were the qverflow occurs */ ++/* [RC 6] the Q were the overflow occurs */ + #define QM_REG_OVFQNUM 0x168058 + /* [R 16] Pause state for physical queues 15-0 */ + #define QM_REG_PAUSESTATE0 0x168410 +@@ -4890,7 +4890,7 @@ + if set, generate pcie_err_attn output when this error is seen. WC \ + */ + #define PXPCS_TL_FUNC345_STAT_ERR_MASTER_ABRT2 \ +- (1 << 3) /* Receive UR Statusfor Function 2. If set, generate \ ++ (1 << 3) /* Receive UR Status for Function 2. If set, generate \ + pcie_err_attn output when this error is seen. WC */ + #define PXPCS_TL_FUNC345_STAT_ERR_CPL_TIMEOUT2 \ + (1 << 2) /* Completer Timeout Status Status for Function 2, if \ +@@ -4986,7 +4986,7 @@ + if set, generate pcie_err_attn output when this error is seen. WC \ + */ + #define PXPCS_TL_FUNC678_STAT_ERR_MASTER_ABRT5 \ +- (1 << 3) /* Receive UR Statusfor Function 5. If set, generate \ ++ (1 << 3) /* Receive UR Status for Function 5. If set, generate \ + pcie_err_attn output when this error is seen. WC */ + #define PXPCS_TL_FUNC678_STAT_ERR_CPL_TIMEOUT5 \ + (1 << 2) /* Completer Timeout Status Status for Function 5, if \ +diff --git a/dpdk/drivers/net/bnx2x/ecore_sp.c b/dpdk/drivers/net/bnx2x/ecore_sp.c +index 0075422eee..c6c3857778 100644 +--- a/dpdk/drivers/net/bnx2x/ecore_sp.c ++++ b/dpdk/drivers/net/bnx2x/ecore_sp.c +@@ -1338,7 +1338,7 @@ static int __ecore_vlan_mac_execute_step(struct bnx2x_softc *sc, + if (rc != ECORE_SUCCESS) { + __ecore_vlan_mac_h_pend(sc, o, *ramrod_flags); + +- /** Calling function should not diffrentiate between this case ++ /** Calling function should not differentiate between this case + * and the case in which there is already a pending ramrod + */ + rc = ECORE_PENDING; +@@ -2246,7 +2246,7 @@ struct ecore_pending_mcast_cmd { + union { + ecore_list_t macs_head; + uint32_t macs_num; /* Needed for DEL command */ +- int next_bin; /* Needed for RESTORE flow with aprox match */ ++ int next_bin; /* Needed for RESTORE flow with approx match */ + } data; + + int done; /* set to TRUE, when the command has been handled, +@@ -3424,7 +3424,7 @@ void ecore_init_mac_credit_pool(struct bnx2x_softc *sc, + } else { + + /* +- * CAM credit is equaly divided between all active functions ++ * CAM credit is equally divided between all active functions + * on the PATH. + */ + if (func_num > 0) { +diff --git a/dpdk/drivers/net/bnx2x/ecore_sp.h b/dpdk/drivers/net/bnx2x/ecore_sp.h +index d58072dac0..1f4d5a3ebe 100644 +--- a/dpdk/drivers/net/bnx2x/ecore_sp.h ++++ b/dpdk/drivers/net/bnx2x/ecore_sp.h +@@ -430,7 +430,7 @@ enum { + RAMROD_RESTORE, + /* Execute the next command now */ + RAMROD_EXEC, +- /* Don't add a new command and continue execution of posponed ++ /* Don't add a new command and continue execution of postponed + * commands. If not set a new command will be added to the + * pending commands list. + */ +@@ -1173,7 +1173,7 @@ struct ecore_rss_config_obj { + /* Last configured indirection table */ + uint8_t ind_table[T_ETH_INDIRECTION_TABLE_SIZE]; + +- /* flags for enabling 4-tupple hash on UDP */ ++ /* flags for enabling 4-tuple hash on UDP */ + uint8_t udp_rss_v4; + uint8_t udp_rss_v6; + +@@ -1285,7 +1285,7 @@ enum ecore_q_type { + #define ECORE_MULTI_TX_COS_E3B0 3 + #define ECORE_MULTI_TX_COS 3 /* Maximum possible */ + #define MAC_PAD (ECORE_ALIGN(ETH_ALEN, sizeof(uint32_t)) - ETH_ALEN) +-/* DMAE channel to be used by FW for timesync workaroun. A driver that sends ++/* DMAE channel to be used by FW for timesync workaround. A driver that sends + * timesync-related ramrods must not use this DMAE command ID. + */ + #define FW_DMAE_CMD_ID 6 +diff --git a/dpdk/drivers/net/bnx2x/elink.c b/dpdk/drivers/net/bnx2x/elink.c +index 2093d8f373..43fbf04ece 100644 +--- a/dpdk/drivers/net/bnx2x/elink.c ++++ b/dpdk/drivers/net/bnx2x/elink.c +@@ -1460,7 +1460,7 @@ static void elink_ets_e3b0_pbf_disabled(const struct elink_params *params) + } + /****************************************************************************** + * Description: +- * E3B0 disable will return basicly the values to init values. ++ * E3B0 disable will return basically the values to init values. + *. + ******************************************************************************/ + static elink_status_t elink_ets_e3b0_disabled(const struct elink_params *params, +@@ -1483,7 +1483,7 @@ static elink_status_t elink_ets_e3b0_disabled(const struct elink_params *params, + + /****************************************************************************** + * Description: +- * Disable will return basicly the values to init values. ++ * Disable will return basically the values to init values. + * + ******************************************************************************/ + elink_status_t elink_ets_disabled(struct elink_params *params, +@@ -1506,7 +1506,7 @@ elink_status_t elink_ets_disabled(struct elink_params *params, + + /****************************************************************************** + * Description +- * Set the COS mappimg to SP and BW until this point all the COS are not ++ * Set the COS mapping to SP and BW until this point all the COS are not + * set as SP or BW. + ******************************************************************************/ + static elink_status_t elink_ets_e3b0_cli_map(const struct elink_params *params, +@@ -1652,7 +1652,7 @@ static elink_status_t elink_ets_e3b0_get_total_bw( + } + ELINK_DEBUG_P0(sc, + "elink_ets_E3B0_config total BW should be 100"); +- /* We can handle a case whre the BW isn't 100 this can happen ++ /* We can handle a case where the BW isn't 100 this can happen + * if the TC are joined. + */ + } +@@ -2608,7 +2608,7 @@ static elink_status_t elink_emac_enable(struct elink_params *params, + REG_WR(sc, NIG_REG_EGRESS_EMAC0_PORT + port * 4, 1); + + #ifdef ELINK_INCLUDE_EMUL +- /* for paladium */ ++ /* for palladium */ + if (CHIP_REV_IS_EMUL(sc)) { + /* Use lane 1 (of lanes 0-3) */ + REG_WR(sc, NIG_REG_XGXS_LANE_SEL_P0 + port * 4, 1); +@@ -2850,7 +2850,7 @@ static void elink_update_pfc_bmac2(struct elink_params *params, + + /* Set Time (based unit is 512 bit time) between automatic + * re-sending of PP packets amd enable automatic re-send of +- * Per-Priroity Packet as long as pp_gen is asserted and ++ * Per-Priority Packet as long as pp_gen is asserted and + * pp_disable is low. + */ + val = 0x8000; +@@ -3369,7 +3369,7 @@ static elink_status_t elink_pbf_update(struct elink_params *params, + } -@@ -1414,6 +1420,11 @@ parse_flow_match(struct netdev *netdev, - memset(&consumed_masks->dl_src, 0, sizeof consumed_masks->dl_src); - consumed_masks->dl_type = 0; + /** +- * elink_get_emac_base - retrive emac base address ++ * elink_get_emac_base - retrieve emac base address + * + * @bp: driver handle + * @mdc_mdio_access: access type +@@ -4518,7 +4518,7 @@ static void elink_warpcore_enable_AN_KR2(struct elink_phy *phy, + elink_cl45_write(sc, phy, reg_set[i].devad, reg_set[i].reg, + reg_set[i].val); + +- /* Start KR2 work-around timer which handles BNX2X8073 link-parner */ ++ /* Start KR2 work-around timer which handles BNX2X8073 link-partner */ + params->link_attr_sync |= LINK_ATTR_SYNC_KR2_ENABLE; + elink_update_link_attr(params, params->link_attr_sync); + } +@@ -7824,7 +7824,7 @@ elink_status_t elink_link_update(struct elink_params *params, + * hence its link is expected to be down + * - SECOND_PHY means that first phy should not be able + * to link up by itself (using configuration) +- * - DEFAULT should be overridden during initialiazation ++ * - DEFAULT should be overridden during initialization + */ + ELINK_DEBUG_P1(sc, "Invalid link indication" + " mpc=0x%x. DISABLING LINK !!!", +@@ -10991,7 +10991,7 @@ static elink_status_t elink_84858_cmd_hdlr(struct elink_phy *phy, + ELINK_DEBUG_P0(sc, "FW cmd failed."); + return ELINK_STATUS_ERROR; + } +- /* Step5: Once the command has completed, read the specficied DATA ++ /* Step5: Once the command has completed, read the specified DATA + * registers for any saved results for the command, if applicable + */ + +diff --git a/dpdk/drivers/net/bnxt/bnxt.h b/dpdk/drivers/net/bnxt/bnxt.h +index 234161053f..76783eb3a1 100644 +--- a/dpdk/drivers/net/bnxt/bnxt.h ++++ b/dpdk/drivers/net/bnxt/bnxt.h +@@ -72,8 +72,7 @@ + #define BROADCOM_DEV_ID_58818_VF 0xd82e + + #define BROADCOM_DEV_957508_N2100 0x5208 +-#define IS_BNXT_DEV_957508_N2100(bp) \ +- ((bp)->pdev->id.subsystem_device_id == BROADCOM_DEV_957508_N2100) ++#define BROADCOM_DEV_957414_N225 0x4145 + + #define BNXT_MAX_MTU 9574 + #define BNXT_NUM_VLANS 2 +@@ -297,7 +296,7 @@ struct bnxt_link_info { + uint8_t link_signal_mode; + uint16_t force_pam4_link_speed; + uint16_t support_pam4_speeds; +- uint16_t auto_pam4_link_speeds; ++ uint16_t auto_pam4_link_speed_mask; + uint16_t support_pam4_auto_speeds; + uint8_t req_signal_mode; + uint8_t module_status; +@@ -580,30 +579,6 @@ struct bnxt_rep_info { + RTE_ETH_RSS_NONFRAG_IPV6_UDP | \ + RTE_ETH_RSS_LEVEL_MASK) + +-#define BNXT_DEV_TX_OFFLOAD_SUPPORT (RTE_ETH_TX_OFFLOAD_IPV4_CKSUM | \ +- RTE_ETH_TX_OFFLOAD_TCP_CKSUM | \ +- RTE_ETH_TX_OFFLOAD_UDP_CKSUM | \ +- RTE_ETH_TX_OFFLOAD_TCP_TSO | \ +- RTE_ETH_TX_OFFLOAD_OUTER_IPV4_CKSUM | \ +- RTE_ETH_TX_OFFLOAD_VXLAN_TNL_TSO | \ +- RTE_ETH_TX_OFFLOAD_GRE_TNL_TSO | \ +- RTE_ETH_TX_OFFLOAD_IPIP_TNL_TSO | \ +- RTE_ETH_TX_OFFLOAD_GENEVE_TNL_TSO | \ +- RTE_ETH_TX_OFFLOAD_QINQ_INSERT | \ +- RTE_ETH_TX_OFFLOAD_MULTI_SEGS) +- +-#define BNXT_DEV_RX_OFFLOAD_SUPPORT (RTE_ETH_RX_OFFLOAD_VLAN_FILTER | \ +- RTE_ETH_RX_OFFLOAD_IPV4_CKSUM | \ +- RTE_ETH_RX_OFFLOAD_UDP_CKSUM | \ +- RTE_ETH_RX_OFFLOAD_TCP_CKSUM | \ +- RTE_ETH_RX_OFFLOAD_OUTER_IPV4_CKSUM | \ +- RTE_ETH_RX_OFFLOAD_OUTER_UDP_CKSUM | \ +- RTE_ETH_RX_OFFLOAD_KEEP_CRC | \ +- RTE_ETH_RX_OFFLOAD_VLAN_EXTEND | \ +- RTE_ETH_RX_OFFLOAD_TCP_LRO | \ +- RTE_ETH_RX_OFFLOAD_SCATTER | \ +- RTE_ETH_RX_OFFLOAD_RSS_HASH) +- + #define BNXT_HWRM_SHORT_REQ_LEN sizeof(struct hwrm_short_input) + + struct bnxt_flow_stat_info { +@@ -672,7 +647,6 @@ struct bnxt { + #define BNXT_FLAG_PORT_STATS BIT(2) + #define BNXT_FLAG_JUMBO BIT(3) + #define BNXT_FLAG_SHORT_CMD BIT(4) +-#define BNXT_FLAG_UPDATE_HASH BIT(5) + #define BNXT_FLAG_PTP_SUPPORTED BIT(6) + #define BNXT_FLAG_MULTI_HOST BIT(7) + #define BNXT_FLAG_EXT_RX_PORT_STATS BIT(8) +@@ -695,9 +669,6 @@ struct bnxt { + #define BNXT_FLAG_FLOW_XSTATS_EN BIT(25) + #define BNXT_FLAG_DFLT_MAC_SET BIT(26) + #define BNXT_FLAG_GFID_ENABLE BIT(27) +-#define BNXT_FLAG_RFS_NEEDS_VNIC BIT(28) +-#define BNXT_FLAG_FLOW_CFA_RFS_RING_TBL_IDX_V2 BIT(29) +-#define BNXT_RFS_NEEDS_VNIC(bp) ((bp)->flags & BNXT_FLAG_RFS_NEEDS_VNIC) + #define BNXT_PF(bp) (!((bp)->flags & BNXT_FLAG_VF)) + #define BNXT_VF(bp) ((bp)->flags & BNXT_FLAG_VF) + #define BNXT_NPAR(bp) ((bp)->flags & BNXT_FLAG_NPAR_PF) +@@ -834,7 +805,7 @@ struct bnxt { + uint16_t max_tx_rings; + uint16_t max_rx_rings; + #define MAX_STINGRAY_RINGS 236U +-#define BNXT_MAX_VF_REP_RINGS 8 ++#define BNXT_MAX_VF_REP_RINGS 8U + + uint16_t max_nq_rings; + uint16_t max_l2_ctx; +@@ -891,6 +862,15 @@ struct bnxt { + uint16_t tx_cfa_action; + struct bnxt_ring_stats *prev_rx_ring_stats; + struct bnxt_ring_stats *prev_tx_ring_stats; ++ ++#define BNXT_MAX_MC_ADDRS ((bp)->max_mcast_addr) ++ struct rte_ether_addr *mcast_addr_list; ++ rte_iova_t mc_list_dma_addr; ++ uint32_t nb_mc_addr; ++ uint32_t max_mcast_addr; /* maximum number of mcast filters supported */ ++ ++ struct rte_eth_rss_conf rss_conf; /* RSS configuration. */ ++ uint16_t tunnel_disable_flag; /* tunnel stateless offloads status */ + }; -+ spec->has_vlan = 0; -+ mask->has_vlan = 1; -+ eth_spec = spec; -+ eth_mask = mask; + static +@@ -1063,5 +1043,8 @@ int bnxt_flow_stats_cnt(struct bnxt *bp); + uint32_t bnxt_get_speed_capabilities(struct bnxt *bp); + int bnxt_flow_ops_get_op(struct rte_eth_dev *dev, + const struct rte_flow_ops **ops); ++int bnxt_dev_start_op(struct rte_eth_dev *eth_dev); ++int bnxt_dev_stop_op(struct rte_eth_dev *eth_dev); ++void bnxt_handle_vf_cfg_change(void *arg); + + #endif +diff --git a/dpdk/drivers/net/bnxt/bnxt_cpr.c b/dpdk/drivers/net/bnxt/bnxt_cpr.c +index a43b22a8f8..e1dcf3ac2f 100644 +--- a/dpdk/drivers/net/bnxt/bnxt_cpr.c ++++ b/dpdk/drivers/net/bnxt/bnxt_cpr.c +@@ -107,6 +107,26 @@ static void bnxt_handle_event_error_report(struct bnxt *bp, + } + } + ++void bnxt_handle_vf_cfg_change(void *arg) ++{ ++ struct bnxt *bp = arg; ++ struct rte_eth_dev *eth_dev = bp->eth_dev; ++ int rc; ++ ++ /* Free and recreate filters with default VLAN */ ++ if (eth_dev->data->dev_started) { ++ rc = bnxt_dev_stop_op(eth_dev); ++ if (rc != 0) { ++ PMD_DRV_LOG(ERR, "Failed to stop Port:%u\n", eth_dev->data->port_id); ++ return; ++ } ++ ++ rc = bnxt_dev_start_op(eth_dev); ++ if (rc != 0) ++ PMD_DRV_LOG(ERR, "Failed to start Port:%u\n", eth_dev->data->port_id); ++ } ++} + - add_flow_pattern(patterns, RTE_FLOW_ITEM_TYPE_ETH, spec, mask, NULL); - } + /* + * Async event handling + */ +@@ -138,8 +158,11 @@ void bnxt_handle_async_event(struct bnxt *bp, + PMD_DRV_LOG(INFO, "Async event: PF driver unloaded\n"); + break; + case HWRM_ASYNC_EVENT_CMPL_EVENT_ID_VF_CFG_CHANGE: +- PMD_DRV_LOG(INFO, "Async event: VF config changed\n"); ++ PMD_DRV_LOG(INFO, "Port %u: VF config change async event\n", port_id); ++ PMD_DRV_LOG(INFO, "event: data1 %#x data2 %#x\n", data1, data2); + bnxt_hwrm_func_qcfg(bp, NULL); ++ if (BNXT_VF(bp)) ++ rte_eal_alarm_set(1, bnxt_handle_vf_cfg_change, (void *)bp); + break; + case HWRM_ASYNC_EVENT_CMPL_EVENT_ID_PORT_CONN_NOT_ALLOWED: + PMD_DRV_LOG(INFO, "Port conn async event\n"); +diff --git a/dpdk/drivers/net/bnxt/bnxt_ethdev.c b/dpdk/drivers/net/bnxt/bnxt_ethdev.c +index f79f33ab4e..517e4b3898 100644 +--- a/dpdk/drivers/net/bnxt/bnxt_ethdev.c ++++ b/dpdk/drivers/net/bnxt/bnxt_ethdev.c +@@ -177,6 +177,7 @@ static int bnxt_restore_vlan_filters(struct bnxt *bp); + static void bnxt_dev_recover(void *arg); + static void bnxt_free_error_recovery_info(struct bnxt *bp); + static void bnxt_free_rep_info(struct bnxt *bp); ++static int bnxt_check_fw_ready(struct bnxt *bp); + + int is_bnxt_in_error(struct bnxt *bp) + { +@@ -368,7 +369,7 @@ static int bnxt_alloc_mem(struct bnxt *bp, bool reconfig) + if (rc) + goto alloc_mem_err; -@@ -1427,8 +1438,14 @@ parse_flow_match(struct netdev *netdev, - spec->tci = match->flow.vlans[0].tci & ~htons(VLAN_CFI); - mask->tci = match->wc.masks.vlans[0].tci & ~htons(VLAN_CFI); +- rc = bnxt_alloc_vnic_attributes(bp); ++ rc = bnxt_alloc_vnic_attributes(bp, reconfig); + if (rc) + goto alloc_mem_err; -- /* Match any protocols. */ -- mask->inner_type = 0; -+ if (eth_spec && eth_mask) { -+ eth_spec->has_vlan = 1; -+ eth_mask->has_vlan = 1; -+ spec->inner_type = eth_spec->type; -+ mask->inner_type = eth_mask->type; -+ eth_spec->type = match->flow.vlans[0].tpid; -+ eth_mask->type = match->wc.masks.vlans[0].tpid; -+ } +@@ -659,6 +660,19 @@ static int bnxt_init_ctx_mem(struct bnxt *bp) + return rc; + } - add_flow_pattern(patterns, RTE_FLOW_ITEM_TYPE_VLAN, spec, mask, NULL); - } -diff --git a/lib/netdev-offload-tc.c b/lib/netdev-offload-tc.c -index 9845e8d3fe..262faf3c62 100644 ---- a/lib/netdev-offload-tc.c -+++ b/lib/netdev-offload-tc.c -@@ -417,11 +417,11 @@ delete_chains_from_netdev(struct netdev *netdev, struct tcf_id *id) - static int - netdev_tc_flow_flush(struct netdev *netdev) ++static inline bool bnxt_force_link_config(struct bnxt *bp) ++{ ++ uint16_t subsystem_device_id = bp->pdev->id.subsystem_device_id; ++ ++ switch (subsystem_device_id) { ++ case BROADCOM_DEV_957508_N2100: ++ case BROADCOM_DEV_957414_N225: ++ return true; ++ default: ++ return false; ++ } ++} ++ + static int bnxt_update_phy_setting(struct bnxt *bp) { -- struct ufid_tc_data *data, *next; -+ struct ufid_tc_data *data; - int err; + struct rte_eth_link new; +@@ -671,11 +685,12 @@ static int bnxt_update_phy_setting(struct bnxt *bp) + } - ovs_mutex_lock(&ufid_lock); -- HMAP_FOR_EACH_SAFE (data, next, tc_to_ufid_node, &tc_to_ufid) { -+ HMAP_FOR_EACH_SAFE (data, tc_to_ufid_node, &tc_to_ufid) { - if (data->netdev != netdev) { - continue; - } -@@ -481,10 +481,10 @@ netdev_tc_flow_dump_destroy(struct netdev_flow_dump *dump) + /* +- * On BCM957508-N2100 adapters, FW will not allow any user other +- * than BMC to shutdown the port. bnxt_get_hwrm_link_config() call +- * always returns link up. Force phy update always in that case. ++ * Device is not obliged link down in certain scenarios, even ++ * when forced. When FW does not allow any user other than BMC ++ * to shutdown the port, bnxt_get_hwrm_link_config() call always ++ * returns link up. Force phy update always in that case. + */ +- if (!new.link_status || IS_BNXT_DEV_957508_N2100(bp)) { ++ if (!new.link_status || bnxt_force_link_config(bp)) { + rc = bnxt_set_hwrm_link_config(bp, true); + if (rc) { + PMD_DRV_LOG(ERR, "Failed to update PHY settings\n"); +@@ -708,7 +723,7 @@ static int bnxt_alloc_prev_ring_stats(struct bnxt *bp) + sizeof(struct bnxt_ring_stats) * + bp->tx_cp_nr_rings, + 0); +- if (bp->prev_tx_ring_stats == NULL) ++ if (bp->tx_cp_nr_rings > 0 && bp->prev_tx_ring_stats == NULL) + goto error; + + return 0; +@@ -786,17 +801,11 @@ static int bnxt_start_nic(struct bnxt *bp) + } + } - static void - parse_flower_rewrite_to_netlink_action(struct ofpbuf *buf, -- struct tc_flower *flower) -+ struct tc_action *action) +- /* default vnic 0 */ +- rc = bnxt_setup_one_vnic(bp, 0); +- if (rc) +- goto err_out; + /* VNIC configuration */ +- if (BNXT_RFS_NEEDS_VNIC(bp)) { +- for (i = 1; i < bp->nr_vnics; i++) { +- rc = bnxt_setup_one_vnic(bp, i); +- if (rc) +- goto err_out; +- } ++ for (i = 0; i < bp->nr_vnics; i++) { ++ rc = bnxt_setup_one_vnic(bp, i); ++ if (rc) ++ goto err_out; + } + + for (j = 0; j < bp->tx_nr_rings; j++) { +@@ -894,6 +903,7 @@ static int bnxt_shutdown_nic(struct bnxt *bp) + + uint32_t bnxt_get_speed_capabilities(struct bnxt *bp) { -- char *mask = (char *) &flower->rewrite.mask; -- char *data = (char *) &flower->rewrite.key; -+ char *mask = (char *) &action->rewrite.mask; -+ char *data = (char *) &action->rewrite.key; ++ uint32_t pam4_link_speed = 0; + uint32_t link_speed = 0; + uint32_t speed_capa = 0; + +@@ -903,8 +913,8 @@ uint32_t bnxt_get_speed_capabilities(struct bnxt *bp) + link_speed = bp->link_info->support_speeds; + + /* If PAM4 is configured, use PAM4 supported speed */ +- if (link_speed == 0 && bp->link_info->support_pam4_speeds > 0) +- link_speed = bp->link_info->support_pam4_speeds; ++ if (bp->link_info->support_pam4_speeds > 0) ++ pam4_link_speed = bp->link_info->support_pam4_speeds; + + if (link_speed & HWRM_PORT_PHY_QCFG_OUTPUT_LINK_SPEED_100MB) + speed_capa |= RTE_ETH_LINK_SPEED_100M; +@@ -926,11 +936,11 @@ uint32_t bnxt_get_speed_capabilities(struct bnxt *bp) + speed_capa |= RTE_ETH_LINK_SPEED_50G; + if (link_speed & HWRM_PORT_PHY_QCFG_OUTPUT_SUPPORT_SPEEDS_100GB) + speed_capa |= RTE_ETH_LINK_SPEED_100G; +- if (link_speed & HWRM_PORT_PHY_QCFG_OUTPUT_SUPPORT_PAM4_SPEEDS_50G) ++ if (pam4_link_speed & HWRM_PORT_PHY_QCFG_OUTPUT_SUPPORT_PAM4_SPEEDS_50G) + speed_capa |= RTE_ETH_LINK_SPEED_50G; +- if (link_speed & HWRM_PORT_PHY_QCFG_OUTPUT_SUPPORT_PAM4_SPEEDS_100G) ++ if (pam4_link_speed & HWRM_PORT_PHY_QCFG_OUTPUT_SUPPORT_PAM4_SPEEDS_100G) + speed_capa |= RTE_ETH_LINK_SPEED_100G; +- if (link_speed & HWRM_PORT_PHY_QCFG_OUTPUT_SUPPORT_PAM4_SPEEDS_200G) ++ if (pam4_link_speed & HWRM_PORT_PHY_QCFG_OUTPUT_SUPPORT_PAM4_SPEEDS_200G) + speed_capa |= RTE_ETH_LINK_SPEED_200G; + + if (bp->link_info->auto_mode == +@@ -954,7 +964,7 @@ static int bnxt_dev_info_get_op(struct rte_eth_dev *eth_dev, + return rc; + + /* MAC Specifics */ +- dev_info->max_mac_addrs = bp->max_l2_ctx; ++ dev_info->max_mac_addrs = RTE_MIN(bp->max_l2_ctx, RTE_ETH_NUM_RECEIVE_MAC_ADDR); + dev_info->max_hash_mac_addrs = 0; + + /* PF/VF specifics */ +@@ -977,16 +987,10 @@ static int bnxt_dev_info_get_op(struct rte_eth_dev *eth_dev, + dev_info->min_rx_bufsize = 1; + dev_info->max_rx_pktlen = BNXT_MAX_PKT_LEN; + +- dev_info->rx_offload_capa = BNXT_DEV_RX_OFFLOAD_SUPPORT; +- if (bp->flags & BNXT_FLAG_PTP_SUPPORTED) +- dev_info->rx_offload_capa |= RTE_ETH_RX_OFFLOAD_TIMESTAMP; +- if (bp->vnic_cap_flags & BNXT_VNIC_CAP_VLAN_RX_STRIP) +- dev_info->rx_offload_capa |= RTE_ETH_RX_OFFLOAD_VLAN_STRIP; ++ dev_info->rx_offload_capa = bnxt_get_rx_port_offloads(bp); + dev_info->tx_queue_offload_capa = RTE_ETH_TX_OFFLOAD_MBUF_FAST_FREE; +- dev_info->tx_offload_capa = BNXT_DEV_TX_OFFLOAD_SUPPORT | ++ dev_info->tx_offload_capa = bnxt_get_tx_port_offloads(bp) | + dev_info->tx_queue_offload_capa; +- if (bp->fw_cap & BNXT_FW_CAP_VLAN_TX_INSERT) +- dev_info->tx_offload_capa |= RTE_ETH_TX_OFFLOAD_VLAN_INSERT; + dev_info->flow_type_rss_offloads = BNXT_ETH_RSS_SUPPORT; + + dev_info->speed_capa = bnxt_get_speed_capabilities(bp); +@@ -1067,6 +1071,7 @@ static int bnxt_dev_configure_op(struct rte_eth_dev *eth_dev) + { + struct bnxt *bp = eth_dev->data->dev_private; + uint64_t rx_offloads = eth_dev->data->dev_conf.rxmode.offloads; ++ struct rte_eth_rss_conf *rss_conf = ð_dev->data->dev_conf.rx_adv_conf.rss_conf; + int rc; + + bp->rx_queues = (void *)eth_dev->data->rx_queues; +@@ -1141,6 +1146,17 @@ static int bnxt_dev_configure_op(struct rte_eth_dev *eth_dev) + rx_offloads |= RTE_ETH_RX_OFFLOAD_RSS_HASH; + eth_dev->data->dev_conf.rxmode.offloads = rx_offloads; + ++ /* application provides the hash key to program */ ++ if (rss_conf->rss_key != NULL) { ++ if (rss_conf->rss_key_len != HW_HASH_KEY_SIZE) ++ PMD_DRV_LOG(WARNING, "port %u RSS key len must be %d bytes long", ++ eth_dev->data->port_id, HW_HASH_KEY_SIZE); ++ else ++ memcpy(bp->rss_conf.rss_key, rss_conf->rss_key, HW_HASH_KEY_SIZE); ++ } ++ bp->rss_conf.rss_key_len = HW_HASH_KEY_SIZE; ++ bp->rss_conf.rss_hf = rss_conf->rss_hf; ++ + bnxt_mtu_set_op(eth_dev, eth_dev->data->mtu); + + return 0; +@@ -1336,6 +1352,11 @@ static int bnxt_handle_if_change_status(struct bnxt *bp) + + /* clear fatal flag so that re-init happens */ + bp->flags &= ~BNXT_FLAG_FATAL_ERROR; ++ ++ rc = bnxt_check_fw_ready(bp); ++ if (rc) ++ return rc; ++ + rc = bnxt_init_resources(bp, true); + + bp->flags &= ~BNXT_FLAG_IF_CHANGE_HOT_FW_RESET_DONE; +@@ -1525,7 +1546,7 @@ static int bnxt_dev_stop(struct rte_eth_dev *eth_dev) + } - for (int type = 0; type < ARRAY_SIZE(set_flower_map); type++) { - char *put = NULL; -@@ -585,8 +585,10 @@ parse_tc_flower_to_stats(struct tc_flower *flower, - } + /* Unload the driver, release resources */ +-static int bnxt_dev_stop_op(struct rte_eth_dev *eth_dev) ++int bnxt_dev_stop_op(struct rte_eth_dev *eth_dev) + { + struct bnxt *bp = eth_dev->data->dev_private; - memset(stats, 0, sizeof *stats); -- stats->n_packets = get_32aligned_u64(&flower->stats.n_packets); -- stats->n_bytes = get_32aligned_u64(&flower->stats.n_bytes); -+ stats->n_packets = get_32aligned_u64(&flower->stats_sw.n_packets); -+ stats->n_packets += get_32aligned_u64(&flower->stats_hw.n_packets); -+ stats->n_bytes = get_32aligned_u64(&flower->stats_sw.n_bytes); -+ stats->n_bytes += get_32aligned_u64(&flower->stats_hw.n_bytes); - stats->used = flower->lastused; +@@ -1541,18 +1562,13 @@ static int bnxt_dev_stop_op(struct rte_eth_dev *eth_dev) + return bnxt_dev_stop(eth_dev); } -@@ -877,7 +879,7 @@ parse_tc_flower_to_match(struct tc_flower *flower, - } - break; - case TC_ACT_PEDIT: { -- parse_flower_rewrite_to_netlink_action(buf, flower); -+ parse_flower_rewrite_to_netlink_action(buf, action); - } - break; - case TC_ACT_ENCAP: { -@@ -1222,8 +1224,8 @@ parse_put_flow_set_masked_action(struct tc_flower *flower, - uint64_t set_stub[1024 / 8]; - struct ofpbuf set_buf = OFPBUF_STUB_INITIALIZER(set_stub); - char *set_data, *set_mask; -- char *key = (char *) &flower->rewrite.key; -- char *mask = (char *) &flower->rewrite.mask; -+ char *key = (char *) &action->rewrite.key; -+ char *mask = (char *) &action->rewrite.mask; - const struct nlattr *attr; - int i, j, type; - size_t size; -@@ -1265,14 +1267,6 @@ parse_put_flow_set_masked_action(struct tc_flower *flower, - } - } +-static int bnxt_dev_start_op(struct rte_eth_dev *eth_dev) ++int bnxt_dev_start_op(struct rte_eth_dev *eth_dev) + { + struct bnxt *bp = eth_dev->data->dev_private; + uint64_t rx_offloads = eth_dev->data->dev_conf.rxmode.offloads; + int vlan_mask = 0; + int rc, retry_cnt = BNXT_IF_CHANGE_RETRY_COUNT; + +- if (!eth_dev->data->nb_tx_queues || !eth_dev->data->nb_rx_queues) { +- PMD_DRV_LOG(ERR, "Queues are not configured yet!\n"); +- return -EINVAL; +- } +- + if (bp->rx_cp_nr_rings > RTE_ETHDEV_QUEUE_STAT_CNTRS) + PMD_DRV_LOG(ERR, + "RxQ cnt %d > RTE_ETHDEV_QUEUE_STAT_CNTRS %d\n", +@@ -1673,6 +1689,7 @@ static int bnxt_dev_close_op(struct rte_eth_dev *eth_dev) + rte_eal_alarm_cancel(bnxt_dev_reset_and_resume, (void *)bp); + rte_eal_alarm_cancel(bnxt_dev_recover, (void *)bp); + bnxt_cancel_fc_thread(bp); ++ rte_eal_alarm_cancel(bnxt_handle_vf_cfg_change, (void *)bp); + + if (eth_dev->data->dev_started) + ret = bnxt_dev_stop(eth_dev); +@@ -1812,6 +1829,14 @@ int bnxt_link_update_op(struct rte_eth_dev *eth_dev, int wait_to_complete) + if (bp->link_info == NULL) + goto out; + ++ /* Only single function PF can bring the phy down. ++ * In certain scenarios, device is not obliged link down even when forced. ++ * When port is stopped, report link down in those cases. ++ */ ++ if (!eth_dev->data->dev_started && ++ (!BNXT_SINGLE_PF(bp) || bnxt_force_link_config(bp))) ++ goto out; ++ + do { + /* Retrieve link info from hardware */ + rc = bnxt_get_hwrm_link_config(bp, &new); +@@ -1829,12 +1854,6 @@ int bnxt_link_update_op(struct rte_eth_dev *eth_dev, int wait_to_complete) + rte_delay_ms(BNXT_LINK_WAIT_INTERVAL); + } while (cnt--); + +- /* Only single function PF can bring phy down. +- * When port is stopped, report link down for VF/MH/NPAR functions. +- */ +- if (!BNXT_SINGLE_PF(bp) && !eth_dev->data->dev_started) +- memset(&new, 0, sizeof(new)); +- + out: + /* Timed out or success */ + if (new.link_status != eth_dev->data->dev_link.link_status || +@@ -2125,11 +2144,6 @@ static int bnxt_rss_hash_update_op(struct rte_eth_dev *eth_dev, + return -EINVAL; + } -- if (!is_all_zeros(&flower->rewrite, sizeof flower->rewrite)) { -- if (flower->rewrite.rewrite == false) { -- flower->rewrite.rewrite = true; -- action->type = TC_ACT_PEDIT; -- flower->action_count++; -- } -- } +- bp->flags |= BNXT_FLAG_UPDATE_HASH; +- memcpy(ð_dev->data->dev_conf.rx_adv_conf.rss_conf, +- rss_conf, +- sizeof(*rss_conf)); - - if (hasmask && !is_all_zeros(set_mask, size)) { - VLOG_DBG_RL(&rl, "unsupported sub attribute of set action type %d", - type); -@@ -1281,6 +1275,8 @@ parse_put_flow_set_masked_action(struct tc_flower *flower, - } + /* Update the default RSS VNIC(s) */ + vnic = BNXT_GET_DEFAULT_VNIC(bp); + vnic->hash_type = bnxt_rte_to_hwrm_hash_types(rss_conf->rss_hf); +@@ -2137,6 +2151,9 @@ static int bnxt_rss_hash_update_op(struct rte_eth_dev *eth_dev, + bnxt_rte_to_hwrm_hash_level(bp, rss_conf->rss_hf, + RTE_ETH_RSS_LEVEL(rss_conf->rss_hf)); + ++ /* Cache the hash function */ ++ bp->rss_conf.rss_hf = rss_conf->rss_hf; ++ + /* + * If hashkey is not specified, use the previously configured + * hashkey +@@ -2152,6 +2169,9 @@ static int bnxt_rss_hash_update_op(struct rte_eth_dev *eth_dev, + } + memcpy(vnic->rss_hash_key, rss_conf->rss_key, rss_conf->rss_key_len); - ofpbuf_uninit(&set_buf); -+ action->type = TC_ACT_PEDIT; -+ flower->action_count++; - return 0; ++ /* Cache the hash key */ ++ memcpy(bp->rss_conf.rss_key, rss_conf->rss_key, HW_HASH_KEY_SIZE); ++ + rss_config: + rc = bnxt_hwrm_vnic_rss_cfg(bp, vnic); + return rc; +@@ -2831,9 +2851,8 @@ bnxt_dev_set_mc_addr_list_op(struct rte_eth_dev *eth_dev, + uint32_t nb_mc_addr) + { + struct bnxt *bp = eth_dev->data->dev_private; +- char *mc_addr_list = (char *)mc_addr_set; + struct bnxt_vnic_info *vnic; +- uint32_t off = 0, i = 0; ++ uint32_t i = 0; + int rc; + + rc = is_bnxt_in_error(bp); +@@ -2842,6 +2861,8 @@ bnxt_dev_set_mc_addr_list_op(struct rte_eth_dev *eth_dev, + + vnic = BNXT_GET_DEFAULT_VNIC(bp); + ++ bp->nb_mc_addr = nb_mc_addr; ++ + if (nb_mc_addr > BNXT_MAX_MC_ADDRS) { + vnic->flags |= BNXT_VNIC_INFO_ALLMULTI; + goto allmulti; +@@ -2849,14 +2870,10 @@ bnxt_dev_set_mc_addr_list_op(struct rte_eth_dev *eth_dev, + + /* TODO Check for Duplicate mcast addresses */ + vnic->flags &= ~BNXT_VNIC_INFO_ALLMULTI; +- for (i = 0; i < nb_mc_addr; i++) { +- memcpy(vnic->mc_list + off, &mc_addr_list[i], +- RTE_ETHER_ADDR_LEN); +- off += RTE_ETHER_ADDR_LEN; +- } ++ for (i = 0; i < nb_mc_addr; i++) ++ rte_ether_addr_copy(&mc_addr_set[i], &bp->mcast_addr_list[i]); + +- vnic->mc_addr_cnt = i; +- if (vnic->mc_addr_cnt) ++ if (bp->nb_mc_addr) + vnic->flags |= BNXT_VNIC_INFO_MCAST; + else + vnic->flags &= ~BNXT_VNIC_INFO_MCAST; +@@ -3003,9 +3020,7 @@ bnxt_tx_burst_mode_get(struct rte_eth_dev *dev, __rte_unused uint16_t queue_id, + + int bnxt_mtu_set_op(struct rte_eth_dev *eth_dev, uint16_t new_mtu) + { +- uint32_t overhead = BNXT_MAX_PKT_LEN - BNXT_MAX_MTU; + struct bnxt *bp = eth_dev->data->dev_private; +- uint32_t new_pkt_size; + uint32_t rc; + uint32_t i; + +@@ -3013,35 +3028,25 @@ int bnxt_mtu_set_op(struct rte_eth_dev *eth_dev, uint16_t new_mtu) + if (rc) + return rc; + ++ /* Return if port is active */ ++ if (eth_dev->data->dev_started) { ++ PMD_DRV_LOG(ERR, "Stop port before changing MTU\n"); ++ return -EPERM; ++ } ++ + /* Exit if receive queues are not configured yet */ + if (!eth_dev->data->nb_rx_queues) +- return rc; +- +- new_pkt_size = new_mtu + overhead; ++ return -ENOTSUP; + +- /* +- * Disallow any MTU change that would require scattered receive support +- * if it is not already enabled. +- */ +- if (eth_dev->data->dev_started && +- !eth_dev->data->scattered_rx && +- (new_pkt_size > +- eth_dev->data->min_rx_buf_size - RTE_PKTMBUF_HEADROOM)) { +- PMD_DRV_LOG(ERR, +- "MTU change would require scattered rx support. "); +- PMD_DRV_LOG(ERR, "Stop port before changing MTU.\n"); +- return -EINVAL; +- } ++ /* Is there a change in mtu setting? */ ++ if (eth_dev->data->mtu == new_mtu) ++ return 0; + + if (new_mtu > RTE_ETHER_MTU) + bp->flags |= BNXT_FLAG_JUMBO; + else + bp->flags &= ~BNXT_FLAG_JUMBO; + +- /* Is there a change in mtu setting? */ +- if (eth_dev->data->mtu == new_mtu) +- return rc; +- + for (i = 0; i < bp->nr_vnics; i++) { + struct bnxt_vnic_info *vnic = &bp->vnic_info[i]; + uint16_t size = 0; +@@ -4264,6 +4269,18 @@ static int bnxt_restore_mac_filters(struct bnxt *bp) + return 0; } -@@ -1541,6 +1537,12 @@ parse_match_ct_state_to_flower(struct tc_flower *flower, struct match *match) - flower->key.ct_state &= ~(TCA_FLOWER_KEY_CT_FLAGS_NEW); - flower->mask.ct_state &= ~(TCA_FLOWER_KEY_CT_FLAGS_NEW); - } ++static int bnxt_restore_mcast_mac_filters(struct bnxt *bp) ++{ ++ int ret = 0; + -+ if (flower->key.ct_state && -+ !(flower->key.ct_state & TCA_FLOWER_KEY_CT_FLAGS_TRACKED)) { -+ flower->key.ct_state |= TCA_FLOWER_KEY_CT_FLAGS_TRACKED; -+ flower->mask.ct_state |= TCA_FLOWER_KEY_CT_FLAGS_TRACKED; -+ } - } ++ ret = bnxt_dev_set_mc_addr_list_op(bp->eth_dev, bp->mcast_addr_list, ++ bp->nb_mc_addr); ++ if (ret) ++ PMD_DRV_LOG(ERR, "Failed to restore multicast MAC addreeses\n"); ++ ++ return ret; ++} ++ + static int bnxt_restore_filters(struct bnxt *bp) + { + struct rte_eth_dev *dev = bp->eth_dev; +@@ -4284,14 +4301,21 @@ static int bnxt_restore_filters(struct bnxt *bp) + if (ret) + return ret; + ++ /* if vlans are already programmed, this can fail with -EEXIST */ + ret = bnxt_restore_vlan_filters(bp); +- /* TODO restore other filters as well */ ++ if (ret && ret != -EEXIST) ++ return ret; ++ ++ ret = bnxt_restore_mcast_mac_filters(bp); ++ if (ret) ++ return ret; ++ + return ret; + } - if (mask->ct_zone) { -@@ -1638,7 +1640,7 @@ netdev_tc_flow_put(struct netdev *netdev, struct match *match, + static int bnxt_check_fw_ready(struct bnxt *bp) + { +- int timeout = bp->fw_reset_max_msecs; ++ int timeout = bp->fw_reset_max_msecs ? : BNXT_MAX_FW_RESET_TIMEOUT; + int rc = 0; - if (mask->vlans[0].tpid && eth_type_vlan(key->vlans[0].tpid)) { - flower.key.encap_eth_type[0] = flower.key.eth_type; -- flower.mask.encap_eth_type[0] = flower.mask.eth_type; -+ flower.mask.encap_eth_type[0] = CONSTANT_HTONS(0xffff); - flower.key.eth_type = key->vlans[0].tpid; - flower.mask.eth_type = mask->vlans[0].tpid; - } -@@ -1841,7 +1843,25 @@ netdev_tc_flow_put(struct netdev *netdev, struct match *match, - VLOG_DBG_RL(&rl, "Can't find netdev for output port %d", port); - return ENODEV; - } -+ -+ if (!netdev_flow_api_equals(netdev, outdev)) { -+ VLOG_DBG_RL(&rl, -+ "Flow API provider mismatch between ingress (%s) " -+ "and egress (%s) ports", -+ netdev_get_name(netdev), netdev_get_name(outdev)); -+ netdev_close(outdev); -+ return EOPNOTSUPP; -+ } -+ - action->out.ifindex_out = netdev_get_ifindex(outdev); -+ if (action->out.ifindex_out < 0) { -+ VLOG_DBG_RL(&rl, -+ "Can't find ifindex for output port %s, error %d", -+ netdev_get_name(outdev), action->out.ifindex_out); -+ netdev_close(outdev); -+ return -action->out.ifindex_out; -+ } -+ - action->out.ingress = is_internal_port(netdev_get_type(outdev)); - action->type = TC_ACT_OUTPUT; - flower.action_count++; -@@ -2015,9 +2035,7 @@ netdev_tc_flow_del(struct netdev *netdev OVS_UNUSED, - if (stats) { - memset(stats, 0, sizeof *stats); - if (!tc_get_flower(&id, &flower)) { -- stats->n_packets = get_32aligned_u64(&flower.stats.n_packets); -- stats->n_bytes = get_32aligned_u64(&flower.stats.n_bytes); -- stats->used = flower.lastused; -+ parse_tc_flower_to_stats(&flower, stats); - } - } + do { +@@ -4345,16 +4369,16 @@ static void bnxt_dev_recover(void *arg) + goto err_start; + } -diff --git a/lib/odp-util.c b/lib/odp-util.c -index 9a705cffa3..2d2a6893c6 100644 ---- a/lib/odp-util.c -+++ b/lib/odp-util.c -@@ -3429,16 +3429,16 @@ format_eth(struct ds *ds, const char *name, const struct eth_addr key, ++ rc = bnxt_restore_filters(bp); ++ if (rc) ++ goto err_start; ++ + rte_eth_fp_ops[bp->eth_dev->data->port_id].rx_pkt_burst = + bp->eth_dev->rx_pkt_burst; + rte_eth_fp_ops[bp->eth_dev->data->port_id].tx_pkt_burst = + bp->eth_dev->tx_pkt_burst; + rte_mb(); - static void - format_be64(struct ds *ds, const char *name, ovs_be64 key, -- const ovs_be64 *mask, bool verbose) -+ const ovs_32aligned_be64 *mask_, bool verbose) +- rc = bnxt_restore_filters(bp); +- if (rc) +- goto err_start; +- + PMD_DRV_LOG(INFO, "Port: %u Recovered from FW reset\n", + bp->eth_dev->data->port_id); + pthread_mutex_unlock(&bp->err_recovery_lock); +@@ -4985,11 +5009,15 @@ static int bnxt_alloc_stats_mem(struct bnxt *bp) + static int bnxt_setup_mac_addr(struct rte_eth_dev *eth_dev) { -- bool mask_empty = mask && !*mask; -+ ovs_be64 mask = mask_ ? get_32aligned_be64(mask_) : htonll(0); + struct bnxt *bp = eth_dev->data->dev_private; ++ size_t max_mac_addr = RTE_MIN(bp->max_l2_ctx, RTE_ETH_NUM_RECEIVE_MAC_ADDR); + int rc = 0; + ++ if (bp->max_l2_ctx > RTE_ETH_NUM_RECEIVE_MAC_ADDR) ++ PMD_DRV_LOG(INFO, "Max number of MAC addrs supported is %d, but will be limited to %d\n", ++ bp->max_l2_ctx, RTE_ETH_NUM_RECEIVE_MAC_ADDR); ++ + eth_dev->data->mac_addrs = rte_zmalloc("bnxt_mac_addr_tbl", +- RTE_ETHER_ADDR_LEN * +- bp->max_l2_ctx, ++ RTE_ETHER_ADDR_LEN * max_mac_addr, + 0); + if (eth_dev->data->mac_addrs == NULL) { + PMD_DRV_LOG(ERR, "Failed to alloc MAC addr tbl\n"); +@@ -5016,6 +5044,23 @@ static int bnxt_setup_mac_addr(struct rte_eth_dev *eth_dev) + /* Copy the permanent MAC from the FUNC_QCAPS response */ + memcpy(ð_dev->data->mac_addrs[0], bp->mac_addr, RTE_ETHER_ADDR_LEN); + ++ /* ++ * Allocate memory to hold multicast mac addresses added. ++ * Used to restore them during reset recovery ++ */ ++ bp->mcast_addr_list = rte_zmalloc("bnxt_mcast_addr_tbl", ++ sizeof(struct rte_ether_addr) * ++ BNXT_MAX_MC_ADDRS, 0); ++ if (bp->mcast_addr_list == NULL) { ++ PMD_DRV_LOG(ERR, "Failed to allocate multicast addr table\n"); ++ return -ENOMEM; ++ } ++ bp->mc_list_dma_addr = rte_malloc_virt2iova(bp->mcast_addr_list); ++ if (bp->mc_list_dma_addr == RTE_BAD_IOVA) { ++ PMD_DRV_LOG(ERR, "Fail to map mcast_addr_list to physical memory\n"); ++ return -ENOMEM; ++ } ++ + return rc; + } -- if (verbose || !mask_empty) { -- bool mask_full = !mask || *mask == OVS_BE64_MAX; -+ if (verbose || mask) { -+ bool mask_full = !mask_ || mask == OVS_BE64_MAX; +@@ -5178,10 +5223,6 @@ static int bnxt_get_config(struct bnxt *bp) + if (rc) + return rc; - ds_put_format(ds, "%s=0x%"PRIx64, name, ntohll(key)); - if (!mask_full) { /* Partially masked. */ -- ds_put_format(ds, "/%#"PRIx64, ntohll(*mask)); -+ ds_put_format(ds, "/%#"PRIx64, ntohll(mask)); - } - ds_put_char(ds, ','); - } -@@ -4630,6 +4630,11 @@ odp_flow_format(const struct nlattr *key, size_t key_len, - ds_put_char(ds, ','); - } - ds_put_cstr(ds, "eth()"); -+ } else if (attr_type == OVS_KEY_ATTR_PACKET_TYPE && is_wildcard) { -+ /* See the above help text, however in the case where the -+ * packet type is not shown, we still need to display the -+ * eth() header if the packets type is wildcarded. */ -+ has_packet_type_key = false; - } - ofpbuf_clear(&ofp); - } -@@ -7161,11 +7166,6 @@ parse_l2_5_onward(const struct nlattr *attrs[OVS_KEY_ATTR_MAX + 1], - } - } - } -- } else if (src_flow->nw_proto == IPPROTO_IGMP -- && src_flow->dl_type == htons(ETH_TYPE_IP)) { -- /* OVS userspace parses the IGMP type, code, and group, but its -- * datapaths do not, so there is always missing information. */ -- return ODP_FIT_TOO_LITTLE; - } - if (is_mask && expected_bit != OVS_KEY_ATTR_UNSPEC) { - if ((flow->tp_src || flow->tp_dst) && flow->nw_proto != 0xff) { -@@ -7188,7 +7188,8 @@ parse_8021q_onward(const struct nlattr *attrs[OVS_KEY_ATTR_MAX + 1], - uint64_t present_attrs, int out_of_range_attr, - uint64_t expected_attrs, struct flow *flow, - const struct nlattr *key, size_t key_len, -- const struct flow *src_flow, char **errorp) -+ const struct flow *src_flow, char **errorp, -+ bool ignore_vlan_limit) - { - static struct vlog_rate_limit rl = VLOG_RATE_LIMIT_INIT(1, 5); - bool is_mask = src_flow != flow; -@@ -7196,9 +7197,11 @@ parse_8021q_onward(const struct nlattr *attrs[OVS_KEY_ATTR_MAX + 1], - const struct nlattr *encap; - enum odp_key_fitness encap_fitness; - enum odp_key_fitness fitness = ODP_FIT_ERROR; -+ int vlan_limit; - int encaps = 0; +- rc = bnxt_hwrm_cfa_adv_flow_mgmt_qcaps(bp); +- if (rc) +- return rc; +- + bnxt_hwrm_port_mac_qcfg(bp); -- while (encaps < flow_vlan_limit && -+ vlan_limit = ignore_vlan_limit ? FLOW_MAX_VLAN_HEADERS : flow_vlan_limit; -+ while (encaps < vlan_limit && - (is_mask - ? (src_flow->vlans[encaps].tci & htons(VLAN_CFI)) != 0 - : eth_type_vlan(flow->dl_type))) { -@@ -7259,6 +7262,14 @@ parse_8021q_onward(const struct nlattr *attrs[OVS_KEY_ATTR_MAX + 1], - } - expected_attrs = 0; + bnxt_hwrm_parent_pf_qcfg(bp); +@@ -5229,6 +5270,25 @@ bnxt_init_locks(struct bnxt *bp) + return err; + } -+ /* For OVS to be backward compatible with newer datapath -+ * implementations, we should ignore out of range attributes. */ -+ if (out_of_range_attr) { -+ VLOG_DBG("Flow key decode found unknown OVS_KEY_ATTR, %d", -+ out_of_range_attr); -+ out_of_range_attr = 0; -+ } ++/* This should be called after we have queried trusted VF cap */ ++static int bnxt_alloc_switch_domain(struct bnxt *bp) ++{ ++ int rc = 0; ++ ++ if (BNXT_PF(bp) || BNXT_VF_IS_TRUSTED(bp)) { ++ rc = rte_eth_switch_domain_alloc(&bp->switch_domain_id); ++ if (rc) ++ PMD_DRV_LOG(ERR, ++ "Failed to alloc switch domain: %d\n", rc); ++ else ++ PMD_DRV_LOG(INFO, ++ "Switch domain allocated %d\n", ++ bp->switch_domain_id); ++ } ++ ++ return rc; ++} + - if (!parse_ethertype(attrs, present_attrs, &expected_attrs, - flow, src_flow, errorp)) { - return ODP_FIT_ERROR; -@@ -7281,7 +7292,7 @@ parse_8021q_onward(const struct nlattr *attrs[OVS_KEY_ATTR_MAX + 1], - static enum odp_key_fitness - odp_flow_key_to_flow__(const struct nlattr *key, size_t key_len, - struct flow *flow, const struct flow *src_flow, -- char **errorp) -+ char **errorp, bool ignore_vlan_limit) + static int bnxt_init_resources(struct bnxt *bp, bool reconfig_dev) { - /* New "struct flow" fields that are visible to the datapath (including all - * data fields) should be translated from equivalent datapath flow fields -@@ -7308,6 +7319,14 @@ odp_flow_key_to_flow__(const struct nlattr *key, size_t key_len, - } - expected_attrs = 0; + int rc = 0; +@@ -5237,6 +5297,10 @@ static int bnxt_init_resources(struct bnxt *bp, bool reconfig_dev) + if (rc) + return rc; + ++ rc = bnxt_alloc_switch_domain(bp); ++ if (rc) ++ return rc; ++ + if (!reconfig_dev) { + rc = bnxt_setup_mac_addr(bp->eth_dev); + if (rc) +@@ -5272,6 +5336,16 @@ static int bnxt_init_resources(struct bnxt *bp, bool reconfig_dev) + } + } -+ /* For OVS to be backward compatible with newer datapath implementations, -+ * we should ignore out of range attributes. */ -+ if (out_of_range_attr) { -+ VLOG_DBG("Flow key decode found unknown OVS_KEY_ATTR, %d", -+ out_of_range_attr); -+ out_of_range_attr = 0; -+ } -+ - /* Metadata. */ - if (present_attrs & (UINT64_C(1) << OVS_KEY_ATTR_RECIRC_ID)) { - flow->recirc_id = nl_attr_get_u32(attrs[OVS_KEY_ATTR_RECIRC_ID]); -@@ -7431,7 +7450,7 @@ odp_flow_key_to_flow__(const struct nlattr *key, size_t key_len, - : eth_type_vlan(src_flow->dl_type)) { - fitness = parse_8021q_onward(attrs, present_attrs, out_of_range_attr, - expected_attrs, flow, key, key_len, -- src_flow, errorp); -+ src_flow, errorp, ignore_vlan_limit); - } else { - if (is_mask) { - /* A missing VLAN mask means exact match on vlan_tci 0 (== no -@@ -7497,7 +7516,7 @@ enum odp_key_fitness - odp_flow_key_to_flow(const struct nlattr *key, size_t key_len, - struct flow *flow, char **errorp) - { -- return odp_flow_key_to_flow__(key, key_len, flow, flow, errorp); -+ return odp_flow_key_to_flow__(key, key_len, flow, flow, errorp, false); ++ if (!reconfig_dev) { ++ bp->rss_conf.rss_key = rte_zmalloc("bnxt_rss_key", ++ HW_HASH_KEY_SIZE, 0); ++ if (bp->rss_conf.rss_key == NULL) { ++ PMD_DRV_LOG(ERR, "port %u cannot allocate RSS hash key memory", ++ bp->eth_dev->data->port_id); ++ return -ENOMEM; ++ } ++ } ++ + rc = bnxt_alloc_mem(bp, reconfig_dev); + if (rc) + return rc; +@@ -5666,24 +5740,6 @@ bnxt_parse_dev_args(struct bnxt *bp, struct rte_devargs *devargs) + return ret; } - /* Converts the 'mask_key_len' bytes of OVS_KEY_ATTR_* attributes in 'mask_key' -@@ -7509,14 +7528,16 @@ odp_flow_key_to_flow(const struct nlattr *key, size_t key_len, - * If 'errorp' is nonnull, this function uses it for detailed error reports: if - * the return value is ODP_FIT_ERROR, it stores a malloc()'d error string in - * '*errorp', otherwise NULL. */ --enum odp_key_fitness --odp_flow_key_to_mask(const struct nlattr *mask_key, size_t mask_key_len, -- struct flow_wildcards *mask, const struct flow *src_flow, -- char **errorp) -+static enum odp_key_fitness -+odp_flow_key_to_mask__(const struct nlattr *mask_key, size_t mask_key_len, -+ struct flow_wildcards *mask, -+ const struct flow *src_flow, -+ char **errorp, bool ignore_vlan_limit) - { - if (mask_key_len) { - return odp_flow_key_to_flow__(mask_key, mask_key_len, -- &mask->masks, src_flow, errorp); -+ &mask->masks, src_flow, errorp, -+ ignore_vlan_limit); - } else { - if (errorp) { - *errorp = NULL; -@@ -7530,6 +7551,15 @@ odp_flow_key_to_mask(const struct nlattr *mask_key, size_t mask_key_len, - } +-static int bnxt_alloc_switch_domain(struct bnxt *bp) +-{ +- int rc = 0; +- +- if (BNXT_PF(bp) || BNXT_VF_IS_TRUSTED(bp)) { +- rc = rte_eth_switch_domain_alloc(&bp->switch_domain_id); +- if (rc) +- PMD_DRV_LOG(ERR, +- "Failed to alloc switch domain: %d\n", rc); +- else +- PMD_DRV_LOG(INFO, +- "Switch domain allocated %d\n", +- bp->switch_domain_id); +- } +- +- return rc; +-} +- + /* Allocate and initialize various fields in bnxt struct that + * need to be allocated/destroyed only once in the lifetime of the driver + */ +@@ -5760,10 +5816,6 @@ static int bnxt_drv_init(struct rte_eth_dev *eth_dev) + if (rc) + return rc; + +- rc = bnxt_alloc_switch_domain(bp); +- if (rc) +- return rc; +- + return rc; } -+enum odp_key_fitness -+odp_flow_key_to_mask(const struct nlattr *mask_key, size_t mask_key_len, -+ struct flow_wildcards *mask, -+ const struct flow *src_flow, char **errorp) -+{ -+ return odp_flow_key_to_mask__(mask_key, mask_key_len, mask, src_flow, -+ errorp, false); -+} -+ - /* Converts the netlink formated key/mask to match. - * Fails if odp_flow_key_from_key/mask and odp_flow_key_key/mask - * disagree on the acceptable form of flow */ -@@ -7540,12 +7570,15 @@ parse_key_and_mask_to_match(const struct nlattr *key, size_t key_len, - { - enum odp_key_fitness fitness; +@@ -5916,6 +5968,10 @@ bnxt_uninit_resources(struct bnxt *bp, bool reconfig_dev) + if (!reconfig_dev) { + bnxt_free_hwrm_resources(bp); + bnxt_free_error_recovery_info(bp); ++ rte_free(bp->mcast_addr_list); ++ bp->mcast_addr_list = NULL; ++ rte_free(bp->rss_conf.rss_key); ++ bp->rss_conf.rss_key = NULL; + } -- fitness = odp_flow_key_to_flow(key, key_len, &match->flow, NULL); -+ fitness = odp_flow_key_to_flow__(key, key_len, &match->flow, &match->flow, -+ NULL, true); - if (fitness) { -- /* This should not happen: it indicates that -- * odp_flow_key_from_flow() and odp_flow_key_to_flow() disagree on -- * the acceptable form of a flow. Log the problem as an error, -- * with enough details to enable debugging. */ -+ /* This will happen when the odp_flow_key_to_flow() function can't -+ * parse the netlink message to a match structure. It will return -+ * ODP_FIT_TOO_LITTLE if there is not enough information to parse the -+ * content successfully, ODP_FIT_TOO_MUCH if there is too much netlink -+ * data and we do not know how to safely ignore it, and ODP_FIT_ERROR -+ * in any other case. */ - static struct vlog_rate_limit rl = VLOG_RATE_LIMIT_INIT(1, 5); + bnxt_uninit_ctx_mem(bp); +@@ -6302,4 +6358,4 @@ bool is_bnxt_supported(struct rte_eth_dev *dev) + RTE_LOG_REGISTER_SUFFIX(bnxt_logtype_driver, driver, NOTICE); + RTE_PMD_REGISTER_PCI(net_bnxt, bnxt_rte_pmd); + RTE_PMD_REGISTER_PCI_TABLE(net_bnxt, bnxt_pci_id_map); +- ++RTE_PMD_REGISTER_KMOD_DEP(net_bnxt, "* igb_uio | uio_pci_generic | vfio-pci"); +diff --git a/dpdk/drivers/net/bnxt/bnxt_filter.c b/dpdk/drivers/net/bnxt/bnxt_filter.c +index 1d08e03b2f..b0c3bbd1b2 100644 +--- a/dpdk/drivers/net/bnxt/bnxt_filter.c ++++ b/dpdk/drivers/net/bnxt/bnxt_filter.c +@@ -99,6 +99,8 @@ void bnxt_free_all_filters(struct bnxt *bp) + bnxt_filter_info, next); + STAILQ_INSERT_TAIL(&bp->free_filter_list, + filter, next); ++ if (filter->vnic) ++ filter->vnic = NULL; + filter = temp_filter; + } + STAILQ_INIT(&vnic->filter); +diff --git a/dpdk/drivers/net/bnxt/bnxt_flow.c b/dpdk/drivers/net/bnxt/bnxt_flow.c +index d062be5525..8bdf2405f0 100644 +--- a/dpdk/drivers/net/bnxt/bnxt_flow.c ++++ b/dpdk/drivers/net/bnxt/bnxt_flow.c +@@ -1074,7 +1074,6 @@ bnxt_update_filter_flags_en(struct bnxt_filter_info *filter, + filter1, filter->fw_l2_filter_id, filter->l2_ref_cnt); + } - if (!VLOG_DROP_ERR(&rl)) { -@@ -7553,20 +7586,18 @@ parse_key_and_mask_to_match(const struct nlattr *key, size_t key_len, +-/* Valid actions supported along with RSS are count and mark. */ + static int + bnxt_validate_rss_action(const struct rte_flow_action actions[]) + { +@@ -1116,13 +1115,55 @@ bnxt_vnic_rss_cfg_update(struct bnxt *bp, + struct rte_flow_error *error) + { + const struct rte_flow_action_rss *rss; +- unsigned int rss_idx, i; ++ unsigned int rss_idx, i, j, fw_idx; + uint16_t hash_type; + uint64_t types; + int rc; + + rss = (const struct rte_flow_action_rss *)act->conf; + ++ /* must specify either all the Rx queues created by application or zero queues */ ++ if (rss->queue_num && vnic->rx_queue_cnt != rss->queue_num) { ++ rte_flow_error_set(error, ++ EINVAL, ++ RTE_FLOW_ERROR_TYPE_ACTION, ++ act, ++ "Incorrect RXQ count"); ++ rc = -rte_errno; ++ goto ret; ++ } ++ ++ /* Validate Rx queues */ ++ for (i = 0; i < rss->queue_num; i++) { ++ PMD_DRV_LOG(DEBUG, "RSS action Queue %d\n", rss->queue[i]); ++ ++ if (rss->queue[i] >= bp->rx_nr_rings || ++ !bp->rx_queues[rss->queue[i]]) { ++ rte_flow_error_set(error, ++ EINVAL, ++ RTE_FLOW_ERROR_TYPE_ACTION, ++ act, ++ "Invalid queue ID for RSS"); ++ rc = -rte_errno; ++ goto ret; ++ } ++ } ++ ++ /* Duplicate queue ids are not supported. */ ++ for (i = 0; i < rss->queue_num; i++) { ++ for (j = i + 1; j < rss->queue_num; j++) { ++ if (rss->queue[i] == rss->queue[j]) { ++ rte_flow_error_set(error, ++ EINVAL, ++ RTE_FLOW_ERROR_TYPE_ACTION, ++ act, ++ "Duplicate queue ID for RSS"); ++ rc = -rte_errno; ++ goto ret; ++ } ++ } ++ } ++ + /* Currently only Toeplitz hash is supported. */ + if (rss->func != RTE_ETH_HASH_FUNCTION_DEFAULT && + rss->func != RTE_ETH_HASH_FUNCTION_TOEPLITZ) { +@@ -1190,28 +1231,22 @@ bnxt_vnic_rss_cfg_update(struct bnxt *bp, + if (rss->queue_num == 0) + goto skip_rss_table; + +- /* Validate Rx queues */ +- for (i = 0; i < rss->queue_num; i++) { +- PMD_DRV_LOG(DEBUG, "RSS action Queue %d\n", rss->queue[i]); +- +- if (rss->queue[i] >= bp->rx_nr_rings || +- !bp->rx_queues[rss->queue[i]]) { +- rte_flow_error_set(error, +- EINVAL, +- RTE_FLOW_ERROR_TYPE_ACTION, +- act, +- "Invalid queue ID for RSS"); +- rc = -rte_errno; +- goto ret; +- } +- } +- + /* Prepare the indirection table */ +- for (rss_idx = 0; rss_idx < HW_HASH_INDEX_SIZE; rss_idx++) { ++ for (rss_idx = 0, fw_idx = 0; rss_idx < HW_HASH_INDEX_SIZE; ++ rss_idx++, fw_idx++) { ++ uint8_t *rxq_state = bp->eth_dev->data->rx_queue_state; + struct bnxt_rx_queue *rxq; + uint32_t idx; + +- idx = rss->queue[rss_idx % rss->queue_num]; ++ for (i = 0; i < bp->rx_cp_nr_rings; i++) { ++ idx = rss->queue[fw_idx % rss->queue_num]; ++ if (rxq_state[idx] != RTE_ETH_QUEUE_STATE_STOPPED) ++ break; ++ fw_idx++; ++ } ++ ++ if (i == bp->rx_cp_nr_rings) ++ return 0; + + if (BNXT_CHIP_P5(bp)) { + rxq = bp->rx_queues[idx]; +@@ -1293,13 +1328,6 @@ bnxt_validate_and_parse_flow(struct rte_eth_dev *dev, + } + PMD_DRV_LOG(DEBUG, "Queue index %d\n", act_q->index); + +- if (use_ntuple && !BNXT_RFS_NEEDS_VNIC(bp)) { +- filter->flags = +- HWRM_CFA_NTUPLE_FILTER_ALLOC_INPUT_FLAGS_DEST_RFS_RING_IDX; +- filter->dst_id = act_q->index; +- goto skip_vnic_alloc; +- } +- + vnic_id = attr->group; + if (!vnic_id) { + PMD_DRV_LOG(DEBUG, "Group id is 0\n"); +@@ -1364,7 +1392,7 @@ bnxt_validate_and_parse_flow(struct rte_eth_dev *dev, + PMD_DRV_LOG(DEBUG, + "Setting vnic ff_idx %d\n", vnic->ff_pool_idx); + filter->dst_id = vnic->fw_vnic_id; +-skip_vnic_alloc: ++ + /* For ntuple filter, create the L2 filter with default VNIC. + * The user specified redirect queue will be set while creating + * the ntuple filter in hardware. +@@ -2063,10 +2091,7 @@ bnxt_flow_create(struct rte_eth_dev *dev, + } + } - ds_init(&s); - odp_flow_format(key, key_len, NULL, 0, NULL, &s, true); -- VLOG_ERR("internal error parsing flow key %s", ds_cstr(&s)); -+ VLOG_ERR("internal error parsing flow key %s (%s)", -+ ds_cstr(&s), odp_key_fitness_to_string(fitness)); - ds_destroy(&s); - } +- if (BNXT_RFS_NEEDS_VNIC(bp)) +- vnic = find_matching_vnic(bp, filter); +- else +- vnic = BNXT_GET_DEFAULT_VNIC(bp); ++ vnic = find_matching_vnic(bp, filter); + done: + if (!ret || update_flow) { + flow->filter = filter; +diff --git a/dpdk/drivers/net/bnxt/bnxt_hwrm.c b/dpdk/drivers/net/bnxt/bnxt_hwrm.c +index f53f8632fe..9c5257309a 100644 +--- a/dpdk/drivers/net/bnxt/bnxt_hwrm.c ++++ b/dpdk/drivers/net/bnxt/bnxt_hwrm.c +@@ -506,8 +506,8 @@ int bnxt_hwrm_cfa_l2_set_rx_mask(struct bnxt *bp, + mask |= HWRM_CFA_L2_SET_RX_MASK_INPUT_MASK_ALL_MCAST; + } else if (vnic->flags & BNXT_VNIC_INFO_MCAST) { + mask |= HWRM_CFA_L2_SET_RX_MASK_INPUT_MASK_MCAST; +- req.num_mc_entries = rte_cpu_to_le_32(vnic->mc_addr_cnt); +- req.mc_tbl_addr = rte_cpu_to_le_64(vnic->mc_list_dma_addr); ++ req.num_mc_entries = rte_cpu_to_le_32(bp->nb_mc_addr); ++ req.mc_tbl_addr = rte_cpu_to_le_64(bp->mc_list_dma_addr); + } + if (vlan_table) { + if (!(mask & HWRM_CFA_L2_SET_RX_MASK_INPUT_MASK_VLAN_NONVLAN)) +@@ -902,18 +902,12 @@ static int __bnxt_hwrm_func_qcaps(struct bnxt *bp) + bp->max_l2_ctx = rte_le_to_cpu_16(resp->max_l2_ctxs); + if (!BNXT_CHIP_P5(bp) && !bp->pdev->max_vfs) + bp->max_l2_ctx += bp->max_rx_em_flows; +- /* TODO: For now, do not support VMDq/RFS on VFs. */ +- if (BNXT_PF(bp)) { +- if (bp->pf->max_vfs) +- bp->max_vnics = 1; +- else +- bp->max_vnics = rte_le_to_cpu_16(resp->max_vnics); +- } else { +- bp->max_vnics = 1; +- } ++ bp->max_vnics = rte_le_to_cpu_16(resp->max_vnics); + PMD_DRV_LOG(DEBUG, "Max l2_cntxts is %d vnics is %d\n", + bp->max_l2_ctx, bp->max_vnics); + bp->max_stat_ctx = rte_le_to_cpu_16(resp->max_stat_ctx); ++ bp->max_mcast_addr = rte_le_to_cpu_32(resp->max_mcast_filters); ++ + if (BNXT_PF(bp)) { + bp->pf->total_vnics = rte_le_to_cpu_16(resp->max_vnics); + if (flags & HWRM_FUNC_QCAPS_OUTPUT_FLAGS_PTP_SUPPORTED) { +@@ -945,6 +939,11 @@ static int __bnxt_hwrm_func_qcaps(struct bnxt *bp) + bp->fw_cap |= BNXT_FW_CAP_VLAN_TX_INSERT; + PMD_DRV_LOG(DEBUG, "VLAN acceleration for TX is enabled\n"); + } ++ ++ bp->tunnel_disable_flag = rte_le_to_cpu_16(resp->tunnel_disable_flag); ++ if (bp->tunnel_disable_flag) ++ PMD_DRV_LOG(DEBUG, "Tunnel parsing capability is disabled, flags : %#x\n", ++ bp->tunnel_disable_flag); + unlock: + HWRM_UNLOCK(); + +@@ -1250,11 +1249,6 @@ int bnxt_hwrm_ver_get(struct bnxt *bp, uint32_t timeout) + else + HWRM_CHECK_RESULT(); + +- if (resp->flags & HWRM_VER_GET_OUTPUT_FLAGS_DEV_NOT_RDY) { +- rc = -EAGAIN; +- goto error; +- } +- + PMD_DRV_LOG(INFO, "%d.%d.%d:%d.%d.%d.%d\n", + resp->hwrm_intf_maj_8b, resp->hwrm_intf_min_8b, + resp->hwrm_intf_upd_8b, resp->hwrm_fw_maj_8b, +@@ -1430,20 +1424,21 @@ static int bnxt_hwrm_port_phy_cfg(struct bnxt *bp, struct bnxt_link_info *conf) + } + } + /* AutoNeg - Advertise speeds specified. */ +- if (conf->auto_link_speed_mask && ++ if ((conf->auto_link_speed_mask || conf->auto_pam4_link_speed_mask) && + !(conf->phy_flags & HWRM_PORT_PHY_CFG_INPUT_FLAGS_FORCE)) { + req.auto_mode = + HWRM_PORT_PHY_CFG_INPUT_AUTO_MODE_SPEED_MASK; +- req.auto_link_speed_mask = +- conf->auto_link_speed_mask; +- if (conf->auto_pam4_link_speeds) { ++ if (conf->auto_pam4_link_speed_mask) { + enables |= + HWRM_PORT_PHY_CFG_IN_EN_AUTO_PAM4_LINK_SPD_MASK; + req.auto_link_pam4_speed_mask = +- conf->auto_pam4_link_speeds; +- } else { ++ rte_cpu_to_le_16(conf->auto_pam4_link_speed_mask); ++ } ++ if (conf->auto_link_speed_mask) { + enables |= + HWRM_PORT_PHY_CFG_IN_EN_AUTO_LINK_SPEED_MASK; ++ req.auto_link_speed_mask = ++ rte_cpu_to_le_16(conf->auto_link_speed_mask); + } + } + if (conf->auto_link_speed && +@@ -1511,12 +1506,12 @@ static int bnxt_hwrm_port_phy_qcfg(struct bnxt *bp, + link_info->phy_ver[1] = resp->phy_min; + link_info->phy_ver[2] = resp->phy_bld; + link_info->link_signal_mode = +- rte_le_to_cpu_16(resp->active_fec_signal_mode); ++ resp->active_fec_signal_mode & HWRM_PORT_PHY_QCFG_OUTPUT_SIGNAL_MODE_MASK; + link_info->force_pam4_link_speed = + rte_le_to_cpu_16(resp->force_pam4_link_speed); + link_info->support_pam4_speeds = + rte_le_to_cpu_16(resp->support_pam4_speeds); +- link_info->auto_pam4_link_speeds = ++ link_info->auto_pam4_link_speed_mask = + rte_le_to_cpu_16(resp->auto_pam4_link_speed_mask); + link_info->module_status = resp->module_status; + HWRM_UNLOCK(); +@@ -1527,7 +1522,7 @@ static int bnxt_hwrm_port_phy_qcfg(struct bnxt *bp, + link_info->support_speeds, link_info->force_link_speed); + PMD_DRV_LOG(DEBUG, "Link Signal:%d,PAM::Auto:%x,Support:%x,Force:%x\n", + link_info->link_signal_mode, +- link_info->auto_pam4_link_speeds, ++ link_info->auto_pam4_link_speed_mask, + link_info->support_pam4_speeds, + link_info->force_pam4_link_speed); + return rc; +@@ -2975,7 +2970,7 @@ static uint16_t bnxt_check_eth_link_autoneg(uint32_t conf_link) + } - return EINVAL; - } + static uint16_t bnxt_parse_eth_link_speed(uint32_t conf_link_speed, +- uint16_t pam4_link) ++ struct bnxt_link_info *link_info) + { + uint16_t eth_link_speed = 0; -- fitness = odp_flow_key_to_mask(mask, mask_len, &match->wc, &match->flow, -- NULL); -+ fitness = odp_flow_key_to_mask__(mask, mask_len, &match->wc, &match->flow, -+ NULL, true); - if (fitness) { -- /* This should not happen: it indicates that -- * odp_flow_key_from_mask() and odp_flow_key_to_mask() -- * disagree on the acceptable form of a mask. Log the problem -- * as an error, with enough details to enable debugging. */ -+ /* This should not happen, see comment above. */ - static struct vlog_rate_limit rl = VLOG_RATE_LIMIT_INIT(1, 5); +@@ -3014,18 +3009,29 @@ static uint16_t bnxt_parse_eth_link_speed(uint32_t conf_link_speed, + HWRM_PORT_PHY_CFG_INPUT_FORCE_LINK_SPEED_40GB; + break; + case RTE_ETH_LINK_SPEED_50G: +- eth_link_speed = pam4_link ? +- HWRM_PORT_PHY_CFG_INPUT_FORCE_PAM4_LINK_SPEED_50GB : +- HWRM_PORT_PHY_CFG_INPUT_FORCE_LINK_SPEED_50GB; ++ if (link_info->support_pam4_speeds & ++ HWRM_PORT_PHY_QCFG_OUTPUT_SUPPORT_PAM4_SPEEDS_50G) { ++ eth_link_speed = HWRM_PORT_PHY_CFG_INPUT_FORCE_PAM4_LINK_SPEED_50GB; ++ link_info->link_signal_mode = BNXT_SIG_MODE_PAM4; ++ } else { ++ eth_link_speed = HWRM_PORT_PHY_CFG_INPUT_FORCE_LINK_SPEED_50GB; ++ link_info->link_signal_mode = BNXT_SIG_MODE_NRZ; ++ } + break; + case RTE_ETH_LINK_SPEED_100G: +- eth_link_speed = pam4_link ? +- HWRM_PORT_PHY_CFG_INPUT_FORCE_PAM4_LINK_SPEED_100GB : +- HWRM_PORT_PHY_CFG_INPUT_FORCE_LINK_SPEED_100GB; ++ if (link_info->support_pam4_speeds & ++ HWRM_PORT_PHY_QCFG_OUTPUT_SUPPORT_PAM4_SPEEDS_100G) { ++ eth_link_speed = HWRM_PORT_PHY_CFG_INPUT_FORCE_PAM4_LINK_SPEED_100GB; ++ link_info->link_signal_mode = BNXT_SIG_MODE_PAM4; ++ } else { ++ eth_link_speed = HWRM_PORT_PHY_CFG_INPUT_FORCE_LINK_SPEED_100GB; ++ link_info->link_signal_mode = BNXT_SIG_MODE_NRZ; ++ } + break; + case RTE_ETH_LINK_SPEED_200G: + eth_link_speed = + HWRM_PORT_PHY_CFG_INPUT_FORCE_PAM4_LINK_SPEED_200GB; ++ link_info->link_signal_mode = BNXT_SIG_MODE_PAM4; + break; + default: + PMD_DRV_LOG(ERR, +@@ -3229,9 +3235,11 @@ int bnxt_set_hwrm_link_config(struct bnxt *bp, bool link_up) + if (!link_up) + goto port_phy_cfg; + ++ /* Get user requested autoneg setting */ + autoneg = bnxt_check_eth_link_autoneg(dev_conf->link_speeds); ++ + if (BNXT_CHIP_P5(bp) && +- dev_conf->link_speeds == RTE_ETH_LINK_SPEED_40G) { ++ dev_conf->link_speeds & RTE_ETH_LINK_SPEED_40G) { + /* 40G is not supported as part of media auto detect. + * The speed should be forced and autoneg disabled + * to configure 40G speed. +@@ -3240,24 +3248,27 @@ int bnxt_set_hwrm_link_config(struct bnxt *bp, bool link_up) + autoneg = 0; + } - if (!VLOG_DROP_ERR(&rl)) { -diff --git a/lib/ofp-actions.c b/lib/ofp-actions.c -index 006837c2e1..c13f97b5c9 100644 ---- a/lib/ofp-actions.c -+++ b/lib/ofp-actions.c -@@ -853,7 +853,9 @@ decode_NXAST_RAW_CONTROLLER2(const struct ext_action_header *eah, - case NXAC2PT_REASON: { - uint8_t u8; - error = ofpprop_parse_u8(&payload, &u8); -- oc->reason = u8; -+ if (!error) { -+ oc->reason = u8; -+ } - break; - } +- /* No auto speeds and no auto_pam4_link. Disable autoneg */ +- if (bp->link_info->auto_link_speed == 0 && +- bp->link_info->link_signal_mode && +- bp->link_info->auto_pam4_link_speeds == 0) ++ /* Override based on current Autoneg setting in PHY for 200G */ ++ if (autoneg == 1 && BNXT_CHIP_P5(bp) && bp->link_info->auto_mode == 0 && ++ bp->link_info->force_pam4_link_speed == ++ HWRM_PORT_PHY_CFG_INPUT_FORCE_PAM4_LINK_SPEED_200GB) { + autoneg = 0; ++ PMD_DRV_LOG(DEBUG, "Disabling autoneg for 200G\n"); ++ } + + speed = bnxt_parse_eth_link_speed(dev_conf->link_speeds, +- bp->link_info->link_signal_mode); ++ bp->link_info); + link_req.phy_flags = HWRM_PORT_PHY_CFG_INPUT_FLAGS_RESET_PHY; + /* Autoneg can be done only when the FW allows. */ +- if (autoneg == 1 && bp->link_info->support_auto_speeds) { ++ if (autoneg == 1 && ++ (bp->link_info->support_auto_speeds || bp->link_info->support_pam4_auto_speeds)) { + link_req.phy_flags |= + HWRM_PORT_PHY_CFG_INPUT_FLAGS_RESTART_AUTONEG; + link_req.auto_link_speed_mask = + bnxt_parse_eth_link_speed_mask(bp, + dev_conf->link_speeds); +- link_req.auto_pam4_link_speeds = +- bp->link_info->auto_pam4_link_speeds; ++ link_req.auto_pam4_link_speed_mask = ++ bp->link_info->auto_pam4_link_speed_mask; + } else { + if (bp->link_info->phy_type == + HWRM_PORT_PHY_QCFG_OUTPUT_PHY_TYPE_BASET || +@@ -3276,21 +3287,21 @@ int bnxt_set_hwrm_link_config(struct bnxt *bp, bool link_up) + else if (bp->link_info->force_pam4_link_speed) + link_req.link_speed = + bp->link_info->force_pam4_link_speed; +- else if (bp->link_info->auto_pam4_link_speeds) ++ else if (bp->link_info->force_link_speed) ++ link_req.link_speed = bp->link_info->force_link_speed; ++ else if (bp->link_info->auto_pam4_link_speed_mask) + link_req.link_speed = +- bp->link_info->auto_pam4_link_speeds; ++ bp->link_info->auto_pam4_link_speed_mask; + else if (bp->link_info->support_pam4_speeds) + link_req.link_speed = + bp->link_info->support_pam4_speeds; +- else if (bp->link_info->force_link_speed) +- link_req.link_speed = bp->link_info->force_link_speed; + else + link_req.link_speed = bp->link_info->auto_link_speed; + /* Auto PAM4 link speed is zero, but auto_link_speed is not + * zero. Use the auto_link_speed. + */ + if (bp->link_info->auto_link_speed != 0 && +- bp->link_info->auto_pam4_link_speeds == 0) ++ bp->link_info->auto_pam4_link_speed_mask == 0) + link_req.link_speed = bp->link_info->auto_link_speed; + } + link_req.duplex = bnxt_parse_eth_link_duplex(dev_conf->link_speeds); +@@ -3491,7 +3502,7 @@ static int bnxt_hwrm_pf_func_cfg(struct bnxt *bp, + rte_cpu_to_le_16(pf_resc->num_hw_ring_grps); + } else if (BNXT_HAS_NQ(bp)) { + enables |= HWRM_FUNC_CFG_INPUT_ENABLES_NUM_MSIX; +- req.num_msix = rte_cpu_to_le_16(bp->max_nq_rings); ++ req.num_msix = rte_cpu_to_le_16(pf_resc->num_nq_rings); + } -@@ -3200,16 +3202,21 @@ set_field_split_str(char *arg, char **key, char **value, char **delim) - { - char *value_end; + req.flags = rte_cpu_to_le_32(bp->pf->func_cfg_flags); +@@ -3504,7 +3515,7 @@ static int bnxt_hwrm_pf_func_cfg(struct bnxt *bp, + req.num_tx_rings = rte_cpu_to_le_16(pf_resc->num_tx_rings); + req.num_rx_rings = rte_cpu_to_le_16(pf_resc->num_rx_rings); + req.num_l2_ctxs = rte_cpu_to_le_16(pf_resc->num_l2_ctxs); +- req.num_vnics = rte_cpu_to_le_16(bp->max_vnics); ++ req.num_vnics = rte_cpu_to_le_16(pf_resc->num_vnics); + req.fid = rte_cpu_to_le_16(0xffff); + req.enables = rte_cpu_to_le_32(enables); + +@@ -3541,14 +3552,12 @@ bnxt_fill_vf_func_cfg_req_new(struct bnxt *bp, + req->min_rx_rings = req->max_rx_rings; + req->max_l2_ctxs = rte_cpu_to_le_16(bp->max_l2_ctx / (num_vfs + 1)); + req->min_l2_ctxs = req->max_l2_ctxs; +- /* TODO: For now, do not support VMDq/RFS on VFs. */ +- req->max_vnics = rte_cpu_to_le_16(1); ++ req->max_vnics = rte_cpu_to_le_16(bp->max_vnics / (num_vfs + 1)); + req->min_vnics = req->max_vnics; + req->max_hw_ring_grps = rte_cpu_to_le_16(bp->max_ring_grps / + (num_vfs + 1)); + req->min_hw_ring_grps = req->max_hw_ring_grps; +- req->flags = +- rte_cpu_to_le_16(HWRM_FUNC_VF_RESOURCE_CFG_INPUT_FLAGS_MIN_GUARANTEED); ++ req->max_msix = rte_cpu_to_le_16(bp->max_nq_rings / (num_vfs + 1)); + } -+ *key = NULL; - *value = arg; -- value_end = strstr(arg, "->"); -- *key = value_end + strlen("->"); - if (delim) { -- *delim = value_end; -+ *delim = NULL; - } + static void +@@ -3608,6 +3617,8 @@ static int bnxt_update_max_resources(struct bnxt *bp, + bp->max_rx_rings -= rte_le_to_cpu_16(resp->alloc_rx_rings); + bp->max_l2_ctx -= rte_le_to_cpu_16(resp->alloc_l2_ctx); + bp->max_ring_grps -= rte_le_to_cpu_16(resp->alloc_hw_ring_grps); ++ bp->max_nq_rings -= rte_le_to_cpu_16(resp->alloc_msix); ++ bp->max_vnics -= rte_le_to_cpu_16(resp->alloc_vnics); + + HWRM_UNLOCK(); + +@@ -3681,6 +3692,8 @@ static int bnxt_query_pf_resources(struct bnxt *bp, + pf_resc->num_rx_rings = rte_le_to_cpu_16(resp->alloc_rx_rings); + pf_resc->num_l2_ctxs = rte_le_to_cpu_16(resp->alloc_l2_ctx); + pf_resc->num_hw_ring_grps = rte_le_to_cpu_32(resp->alloc_hw_ring_grps); ++ pf_resc->num_nq_rings = rte_le_to_cpu_32(resp->alloc_msix); ++ pf_resc->num_vnics = rte_le_to_cpu_16(resp->alloc_vnics); + bp->pf->evb_mode = resp->evb_mode; + + HWRM_UNLOCK(); +@@ -3701,6 +3714,8 @@ bnxt_calculate_pf_resources(struct bnxt *bp, + pf_resc->num_rx_rings = bp->max_rx_rings; + pf_resc->num_l2_ctxs = bp->max_l2_ctx; + pf_resc->num_hw_ring_grps = bp->max_ring_grps; ++ pf_resc->num_nq_rings = bp->max_nq_rings; ++ pf_resc->num_vnics = bp->max_vnics; + + return; + } +@@ -3719,6 +3734,10 @@ bnxt_calculate_pf_resources(struct bnxt *bp, + bp->max_l2_ctx % (num_vfs + 1); + pf_resc->num_hw_ring_grps = bp->max_ring_grps / (num_vfs + 1) + + bp->max_ring_grps % (num_vfs + 1); ++ pf_resc->num_nq_rings = bp->max_nq_rings / (num_vfs + 1) + ++ bp->max_nq_rings % (num_vfs + 1); ++ pf_resc->num_vnics = bp->max_vnics / (num_vfs + 1) + ++ bp->max_vnics % (num_vfs + 1); + } -+ value_end = strstr(arg, "->"); - if (!value_end) { - return xasprintf("%s: missing `->'", arg); - } -+ -+ *key = value_end + strlen("->"); -+ if (delim) { -+ *delim = value_end; -+ } - if (strlen(value_end) <= strlen("->")) { - return xasprintf("%s: missing field name following `->'", arg); - } -diff --git a/lib/ofp-msgs.c b/lib/ofp-msgs.c -index fec54f75f8..93aa812978 100644 ---- a/lib/ofp-msgs.c -+++ b/lib/ofp-msgs.c -@@ -1123,8 +1123,8 @@ ofpmp_partial_error(struct hmap *assembler, struct ofpmp_partial *p, - void - ofpmp_assembler_clear(struct hmap *assembler) + int bnxt_hwrm_allocate_pf_only(struct bnxt *bp) +@@ -3727,7 +3746,7 @@ int bnxt_hwrm_allocate_pf_only(struct bnxt *bp) + int rc; + + if (!BNXT_PF(bp)) { +- PMD_DRV_LOG(ERR, "Attempt to allcoate VFs on a VF!\n"); ++ PMD_DRV_LOG(ERR, "Attempt to allocate VFs on a VF!\n"); + return -EINVAL; + } + +@@ -3894,6 +3913,8 @@ bnxt_update_pf_resources(struct bnxt *bp, + bp->max_tx_rings = pf_resc->num_tx_rings; + bp->max_rx_rings = pf_resc->num_rx_rings; + bp->max_ring_grps = pf_resc->num_hw_ring_grps; ++ bp->max_nq_rings = pf_resc->num_nq_rings; ++ bp->max_vnics = pf_resc->num_vnics; + } + + static int32_t +@@ -6106,38 +6127,6 @@ int bnxt_hwrm_cfa_pair_free(struct bnxt *bp, struct bnxt_representor *rep_bp) + return rc; + } + +-int bnxt_hwrm_cfa_adv_flow_mgmt_qcaps(struct bnxt *bp) +-{ +- struct hwrm_cfa_adv_flow_mgnt_qcaps_output *resp = +- bp->hwrm_cmd_resp_addr; +- struct hwrm_cfa_adv_flow_mgnt_qcaps_input req = {0}; +- uint32_t flags = 0; +- int rc = 0; +- +- if (!(bp->fw_cap & BNXT_FW_CAP_ADV_FLOW_MGMT)) +- return 0; +- +- if (!(BNXT_PF(bp) || BNXT_VF_IS_TRUSTED(bp))) { +- PMD_DRV_LOG(DEBUG, +- "Not a PF or trusted VF. Command not supported\n"); +- return 0; +- } +- +- HWRM_PREP(&req, HWRM_CFA_ADV_FLOW_MGNT_QCAPS, BNXT_USE_CHIMP_MB); +- rc = bnxt_hwrm_send_message(bp, &req, sizeof(req), BNXT_USE_CHIMP_MB); +- +- HWRM_CHECK_RESULT(); +- flags = rte_le_to_cpu_32(resp->flags); +- HWRM_UNLOCK(); +- +- if (flags & HWRM_CFA_ADV_FLOW_MGNT_QCAPS_RFS_RING_TBL_IDX_V2_SUPPORTED) +- bp->flags |= BNXT_FLAG_FLOW_CFA_RFS_RING_TBL_IDX_V2; +- else +- bp->flags |= BNXT_FLAG_RFS_NEEDS_VNIC; +- +- return rc; +-} +- + int bnxt_hwrm_fw_echo_reply(struct bnxt *bp, uint32_t echo_req_data1, + uint32_t echo_req_data2) { -- struct ofpmp_partial *p, *next; -- HMAP_FOR_EACH_SAFE (p, next, hmap_node, assembler) { -+ struct ofpmp_partial *p; -+ HMAP_FOR_EACH_SAFE (p, hmap_node, assembler) { - ofpmp_partial_destroy(assembler, p); - } +@@ -6175,10 +6164,6 @@ int bnxt_hwrm_poll_ver_get(struct bnxt *bp) + rc = bnxt_hwrm_send_message(bp, &req, sizeof(req), BNXT_USE_CHIMP_MB); + + HWRM_CHECK_RESULT_SILENT(); +- +- if (resp->flags & HWRM_VER_GET_OUTPUT_FLAGS_DEV_NOT_RDY) +- rc = -EAGAIN; +- + HWRM_UNLOCK(); + + return rc; +@@ -6262,3 +6247,26 @@ int bnxt_hwrm_config_host_mtu(struct bnxt *bp) + + return rc; } -@@ -1290,8 +1290,8 @@ ofpmp_assembler_execute(struct hmap *assembler, struct ofpbuf *msg, - * on either side by parts with 0-byte bodies. We remove the 0-byte - * ones here to simplify processing later. - */ -- struct ofpbuf *b, *next; -- LIST_FOR_EACH_SAFE (b, next, list_node, out) { -+ struct ofpbuf *b; -+ LIST_FOR_EACH_SAFE (b, list_node, out) { - if (b->size <= min_len && !ovs_list_is_short(out)) { - ovs_list_remove(&b->list_node); - ofpbuf_delete(b); -diff --git a/lib/ofp-packet.c b/lib/ofp-packet.c -index 4579548ee1..9485ddfc93 100644 ---- a/lib/ofp-packet.c -+++ b/lib/ofp-packet.c -@@ -133,7 +133,9 @@ decode_nx_packet_in2(const struct ofp_header *oh, bool loose, - case NXPINT_FULL_LEN: { - uint32_t u32; - error = ofpprop_parse_u32(&payload, &u32); -- *total_len = u32; -+ if (!error) { -+ *total_len = u32; -+ } - break; - } ++ ++int ++bnxt_vnic_rss_clear_p5(struct bnxt *bp, struct bnxt_vnic_info *vnic) ++{ ++ struct hwrm_vnic_rss_cfg_output *resp = bp->hwrm_cmd_resp_addr; ++ struct hwrm_vnic_rss_cfg_input req = {0}; ++ int nr_ctxs = vnic->num_lb_ctxts; ++ int i, rc = 0; ++ ++ for (i = 0; i < nr_ctxs; i++) { ++ HWRM_PREP(&req, HWRM_VNIC_RSS_CFG, BNXT_USE_CHIMP_MB); ++ ++ req.rss_ctx_idx = rte_cpu_to_le_16(vnic->fw_grp_ids[i]); ++ req.vnic_id = rte_cpu_to_le_16(vnic->fw_vnic_id); ++ ++ rc = bnxt_hwrm_send_message(bp, &req, sizeof(req), BNXT_USE_CHIMP_MB); ++ ++ HWRM_CHECK_RESULT(); ++ HWRM_UNLOCK(); ++ } ++ ++ return rc; ++} +diff --git a/dpdk/drivers/net/bnxt/bnxt_hwrm.h b/dpdk/drivers/net/bnxt/bnxt_hwrm.h +index f8f0556201..a82d9fb3ef 100644 +--- a/dpdk/drivers/net/bnxt/bnxt_hwrm.h ++++ b/dpdk/drivers/net/bnxt/bnxt_hwrm.h +@@ -58,9 +58,6 @@ struct hwrm_func_qstats_output; + #define HWRM_PORT_PHY_CFG_IN_EN_AUTO_LINK_SPEED_MASK \ + HWRM_PORT_PHY_CFG_INPUT_ENABLES_AUTO_LINK_SPEED_MASK + +-#define HWRM_CFA_ADV_FLOW_MGNT_QCAPS_RFS_RING_TBL_IDX_V2_SUPPORTED \ +- HWRM_CFA_ADV_FLOW_MGNT_QCAPS_OUTPUT_FLAGS_RFS_RING_TBL_IDX_V2_SUPPORTED +- + #define HWRM_SPEC_CODE_1_8_4 0x10804 + #define HWRM_SPEC_CODE_1_9_0 0x10900 + #define HWRM_SPEC_CODE_1_9_2 0x10902 +@@ -117,11 +114,36 @@ struct bnxt_pf_resource_info { + uint16_t num_rx_rings; + uint16_t num_cp_rings; + uint16_t num_l2_ctxs; ++ uint16_t num_nq_rings; ++ uint16_t num_vnics; + uint32_t num_hw_ring_grps; + }; -@@ -152,7 +154,9 @@ decode_nx_packet_in2(const struct ofp_header *oh, bool loose, - case NXPINT_REASON: { - uint8_t reason; - error = ofpprop_parse_u8(&payload, &reason); -- pin->reason = reason; -+ if (!error) { -+ pin->reason = reason; -+ } - break; - } + #define BNXT_CTX_VAL_INVAL 0xFFFF -@@ -883,7 +887,9 @@ ofputil_decode_packet_in_private(const struct ofp_header *oh, bool loose, - case NXCPT_ODP_PORT: { - uint32_t value; - error = ofpprop_parse_u32(&payload, &value); -- pin->odp_port = u32_to_odp(value); -+ if (!error) { -+ pin->odp_port = u32_to_odp(value); -+ } - break; - } ++#define BNXT_TUNNELED_OFFLOADS_CAP_VXLAN_EN(bp) \ ++ (!((bp)->tunnel_disable_flag & HWRM_FUNC_QCAPS_OUTPUT_TUNNEL_DISABLE_FLAG_DISABLE_VXLAN)) ++#define BNXT_TUNNELED_OFFLOADS_CAP_NGE_EN(bp) \ ++ (!((bp)->tunnel_disable_flag & HWRM_FUNC_QCAPS_OUTPUT_TUNNEL_DISABLE_FLAG_DISABLE_NGE)) ++#define BNXT_TUNNELED_OFFLOADS_CAP_GRE_EN(bp) \ ++ (!((bp)->tunnel_disable_flag & HWRM_FUNC_QCAPS_OUTPUT_TUNNEL_DISABLE_FLAG_DISABLE_GRE)) ++#define BNXT_TUNNELED_OFFLOADS_CAP_IPINIP_EN(bp) \ ++ (!((bp)->tunnel_disable_flag & HWRM_FUNC_QCAPS_OUTPUT_TUNNEL_DISABLE_FLAG_DISABLE_IPINIP)) ++ ++/* ++ * If the device supports VXLAN, GRE, IPIP and GENEVE tunnel parsing, then report ++ * RTE_ETH_RX_OFFLOAD_OUTER_IPV4_CKSUM, RTE_ETH_RX_OFFLOAD_OUTER_UDP_CKSUM and ++ * RTE_ETH_TX_OFFLOAD_OUTER_IPV4_CKSUM in the Rx/Tx offload capabilities of the device. ++ */ ++#define BNXT_TUNNELED_OFFLOADS_CAP_ALL_EN(bp) \ ++ (BNXT_TUNNELED_OFFLOADS_CAP_VXLAN_EN(bp) && \ ++ BNXT_TUNNELED_OFFLOADS_CAP_NGE_EN(bp) && \ ++ BNXT_TUNNELED_OFFLOADS_CAP_GRE_EN(bp) && \ ++ BNXT_TUNNELED_OFFLOADS_CAP_IPINIP_EN(bp)) ++ ++#define BNXT_SIG_MODE_NRZ HWRM_PORT_PHY_QCFG_OUTPUT_SIGNAL_MODE_NRZ ++#define BNXT_SIG_MODE_PAM4 HWRM_PORT_PHY_QCFG_OUTPUT_SIGNAL_MODE_PAM4 ++ + int bnxt_hwrm_cfa_l2_clear_rx_mask(struct bnxt *bp, + struct bnxt_vnic_info *vnic); + int bnxt_hwrm_cfa_l2_set_rx_mask(struct bnxt *bp, struct bnxt_vnic_info *vnic, +@@ -296,7 +318,6 @@ int bnxt_hwrm_first_vf_id_query(struct bnxt *bp, uint16_t fid, + int bnxt_hwrm_cfa_pair_exists(struct bnxt *bp, struct bnxt_representor *rep_bp); + int bnxt_hwrm_cfa_pair_alloc(struct bnxt *bp, struct bnxt_representor *rep); + int bnxt_hwrm_cfa_pair_free(struct bnxt *bp, struct bnxt_representor *rep); +-int bnxt_hwrm_cfa_adv_flow_mgmt_qcaps(struct bnxt *bp); + int bnxt_hwrm_fw_echo_reply(struct bnxt *bp, uint32_t echo_req_data1, + uint32_t echo_req_data2); + int bnxt_hwrm_poll_ver_get(struct bnxt *bp); +@@ -310,4 +331,5 @@ int bnxt_hwrm_stat_ctx_alloc(struct bnxt *bp, struct bnxt_cp_ring_info *cpr); + void bnxt_free_hwrm_tx_ring(struct bnxt *bp, int queue_index); + int bnxt_alloc_hwrm_tx_ring(struct bnxt *bp, int queue_index); + int bnxt_hwrm_config_host_mtu(struct bnxt *bp); ++int bnxt_vnic_rss_clear_p5(struct bnxt *bp, struct bnxt_vnic_info *vnic); + #endif +diff --git a/dpdk/drivers/net/bnxt/bnxt_reps.c b/dpdk/drivers/net/bnxt/bnxt_reps.c +index 22b76b72b9..299b4c24a8 100644 +--- a/dpdk/drivers/net/bnxt/bnxt_reps.c ++++ b/dpdk/drivers/net/bnxt/bnxt_reps.c +@@ -35,16 +35,20 @@ static const struct eth_dev_ops bnxt_rep_dev_ops = { + uint16_t + bnxt_vfr_recv(uint16_t port_id, uint16_t queue_id, struct rte_mbuf *mbuf) + { +- struct rte_mbuf **prod_rx_buf; ++ struct bnxt_representor *vfr_bp = NULL; + struct bnxt_rx_ring_info *rep_rxr; +- struct bnxt_rx_queue *rep_rxq; + struct rte_eth_dev *vfr_eth_dev; +- struct bnxt_representor *vfr_bp; ++ struct rte_mbuf **prod_rx_buf; ++ struct bnxt_rx_queue *rep_rxq; + uint16_t mask; + uint8_t que; + + vfr_eth_dev = &rte_eth_devices[port_id]; +- vfr_bp = vfr_eth_dev->data->dev_private; ++ vfr_bp = vfr_eth_dev ? vfr_eth_dev->data->dev_private : NULL; ++ ++ if (unlikely(vfr_bp == NULL)) ++ return 1; ++ + /* If rxq_id happens to be > nr_rings, use ring 0 */ + que = queue_id < vfr_bp->rx_nr_rings ? queue_id : 0; + rep_rxq = vfr_bp->rx_queues[que]; +@@ -545,7 +549,10 @@ int bnxt_rep_dev_info_get_op(struct rte_eth_dev *eth_dev, + dev_info->max_mac_addrs = parent_bp->max_l2_ctx; + dev_info->max_hash_mac_addrs = 0; + +- max_rx_rings = BNXT_MAX_VF_REP_RINGS; ++ max_rx_rings = parent_bp->rx_nr_rings ? ++ RTE_MIN(parent_bp->rx_nr_rings, BNXT_MAX_VF_REP_RINGS) : ++ BNXT_MAX_VF_REP_RINGS; ++ + /* For the sake of symmetry, max_rx_queues = max_tx_queues */ + dev_info->max_rx_queues = max_rx_rings; + dev_info->max_tx_queues = max_rx_rings; +@@ -561,10 +568,8 @@ int bnxt_rep_dev_info_get_op(struct rte_eth_dev *eth_dev, + dev_info->min_rx_bufsize = 1; + dev_info->max_rx_pktlen = BNXT_MAX_PKT_LEN; + +- dev_info->rx_offload_capa = BNXT_DEV_RX_OFFLOAD_SUPPORT; +- if (parent_bp->flags & BNXT_FLAG_PTP_SUPPORTED) +- dev_info->rx_offload_capa |= RTE_ETH_RX_OFFLOAD_TIMESTAMP; +- dev_info->tx_offload_capa = BNXT_DEV_TX_OFFLOAD_SUPPORT; ++ dev_info->rx_offload_capa = bnxt_get_rx_port_offloads(parent_bp); ++ dev_info->tx_offload_capa = bnxt_get_tx_port_offloads(parent_bp); + dev_info->flow_type_rss_offloads = BNXT_ETH_RSS_SUPPORT; + + dev_info->switch_info.name = eth_dev->device->name; +@@ -626,10 +631,10 @@ int bnxt_rep_rx_queue_setup_op(struct rte_eth_dev *eth_dev, + struct rte_mbuf **buf_ring; + int rc = 0; + +- if (queue_idx >= BNXT_MAX_VF_REP_RINGS) { ++ if (queue_idx >= rep_bp->rx_nr_rings) { + PMD_DRV_LOG(ERR, + "Cannot create Rx ring %d. %d rings available\n", +- queue_idx, BNXT_MAX_VF_REP_RINGS); ++ queue_idx, rep_bp->rx_nr_rings); + return -EINVAL; + } -diff --git a/lib/ofpbuf.c b/lib/ofpbuf.c -index 271105bdea..879275a7a3 100644 ---- a/lib/ofpbuf.c -+++ b/lib/ofpbuf.c -@@ -426,6 +426,10 @@ void - ofpbuf_reserve(struct ofpbuf *b, size_t size) +@@ -726,10 +731,10 @@ int bnxt_rep_tx_queue_setup_op(struct rte_eth_dev *eth_dev, + struct bnxt_tx_queue *parent_txq, *txq; + struct bnxt_vf_rep_tx_queue *vfr_txq; + +- if (queue_idx >= BNXT_MAX_VF_REP_RINGS) { ++ if (queue_idx >= rep_bp->rx_nr_rings) { + PMD_DRV_LOG(ERR, + "Cannot create Tx rings %d. %d rings available\n", +- queue_idx, BNXT_MAX_VF_REP_RINGS); ++ queue_idx, rep_bp->rx_nr_rings); + return -EINVAL; + } + +@@ -802,10 +807,10 @@ int bnxt_rep_stats_get_op(struct rte_eth_dev *eth_dev, + struct rte_eth_stats *stats) { - ovs_assert(!b->size); -+ -+ if (!size) { -+ return; -+ } - ofpbuf_prealloc_tailroom(b, size); - b->data = (char*)b->data + size; - } -@@ -436,6 +440,10 @@ ofpbuf_reserve(struct ofpbuf *b, size_t size) - void * - ofpbuf_push_uninit(struct ofpbuf *b, size_t size) + struct bnxt_representor *rep_bp = eth_dev->data->dev_private; +- int i; ++ unsigned int i; + + memset(stats, 0, sizeof(*stats)); +- for (i = 0; i < BNXT_MAX_VF_REP_RINGS; i++) { ++ for (i = 0; i < rep_bp->rx_nr_rings; i++) { + stats->obytes += rep_bp->tx_bytes[i]; + stats->opackets += rep_bp->tx_pkts[i]; + stats->ibytes += rep_bp->rx_bytes[i]; +@@ -825,9 +830,9 @@ int bnxt_rep_stats_get_op(struct rte_eth_dev *eth_dev, + int bnxt_rep_stats_reset_op(struct rte_eth_dev *eth_dev) { -+ if (!size) { -+ return b->data; -+ } + struct bnxt_representor *rep_bp = eth_dev->data->dev_private; +- int i; ++ unsigned int i; + +- for (i = 0; i < BNXT_MAX_VF_REP_RINGS; i++) { ++ for (i = 0; i < rep_bp->rx_nr_rings; i++) { + rep_bp->tx_pkts[i] = 0; + rep_bp->tx_bytes[i] = 0; + rep_bp->rx_pkts[i] = 0; +diff --git a/dpdk/drivers/net/bnxt/bnxt_ring.c b/dpdk/drivers/net/bnxt/bnxt_ring.c +index dc437f314e..4cdbb177d9 100644 +--- a/dpdk/drivers/net/bnxt/bnxt_ring.c ++++ b/dpdk/drivers/net/bnxt/bnxt_ring.c +@@ -752,6 +752,8 @@ int bnxt_alloc_hwrm_rings(struct bnxt *bp) + rc = bnxt_alloc_hwrm_rx_ring(bp, i); + if (rc) + goto err_out; ++ bnxt_hwrm_set_ring_coal(bp, &coal, ++ rxq->cp_ring->cp_ring_struct->fw_ring_id); + } + + /* If something is wrong with Rx ring alloc, skip Tx ring alloc */ +@@ -851,6 +853,7 @@ int bnxt_alloc_async_ring_struct(struct bnxt *bp) + ring->ring_mask = ring->ring_size - 1; + ring->vmem_size = 0; + ring->vmem = NULL; ++ ring->fw_ring_id = INVALID_HW_RING_ID; + + bp->async_cp_ring = cpr; + cpr->cp_ring_struct = ring; +diff --git a/dpdk/drivers/net/bnxt/bnxt_rxq.c b/dpdk/drivers/net/bnxt/bnxt_rxq.c +index 1456f8b54f..4f2e0e7376 100644 +--- a/dpdk/drivers/net/bnxt/bnxt_rxq.c ++++ b/dpdk/drivers/net/bnxt/bnxt_rxq.c +@@ -20,6 +20,32 @@ + * RX Queues + */ + ++uint64_t bnxt_get_rx_port_offloads(struct bnxt *bp) ++{ ++ uint64_t rx_offload_capa; ++ ++ rx_offload_capa = RTE_ETH_RX_OFFLOAD_IPV4_CKSUM | ++ RTE_ETH_RX_OFFLOAD_UDP_CKSUM | ++ RTE_ETH_RX_OFFLOAD_TCP_CKSUM | ++ RTE_ETH_RX_OFFLOAD_KEEP_CRC | ++ RTE_ETH_RX_OFFLOAD_VLAN_FILTER | ++ RTE_ETH_RX_OFFLOAD_VLAN_EXTEND | ++ RTE_ETH_RX_OFFLOAD_TCP_LRO | ++ RTE_ETH_RX_OFFLOAD_SCATTER | ++ RTE_ETH_RX_OFFLOAD_RSS_HASH; ++ ++ if (bp->flags & BNXT_FLAG_PTP_SUPPORTED) ++ rx_offload_capa |= RTE_ETH_RX_OFFLOAD_TIMESTAMP; ++ if (bp->vnic_cap_flags & BNXT_VNIC_CAP_VLAN_RX_STRIP) ++ rx_offload_capa |= RTE_ETH_RX_OFFLOAD_VLAN_STRIP; ++ ++ if (BNXT_TUNNELED_OFFLOADS_CAP_ALL_EN(bp)) ++ rx_offload_capa |= RTE_ETH_RX_OFFLOAD_OUTER_IPV4_CKSUM | ++ RTE_ETH_RX_OFFLOAD_OUTER_UDP_CKSUM; ++ ++ return rx_offload_capa; ++} + - ofpbuf_prealloc_headroom(b, size); - b->data = (char*)b->data - size; - b->size += size; -diff --git a/lib/ovs-lldp.c b/lib/ovs-lldp.c -index 162311fa45..2d13e971ed 100644 ---- a/lib/ovs-lldp.c -+++ b/lib/ovs-lldp.c -@@ -559,9 +559,9 @@ aa_mapping_unregister_mapping(struct lldp *lldp, - struct lldpd_hardware *hw, - struct aa_mapping_internal *m) + /* Determine whether the current configuration needs aggregation ring in HW. */ + int bnxt_need_agg_ring(struct rte_eth_dev *eth_dev) { -- struct lldpd_aa_isid_vlan_maps_tlv *lm, *lm_next; -+ struct lldpd_aa_isid_vlan_maps_tlv *lm; +@@ -40,6 +66,7 @@ void bnxt_free_rxq_stats(struct bnxt_rx_queue *rxq) + int bnxt_mq_rx_configure(struct bnxt *bp) + { + struct rte_eth_conf *dev_conf = &bp->eth_dev->data->dev_conf; ++ struct rte_eth_rss_conf *rss = &bp->rss_conf; + const struct rte_eth_vmdq_rx_conf *conf = + &dev_conf->rx_adv_conf.vmdq_rx_conf; + unsigned int i, j, nb_q_per_grp = 1, ring_idx = 0; +@@ -147,32 +174,19 @@ int bnxt_mq_rx_configure(struct bnxt *bp) -- LIST_FOR_EACH_SAFE (lm, lm_next, m_entries, -+ LIST_FOR_EACH_SAFE (lm, m_entries, - &hw->h_lport.p_isid_vlan_maps) { - uint32_t isid = lm->isid_vlan_data.isid; + bp->rx_num_qs_per_vnic = nb_q_per_grp; -@@ -738,6 +738,14 @@ lldp_put_packet(struct lldp *lldp, struct dp_packet *packet, - ovs_mutex_unlock(&mutex); - } +- if (dev_conf->rxmode.mq_mode & RTE_ETH_MQ_RX_RSS_FLAG) { +- struct rte_eth_rss_conf *rss = &dev_conf->rx_adv_conf.rss_conf; +- +- if (bp->flags & BNXT_FLAG_UPDATE_HASH) +- bp->flags &= ~BNXT_FLAG_UPDATE_HASH; +- +- for (i = 0; i < bp->nr_vnics; i++) { +- uint32_t lvl = RTE_ETH_RSS_LEVEL(rss->rss_hf); +- +- vnic = &bp->vnic_info[i]; +- vnic->hash_type = +- bnxt_rte_to_hwrm_hash_types(rss->rss_hf); +- vnic->hash_mode = +- bnxt_rte_to_hwrm_hash_level(bp, +- rss->rss_hf, +- lvl); +- +- /* +- * Use the supplied key if the key length is +- * acceptable and the rss_key is not NULL +- */ +- if (rss->rss_key && +- rss->rss_key_len <= HW_HASH_KEY_SIZE) +- memcpy(vnic->rss_hash_key, +- rss->rss_key, rss->rss_key_len); +- } ++ for (i = 0; i < bp->nr_vnics; i++) { ++ uint32_t lvl = RTE_ETH_RSS_LEVEL(rss->rss_hf); ++ ++ vnic = &bp->vnic_info[i]; ++ vnic->hash_type = bnxt_rte_to_hwrm_hash_types(rss->rss_hf); ++ vnic->hash_mode = bnxt_rte_to_hwrm_hash_level(bp, rss->rss_hf, lvl); ++ ++ /* ++ * Use the supplied key if the key length is ++ * acceptable and the rss_key is not NULL ++ */ ++ if (rss->rss_key && rss->rss_key_len <= HW_HASH_KEY_SIZE) ++ memcpy(vnic->rss_hash_key, rss->rss_key, rss->rss_key_len); + } -+/* Is LLDP enabled? -+ */ -+bool -+lldp_is_enabled(struct lldp *lldp) + return rc; +@@ -214,7 +228,9 @@ void bnxt_rx_queue_release_mbufs(struct bnxt_rx_queue *rxq) + } + } + /* Free up mbufs in Agg ring */ +- if (!bnxt_need_agg_ring(rxq->bp->eth_dev)) ++ if (rxq->bp == NULL || ++ rxq->bp->eth_dev == NULL || ++ !bnxt_need_agg_ring(rxq->bp->eth_dev)) + return; + + sw_ring = rxq->rx_ring->ag_buf_ring; +@@ -387,10 +403,6 @@ int bnxt_rx_queue_setup_op(struct rte_eth_dev *eth_dev, + rxq->rx_started = rxq->rx_deferred_start ? false : true; + rxq->vnic = BNXT_GET_DEFAULT_VNIC(bp); + +- /* Configure mtu if it is different from what was configured before */ +- if (!queue_idx) +- bnxt_mtu_set_op(eth_dev, eth_dev->data->mtu); +- + return 0; + err: + bnxt_rx_queue_release_op(eth_dev, queue_idx); +@@ -472,10 +484,11 @@ int bnxt_rx_queue_start(struct rte_eth_dev *dev, uint16_t rx_queue_id) + if (rc) + return rc; + +- if (BNXT_CHIP_P5(bp)) { +- /* Reconfigure default receive ring and MRU. */ +- bnxt_hwrm_vnic_cfg(bp, rxq->vnic); +- } ++ if (BNXT_HAS_RING_GRPS(bp)) ++ rxq->vnic->dflt_ring_grp = bp->grp_info[rx_queue_id].fw_grp_id; ++ /* Reconfigure default receive ring and MRU. */ ++ bnxt_hwrm_vnic_cfg(bp, rxq->vnic); ++ + PMD_DRV_LOG(INFO, "Rx queue started %d\n", rx_queue_id); + + if (dev_conf->rxmode.mq_mode & RTE_ETH_MQ_RX_RSS_FLAG) { +@@ -574,6 +587,9 @@ int bnxt_rx_queue_stop(struct rte_eth_dev *dev, uint16_t rx_queue_id) + if (active_queue_cnt == 0) { + uint16_t saved_mru = vnic->mru; + ++ /* clear RSS setting on vnic. */ ++ bnxt_vnic_rss_clear_p5(bp, vnic); ++ + vnic->mru = 0; + /* Reconfigure default receive ring and MRU. */ + bnxt_hwrm_vnic_cfg(bp, vnic); +diff --git a/dpdk/drivers/net/bnxt/bnxt_rxq.h b/dpdk/drivers/net/bnxt/bnxt_rxq.h +index 0331c23810..287df8dff3 100644 +--- a/dpdk/drivers/net/bnxt/bnxt_rxq.h ++++ b/dpdk/drivers/net/bnxt/bnxt_rxq.h +@@ -65,4 +65,5 @@ int bnxt_rx_queue_stop(struct rte_eth_dev *dev, + void bnxt_rx_queue_release_mbufs(struct bnxt_rx_queue *rxq); + int bnxt_need_agg_ring(struct rte_eth_dev *eth_dev); + void bnxt_free_rxq_mem(struct bnxt_rx_queue *rxq); ++uint64_t bnxt_get_rx_port_offloads(struct bnxt *bp); + #endif +diff --git a/dpdk/drivers/net/bnxt/bnxt_rxr.c b/dpdk/drivers/net/bnxt/bnxt_rxr.c +index 44247d7200..b60c2470f3 100644 +--- a/dpdk/drivers/net/bnxt/bnxt_rxr.c ++++ b/dpdk/drivers/net/bnxt/bnxt_rxr.c +@@ -824,6 +824,9 @@ void bnxt_set_mark_in_mbuf(struct bnxt *bp, + { + uint32_t cfa_code = 0; + ++ if (unlikely(bp->mark_table == NULL)) ++ return; ++ + cfa_code = rte_le_to_cpu_16(rxcmp1->cfa_code); + if (!cfa_code) + return; +@@ -1408,6 +1411,9 @@ int bnxt_flush_rx_cmp(struct bnxt_cp_ring_info *cpr) + cons = RING_CMP(cpr->cp_ring_struct, raw_cons); + rxcmp = (struct rx_pkt_cmpl *)&cpr->cp_desc_ring[cons]; + ++ if (!bnxt_cpr_cmp_valid(rxcmp, raw_cons, ring_mask + 1)) ++ break; ++ + if (CMP_TYPE(rxcmp) == CMPL_BASE_TYPE_HWRM_DONE) + return 1; + +diff --git a/dpdk/drivers/net/bnxt/bnxt_rxtx_vec_neon.c b/dpdk/drivers/net/bnxt/bnxt_rxtx_vec_neon.c +index f15e2d3b4e..611fbadb08 100644 +--- a/dpdk/drivers/net/bnxt/bnxt_rxtx_vec_neon.c ++++ b/dpdk/drivers/net/bnxt/bnxt_rxtx_vec_neon.c +@@ -235,25 +235,38 @@ recv_burst_vec_neon(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t nb_pkts) + } + + /* +- * Load the four current descriptors into SSE registers in +- * reverse order to ensure consistent state. ++ * Load the four current descriptors into NEON registers. ++ * IO barriers are used to ensure consistent state. + */ + rxcmp1[3] = vld1q_u32((void *)&cpr->cp_desc_ring[cons + 7]); + rte_io_rmb(); ++ /* Reload lower 64b of descriptors to make it ordered after info3_v. */ ++ rxcmp1[3] = vreinterpretq_u32_u64(vld1q_lane_u64 ++ ((void *)&cpr->cp_desc_ring[cons + 7], ++ vreinterpretq_u64_u32(rxcmp1[3]), 0)); + rxcmp[3] = vld1q_u32((void *)&cpr->cp_desc_ring[cons + 6]); + + rxcmp1[2] = vld1q_u32((void *)&cpr->cp_desc_ring[cons + 5]); + rte_io_rmb(); ++ rxcmp1[2] = vreinterpretq_u32_u64(vld1q_lane_u64 ++ ((void *)&cpr->cp_desc_ring[cons + 5], ++ vreinterpretq_u64_u32(rxcmp1[2]), 0)); + rxcmp[2] = vld1q_u32((void *)&cpr->cp_desc_ring[cons + 4]); + + t1 = vreinterpretq_u64_u32(vzip2q_u32(rxcmp1[2], rxcmp1[3])); + + rxcmp1[1] = vld1q_u32((void *)&cpr->cp_desc_ring[cons + 3]); + rte_io_rmb(); ++ rxcmp1[1] = vreinterpretq_u32_u64(vld1q_lane_u64 ++ ((void *)&cpr->cp_desc_ring[cons + 3], ++ vreinterpretq_u64_u32(rxcmp1[1]), 0)); + rxcmp[1] = vld1q_u32((void *)&cpr->cp_desc_ring[cons + 2]); + + rxcmp1[0] = vld1q_u32((void *)&cpr->cp_desc_ring[cons + 1]); + rte_io_rmb(); ++ rxcmp1[0] = vreinterpretq_u32_u64(vld1q_lane_u64 ++ ((void *)&cpr->cp_desc_ring[cons + 1], ++ vreinterpretq_u64_u32(rxcmp1[0]), 0)); + rxcmp[0] = vld1q_u32((void *)&cpr->cp_desc_ring[cons + 0]); + + t0 = vreinterpretq_u64_u32(vzip2q_u32(rxcmp1[0], rxcmp1[1])); +diff --git a/dpdk/drivers/net/bnxt/bnxt_stats.c b/dpdk/drivers/net/bnxt/bnxt_stats.c +index 991eafc644..208aa5616d 100644 +--- a/dpdk/drivers/net/bnxt/bnxt_stats.c ++++ b/dpdk/drivers/net/bnxt/bnxt_stats.c +@@ -741,7 +741,7 @@ int bnxt_dev_xstats_get_op(struct rte_eth_dev *eth_dev, + (bp->fw_tx_port_stats_ext_size / + stat_size)); + +- memset(xstats, 0, sizeof(*xstats)); ++ memset(xstats, 0, sizeof(*xstats) * n); + + count = 0; + for (i = 0; i < RTE_DIM(bnxt_rx_stats_strings); i++) { +@@ -846,7 +846,7 @@ int bnxt_flow_stats_cnt(struct bnxt *bp) + + int bnxt_dev_xstats_get_names_op(struct rte_eth_dev *eth_dev, + struct rte_eth_xstat_name *xstats_names, +- __rte_unused unsigned int limit) ++ unsigned int size) + { + struct bnxt *bp = (struct bnxt *)eth_dev->data->dev_private; + const unsigned int stat_cnt = RTE_DIM(bnxt_rx_stats_strings) + +@@ -862,63 +862,62 @@ int bnxt_dev_xstats_get_names_op(struct rte_eth_dev *eth_dev, + if (rc) + return rc; + +- if (xstats_names != NULL) { +- count = 0; ++ if (xstats_names == NULL || size < stat_cnt) ++ return stat_cnt; + +- for (i = 0; i < RTE_DIM(bnxt_rx_stats_strings); i++) { +- strlcpy(xstats_names[count].name, +- bnxt_rx_stats_strings[i].name, +- sizeof(xstats_names[count].name)); +- count++; +- } ++ for (i = 0; i < RTE_DIM(bnxt_rx_stats_strings); i++) { ++ strlcpy(xstats_names[count].name, ++ bnxt_rx_stats_strings[i].name, ++ sizeof(xstats_names[count].name)); ++ count++; ++ } + +- for (i = 0; i < RTE_DIM(bnxt_tx_stats_strings); i++) { +- strlcpy(xstats_names[count].name, +- bnxt_tx_stats_strings[i].name, +- sizeof(xstats_names[count].name)); +- count++; +- } ++ for (i = 0; i < RTE_DIM(bnxt_tx_stats_strings); i++) { ++ strlcpy(xstats_names[count].name, ++ bnxt_tx_stats_strings[i].name, ++ sizeof(xstats_names[count].name)); ++ count++; ++ } + +- for (i = 0; i < RTE_DIM(bnxt_func_stats_strings); i++) { +- strlcpy(xstats_names[count].name, +- bnxt_func_stats_strings[i].name, +- sizeof(xstats_names[count].name)); +- count++; +- } ++ for (i = 0; i < RTE_DIM(bnxt_func_stats_strings); i++) { ++ strlcpy(xstats_names[count].name, ++ bnxt_func_stats_strings[i].name, ++ sizeof(xstats_names[count].name)); ++ count++; ++ } + +- for (i = 0; i < RTE_DIM(bnxt_rx_ext_stats_strings); i++) { +- strlcpy(xstats_names[count].name, +- bnxt_rx_ext_stats_strings[i].name, +- sizeof(xstats_names[count].name)); ++ for (i = 0; i < RTE_DIM(bnxt_rx_ext_stats_strings); i++) { ++ strlcpy(xstats_names[count].name, ++ bnxt_rx_ext_stats_strings[i].name, ++ sizeof(xstats_names[count].name)); + +- count++; +- } ++ count++; ++ } + +- for (i = 0; i < RTE_DIM(bnxt_tx_ext_stats_strings); i++) { +- strlcpy(xstats_names[count].name, +- bnxt_tx_ext_stats_strings[i].name, +- sizeof(xstats_names[count].name)); ++ for (i = 0; i < RTE_DIM(bnxt_tx_ext_stats_strings); i++) { ++ strlcpy(xstats_names[count].name, ++ bnxt_tx_ext_stats_strings[i].name, ++ sizeof(xstats_names[count].name)); + +- count++; +- } ++ count++; ++ } + +- if (bp->fw_cap & BNXT_FW_CAP_ADV_FLOW_COUNTERS && +- bp->fw_cap & BNXT_FW_CAP_ADV_FLOW_MGMT && +- BNXT_FLOW_XSTATS_EN(bp)) { +- for (i = 0; i < bp->max_l2_ctx; i++) { +- char buf[RTE_ETH_XSTATS_NAME_SIZE]; ++ if (bp->fw_cap & BNXT_FW_CAP_ADV_FLOW_COUNTERS && ++ bp->fw_cap & BNXT_FW_CAP_ADV_FLOW_MGMT && ++ BNXT_FLOW_XSTATS_EN(bp)) { ++ for (i = 0; i < bp->max_l2_ctx; i++) { ++ char buf[RTE_ETH_XSTATS_NAME_SIZE]; + +- sprintf(buf, "flow_%d_bytes", i); +- strlcpy(xstats_names[count].name, buf, +- sizeof(xstats_names[count].name)); +- count++; ++ sprintf(buf, "flow_%d_bytes", i); ++ strlcpy(xstats_names[count].name, buf, ++ sizeof(xstats_names[count].name)); ++ count++; + +- sprintf(buf, "flow_%d_packets", i); +- strlcpy(xstats_names[count].name, buf, +- sizeof(xstats_names[count].name)); ++ sprintf(buf, "flow_%d_packets", i); ++ strlcpy(xstats_names[count].name, buf, ++ sizeof(xstats_names[count].name)); + +- count++; +- } ++ count++; + } + } + +diff --git a/dpdk/drivers/net/bnxt/bnxt_txq.c b/dpdk/drivers/net/bnxt/bnxt_txq.c +index 72a55ea643..c8745add5e 100644 +--- a/dpdk/drivers/net/bnxt/bnxt_txq.c ++++ b/dpdk/drivers/net/bnxt/bnxt_txq.c +@@ -17,6 +17,35 @@ + * TX Queues + */ + ++uint64_t bnxt_get_tx_port_offloads(struct bnxt *bp) +{ -+ return lldp ? lldp->enabled : false; ++ uint64_t tx_offload_capa; ++ ++ tx_offload_capa = RTE_ETH_TX_OFFLOAD_IPV4_CKSUM | ++ RTE_ETH_TX_OFFLOAD_UDP_CKSUM | ++ RTE_ETH_TX_OFFLOAD_TCP_CKSUM | ++ RTE_ETH_TX_OFFLOAD_TCP_TSO | ++ RTE_ETH_TX_OFFLOAD_QINQ_INSERT | ++ RTE_ETH_TX_OFFLOAD_MULTI_SEGS; ++ ++ if (bp->fw_cap & BNXT_FW_CAP_VLAN_TX_INSERT) ++ tx_offload_capa |= RTE_ETH_TX_OFFLOAD_VLAN_INSERT; ++ ++ if (BNXT_TUNNELED_OFFLOADS_CAP_ALL_EN(bp)) ++ tx_offload_capa |= RTE_ETH_TX_OFFLOAD_OUTER_IPV4_CKSUM; ++ ++ if (BNXT_TUNNELED_OFFLOADS_CAP_VXLAN_EN(bp)) ++ tx_offload_capa |= RTE_ETH_TX_OFFLOAD_VXLAN_TNL_TSO; ++ if (BNXT_TUNNELED_OFFLOADS_CAP_GRE_EN(bp)) ++ tx_offload_capa |= RTE_ETH_TX_OFFLOAD_GRE_TNL_TSO; ++ if (BNXT_TUNNELED_OFFLOADS_CAP_NGE_EN(bp)) ++ tx_offload_capa |= RTE_ETH_TX_OFFLOAD_GENEVE_TNL_TSO; ++ if (BNXT_TUNNELED_OFFLOADS_CAP_IPINIP_EN(bp)) ++ tx_offload_capa |= RTE_ETH_TX_OFFLOAD_IPIP_TNL_TSO; ++ ++ return tx_offload_capa; +} + - /* Configures the LLDP stack. - */ - bool -@@ -953,8 +961,8 @@ lldp_ref(const struct lldp *lldp_) - void - lldp_destroy_dummy(struct lldp *lldp) + void bnxt_free_txq_stats(struct bnxt_tx_queue *txq) { -- struct lldpd_hardware *hw, *hw_next; -- struct lldpd_chassis *chassis, *chassis_next; -+ struct lldpd_hardware *hw; -+ struct lldpd_chassis *chassis; - struct lldpd *cfg; + if (txq && txq->cp_ring && txq->cp_ring->hw_stats) +diff --git a/dpdk/drivers/net/bnxt/bnxt_txq.h b/dpdk/drivers/net/bnxt/bnxt_txq.h +index 67fd4cbebb..f3a03812ad 100644 +--- a/dpdk/drivers/net/bnxt/bnxt_txq.h ++++ b/dpdk/drivers/net/bnxt/bnxt_txq.h +@@ -43,4 +43,5 @@ int bnxt_tx_queue_setup_op(struct rte_eth_dev *eth_dev, + uint16_t nb_desc, + unsigned int socket_id, + const struct rte_eth_txconf *tx_conf); ++uint64_t bnxt_get_tx_port_offloads(struct bnxt *bp); + #endif +diff --git a/dpdk/drivers/net/bnxt/bnxt_txr.c b/dpdk/drivers/net/bnxt/bnxt_txr.c +index e2b7e40571..3b8f2382f9 100644 +--- a/dpdk/drivers/net/bnxt/bnxt_txr.c ++++ b/dpdk/drivers/net/bnxt/bnxt_txr.c +@@ -602,6 +602,9 @@ int bnxt_flush_tx_cmp(struct bnxt_cp_ring_info *cpr) + cons = RING_CMPL(ring_mask, raw_cons); + txcmp = (struct tx_cmpl *)&cp_desc_ring[cons]; + ++ if (!bnxt_cpr_cmp_valid(txcmp, raw_cons, ring_mask + 1)) ++ break; ++ + opaque = rte_cpu_to_le_32(txcmp->opaque); + raw_cons = NEXT_RAW_CMP(raw_cons); + +diff --git a/dpdk/drivers/net/bnxt/bnxt_vnic.c b/dpdk/drivers/net/bnxt/bnxt_vnic.c +index c63cf4b943..b3c03a2af5 100644 +--- a/dpdk/drivers/net/bnxt/bnxt_vnic.c ++++ b/dpdk/drivers/net/bnxt/bnxt_vnic.c +@@ -98,23 +98,16 @@ void bnxt_free_vnic_attributes(struct bnxt *bp) + + for (i = 0; i < bp->max_vnics; i++) { + vnic = &bp->vnic_info[i]; +- if (vnic->rss_table) { +- /* 'Unreserve' the rss_table */ +- /* N/A */ +- +- vnic->rss_table = NULL; +- } +- +- if (vnic->rss_hash_key) { +- /* 'Unreserve' the rss_hash_key */ +- /* N/A */ +- ++ if (vnic->rss_mz != NULL) { ++ rte_memzone_free(vnic->rss_mz); ++ vnic->rss_mz = NULL; + vnic->rss_hash_key = NULL; ++ vnic->rss_table = NULL; + } + } + } - if (!lldp) { -@@ -963,13 +971,13 @@ lldp_destroy_dummy(struct lldp *lldp) +-int bnxt_alloc_vnic_attributes(struct bnxt *bp) ++int bnxt_alloc_vnic_attributes(struct bnxt *bp, bool reconfig) + { + struct bnxt_vnic_info *vnic; + struct rte_pci_device *pdev = bp->pdev; +@@ -122,12 +115,10 @@ int bnxt_alloc_vnic_attributes(struct bnxt *bp) + char mz_name[RTE_MEMZONE_NAMESIZE]; + uint32_t entry_length; + size_t rss_table_size; +- uint16_t max_vnics; + int i; + rte_iova_t mz_phys_addr; + +- entry_length = HW_HASH_KEY_SIZE + +- BNXT_MAX_MC_ADDRS * RTE_ETHER_ADDR_LEN; ++ entry_length = HW_HASH_KEY_SIZE; + + if (BNXT_CHIP_P5(bp)) + rss_table_size = BNXT_RSS_TBL_SIZE_P5 * +@@ -137,43 +128,42 @@ int bnxt_alloc_vnic_attributes(struct bnxt *bp) + + entry_length = RTE_CACHE_LINE_ROUNDUP(entry_length + rss_table_size); + +- max_vnics = bp->max_vnics; +- snprintf(mz_name, RTE_MEMZONE_NAMESIZE, +- "bnxt_" PCI_PRI_FMT "_vnicattr", pdev->addr.domain, +- pdev->addr.bus, pdev->addr.devid, pdev->addr.function); +- mz_name[RTE_MEMZONE_NAMESIZE - 1] = 0; +- mz = rte_memzone_lookup(mz_name); +- if (!mz) { +- mz = rte_memzone_reserve(mz_name, +- entry_length * max_vnics, +- bp->eth_dev->device->numa_node, +- RTE_MEMZONE_2MB | +- RTE_MEMZONE_SIZE_HINT_ONLY | +- RTE_MEMZONE_IOVA_CONTIG); +- if (!mz) +- return -ENOMEM; +- } +- mz_phys_addr = mz->iova; +- +- for (i = 0; i < max_vnics; i++) { ++ for (i = 0; i < bp->max_vnics; i++) { + vnic = &bp->vnic_info[i]; + ++ snprintf(mz_name, RTE_MEMZONE_NAMESIZE, ++ "bnxt_" PCI_PRI_FMT "_vnicattr_%d", pdev->addr.domain, ++ pdev->addr.bus, pdev->addr.devid, pdev->addr.function, i); ++ mz_name[RTE_MEMZONE_NAMESIZE - 1] = 0; ++ mz = rte_memzone_lookup(mz_name); ++ if (mz == NULL) { ++ mz = rte_memzone_reserve(mz_name, ++ entry_length, ++ bp->eth_dev->device->numa_node, ++ RTE_MEMZONE_2MB | ++ RTE_MEMZONE_SIZE_HINT_ONLY | ++ RTE_MEMZONE_IOVA_CONTIG); ++ if (mz == NULL) { ++ PMD_DRV_LOG(ERR, "Cannot allocate bnxt vnic_attributes memory\n"); ++ return -ENOMEM; ++ } ++ } ++ vnic->rss_mz = mz; ++ mz_phys_addr = mz->iova; ++ + /* Allocate rss table and hash key */ +- vnic->rss_table = +- (void *)((char *)mz->addr + (entry_length * i)); ++ vnic->rss_table = (void *)((char *)mz->addr); ++ vnic->rss_table_dma_addr = mz_phys_addr; + memset(vnic->rss_table, -1, entry_length); + +- vnic->rss_table_dma_addr = mz_phys_addr + (entry_length * i); +- vnic->rss_hash_key = (void *)((char *)vnic->rss_table + +- rss_table_size); +- +- vnic->rss_hash_key_dma_addr = vnic->rss_table_dma_addr + +- rss_table_size; +- vnic->mc_list = (void *)((char *)vnic->rss_hash_key + +- HW_HASH_KEY_SIZE); +- vnic->mc_list_dma_addr = vnic->rss_hash_key_dma_addr + +- HW_HASH_KEY_SIZE; +- bnxt_prandom_bytes(vnic->rss_hash_key, HW_HASH_KEY_SIZE); ++ vnic->rss_hash_key = (void *)((char *)vnic->rss_table + rss_table_size); ++ vnic->rss_hash_key_dma_addr = vnic->rss_table_dma_addr + rss_table_size; ++ if (!reconfig) { ++ bnxt_prandom_bytes(vnic->rss_hash_key, HW_HASH_KEY_SIZE); ++ memcpy(bp->rss_conf.rss_key, vnic->rss_hash_key, HW_HASH_KEY_SIZE); ++ } else { ++ memcpy(vnic->rss_hash_key, bp->rss_conf.rss_key, HW_HASH_KEY_SIZE); ++ } + } - cfg = lldp->lldpd; + return 0; +diff --git a/dpdk/drivers/net/bnxt/bnxt_vnic.h b/dpdk/drivers/net/bnxt/bnxt_vnic.h +index 37b452f281..9055b93c4b 100644 +--- a/dpdk/drivers/net/bnxt/bnxt_vnic.h ++++ b/dpdk/drivers/net/bnxt/bnxt_vnic.h +@@ -28,14 +28,11 @@ struct bnxt_vnic_info { + uint16_t mru; + uint16_t hash_type; + uint8_t hash_mode; ++ const struct rte_memzone *rss_mz; + rte_iova_t rss_table_dma_addr; + uint16_t *rss_table; + rte_iova_t rss_hash_key_dma_addr; + void *rss_hash_key; +- rte_iova_t mc_list_dma_addr; +- char *mc_list; +- uint32_t mc_addr_cnt; +-#define BNXT_MAX_MC_ADDRS 16 + uint32_t flags; + #define BNXT_VNIC_INFO_PROMISC (1 << 0) + #define BNXT_VNIC_INFO_ALLMULTI (1 << 1) +@@ -64,7 +61,7 @@ int bnxt_free_vnic(struct bnxt *bp, struct bnxt_vnic_info *vnic, + struct bnxt_vnic_info *bnxt_alloc_vnic(struct bnxt *bp); + void bnxt_free_all_vnics(struct bnxt *bp); + void bnxt_free_vnic_attributes(struct bnxt *bp); +-int bnxt_alloc_vnic_attributes(struct bnxt *bp); ++int bnxt_alloc_vnic_attributes(struct bnxt *bp, bool reconfig); + void bnxt_free_vnic_mem(struct bnxt *bp); + int bnxt_alloc_vnic_mem(struct bnxt *bp); + int bnxt_vnic_grp_alloc(struct bnxt *bp, struct bnxt_vnic_info *vnic); +diff --git a/dpdk/drivers/net/bnxt/hsi_struct_def_dpdk.h b/dpdk/drivers/net/bnxt/hsi_struct_def_dpdk.h +index 88624f8129..f1fd5c8c6e 100644 +--- a/dpdk/drivers/net/bnxt/hsi_struct_def_dpdk.h ++++ b/dpdk/drivers/net/bnxt/hsi_struct_def_dpdk.h +@@ -12221,7 +12221,66 @@ struct hwrm_func_qcaps_output { + * function call for allocating Key Contexts. + */ + uint16_t max_key_ctxs_alloc; +- uint8_t unused_1[7]; ++ uint32_t flags_ext2; ++ /* ++ * When this bit is '1', it indicates that FW will support ++ * timestamping on all RX packets, not just PTP type packets. ++ */ ++ #define HWRM_FUNC_QCAPS_OUTPUT_FLAGS_EXT2_RX_ALL_PKTS_TIMESTAMPS_SUPPORTED \ ++ UINT32_C(0x1) ++ /* When this bit is '1', it indicates that HW and FW support QUIC. */ ++ #define HWRM_FUNC_QCAPS_OUTPUT_FLAGS_EXT2_QUIC_SUPPORTED \ ++ UINT32_C(0x2) ++ uint16_t tunnel_disable_flag; ++ /* ++ * When this bit is '1', it indicates that the VXLAN parsing ++ * is disabled in hardware ++ */ ++ #define HWRM_FUNC_QCAPS_OUTPUT_TUNNEL_DISABLE_FLAG_DISABLE_VXLAN \ ++ UINT32_C(0x1) ++ /* ++ * When this bit is '1', it indicates that the NGE parsing ++ * is disabled in hardware ++ */ ++ #define HWRM_FUNC_QCAPS_OUTPUT_TUNNEL_DISABLE_FLAG_DISABLE_NGE \ ++ UINT32_C(0x2) ++ /* ++ * When this bit is '1', it indicates that the NVGRE parsing ++ * is disabled in hardware ++ */ ++ #define HWRM_FUNC_QCAPS_OUTPUT_TUNNEL_DISABLE_FLAG_DISABLE_NVGRE \ ++ UINT32_C(0x4) ++ /* ++ * When this bit is '1', it indicates that the L2GRE parsing ++ * is disabled in hardware ++ */ ++ #define HWRM_FUNC_QCAPS_OUTPUT_TUNNEL_DISABLE_FLAG_DISABLE_L2GRE \ ++ UINT32_C(0x8) ++ /* ++ * When this bit is '1', it indicates that the GRE parsing ++ * is disabled in hardware ++ */ ++ #define HWRM_FUNC_QCAPS_OUTPUT_TUNNEL_DISABLE_FLAG_DISABLE_GRE \ ++ UINT32_C(0x10) ++ /* ++ * When this bit is '1', it indicates that the IPINIP parsing ++ * is disabled in hardware ++ */ ++ #define HWRM_FUNC_QCAPS_OUTPUT_TUNNEL_DISABLE_FLAG_DISABLE_IPINIP \ ++ UINT32_C(0x20) ++ /* ++ * When this bit is '1', it indicates that the MPLS parsing ++ * is disabled in hardware ++ */ ++ #define HWRM_FUNC_QCAPS_OUTPUT_TUNNEL_DISABLE_FLAG_DISABLE_MPLS \ ++ UINT32_C(0x40) ++ /* ++ * When this bit is '1', it indicates that the PPPOE parsing ++ * is disabled in hardware ++ */ ++ #define HWRM_FUNC_QCAPS_OUTPUT_TUNNEL_DISABLE_FLAG_DISABLE_PPPOE \ ++ UINT32_C(0x80) ++ uint8_t unused_1; + /* + * This field is used in Output records to indicate that the output + * is completely written to RAM. This field should be read as '1' +diff --git a/dpdk/drivers/net/bnxt/tf_core/tf_session.c b/dpdk/drivers/net/bnxt/tf_core/tf_session.c +index 9f849a0a76..c30c0e7029 100644 +--- a/dpdk/drivers/net/bnxt/tf_core/tf_session.c ++++ b/dpdk/drivers/net/bnxt/tf_core/tf_session.c +@@ -230,10 +230,12 @@ tf_session_create(struct tf *tfp, + "FW Session close failed, rc:%s\n", + strerror(-rc)); + } ++ if (tfp->session) { ++ tfp_free(tfp->session->core_data); ++ tfp_free(tfp->session); ++ tfp->session = NULL; ++ } + +- tfp_free(tfp->session->core_data); +- tfp_free(tfp->session); +- tfp->session = NULL; + return rc; + } -- LIST_FOR_EACH_SAFE (hw, hw_next, h_entries, &cfg->g_hardware) { -+ LIST_FOR_EACH_SAFE (hw, h_entries, &cfg->g_hardware) { - ovs_list_remove(&hw->h_entries); - free(hw->h_lport.p_lastframe); - free(hw); - } +diff --git a/dpdk/drivers/net/bnxt/tf_core/tfp.c b/dpdk/drivers/net/bnxt/tf_core/tfp.c +index a4b0934610..a967a9ccf2 100644 +--- a/dpdk/drivers/net/bnxt/tf_core/tfp.c ++++ b/dpdk/drivers/net/bnxt/tf_core/tfp.c +@@ -52,7 +52,7 @@ tfp_send_msg_direct(struct bnxt *bp, + } -- LIST_FOR_EACH_SAFE (chassis, chassis_next, list, &cfg->g_chassis) { -+ LIST_FOR_EACH_SAFE (chassis, list, &cfg->g_chassis) { - ovs_list_remove(&chassis->list); - free(chassis); - } -diff --git a/lib/ovs-lldp.h b/lib/ovs-lldp.h -index 0e536e8c27..661ac4e18a 100644 ---- a/lib/ovs-lldp.h -+++ b/lib/ovs-lldp.h -@@ -86,6 +86,7 @@ void lldp_run(struct lldpd *cfg); - bool lldp_should_send_packet(struct lldp *cfg); - bool lldp_should_process_flow(struct lldp *lldp, const struct flow *flow); - bool lldp_configure(struct lldp *lldp, const struct smap *cfg); -+bool lldp_is_enabled(struct lldp *lldp); - void lldp_process_packet(struct lldp *cfg, const struct dp_packet *); - void lldp_put_packet(struct lldp *lldp, struct dp_packet *packet, - const struct eth_addr eth_src); -diff --git a/lib/ovs-numa.h b/lib/ovs-numa.h -index ecc251a7ff..83bd10cca5 100644 ---- a/lib/ovs-numa.h -+++ b/lib/ovs-numa.h -@@ -68,9 +68,9 @@ void ovs_numa_dump_destroy(struct ovs_numa_dump *); - int ovs_numa_thread_setaffinity_core(unsigned core_id); + /** +- * Allocates zero'ed memory from the heap. ++ * Allocates zeroed memory from the heap. + * + * Returns success or failure code. + */ +diff --git a/dpdk/drivers/net/bnxt/tf_core/tfp.h b/dpdk/drivers/net/bnxt/tf_core/tfp.h +index dd0a347058..5a99c7a06e 100644 +--- a/dpdk/drivers/net/bnxt/tf_core/tfp.h ++++ b/dpdk/drivers/net/bnxt/tf_core/tfp.h +@@ -150,7 +150,7 @@ tfp_msg_hwrm_oem_cmd(struct tf *tfp, + uint32_t max_flows); + + /** +- * Allocates zero'ed memory from the heap. ++ * Allocates zeroed memory from the heap. + * + * NOTE: Also performs virt2phy address conversion by default thus is + * can be expensive to invoke. +diff --git a/dpdk/drivers/net/bnxt/tf_ulp/ulp_rte_parser.c b/dpdk/drivers/net/bnxt/tf_ulp/ulp_rte_parser.c +index f4274dd634..9edf3e8799 100644 +--- a/dpdk/drivers/net/bnxt/tf_ulp/ulp_rte_parser.c ++++ b/dpdk/drivers/net/bnxt/tf_ulp/ulp_rte_parser.c +@@ -1096,7 +1096,7 @@ ulp_rte_ipv4_hdr_handler(const struct rte_flow_item *item, + hdr.fragment_offset), + ulp_deference_struct(ipv4_mask, + hdr.fragment_offset), +- ULP_PRSR_ACT_DEFAULT); ++ ULP_PRSR_ACT_MASK_IGNORE); + + size = sizeof(((struct rte_flow_item_ipv4 *)NULL)->hdr.time_to_live); + ulp_rte_prsr_fld_mask(params, &idx, size, +diff --git a/dpdk/drivers/net/bonding/eth_bond_8023ad_private.h b/dpdk/drivers/net/bonding/eth_bond_8023ad_private.h +index 9b5738afee..a5e1fffea1 100644 +--- a/dpdk/drivers/net/bonding/eth_bond_8023ad_private.h ++++ b/dpdk/drivers/net/bonding/eth_bond_8023ad_private.h +@@ -20,7 +20,7 @@ + /** Maximum number of LACP packets from one slave queued in TX ring. */ + #define BOND_MODE_8023AX_SLAVE_TX_PKTS 1 + /** +- * Timeouts deffinitions (5.4.4 in 802.1AX documentation). ++ * Timeouts definitions (5.4.4 in 802.1AX documentation). + */ + #define BOND_8023AD_FAST_PERIODIC_MS 900 + #define BOND_8023AD_SLOW_PERIODIC_MS 29000 +diff --git a/dpdk/drivers/net/bonding/eth_bond_private.h b/dpdk/drivers/net/bonding/eth_bond_private.h +index 8b104b6391..8222e3cd38 100644 +--- a/dpdk/drivers/net/bonding/eth_bond_private.h ++++ b/dpdk/drivers/net/bonding/eth_bond_private.h +@@ -139,7 +139,7 @@ struct bond_dev_private { + + uint16_t slave_count; /**< Number of bonded slaves */ + struct bond_slave_details slaves[RTE_MAX_ETHPORTS]; +- /**< Arary of bonded slaves details */ ++ /**< Array of bonded slaves details */ + + struct mode8023ad_private mode4; + uint16_t tlb_slaves_order[RTE_MAX_ETHPORTS]; +@@ -240,12 +240,16 @@ slave_remove_mac_addresses(struct rte_eth_dev *bonded_eth_dev, + uint16_t slave_port_id); - #define FOR_EACH_CORE_ON_DUMP(ITER, DUMP) \ -- HMAP_FOR_EACH((ITER), hmap_node, &(DUMP)->cores) -+ HMAP_FOR_EACH (ITER, hmap_node, &(DUMP)->cores) + int +-bond_ethdev_mode_set(struct rte_eth_dev *eth_dev, int mode); ++bond_ethdev_mode_set(struct rte_eth_dev *eth_dev, uint8_t mode); - #define FOR_EACH_NUMA_ON_DUMP(ITER, DUMP) \ -- HMAP_FOR_EACH((ITER), hmap_node, &(DUMP)->numas) -+ HMAP_FOR_EACH (ITER, hmap_node, &(DUMP)->numas) + int + slave_configure(struct rte_eth_dev *bonded_eth_dev, + struct rte_eth_dev *slave_eth_dev); - #endif /* ovs-numa.h */ -diff --git a/lib/ovs-rcu.c b/lib/ovs-rcu.c -index 1866bd3088..946aa04d18 100644 ---- a/lib/ovs-rcu.c -+++ b/lib/ovs-rcu.c -@@ -444,3 +444,40 @@ ovsrcu_init_module(void) - ovsthread_once_done(&once); - } ++int ++slave_start(struct rte_eth_dev *bonded_eth_dev, ++ struct rte_eth_dev *slave_eth_dev); ++ + void + slave_remove(struct bond_dev_private *internals, + struct rte_eth_dev *slave_eth_dev); +diff --git a/dpdk/drivers/net/bonding/rte_eth_bond_8023ad.c b/dpdk/drivers/net/bonding/rte_eth_bond_8023ad.c +index ca50583d62..b3cddd8a20 100644 +--- a/dpdk/drivers/net/bonding/rte_eth_bond_8023ad.c ++++ b/dpdk/drivers/net/bonding/rte_eth_bond_8023ad.c +@@ -243,7 +243,7 @@ record_default(struct port *port) + { + /* Record default parameters for partner. Partner admin parameters + * are not implemented so set them to arbitrary default (last known) and +- * mark actor that parner is in defaulted state. */ ++ * mark actor that partner is in defaulted state. */ + port->partner_state = STATE_LACP_ACTIVE; + ACTOR_STATE_SET(port, DEFAULTED); + } +@@ -300,7 +300,7 @@ rx_machine(struct bond_dev_private *internals, uint16_t slave_id, + MODE4_DEBUG("LACP -> CURRENT\n"); + BOND_PRINT_LACP(lacp); + /* Update selected flag. If partner parameters are defaulted assume they +- * are match. If not defaulted compare LACP actor with ports parner ++ * are match. If not defaulted compare LACP actor with ports partner + * params. */ + if (!ACTOR_STATE(port, DEFAULTED) && + (ACTOR_STATE(port, AGGREGATION) != PARTNER_STATE(port, AGGREGATION) +@@ -399,16 +399,16 @@ periodic_machine(struct bond_dev_private *internals, uint16_t slave_id) + PARTNER_STATE(port, LACP_ACTIVE); + + uint8_t is_partner_fast, was_partner_fast; +- /* No periodic is on BEGIN, LACP DISABLE or when both sides are pasive */ ++ /* No periodic is on BEGIN, LACP DISABLE or when both sides are passive */ + if (SM_FLAG(port, BEGIN) || !SM_FLAG(port, LACP_ENABLED) || !active) { + timer_cancel(&port->periodic_timer); + timer_force_expired(&port->tx_machine_timer); + SM_FLAG_CLR(port, PARTNER_SHORT_TIMEOUT); + + MODE4_DEBUG("-> NO_PERIODIC ( %s%s%s)\n", +- SM_FLAG(port, BEGIN) ? "begind " : "", ++ SM_FLAG(port, BEGIN) ? "begin " : "", + SM_FLAG(port, LACP_ENABLED) ? "" : "LACP disabled ", +- active ? "LACP active " : "LACP pasive "); ++ active ? "LACP active " : "LACP passive "); + return; + } + +@@ -495,10 +495,10 @@ mux_machine(struct bond_dev_private *internals, uint16_t slave_id) + if ((ACTOR_STATE(port, DISTRIBUTING) || ACTOR_STATE(port, COLLECTING)) && + !PARTNER_STATE(port, SYNCHRONIZATION)) { + /* If in COLLECTING or DISTRIBUTING state and partner becomes out of +- * sync transit to ATACHED state. */ ++ * sync transit to ATTACHED state. */ + ACTOR_STATE_CLR(port, DISTRIBUTING); + ACTOR_STATE_CLR(port, COLLECTING); +- /* Clear actor sync to activate transit ATACHED in condition bellow */ ++ /* Clear actor sync to activate transit ATTACHED in condition bellow */ + ACTOR_STATE_CLR(port, SYNCHRONIZATION); + MODE4_DEBUG("Out of sync -> ATTACHED\n"); + } +@@ -696,7 +696,7 @@ selection_logic(struct bond_dev_private *internals, uint16_t slave_id) + /* Search for aggregator suitable for this port */ + for (i = 0; i < slaves_count; ++i) { + agg = &bond_mode_8023ad_ports[slaves[i]]; +- /* Skip ports that are not aggreagators */ ++ /* Skip ports that are not aggregators */ + if (agg->aggregator_port_id != slaves[i]) + continue; + +@@ -921,7 +921,7 @@ bond_mode_8023ad_periodic_cb(void *arg) + + SM_FLAG_SET(port, BEGIN); + +- /* LACP is disabled on half duples or link is down */ ++ /* LACP is disabled on half duplex or link is down */ + if (SM_FLAG(port, LACP_ENABLED)) { + /* If port was enabled set it to BEGIN state */ + SM_FLAG_CLR(port, LACP_ENABLED); +@@ -1069,7 +1069,7 @@ bond_mode_8023ad_activate_slave(struct rte_eth_dev *bond_dev, + port->partner_state = STATE_LACP_ACTIVE | STATE_AGGREGATION; + port->sm_flags = SM_FLAGS_BEGIN; + +- /* use this port as agregator */ ++ /* use this port as aggregator */ + port->aggregator_port_id = slave_id; + + if (bond_mode_8023ad_register_lacp_mac(slave_id) < 0) { +diff --git a/dpdk/drivers/net/bonding/rte_eth_bond_8023ad.h b/dpdk/drivers/net/bonding/rte_eth_bond_8023ad.h +index 11a71a55e5..7eb392f8c8 100644 +--- a/dpdk/drivers/net/bonding/rte_eth_bond_8023ad.h ++++ b/dpdk/drivers/net/bonding/rte_eth_bond_8023ad.h +@@ -68,7 +68,7 @@ struct port_params { + struct rte_ether_addr system; + /**< System ID - Slave MAC address, same as bonding MAC address */ + uint16_t key; +- /**< Speed information (implementation dependednt) and duplex. */ ++ /**< Speed information (implementation dependent) and duplex. */ + uint16_t port_priority; + /**< Priority of this (unused in current implementation) */ + uint16_t port_number; +@@ -317,7 +317,7 @@ rte_eth_bond_8023ad_dedicated_queues_disable(uint16_t port_id); + * @param port_id Bonding device id + * + * @return +- * agregator mode on success, negative value otherwise ++ * aggregator mode on success, negative value otherwise + */ + int + rte_eth_bond_8023ad_agg_selection_get(uint16_t port_id); +diff --git a/dpdk/drivers/net/bonding/rte_eth_bond_alb.h b/dpdk/drivers/net/bonding/rte_eth_bond_alb.h +index 386e70c594..4e9aeda9bc 100644 +--- a/dpdk/drivers/net/bonding/rte_eth_bond_alb.h ++++ b/dpdk/drivers/net/bonding/rte_eth_bond_alb.h +@@ -96,7 +96,7 @@ bond_mode_alb_arp_xmit(struct rte_ether_hdr *eth_h, uint16_t offset, + * @param internals Bonding data. + * + * @return +- * Index of slawe on which packet should be sent. ++ * Index of slave on which packet should be sent. + */ + uint16_t + bond_mode_alb_arp_upd(struct client_data *client_info, +diff --git a/dpdk/drivers/net/bonding/rte_eth_bond_api.c b/dpdk/drivers/net/bonding/rte_eth_bond_api.c +index 84943cffe2..919c580fb8 100644 +--- a/dpdk/drivers/net/bonding/rte_eth_bond_api.c ++++ b/dpdk/drivers/net/bonding/rte_eth_bond_api.c +@@ -375,7 +375,7 @@ eth_bond_slave_inherit_dev_info_rx_next(struct bond_dev_private *internals, + * value. Thus, the new internal value of default Rx queue offloads + * has to be masked by rx_queue_offload_capa to make sure that only + * commonly supported offloads are preserved from both the previous +- * value and the value being inhereted from the new slave device. ++ * value and the value being inherited from the new slave device. + */ + rxconf_i->offloads = (rxconf_i->offloads | rxconf->offloads) & + internals->rx_queue_offload_capa; +@@ -413,7 +413,7 @@ eth_bond_slave_inherit_dev_info_tx_next(struct bond_dev_private *internals, + * value. Thus, the new internal value of default Tx queue offloads + * has to be masked by tx_queue_offload_capa to make sure that only + * commonly supported offloads are preserved from both the previous +- * value and the value being inhereted from the new slave device. ++ * value and the value being inherited from the new slave device. + */ + txconf_i->offloads = (txconf_i->offloads | txconf->offloads) & + internals->tx_queue_offload_capa; +@@ -566,6 +566,12 @@ __eth_bond_slave_add_lock_free(uint16_t bonded_port_id, uint16_t slave_port_id) + slave_port_id); + return -1; + } ++ if (slave_start(bonded_eth_dev, slave_eth_dev) != 0) { ++ internals->slave_count--; ++ RTE_BOND_LOG(ERR, "rte_bond_slaves_start: port=%d", ++ slave_port_id); ++ return -1; ++ } + } + + /* Update all slave devices MACs */ +@@ -668,7 +674,7 @@ __eth_bond_slave_remove_lock_free(uint16_t bonded_port_id, + } + + if (slave_idx < 0) { +- RTE_BOND_LOG(ERR, "Couldn't find slave in port list, slave count %d", ++ RTE_BOND_LOG(ERR, "Couldn't find slave in port list, slave count %u", + internals->slave_count); + return -1; + } +diff --git a/dpdk/drivers/net/bonding/rte_eth_bond_pmd.c b/dpdk/drivers/net/bonding/rte_eth_bond_pmd.c +index 84f4900ee5..9b3acde46c 100644 +--- a/dpdk/drivers/net/bonding/rte_eth_bond_pmd.c ++++ b/dpdk/drivers/net/bonding/rte_eth_bond_pmd.c +@@ -1318,7 +1318,7 @@ bond_ethdev_tx_burst_broadcast(void *queue, struct rte_mbuf **bufs, + + /* Increment reference count on mbufs */ + for (i = 0; i < nb_pkts; i++) +- rte_mbuf_refcnt_update(bufs[i], num_of_slaves - 1); ++ rte_pktmbuf_refcnt_update(bufs[i], num_of_slaves - 1); + + /* Transmit burst on each active slave */ + for (i = 0; i < num_of_slaves; i++) { +@@ -1554,7 +1554,7 @@ mac_address_slaves_update(struct rte_eth_dev *bonded_eth_dev) } + + int +-bond_ethdev_mode_set(struct rte_eth_dev *eth_dev, int mode) ++bond_ethdev_mode_set(struct rte_eth_dev *eth_dev, uint8_t mode) + { + struct bond_dev_private *internals; + +@@ -1678,14 +1678,10 @@ int + slave_configure(struct rte_eth_dev *bonded_eth_dev, + struct rte_eth_dev *slave_eth_dev) + { +- struct bond_rx_queue *bd_rx_q; +- struct bond_tx_queue *bd_tx_q; + uint16_t nb_rx_queues; + uint16_t nb_tx_queues; + + int errval; +- uint16_t q_id; +- struct rte_flow_error flow_error; + + struct bond_dev_private *internals = bonded_eth_dev->data->dev_private; + +@@ -1711,19 +1707,32 @@ slave_configure(struct rte_eth_dev *bonded_eth_dev, + bonded_eth_dev->data->dev_conf.rx_adv_conf.rss_conf.rss_hf; + slave_eth_dev->data->dev_conf.rxmode.mq_mode = + bonded_eth_dev->data->dev_conf.rxmode.mq_mode; ++ } else { ++ slave_eth_dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key_len = 0; ++ slave_eth_dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key = NULL; ++ slave_eth_dev->data->dev_conf.rx_adv_conf.rss_conf.rss_hf = 0; ++ slave_eth_dev->data->dev_conf.rxmode.mq_mode = ++ bonded_eth_dev->data->dev_conf.rxmode.mq_mode; + } + +- if (bonded_eth_dev->data->dev_conf.rxmode.offloads & +- RTE_ETH_RX_OFFLOAD_VLAN_FILTER) +- slave_eth_dev->data->dev_conf.rxmode.offloads |= +- RTE_ETH_RX_OFFLOAD_VLAN_FILTER; +- else +- slave_eth_dev->data->dev_conf.rxmode.offloads &= +- ~RTE_ETH_RX_OFFLOAD_VLAN_FILTER; +- + slave_eth_dev->data->dev_conf.rxmode.mtu = + bonded_eth_dev->data->dev_conf.rxmode.mtu; + ++ slave_eth_dev->data->dev_conf.txmode.offloads |= ++ bonded_eth_dev->data->dev_conf.txmode.offloads; + -+static void -+ovsrcu_barrier_func(void *seq_) -+{ -+ struct seq *seq = (struct seq *) seq_; -+ seq_change(seq); -+} ++ slave_eth_dev->data->dev_conf.txmode.offloads &= ++ (bonded_eth_dev->data->dev_conf.txmode.offloads | ++ ~internals->tx_offload_capa); + -+/* Similar to the kernel rcu_barrier, ovsrcu_barrier waits for all outstanding -+ * RCU callbacks to complete. However, unlike the kernel rcu_barrier, which -+ * might return immediately if there are no outstanding RCU callbacks, -+ * this API will at least wait for a grace period. -+ * -+ * Another issue the caller might need to know is that the barrier is just -+ * for "one-shot", i.e. if inside some RCU callbacks, another RCU callback is -+ * registered, this API only guarantees the first round of RCU callbacks have -+ * been executed after it returns. -+ */ -+void -+ovsrcu_barrier(void) -+{ -+ struct seq *seq = seq_create(); -+ /* First let all threads flush their cbsets. */ -+ ovsrcu_synchronize(); ++ slave_eth_dev->data->dev_conf.rxmode.offloads |= ++ bonded_eth_dev->data->dev_conf.rxmode.offloads; + -+ /* Then register a new cbset, ensure this cbset -+ * is at the tail of the global list. */ -+ uint64_t seqno = seq_read(seq); -+ ovsrcu_postpone__(ovsrcu_barrier_func, (void *) seq); ++ slave_eth_dev->data->dev_conf.rxmode.offloads &= ++ (bonded_eth_dev->data->dev_conf.rxmode.offloads | ++ ~internals->rx_offload_capa); + -+ do { -+ seq_wait(seq, seqno); -+ poll_block(); -+ } while (seqno == seq_read(seq)); + -+ seq_destroy(seq); -+} -diff --git a/lib/ovs-rcu.h b/lib/ovs-rcu.h -index ecc4c92010..8b397b7fb0 100644 ---- a/lib/ovs-rcu.h -+++ b/lib/ovs-rcu.h -@@ -155,6 +155,19 @@ - * port_delete(id); - * } - * -+ * Use ovsrcu_barrier() to wait for all the outstanding RCU callbacks to -+ * finish. This is useful when you have to destroy some resources however -+ * these resources are referenced in the outstanding RCU callbacks. -+ * -+ * void rcu_cb(void *A) { -+ * do_something(A); -+ * } -+ * -+ * void destroy_A() { -+ * ovsrcu_postpone(rcu_cb, A); // will use A later -+ * ovsrcu_barrier(); // wait for rcu_cb done -+ * do_destroy_A(); // free A -+ * } - */ + nb_rx_queues = bonded_eth_dev->data->nb_rx_queues; + nb_tx_queues = bonded_eth_dev->data->nb_tx_queues; - #include "compiler.h" -@@ -310,4 +323,6 @@ void ovsrcu_synchronize(void); +@@ -1734,14 +1743,6 @@ slave_configure(struct rte_eth_dev *bonded_eth_dev, + } + } - void ovsrcu_exit(void); +- errval = rte_eth_dev_set_mtu(slave_eth_dev->data->port_id, +- bonded_eth_dev->data->mtu); +- if (errval != 0 && errval != -ENOTSUP) { +- RTE_BOND_LOG(ERR, "rte_eth_dev_set_mtu: port %u, err (%d)", +- slave_eth_dev->data->port_id, errval); +- return errval; +- } +- + /* Configure device */ + errval = rte_eth_dev_configure(slave_eth_dev->data->port_id, + nb_rx_queues, nb_tx_queues, +@@ -1752,6 +1753,27 @@ slave_configure(struct rte_eth_dev *bonded_eth_dev, + return errval; + } -+void ovsrcu_barrier(void); ++ errval = rte_eth_dev_set_mtu(slave_eth_dev->data->port_id, ++ bonded_eth_dev->data->mtu); ++ if (errval != 0 && errval != -ENOTSUP) { ++ RTE_BOND_LOG(ERR, "rte_eth_dev_set_mtu: port %u, err (%d)", ++ slave_eth_dev->data->port_id, errval); ++ return errval; ++ } ++ return 0; ++} + - #endif /* ovs-rcu.h */ -diff --git a/lib/ovs-router.c b/lib/ovs-router.c -index 09b81c6e5a..5d0fbd503e 100644 ---- a/lib/ovs-router.c -+++ b/lib/ovs-router.c -@@ -164,9 +164,10 @@ static void rt_init_match(struct match *match, uint32_t mark, - match->flow.pkt_mark = mark; - } - --static int --get_src_addr(const struct in6_addr *ip6_dst, -- const char output_bridge[], struct in6_addr *psrc) +int -+ovs_router_get_netdev_source_address(const struct in6_addr *ip6_dst, -+ const char output_bridge[], -+ struct in6_addr *psrc) - { - struct in6_addr *mask, *addr6; - int err, n_in6, i, max_plen = -1; -@@ -235,9 +236,11 @@ ovs_router_insert__(uint32_t mark, uint8_t priority, bool local, - p->plen = plen; - p->local = local; - p->priority = priority; -- err = get_src_addr(ip6_dst, output_bridge, &p->src_addr); -+ err = ovs_router_get_netdev_source_address(ip6_dst, output_bridge, -+ &p->src_addr); - if (err && ipv6_addr_is_set(gw)) { -- err = get_src_addr(gw, output_bridge, &p->src_addr); -+ err = ovs_router_get_netdev_source_address(gw, output_bridge, -+ &p->src_addr); - } - if (err) { - struct ds ds = DS_EMPTY_INITIALIZER; -diff --git a/lib/ovs-router.h b/lib/ovs-router.h -index 34ea163eef..d8ce3c00de 100644 ---- a/lib/ovs-router.h -+++ b/lib/ovs-router.h -@@ -37,6 +37,10 @@ void ovs_router_flush(void); ++slave_start(struct rte_eth_dev *bonded_eth_dev, ++ struct rte_eth_dev *slave_eth_dev) ++{ ++ int errval = 0; ++ struct bond_rx_queue *bd_rx_q; ++ struct bond_tx_queue *bd_tx_q; ++ uint16_t q_id; ++ struct rte_flow_error flow_error; ++ struct bond_dev_private *internals = bonded_eth_dev->data->dev_private; ++ + /* Setup Rx Queues */ + for (q_id = 0; q_id < bonded_eth_dev->data->nb_rx_queues; q_id++) { + bd_rx_q = (struct bond_rx_queue *)bonded_eth_dev->data->rx_queues[q_id]; +@@ -1799,10 +1821,13 @@ slave_configure(struct rte_eth_dev *bonded_eth_dev, + return errval; + } + +- if (internals->mode4.dedicated_queues.flow[slave_eth_dev->data->port_id] != NULL) +- rte_flow_destroy(slave_eth_dev->data->port_id, ++ if (internals->mode4.dedicated_queues.flow[slave_eth_dev->data->port_id] != NULL) { ++ errval = rte_flow_destroy(slave_eth_dev->data->port_id, + internals->mode4.dedicated_queues.flow[slave_eth_dev->data->port_id], + &flow_error); ++ RTE_BOND_LOG(ERR, "bond_ethdev_8023ad_flow_destroy: port=%d, err (%d)", ++ slave_eth_dev->data->port_id, errval); ++ } + + errval = bond_ethdev_8023ad_flow_set(bonded_eth_dev, + slave_eth_dev->data->port_id); +@@ -1994,6 +2019,13 @@ bond_ethdev_start(struct rte_eth_dev *eth_dev) + internals->slaves[i].port_id); + goto out_err; + } ++ if (slave_start(eth_dev, slave_ethdev) != 0) { ++ RTE_BOND_LOG(ERR, ++ "bonded port (%d) failed to start slave device (%d)", ++ eth_dev->data->port_id, ++ internals->slaves[i].port_id); ++ goto out_err; ++ } + /* We will need to poll for link status if any slave doesn't + * support interrupts + */ +@@ -2092,18 +2124,20 @@ bond_ethdev_stop(struct rte_eth_dev *eth_dev) + internals->link_status_polling_enabled = 0; + for (i = 0; i < internals->slave_count; i++) { + uint16_t slave_id = internals->slaves[i].port_id; ++ ++ internals->slaves[i].last_link_status = 0; ++ ret = rte_eth_dev_stop(slave_id); ++ if (ret != 0) { ++ RTE_BOND_LOG(ERR, "Failed to stop device on port %u", ++ slave_id); ++ return ret; ++ } ++ ++ /* active slaves need to be deactivated. */ + if (find_slave_by_id(internals->active_slaves, + internals->active_slave_count, slave_id) != +- internals->active_slave_count) { +- internals->slaves[i].last_link_status = 0; +- ret = rte_eth_dev_stop(slave_id); +- if (ret != 0) { +- RTE_BOND_LOG(ERR, "Failed to stop device on port %u", +- slave_id); +- return ret; +- } ++ internals->active_slave_count) + deactivate_slave(eth_dev, slave_id); +- } + } - void ovs_router_disable_system_routing_table(void); + return 0; +@@ -2128,6 +2162,7 @@ bond_ethdev_close(struct rte_eth_dev *dev) + RTE_BOND_LOG(ERR, "Failed to stop device on port %u", + port_id); + skipped++; ++ continue; + } + + if (rte_eth_bond_slave_remove(bond_port_id, port_id) != 0) { +@@ -2684,6 +2719,39 @@ bond_ethdev_promiscuous_disable(struct rte_eth_dev *dev) + return ret; + } -+int ovs_router_get_netdev_source_address(const struct in6_addr *ip6_dst, -+ const char output_bridge[], -+ struct in6_addr *psrc); ++static int ++bond_ethdev_promiscuous_update(struct rte_eth_dev *dev) ++{ ++ struct bond_dev_private *internals = dev->data->dev_private; ++ uint16_t port_id = internals->current_primary_port; ++ ++ switch (internals->mode) { ++ case BONDING_MODE_ROUND_ROBIN: ++ case BONDING_MODE_BALANCE: ++ case BONDING_MODE_BROADCAST: ++ case BONDING_MODE_8023AD: ++ /* As promiscuous mode is propagated to all slaves for these ++ * mode, no need to update for bonding device. ++ */ ++ break; ++ case BONDING_MODE_ACTIVE_BACKUP: ++ case BONDING_MODE_TLB: ++ case BONDING_MODE_ALB: ++ default: ++ /* As promiscuous mode is propagated only to primary slave ++ * for these mode. When active/standby switchover, promiscuous ++ * mode should be set to new primary slave according to bonding ++ * device. ++ */ ++ if (rte_eth_promiscuous_get(internals->port_id) == 1) ++ rte_eth_promiscuous_enable(port_id); ++ else ++ rte_eth_promiscuous_disable(port_id); ++ } ++ ++ return 0; ++} + - #ifdef __cplusplus - } - #endif -diff --git a/lib/ovsdb-cs.c b/lib/ovsdb-cs.c -index dead31275d..9713c7dc7c 100644 ---- a/lib/ovsdb-cs.c -+++ b/lib/ovsdb-cs.c -@@ -900,8 +900,8 @@ ovsdb_cs_db_get_table(struct ovsdb_cs_db *db, const char *table) - static void - ovsdb_cs_db_destroy_tables(struct ovsdb_cs_db *db) - { -- struct ovsdb_cs_db_table *table, *next; -- HMAP_FOR_EACH_SAFE (table, next, hmap_node, &db->tables) { -+ struct ovsdb_cs_db_table *table; -+ HMAP_FOR_EACH_SAFE (table, hmap_node, &db->tables) { - json_destroy(table->ack_cond); - json_destroy(table->req_cond); - json_destroy(table->new_cond); -@@ -1793,8 +1793,8 @@ ovsdb_cs_update_server_row(struct server_row *row, - static void - ovsdb_cs_clear_server_rows(struct ovsdb_cs *cs) + static int + bond_ethdev_allmulticast_enable(struct rte_eth_dev *eth_dev) { -- struct server_row *row, *next; -- HMAP_FOR_EACH_SAFE (row, next, hmap_node, &cs->server_rows) { -+ struct server_row *row; -+ HMAP_FOR_EACH_SAFE (row, hmap_node, &cs->server_rows) { - ovsdb_cs_delete_server_row(cs, row); - } +@@ -2797,6 +2865,39 @@ bond_ethdev_allmulticast_disable(struct rte_eth_dev *eth_dev) + return ret; } -@@ -2128,9 +2128,9 @@ void - ovsdb_cs_free_schema(struct shash *schema) + ++static int ++bond_ethdev_allmulticast_update(struct rte_eth_dev *dev) ++{ ++ struct bond_dev_private *internals = dev->data->dev_private; ++ uint16_t port_id = internals->current_primary_port; ++ ++ switch (internals->mode) { ++ case BONDING_MODE_ROUND_ROBIN: ++ case BONDING_MODE_BALANCE: ++ case BONDING_MODE_BROADCAST: ++ case BONDING_MODE_8023AD: ++ /* As allmulticast mode is propagated to all slaves for these ++ * mode, no need to update for bonding device. ++ */ ++ break; ++ case BONDING_MODE_ACTIVE_BACKUP: ++ case BONDING_MODE_TLB: ++ case BONDING_MODE_ALB: ++ default: ++ /* As allmulticast mode is propagated only to primary slave ++ * for these mode. When active/standby switchover, allmulticast ++ * mode should be set to new primary slave according to bonding ++ * device. ++ */ ++ if (rte_eth_allmulticast_get(internals->port_id) == 1) ++ rte_eth_allmulticast_enable(port_id); ++ else ++ rte_eth_allmulticast_disable(port_id); ++ } ++ ++ return 0; ++} ++ + static void + bond_ethdev_delayed_lsc_propagation(void *arg) { - if (schema) { -- struct shash_node *node, *next; -+ struct shash_node *node; +@@ -2886,6 +2987,8 @@ bond_ethdev_lsc_event_callback(uint16_t port_id, enum rte_eth_event_type type, + lsc_flag = 1; + + mac_address_slaves_update(bonded_eth_dev); ++ bond_ethdev_promiscuous_update(bonded_eth_dev); ++ bond_ethdev_allmulticast_update(bonded_eth_dev); + } + + activate_slave(bonded_eth_dev, port_id); +@@ -2915,6 +3018,8 @@ bond_ethdev_lsc_event_callback(uint16_t port_id, enum rte_eth_event_type type, + else + internals->current_primary_port = internals->primary_port; + mac_address_slaves_update(bonded_eth_dev); ++ bond_ethdev_promiscuous_update(bonded_eth_dev); ++ bond_ethdev_allmulticast_update(bonded_eth_dev); + } + } -- SHASH_FOR_EACH_SAFE (node, next, schema) { -+ SHASH_FOR_EACH_SAFE (node, schema) { - struct sset *sset = node->data; - sset_destroy(sset); - free(sset); -diff --git a/lib/ovsdb-data.c b/lib/ovsdb-data.c -index 6b1c20ff85..61ad7679a6 100644 ---- a/lib/ovsdb-data.c -+++ b/lib/ovsdb-data.c -@@ -1957,6 +1957,19 @@ ovsdb_datum_add_unsafe(struct ovsdb_datum *datum, - } +@@ -3293,7 +3398,7 @@ bond_alloc(struct rte_vdev_device *dev, uint8_t mode) + /* Set mode 4 default configuration */ + bond_mode_8023ad_setup(eth_dev, NULL); + if (bond_ethdev_mode_set(eth_dev, mode)) { +- RTE_BOND_LOG(ERR, "Failed to set bonded device %d mode to %d", ++ RTE_BOND_LOG(ERR, "Failed to set bonded device %u mode to %u", + eth_dev->data->port_id, mode); + goto err; + } +@@ -3483,6 +3588,7 @@ bond_ethdev_configure(struct rte_eth_dev *dev) + const char *name = dev->device->name; + struct bond_dev_private *internals = dev->data->dev_private; + struct rte_kvargs *kvlist = internals->kvlist; ++ uint64_t offloads; + int arg_count; + uint16_t port_id = dev - rte_eth_devices; + uint8_t agg_mode; +@@ -3504,6 +3610,11 @@ bond_ethdev_configure(struct rte_eth_dev *dev) + if (dev->data->dev_conf.rxmode.mq_mode & RTE_ETH_MQ_RX_RSS) { + struct rte_eth_rss_conf *rss_conf = + &dev->data->dev_conf.rx_adv_conf.rss_conf; ++ ++ if (internals->rss_key_len == 0) { ++ internals->rss_key_len = sizeof(default_rss_key); ++ } ++ + if (rss_conf->rss_key != NULL) { + if (internals->rss_key_len > rss_conf->rss_key_len) { + RTE_BOND_LOG(ERR, "Invalid rss key length(%u)", +@@ -3515,13 +3626,18 @@ bond_ethdev_configure(struct rte_eth_dev *dev) + internals->rss_key_len); + } else { + if (internals->rss_key_len > sizeof(default_rss_key)) { +- RTE_BOND_LOG(ERR, +- "There is no suitable default hash key"); +- return -EINVAL; ++ /* ++ * If the rss_key includes standard_rss_key and ++ * extended_hash_key, the rss key length will be ++ * larger than default rss key length, so it should ++ * re-calculate the hash key. ++ */ ++ for (i = 0; i < internals->rss_key_len; i++) ++ internals->rss_key[i] = (uint8_t)rte_rand(); ++ } else { ++ memcpy(internals->rss_key, default_rss_key, ++ internals->rss_key_len); + } +- +- memcpy(internals->rss_key, default_rss_key, +- internals->rss_key_len); + } + + for (i = 0; i < RTE_DIM(internals->reta_conf); i++) { +@@ -3533,6 +3649,16 @@ bond_ethdev_configure(struct rte_eth_dev *dev) + } + } + ++ offloads = dev->data->dev_conf.txmode.offloads; ++ if ((offloads & RTE_ETH_TX_OFFLOAD_MBUF_FAST_FREE) && ++ (internals->mode == BONDING_MODE_8023AD || ++ internals->mode == BONDING_MODE_BROADCAST)) { ++ RTE_BOND_LOG(WARNING, ++ "bond mode broadcast & 8023AD don't support MBUF_FAST_FREE offload, force disable it."); ++ offloads &= ~RTE_ETH_TX_OFFLOAD_MBUF_FAST_FREE; ++ dev->data->dev_conf.txmode.offloads = offloads; ++ } ++ + /* set the max_rx_pktlen */ + internals->max_rx_pktlen = internals->candidate_max_rx_pktlen; + +@@ -3765,6 +3891,18 @@ bond_ethdev_configure(struct rte_eth_dev *dev) + return -1; + } + ++ /* configure slaves so we can pass mtu setting */ ++ for (i = 0; i < internals->slave_count; i++) { ++ struct rte_eth_dev *slave_ethdev = ++ &(rte_eth_devices[internals->slaves[i].port_id]); ++ if (slave_configure(dev, slave_ethdev) != 0) { ++ RTE_BOND_LOG(ERR, ++ "bonded port (%d) failed to configure slave device (%d)", ++ dev->data->port_id, ++ internals->slaves[i].port_id); ++ return -1; ++ } ++ } + return 0; } -+void -+ovsdb_datum_add_from_index_unsafe(struct ovsdb_datum *dst, -+ const struct ovsdb_datum *src, -+ size_t idx, -+ const struct ovsdb_type *type) +diff --git a/dpdk/drivers/net/cnxk/cn10k_ethdev.c b/dpdk/drivers/net/cnxk/cn10k_ethdev.c +index 8378cbffc2..c36b858110 100644 +--- a/dpdk/drivers/net/cnxk/cn10k_ethdev.c ++++ b/dpdk/drivers/net/cnxk/cn10k_ethdev.c +@@ -313,6 +313,12 @@ cn10k_nix_configure(struct rte_eth_dev *eth_dev) + if (rc) + return rc; + ++ if (dev->tx_offloads & RTE_ETH_TX_OFFLOAD_SECURITY || ++ dev->rx_offloads & RTE_ETH_RX_OFFLOAD_SECURITY) { ++ /* Register callback to handle security error work */ ++ roc_nix_inl_cb_register(cn10k_eth_sec_sso_work_cb, NULL); ++ } ++ + /* Update offload flags */ + dev->rx_offload_flags = nix_rx_offload_flags(eth_dev); + dev->tx_offload_flags = nix_tx_offload_flags(eth_dev); +diff --git a/dpdk/drivers/net/cnxk/cn10k_ethdev.h b/dpdk/drivers/net/cnxk/cn10k_ethdev.h +index c2a46ad7ec..13403e14c4 100644 +--- a/dpdk/drivers/net/cnxk/cn10k_ethdev.h ++++ b/dpdk/drivers/net/cnxk/cn10k_ethdev.h +@@ -53,7 +53,7 @@ struct cn10k_outb_priv_data { + void *userdata; + /* Rlen computation data */ + struct cnxk_ipsec_outb_rlens rlens; +- /* Back pinter to eth sec session */ ++ /* Back pointer to eth sec session */ + struct cnxk_eth_sec_sess *eth_sec; + /* SA index */ + uint32_t sa_idx; +@@ -82,6 +82,9 @@ void cn10k_eth_set_tx_function(struct rte_eth_dev *eth_dev); + /* Security context setup */ + void cn10k_eth_sec_ops_override(void); + ++/* SSO Work callback */ ++void cn10k_eth_sec_sso_work_cb(uint64_t *gw, void *args); ++ + #define LMT_OFF(lmt_addr, lmt_num, offset) \ + (void *)((uintptr_t)(lmt_addr) + \ + ((uint64_t)(lmt_num) << ROC_LMT_LINE_SIZE_LOG2) + (offset)) +diff --git a/dpdk/drivers/net/cnxk/cn10k_ethdev_sec.c b/dpdk/drivers/net/cnxk/cn10k_ethdev_sec.c +index 235c16840e..f84a0fe80c 100644 +--- a/dpdk/drivers/net/cnxk/cn10k_ethdev_sec.c ++++ b/dpdk/drivers/net/cnxk/cn10k_ethdev_sec.c +@@ -138,7 +138,21 @@ static const struct rte_security_capability cn10k_eth_sec_capabilities[] = { + } + }; + +-static void ++static inline void ++cnxk_pktmbuf_free_no_cache(struct rte_mbuf *mbuf) +{ -+ const union ovsdb_atom *key = &src->keys[idx]; -+ const union ovsdb_atom *value = type->value.type != OVSDB_TYPE_VOID -+ ? &src->values[idx] -+ : NULL; -+ ovsdb_datum_add_unsafe(dst, key, value, type, NULL); ++ struct rte_mbuf *next; ++ ++ if (!mbuf) ++ return; ++ do { ++ next = mbuf->next; ++ roc_npa_aura_op_free(mbuf->pool->pool_id, 1, (rte_iova_t)mbuf); ++ mbuf = next; ++ } while (mbuf != NULL); +} + - /* Adds 'n' atoms starting from index 'start_idx' from 'src' to the end of - * 'dst'. 'dst' should have enough memory allocated to hold the additional - * 'n' atoms. Atoms are not cloned, i.e. 'dst' will reference the same data. -@@ -2165,12 +2178,10 @@ ovsdb_datum_added_removed(struct ovsdb_datum *added, - int c = ovsdb_atom_compare_3way(&old->keys[oi], &new->keys[ni], - type->key.type); - if (c < 0) { -- ovsdb_datum_add_unsafe(removed, &old->keys[oi], &old->values[oi], -- type, NULL); -+ ovsdb_datum_add_from_index_unsafe(removed, old, oi, type); - oi++; - } else if (c > 0) { -- ovsdb_datum_add_unsafe(added, &new->keys[ni], &new->values[ni], -- type, NULL); -+ ovsdb_datum_add_from_index_unsafe(added, new, ni, type); - ni++; - } else { - if (type->value.type != OVSDB_TYPE_VOID && -@@ -2186,13 +2197,11 @@ ovsdb_datum_added_removed(struct ovsdb_datum *added, - } - - for (; oi < old->n; oi++) { -- ovsdb_datum_add_unsafe(removed, &old->keys[oi], &old->values[oi], -- type, NULL); -+ ovsdb_datum_add_from_index_unsafe(removed, old, oi, type); - } ++void + cn10k_eth_sec_sso_work_cb(uint64_t *gw, void *args) + { + struct rte_eth_event_ipsec_desc desc; +@@ -148,6 +162,7 @@ cn10k_eth_sec_sso_work_cb(uint64_t *gw, void *args) + struct cpt_cn10k_res_s *res; + struct rte_eth_dev *eth_dev; + struct cnxk_eth_dev *dev; ++ static uint64_t warn_cnt; + uint16_t dlen_adj, rlen; + struct rte_mbuf *mbuf; + uintptr_t sa_base; +@@ -161,7 +176,7 @@ cn10k_eth_sec_sso_work_cb(uint64_t *gw, void *args) + /* Event from inbound inline dev due to IPSEC packet bad L4 */ + mbuf = (struct rte_mbuf *)(gw[1] - sizeof(struct rte_mbuf)); + plt_nix_dbg("Received mbuf %p from inline dev inbound", mbuf); +- rte_pktmbuf_free(mbuf); ++ cnxk_pktmbuf_free_no_cache(mbuf); + return; + case RTE_EVENT_TYPE_CPU: + /* Check for subtype */ +@@ -212,17 +227,29 @@ cn10k_eth_sec_sso_work_cb(uint64_t *gw, void *args) + case ROC_IE_OT_UCC_ERR_SA_OVERFLOW: + desc.subtype = RTE_ETH_EVENT_IPSEC_ESN_OVERFLOW; + break; ++ case ROC_IE_OT_UCC_ERR_PKT_IP: ++ warn_cnt++; ++ if (warn_cnt % 10000 == 0) ++ plt_warn("Outbound error, bad ip pkt, mbuf %p," ++ " sa_index %u (total warnings %" PRIu64 ")", ++ mbuf, sess_priv.sa_idx, warn_cnt); ++ desc.subtype = RTE_ETH_EVENT_IPSEC_UNKNOWN; ++ break; + default: +- plt_warn("Outbound error, mbuf %p, sa_index %u, " +- "compcode %x uc %x", mbuf, sess_priv.sa_idx, +- res->compcode, res->uc_compcode); ++ warn_cnt++; ++ if (warn_cnt % 10000 == 0) ++ plt_warn("Outbound error, mbuf %p, sa_index %u," ++ " compcode %x uc %x," ++ " (total warnings %" PRIu64 ")", ++ mbuf, sess_priv.sa_idx, res->compcode, ++ res->uc_compcode, warn_cnt); + desc.subtype = RTE_ETH_EVENT_IPSEC_UNKNOWN; + break; + } - for (; ni < new->n; ni++) { -- ovsdb_datum_add_unsafe(added, &new->keys[ni], &new->values[ni], -- type, NULL); -+ ovsdb_datum_add_from_index_unsafe(added, new, ni, type); - } + desc.metadata = (uint64_t)priv->userdata; + rte_eth_dev_callback_process(eth_dev, RTE_ETH_EVENT_IPSEC, &desc); +- rte_pktmbuf_free(mbuf); ++ cnxk_pktmbuf_free_no_cache(mbuf); } -@@ -2228,12 +2237,10 @@ ovsdb_datum_diff(struct ovsdb_datum *diff, - int c = ovsdb_atom_compare_3way(&old->keys[oi], &new->keys[ni], - type->key.type); - if (c < 0) { -- ovsdb_datum_add_unsafe(diff, &old->keys[oi], &old->values[oi], -- type, NULL); -+ ovsdb_datum_add_from_index_unsafe(diff, old, oi, type); - oi++; - } else if (c > 0) { -- ovsdb_datum_add_unsafe(diff, &new->keys[ni], &new->values[ni], -- type, NULL); -+ ovsdb_datum_add_from_index_unsafe(diff, new, ni, type); - ni++; - } else { - if (type->value.type != OVSDB_TYPE_VOID && -@@ -2247,13 +2254,11 @@ ovsdb_datum_diff(struct ovsdb_datum *diff, - } - - for (; oi < old->n; oi++) { -- ovsdb_datum_add_unsafe(diff, &old->keys[oi], &old->values[oi], -- type, NULL); -+ ovsdb_datum_add_from_index_unsafe(diff, old, oi, type); - } + static int +@@ -249,9 +276,6 @@ cn10k_eth_sec_session_create(void *device, + if (rte_security_dynfield_register() < 0) + return -ENOTSUP; - for (; ni < new->n; ni++) { -- ovsdb_datum_add_unsafe(diff, &new->keys[ni], &new->values[ni], -- type, NULL); -+ ovsdb_datum_add_from_index_unsafe(diff, new, ni, type); - } +- if (rte_eal_process_type() == RTE_PROC_PRIMARY) +- roc_nix_inl_cb_register(cn10k_eth_sec_sso_work_cb, NULL); +- + ipsec = &conf->ipsec; + crypto = conf->crypto_xform; + inbound = !!(ipsec->direction == RTE_SECURITY_IPSEC_SA_DIR_INGRESS); +diff --git a/dpdk/drivers/net/cnxk/cn10k_rx.c b/dpdk/drivers/net/cnxk/cn10k_rx.c +index 5d603514c0..94b0dfcde7 100644 +--- a/dpdk/drivers/net/cnxk/cn10k_rx.c ++++ b/dpdk/drivers/net/cnxk/cn10k_rx.c +@@ -31,6 +31,10 @@ pick_rx_func(struct rte_eth_dev *eth_dev, + [!!(dev->rx_offload_flags & NIX_RX_OFFLOAD_PTYPE_F)] + [!!(dev->rx_offload_flags & NIX_RX_OFFLOAD_RSS_F)]; + ++ if (eth_dev->data->dev_started) ++ rte_eth_fp_ops[eth_dev->data->port_id].rx_pkt_burst = ++ eth_dev->rx_pkt_burst; ++ + rte_atomic_thread_fence(__ATOMIC_RELEASE); } -diff --git a/lib/ovsdb-data.h b/lib/ovsdb-data.h -index 47115a7b85..ba5d179a65 100644 ---- a/lib/ovsdb-data.h -+++ b/lib/ovsdb-data.h -@@ -280,6 +280,10 @@ void ovsdb_datum_add_unsafe(struct ovsdb_datum *, - const union ovsdb_atom *value, - const struct ovsdb_type *, - const union ovsdb_atom *range_end_atom); -+void ovsdb_datum_add_from_index_unsafe(struct ovsdb_datum *dst, -+ const struct ovsdb_datum *src, -+ size_t idx, -+ const struct ovsdb_type *type); - - /* Transactions with named-uuid row names. */ - struct json *ovsdb_datum_to_json_with_row_names(const struct ovsdb_datum *, -diff --git a/lib/ovsdb-idl.c b/lib/ovsdb-idl.c -index c19128d55c..882ede7559 100644 ---- a/lib/ovsdb-idl.c -+++ b/lib/ovsdb-idl.c -@@ -389,25 +389,25 @@ ovsdb_idl_clear(struct ovsdb_idl *db) - */ - for (size_t i = 0; i < db->class_->n_tables; i++) { - struct ovsdb_idl_table *table = &db->tables[i]; -- struct ovsdb_idl_row *row, *next_row; -+ struct ovsdb_idl_row *row; +diff --git a/dpdk/drivers/net/cnxk/cn10k_rx.h b/dpdk/drivers/net/cnxk/cn10k_rx.h +index fe408907a6..5806392322 100644 +--- a/dpdk/drivers/net/cnxk/cn10k_rx.h ++++ b/dpdk/drivers/net/cnxk/cn10k_rx.h +@@ -363,7 +363,13 @@ cn10k_nix_cqe_to_mbuf(const struct nix_cqe_hdr_s *cq, const uint32_t tag, + *(uint64_t *)(&mbuf->rearm_data) = val; + + if (flag & NIX_RX_MULTI_SEG_F) +- nix_cqe_xtract_mseg(rx, mbuf, val, flag); ++ /* ++ * For multi segment packets, mbuf length correction according ++ * to Rx timestamp length will be handled later during ++ * timestamp data process. ++ * Hence, flag argument is not required. ++ */ ++ nix_cqe_xtract_mseg(rx, mbuf, val, 0); + else + mbuf->next = NULL; + } +@@ -451,7 +457,6 @@ cn10k_nix_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t pkts, + flags); + cnxk_nix_mbuf_to_tstamp(mbuf, rxq->tstamp, + (flags & NIX_RX_OFFLOAD_TSTAMP_F), +- (flags & NIX_RX_MULTI_SEG_F), + (uint64_t *)((uint8_t *)mbuf + + data_off)); + rx_pkts[packets++] = mbuf; +@@ -481,10 +486,11 @@ cn10k_nix_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t pkts, + plt_write64((wdata | nb_pkts), rxq->cq_door); + + /* Free remaining meta buffers if any */ +- if (flags & NIX_RX_OFFLOAD_SECURITY_F && loff) { ++ if (flags & NIX_RX_OFFLOAD_SECURITY_F && loff) + nix_sec_flush_meta(laddr, lmt_id + lnum, loff, aura_handle); +- plt_io_wmb(); +- } ++ ++ if (flags & NIX_RX_OFFLOAD_SECURITY_F) ++ rte_io_wmb(); + + return nb_pkts; + } +diff --git a/dpdk/drivers/net/cnxk/cn10k_tx.c b/dpdk/drivers/net/cnxk/cn10k_tx.c +index 5e6c5ee111..4e1abf7804 100644 +--- a/dpdk/drivers/net/cnxk/cn10k_tx.c ++++ b/dpdk/drivers/net/cnxk/cn10k_tx.c +@@ -37,6 +37,10 @@ pick_tx_func(struct rte_eth_dev *eth_dev, + [!!(dev->tx_offload_flags & NIX_TX_OFFLOAD_VLAN_QINQ_F)] + [!!(dev->tx_offload_flags & NIX_TX_OFFLOAD_OL3_OL4_CSUM_F)] + [!!(dev->tx_offload_flags & NIX_TX_OFFLOAD_L3_L4_CSUM_F)]; ++ ++ if (eth_dev->data->dev_started) ++ rte_eth_fp_ops[eth_dev->data->port_id].tx_pkt_burst = ++ eth_dev->tx_pkt_burst; + } - if (hmap_is_empty(&table->rows)) { - continue; - } + void +diff --git a/dpdk/drivers/net/cnxk/cn10k_tx.h b/dpdk/drivers/net/cnxk/cn10k_tx.h +index 873e1871f9..6704d2d655 100644 +--- a/dpdk/drivers/net/cnxk/cn10k_tx.h ++++ b/dpdk/drivers/net/cnxk/cn10k_tx.h +@@ -736,7 +736,7 @@ cn10k_nix_xmit_prepare_tstamp(uintptr_t lmt_addr, const uint64_t *cmd, + /* Retrieving the default desc values */ + lmt[off] = cmd[2]; + +- /* Using compiler barier to avoid voilation of C ++ /* Using compiler barrier to avoid violation of C + * aliasing rules. + */ + rte_compiler_barrier(); +@@ -745,7 +745,7 @@ cn10k_nix_xmit_prepare_tstamp(uintptr_t lmt_addr, const uint64_t *cmd, + /* Packets for which RTE_MBUF_F_TX_IEEE1588_TMST is not set, tx tstamp + * should not be recorded, hence changing the alg type to + * NIX_SENDMEMALG_SET and also changing send mem addr field to +- * next 8 bytes as it corrpt the actual tx tstamp registered ++ * next 8 bytes as it corrupts the actual Tx tstamp registered + * address. + */ + send_mem->w0.subdc = NIX_SUBDC_MEM; +@@ -849,8 +849,8 @@ cn10k_nix_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts, uint16_t pkts, + uintptr_t pa, lbase = txq->lmt_base; + uint16_t lmt_id, burst, left, i; + uintptr_t c_lbase = lbase; ++ uint64_t lso_tun_fmt = 0; + rte_iova_t c_io_addr; +- uint64_t lso_tun_fmt; + uint16_t c_lmt_id; + uint64_t sa_base; + uintptr_t laddr; +@@ -976,9 +976,9 @@ cn10k_nix_xmit_pkts_mseg(void *tx_queue, struct rte_mbuf **tx_pkts, + uint16_t segdw, lmt_id, burst, left, i; + uint8_t lnum, c_lnum, c_loff; + uintptr_t c_lbase = lbase; ++ uint64_t lso_tun_fmt = 0; + uint64_t data0, data1; + rte_iova_t c_io_addr; +- uint64_t lso_tun_fmt; + uint8_t shft, c_shft; + __uint128_t data128; + uint16_t c_lmt_id; +@@ -2254,7 +2254,7 @@ cn10k_nix_xmit_pkts_vector(void *tx_queue, struct rte_mbuf **tx_pkts, + } + + if (flags & NIX_TX_OFFLOAD_TSTAMP_F) { +- /* Tx ol_flag for timestam. */ ++ /* Tx ol_flag for timestamp. */ + const uint64x2_t olf = {RTE_MBUF_F_TX_IEEE1588_TMST, + RTE_MBUF_F_TX_IEEE1588_TMST}; + /* Set send mem alg to SUB. */ +diff --git a/dpdk/drivers/net/cnxk/cn9k_rx.c b/dpdk/drivers/net/cnxk/cn9k_rx.c +index 8d504c4a6d..60baf10b39 100644 +--- a/dpdk/drivers/net/cnxk/cn9k_rx.c ++++ b/dpdk/drivers/net/cnxk/cn9k_rx.c +@@ -31,6 +31,10 @@ pick_rx_func(struct rte_eth_dev *eth_dev, + [!!(dev->rx_offload_flags & NIX_RX_OFFLOAD_PTYPE_F)] + [!!(dev->rx_offload_flags & NIX_RX_OFFLOAD_RSS_F)]; + ++ if (eth_dev->data->dev_started) ++ rte_eth_fp_ops[eth_dev->data->port_id].rx_pkt_burst = ++ eth_dev->rx_pkt_burst; ++ + rte_atomic_thread_fence(__ATOMIC_RELEASE); + } -- HMAP_FOR_EACH_SAFE (row, next_row, hmap_node, &table->rows) { -- struct ovsdb_idl_arc *arc, *next_arc; -+ HMAP_FOR_EACH_SAFE (row, hmap_node, &table->rows) { -+ struct ovsdb_idl_arc *arc; +diff --git a/dpdk/drivers/net/cnxk/cn9k_rx.h b/dpdk/drivers/net/cnxk/cn9k_rx.h +index 225bb4197c..848df4190c 100644 +--- a/dpdk/drivers/net/cnxk/cn9k_rx.h ++++ b/dpdk/drivers/net/cnxk/cn9k_rx.h +@@ -341,14 +341,19 @@ cn9k_nix_cqe_to_mbuf(const struct nix_cqe_hdr_s *cq, const uint32_t tag, + ol_flags = + nix_update_match_id(rx->cn9k.match_id, ol_flags, mbuf); + ++ mbuf->ol_flags = ol_flags; ++ *(uint64_t *)(&mbuf->rearm_data) = val; + mbuf->pkt_len = len; + mbuf->data_len = len; +- *(uint64_t *)(&mbuf->rearm_data) = val; +- +- mbuf->ol_flags = ol_flags; + + if (flag & NIX_RX_MULTI_SEG_F) +- nix_cqe_xtract_mseg(rx, mbuf, val, flag); ++ /* ++ * For multi segment packets, mbuf length correction according ++ * to Rx timestamp length will be handled later during ++ * timestamp data process. ++ * Hence, flag argument is not required. ++ */ ++ nix_cqe_xtract_mseg(rx, mbuf, val, 0); + else + mbuf->next = NULL; + } +@@ -413,7 +418,6 @@ cn9k_nix_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t pkts, + flags); + cnxk_nix_mbuf_to_tstamp(mbuf, rxq->tstamp, + (flags & NIX_RX_OFFLOAD_TSTAMP_F), +- (flags & NIX_RX_MULTI_SEG_F), + (uint64_t *)((uint8_t *)mbuf + + data_off)); + rx_pkts[packets++] = mbuf; +diff --git a/dpdk/drivers/net/cnxk/cn9k_tx.c b/dpdk/drivers/net/cnxk/cn9k_tx.c +index f3f19fed97..f560286c97 100644 +--- a/dpdk/drivers/net/cnxk/cn9k_tx.c ++++ b/dpdk/drivers/net/cnxk/cn9k_tx.c +@@ -36,6 +36,10 @@ pick_tx_func(struct rte_eth_dev *eth_dev, + [!!(dev->tx_offload_flags & NIX_TX_OFFLOAD_VLAN_QINQ_F)] + [!!(dev->tx_offload_flags & NIX_TX_OFFLOAD_OL3_OL4_CSUM_F)] + [!!(dev->tx_offload_flags & NIX_TX_OFFLOAD_L3_L4_CSUM_F)]; ++ ++ if (eth_dev->data->dev_started) ++ rte_eth_fp_ops[eth_dev->data->port_id].tx_pkt_burst = ++ eth_dev->tx_pkt_burst; + } - if (!ovsdb_idl_row_is_orphan(row)) { - ovsdb_idl_remove_from_indexes(row); - ovsdb_idl_row_unparse(row); - } -- LIST_FOR_EACH_SAFE (arc, next_arc, src_node, &row->src_arcs) { -+ LIST_FOR_EACH_SAFE (arc, src_node, &row->src_arcs) { - ovs_list_remove(&arc->src_node); - ovs_list_remove(&arc->dst_node); - free(arc); - } -- LIST_FOR_EACH_SAFE (arc, next_arc, dst_node, &row->dst_arcs) { -+ LIST_FOR_EACH_SAFE (arc, dst_node, &row->dst_arcs) { - ovs_list_remove(&arc->src_node); - ovs_list_remove(&arc->dst_node); - free(arc); -@@ -1041,8 +1041,8 @@ ovsdb_idl_condition_destroy(struct ovsdb_idl_condition *cond) void - ovsdb_idl_condition_clear(struct ovsdb_idl_condition *cond) +diff --git a/dpdk/drivers/net/cnxk/cn9k_tx.h b/dpdk/drivers/net/cnxk/cn9k_tx.h +index 435dde1317..8167313a15 100644 +--- a/dpdk/drivers/net/cnxk/cn9k_tx.h ++++ b/dpdk/drivers/net/cnxk/cn9k_tx.h +@@ -304,7 +304,7 @@ cn9k_nix_xmit_prepare_tstamp(uint64_t *cmd, const uint64_t *send_mem_desc, + /* Retrieving the default desc values */ + cmd[off] = send_mem_desc[6]; + +- /* Using compiler barier to avoid voilation of C ++ /* Using compiler barrier to avoid violation of C + * aliasing rules. + */ + rte_compiler_barrier(); +@@ -313,7 +313,7 @@ cn9k_nix_xmit_prepare_tstamp(uint64_t *cmd, const uint64_t *send_mem_desc, + /* Packets for which RTE_MBUF_F_TX_IEEE1588_TMST is not set, tx tstamp + * should not be recorded, hence changing the alg type to + * NIX_SENDMEMALG_SET and also changing send mem addr field to +- * next 8 bytes as it corrpt the actual tx tstamp registered ++ * next 8 bytes as it corrupts the actual Tx tstamp registered + * address. + */ + send_mem->w0.cn9k.alg = +@@ -465,8 +465,8 @@ cn9k_nix_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts, uint16_t pkts, { -- struct ovsdb_idl_clause *clause, *next; -- HMAP_FOR_EACH_SAFE (clause, next, hmap_node, &cond->clauses) { -+ struct ovsdb_idl_clause *clause; -+ HMAP_FOR_EACH_SAFE (clause, hmap_node, &cond->clauses) { - hmap_remove(&cond->clauses, &clause->hmap_node); - ovsdb_idl_clause_destroy(clause); - } -@@ -1345,9 +1345,9 @@ ovsdb_idl_track_clear__(struct ovsdb_idl *idl, bool flush_all) - struct ovsdb_idl_table *table = &idl->tables[i]; - - if (!ovs_list_is_empty(&table->track_list)) { -- struct ovsdb_idl_row *row, *next; -+ struct ovsdb_idl_row *row; + struct cn9k_eth_txq *txq = tx_queue; + const rte_iova_t io_addr = txq->io_addr; ++ uint64_t lso_tun_fmt = 0; + void *lmt_addr = txq->lmt_addr; +- uint64_t lso_tun_fmt; + uint16_t i; + + NIX_XMIT_FC_OR_RETURN(txq, pkts); +@@ -506,8 +506,8 @@ cn9k_nix_xmit_pkts_mseg(void *tx_queue, struct rte_mbuf **tx_pkts, + { + struct cn9k_eth_txq *txq = tx_queue; + const rte_iova_t io_addr = txq->io_addr; ++ uint64_t lso_tun_fmt = 0; + void *lmt_addr = txq->lmt_addr; +- uint64_t lso_tun_fmt; + uint16_t segdw; + uint64_t i; + +@@ -1531,7 +1531,7 @@ cn9k_nix_xmit_pkts_vector(void *tx_queue, struct rte_mbuf **tx_pkts, + } + + if (flags & NIX_TX_OFFLOAD_TSTAMP_F) { +- /* Tx ol_flag for timestam. */ ++ /* Tx ol_flag for timestamp. */ + const uint64x2_t olf = {RTE_MBUF_F_TX_IEEE1588_TMST, + RTE_MBUF_F_TX_IEEE1588_TMST}; + /* Set send mem alg to SUB. */ +diff --git a/dpdk/drivers/net/cnxk/cnxk_ethdev.c b/dpdk/drivers/net/cnxk/cnxk_ethdev.c +index 74f625553d..a1da90be57 100644 +--- a/dpdk/drivers/net/cnxk/cnxk_ethdev.c ++++ b/dpdk/drivers/net/cnxk/cnxk_ethdev.c +@@ -3,6 +3,8 @@ + */ + #include -- LIST_FOR_EACH_SAFE(row, next, track_node, &table->track_list) { -+ LIST_FOR_EACH_SAFE (row, track_node, &table->track_list) { - if (row->updated) { - free(row->updated); - row->updated = NULL; -@@ -1480,9 +1480,9 @@ ovsdb_idl_parse_update(struct ovsdb_idl *idl, - static void - ovsdb_idl_reparse_deleted(struct ovsdb_idl *db) ++#include ++ + static inline uint64_t + nix_get_rx_offload_capa(struct cnxk_eth_dev *dev) { -- struct ovsdb_idl_row *row, *next; -+ struct ovsdb_idl_row *row; +@@ -177,7 +179,7 @@ nix_meter_fini(struct cnxk_eth_dev *dev) + struct roc_nix *nix = &dev->nix; + struct roc_nix_rq *rq; + uint32_t i; +- int rc; ++ int rc = 0; + + RTE_TAILQ_FOREACH_SAFE(mtr, fms, next, next_mtr) { + for (i = 0; i < mtr->rq_num; i++) { +@@ -597,6 +599,13 @@ cnxk_nix_rx_queue_setup(struct rte_eth_dev *eth_dev, uint16_t qid, + rxq_sp->qconf.mp = mp; + + if (dev->rx_offloads & RTE_ETH_RX_OFFLOAD_SECURITY) { ++ /* Pass a tagmask used to handle error packets in inline device. ++ * Ethdev rq's tag_mask field will be overwritten later ++ * when sso is setup. ++ */ ++ rq->tag_mask = ++ 0x0FF00000 | ((uint32_t)RTE_EVENT_TYPE_ETHDEV << 28); ++ + /* Setup rq reference for inline dev if present */ + rc = roc_nix_inl_dev_rq_get(rq); + if (rc) +@@ -1122,6 +1131,10 @@ cnxk_nix_configure(struct rte_eth_dev *eth_dev) + goto fail_configure; + } -- LIST_FOR_EACH_SAFE (row, next, track_node, &db->deleted_untracked_rows) { -+ LIST_FOR_EACH_SAFE (row, track_node, &db->deleted_untracked_rows) { - ovsdb_idl_row_untrack_change(row); - add_tracked_change_for_references(row); - ovsdb_idl_row_reparse_backrefs(row); -@@ -1906,8 +1906,8 @@ ovsdb_idl_index_create2(struct ovsdb_idl *idl, - static void - ovsdb_idl_destroy_indexes(struct ovsdb_idl_table *table) ++ /* Check if ptp is enable in PF owning this VF*/ ++ if (!roc_nix_is_pf(nix) && (!roc_nix_is_sdp(nix))) ++ dev->ptp_en = roc_nix_ptp_is_enable(nix); ++ + dev->npc.channel = roc_nix_get_base_chan(nix); + + nb_rxq = data->nb_rx_queues; +diff --git a/dpdk/drivers/net/cnxk/cnxk_ethdev.h b/dpdk/drivers/net/cnxk/cnxk_ethdev.h +index 5bfda3d815..480cc6dfa4 100644 +--- a/dpdk/drivers/net/cnxk/cnxk_ethdev.h ++++ b/dpdk/drivers/net/cnxk/cnxk_ethdev.h +@@ -685,14 +685,11 @@ cnxk_nix_timestamp_dynfield(struct rte_mbuf *mbuf, + static __rte_always_inline void + cnxk_nix_mbuf_to_tstamp(struct rte_mbuf *mbuf, + struct cnxk_timesync_info *tstamp, +- const uint8_t ts_enable, const uint8_t mseg_enable, +- uint64_t *tstamp_ptr) ++ const uint8_t ts_enable, uint64_t *tstamp_ptr) { -- struct ovsdb_idl_index *index, *next; -- LIST_FOR_EACH_SAFE (index, next, node, &table->indexes) { -+ struct ovsdb_idl_index *index; -+ LIST_FOR_EACH_SAFE (index, node, &table->indexes) { - skiplist_destroy(index->skiplist, NULL); - free(index->columns); - free(index); -@@ -2145,12 +2145,12 @@ ovsdb_idl_row_clear_new(struct ovsdb_idl_row *row) - static void - ovsdb_idl_row_clear_arcs(struct ovsdb_idl_row *row, bool destroy_dsts) + if (ts_enable) { +- if (!mseg_enable) { +- mbuf->pkt_len -= CNXK_NIX_TIMESYNC_RX_OFFSET; +- mbuf->data_len -= CNXK_NIX_TIMESYNC_RX_OFFSET; +- } ++ mbuf->pkt_len -= CNXK_NIX_TIMESYNC_RX_OFFSET; ++ mbuf->data_len -= CNXK_NIX_TIMESYNC_RX_OFFSET; + + /* Reading the rx timestamp inserted by CGX, viz at + * starting of the packet data. +diff --git a/dpdk/drivers/net/cnxk/cnxk_ethdev_mtr.c b/dpdk/drivers/net/cnxk/cnxk_ethdev_mtr.c +index 39d8563826..b6ccccdc39 100644 +--- a/dpdk/drivers/net/cnxk/cnxk_ethdev_mtr.c ++++ b/dpdk/drivers/net/cnxk/cnxk_ethdev_mtr.c +@@ -277,15 +277,54 @@ cnxk_nix_mtr_profile_delete(struct rte_eth_dev *eth_dev, uint32_t profile_id, + return 0; + } + ++static int ++update_mtr_err(uint32_t act_color, struct rte_mtr_error *error, bool action) ++{ ++ const char *str = NULL; ++ switch (act_color) { ++ case RTE_COLOR_GREEN: ++ if (action) { ++ str = "Green action is not valid"; ++ goto notsup; ++ } else { ++ str = "Green action is null"; ++ goto notvalid; ++ } ++ break; ++ case RTE_COLOR_YELLOW: ++ if (action) { ++ str = "Yellow action is not valid"; ++ goto notsup; ++ } else { ++ str = "Yellow action is null"; ++ goto notvalid; ++ } ++ break; ++ case RTE_COLOR_RED: ++ if (action) { ++ str = "Red action is not valid"; ++ goto notsup; ++ } else { ++ str = "Red action is null"; ++ goto notvalid; ++ } ++ break; ++ } ++notsup: ++ return -rte_mtr_error_set(error, ENOTSUP, ++ RTE_MTR_ERROR_TYPE_METER_POLICY, NULL, str); ++notvalid: ++ return -rte_mtr_error_set(error, EINVAL, ++ RTE_MTR_ERROR_TYPE_METER_POLICY, NULL, str); ++} ++ + static int + cnxk_nix_mtr_policy_validate(struct rte_eth_dev *dev, + struct rte_mtr_meter_policy_params *policy, + struct rte_mtr_error *error) { -- struct ovsdb_idl_arc *arc, *next; -+ struct ovsdb_idl_arc *arc; +- static const char *const action_color[] = {"Green", "Yellow", "Red"}; + bool supported[RTE_COLORS] = {false, false, false}; + const struct rte_flow_action *action; +- char message[1024]; + uint32_t i; + + RTE_SET_USED(dev); +@@ -304,21 +343,11 @@ cnxk_nix_mtr_policy_validate(struct rte_eth_dev *dev, + if (action->type == RTE_FLOW_ACTION_TYPE_DROP) + supported[i] = true; + +- if (!supported[i]) { +- sprintf(message, +- "%s action is not valid", +- action_color[i]); +- return -rte_mtr_error_set(error, +- ENOTSUP, +- RTE_MTR_ERROR_TYPE_METER_POLICY, NULL, +- message); +- } ++ if (!supported[i]) ++ return update_mtr_err(i, error, true); + } + } else { +- sprintf(message, "%s action is null", action_color[i]); +- return -rte_mtr_error_set(error, EINVAL, +- RTE_MTR_ERROR_TYPE_METER_POLICY, NULL, +- message); ++ return update_mtr_err(i, error, false); + } + } - /* Delete all forward arcs. If 'destroy_dsts', destroy any orphaned rows - * that this causes to be unreferenced. - */ -- LIST_FOR_EACH_SAFE (arc, next, src_node, &row->src_arcs) { -+ LIST_FOR_EACH_SAFE (arc, src_node, &row->src_arcs) { - ovs_list_remove(&arc->dst_node); - if (destroy_dsts - && ovsdb_idl_row_is_orphan(arc->dst) -@@ -2166,7 +2166,7 @@ ovsdb_idl_row_clear_arcs(struct ovsdb_idl_row *row, bool destroy_dsts) - static void - ovsdb_idl_row_reparse_backrefs(struct ovsdb_idl_row *row) +diff --git a/dpdk/drivers/net/cnxk/cnxk_ethdev_ops.c b/dpdk/drivers/net/cnxk/cnxk_ethdev_ops.c +index ce5f1f7240..f1d13c5004 100644 +--- a/dpdk/drivers/net/cnxk/cnxk_ethdev_ops.c ++++ b/dpdk/drivers/net/cnxk/cnxk_ethdev_ops.c +@@ -517,7 +517,8 @@ cnxk_nix_allmulticast_enable(struct rte_eth_dev *eth_dev) { -- struct ovsdb_idl_arc *arc, *next; -+ struct ovsdb_idl_arc *arc; + struct cnxk_eth_dev *dev = cnxk_eth_pmd_priv(eth_dev); - /* This is trickier than it looks. ovsdb_idl_row_clear_arcs() will destroy - * 'arc', so we need to use the "safe" variant of list traversal. However, -@@ -2178,7 +2178,7 @@ ovsdb_idl_row_reparse_backrefs(struct ovsdb_idl_row *row) - * (If duplicate arcs were possible then we would need to make sure that - * 'next' didn't also point into 'arc''s destination, but we forbid - * duplicate arcs.) */ -- LIST_FOR_EACH_SAFE (arc, next, dst_node, &row->dst_arcs) { -+ LIST_FOR_EACH_SAFE (arc, dst_node, &row->dst_arcs) { - struct ovsdb_idl_row *ref = arc->src; +- return roc_nix_npc_mcast_config(&dev->nix, true, false); ++ return roc_nix_npc_mcast_config(&dev->nix, true, ++ eth_dev->data->promiscuous); + } - ovsdb_idl_row_unparse(ref); -@@ -2329,9 +2329,9 @@ ovsdb_idl_row_destroy_postprocess(struct ovsdb_idl *idl) - struct ovsdb_idl_table *table = &idl->tables[i]; + int +@@ -746,6 +747,8 @@ cnxk_nix_reta_update(struct rte_eth_dev *eth_dev, + goto fail; + } - if (!ovs_list_is_empty(&table->track_list)) { -- struct ovsdb_idl_row *row, *next; -+ struct ovsdb_idl_row *row; ++ roc_nix_rss_reta_get(nix, 0, reta); ++ + /* Copy RETA table */ + for (i = 0; i < (int)(dev->nix.reta_sz / RTE_ETH_RETA_GROUP_SIZE); i++) { + for (j = 0; j < RTE_ETH_RETA_GROUP_SIZE; j++) { +diff --git a/dpdk/drivers/net/cnxk/cnxk_ethdev_telemetry.c b/dpdk/drivers/net/cnxk/cnxk_ethdev_telemetry.c +index 83bc65848c..4fd9048643 100644 +--- a/dpdk/drivers/net/cnxk/cnxk_ethdev_telemetry.c ++++ b/dpdk/drivers/net/cnxk/cnxk_ethdev_telemetry.c +@@ -49,6 +49,8 @@ ethdev_tel_handle_info(const char *cmd __rte_unused, + rte_tel_data_add_dict_int(d, "n_ports", n_ports); + + i_data = rte_tel_data_alloc(); ++ if (i_data == NULL) ++ return -ENOMEM; + rte_tel_data_start_array(i_data, RTE_TEL_U64_VAL); + + for (i = 0; i < RTE_MAX_ETHPORTS; i++) { +diff --git a/dpdk/drivers/net/cnxk/cnxk_ptp.c b/dpdk/drivers/net/cnxk/cnxk_ptp.c +index 139fea256c..359f9a30ae 100644 +--- a/dpdk/drivers/net/cnxk/cnxk_ptp.c ++++ b/dpdk/drivers/net/cnxk/cnxk_ptp.c +@@ -12,7 +12,7 @@ cnxk_nix_read_clock(struct rte_eth_dev *eth_dev, uint64_t *clock) + /* This API returns the raw PTP HI clock value. Since LFs do not + * have direct access to PTP registers and it requires mbox msg + * to AF for this value. In fastpath reading this value for every +- * packet (which involes mbox call) becomes very expensive, hence ++ * packet (which involves mbox call) becomes very expensive, hence + * we should be able to derive PTP HI clock value from tsc by + * using freq_mult and clk_delta calculated during configure stage. + */ +diff --git a/dpdk/drivers/net/cnxk/cnxk_rte_flow.c b/dpdk/drivers/net/cnxk/cnxk_rte_flow.c +index b08d7c34fa..32166ae764 100644 +--- a/dpdk/drivers/net/cnxk/cnxk_rte_flow.c ++++ b/dpdk/drivers/net/cnxk/cnxk_rte_flow.c +@@ -297,7 +297,14 @@ cnxk_flow_validate(struct rte_eth_dev *eth_dev, + return rc; + } -- LIST_FOR_EACH_SAFE(row, next, track_node, &table->track_list) { -+ LIST_FOR_EACH_SAFE (row, track_node, &table->track_list) { - if (!ovsdb_idl_track_is_set(row->table)) { - ovs_list_remove(&row->track_node); - ovsdb_idl_row_unparse(row); -@@ -2729,7 +2729,7 @@ ovsdb_idl_txn_increment(struct ovsdb_idl_txn *txn, - void - ovsdb_idl_txn_destroy(struct ovsdb_idl_txn *txn) - { -- struct ovsdb_idl_txn_insert *insert, *next; -+ struct ovsdb_idl_txn_insert *insert; +- return roc_npc_flow_parse(npc, &in_attr, in_pattern, in_actions, &flow); ++ rc = roc_npc_flow_parse(npc, &in_attr, in_pattern, in_actions, &flow); ++ ++ if (rc) { ++ rte_flow_error_set(error, 0, rc, NULL, ++ "Flow validation failed"); ++ return rc; ++ } ++ return 0; + } - if (txn->status == TXN_INCOMPLETE) { - ovsdb_cs_forget_transaction(txn->idl->cs, txn->request_id); -@@ -2739,7 +2739,7 @@ ovsdb_idl_txn_destroy(struct ovsdb_idl_txn *txn) - ovsdb_idl_txn_abort(txn); - ds_destroy(&txn->comment); - free(txn->error); -- HMAP_FOR_EACH_SAFE (insert, next, hmap_node, &txn->inserted_rows) { -+ HMAP_FOR_EACH_SAFE (insert, hmap_node, &txn->inserted_rows) { - free(insert); - } - hmap_destroy(&txn->inserted_rows); -@@ -2824,7 +2824,7 @@ substitute_uuids(struct json *json, const struct ovsdb_idl_txn *txn) - static void - ovsdb_idl_txn_disassemble(struct ovsdb_idl_txn *txn) - { -- struct ovsdb_idl_row *row, *next; -+ struct ovsdb_idl_row *row; + struct roc_npc_flow * +diff --git a/dpdk/drivers/net/cxgbe/base/adapter.h b/dpdk/drivers/net/cxgbe/base/adapter.h +index 1c7c8afe16..97963422bf 100644 +--- a/dpdk/drivers/net/cxgbe/base/adapter.h ++++ b/dpdk/drivers/net/cxgbe/base/adapter.h +@@ -291,8 +291,6 @@ struct sge { + u32 fl_starve_thres; /* Free List starvation threshold */ + }; - /* This must happen early. Otherwise, ovsdb_idl_row_parse() will call an - * ovsdb_idl_column's 'parse' function, which will call -@@ -2832,7 +2832,7 @@ ovsdb_idl_txn_disassemble(struct ovsdb_idl_txn *txn) - * transaction and fail to update the graph. */ - txn->idl->txn = NULL; +-#define T4_OS_NEEDS_MBOX_LOCKING 1 +- + /* + * OS Lock/List primitives for those interfaces in the Common Code which + * need this. +diff --git a/dpdk/drivers/net/cxgbe/base/t4_hw.c b/dpdk/drivers/net/cxgbe/base/t4_hw.c +index cdcd7e5510..645833765a 100644 +--- a/dpdk/drivers/net/cxgbe/base/t4_hw.c ++++ b/dpdk/drivers/net/cxgbe/base/t4_hw.c +@@ -263,17 +263,6 @@ static void fw_asrt(struct adapter *adap, u32 mbox_addr) + + #define X_CIM_PF_NOACCESS 0xeeeeeeee + +-/* +- * If the Host OS Driver needs locking arround accesses to the mailbox, this +- * can be turned on via the T4_OS_NEEDS_MBOX_LOCKING CPP define ... +- */ +-/* makes single-statement usage a bit cleaner ... */ +-#ifdef T4_OS_NEEDS_MBOX_LOCKING +-#define T4_OS_MBOX_LOCKING(x) x +-#else +-#define T4_OS_MBOX_LOCKING(x) do {} while (0) +-#endif +- + /** + * t4_wr_mbox_meat_timeout - send a command to FW through the given mailbox + * @adap: the adapter +@@ -314,28 +303,17 @@ int t4_wr_mbox_meat_timeout(struct adapter *adap, int mbox, + 1, 1, 3, 5, 10, 10, 20, 50, 100 + }; + +- u32 v; +- u64 res; +- int i, ms; +- unsigned int delay_idx; +- __be64 *temp = (__be64 *)malloc(size * sizeof(char)); +- __be64 *p = temp; + u32 data_reg = PF_REG(mbox, A_CIM_PF_MAILBOX_DATA); + u32 ctl_reg = PF_REG(mbox, A_CIM_PF_MAILBOX_CTRL); +- u32 ctl; +- struct mbox_entry entry; +- u32 pcie_fw = 0; +- +- if (!temp) +- return -ENOMEM; ++ struct mbox_entry *entry; ++ u32 v, ctl, pcie_fw = 0; ++ unsigned int delay_idx; ++ const __be64 *p; ++ int i, ms, ret; ++ u64 res; + +- if ((size & 15) || size > MBOX_LEN) { +- free(temp); ++ if ((size & 15) != 0 || size > MBOX_LEN) + return -EINVAL; +- } +- +- memset(p, 0, size); +- memcpy(p, (const __be64 *)cmd, size); -- HMAP_FOR_EACH_SAFE (row, next, txn_node, &txn->txn_rows) { -+ HMAP_FOR_EACH_SAFE (row, txn_node, &txn->txn_rows) { - enum { INSERTED, MODIFIED, DELETED } op - = (!row->new_datum ? DELETED - : !row->old_datum ? INSERTED -diff --git a/lib/ovsdb-map-op.c b/lib/ovsdb-map-op.c -index 7b90ba84f9..795066e8ef 100644 ---- a/lib/ovsdb-map-op.c -+++ b/lib/ovsdb-map-op.c -@@ -91,8 +91,8 @@ map_op_list_create(void) - void - map_op_list_destroy(struct map_op_list *list, const struct ovsdb_type *type) - { -- struct map_op *map_op, *next; -- HMAP_FOR_EACH_SAFE (map_op, next, node, &list->hmap) { -+ struct map_op *map_op; -+ HMAP_FOR_EACH_SAFE (map_op, node, &list->hmap) { - map_op_destroy(map_op, type); - } - hmap_destroy(&list->hmap); -diff --git a/lib/ovsdb-set-op.c b/lib/ovsdb-set-op.c -index 62c4621181..321043282e 100644 ---- a/lib/ovsdb-set-op.c -+++ b/lib/ovsdb-set-op.c -@@ -90,8 +90,8 @@ set_op_list_create(void) - void - set_op_list_destroy(struct set_op_list *list, const struct ovsdb_type *type) - { -- struct set_op *set_op, *next; -- HMAP_FOR_EACH_SAFE (set_op, next, node, &list->hmap) { -+ struct set_op *set_op; -+ HMAP_FOR_EACH_SAFE (set_op, node, &list->hmap) { - set_op_destroy(set_op, type); - } - hmap_destroy(&list->hmap); -diff --git a/lib/packets.c b/lib/packets.c -index d0fba81766..874066e3c6 100644 ---- a/lib/packets.c -+++ b/lib/packets.c -@@ -427,9 +427,9 @@ add_mpls(struct dp_packet *packet, ovs_be16 ethtype, ovs_be32 lse, - } + /* + * If we have a negative timeout, that implies that we can't sleep. +@@ -345,14 +323,17 @@ int t4_wr_mbox_meat_timeout(struct adapter *adap, int mbox, + timeout = -timeout; + } - if (!l3_encap) { -- ovs_be32 *header = dp_packet_push_uninit(packet, MPLS_HLEN); -+ struct mpls_hdr *header = dp_packet_push_uninit(packet, MPLS_HLEN); +-#ifdef T4_OS_NEEDS_MBOX_LOCKING ++ entry = t4_os_alloc(sizeof(*entry)); ++ if (entry == NULL) ++ return -ENOMEM; ++ + /* + * Queue ourselves onto the mailbox access list. When our entry is at + * the front of the list, we have rights to access the mailbox. So we + * wait [for a while] till we're at the front [or bail out with an + * EBUSY] ... + */ +- t4_os_atomic_add_tail(&entry, &adap->mbox_list, &adap->mbox_lock); ++ t4_os_atomic_add_tail(entry, &adap->mbox_list, &adap->mbox_lock); + + delay_idx = 0; + ms = delay[0]; +@@ -367,18 +348,18 @@ int t4_wr_mbox_meat_timeout(struct adapter *adap, int mbox, + */ + pcie_fw = t4_read_reg(adap, A_PCIE_FW); + if (i > 4 * timeout || (pcie_fw & F_PCIE_FW_ERR)) { +- t4_os_atomic_list_del(&entry, &adap->mbox_list, ++ t4_os_atomic_list_del(entry, &adap->mbox_list, + &adap->mbox_lock); + t4_report_fw_error(adap); +- free(temp); +- return (pcie_fw & F_PCIE_FW_ERR) ? -ENXIO : -EBUSY; ++ ret = ((pcie_fw & F_PCIE_FW_ERR) != 0) ? -ENXIO : -EBUSY; ++ goto out_free; + } + + /* + * If we're at the head, break out and start the mailbox + * protocol. + */ +- if (t4_os_list_first_entry(&adap->mbox_list) == &entry) ++ if (t4_os_list_first_entry(&adap->mbox_list) == entry) + break; + + /* +@@ -393,7 +374,6 @@ int t4_wr_mbox_meat_timeout(struct adapter *adap, int mbox, + rte_delay_ms(ms); + } + } +-#endif /* T4_OS_NEEDS_MBOX_LOCKING */ + + /* + * Attempt to gain access to the mailbox. +@@ -410,12 +390,11 @@ int t4_wr_mbox_meat_timeout(struct adapter *adap, int mbox, + * mailbox atomic access list and report the error to our caller. + */ + if (v != X_MBOWNER_PL) { +- T4_OS_MBOX_LOCKING(t4_os_atomic_list_del(&entry, +- &adap->mbox_list, +- &adap->mbox_lock)); ++ t4_os_atomic_list_del(entry, &adap->mbox_list, ++ &adap->mbox_lock); + t4_report_fw_error(adap); +- free(temp); +- return (v == X_MBOWNER_FW ? -EBUSY : -ETIMEDOUT); ++ ret = (v == X_MBOWNER_FW) ? -EBUSY : -ETIMEDOUT; ++ goto out_free; + } -- *header = lse; -+ put_16aligned_be32(&header->mpls_lse, lse); - packet->l2_5_ofs = 0; - packet->packet_type = PACKET_TYPE_BE(OFPHTN_ETHERTYPE, - ntohs(ethtype)); -diff --git a/lib/pcap-file.c b/lib/pcap-file.c -index 41835f6f4d..3ed7ea4880 100644 ---- a/lib/pcap-file.c -+++ b/lib/pcap-file.c -@@ -344,9 +344,9 @@ tcp_reader_open(void) - void - tcp_reader_close(struct tcp_reader *r) - { -- struct tcp_stream *stream, *next_stream; -+ struct tcp_stream *stream; + /* +@@ -441,7 +420,7 @@ int t4_wr_mbox_meat_timeout(struct adapter *adap, int mbox, + /* + * Copy in the new mailbox command and send it on its way ... + */ +- for (i = 0; i < size; i += 8, p++) ++ for (i = 0, p = cmd; i < size; i += 8, p++) + t4_write_reg64(adap, data_reg + i, be64_to_cpu(*p)); + + CXGBE_DEBUG_MBOX(adap, "%s: mbox %u: %016llx %016llx %016llx %016llx " +@@ -512,11 +491,10 @@ int t4_wr_mbox_meat_timeout(struct adapter *adap, int mbox, + get_mbox_rpl(adap, rpl, size / 8, data_reg); + } + t4_write_reg(adap, ctl_reg, V_MBOWNER(X_MBOWNER_NONE)); +- T4_OS_MBOX_LOCKING( +- t4_os_atomic_list_del(&entry, &adap->mbox_list, +- &adap->mbox_lock)); +- free(temp); +- return -G_FW_CMD_RETVAL((int)res); ++ t4_os_atomic_list_del(entry, &adap->mbox_list, ++ &adap->mbox_lock); ++ ret = -G_FW_CMD_RETVAL((int)res); ++ goto out_free; + } + } -- HMAP_FOR_EACH_SAFE (stream, next_stream, hmap_node, &r->streams) { -+ HMAP_FOR_EACH_SAFE (stream, hmap_node, &r->streams) { - tcp_stream_destroy(r, stream); - } - hmap_destroy(&r->streams); -diff --git a/lib/perf-counter.c b/lib/perf-counter.c -index e4eca58d03..6952fcb594 100644 ---- a/lib/perf-counter.c -+++ b/lib/perf-counter.c -@@ -178,14 +178,14 @@ perf_counters_clear(void) - void - perf_counters_destroy(void) - { -- struct shash_node *node, *next; -+ struct shash_node *node; +@@ -527,12 +505,13 @@ int t4_wr_mbox_meat_timeout(struct adapter *adap, int mbox, + */ + dev_err(adap, "command %#x in mailbox %d timed out\n", + *(const u8 *)cmd, mbox); +- T4_OS_MBOX_LOCKING(t4_os_atomic_list_del(&entry, +- &adap->mbox_list, +- &adap->mbox_lock)); ++ t4_os_atomic_list_del(entry, &adap->mbox_list, &adap->mbox_lock); + t4_report_fw_error(adap); +- free(temp); +- return (pcie_fw & F_PCIE_FW_ERR) ? -ENXIO : -ETIMEDOUT; ++ ret = ((pcie_fw & F_PCIE_FW_ERR) != 0) ? -ENXIO : -ETIMEDOUT; ++ ++out_free: ++ t4_os_free(entry); ++ return ret; + } - if (fd__ != -1) { - ioctl(fd__, PERF_EVENT_IOC_DISABLE, 0); - close(fd__); - } + int t4_wr_mbox_meat(struct adapter *adap, int mbox, const void *cmd, int size, +diff --git a/dpdk/drivers/net/cxgbe/base/t4vf_hw.c b/dpdk/drivers/net/cxgbe/base/t4vf_hw.c +index 561d759dbc..7dbd4deb79 100644 +--- a/dpdk/drivers/net/cxgbe/base/t4vf_hw.c ++++ b/dpdk/drivers/net/cxgbe/base/t4vf_hw.c +@@ -83,7 +83,7 @@ int t4vf_wr_mbox_core(struct adapter *adapter, + + u32 mbox_ctl = T4VF_CIM_BASE_ADDR + A_CIM_VF_EXT_MAILBOX_CTRL; + __be64 cmd_rpl[MBOX_LEN / 8]; +- struct mbox_entry entry; ++ struct mbox_entry *entry; + unsigned int delay_idx; + u32 v, mbox_data; + const __be64 *p; +@@ -106,13 +106,17 @@ int t4vf_wr_mbox_core(struct adapter *adapter, + size > NUM_CIM_VF_MAILBOX_DATA_INSTANCES * 4) + return -EINVAL; + ++ entry = t4_os_alloc(sizeof(*entry)); ++ if (entry == NULL) ++ return -ENOMEM; ++ + /* + * Queue ourselves onto the mailbox access list. When our entry is at + * the front of the list, we have rights to access the mailbox. So we + * wait [for a while] till we're at the front [or bail out with an + * EBUSY] ... + */ +- t4_os_atomic_add_tail(&entry, &adapter->mbox_list, &adapter->mbox_lock); ++ t4_os_atomic_add_tail(entry, &adapter->mbox_list, &adapter->mbox_lock); + + delay_idx = 0; + ms = delay[0]; +@@ -125,17 +129,17 @@ int t4vf_wr_mbox_core(struct adapter *adapter, + * contend on access to the mailbox ... + */ + if (i > (2 * FW_CMD_MAX_TIMEOUT)) { +- t4_os_atomic_list_del(&entry, &adapter->mbox_list, ++ t4_os_atomic_list_del(entry, &adapter->mbox_list, + &adapter->mbox_lock); + ret = -EBUSY; +- return ret; ++ goto out_free; + } + + /* + * If we're at the head, break out and start the mailbox + * protocol. + */ +- if (t4_os_list_first_entry(&adapter->mbox_list) == &entry) ++ if (t4_os_list_first_entry(&adapter->mbox_list) == entry) + break; + + /* +@@ -160,10 +164,10 @@ int t4vf_wr_mbox_core(struct adapter *adapter, + v = G_MBOWNER(t4_read_reg(adapter, mbox_ctl)); + + if (v != X_MBOWNER_PL) { +- t4_os_atomic_list_del(&entry, &adapter->mbox_list, ++ t4_os_atomic_list_del(entry, &adapter->mbox_list, + &adapter->mbox_lock); + ret = (v == X_MBOWNER_FW) ? -EBUSY : -ETIMEDOUT; +- return ret; ++ goto out_free; + } -- SHASH_FOR_EACH_SAFE (node, next, &perf_counters) { -+ SHASH_FOR_EACH_SAFE (node, &perf_counters) { - shash_delete(&perf_counters, node); - } + /* +@@ -224,7 +228,7 @@ int t4vf_wr_mbox_core(struct adapter *adapter, + get_mbox_rpl(adapter, cmd_rpl, size / 8, mbox_data); + t4_write_reg(adapter, mbox_ctl, + V_MBOWNER(X_MBOWNER_NONE)); +- t4_os_atomic_list_del(&entry, &adapter->mbox_list, ++ t4_os_atomic_list_del(entry, &adapter->mbox_list, + &adapter->mbox_lock); + + /* return value in high-order host-endian word */ +@@ -236,7 +240,8 @@ int t4vf_wr_mbox_core(struct adapter *adapter, + & F_FW_CMD_REQUEST) == 0); + memcpy(rpl, cmd_rpl, size); + } +- return -((int)G_FW_CMD_RETVAL(v)); ++ ret = -((int)G_FW_CMD_RETVAL(v)); ++ goto out_free; + } + } -diff --git a/lib/poll-loop.c b/lib/poll-loop.c -index 4e751ff2c7..70fabeb8a3 100644 ---- a/lib/poll-loop.c -+++ b/lib/poll-loop.c -@@ -298,9 +298,9 @@ log_wakeup(const char *where, const struct pollfd *pollfd, int timeout) - static void - free_poll_nodes(struct poll_loop *loop) +@@ -246,8 +251,11 @@ int t4vf_wr_mbox_core(struct adapter *adapter, + dev_err(adapter, "command %#x timed out\n", + *(const u8 *)cmd); + dev_err(adapter, " Control = %#x\n", t4_read_reg(adapter, mbox_ctl)); +- t4_os_atomic_list_del(&entry, &adapter->mbox_list, &adapter->mbox_lock); ++ t4_os_atomic_list_del(entry, &adapter->mbox_list, &adapter->mbox_lock); + ret = -ETIMEDOUT; ++ ++out_free: ++ t4_os_free(entry); + return ret; + } + +diff --git a/dpdk/drivers/net/cxgbe/cxgbe_flow.c b/dpdk/drivers/net/cxgbe/cxgbe_flow.c +index edcbba9d7c..6e460dfe2e 100644 +--- a/dpdk/drivers/net/cxgbe/cxgbe_flow.c ++++ b/dpdk/drivers/net/cxgbe/cxgbe_flow.c +@@ -1378,7 +1378,7 @@ cxgbe_flow_validate(struct rte_eth_dev *dev, + } + + /* +- * @ret : > 0 filter destroyed succsesfully ++ * @ret : > 0 filter destroyed successfully + * < 0 error destroying filter + * == 1 filter not active / not found + */ +diff --git a/dpdk/drivers/net/cxgbe/cxgbevf_main.c b/dpdk/drivers/net/cxgbe/cxgbevf_main.c +index f639612ae4..d0c93f8ac3 100644 +--- a/dpdk/drivers/net/cxgbe/cxgbevf_main.c ++++ b/dpdk/drivers/net/cxgbe/cxgbevf_main.c +@@ -44,7 +44,7 @@ static void size_nports_qsets(struct adapter *adapter) + */ + pmask_nports = hweight32(adapter->params.vfres.pmask); + if (pmask_nports < adapter->params.nports) { +- dev_warn(adapter->pdev_dev, "only using %d of %d provissioned" ++ dev_warn(adapter->pdev_dev, "only using %d of %d provisioned" + " virtual interfaces; limited by Port Access Rights" + " mask %#x\n", pmask_nports, adapter->params.nports, + adapter->params.vfres.pmask); +diff --git a/dpdk/drivers/net/cxgbe/sge.c b/dpdk/drivers/net/cxgbe/sge.c +index f623f3e684..566cd48406 100644 +--- a/dpdk/drivers/net/cxgbe/sge.c ++++ b/dpdk/drivers/net/cxgbe/sge.c +@@ -211,7 +211,7 @@ static inline unsigned int fl_cap(const struct sge_fl *fl) + * @fl: the Free List + * + * Tests specified Free List to see whether the number of buffers +- * available to the hardware has falled below our "starvation" ++ * available to the hardware has fallen below our "starvation" + * threshold. + */ + static inline bool fl_starving(const struct adapter *adapter, +@@ -678,7 +678,7 @@ static void write_sgl(struct rte_mbuf *mbuf, struct sge_txq *q, + * @q: the Tx queue + * @n: number of new descriptors to give to HW + * +- * Ring the doorbel for a Tx queue. ++ * Ring the doorbell for a Tx queue. + */ + static inline void ring_tx_db(struct adapter *adap, struct sge_txq *q) { -- struct poll_node *node, *next; -+ struct poll_node *node; +@@ -789,9 +789,9 @@ static inline void txq_advance(struct sge_txq *q, unsigned int n) -- HMAP_FOR_EACH_SAFE (node, next, hmap_node, &loop->poll_nodes) { -+ HMAP_FOR_EACH_SAFE (node, hmap_node, &loop->poll_nodes) { - hmap_remove(&loop->poll_nodes, &node->hmap_node); - #ifdef _WIN32 - if (node->wevent && node->pollfd.fd) { -diff --git a/lib/rculist.h b/lib/rculist.h -index 1072b87af2..c0d77acf94 100644 ---- a/lib/rculist.h -+++ b/lib/rculist.h -@@ -365,35 +365,57 @@ rculist_is_singleton_protected(const struct rculist *list) - return list_next == list->prev && list_next != list; + #define MAX_COALESCE_LEN 64000 + +-static inline int wraps_around(struct sge_txq *q, int ndesc) ++static inline bool wraps_around(struct sge_txq *q, int ndesc) + { +- return (q->pidx + ndesc) > q->size ? 1 : 0; ++ return (q->pidx + ndesc) > q->size ? true : false; } --#define RCULIST_FOR_EACH(ITER, MEMBER, RCULIST) \ -- for (INIT_CONTAINER(ITER, rculist_next(RCULIST), MEMBER); \ -- &(ITER)->MEMBER != (RCULIST); \ -- ASSIGN_CONTAINER(ITER, rculist_next(&(ITER)->MEMBER), MEMBER)) --#define RCULIST_FOR_EACH_CONTINUE(ITER, MEMBER, RCULIST) \ -- for (ASSIGN_CONTAINER(ITER, rculist_next(&(ITER)->MEMBER), MEMBER); \ -- &(ITER)->MEMBER != (RCULIST); \ -- ASSIGN_CONTAINER(ITER, rculist_next(&(ITER)->MEMBER), MEMBER)) -- --#define RCULIST_FOR_EACH_REVERSE_PROTECTED(ITER, MEMBER, RCULIST) \ -- for (INIT_CONTAINER(ITER, (RCULIST)->prev, MEMBER); \ -- &(ITER)->MEMBER != (RCULIST); \ -- ASSIGN_CONTAINER(ITER, (ITER)->MEMBER.prev, MEMBER)) --#define RCULIST_FOR_EACH_REVERSE_PROTECTED_CONTINUE(ITER, MEMBER, RCULIST) \ -- for (ASSIGN_CONTAINER(ITER, (ITER)->MEMBER.prev, MEMBER); \ -- &(ITER)->MEMBER != (RCULIST); \ -- ASSIGN_CONTAINER(ITER, (ITER)->MEMBER.prev, MEMBER)) + static void tx_timer_cb(void *data) +@@ -842,7 +842,6 @@ static inline void ship_tx_pkt_coalesce_wr(struct adapter *adap, + + /* fill the pkts WR header */ + wr = (void *)&q->desc[q->pidx]; +- wr->op_pkd = htonl(V_FW_WR_OP(FW_ETH_TX_PKTS2_WR)); + vmwr = (void *)&q->desc[q->pidx]; + + wr_mid = V_FW_WR_LEN16(DIV_ROUND_UP(q->coalesce.flits, 2)); +@@ -852,8 +851,11 @@ static inline void ship_tx_pkt_coalesce_wr(struct adapter *adap, + wr->npkt = q->coalesce.idx; + wr->r3 = 0; + if (is_pf4(adap)) { +- wr->op_pkd = htonl(V_FW_WR_OP(FW_ETH_TX_PKTS2_WR)); + wr->type = q->coalesce.type; ++ if (likely(wr->type != 0)) ++ wr->op_pkd = htonl(V_FW_WR_OP(FW_ETH_TX_PKTS2_WR)); ++ else ++ wr->op_pkd = htonl(V_FW_WR_OP(FW_ETH_TX_PKTS_WR)); + } else { + wr->op_pkd = htonl(V_FW_WR_OP(FW_ETH_TX_PKTS_VM_WR)); + vmwr->r4 = 0; +@@ -877,7 +879,7 @@ static inline void ship_tx_pkt_coalesce_wr(struct adapter *adap, + } + + /** +- * should_tx_packet_coalesce - decides wether to coalesce an mbuf or not ++ * should_tx_packet_coalesce - decides whether to coalesce an mbuf or not + * @txq: tx queue where the mbuf is sent + * @mbuf: mbuf to be sent + * @nflits: return value for number of flits needed +@@ -932,13 +934,16 @@ static inline int should_tx_packet_coalesce(struct sge_eth_txq *txq, + ndesc = DIV_ROUND_UP(q->coalesce.flits + flits, 8); + credits = txq_avail(q) - ndesc; + ++ if (unlikely(wraps_around(q, ndesc))) ++ return 0; ++ + /* If we are wrapping or this is last mbuf then, send the + * already coalesced mbufs and let the non-coalesce pass + * handle the mbuf. + */ +- if (unlikely(credits < 0 || wraps_around(q, ndesc))) { ++ if (unlikely(credits < 0)) { + ship_tx_pkt_coalesce_wr(adap, txq); +- return 0; ++ return -EBUSY; + } + + /* If the max coalesce len or the max WR len is reached +@@ -962,8 +967,12 @@ static inline int should_tx_packet_coalesce(struct sge_eth_txq *txq, + ndesc = flits_to_desc(q->coalesce.flits + flits); + credits = txq_avail(q) - ndesc; + +- if (unlikely(credits < 0 || wraps_around(q, ndesc))) ++ if (unlikely(wraps_around(q, ndesc))) + return 0; ++ ++ if (unlikely(credits < 0)) ++ return -EBUSY; ++ + q->coalesce.flits += wr_size / sizeof(__be64); + q->coalesce.type = type; + q->coalesce.ptr = (unsigned char *)&q->desc[q->pidx] + +@@ -1106,7 +1115,7 @@ int t4_eth_xmit(struct sge_eth_txq *txq, struct rte_mbuf *mbuf, + unsigned int flits, ndesc, cflits; + int l3hdr_len, l4hdr_len, eth_xtra_len; + int len, last_desc; +- int credits; ++ int should_coal, credits; + u32 wr_mid; + u64 cntrl, *end; + bool v6; +@@ -1138,9 +1147,9 @@ int t4_eth_xmit(struct sge_eth_txq *txq, struct rte_mbuf *mbuf, + /* align the end of coalesce WR to a 512 byte boundary */ + txq->q.coalesce.max = (8 - (txq->q.pidx & 7)) * 8; + +- if (!((m->ol_flags & RTE_MBUF_F_TX_TCP_SEG) || +- m->pkt_len > RTE_ETHER_MAX_LEN)) { +- if (should_tx_packet_coalesce(txq, mbuf, &cflits, adap)) { ++ if ((m->ol_flags & RTE_MBUF_F_TX_TCP_SEG) == 0) { ++ should_coal = should_tx_packet_coalesce(txq, mbuf, &cflits, adap); ++ if (should_coal > 0) { + if (unlikely(map_mbuf(mbuf, addr) < 0)) { + dev_warn(adap, "%s: mapping err for coalesce\n", + __func__); +@@ -1149,8 +1158,8 @@ int t4_eth_xmit(struct sge_eth_txq *txq, struct rte_mbuf *mbuf, + } + return tx_do_packet_coalesce(txq, mbuf, cflits, adap, + pi, addr, nb_pkts); +- } else { +- return -EBUSY; ++ } else if (should_coal < 0) { ++ return should_coal; + } + } + +@@ -1197,8 +1206,7 @@ int t4_eth_xmit(struct sge_eth_txq *txq, struct rte_mbuf *mbuf, + end = (u64 *)vmwr + flits; + } + +- len = 0; +- len += sizeof(*cpl); ++ len = sizeof(*cpl); + + /* Coalescing skipped and we send through normal path */ + if (!(m->ol_flags & RTE_MBUF_F_TX_TCP_SEG)) { +@@ -1846,7 +1854,7 @@ int t4_sge_alloc_rxq(struct adapter *adap, struct sge_rspq *iq, bool fwevtq, + * for its status page) along with the associated software + * descriptor ring. The free list size needs to be a multiple + * of the Egress Queue Unit and at least 2 Egress Units larger +- * than the SGE's Egress Congrestion Threshold ++ * than the SGE's Egress Congestion Threshold + * (fl_starve_thres - 1). + */ + if (fl->size < s->fl_starve_thres - 1 + 2 * 8) +@@ -1910,7 +1918,7 @@ int t4_sge_alloc_rxq(struct adapter *adap, struct sge_rspq *iq, bool fwevtq, + iq->stat = (void *)&iq->desc[iq->size * 8]; + iq->eth_dev = eth_dev; + iq->handler = hnd; +- iq->port_id = pi->pidx; ++ iq->port_id = eth_dev->data->port_id; + iq->mb_pool = mp; + + /* set offset to -1 to distinguish ingress queues without FL */ +diff --git a/dpdk/drivers/net/dpaa/dpaa_ethdev.c b/dpdk/drivers/net/dpaa/dpaa_ethdev.c +index e49f765434..9847ca1be1 100644 +--- a/dpdk/drivers/net/dpaa/dpaa_ethdev.c ++++ b/dpdk/drivers/net/dpaa/dpaa_ethdev.c +@@ -1030,7 +1030,7 @@ int dpaa_eth_rx_queue_setup(struct rte_eth_dev *dev, uint16_t queue_idx, + QM_FQCTRL_CTXASTASHING | + QM_FQCTRL_PREFERINCACHE; + opts.fqd.context_a.stashing.exclusive = 0; +- /* In muticore scenario stashing becomes a bottleneck on LS1046. ++ /* In multicore scenario stashing becomes a bottleneck on LS1046. + * So do not enable stashing in this case + */ + if (dpaa_svr_family != SVR_LS1046A_FAMILY) +@@ -1201,23 +1201,17 @@ int + dpaa_eth_eventq_detach(const struct rte_eth_dev *dev, + int eth_rx_queue_id) + { +- struct qm_mcc_initfq opts; ++ struct qm_mcc_initfq opts = {0}; + int ret; + u32 flags = 0; + struct dpaa_if *dpaa_intf = dev->data->dev_private; + struct qman_fq *rxq = &dpaa_intf->rx_queues[eth_rx_queue_id]; + +- dpaa_poll_queue_default_config(&opts); - --#define RCULIST_FOR_EACH_PROTECTED(ITER, MEMBER, RCULIST) \ -- for (INIT_CONTAINER(ITER, rculist_next_protected(RCULIST), MEMBER); \ -- &(ITER)->MEMBER != (RCULIST); \ -- ASSIGN_CONTAINER(ITER, rculist_next_protected(&(ITER)->MEMBER), \ -- MEMBER)) +- if (dpaa_intf->cgr_rx) { +- opts.we_mask |= QM_INITFQ_WE_CGID; +- opts.fqd.cgid = dpaa_intf->cgr_rx[eth_rx_queue_id].cgrid; +- opts.fqd.fq_ctrl |= QM_FQCTRL_CGE; +- } - --#define RCULIST_FOR_EACH_SAFE_PROTECTED(ITER, NEXT, MEMBER, RCULIST) \ -- for (INIT_CONTAINER(ITER, rculist_next_protected(RCULIST), MEMBER); \ -- (&(ITER)->MEMBER != (RCULIST) \ -- ? INIT_CONTAINER(NEXT, rculist_next_protected(&(ITER)->MEMBER), \ -- MEMBER), 1 : 0); \ -- (ITER) = (NEXT)) -+#define RCULIST_FOR_EACH(ITER, MEMBER, RCULIST) \ -+ for (INIT_MULTIVAR(ITER, MEMBER, rculist_next(RCULIST), \ -+ const struct rculist); \ -+ CONDITION_MULTIVAR(ITER, MEMBER, ITER_VAR(ITER) != (RCULIST)); \ -+ UPDATE_MULTIVAR(ITER, rculist_next(ITER_VAR(ITER)))) -+ -+#define RCULIST_FOR_EACH_CONTINUE(ITER, MEMBER, RCULIST) \ -+ for (INIT_MULTIVAR(ITER, MEMBER, rculist_next(&(ITER)->MEMBER), \ -+ const struct rculist); \ -+ CONDITION_MULTIVAR(ITER, MEMBER, ITER_VAR(ITER) != (RCULIST)); \ -+ UPDATE_MULTIVAR(ITER, rculist_next(ITER_VAR(ITER)))) -+ -+#define RCULIST_FOR_EACH_REVERSE_PROTECTED(ITER, MEMBER, RCULIST) \ -+ for (INIT_MULTIVAR(ITER, MEMBER, (RCULIST)->prev, struct rculist); \ -+ CONDITION_MULTIVAR(ITER, MEMBER, ITER_VAR(ITER) != (RCULIST)); \ -+ UPDATE_MULTIVAR(ITER, ITER_VAR(VAR).prev)) -+ -+#define RCULIST_FOR_EACH_REVERSE_PROTECTED_CONTINUE(ITER, MEMBER, RCULIST) \ -+ for (INIT_MULTIVAR(ITER, MEMBER, (ITER)->MEMBER.prev, struct rculist); \ -+ CONDITION_MULTIVAR(ITER, MEMBER, ITER_VAR(ITER) != (RCULIST)); \ -+ UPDATE_MULTIVAR(ITER, ITER_VAR(VAR).prev)) -+ -+#define RCULIST_FOR_EACH_PROTECTED(ITER, MEMBER, RCULIST) \ -+ for (INIT_MULTIVAR(ITER, MEMBER, rculist_next_protected(RCULIST), \ -+ struct rculist); \ -+ CONDITION_MULTIVAR(ITER, MEMBER, ITER_VAR(ITER) != (RCULIST)); \ -+ UPDATE_MULTIVAR(ITER, rculist_next_protected(ITER_VAR(ITER))) \ -+ -+#define RCULIST_FOR_EACH_SAFE_SHORT_PROTECTED(ITER, MEMBER, RCULIST) \ -+ for (INIT_MULTIVAR_SAFE_SHORT(ITER, MEMBER, \ -+ rculist_next_protected(RCULIST), \ -+ struct rculist); \ -+ CONDITION_MULTIVAR_SAFE_SHORT(ITER, MEMBER, \ -+ ITER_VAR(ITER) != (RCULIST), \ -+ ITER_NEXT_VAR(ITER) = rculist_next_protected(ITER_VAR(VAR))); \ -+ UPDATE_MULTIVAR_SHORT(ITER)) -+ -+#define RCULIST_FOR_EACH_SAFE_LONG_PROTECTED(ITER, NEXT, MEMBER, RCULIST) \ -+ for (INIT_MULTIVAR_SAFE_LONG(ITER, NEXT, MEMBER, \ -+ rculist_next_protected(RCULIST) \ -+ struct rculist); \ -+ CONDITION_MULTIVAR_SAFE_LONG(VAR, NEXT, MEMBER \ -+ ITER_VAR(ITER) != (RCULIST), \ -+ ITER_VAR(NEXT) = rculist_next_protected(ITER_VAR(VAR)), \ -+ ITER_VAR(NEXT) != (RCULIST)); \ -+ UPDATE_MULTIVAR_LONG(ITER)) -+ -+#define RCULIST_FOR_EACH_SAFE_PROTECTED(...) \ -+ OVERLOAD_SAFE_MACRO(RCULIST_FOR_EACH_SAFE_LONG_PROTECTED, \ -+ RCULIST_FOR_EACH_SAFE_SHORT_PROTECTED, \ -+ 4, __VA_ARGS__) ++ qman_retire_fq(rxq, NULL); ++ qman_oos_fq(rxq); + ret = qman_init_fq(rxq, flags, &opts); + if (ret) { +- DPAA_PMD_ERR("init rx fqid %d failed with ret: %d", ++ DPAA_PMD_ERR("detach rx fqid %d failed with ret: %d", + rxq->fqid, ret); + } + +@@ -1866,7 +1860,7 @@ dpaa_dev_init(struct rte_eth_dev *eth_dev) + + dpaa_intf->name = dpaa_device->name; + +- /* save fman_if & cfg in the interface struture */ ++ /* save fman_if & cfg in the interface structure */ + eth_dev->process_private = fman_intf; + dpaa_intf->ifid = dev_id; + dpaa_intf->cfg = cfg; +@@ -2169,7 +2163,7 @@ rte_dpaa_probe(struct rte_dpaa_driver *dpaa_drv, + if (dpaa_svr_family == SVR_LS1043A_FAMILY) + dpaa_push_mode_max_queue = 0; + +- /* if push mode queues to be enabled. Currenly we are allowing ++ /* if push mode queues to be enabled. Currently we are allowing + * only one queue per thread. + */ + if (getenv("DPAA_PUSH_QUEUES_NUMBER")) { +diff --git a/dpdk/drivers/net/dpaa/dpaa_rxtx.c b/dpdk/drivers/net/dpaa/dpaa_rxtx.c +index ffac6ce3e2..956fe946fa 100644 +--- a/dpdk/drivers/net/dpaa/dpaa_rxtx.c ++++ b/dpdk/drivers/net/dpaa/dpaa_rxtx.c +@@ -600,8 +600,8 @@ void dpaa_rx_cb_prepare(struct qm_dqrr_entry *dq, void **bufs) + void *ptr = rte_dpaa_mem_ptov(qm_fd_addr(&dq->fd)); + + /* In case of LS1046, annotation stashing is disabled due to L2 cache +- * being bottleneck in case of multicore scanario for this platform. +- * So we prefetch the annoation beforehand, so that it is available ++ * being bottleneck in case of multicore scenario for this platform. ++ * So we prefetch the annotation beforehand, so that it is available + * in cache when accessed. + */ + rte_prefetch0((void *)((uint8_t *)ptr + DEFAULT_RX_ICEOF)); +diff --git a/dpdk/drivers/net/dpaa/fmlib/fm_ext.h b/dpdk/drivers/net/dpaa/fmlib/fm_ext.h +index 27c9fb471e..8e7153bdaf 100644 +--- a/dpdk/drivers/net/dpaa/fmlib/fm_ext.h ++++ b/dpdk/drivers/net/dpaa/fmlib/fm_ext.h +@@ -176,7 +176,7 @@ typedef struct t_fm_prs_result { + #define FM_FD_ERR_PRS_HDR_ERR 0x00000020 + /**< Header error was identified during parsing */ + #define FM_FD_ERR_BLOCK_LIMIT_EXCEEDED 0x00000008 +- /**< Frame parsed beyind 256 first bytes */ ++ /**< Frame parsed beyond 256 first bytes */ + + #define FM_FD_TX_STATUS_ERR_MASK (FM_FD_ERR_UNSUPPORTED_FORMAT | \ + FM_FD_ERR_LENGTH | \ +diff --git a/dpdk/drivers/net/dpaa/fmlib/fm_pcd_ext.h b/dpdk/drivers/net/dpaa/fmlib/fm_pcd_ext.h +index 8be3885fbc..3802b42916 100644 +--- a/dpdk/drivers/net/dpaa/fmlib/fm_pcd_ext.h ++++ b/dpdk/drivers/net/dpaa/fmlib/fm_pcd_ext.h +@@ -276,7 +276,7 @@ typedef struct ioc_fm_pcd_counters_params_t { + } ioc_fm_pcd_counters_params_t; + + /* +- * @Description structure for FM exception definitios ++ * @Description structure for FM exception definitions + */ + typedef struct ioc_fm_pcd_exception_params_t { + ioc_fm_pcd_exceptions exception; /**< The requested exception */ +@@ -883,7 +883,7 @@ typedef enum ioc_fm_pcd_manip_hdr_rmv_specific_l2 { + e_IOC_FM_PCD_MANIP_HDR_RMV_ETHERNET, /**< Ethernet/802.3 MAC */ + e_IOC_FM_PCD_MANIP_HDR_RMV_STACKED_QTAGS, /**< stacked QTags */ + e_IOC_FM_PCD_MANIP_HDR_RMV_ETHERNET_AND_MPLS, +- /**< MPLS and Ethernet/802.3 MAC header unitl the header ++ /**< MPLS and Ethernet/802.3 MAC header until the header + * which follows the MPLS header + */ + e_IOC_FM_PCD_MANIP_HDR_RMV_MPLS +@@ -3293,7 +3293,7 @@ typedef struct ioc_fm_pcd_cc_tbl_get_stats_t { + /* + * @Function fm_pcd_net_env_characteristics_delete + * +- * @Description Deletes a set of Network Environment Charecteristics. ++ * @Description Deletes a set of Network Environment Characteristics. + * + * @Param[in] ioc_fm_obj_t The id of a Network Environment object. + * +@@ -3493,7 +3493,7 @@ typedef struct ioc_fm_pcd_cc_tbl_get_stats_t { + * @Return 0 on success; Error code otherwise. + * + * @Cautions Allowed only following fm_pcd_match_table_set() not only of +- * the relevnt node but also the node that points to this node. ++ * the relevant node but also the node that points to this node. + */ + #define FM_PCD_IOC_MATCH_TABLE_MODIFY_KEY_AND_NEXT_ENGINE \ + _IOW(FM_IOC_TYPE_BASE, FM_PCD_IOC_NUM(35), \ +diff --git a/dpdk/drivers/net/dpaa/fmlib/fm_port_ext.h b/dpdk/drivers/net/dpaa/fmlib/fm_port_ext.h +index 6f5479fbe1..bb2e00222e 100644 +--- a/dpdk/drivers/net/dpaa/fmlib/fm_port_ext.h ++++ b/dpdk/drivers/net/dpaa/fmlib/fm_port_ext.h +@@ -498,7 +498,7 @@ typedef struct ioc_fm_port_pcd_prs_params_t { + /**< Number of bytes from beginning of packet to start parsing + */ + ioc_net_header_type first_prs_hdr; +- /**< The type of the first header axpected at 'parsing_offset' ++ /**< The type of the first header expected at 'parsing_offset' + */ + bool include_in_prs_statistics; + /**< TRUE to include this port in the parser statistics */ +@@ -524,7 +524,7 @@ typedef struct ioc_fm_port_pcd_prs_params_t { + } ioc_fm_port_pcd_prs_params_t; + + /* +- * @Description A structure for defining coarse alassification parameters ++ * @Description A structure for defining coarse classification parameters + * (Must match t_fm_portPcdCcParams defined in fm_port_ext.h) + */ + typedef struct ioc_fm_port_pcd_cc_params_t { +@@ -602,7 +602,7 @@ typedef struct ioc_fm_pcd_prs_start_t { + /**< Number of bytes from beginning of packet to start parsing + */ + ioc_net_header_type first_prs_hdr; +- /**< The type of the first header axpected at 'parsing_offset' ++ /**< The type of the first header expected at 'parsing_offset' + */ + } ioc_fm_pcd_prs_start_t; + +@@ -1356,7 +1356,7 @@ typedef uint32_t fm_port_frame_err_select_t; + #define FM_PORT_FRM_ERR_PRS_HDR_ERR FM_FD_ERR_PRS_HDR_ERR + /**< Header error was identified during parsing */ + #define FM_PORT_FRM_ERR_BLOCK_LIMIT_EXCEEDED FM_FD_ERR_BLOCK_LIMIT_EXCEEDED +- /**< Frame parsed beyind 256 first bytes */ ++ /**< Frame parsed beyond 256 first bytes */ + #define FM_PORT_FRM_ERR_PROCESS_TIMEOUT 0x00000001 + /**< FPM Frame Processing Timeout Exceeded */ + /* @} */ +@@ -1390,7 +1390,7 @@ typedef void (t_fm_port_exception_callback) (t_handle h_app, + * @Param[in] length length of received data + * @Param[in] status receive status and errors + * @Param[in] position position of buffer in frame +- * @Param[in] h_buf_context A handle of the user acossiated with this buffer ++ * @Param[in] h_buf_context A handle of the user associated with this buffer + * + * @Retval e_RX_STORE_RESPONSE_CONTINUE + * order the driver to continue Rx operation for all ready data. +@@ -1414,7 +1414,7 @@ typedef e_rx_store_response(t_fm_port_im_rx_store_callback) (t_handle h_app, + * @Param[in] p_data A pointer to data received + * @Param[in] status transmit status and errors + * @Param[in] last_buffer is last buffer in frame +- * @Param[in] h_buf_context A handle of the user acossiated with this buffer ++ * @Param[in] h_buf_context A handle of the user associated with this buffer + */ + typedef void (t_fm_port_im_tx_conf_callback) (t_handle h_app, + uint8_t *p_data, +@@ -2585,7 +2585,7 @@ typedef struct t_fm_port_congestion_grps { + bool pfc_prio_enable[FM_NUM_CONG_GRPS][FM_MAX_PFC_PRIO]; + /**< a matrix that represents the map between the CG ids + * defined in 'congestion_grps_to_consider' to the +- * priorties mapping array. ++ * priorities mapping array. + */ + } t_fm_port_congestion_grps; + +diff --git a/dpdk/drivers/net/dpaa2/dpaa2_ethdev.c b/dpdk/drivers/net/dpaa2/dpaa2_ethdev.c +index a3706439d5..b875139689 100644 +--- a/dpdk/drivers/net/dpaa2/dpaa2_ethdev.c ++++ b/dpdk/drivers/net/dpaa2/dpaa2_ethdev.c +@@ -143,7 +143,7 @@ dpaa2_vlan_offload_set(struct rte_eth_dev *dev, int mask) + PMD_INIT_FUNC_TRACE(); + + if (mask & RTE_ETH_VLAN_FILTER_MASK) { +- /* VLAN Filter not avaialble */ ++ /* VLAN Filter not available */ + if (!priv->max_vlan_filters) { + DPAA2_PMD_INFO("VLAN filter not available"); + return -ENOTSUP; +@@ -395,6 +395,8 @@ dpaa2_alloc_rx_tx_queues(struct rte_eth_dev *dev) + if (dpaa2_enable_err_queue) { + priv->rx_err_vq = rte_zmalloc("dpni_rx_err", + sizeof(struct dpaa2_queue), 0); ++ if (!priv->rx_err_vq) ++ goto fail; + + dpaa2_q = (struct dpaa2_queue *)priv->rx_err_vq; + dpaa2_q->q_storage = rte_malloc("err_dq_storage", +@@ -916,7 +918,7 @@ dpaa2_dev_tx_queue_setup(struct rte_eth_dev *dev, + cong_notif_cfg.units = DPNI_CONGESTION_UNIT_FRAMES; + cong_notif_cfg.threshold_entry = nb_tx_desc; + /* Notify that the queue is not congested when the data in +- * the queue is below this thershold.(90% of value) ++ * the queue is below this threshold.(90% of value) + */ + cong_notif_cfg.threshold_exit = (nb_tx_desc * 9) / 10; + cong_notif_cfg.message_ctx = 0; +@@ -1058,7 +1060,7 @@ dpaa2_supported_ptypes_get(struct rte_eth_dev *dev) + * Dpaa2 link Interrupt handler + * + * @param param +- * The address of parameter (struct rte_eth_dev *) regsitered before. ++ * The address of parameter (struct rte_eth_dev *) registered before. + * + * @return + * void +@@ -1252,7 +1254,12 @@ dpaa2_dev_stop(struct rte_eth_dev *dev) + struct fsl_mc_io *dpni = (struct fsl_mc_io *)dev->process_private; + int ret; + struct rte_eth_link link; +- struct rte_intr_handle *intr_handle = dev->intr_handle; ++ struct rte_device *rdev = dev->device; ++ struct rte_intr_handle *intr_handle; ++ struct rte_dpaa2_device *dpaa2_dev; ++ ++ dpaa2_dev = container_of(rdev, struct rte_dpaa2_device, device); ++ intr_handle = dpaa2_dev->intr_handle; + + PMD_INIT_FUNC_TRACE(); + +@@ -2236,7 +2243,7 @@ int dpaa2_eth_eventq_attach(const struct rte_eth_dev *dev, + ocfg.oa = 1; + /* Late arrival window size disabled */ + ocfg.olws = 0; +- /* ORL resource exhaustaion advance NESN disabled */ ++ /* ORL resource exhaustion advance NESN disabled */ + ocfg.oeane = 0; + /* Loose ordering enabled */ + ocfg.oloe = 1; +@@ -2720,13 +2727,13 @@ dpaa2_dev_init(struct rte_eth_dev *eth_dev) + } + eth_dev->tx_pkt_burst = dpaa2_dev_tx; + +- /*Init fields w.r.t. classficaition*/ ++ /* Init fields w.r.t. classification */ + memset(&priv->extract.qos_key_extract, 0, + sizeof(struct dpaa2_key_extract)); + priv->extract.qos_extract_param = (size_t)rte_malloc(NULL, 256, 64); + if (!priv->extract.qos_extract_param) { + DPAA2_PMD_ERR(" Error(%d) in allocation resources for flow " +- " classificaiton ", ret); ++ " classification ", ret); + goto init_err; + } + priv->extract.qos_key_extract.key_info.ipv4_src_offset = +@@ -2744,7 +2751,7 @@ dpaa2_dev_init(struct rte_eth_dev *eth_dev) + priv->extract.tc_extract_param[i] = + (size_t)rte_malloc(NULL, 256, 64); + if (!priv->extract.tc_extract_param[i]) { +- DPAA2_PMD_ERR(" Error(%d) in allocation resources for flow classificaiton", ++ DPAA2_PMD_ERR(" Error(%d) in allocation resources for flow classification", + ret); + goto init_err; + } +diff --git a/dpdk/drivers/net/dpaa2/dpaa2_ethdev.h b/dpdk/drivers/net/dpaa2/dpaa2_ethdev.h +index c5e9267bf0..fd4eabed4e 100644 +--- a/dpdk/drivers/net/dpaa2/dpaa2_ethdev.h ++++ b/dpdk/drivers/net/dpaa2/dpaa2_ethdev.h +@@ -62,7 +62,7 @@ + /* Disable RX tail drop, default is enable */ + #define DPAA2_RX_TAILDROP_OFF 0x04 + /* Tx confirmation enabled */ +-#define DPAA2_TX_CONF_ENABLE 0x08 ++#define DPAA2_TX_CONF_ENABLE 0x06 + + #define DPAA2_RSS_OFFLOAD_ALL ( \ + RTE_ETH_RSS_L2_PAYLOAD | \ +@@ -117,7 +117,7 @@ extern int dpaa2_timestamp_dynfield_offset; + + #define DPAA2_FLOW_MAX_KEY_SIZE 16 + +-/*Externaly defined*/ ++/* Externally defined */ + extern const struct rte_flow_ops dpaa2_flow_ops; + + extern const struct rte_tm_ops dpaa2_tm_ops; +diff --git a/dpdk/drivers/net/dpaa2/dpaa2_flow.c b/dpdk/drivers/net/dpaa2/dpaa2_flow.c +index 84fe37a7c0..bf55eb70a3 100644 +--- a/dpdk/drivers/net/dpaa2/dpaa2_flow.c ++++ b/dpdk/drivers/net/dpaa2/dpaa2_flow.c +@@ -1451,7 +1451,7 @@ dpaa2_configure_flow_generic_ip( + flow, pattern, &local_cfg, + device_configured, group); + if (ret) { +- DPAA2_PMD_ERR("IP discrimation failed!"); ++ DPAA2_PMD_ERR("IP discrimination failed!"); + return -1; + } + +@@ -3349,7 +3349,7 @@ dpaa2_flow_verify_action( + (actions[j].conf); + if (rss_conf->queue_num > priv->dist_queues) { + DPAA2_PMD_ERR( +- "RSS number exceeds the distrbution size"); ++ "RSS number exceeds the distribution size"); + return -ENOTSUP; + } + for (i = 0; i < (int)rss_conf->queue_num; i++) { +@@ -3596,7 +3596,7 @@ dpaa2_generic_flow_set(struct rte_flow *flow, + qos_cfg.keep_entries = true; + qos_cfg.key_cfg_iova = + (size_t)priv->extract.qos_extract_param; +- /* QoS table is effecitive for multiple TCs.*/ ++ /* QoS table is effective for multiple TCs. */ + if (priv->num_rx_tc > 1) { + ret = dpni_set_qos_table(dpni, CMD_PRI_LOW, + priv->token, &qos_cfg); +@@ -3655,7 +3655,7 @@ dpaa2_generic_flow_set(struct rte_flow *flow, + 0, 0); + if (ret < 0) { + DPAA2_PMD_ERR( +- "Error in addnig entry to QoS table(%d)", ret); ++ "Error in adding entry to QoS table(%d)", ret); + return ret; + } + } +diff --git a/dpdk/drivers/net/dpaa2/dpaa2_mux.c b/dpdk/drivers/net/dpaa2/dpaa2_mux.c +index d347f4df51..54f53b7ea0 100644 +--- a/dpdk/drivers/net/dpaa2/dpaa2_mux.c ++++ b/dpdk/drivers/net/dpaa2/dpaa2_mux.c +@@ -95,7 +95,7 @@ rte_pmd_dpaa2_mux_flow_create(uint32_t dpdmux_id, + mask_iova = (void *)((size_t)key_iova + DIST_PARAM_IOVA_SIZE); + + /* Currently taking only IP protocol as an extract type. +- * This can be exended to other fields using pattern->type. ++ * This can be extended to other fields using pattern->type. + */ + memset(&kg_cfg, 0, sizeof(struct dpkg_profile_cfg)); + +@@ -296,7 +296,7 @@ dpaa2_create_dpdmux_device(int vdev_fd __rte_unused, + } + + ret = dpdmux_if_set_default(&dpdmux_dev->dpdmux, CMD_PRI_LOW, +- dpdmux_dev->token, 1); ++ dpdmux_dev->token, attr.default_if); + if (ret) { + DPAA2_PMD_ERR("setting default interface failed in %s", + __func__); +diff --git a/dpdk/drivers/net/dpaa2/dpaa2_ptp.c b/dpdk/drivers/net/dpaa2/dpaa2_ptp.c +index 8d79e39244..3a4536dd69 100644 +--- a/dpdk/drivers/net/dpaa2/dpaa2_ptp.c ++++ b/dpdk/drivers/net/dpaa2/dpaa2_ptp.c +@@ -111,10 +111,12 @@ int dpaa2_timesync_read_tx_timestamp(struct rte_eth_dev *dev, + { + struct dpaa2_dev_priv *priv = dev->data->dev_private; + +- if (priv->next_tx_conf_queue) +- dpaa2_dev_tx_conf(priv->next_tx_conf_queue); +- else ++ if (priv->next_tx_conf_queue) { ++ while (!priv->tx_timestamp) ++ dpaa2_dev_tx_conf(priv->next_tx_conf_queue); ++ } else { + return -1; ++ } + *timestamp = rte_ns_to_timespec(priv->tx_timestamp); + + return 0; +diff --git a/dpdk/drivers/net/dpaa2/dpaa2_rxtx.c b/dpdk/drivers/net/dpaa2/dpaa2_rxtx.c +index c65589a5f3..9fb6c5f91d 100644 +--- a/dpdk/drivers/net/dpaa2/dpaa2_rxtx.c ++++ b/dpdk/drivers/net/dpaa2/dpaa2_rxtx.c +@@ -140,8 +140,10 @@ dpaa2_dev_rx_parse_slow(struct rte_mbuf *mbuf, + annotation->word3, annotation->word4); + + #if defined(RTE_LIBRTE_IEEE1588) +- if (BIT_ISSET_AT_POS(annotation->word1, DPAA2_ETH_FAS_PTP)) ++ if (BIT_ISSET_AT_POS(annotation->word1, DPAA2_ETH_FAS_PTP)) { + mbuf->ol_flags |= RTE_MBUF_F_RX_IEEE1588_PTP; ++ mbuf->ol_flags |= RTE_MBUF_F_RX_IEEE1588_TMST; ++ } + #endif + + if (BIT_ISSET_AT_POS(annotation->word3, L2_VLAN_1_PRESENT)) { +@@ -714,7 +716,7 @@ dpaa2_dev_prefetch_rx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts) + rte_prefetch0((void *)(size_t)(dq_storage + 1)); + + /* Prepare next pull descriptor. This will give space for the +- * prefething done on DQRR entries ++ * prefetching done on DQRR entries + */ + q_storage->toggle ^= 1; + dq_storage1 = q_storage->dq_storage[q_storage->toggle]; +@@ -769,7 +771,10 @@ dpaa2_dev_prefetch_rx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts) + else + bufs[num_rx] = eth_fd_to_mbuf(fd, eth_data->port_id); + #if defined(RTE_LIBRTE_IEEE1588) +- priv->rx_timestamp = *dpaa2_timestamp_dynfield(bufs[num_rx]); ++ if (bufs[num_rx]->ol_flags & PKT_RX_IEEE1588_TMST) { ++ priv->rx_timestamp = ++ *dpaa2_timestamp_dynfield(bufs[num_rx]); ++ } + #endif + + if (eth_data->dev_conf.rxmode.offloads & +@@ -986,6 +991,13 @@ dpaa2_dev_rx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts) + bufs[num_rx] = eth_fd_to_mbuf(fd, + eth_data->port_id); + ++#if defined(RTE_LIBRTE_IEEE1588) ++ if (bufs[num_rx]->ol_flags & PKT_RX_IEEE1588_TMST) { ++ priv->rx_timestamp = ++ *dpaa2_timestamp_dynfield(bufs[num_rx]); ++ } ++#endif + + if (eth_data->dev_conf.rxmode.offloads & + RTE_ETH_RX_OFFLOAD_VLAN_STRIP) { + rte_vlan_strip(bufs[num_rx]); +@@ -1021,6 +1033,8 @@ uint16_t dpaa2_dev_tx_conf(void *queue) + struct rte_eth_dev_data *eth_data = dpaa2_q->eth_data; + struct dpaa2_dev_priv *priv = eth_data->dev_private; + struct dpaa2_annot_hdr *annotation; ++ void *v_addr; ++ struct rte_mbuf *mbuf; + #endif - #endif /* rculist.h */ -diff --git a/lib/reconnect.c b/lib/reconnect.c -index a929ddfd2d..89a0bcaf95 100644 ---- a/lib/reconnect.c -+++ b/lib/reconnect.c -@@ -75,7 +75,8 @@ struct reconnect { + if (unlikely(!DPAA2_PER_LCORE_DPIO)) { +@@ -1105,10 +1119,16 @@ uint16_t dpaa2_dev_tx_conf(void *queue) + num_tx_conf++; + num_pulled++; + #if defined(RTE_LIBRTE_IEEE1588) +- annotation = (struct dpaa2_annot_hdr *)((size_t) +- DPAA2_IOVA_TO_VADDR(DPAA2_GET_FD_ADDR(fd)) + +- DPAA2_FD_PTA_SIZE); +- priv->tx_timestamp = annotation->word2; ++ v_addr = DPAA2_IOVA_TO_VADDR(DPAA2_GET_FD_ADDR(fd)); ++ mbuf = DPAA2_INLINE_MBUF_FROM_BUF(v_addr, ++ rte_dpaa2_bpid_info[DPAA2_GET_FD_BPID(fd)].meta_data_size); ++ ++ if (mbuf->ol_flags & PKT_TX_IEEE1588_TMST) { ++ annotation = (struct dpaa2_annot_hdr *)((size_t) ++ DPAA2_IOVA_TO_VADDR(DPAA2_GET_FD_ADDR(fd)) + ++ DPAA2_FD_PTA_SIZE); ++ priv->tx_timestamp = annotation->word2; ++ } + #endif + } while (pending); + +@@ -1184,8 +1204,11 @@ dpaa2_dev_tx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts) + * corresponding to last packet transmitted for reading + * the timestamp + */ +- priv->next_tx_conf_queue = dpaa2_q->tx_conf_queue; +- dpaa2_dev_tx_conf(dpaa2_q->tx_conf_queue); ++ if ((*bufs)->ol_flags & PKT_TX_IEEE1588_TMST) { ++ priv->next_tx_conf_queue = dpaa2_q->tx_conf_queue; ++ dpaa2_dev_tx_conf(dpaa2_q->tx_conf_queue); ++ priv->tx_timestamp = 0; ++ } + #endif - static void reconnect_transition__(struct reconnect *, long long int now, - enum state state); --static long long int reconnect_deadline__(const struct reconnect *); -+static long long int reconnect_deadline__(const struct reconnect *, -+ long long int now); - static bool reconnect_may_retry(struct reconnect *); + /*Prepare enqueue descriptor*/ +@@ -1510,7 +1533,7 @@ dpaa2_dev_tx_ordered(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts) + if (*dpaa2_seqn(*bufs)) { + /* Use only queue 0 for Tx in case of atomic/ + * ordered packets as packets can get unordered +- * when being tranmitted out from the interface ++ * when being transmitted out from the interface + */ + dpaa2_set_enqueue_descriptor(order_sendq, + (*bufs), +@@ -1738,7 +1761,7 @@ dpaa2_dev_loopback_rx(void *queue, + rte_prefetch0((void *)(size_t)(dq_storage + 1)); + + /* Prepare next pull descriptor. This will give space for the +- * prefething done on DQRR entries ++ * prefetching done on DQRR entries + */ + q_storage->toggle ^= 1; + dq_storage1 = q_storage->dq_storage[q_storage->toggle]; +diff --git a/dpdk/drivers/net/dpaa2/mc/fsl_dpni.h b/dpdk/drivers/net/dpaa2/mc/fsl_dpni.h +index 469ab9b3d4..3b9bffeed7 100644 +--- a/dpdk/drivers/net/dpaa2/mc/fsl_dpni.h ++++ b/dpdk/drivers/net/dpaa2/mc/fsl_dpni.h +@@ -93,7 +93,7 @@ struct fsl_mc_io; + */ + #define DPNI_OPT_OPR_PER_TC 0x000080 + /** +- * All Tx traffic classes will use a single sender (ignore num_queueus for tx) ++ * All Tx traffic classes will use a single sender (ignore num_queues for tx) + */ + #define DPNI_OPT_SINGLE_SENDER 0x000100 + /** +@@ -617,7 +617,7 @@ int dpni_get_tx_data_offset(struct fsl_mc_io *mc_io, + * @page_3.ceetm_reject_bytes: Cumulative count of the number of bytes in all + * frames whose enqueue was rejected + * @page_3.ceetm_reject_frames: Cumulative count of all frame enqueues rejected +- * @page_4: congestion point drops for seleted TC ++ * @page_4: congestion point drops for selected TC + * @page_4.cgr_reject_frames: number of rejected frames due to congestion point + * @page_4.cgr_reject_bytes: number of rejected bytes due to congestion point + * @page_5: policer statistics per TC +@@ -1417,7 +1417,7 @@ int dpni_get_tx_confirmation_mode(struct fsl_mc_io *mc_io, + * dpkg_prepare_key_cfg() + * @discard_on_miss: Set to '1' to discard frames in case of no match (miss); + * '0' to use the 'default_tc' in such cases +- * @keep_entries: if set to one will not delele existing table entries. This ++ * @keep_entries: if set to one will not delete existing table entries. This + * option will work properly only for dpni objects created with + * DPNI_OPT_HAS_KEY_MASKING option. All previous QoS entries must + * be compatible with new key composition rule. +@@ -1516,7 +1516,7 @@ int dpni_clear_qos_table(struct fsl_mc_io *mc_io, + * @flow_id: Identifies the Rx queue used for matching traffic. Supported + * values are in range 0 to num_queue-1. + * @redirect_obj_token: token that identifies the object where frame is +- * redirected when this rule is hit. This paraneter is used only when one of the ++ * redirected when this rule is hit. This parameter is used only when one of the + * flags DPNI_FS_OPT_REDIRECT_TO_DPNI_RX or DPNI_FS_OPT_REDIRECT_TO_DPNI_TX is + * set. + * The token is obtained using dpni_open() API call. The object must stay +@@ -1797,7 +1797,7 @@ int dpni_load_sw_sequence(struct fsl_mc_io *mc_io, + struct dpni_load_ss_cfg *cfg); + + /** +- * dpni_eanble_sw_sequence() - Enables a software sequence in the parser ++ * dpni_enable_sw_sequence() - Enables a software sequence in the parser + * profile + * corresponding to the ingress or egress of the DPNI. + * @mc_io: Pointer to MC portal's I/O object +diff --git a/dpdk/drivers/net/e1000/e1000_ethdev.h b/dpdk/drivers/net/e1000/e1000_ethdev.h +index a548ae2ccb..718a9746ed 100644 +--- a/dpdk/drivers/net/e1000/e1000_ethdev.h ++++ b/dpdk/drivers/net/e1000/e1000_ethdev.h +@@ -103,7 +103,7 @@ + * Maximum number of Ring Descriptors. + * + * Since RDLEN/TDLEN should be multiple of 128 bytes, the number of ring +- * desscriptors should meet the following condition: ++ * descriptors should meet the following condition: + * (num_ring_desc * sizeof(struct e1000_rx/tx_desc)) % 128 == 0 + */ + #define E1000_MIN_RING_DESC 32 +@@ -252,7 +252,7 @@ struct igb_rte_flow_rss_conf { + }; - static const char * -@@ -539,7 +540,7 @@ reconnect_transition__(struct reconnect *fsm, long long int now, + /* +- * Structure to store filters'info. ++ * Structure to store filters' info. + */ + struct e1000_filter_info { + uint8_t ethertype_mask; /* Bit mask for every used ethertype filter */ +diff --git a/dpdk/drivers/net/e1000/em_ethdev.c b/dpdk/drivers/net/e1000/em_ethdev.c +index 31c4870086..794496abfc 100644 +--- a/dpdk/drivers/net/e1000/em_ethdev.c ++++ b/dpdk/drivers/net/e1000/em_ethdev.c +@@ -1058,8 +1058,8 @@ eth_em_infos_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info) + + /* + * Starting with 631xESB hw supports 2 TX/RX queues per port. +- * Unfortunatelly, all these nics have just one TX context. +- * So we have few choises for TX: ++ * Unfortunately, all these nics have just one TX context. ++ * So we have few choices for TX: + * - Use just one TX queue. + * - Allow cksum offload only for one TX queue. + * - Don't allow TX cksum offload at all. +@@ -1068,7 +1068,7 @@ eth_em_infos_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info) + * (Multiple Receive Queues are mutually exclusive with UDP + * fragmentation and are not supported when a legacy receive + * descriptor format is used). +- * Which means separate RX routinies - as legacy nics (82540, 82545) ++ * Which means separate RX routines - as legacy nics (82540, 82545) + * don't support extended RXD. + * To avoid it we support just one RX queue for now (no RSS). + */ +@@ -1558,7 +1558,7 @@ eth_em_interrupt_get_status(struct rte_eth_dev *dev) } - static long long int --reconnect_deadline__(const struct reconnect *fsm) -+reconnect_deadline__(const struct reconnect *fsm, long long int now) - { - ovs_assert(fsm->state_entered != LLONG_MIN); - switch (fsm->state) { -@@ -557,8 +558,18 @@ reconnect_deadline__(const struct reconnect *fsm) - if (fsm->probe_interval) { - long long int base = MAX(fsm->last_activity, fsm->state_entered); - long long int expiration = base + fsm->probe_interval; -- if (fsm->last_receive_attempt >= expiration) { -+ if (now < expiration || fsm->last_receive_attempt >= expiration) { -+ /* We still have time before the expiration or the time has -+ * already passed and there was no activity. In the first case -+ * we need to wait for the expiration, in the second - we're -+ * already past the deadline. */ - return expiration; -+ } else { -+ /* Time has already passed, but we didn't attempt to receive -+ * anything. We need to wake up and try to receive even if -+ * nothing is pending, so we can update the expiration time or -+ * transition to a different state. */ -+ return now + 1; - } - } - return LLONG_MAX; -@@ -566,8 +577,10 @@ reconnect_deadline__(const struct reconnect *fsm) - case S_IDLE: - if (fsm->probe_interval) { - long long int expiration = fsm->state_entered + fsm->probe_interval; -- if (fsm->last_receive_attempt >= expiration) { -+ if (now < expiration || fsm->last_receive_attempt >= expiration) { - return expiration; -+ } else { -+ return now + 1; - } - } - return LLONG_MAX; -@@ -618,7 +631,7 @@ reconnect_deadline__(const struct reconnect *fsm) - enum reconnect_action - reconnect_run(struct reconnect *fsm, long long int now) - { -- if (now >= reconnect_deadline__(fsm)) { -+ if (now >= reconnect_deadline__(fsm, now)) { - switch (fsm->state) { - case S_VOID: - return 0; -@@ -671,7 +684,7 @@ reconnect_wait(struct reconnect *fsm, long long int now) - int - reconnect_timeout(struct reconnect *fsm, long long int now) - { -- long long int deadline = reconnect_deadline__(fsm); -+ long long int deadline = reconnect_deadline__(fsm, now); - if (deadline != LLONG_MAX) { - long long int remaining = deadline - now; - return MAX(0, MIN(INT_MAX, remaining)); -diff --git a/lib/seq.c b/lib/seq.c -index 6581cb06ba..99e5bf8bd1 100644 ---- a/lib/seq.c -+++ b/lib/seq.c -@@ -297,9 +297,9 @@ static void - seq_thread_woke(struct seq_thread *thread) - OVS_REQUIRES(seq_mutex) - { -- struct seq_waiter *waiter, *next_waiter; -+ struct seq_waiter *waiter; + /* +- * It executes link_update after knowing an interrupt is prsent. ++ * It executes link_update after knowing an interrupt is present. + * + * @param dev + * Pointer to struct rte_eth_dev. +@@ -1616,7 +1616,7 @@ eth_em_interrupt_action(struct rte_eth_dev *dev, + * @param handle + * Pointer to interrupt handle. + * @param param +- * The address of parameter (struct rte_eth_dev *) regsitered before. ++ * The address of parameter (struct rte_eth_dev *) registered before. + * + * @return + * void +diff --git a/dpdk/drivers/net/e1000/em_rxtx.c b/dpdk/drivers/net/e1000/em_rxtx.c +index 39262502bb..cea5b490ba 100644 +--- a/dpdk/drivers/net/e1000/em_rxtx.c ++++ b/dpdk/drivers/net/e1000/em_rxtx.c +@@ -141,7 +141,7 @@ union em_vlan_macip { + struct em_ctx_info { + uint64_t flags; /**< ol_flags related to context build. */ + uint32_t cmp_mask; /**< compare mask */ +- union em_vlan_macip hdrlen; /**< L2 and L3 header lenghts */ ++ union em_vlan_macip hdrlen; /**< L2 and L3 header lengths */ + }; -- LIST_FOR_EACH_SAFE (waiter, next_waiter, list_node, &thread->waiters) { -+ LIST_FOR_EACH_SAFE (waiter, list_node, &thread->waiters) { - ovs_assert(waiter->thread == thread); - seq_waiter_destroy(waiter); - } -@@ -319,9 +319,9 @@ static void - seq_wake_waiters(struct seq *seq) - OVS_REQUIRES(seq_mutex) - { -- struct seq_waiter *waiter, *next_waiter; -+ struct seq_waiter *waiter; + /** +@@ -829,7 +829,7 @@ eth_em_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, + * register. + * Update the RDT with the value of the last processed RX descriptor + * minus 1, to guarantee that the RDT register is never equal to the +- * RDH register, which creates a "full" ring situtation from the ++ * RDH register, which creates a "full" ring situation from the + * hardware point of view... + */ + nb_hold = (uint16_t) (nb_hold + rxq->nb_rx_hold); +@@ -1074,7 +1074,7 @@ eth_em_recv_scattered_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, + * register. + * Update the RDT with the value of the last processed RX descriptor + * minus 1, to guarantee that the RDT register is never equal to the +- * RDH register, which creates a "full" ring situtation from the ++ * RDH register, which creates a "full" ring situation from the + * hardware point of view... + */ + nb_hold = (uint16_t) (nb_hold + rxq->nb_rx_hold); +diff --git a/dpdk/drivers/net/e1000/igb_ethdev.c b/dpdk/drivers/net/e1000/igb_ethdev.c +index 3ee16c15fe..a9c18b27e8 100644 +--- a/dpdk/drivers/net/e1000/igb_ethdev.c ++++ b/dpdk/drivers/net/e1000/igb_ethdev.c +@@ -1149,7 +1149,7 @@ eth_igb_configure(struct rte_eth_dev *dev) + if (dev->data->dev_conf.rxmode.mq_mode & RTE_ETH_MQ_RX_RSS_FLAG) + dev->data->dev_conf.rxmode.offloads |= RTE_ETH_RX_OFFLOAD_RSS_HASH; + +- /* multipe queue mode checking */ ++ /* multiple queue mode checking */ + ret = igb_check_mq_mode(dev); + if (ret != 0) { + PMD_DRV_LOG(ERR, "igb_check_mq_mode fails with %d.", +@@ -1265,7 +1265,7 @@ eth_igb_start(struct rte_eth_dev *dev) + } + } -- HMAP_FOR_EACH_SAFE (waiter, next_waiter, hmap_node, &seq->waiters) { -+ HMAP_FOR_EACH_SAFE (waiter, hmap_node, &seq->waiters) { - latch_set(&waiter->thread->latch); - seq_waiter_destroy(waiter); - } -diff --git a/lib/shash.c b/lib/shash.c -index a8433629ab..a7b2c64582 100644 ---- a/lib/shash.c -+++ b/lib/shash.c -@@ -68,9 +68,9 @@ shash_moved(struct shash *sh) - void - shash_clear(struct shash *sh) - { -- struct shash_node *node, *next; -+ struct shash_node *node; +- /* confiugre msix for rx interrupt */ ++ /* configure MSI-X for Rx interrupt */ + eth_igb_configure_msix_intr(dev); -- SHASH_FOR_EACH_SAFE (node, next, sh) { -+ SHASH_FOR_EACH_SAFE (node, sh) { - hmap_remove(&sh->map, &node->node); - free(node->name); - free(node); -@@ -81,9 +81,9 @@ shash_clear(struct shash *sh) - void - shash_clear_free_data(struct shash *sh) - { -- struct shash_node *node, *next; -+ struct shash_node *node; + /* Configure for OS presence */ +@@ -2819,7 +2819,7 @@ eth_igb_interrupt_get_status(struct rte_eth_dev *dev) + } -- SHASH_FOR_EACH_SAFE (node, next, sh) { -+ SHASH_FOR_EACH_SAFE (node, sh) { - hmap_remove(&sh->map, &node->node); - free(node->data); - free(node->name); -diff --git a/lib/simap.c b/lib/simap.c -index f404ece677..0ee08d74d5 100644 ---- a/lib/simap.c -+++ b/lib/simap.c -@@ -63,9 +63,9 @@ simap_moved(struct simap *simap) - void - simap_clear(struct simap *simap) + /* +- * It executes link_update after knowing an interrupt is prsent. ++ * It executes link_update after knowing an interrupt is present. + * + * @param dev + * Pointer to struct rte_eth_dev. +@@ -2889,7 +2889,7 @@ eth_igb_interrupt_action(struct rte_eth_dev *dev, + * @param handle + * Pointer to interrupt handle. + * @param param +- * The address of parameter (struct rte_eth_dev *) regsitered before. ++ * The address of parameter (struct rte_eth_dev *) registered before. + * + * @return + * void +@@ -3787,7 +3787,7 @@ igb_inject_2uple_filter(struct rte_eth_dev *dev, + * + * @param + * dev: Pointer to struct rte_eth_dev. +- * ntuple_filter: ponter to the filter that will be added. ++ * ntuple_filter: pointer to the filter that will be added. + * + * @return + * - On success, zero. +@@ -3868,7 +3868,7 @@ igb_delete_2tuple_filter(struct rte_eth_dev *dev, + * + * @param + * dev: Pointer to struct rte_eth_dev. +- * ntuple_filter: ponter to the filter that will be removed. ++ * ntuple_filter: pointer to the filter that will be removed. + * + * @return + * - On success, zero. +@@ -4226,7 +4226,7 @@ igb_inject_5tuple_filter_82576(struct rte_eth_dev *dev, + * + * @param + * dev: Pointer to struct rte_eth_dev. +- * ntuple_filter: ponter to the filter that will be added. ++ * ntuple_filter: pointer to the filter that will be added. + * + * @return + * - On success, zero. +@@ -4313,7 +4313,7 @@ igb_delete_5tuple_filter_82576(struct rte_eth_dev *dev, + * + * @param + * dev: Pointer to struct rte_eth_dev. +- * ntuple_filter: ponter to the filter that will be removed. ++ * ntuple_filter: pointer to the filter that will be removed. + * + * @return + * - On success, zero. +@@ -4831,7 +4831,7 @@ igb_timesync_disable(struct rte_eth_dev *dev) + /* Disable L2 filtering of IEEE1588/802.1AS Ethernet frame types. */ + E1000_WRITE_REG(hw, E1000_ETQF(E1000_ETQF_FILTER_1588), 0); + +- /* Stop incrementating the System Time registers. */ ++ /* Stop incrementing the System Time registers. */ + E1000_WRITE_REG(hw, E1000_TIMINCA, 0); + + return 0; +diff --git a/dpdk/drivers/net/e1000/igb_flow.c b/dpdk/drivers/net/e1000/igb_flow.c +index e72376f69c..e46697b6a1 100644 +--- a/dpdk/drivers/net/e1000/igb_flow.c ++++ b/dpdk/drivers/net/e1000/igb_flow.c +@@ -57,7 +57,7 @@ struct igb_flex_filter_list igb_filter_flex_list; + struct igb_rss_filter_list igb_filter_rss_list; + + /** +- * Please aware there's an asumption for all the parsers. ++ * Please be aware there's an assumption for all the parsers. + * rte_flow_item is using big endian, rte_flow_attr and + * rte_flow_action are using CPU order. + * Because the pattern is used to describe the packets, +@@ -1608,7 +1608,7 @@ igb_flow_create(struct rte_eth_dev *dev, + + /** + * Check if the flow rule is supported by igb. +- * It only checkes the format. Don't guarantee the rule can be programmed into ++ * It only checks the format. Don't guarantee the rule can be programmed into + * the HW. Because there can be no enough room for the rule. + */ + static int +diff --git a/dpdk/drivers/net/e1000/igb_pf.c b/dpdk/drivers/net/e1000/igb_pf.c +index fe355ef6b3..3f3fd0d61e 100644 +--- a/dpdk/drivers/net/e1000/igb_pf.c ++++ b/dpdk/drivers/net/e1000/igb_pf.c +@@ -155,7 +155,7 @@ int igb_pf_host_configure(struct rte_eth_dev *eth_dev) + else + E1000_WRITE_REG(hw, E1000_DTXSWC, E1000_DTXSWC_VMDQ_LOOPBACK_EN); + +- /* clear VMDq map to perment rar 0 */ ++ /* clear VMDq map to permanent rar 0 */ + rah = E1000_READ_REG(hw, E1000_RAH(0)); + rah &= ~ (0xFF << E1000_RAH_POOLSEL_SHIFT); + E1000_WRITE_REG(hw, E1000_RAH(0), rah); +diff --git a/dpdk/drivers/net/e1000/igb_rxtx.c b/dpdk/drivers/net/e1000/igb_rxtx.c +index 4a311a7b18..f32dee46df 100644 +--- a/dpdk/drivers/net/e1000/igb_rxtx.c ++++ b/dpdk/drivers/net/e1000/igb_rxtx.c +@@ -150,7 +150,7 @@ union igb_tx_offload { + (TX_MACIP_LEN_CMP_MASK | TX_TCP_LEN_CMP_MASK | TX_TSO_MSS_CMP_MASK) + + /** +- * Strucutre to check if new context need be built ++ * Structure to check if new context need be built + */ + struct igb_advctx_info { + uint64_t flags; /**< ol_flags related to context build. */ +@@ -967,7 +967,7 @@ eth_igb_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, + * register. + * Update the RDT with the value of the last processed RX descriptor + * minus 1, to guarantee that the RDT register is never equal to the +- * RDH register, which creates a "full" ring situtation from the ++ * RDH register, which creates a "full" ring situation from the + * hardware point of view... + */ + nb_hold = (uint16_t) (nb_hold + rxq->nb_rx_hold); +@@ -1229,7 +1229,7 @@ eth_igb_recv_scattered_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, + * register. + * Update the RDT with the value of the last processed RX descriptor + * minus 1, to guarantee that the RDT register is never equal to the +- * RDH register, which creates a "full" ring situtation from the ++ * RDH register, which creates a "full" ring situation from the + * hardware point of view... + */ + nb_hold = (uint16_t) (nb_hold + rxq->nb_rx_hold); +@@ -1252,7 +1252,7 @@ eth_igb_recv_scattered_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, + * Maximum number of Ring Descriptors. + * + * Since RDLEN/TDLEN should be multiple of 128bytes, the number of ring +- * desscriptors should meet the following condition: ++ * descriptors should meet the following condition: + * (num_ring_desc * sizeof(struct e1000_rx/tx_desc)) % 128 == 0 + */ + +@@ -1350,7 +1350,7 @@ igb_tx_done_cleanup(struct igb_tx_queue *txq, uint32_t free_cnt) + sw_ring[tx_id].last_id = tx_id; + } + +- /* Move to next segemnt. */ ++ /* Move to next segment. */ + tx_id = sw_ring[tx_id].next_id; + + } while (tx_id != tx_next); +@@ -1383,7 +1383,7 @@ igb_tx_done_cleanup(struct igb_tx_queue *txq, uint32_t free_cnt) + + /* Walk the list and find the next mbuf, if any. */ + do { +- /* Move to next segemnt. */ ++ /* Move to next segment. */ + tx_id = sw_ring[tx_id].next_id; + + if (sw_ring[tx_id].mbuf) +@@ -2146,7 +2146,7 @@ igb_vmdq_rx_hw_configure(struct rte_eth_dev *dev) + + igb_rss_disable(dev); + +- /* RCTL: eanble VLAN filter */ ++ /* RCTL: enable VLAN filter */ + rctl = E1000_READ_REG(hw, E1000_RCTL); + rctl |= E1000_RCTL_VFE; + E1000_WRITE_REG(hw, E1000_RCTL, rctl); +diff --git a/dpdk/drivers/net/ena/ena_ethdev.c b/dpdk/drivers/net/ena/ena_ethdev.c +index 634c97acf6..770b101688 100644 +--- a/dpdk/drivers/net/ena/ena_ethdev.c ++++ b/dpdk/drivers/net/ena/ena_ethdev.c +@@ -38,11 +38,6 @@ + + #define ENA_PTYPE_HAS_HASH (RTE_PTYPE_L4_TCP | RTE_PTYPE_L4_UDP) + +-enum ethtool_stringset { +- ETH_SS_TEST = 0, +- ETH_SS_STATS, +-}; +- + struct ena_stats { + char name[ETH_GSTRING_LEN]; + int stat_offset; +@@ -280,6 +275,15 @@ static const struct eth_dev_ops ena_dev_ops = { + .rss_hash_conf_get = ena_rss_hash_conf_get, + }; + ++static inline void ena_trigger_reset(struct ena_adapter *adapter, ++ enum ena_regs_reset_reason_types reason) ++{ ++ if (likely(!adapter->trigger_reset)) { ++ adapter->reset_reason = reason; ++ adapter->trigger_reset = true; ++ } ++} ++ + static inline void ena_rx_mbuf_prepare(struct rte_mbuf *mbuf, + struct ena_com_rx_ctx *ena_rx_ctx, + bool fill_hash) +@@ -306,7 +310,13 @@ static inline void ena_rx_mbuf_prepare(struct rte_mbuf *mbuf, + ol_flags |= RTE_MBUF_F_RX_L4_CKSUM_UNKNOWN; + else + if (unlikely(ena_rx_ctx->l4_csum_err)) +- ol_flags |= RTE_MBUF_F_RX_L4_CKSUM_BAD; ++ /* ++ * For the L4 Rx checksum offload the HW may indicate ++ * bad checksum although it's valid. Because of that, ++ * we're setting the UNKNOWN flag to let the app ++ * re-verify the checksum. ++ */ ++ ol_flags |= RTE_MBUF_F_RX_L4_CKSUM_UNKNOWN; + else + ol_flags |= RTE_MBUF_F_RX_L4_CKSUM_GOOD; + +@@ -344,6 +354,8 @@ static inline void ena_tx_mbuf_prepare(struct rte_mbuf *mbuf, + + if (mbuf->ol_flags & RTE_MBUF_F_TX_IPV6) { + ena_tx_ctx->l3_proto = ENA_ETH_IO_L3_PROTO_IPV6; ++ /* For the IPv6 packets, DF always needs to be true. */ ++ ena_tx_ctx->df = 1; + } else { + ena_tx_ctx->l3_proto = ENA_ETH_IO_L3_PROTO_IPV4; + +@@ -351,7 +363,7 @@ static inline void ena_tx_mbuf_prepare(struct rte_mbuf *mbuf, + if (mbuf->packet_type & + (RTE_PTYPE_L4_NONFRAG + | RTE_PTYPE_INNER_L4_NONFRAG)) +- ena_tx_ctx->df = true; ++ ena_tx_ctx->df = 1; + } + + /* check if L4 checksum is needed */ +@@ -399,8 +411,7 @@ static int validate_tx_req_id(struct ena_ring *tx_ring, u16 req_id) + + /* Trigger device reset */ + ++tx_ring->tx_stats.bad_req_id; +- tx_ring->adapter->reset_reason = ENA_REGS_RESET_INV_TX_REQ_ID; +- tx_ring->adapter->trigger_reset = true; ++ ena_trigger_reset(tx_ring->adapter, ENA_REGS_RESET_INV_TX_REQ_ID); + return -EFAULT; + } + +@@ -1408,7 +1419,7 @@ static int ena_populate_rx_queue(struct ena_ring *rxq, unsigned int count) + ++rxq->rx_stats.refill_partial; + } + +- /* When we submitted free recources to device... */ ++ /* When we submitted free resources to device... */ + if (likely(i > 0)) { + /* ...let HW know that it can fill buffers with data. */ + ena_com_write_sq_doorbell(rxq->ena_com_io_sq); +@@ -1529,8 +1540,7 @@ static void check_for_missing_keep_alive(struct ena_adapter *adapter) + if (unlikely((rte_get_timer_cycles() - adapter->timestamp_wd) >= + adapter->keep_alive_timeout)) { + PMD_DRV_LOG(ERR, "Keep alive timeout\n"); +- adapter->reset_reason = ENA_REGS_RESET_KEEP_ALIVE_TO; +- adapter->trigger_reset = true; ++ ena_trigger_reset(adapter, ENA_REGS_RESET_KEEP_ALIVE_TO); + ++adapter->dev_stats.wd_expired; + } + } +@@ -1540,8 +1550,7 @@ static void check_for_admin_com_state(struct ena_adapter *adapter) { -- struct simap_node *node, *next; -+ struct simap_node *node; + if (unlikely(!ena_com_get_admin_running_state(&adapter->ena_dev))) { + PMD_DRV_LOG(ERR, "ENA admin queue is not in running state\n"); +- adapter->reset_reason = ENA_REGS_RESET_ADMIN_TO; +- adapter->trigger_reset = true; ++ ena_trigger_reset(adapter, ENA_REGS_RESET_ADMIN_TO); + } + } -- SIMAP_FOR_EACH_SAFE (node, next, simap) { -+ SIMAP_FOR_EACH_SAFE (node, simap) { - hmap_remove(&simap->map, &node->node); - free(node->name); - free(node); -diff --git a/lib/simap.h b/lib/simap.h -index 5e646e6607..8db7bea7c9 100644 ---- a/lib/simap.h -+++ b/lib/simap.h -@@ -41,12 +41,22 @@ struct simap_node { - BUILD_ASSERT_TYPE(SIMAP_NODE, struct simap_node *), \ - BUILD_ASSERT_TYPE(SIMAP, struct simap *)) +@@ -1632,6 +1641,9 @@ static void ena_timer_wd_callback(__rte_unused struct rte_timer *timer, + struct rte_eth_dev *dev = arg; + struct ena_adapter *adapter = dev->data->dev_private; --#define SIMAP_FOR_EACH_SAFE(SIMAP_NODE, NEXT, SIMAP) \ -- HMAP_FOR_EACH_SAFE_INIT (SIMAP_NODE, NEXT, node, &(SIMAP)->map, \ -+#define SIMAP_FOR_EACH_SAFE_SHORT(SIMAP_NODE, SIMAP) \ -+ HMAP_FOR_EACH_SAFE_SHORT_INIT (SIMAP_NODE, node, &(SIMAP)->map, \ - BUILD_ASSERT_TYPE(SIMAP_NODE, struct simap_node *), \ -- BUILD_ASSERT_TYPE(NEXT, struct simap_node *), \ - BUILD_ASSERT_TYPE(SIMAP, struct simap *)) ++ if (unlikely(adapter->trigger_reset)) ++ return; ++ + check_for_missing_keep_alive(adapter); + check_for_admin_com_state(adapter); + check_for_tx_completions(adapter); +@@ -1682,6 +1694,13 @@ ena_set_queues_placement_policy(struct ena_adapter *adapter, + return 0; + } -+#define SIMAP_FOR_EACH_SAFE_LONG(SIMAP_NODE, NEXT, SIMAP) \ -+ HMAP_FOR_EACH_SAFE_LONG_INIT (SIMAP_NODE, NEXT, node, &(SIMAP)->map, \ -+ BUILD_ASSERT_TYPE(SIMAP_NODE, struct simap_node *), \ -+ BUILD_ASSERT_TYPE(NEXT, struct simap_node *), \ -+ BUILD_ASSERT_TYPE(SIMAP, struct simap *)) ++ if (adapter->dev_mem_base == NULL) { ++ PMD_DRV_LOG(ERR, ++ "LLQ is advertised as supported, but device doesn't expose mem bar\n"); ++ ena_dev->tx_mem_queue_type = ENA_ADMIN_PLACEMENT_POLICY_HOST; ++ return 0; ++ } ++ + rc = ena_com_config_dev_mode(ena_dev, llq, llq_default_configurations); + if (unlikely(rc)) { + PMD_INIT_LOG(WARNING, +@@ -1694,13 +1713,6 @@ ena_set_queues_placement_policy(struct ena_adapter *adapter, + if (ena_dev->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_HOST) + return 0; + +- if (!adapter->dev_mem_base) { +- PMD_DRV_LOG(ERR, +- "Unable to access LLQ BAR resource. Fallback to host mode policy.\n"); +- ena_dev->tx_mem_queue_type = ENA_ADMIN_PLACEMENT_POLICY_HOST; +- return 0; +- } +- + ena_dev->mem_bar = adapter->dev_mem_base; + + return 0; +@@ -2028,9 +2040,6 @@ static int ena_dev_configure(struct rte_eth_dev *dev) + */ + adapter->tx_cleanup_stall_delay = adapter->missing_tx_completion_to / 2; + +- adapter->tx_selected_offloads = dev->data->dev_conf.txmode.offloads; +- adapter->rx_selected_offloads = dev->data->dev_conf.rxmode.offloads; +- + return 0; + } + +@@ -2325,14 +2334,13 @@ static uint16_t eth_ena_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, + rc); + if (rc == ENA_COM_NO_SPACE) { + ++rx_ring->rx_stats.bad_desc_num; +- rx_ring->adapter->reset_reason = +- ENA_REGS_RESET_TOO_MANY_RX_DESCS; ++ ena_trigger_reset(rx_ring->adapter, ++ ENA_REGS_RESET_TOO_MANY_RX_DESCS); + } else { + ++rx_ring->rx_stats.bad_req_id; +- rx_ring->adapter->reset_reason = +- ENA_REGS_RESET_INV_RX_REQ_ID; ++ ena_trigger_reset(rx_ring->adapter, ++ ENA_REGS_RESET_INV_RX_REQ_ID); + } +- rx_ring->adapter->trigger_reset = true; + return 0; + } + +@@ -2732,9 +2740,8 @@ static int ena_xmit_mbuf(struct ena_ring *tx_ring, struct rte_mbuf *mbuf) + if (unlikely(rc)) { + PMD_DRV_LOG(ERR, "Failed to prepare Tx buffers, rc: %d\n", rc); + ++tx_ring->tx_stats.prepare_ctx_err; +- tx_ring->adapter->reset_reason = +- ENA_REGS_RESET_DRIVER_INVALID_STATE; +- tx_ring->adapter->trigger_reset = true; ++ ena_trigger_reset(tx_ring->adapter, ++ ENA_REGS_RESET_DRIVER_INVALID_STATE); + return rc; + } + +diff --git a/dpdk/drivers/net/ena/ena_ethdev.h b/dpdk/drivers/net/ena/ena_ethdev.h +index 865e1241e0..0632f793d0 100644 +--- a/dpdk/drivers/net/ena/ena_ethdev.h ++++ b/dpdk/drivers/net/ena/ena_ethdev.h +@@ -42,7 +42,7 @@ + + /* While processing submitted and completed descriptors (rx and tx path + * respectively) in a loop it is desired to: +- * - perform batch submissions while populating sumbissmion queue ++ * - perform batch submissions while populating submission queue + * - avoid blocking transmission of other packets during cleanup phase + * Hence the utilization ratio of 1/8 of a queue size or max value if the size + * of the ring is very big - like 8k Rx rings. +@@ -280,11 +280,6 @@ struct ena_adapter { + struct ena_driver_stats *drv_stats; + enum ena_adapter_state state; + +- uint64_t tx_supported_offloads; +- uint64_t tx_selected_offloads; +- uint64_t rx_supported_offloads; +- uint64_t rx_selected_offloads; +- + bool link_status; + + enum ena_regs_reset_reason_types reset_reason; +diff --git a/dpdk/drivers/net/ena/ena_rss.c b/dpdk/drivers/net/ena/ena_rss.c +index be4007e3f3..8193eaf6fc 100644 +--- a/dpdk/drivers/net/ena/ena_rss.c ++++ b/dpdk/drivers/net/ena/ena_rss.c +@@ -51,15 +51,14 @@ void ena_rss_key_fill(void *key, size_t size) + static uint8_t default_key[ENA_HASH_KEY_SIZE]; + size_t i; + +- RTE_ASSERT(size <= ENA_HASH_KEY_SIZE); +- + if (!key_generated) { +- for (i = 0; i < ENA_HASH_KEY_SIZE; ++i) ++ for (i = 0; i < RTE_DIM(default_key); ++i) + default_key[i] = rte_rand() & 0xff; + key_generated = true; + } + +- rte_memcpy(key, default_key, size); ++ RTE_ASSERT(size <= sizeof(default_key)); ++ rte_memcpy(key, default_key, RTE_MIN(size, sizeof(default_key))); + } + + int ena_rss_reta_update(struct rte_eth_dev *dev, +diff --git a/dpdk/drivers/net/enetfec/enet_ethdev.c b/dpdk/drivers/net/enetfec/enet_ethdev.c +index 714f8ac7ec..c938e58204 100644 +--- a/dpdk/drivers/net/enetfec/enet_ethdev.c ++++ b/dpdk/drivers/net/enetfec/enet_ethdev.c +@@ -2,9 +2,12 @@ + * Copyright 2020-2021 NXP + */ + ++#include + -+#define SIMAP_FOR_EACH_SAFE(...) \ -+ OVERLOAD_SAFE_MACRO(SIMAP_FOR_EACH_SAFE_LONG, \ -+ SIMAP_FOR_EACH_SAFE_SHORT, \ -+ 3, __VA_ARGS__) + #include + #include + #include + - void simap_init(struct simap *); - void simap_destroy(struct simap *); - void simap_swap(struct simap *, struct simap *); -diff --git a/lib/smap.c b/lib/smap.c -index e82261497c..b23eeb52d3 100644 ---- a/lib/smap.c -+++ b/lib/smap.c -@@ -185,9 +185,9 @@ smap_steal(struct smap *smap, struct smap_node *node, - void - smap_clear(struct smap *smap) - { -- struct smap_node *node, *next; -+ struct smap_node *node; + #include "enet_pmd_logs.h" + #include "enet_ethdev.h" + #include "enet_regs.h" +@@ -454,6 +457,12 @@ enetfec_rx_queue_setup(struct rte_eth_dev *dev, + return -EINVAL; + } -- SMAP_FOR_EACH_SAFE (node, next, smap) { -+ SMAP_FOR_EACH_SAFE (node, smap) { - smap_remove_node(smap, node); - } ++ if (queue_idx >= ENETFEC_MAX_Q) { ++ ENETFEC_PMD_ERR("Invalid queue id %" PRIu16 ", max %d\n", ++ queue_idx, ENETFEC_MAX_Q); ++ return -EINVAL; ++ } ++ + /* allocate receive queue */ + rxq = rte_zmalloc(NULL, sizeof(*rxq), RTE_CACHE_LINE_SIZE); + if (rxq == NULL) { +diff --git a/dpdk/drivers/net/enetfec/enet_regs.h b/dpdk/drivers/net/enetfec/enet_regs.h +index a300c6f8bc..c9400957f8 100644 +--- a/dpdk/drivers/net/enetfec/enet_regs.h ++++ b/dpdk/drivers/net/enetfec/enet_regs.h +@@ -12,7 +12,7 @@ + #define RX_BD_CR ((ushort)0x0004) /* CRC or Frame error */ + #define RX_BD_SH ((ushort)0x0008) /* Reserved */ + #define RX_BD_NO ((ushort)0x0010) /* Rcvd non-octet aligned frame */ +-#define RX_BD_LG ((ushort)0x0020) /* Rcvd frame length voilation */ ++#define RX_BD_LG ((ushort)0x0020) /* Rcvd frame length violation */ + #define RX_BD_FIRST ((ushort)0x0400) /* Reserved */ + #define RX_BD_LAST ((ushort)0x0800) /* last buffer in the frame */ + #define RX_BD_INT 0x00800000 +diff --git a/dpdk/drivers/net/enic/enic_flow.c b/dpdk/drivers/net/enic/enic_flow.c +index 33147169ba..cf51793cfe 100644 +--- a/dpdk/drivers/net/enic/enic_flow.c ++++ b/dpdk/drivers/net/enic/enic_flow.c +@@ -405,7 +405,7 @@ enic_copy_item_ipv4_v1(struct copy_item_args *arg) + return ENOTSUP; + } + +- /* check that the suppied mask exactly matches capabilty */ ++ /* check that the supplied mask exactly matches capability */ + if (!mask_exact_match((const uint8_t *)&supported_mask, + (const uint8_t *)item->mask, sizeof(*mask))) { + ENICPMD_LOG(ERR, "IPv4 exact match mask"); +@@ -443,7 +443,7 @@ enic_copy_item_udp_v1(struct copy_item_args *arg) + return ENOTSUP; + } + +- /* check that the suppied mask exactly matches capabilty */ ++ /* check that the supplied mask exactly matches capability */ + if (!mask_exact_match((const uint8_t *)&supported_mask, + (const uint8_t *)item->mask, sizeof(*mask))) { + ENICPMD_LOG(ERR, "UDP exact match mask"); +@@ -482,7 +482,7 @@ enic_copy_item_tcp_v1(struct copy_item_args *arg) + return ENOTSUP; + } + +- /* check that the suppied mask exactly matches capabilty */ ++ /* check that the supplied mask exactly matches capability */ + if (!mask_exact_match((const uint8_t *)&supported_mask, + (const uint8_t *)item->mask, sizeof(*mask))) { + ENICPMD_LOG(ERR, "TCP exact match mask"); +@@ -1044,14 +1044,14 @@ fixup_l5_layer(struct enic *enic, struct filter_generic_1 *gp, } -diff --git a/lib/smap.h b/lib/smap.h -index a921159667..2fe6c540a7 100644 ---- a/lib/smap.h -+++ b/lib/smap.h -@@ -45,13 +45,24 @@ struct smap_node { - BUILD_ASSERT_TYPE(SMAP_NODE, struct smap_node *), \ - BUILD_ASSERT_TYPE(SMAP, struct smap *)) --#define SMAP_FOR_EACH_SAFE(SMAP_NODE, NEXT, SMAP) \ -- HMAP_FOR_EACH_SAFE_INIT ( \ -+#define SMAP_FOR_EACH_SAFE_SHORT(SMAP_NODE, SMAP) \ -+ HMAP_FOR_EACH_SAFE_SHORT_INIT ( \ -+ SMAP_NODE, node, &(SMAP)->map, \ -+ BUILD_ASSERT_TYPE(SMAP_NODE, struct smap_node *), \ -+ BUILD_ASSERT_TYPE(SMAP, struct smap *)) -+ -+#define SMAP_FOR_EACH_SAFE_LONG(SMAP_NODE, NEXT, SMAP) \ -+ HMAP_FOR_EACH_SAFE_LONG_INIT ( \ - SMAP_NODE, NEXT, node, &(SMAP)->map, \ - BUILD_ASSERT_TYPE(SMAP_NODE, struct smap_node *), \ - BUILD_ASSERT_TYPE(NEXT, struct smap_node *), \ - BUILD_ASSERT_TYPE(SMAP, struct smap *)) + /** +- * Build the intenal enic filter structure from the provided pattern. The ++ * Build the internal enic filter structure from the provided pattern. The + * pattern is validated as the items are copied. + * + * @param pattern[in] + * @param items_info[in] + * Info about this NICs item support, like valid previous items. + * @param enic_filter[out] +- * NIC specfilc filters derived from the pattern. ++ * NIC specific filters derived from the pattern. + * @param error[out] + */ + static int +@@ -1123,12 +1123,12 @@ enic_copy_filter(const struct rte_flow_item pattern[], + } -+#define SMAP_FOR_EACH_SAFE(...) \ -+ OVERLOAD_SAFE_MACRO(SMAP_FOR_EACH_SAFE_LONG, \ -+ SMAP_FOR_EACH_SAFE_SHORT, \ -+ 3, __VA_ARGS__) -+ - /* Initializer for an immutable struct smap 'SMAP' that contains one or two - * key-value pairs, e.g. + /** +- * Build the intenal version 1 NIC action structure from the provided pattern. ++ * Build the internal version 1 NIC action structure from the provided pattern. + * The pattern is validated as the items are copied. * -diff --git a/lib/socket-util.c b/lib/socket-util.c -index 4f1ffecf5d..38705cc51e 100644 ---- a/lib/socket-util.c -+++ b/lib/socket-util.c -@@ -62,7 +62,8 @@ static bool parse_sockaddr_components(struct sockaddr_storage *ss, - const char *port_s, - uint16_t default_port, - const char *s, -- bool resolve_host); -+ bool resolve_host, -+ bool *dns_failure); + * @param actions[in] + * @param enic_action[out] +- * NIC specfilc actions derived from the actions. ++ * NIC specific actions derived from the actions. + * @param error[out] + */ + static int +@@ -1170,12 +1170,12 @@ enic_copy_action_v1(__rte_unused struct enic *enic, + } - /* Sets 'fd' to non-blocking mode. Returns 0 if successful, otherwise a - * positive errno value. */ -@@ -438,7 +439,7 @@ parse_sockaddr_components_dns(struct sockaddr_storage *ss OVS_UNUSED, - dns_resolve(host_s, &tmp_host_s); - if (tmp_host_s != NULL) { - parse_sockaddr_components(ss, tmp_host_s, port_s, -- default_port, s, false); -+ default_port, s, false, NULL); - free(tmp_host_s); - return true; - } -@@ -450,11 +451,15 @@ parse_sockaddr_components(struct sockaddr_storage *ss, - char *host_s, - const char *port_s, uint16_t default_port, - const char *s, -- bool resolve_host) -+ bool resolve_host, bool *dns_failure) - { - struct sockaddr_in *sin = sin_cast(sa_cast(ss)); - int port; + /** +- * Build the intenal version 2 NIC action structure from the provided pattern. ++ * Build the internal version 2 NIC action structure from the provided pattern. + * The pattern is validated as the items are copied. + * + * @param actions[in] + * @param enic_action[out] +- * NIC specfilc actions derived from the actions. ++ * NIC specific actions derived from the actions. + * @param error[out] + */ + static int +diff --git a/dpdk/drivers/net/enic/enic_fm_flow.c b/dpdk/drivers/net/enic/enic_fm_flow.c +index ae43f36bc0..ab73cd8530 100644 +--- a/dpdk/drivers/net/enic/enic_fm_flow.c ++++ b/dpdk/drivers/net/enic/enic_fm_flow.c +@@ -721,7 +721,7 @@ enic_fm_copy_item_gtp(struct copy_item_args *arg) + } -+ if (dns_failure) { -+ *dns_failure = false; -+ } -+ - if (port_s && port_s[0]) { - if (!str_to_int(port_s, 10, &port) || port < 0 || port > 65535) { - VLOG_ERR("%s: bad port number \"%s\"", s, port_s); -@@ -501,10 +506,15 @@ parse_sockaddr_components(struct sockaddr_storage *ss, - return true; + /* NIC does not support GTP tunnels. No Items are allowed after this. +- * This prevents the specificaiton of further items. ++ * This prevents the specification of further items. + */ + arg->header_level = 0; + +@@ -733,7 +733,7 @@ enic_fm_copy_item_gtp(struct copy_item_args *arg) + + /* + * Use the raw L4 buffer to match GTP as fm_header_set does not have +- * GTP header. UDP dst port must be specifiec. Using the raw buffer ++ * GTP header. UDP dst port must be specific. Using the raw buffer + * does not affect such UDP item, since we skip UDP in the raw buffer. + */ + fm_data->fk_header_select |= FKH_L4RAW; +@@ -1846,7 +1846,7 @@ enic_fm_dump_tcam_actions(const struct fm_action *fm_action) + /* Remove trailing comma */ + if (buf[0]) + *(bp - 1) = '\0'; +- ENICPMD_LOG(DEBUG, " Acions: %s", buf); ++ ENICPMD_LOG(DEBUG, " Actions: %s", buf); + } - resolve: -- if (resolve_host && parse_sockaddr_components_dns(ss, host_s, port_s, -- default_port, s)) { -- return true; -- } else if (!resolve_host) { -+ if (resolve_host) { -+ if (parse_sockaddr_components_dns(ss, host_s, port_s, -+ default_port, s)) { -+ return true; -+ } -+ if (dns_failure) { -+ *dns_failure = true; -+ } -+ } else { - VLOG_ERR("%s: bad IP address \"%s\"", s, host_s); - } - exit: -@@ -521,10 +531,12 @@ exit: - * It resolves the host if 'resolve_host' is true. + static int +@@ -2364,7 +2364,7 @@ enic_action_handle_get(struct enic_flowman *fm, struct fm_action *action_in, + if (ret < 0 && ret != -ENOENT) + return rte_flow_error_set(error, -ret, + RTE_FLOW_ERROR_TYPE_UNSPECIFIED, +- NULL, "enic: rte_hash_lookup(aciton)"); ++ NULL, "enic: rte_hash_lookup(action)"); + + if (ret == -ENOENT) { + /* Allocate a new action on the NIC. */ +@@ -2372,11 +2372,11 @@ enic_action_handle_get(struct enic_flowman *fm, struct fm_action *action_in, + memcpy(fma, action_in, sizeof(*fma)); + + ah = calloc(1, sizeof(*ah)); +- memcpy(&ah->key, action_in, sizeof(struct fm_action)); + if (ah == NULL) + return rte_flow_error_set(error, ENOMEM, + RTE_FLOW_ERROR_TYPE_HANDLE, + NULL, "enic: calloc(fm-action)"); ++ memcpy(&ah->key, action_in, sizeof(struct fm_action)); + args[0] = FM_ACTION_ALLOC; + args[1] = fm->cmd.pa; + ret = flowman_cmd(fm, args, 2); +@@ -2435,7 +2435,7 @@ __enic_fm_flow_add_entry(struct enic_flowman *fm, + + ENICPMD_FUNC_TRACE(); + +- /* Get or create an aciton handle. */ ++ /* Get or create an action handle. */ + ret = enic_action_handle_get(fm, action_in, error, &ah); + if (ret) + return ret; +diff --git a/dpdk/drivers/net/enic/enic_main.c b/dpdk/drivers/net/enic/enic_main.c +index 7f84b5f935..97d97ea793 100644 +--- a/dpdk/drivers/net/enic/enic_main.c ++++ b/dpdk/drivers/net/enic/enic_main.c +@@ -1137,7 +1137,7 @@ int enic_disable(struct enic *enic) + } + + /* If we were using interrupts, set the interrupt vector to -1 +- * to disable interrupts. We are not disabling link notifcations, ++ * to disable interrupts. We are not disabling link notifications, + * though, as we want the polling of link status to continue working. + */ + if (enic->rte_dev->data->dev_conf.intr_conf.lsc) +diff --git a/dpdk/drivers/net/enic/enic_rxtx.c b/dpdk/drivers/net/enic/enic_rxtx.c +index c44715bfd0..33e96b480e 100644 +--- a/dpdk/drivers/net/enic/enic_rxtx.c ++++ b/dpdk/drivers/net/enic/enic_rxtx.c +@@ -653,7 +653,7 @@ static void enqueue_simple_pkts(struct rte_mbuf **pkts, + * The app should not send oversized + * packets. tx_pkt_prepare includes a check as + * well. But some apps ignore the device max size and +- * tx_pkt_prepare. Oversized packets cause WQ errrors ++ * tx_pkt_prepare. Oversized packets cause WQ errors + * and the NIC ends up disabling the whole WQ. So + * truncate packets.. + */ +diff --git a/dpdk/drivers/net/failsafe/failsafe.c b/dpdk/drivers/net/failsafe/failsafe.c +index 3c754a5f66..05cf533896 100644 +--- a/dpdk/drivers/net/failsafe/failsafe.c ++++ b/dpdk/drivers/net/failsafe/failsafe.c +@@ -308,8 +308,8 @@ fs_rte_eth_free(const char *name) + if (dev == NULL) + return 0; /* port already released */ + ret = failsafe_eth_dev_close(dev); +- rte_eth_dev_release_port(dev); + rte_intr_instance_free(PRIV(dev)->intr_handle); ++ rte_eth_dev_release_port(dev); + return ret; + } + +diff --git a/dpdk/drivers/net/fm10k/fm10k.h b/dpdk/drivers/net/fm10k/fm10k.h +index 7cfa29faa8..17a7056c45 100644 +--- a/dpdk/drivers/net/fm10k/fm10k.h ++++ b/dpdk/drivers/net/fm10k/fm10k.h +@@ -44,7 +44,7 @@ + #define FM10K_TX_MAX_MTU_SEG UINT8_MAX + + /* +- * byte aligment for HW RX data buffer ++ * byte alignment for HW RX data buffer + * Datasheet requires RX buffer addresses shall either be 512-byte aligned or + * be 8-byte aligned but without crossing host memory pages (4KB alignment + * boundaries). Satisfy first option. +diff --git a/dpdk/drivers/net/fm10k/fm10k_ethdev.c b/dpdk/drivers/net/fm10k/fm10k_ethdev.c +index 43e1d13431..8bbd8b445d 100644 +--- a/dpdk/drivers/net/fm10k/fm10k_ethdev.c ++++ b/dpdk/drivers/net/fm10k/fm10k_ethdev.c +@@ -290,7 +290,7 @@ rx_queue_free(struct fm10k_rx_queue *q) + } + + /* +- * disable RX queue, wait unitl HW finished necessary flush operation ++ * disable RX queue, wait until HW finished necessary flush operation + */ + static inline int + rx_queue_disable(struct fm10k_hw *hw, uint16_t qnum) +@@ -379,7 +379,7 @@ tx_queue_free(struct fm10k_tx_queue *q) + } + + /* +- * disable TX queue, wait unitl HW finished necessary flush operation ++ * disable TX queue, wait until HW finished necessary flush operation + */ + static inline int + tx_queue_disable(struct fm10k_hw *hw, uint16_t qnum) +@@ -453,7 +453,7 @@ fm10k_dev_configure(struct rte_eth_dev *dev) + if (dev->data->dev_conf.rxmode.mq_mode & RTE_ETH_MQ_RX_RSS_FLAG) + dev->data->dev_conf.rxmode.offloads |= RTE_ETH_RX_OFFLOAD_RSS_HASH; + +- /* multipe queue mode checking */ ++ /* multiple queue mode checking */ + ret = fm10k_check_mq_mode(dev); + if (ret != 0) { + PMD_DRV_LOG(ERR, "fm10k_check_mq_mode fails with %d.", +@@ -2553,7 +2553,7 @@ fm10k_dev_handle_fault(struct fm10k_hw *hw, uint32_t eicr) + * @param handle + * Pointer to interrupt handle. + * @param param +- * The address of parameter (struct rte_eth_dev *) regsitered before. ++ * The address of parameter (struct rte_eth_dev *) registered before. * - * On success, returns true and stores the parsed remote address into '*ss'. -- * On failure, logs an error, stores zeros into '*ss', and returns false. */ -+ * On failure, logs an error, stores zeros into '*ss', and returns false, -+ * '*dns_failure' indicates if the host resolution failed. */ - bool - inet_parse_active(const char *target_, int default_port, -- struct sockaddr_storage *ss, bool resolve_host) -+ struct sockaddr_storage *ss, -+ bool resolve_host, bool *dns_failure) - { - char *target = xstrdup(target_); - char *port, *host; -@@ -539,7 +551,7 @@ inet_parse_active(const char *target_, int default_port, - ok = false; - } else { - ok = parse_sockaddr_components(ss, host, port, default_port, -- target_, resolve_host); -+ target_, resolve_host, dns_failure); - } - if (!ok) { - memset(ss, 0, sizeof *ss); -@@ -576,7 +588,7 @@ inet_open_active(int style, const char *target, int default_port, - int error; + * @return + * void +@@ -2676,7 +2676,7 @@ fm10k_dev_interrupt_handler_pf(void *param) + * @param handle + * Pointer to interrupt handle. + * @param param +- * The address of parameter (struct rte_eth_dev *) regsitered before. ++ * The address of parameter (struct rte_eth_dev *) registered before. + * + * @return + * void +@@ -3034,7 +3034,7 @@ fm10k_params_init(struct rte_eth_dev *dev) + struct fm10k_dev_info *info = + FM10K_DEV_PRIVATE_TO_INFO(dev->data->dev_private); + +- /* Inialize bus info. Normally we would call fm10k_get_bus_info(), but ++ /* Initialize bus info. Normally we would call fm10k_get_bus_info(), but + * there is no way to get link status without reading BAR4. Until this + * works, assume we have maximum bandwidth. + * @todo - fix bus info +diff --git a/dpdk/drivers/net/fm10k/fm10k_rxtx_vec.c b/dpdk/drivers/net/fm10k/fm10k_rxtx_vec.c +index 1269250e23..10ce5a7582 100644 +--- a/dpdk/drivers/net/fm10k/fm10k_rxtx_vec.c ++++ b/dpdk/drivers/net/fm10k/fm10k_rxtx_vec.c +@@ -212,7 +212,7 @@ fm10k_rx_vec_condition_check(struct rte_eth_dev *dev) + struct rte_eth_fdir_conf *fconf = &dev->data->dev_conf.fdir_conf; + + #ifndef RTE_FM10K_RX_OLFLAGS_ENABLE +- /* whithout rx ol_flags, no VP flag report */ ++ /* without rx ol_flags, no VP flag report */ + if (rxmode->offloads & RTE_ETH_RX_OFFLOAD_VLAN_EXTEND) + return -1; + #endif +@@ -239,7 +239,7 @@ fm10k_rxq_vec_setup(struct fm10k_rx_queue *rxq) + struct rte_mbuf mb_def = { .buf_addr = 0 }; /* zeroed mbuf */ + + mb_def.nb_segs = 1; +- /* data_off will be ajusted after new mbuf allocated for 512-byte ++ /* data_off will be adjusted after new mbuf allocated for 512-byte + * alignment. + */ + mb_def.data_off = RTE_PKTMBUF_HEADROOM; +@@ -410,7 +410,7 @@ fm10k_recv_raw_pkts_vec(void *rx_queue, struct rte_mbuf **rx_pkts, + if (!(rxdp->d.staterr & FM10K_RXD_STATUS_DD)) + return 0; - /* Parse. */ -- if (!inet_parse_active(target, default_port, &ss, true)) { -+ if (!inet_parse_active(target, default_port, &ss, true, NULL)) { - error = EAFNOSUPPORT; - goto exit; - } -@@ -660,7 +672,7 @@ inet_parse_passive(const char *target_, int default_port, - ok = false; - } else { - ok = parse_sockaddr_components(ss, host, port, default_port, -- target_, true); -+ target_, true, NULL); - } - if (!ok) { - memset(ss, 0, sizeof *ss); -@@ -783,7 +795,8 @@ inet_parse_address(const char *target_, struct sockaddr_storage *ss) - { - char *target = xstrdup(target_); - char *host = unbracket(target); -- bool ok = parse_sockaddr_components(ss, host, NULL, 0, target_, false); -+ bool ok = parse_sockaddr_components(ss, host, NULL, 0, -+ target_, false, NULL); - if (!ok) { - memset(ss, 0, sizeof *ss); - } -diff --git a/lib/socket-util.h b/lib/socket-util.h -index 9ccb7d4cc4..bf66393df9 100644 ---- a/lib/socket-util.h -+++ b/lib/socket-util.h -@@ -49,7 +49,8 @@ ovs_be32 guess_netmask(ovs_be32 ip); - void inet_parse_host_port_tokens(char *s, char **hostp, char **portp); - void inet_parse_port_host_tokens(char *s, char **portp, char **hostp); - bool inet_parse_active(const char *target, int default_port, -- struct sockaddr_storage *ssp, bool resolve_host); -+ struct sockaddr_storage *ssp, -+ bool resolve_host, bool *dns_failure); - int inet_open_active(int style, const char *target, int default_port, - struct sockaddr_storage *ssp, int *fdp, uint8_t dscp); +- /* Vecotr RX will process 4 packets at a time, strip the unaligned ++ /* Vector RX will process 4 packets at a time, strip the unaligned + * tails in case it's not multiple of 4. + */ + nb_pkts = RTE_ALIGN_FLOOR(nb_pkts, RTE_FM10K_DESCS_PER_LOOP); +@@ -481,7 +481,7 @@ fm10k_recv_raw_pkts_vec(void *rx_queue, struct rte_mbuf **rx_pkts, + _mm_storeu_si128((__m128i *)&rx_pkts[pos], mbp1); + + #if defined(RTE_ARCH_X86_64) +- /* B.1 load 2 64 bit mbuf poitns */ ++ /* B.1 load 2 64 bit mbuf points */ + mbp2 = _mm_loadu_si128((__m128i *)&mbufp[pos+2]); + #endif -diff --git a/lib/sset.c b/lib/sset.c -index b2e3f43ec9..6fbaa9d60d 100644 ---- a/lib/sset.c -+++ b/lib/sset.c -@@ -212,9 +212,9 @@ sset_add_array(struct sset *set, char **names, size_t n) - void - sset_clear(struct sset *set) +@@ -573,7 +573,7 @@ fm10k_recv_raw_pkts_vec(void *rx_queue, struct rte_mbuf **rx_pkts, + + fm10k_desc_to_pktype_v(descs0, &rx_pkts[pos]); + +- /* C.4 calc avaialbe number of desc */ ++ /* C.4 calc available number of desc */ + var = __builtin_popcountll(_mm_cvtsi128_si64(staterr)); + nb_pkts_recd += var; + if (likely(var != RTE_FM10K_DESCS_PER_LOOP)) +diff --git a/dpdk/drivers/net/hinic/hinic_pmd_ethdev.c b/dpdk/drivers/net/hinic/hinic_pmd_ethdev.c +index 1853511c3b..e8d9aaba84 100644 +--- a/dpdk/drivers/net/hinic/hinic_pmd_ethdev.c ++++ b/dpdk/drivers/net/hinic/hinic_pmd_ethdev.c +@@ -255,7 +255,7 @@ static int hinic_vlan_offload_set(struct rte_eth_dev *dev, int mask); + * Interrupt handler triggered by NIC for handling + * specific event. + * +- * @param: The address of parameter (struct rte_eth_dev *) regsitered before. ++ * @param: The address of parameter (struct rte_eth_dev *) registered before. + */ + static void hinic_dev_interrupt_handler(void *param) { -- const char *name, *next; -+ const char *name; +@@ -336,7 +336,7 @@ static int hinic_dev_configure(struct rte_eth_dev *dev) + return err; + } -- SSET_FOR_EACH_SAFE (name, next, set) { -+ SSET_FOR_EACH_SAFE (name, set) { - sset_delete(set, SSET_NODE_FROM_NAME(name)); - } +- /* init vlan offoad */ ++ /* init VLAN offload */ + err = hinic_vlan_offload_set(dev, + RTE_ETH_VLAN_STRIP_MASK | RTE_ETH_VLAN_FILTER_MASK); + if (err) { +diff --git a/dpdk/drivers/net/hinic/hinic_pmd_ethdev.h b/dpdk/drivers/net/hinic/hinic_pmd_ethdev.h +index 5eca8b10b9..8e6251f69f 100644 +--- a/dpdk/drivers/net/hinic/hinic_pmd_ethdev.h ++++ b/dpdk/drivers/net/hinic/hinic_pmd_ethdev.h +@@ -170,7 +170,7 @@ struct tag_tcam_key_mem { + /* + * tunnel packet, mask must be 0xff, spec value is 1; + * normal packet, mask must be 0, spec value is 0; +- * if tunnal packet, ucode use ++ * if tunnel packet, ucode use + * sip/dip/protocol/src_port/dst_dport from inner packet + */ + u32 tunnel_flag:8; +diff --git a/dpdk/drivers/net/hinic/hinic_pmd_flow.c b/dpdk/drivers/net/hinic/hinic_pmd_flow.c +index d71a42afbd..2cf24ebcf6 100644 +--- a/dpdk/drivers/net/hinic/hinic_pmd_flow.c ++++ b/dpdk/drivers/net/hinic/hinic_pmd_flow.c +@@ -734,7 +734,7 @@ static int hinic_check_ntuple_item_ele(const struct rte_flow_item *item, + * END + * other members in mask and spec should set to 0x00. + * item->last should be NULL. +- * Please aware there's an asumption for all the parsers. ++ * Please be aware there's an assumption for all the parsers. + * rte_flow_item is using big endian, rte_flow_attr and + * rte_flow_action are using CPU order. + * Because the pattern is used to describe the packets, +@@ -1630,7 +1630,7 @@ static int hinic_parse_fdir_filter(struct rte_eth_dev *dev, + + /** + * Check if the flow rule is supported by nic. +- * It only checkes the format. Don't guarantee the rule can be programmed into ++ * It only checks the format. Don't guarantee the rule can be programmed into + * the HW. Because there can be no enough room for the rule. + */ + static int hinic_flow_validate(struct rte_eth_dev *dev, +diff --git a/dpdk/drivers/net/hinic/hinic_pmd_tx.c b/dpdk/drivers/net/hinic/hinic_pmd_tx.c +index 2688817f37..f09b1a6e1e 100644 +--- a/dpdk/drivers/net/hinic/hinic_pmd_tx.c ++++ b/dpdk/drivers/net/hinic/hinic_pmd_tx.c +@@ -1144,7 +1144,7 @@ u16 hinic_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts, u16 nb_pkts) + mbuf_pkt = *tx_pkts++; + queue_info = 0; + +- /* 1. parse sge and tx offlod info from mbuf */ ++ /* 1. parse sge and tx offload info from mbuf */ + if (unlikely(!hinic_get_sge_txoff_info(mbuf_pkt, + &sqe_info, &off_info))) { + txq->txq_stats.off_errs++; +diff --git a/dpdk/drivers/net/hns3/hns3_cmd.c b/dpdk/drivers/net/hns3/hns3_cmd.c +index 2ce59d8de6..3495e2acc1 100644 +--- a/dpdk/drivers/net/hns3/hns3_cmd.c ++++ b/dpdk/drivers/net/hns3/hns3_cmd.c +@@ -466,7 +466,7 @@ hns3_mask_capability(struct hns3_hw *hw, + for (i = 0; i < MAX_CAPS_BIT; i++) { + if (!(caps_masked & BIT_ULL(i))) + continue; +- hns3_info(hw, "mask capabiliy: id-%u, name-%s.", ++ hns3_info(hw, "mask capability: id-%u, name-%s.", + i, hns3_get_caps_name(i)); + } } -@@ -312,7 +312,9 @@ sset_at_position(const struct sset *set, struct sset_position *pos) - struct hmap_node *hmap_node; +@@ -635,39 +635,6 @@ hns3_firmware_compat_config(struct hns3_hw *hw, bool is_init) + struct hns3_cmd_desc desc; + uint32_t compat = 0; + +-#if defined(RTE_HNS3_ONLY_1630_FPGA) +- /* If resv reg enabled phy driver of imp is not configured, driver +- * will use temporary phy driver. +- */ +- struct rte_pci_device *pci_dev; +- struct rte_eth_dev *eth_dev; +- uint8_t revision; +- int ret; +- +- eth_dev = &rte_eth_devices[hw->data->port_id]; +- pci_dev = RTE_ETH_DEV_TO_PCI(eth_dev); +- /* Get PCI revision id */ +- ret = rte_pci_read_config(pci_dev, &revision, HNS3_PCI_REVISION_ID_LEN, +- HNS3_PCI_REVISION_ID); +- if (ret != HNS3_PCI_REVISION_ID_LEN) { +- PMD_INIT_LOG(ERR, "failed to read pci revision id, ret = %d", +- ret); +- return -EIO; +- } +- if (revision == PCI_REVISION_ID_HIP09_A) { +- struct hns3_pf *pf = HNS3_DEV_HW_TO_PF(hw); +- if (hns3_dev_get_support(hw, COPPER) == 0 || pf->is_tmp_phy) { +- PMD_INIT_LOG(ERR, "***use temp phy driver in dpdk***"); +- pf->is_tmp_phy = true; +- hns3_set_bit(hw->capability, +- HNS3_DEV_SUPPORT_COPPER_B, 1); +- return 0; +- } +- +- PMD_INIT_LOG(ERR, "***use phy driver in imp***"); +- } +-#endif +- + hns3_cmd_setup_basic_desc(&desc, HNS3_OPC_FIRMWARE_COMPAT_CFG, false); + req = (struct hns3_firmware_compat_cmd *)desc.data; - hmap_node = hmap_at_position(&set->map, &pos->pos); -- return SSET_NODE_FROM_HMAP_NODE(hmap_node); -+ return hmap_node -+ ? SSET_NODE_FROM_HMAP_NODE(hmap_node) -+ : NULL; +@@ -736,7 +703,7 @@ hns3_cmd_init(struct hns3_hw *hw) + return 0; + + /* +- * Requiring firmware to enable some features, firber port can still ++ * Requiring firmware to enable some features, fiber port can still + * work without it, but copper port can't work because the firmware + * fails to take over the PHY. + */ +diff --git a/dpdk/drivers/net/hns3/hns3_cmd.h b/dpdk/drivers/net/hns3/hns3_cmd.h +index 81bc9e9d98..82c999061d 100644 +--- a/dpdk/drivers/net/hns3/hns3_cmd.h ++++ b/dpdk/drivers/net/hns3/hns3_cmd.h +@@ -323,7 +323,7 @@ enum HNS3_CAPS_BITS { + HNS3_CAPS_UDP_TUNNEL_CSUM_B, + HNS3_CAPS_RAS_IMP_B, + HNS3_CAPS_RXD_ADV_LAYOUT_B = 15, +- HNS3_CAPS_TM_B = 17, ++ HNS3_CAPS_TM_B = 19, + }; + + /* Capabilities of VF dependent on the PF */ +@@ -603,7 +603,6 @@ struct hns3_cfg_gro_status_cmd { + + #define HNS3_RSS_HASH_KEY_OFFSET_B 4 + +-#define HNS3_RSS_CFG_TBL_SIZE 16 + #define HNS3_RSS_HASH_KEY_NUM 16 + /* Configure the algorithm mode and Hash Key, opcode:0x0D01 */ + struct hns3_rss_generic_config_cmd { +diff --git a/dpdk/drivers/net/hns3/hns3_common.c b/dpdk/drivers/net/hns3/hns3_common.c +index eac2aa1040..78158401f2 100644 +--- a/dpdk/drivers/net/hns3/hns3_common.c ++++ b/dpdk/drivers/net/hns3/hns3_common.c +@@ -216,7 +216,7 @@ hns3_parse_mbx_time_limit(const char *key, const char *value, void *extra_args) + + /* + * 500ms is empirical value in process of mailbox communication. If +- * the delay value is set to one lower thanthe empirical value, mailbox ++ * the delay value is set to one lower than the empirical value, mailbox + * communication may fail. + */ + if (val > HNS3_MBX_DEF_TIME_LIMIT_MS && val <= UINT16_MAX) +@@ -236,6 +236,12 @@ hns3_parse_devargs(struct rte_eth_dev *dev) + uint64_t dev_caps_mask = 0; + struct rte_kvargs *kvlist; + ++ /* Set default value of runtime config parameters. */ ++ hns->rx_func_hint = HNS3_IO_FUNC_HINT_NONE; ++ hns->tx_func_hint = HNS3_IO_FUNC_HINT_NONE; ++ hns->dev_caps_mask = 0; ++ hns->mbx_time_limit_ms = HNS3_MBX_DEF_TIME_LIMIT_MS; ++ + if (dev->device->devargs == NULL) + return; + +@@ -603,7 +609,7 @@ hns3_init_ring_with_vector(struct hns3_hw *hw) + hw->intr_tqps_num = RTE_MIN(vec, hw->tqps_num); + for (i = 0; i < hw->intr_tqps_num; i++) { + /* +- * Set gap limiter/rate limiter/quanity limiter algorithm ++ * Set gap limiter/rate limiter/quantity limiter algorithm + * configuration for interrupt coalesce of queue's interrupt. + */ + hns3_set_queue_intr_gl(hw, i, HNS3_RING_GL_RX, +diff --git a/dpdk/drivers/net/hns3/hns3_dcb.c b/dpdk/drivers/net/hns3/hns3_dcb.c +index 3d0159d787..e4417e87fd 100644 +--- a/dpdk/drivers/net/hns3/hns3_dcb.c ++++ b/dpdk/drivers/net/hns3/hns3_dcb.c +@@ -25,7 +25,7 @@ + * IR(Mbps) = ------------------------- * CLOCK(1000Mbps) + * Tick * (2 ^ IR_s) + * +- * @return: 0: calculate sucessful, negative: fail ++ * @return: 0: calculate successful, negative: fail + */ + static int + hns3_shaper_para_calc(struct hns3_hw *hw, uint32_t ir, uint8_t shaper_level, +@@ -36,8 +36,8 @@ hns3_shaper_para_calc(struct hns3_hw *hw, uint32_t ir, uint8_t shaper_level, + #define DIVISOR_IR_B_126 (126 * DIVISOR_CLK) + + const uint16_t tick_array[HNS3_SHAPER_LVL_CNT] = { +- 6 * 256, /* Prioriy level */ +- 6 * 32, /* Prioriy group level */ ++ 6 * 256, /* Priority level */ ++ 6 * 32, /* Priority group level */ + 6 * 8, /* Port level */ + 6 * 256 /* Qset level */ + }; +@@ -1532,7 +1532,7 @@ hns3_dcb_hw_configure(struct hns3_adapter *hns) + + ret = hns3_dcb_schd_setup_hw(hw); + if (ret) { +- hns3_err(hw, "dcb schdule configure failed! ret = %d", ret); ++ hns3_err(hw, "dcb schedule configure failed! ret = %d", ret); + return ret; + } + +@@ -1737,7 +1737,7 @@ hns3_get_fc_mode(struct hns3_hw *hw, enum rte_eth_fc_mode mode) + * hns3_dcb_pfc_enable - Enable priority flow control + * @dev: pointer to ethernet device + * +- * Configures the pfc settings for one porority. ++ * Configures the pfc settings for one priority. + */ + int + hns3_dcb_pfc_enable(struct rte_eth_dev *dev, struct rte_eth_pfc_conf *pfc_conf) +diff --git a/dpdk/drivers/net/hns3/hns3_ethdev.c b/dpdk/drivers/net/hns3/hns3_ethdev.c +index 0bd12907d8..40a33549e0 100644 +--- a/dpdk/drivers/net/hns3/hns3_ethdev.c ++++ b/dpdk/drivers/net/hns3/hns3_ethdev.c +@@ -227,17 +227,11 @@ hns3_check_event_cause(struct hns3_adapter *hns, uint32_t *clearval) + return ret; } - /* Replaces 'a' by the intersection of 'a' and 'b'. That is, removes from 'a' -@@ -320,9 +322,9 @@ sset_at_position(const struct sset *set, struct sset_position *pos) - void - sset_intersect(struct sset *a, const struct sset *b) +-static bool +-hns3_is_1588_event_type(uint32_t event_type) +-{ +- return (event_type == HNS3_VECTOR0_EVENT_PTP); +-} +- + static void + hns3_clear_event_cause(struct hns3_hw *hw, uint32_t event_type, uint32_t regclr) { -- const char *name, *next; -+ const char *name; + if (event_type == HNS3_VECTOR0_EVENT_RST || +- hns3_is_1588_event_type(event_type)) ++ event_type == HNS3_VECTOR0_EVENT_PTP) + hns3_write_dev(hw, HNS3_MISC_RESET_STS_REG, regclr); + else if (event_type == HNS3_VECTOR0_EVENT_MBX) + hns3_write_dev(hw, HNS3_VECTOR0_CMDQ_SRC_REG, regclr); +@@ -324,7 +318,7 @@ hns3_interrupt_handler(void *param) + hns3_schedule_reset(hns); + } else if (event_cause == HNS3_VECTOR0_EVENT_MBX) { + hns3_dev_handle_mbx_msg(hw); +- } else { ++ } else if (event_cause != HNS3_VECTOR0_EVENT_PTP) { + hns3_warn(hw, "received unknown event: vector0_int_stat:0x%x " + "ras_int_stat:0x%x cmdq_int_stat:0x%x", + vector0_int, ras_int, cmdq_int); +@@ -574,7 +568,7 @@ hns3_set_vlan_rx_offload_cfg(struct hns3_adapter *hns, + hns3_set_bit(req->vport_vlan_cfg, HNS3_SHOW_TAG2_EN_B, + vcfg->vlan2_vlan_prionly ? 1 : 0); + +- /* firmwall will ignore this configuration for PCI_REVISION_ID_HIP08 */ ++ /* firmware will ignore this configuration for PCI_REVISION_ID_HIP08 */ + hns3_set_bit(req->vport_vlan_cfg, HNS3_DISCARD_TAG1_EN_B, + vcfg->strip_tag1_discard_en ? 1 : 0); + hns3_set_bit(req->vport_vlan_cfg, HNS3_DISCARD_TAG2_EN_B, +@@ -784,7 +778,7 @@ hns3_set_vlan_tx_offload_cfg(struct hns3_adapter *hns, + vcfg->insert_tag2_en ? 1 : 0); + hns3_set_bit(req->vport_vlan_cfg, HNS3_CFG_NIC_ROCE_SEL_B, 0); + +- /* firmwall will ignore this configuration for PCI_REVISION_ID_HIP08 */ ++ /* firmware will ignore this configuration for PCI_REVISION_ID_HIP08 */ + hns3_set_bit(req->vport_vlan_cfg, HNS3_TAG_SHIFT_MODE_EN_B, + vcfg->tag_shift_mode_en ? 1 : 0); + +@@ -2033,11 +2027,9 @@ hns3_dev_configure(struct rte_eth_dev *dev) + goto cfg_err; + } -- SSET_FOR_EACH_SAFE (name, next, a) { -+ SSET_FOR_EACH_SAFE (name, a) { - if (!sset_contains(b, name)) { - sset_delete(a, SSET_NODE_FROM_NAME(name)); - } -diff --git a/lib/sset.h b/lib/sset.h -index f0bb8b5344..214d6fb41c 100644 ---- a/lib/sset.h -+++ b/lib/sset.h -@@ -87,13 +87,26 @@ void sset_intersect(struct sset *, const struct sset *); - NAME != NULL; \ - (NAME) = SSET_NEXT(SSET, NAME)) +- /* When RSS is not configured, redirect the packet queue 0 */ + if ((uint32_t)mq_mode & RTE_ETH_MQ_RX_RSS_FLAG) { + conf->rxmode.offloads |= RTE_ETH_RX_OFFLOAD_RSS_HASH; + rss_conf = conf->rx_adv_conf.rss_conf; +- hw->rss_dis_flag = false; + ret = hns3_dev_rss_hash_update(dev, &rss_conf); + if (ret) + goto cfg_err; +@@ -2093,7 +2085,6 @@ static int + hns3_config_mtu(struct hns3_hw *hw, uint16_t mps) + { + struct hns3_adapter *hns = HNS3_DEV_HW_TO_ADAPTER(hw); +- uint16_t original_mps = hns->pf.mps; + int err; + int ret; --#define SSET_FOR_EACH_SAFE(NAME, NEXT, SSET) \ -+#define SSET_FOR_EACH_SAFE_LONG(NAME, NEXT, SSET) \ - for ((NAME) = SSET_FIRST(SSET); \ - (NAME != NULL \ - ? (NEXT) = SSET_NEXT(SSET, NAME), true \ - : false); \ - (NAME) = (NEXT)) +@@ -2103,22 +2094,20 @@ hns3_config_mtu(struct hns3_hw *hw, uint16_t mps) + return ret; + } -+#define SSET_FOR_EACH_SAFE_SHORT(NAME, SSET) \ -+ for (const char * NAME__next = \ -+ ((NAME) = SSET_FIRST(SSET), NULL); \ -+ (NAME != NULL \ -+ ? (NAME__next = SSET_NEXT(SSET, NAME), true) \ -+ : (NAME__next = NULL, false)); \ -+ (NAME) = NAME__next) +- hns->pf.mps = mps; + ret = hns3_buffer_alloc(hw); + if (ret) { + hns3_err(hw, "failed to allocate buffer, ret = %d", ret); + goto rollback; + } + ++ hns->pf.mps = mps; + -+#define SSET_FOR_EACH_SAFE(...) \ -+ OVERLOAD_SAFE_MACRO(SSET_FOR_EACH_SAFE_LONG, \ -+ SSET_FOR_EACH_SAFE_SHORT, \ -+ 3, __VA_ARGS__) + return 0; + + rollback: +- err = hns3_set_mac_mtu(hw, original_mps); +- if (err) { ++ err = hns3_set_mac_mtu(hw, hns->pf.mps); ++ if (err) + hns3_err(hw, "fail to rollback MTU, err = %d", err); +- return ret; +- } +- hns->pf.mps = original_mps; + + return ret; + } +@@ -2767,6 +2756,10 @@ hns3_get_capability(struct hns3_hw *hw) + } + hw->revision = revision; + ++ ret = hns3_query_mac_stats_reg_num(hw); ++ if (ret) ++ return ret; + - const char **sset_array(const struct sset *); - const char **sset_sort(const struct sset *); - -diff --git a/lib/stopwatch.c b/lib/stopwatch.c -index 1c71df1a12..ec567603b1 100644 ---- a/lib/stopwatch.c -+++ b/lib/stopwatch.c -@@ -464,7 +464,7 @@ stopwatch_thread(void *ign OVS_UNUSED) - static void - stopwatch_exit(void) + if (revision < PCI_REVISION_ID_HIP09_A) { + hns3_set_default_dev_specifications(hw); + hw->intr.mapping_mode = HNS3_INTR_MAPPING_VEC_RSV_ONE; +@@ -2820,11 +2813,8 @@ hns3_check_media_type(struct hns3_hw *hw, uint8_t media_type) + } + break; + case HNS3_MEDIA_TYPE_FIBER: +- ret = 0; +- break; + case HNS3_MEDIA_TYPE_BACKPLANE: +- PMD_INIT_LOG(ERR, "Media type is Backplane, not supported."); +- ret = -EOPNOTSUPP; ++ ret = 0; + break; + default: + PMD_INIT_LOG(ERR, "Unknown media type = %u!", media_type); +@@ -2855,7 +2845,6 @@ hns3_get_board_configuration(struct hns3_hw *hw) + + hw->mac.media_type = cfg.media_type; + hw->rss_size_max = cfg.rss_size_max; +- hw->rss_dis_flag = false; + memcpy(hw->mac.mac_addr, cfg.mac_addr, RTE_ETHER_ADDR_LEN); + hw->mac.phy_addr = cfg.phy_addr; + hw->num_tx_desc = cfg.tqp_desc_num; +@@ -3420,7 +3409,7 @@ hns3_only_alloc_priv_buff(struct hns3_hw *hw, + * hns3_rx_buffer_calc: calculate the rx private buffer size for all TCs + * @hw: pointer to struct hns3_hw + * @buf_alloc: pointer to buffer calculation data +- * @return: 0: calculate sucessful, negative: fail ++ * @return: 0: calculate successful, negative: fail + */ + static int + hns3_rx_buffer_calc(struct hns3_hw *hw, struct hns3_pkt_buf_alloc *buf_alloc) +@@ -4280,14 +4269,11 @@ hns3_update_link_info(struct rte_eth_dev *eth_dev) { -- struct shash_node *node, *node_next; -+ struct shash_node *node; - struct stopwatch_packet *pkt = stopwatch_packet_create(OP_SHUTDOWN); - stopwatch_packet_write(pkt); - xpthread_join(stopwatch_thread_id, NULL); -@@ -473,7 +473,7 @@ stopwatch_exit(void) - * other competing thread. We are now the sole owners - * of all data in the file. - */ -- SHASH_FOR_EACH_SAFE (node, node_next, &stopwatches) { -+ SHASH_FOR_EACH_SAFE (node, &stopwatches) { - struct stopwatch *sw = node->data; - shash_delete(&stopwatches, node); - free(sw); -diff --git a/lib/stream.c b/lib/stream.c -index fcaddf10ad..71039e24f1 100644 ---- a/lib/stream.c -+++ b/lib/stream.c -@@ -788,7 +788,7 @@ stream_parse_target_with_default_port(const char *target, int default_port, - struct sockaddr_storage *ss) - { - return ((!strncmp(target, "tcp:", 4) || !strncmp(target, "ssl:", 4)) -- && inet_parse_active(target + 4, default_port, ss, true)); -+ && inet_parse_active(target + 4, default_port, ss, true, NULL)); + struct hns3_adapter *hns = eth_dev->data->dev_private; + struct hns3_hw *hw = &hns->hw; +- int ret = 0; + + if (hw->mac.media_type == HNS3_MEDIA_TYPE_COPPER) +- ret = hns3_update_copper_link_info(hw); +- else if (hw->mac.media_type == HNS3_MEDIA_TYPE_FIBER) +- ret = hns3_update_fiber_link_info(hw); ++ return hns3_update_copper_link_info(hw); + +- return ret; ++ return hns3_update_fiber_link_info(hw); } - /* Attempts to guess the content type of a stream whose first few bytes were -diff --git a/lib/tc.c b/lib/tc.c -index adb2d3182a..6301e30365 100644 ---- a/lib/tc.c -+++ b/lib/tc.c -@@ -568,16 +568,17 @@ nl_parse_flower_vlan(struct nlattr **attrs, struct tc_flower *flower) - - flower->key.encap_eth_type[0] = - nl_attr_get_be16(attrs[TCA_FLOWER_KEY_ETH_TYPE]); -+ flower->mask.encap_eth_type[0] = CONSTANT_HTONS(0xffff); + static int +@@ -4396,10 +4382,12 @@ hns3_service_handler(void *param) + struct hns3_adapter *hns = eth_dev->data->dev_private; + struct hns3_hw *hw = &hns->hw; + +- if (!hns3_is_reset_pending(hns)) ++ if (!hns3_is_reset_pending(hns)) { + hns3_update_linkstatus_and_event(hw, true); +- else ++ hns3_update_hw_stats(hw); ++ } else { + hns3_warn(hw, "Cancel the query when reset is pending"); ++ } + + rte_eal_alarm_set(HNS3_SERVICE_INTERVAL, hns3_service_handler, eth_dev); + } +@@ -4410,6 +4398,10 @@ hns3_init_hardware(struct hns3_adapter *hns) + struct hns3_hw *hw = &hns->hw; + int ret; + ++ /* ++ * All queue-related HW operations must be performed after the TCAM ++ * table is configured. ++ */ + ret = hns3_map_tqp(hw); + if (ret) { + PMD_INIT_LOG(ERR, "Failed to map tqp: %d", ret); +@@ -4550,14 +4542,14 @@ hns3_set_firber_default_support_speed(struct hns3_hw *hw) + } - if (attrs[TCA_FLOWER_KEY_VLAN_ID]) { - flower->key.vlan_id[0] = - nl_attr_get_u16(attrs[TCA_FLOWER_KEY_VLAN_ID]); -- flower->mask.vlan_id[0] = 0xffff; -+ flower->mask.vlan_id[0] = VLAN_VID_MASK >> VLAN_VID_SHIFT; - } - if (attrs[TCA_FLOWER_KEY_VLAN_PRIO]) { - flower->key.vlan_prio[0] = - nl_attr_get_u8(attrs[TCA_FLOWER_KEY_VLAN_PRIO]); -- flower->mask.vlan_prio[0] = 0xff; -+ flower->mask.vlan_prio[0] = VLAN_PCP_MASK >> VLAN_PCP_SHIFT; - } + /* +- * Validity of supported_speed for firber and copper media type can be ++ * Validity of supported_speed for fiber and copper media type can be + * guaranteed by the following policy: + * Copper: + * Although the initialization of the phy in the firmware may not be + * completed, the firmware can guarantees that the supported_speed is + * an valid value. + * Firber: +- * If the version of firmware supports the acitive query way of the ++ * If the version of firmware supports the active query way of the + * HNS3_OPC_GET_SFP_INFO opcode, the supported_speed can be obtained + * through it. If unsupported, use the SFP's speed as the value of the + * supported_speed. +@@ -4574,11 +4566,13 @@ hns3_get_port_supported_speed(struct rte_eth_dev *eth_dev) + if (ret) + return ret; + +- if (mac->media_type == HNS3_MEDIA_TYPE_FIBER) { ++ if (mac->media_type == HNS3_MEDIA_TYPE_FIBER || ++ mac->media_type == HNS3_MEDIA_TYPE_BACKPLANE) { + /* + * Some firmware does not support the report of supported_speed, +- * and only report the effective speed of SFP. In this case, it +- * is necessary to use the SFP's speed as the supported_speed. ++ * and only report the effective speed of SFP/backplane. In this ++ * case, it is necessary to use the SFP/backplane's speed as the ++ * supported_speed. + */ + if (mac->supported_speed == 0) + mac->supported_speed = +@@ -4650,13 +4644,6 @@ hns3_init_pf(struct rte_eth_dev *eth_dev) + goto err_cmd_init; + } - if (!attrs[TCA_FLOWER_KEY_VLAN_ETH_TYPE]) { -@@ -590,17 +591,18 @@ nl_parse_flower_vlan(struct nlattr **attrs, struct tc_flower *flower) - } +- /* Hardware statistics of imissed registers cleared. */ +- ret = hns3_update_imissed_stats(hw, true); +- if (ret) { +- hns3_err(hw, "clear imissed stats failed, ret = %d", ret); +- goto err_cmd_init; +- } +- + hns3_config_all_msix_error(hw, true); - flower->key.encap_eth_type[1] = flower->key.encap_eth_type[0]; -+ flower->mask.encap_eth_type[1] = CONSTANT_HTONS(0xffff); - flower->key.encap_eth_type[0] = encap_ethtype; + ret = rte_intr_callback_register(pci_dev->intr_handle, +@@ -4682,7 +4669,7 @@ hns3_init_pf(struct rte_eth_dev *eth_dev) + goto err_get_config; + } - if (attrs[TCA_FLOWER_KEY_CVLAN_ID]) { - flower->key.vlan_id[1] = - nl_attr_get_u16(attrs[TCA_FLOWER_KEY_CVLAN_ID]); -- flower->mask.vlan_id[1] = 0xffff; -+ flower->mask.vlan_id[1] = VLAN_VID_MASK >> VLAN_VID_SHIFT; - } - if (attrs[TCA_FLOWER_KEY_CVLAN_PRIO]) { - flower->key.vlan_prio[1] = - nl_attr_get_u8(attrs[TCA_FLOWER_KEY_CVLAN_PRIO]); -- flower->mask.vlan_prio[1] = 0xff; -+ flower->mask.vlan_prio[1] = VLAN_PCP_MASK >> VLAN_PCP_SHIFT; - } +- ret = hns3_tqp_stats_init(hw); ++ ret = hns3_stats_init(hw); + if (ret) + goto err_get_config; + +@@ -4728,7 +4715,7 @@ hns3_init_pf(struct rte_eth_dev *eth_dev) + err_fdir: + hns3_uninit_umv_space(hw); + err_init_hw: +- hns3_tqp_stats_uninit(hw); ++ hns3_stats_uninit(hw); + err_get_config: + hns3_pf_disable_irq0(hw); + rte_intr_disable(pci_dev->intr_handle); +@@ -4762,7 +4749,7 @@ hns3_uninit_pf(struct rte_eth_dev *eth_dev) + hns3_flow_uninit(eth_dev); + hns3_fdir_filter_uninit(hns); + hns3_uninit_umv_space(hw); +- hns3_tqp_stats_uninit(hw); ++ hns3_stats_uninit(hw); + hns3_config_mac_tnl_int(hw, false); + hns3_pf_disable_irq0(hw); + rte_intr_disable(pci_dev->intr_handle); +@@ -4847,7 +4834,7 @@ hns3_check_port_speed(struct hns3_hw *hw, uint32_t link_speeds) + + if (mac->media_type == HNS3_MEDIA_TYPE_COPPER) + speed_bit = hns3_convert_link_speeds2bitmap_copper(link_speeds); +- else if (mac->media_type == HNS3_MEDIA_TYPE_FIBER) ++ else + speed_bit = hns3_convert_link_speeds2bitmap_fiber(link_speeds); + + if (!(speed_bit & supported_speed)) { +@@ -4991,32 +4978,35 @@ hns3_set_fiber_port_link_speed(struct hns3_hw *hw, + return hns3_cfg_mac_speed_dup(hw, cfg->speed, cfg->duplex); } -@@ -937,24 +939,21 @@ nl_parse_flower_ip(struct nlattr **attrs, struct tc_flower *flower) { - key->icmp_code = - nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV4_CODE]); - mask->icmp_code = -- nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV4_CODE]); -+ nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV4_CODE_MASK]); - } - if (attrs[TCA_FLOWER_KEY_ICMPV4_TYPE_MASK]) { -- key->icmp_type = -- nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV4_TYPE_MASK]); -+ key->icmp_type = nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV4_TYPE]); - mask->icmp_type = - nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV4_TYPE_MASK]); - } - } else if (ip_proto == IPPROTO_ICMPV6) { - if (attrs[TCA_FLOWER_KEY_ICMPV6_CODE_MASK]) { -- key->icmp_code = -- nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV6_CODE]); -+ key->icmp_code = nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV6_CODE]); - mask->icmp_code = -- nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV6_CODE]); -+ nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV6_CODE_MASK]); - } - if (attrs[TCA_FLOWER_KEY_ICMPV6_TYPE_MASK]) { -- key->icmp_type = -- nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV6_TYPE_MASK]); -+ key->icmp_type = nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV6_TYPE]); - mask->icmp_type = - nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV6_TYPE_MASK]); - } -@@ -1006,14 +1005,14 @@ static const struct nl_policy pedit_policy[] = { ++static const char * ++hns3_get_media_type_name(uint8_t media_type) ++{ ++ if (media_type == HNS3_MEDIA_TYPE_FIBER) ++ return "fiber"; ++ else if (media_type == HNS3_MEDIA_TYPE_COPPER) ++ return "copper"; ++ else if (media_type == HNS3_MEDIA_TYPE_BACKPLANE) ++ return "backplane"; ++ else ++ return "unknown"; ++} ++ static int - nl_parse_act_pedit(struct nlattr *options, struct tc_flower *flower) + hns3_set_port_link_speed(struct hns3_hw *hw, + struct hns3_set_link_speed_cfg *cfg) { -- struct tc_action *action; -+ struct tc_action *action = &flower->actions[flower->action_count++]; - struct nlattr *pe_attrs[ARRAY_SIZE(pedit_policy)]; - const struct tc_pedit *pe; - const struct tc_pedit_key *keys; - const struct nlattr *nla, *keys_ex, *ex_type; - const void *keys_attr; -- char *rewrite_key = (void *) &flower->rewrite.key; -- char *rewrite_mask = (void *) &flower->rewrite.mask; -+ char *rewrite_key = (void *) &action->rewrite.key; -+ char *rewrite_mask = (void *) &action->rewrite.mask; - size_t keys_ex_size, left; - int type, i = 0, err; + int ret; -@@ -1092,7 +1091,6 @@ nl_parse_act_pedit(struct nlattr *options, struct tc_flower *flower) - i++; - } +- if (hw->mac.media_type == HNS3_MEDIA_TYPE_COPPER) { +-#if defined(RTE_HNS3_ONLY_1630_FPGA) +- struct hns3_pf *pf = HNS3_DEV_HW_TO_PF(hw); +- if (pf->is_tmp_phy) +- return 0; +-#endif +- ++ if (hw->mac.media_type == HNS3_MEDIA_TYPE_COPPER) + ret = hns3_set_copper_port_link_speed(hw, cfg); +- if (ret) { +- hns3_err(hw, "failed to set copper port link speed," +- "ret = %d.", ret); +- return ret; +- } +- } else if (hw->mac.media_type == HNS3_MEDIA_TYPE_FIBER) { ++ else + ret = hns3_set_fiber_port_link_speed(hw, cfg); +- if (ret) { +- hns3_err(hw, "failed to set fiber port link speed," +- "ret = %d.", ret); +- return ret; +- } ++ ++ if (ret) { ++ hns3_err(hw, "failed to set %s port link speed, ret = %d.", ++ hns3_get_media_type_name(hw->mac.media_type), ++ ret); ++ return ret; + } -- action = &flower->actions[flower->action_count++]; - action->type = TC_ACT_PEDIT; + return 0; +@@ -5327,7 +5317,7 @@ hns3_get_autoneg_fc_mode(struct hns3_hw *hw) + + /* + * Flow control auto-negotiation is not supported for fiber and +- * backpalne media type. ++ * backplane media type. + */ + case HNS3_MEDIA_TYPE_FIBER: + case HNS3_MEDIA_TYPE_BACKPLANE: +@@ -5579,15 +5569,15 @@ hns3_reinit_dev(struct hns3_adapter *hns) + return ret; + } - return 0; -@@ -1314,8 +1312,8 @@ nl_parse_act_gact(struct nlattr *options, struct tc_flower *flower) - struct nlattr *gact_attrs[ARRAY_SIZE(gact_policy)]; - const struct tc_gact *p; - struct nlattr *gact_parms; -- const struct tcf_t *tm; - struct tc_action *action; -+ struct tcf_t tm; +- ret = hns3_reset_all_tqps(hns); ++ ret = hns3_init_hardware(hns); + if (ret) { +- hns3_err(hw, "Failed to reset all queues: %d", ret); ++ hns3_err(hw, "Failed to init hardware: %d", ret); + return ret; + } - if (!nl_parse_nested(options, gact_policy, gact_attrs, - ARRAY_SIZE(gact_policy))) { -@@ -1335,8 +1333,9 @@ nl_parse_act_gact(struct nlattr *options, struct tc_flower *flower) - return EINVAL; - } +- ret = hns3_init_hardware(hns); ++ ret = hns3_reset_all_tqps(hns); + if (ret) { +- hns3_err(hw, "Failed to init hardware: %d", ret); ++ hns3_err(hw, "Failed to reset all queues: %d", ret); + return ret; + } -- tm = nl_attr_get_unspec(gact_attrs[TCA_GACT_TM], sizeof *tm); -- nl_parse_tcf(tm, flower); -+ memcpy(&tm, nl_attr_get_unspec(gact_attrs[TCA_GACT_TM], sizeof tm), -+ sizeof tm); -+ nl_parse_tcf(&tm, flower); +@@ -6191,7 +6181,7 @@ hns3_fec_get_internal(struct hns3_hw *hw, uint32_t *fec_capa) + } - return 0; - } -@@ -1357,9 +1356,9 @@ nl_parse_act_mirred(struct nlattr *options, struct tc_flower *flower) - struct nlattr *mirred_attrs[ARRAY_SIZE(mirred_policy)]; - const struct tc_mirred *m; - const struct nlattr *mirred_parms; -- const struct tcf_t *tm; - struct nlattr *mirred_tm; - struct tc_action *action; -+ struct tcf_t tm; + /* +- * FEC mode order defined in hns3 hardware is inconsistend with ++ * FEC mode order defined in hns3 hardware is inconsistent with + * that defined in the ethdev library. So the sequence needs + * to be converted. + */ +diff --git a/dpdk/drivers/net/hns3/hns3_ethdev.h b/dpdk/drivers/net/hns3/hns3_ethdev.h +index aa45b31261..134a33ee2f 100644 +--- a/dpdk/drivers/net/hns3/hns3_ethdev.h ++++ b/dpdk/drivers/net/hns3/hns3_ethdev.h +@@ -126,7 +126,7 @@ struct hns3_tc_info { + uint8_t tc_sch_mode; /* 0: sp; 1: dwrr */ + uint8_t pgid; + uint32_t bw_limit; +- uint8_t up_to_tc_map; /* user priority maping on the TC */ ++ uint8_t up_to_tc_map; /* user priority mapping on the TC */ + }; - if (!nl_parse_nested(options, mirred_policy, mirred_attrs, - ARRAY_SIZE(mirred_policy))) { -@@ -1387,8 +1386,8 @@ nl_parse_act_mirred(struct nlattr *options, struct tc_flower *flower) - action->type = TC_ACT_OUTPUT; + struct hns3_dcb_info { +@@ -502,8 +502,15 @@ struct hns3_hw { + struct hns3_tqp_stats tqp_stats; + /* Include Mac stats | Rx stats | Tx stats */ + struct hns3_mac_stats mac_stats; ++ uint32_t mac_stats_reg_num; + struct hns3_rx_missed_stats imissed_stats; + uint64_t oerror_stats; ++ /* ++ * The lock is used to protect statistics update in stats APIs and ++ * periodic task. ++ */ ++ rte_spinlock_t stats_lock; ++ + uint32_t fw_version; + uint16_t pf_vf_if_version; /* version of communication interface */ + +@@ -523,7 +530,6 @@ struct hns3_hw { + + /* The configuration info of RSS */ + struct hns3_rss_conf rss_info; +- bool rss_dis_flag; /* disable rss flag. true: disable, false: enable */ + uint16_t rss_ind_tbl_size; + uint16_t rss_key_size; + +@@ -571,12 +577,12 @@ struct hns3_hw { + /* + * vlan mode. + * value range: +- * HNS3_SW_SHIFT_AND_DISCARD_MODE/HNS3_HW_SHFIT_AND_DISCARD_MODE ++ * HNS3_SW_SHIFT_AND_DISCARD_MODE/HNS3_HW_SHIFT_AND_DISCARD_MODE + * + * - HNS3_SW_SHIFT_AND_DISCARD_MODE + * For some versions of hardware network engine, because of the + * hardware limitation, PMD needs to detect the PVID status +- * to work with haredware to implement PVID-related functions. ++ * to work with hardware to implement PVID-related functions. + * For example, driver need discard the stripped PVID tag to ensure + * the PVID will not report to mbuf and shift the inserted VLAN tag + * to avoid port based VLAN covering it. +@@ -724,7 +730,7 @@ enum hns3_mp_req_type { + HNS3_MP_REQ_MAX + }; - mirred_tm = mirred_attrs[TCA_MIRRED_TM]; -- tm = nl_attr_get_unspec(mirred_tm, sizeof *tm); -- nl_parse_tcf(tm, flower); -+ memcpy(&tm, nl_attr_get_unspec(mirred_tm, sizeof tm), sizeof tm); -+ nl_parse_tcf(&tm, flower); +-/* Pameters for IPC. */ ++/* Parameters for IPC. */ + struct hns3_mp_param { + enum hns3_mp_req_type type; + int port_id; +diff --git a/dpdk/drivers/net/hns3/hns3_ethdev_vf.c b/dpdk/drivers/net/hns3/hns3_ethdev_vf.c +index 805abd4543..0af4dcb324 100644 +--- a/dpdk/drivers/net/hns3/hns3_ethdev_vf.c ++++ b/dpdk/drivers/net/hns3/hns3_ethdev_vf.c +@@ -242,7 +242,7 @@ hns3vf_set_default_mac_addr(struct rte_eth_dev *dev, + if (ret == -EPERM) { + hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, + old_addr); +- hns3_warn(hw, "Has permanet mac addr(%s) for vf", ++ hns3_warn(hw, "Has permanent mac addr(%s) for vf", + mac_str); + } else { + hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, +@@ -318,7 +318,7 @@ hns3vf_set_promisc_mode(struct hns3_hw *hw, bool en_bc_pmc, + * 1. The promiscuous/allmulticast mode can be configured successfully + * only based on the trusted VF device. If based on the non trusted + * VF device, configuring promiscuous/allmulticast mode will fail. +- * The hns3 VF device can be confiruged as trusted device by hns3 PF ++ * The hns3 VF device can be configured as trusted device by hns3 PF + * kernel ethdev driver on the host by the following command: + * "ip link set vf turst on" + * 2. After the promiscuous mode is configured successfully, hns3 VF PMD +@@ -330,7 +330,7 @@ hns3vf_set_promisc_mode(struct hns3_hw *hw, bool en_bc_pmc, + * filter is still effective even in promiscuous mode. If upper + * applications don't call rte_eth_dev_vlan_filter API function to + * set vlan based on VF device, hns3 VF PMD will can't receive +- * the packets with vlan tag in promiscuoue mode. ++ * the packets with vlan tag in promiscuous mode. + */ + hns3_cmd_setup_basic_desc(&desc, HNS3_OPC_MBX_VF_TO_PF, false); + req->msg[0] = HNS3_MBX_SET_PROMISC_MODE; +@@ -496,7 +496,6 @@ hns3vf_dev_configure(struct rte_eth_dev *dev) + /* When RSS is not configured, redirect the packet queue 0 */ + if ((uint32_t)mq_mode & RTE_ETH_MQ_RX_RSS_FLAG) { + conf->rxmode.offloads |= RTE_ETH_RX_OFFLOAD_RSS_HASH; +- hw->rss_dis_flag = false; + rss_conf = conf->rx_adv_conf.rss_conf; + ret = hns3_dev_rss_hash_update(dev, &rss_conf); + if (ret) +@@ -780,6 +779,14 @@ hns3vf_get_push_lsc_cap(struct hns3_hw *hw) + + while (remain_ms > 0) { + rte_delay_ms(HNS3_POLL_RESPONE_MS); ++ /* ++ * The probe process may perform in interrupt thread context. ++ * For example, users attach a device in the secondary process. ++ * At the moment, the handling mailbox task will be blocked. So ++ * driver has to actively handle the HNS3_MBX_LINK_STAT_CHANGE ++ * mailbox from PF driver to get this capability. ++ */ ++ hns3_dev_handle_mbx_msg(hw); + if (__atomic_load_n(&vf->pf_push_lsc_cap, __ATOMIC_ACQUIRE) != + HNS3_PF_PUSH_LSC_CAP_UNKNOWN) + break; +@@ -1031,7 +1038,6 @@ hns3vf_get_configuration(struct hns3_hw *hw) + int ret; + + hw->mac.media_type = HNS3_MEDIA_TYPE_NONE; +- hw->rss_dis_flag = false; + + /* Get device capability */ + ret = hns3vf_get_capability(hw); +@@ -1385,10 +1391,12 @@ hns3vf_service_handler(void *param) + * Before querying the link status, check whether there is a reset + * pending, and if so, abandon the query. + */ +- if (!hns3vf_is_reset_pending(hns)) ++ if (!hns3vf_is_reset_pending(hns)) { + hns3vf_request_link_info(hw); +- else ++ hns3_update_hw_stats(hw); ++ } else { + hns3_warn(hw, "Cancel the query when reset is pending"); ++ } + + rte_eal_alarm_set(HNS3VF_SERVICE_INTERVAL, hns3vf_service_handler, + eth_dev); +@@ -1558,17 +1566,10 @@ hns3vf_init_vf(struct rte_eth_dev *eth_dev) + goto err_get_config; + } - return 0; +- ret = hns3_tqp_stats_init(hw); ++ ret = hns3_stats_init(hw); + if (ret) + goto err_get_config; + +- /* Hardware statistics of imissed registers cleared. */ +- ret = hns3_update_imissed_stats(hw, true); +- if (ret) { +- hns3_err(hw, "clear imissed stats failed, ret = %d", ret); +- goto err_set_tc_queue; +- } +- + ret = hns3vf_set_tc_queue_mapping(hns, hw->tqps_num, hw->tqps_num); + if (ret) { + PMD_INIT_LOG(ERR, "failed to set tc info, ret = %d.", ret); +@@ -1596,7 +1597,7 @@ hns3vf_init_vf(struct rte_eth_dev *eth_dev) + return 0; + + err_set_tc_queue: +- hns3_tqp_stats_uninit(hw); ++ hns3_stats_uninit(hw); + + err_get_config: + hns3vf_disable_irq0(hw); +@@ -1627,7 +1628,7 @@ hns3vf_uninit_vf(struct rte_eth_dev *eth_dev) + (void)hns3vf_set_alive(hw, false); + (void)hns3vf_set_promisc_mode(hw, false, false, false); + hns3_flow_uninit(eth_dev); +- hns3_tqp_stats_uninit(hw); ++ hns3_stats_uninit(hw); + hns3vf_disable_irq0(hw); + rte_intr_disable(pci_dev->intr_handle); + hns3_intr_unregister(pci_dev->intr_handle, hns3vf_interrupt_handler, +@@ -1925,6 +1926,7 @@ hns3vf_is_reset_pending(struct hns3_adapter *hns) + static int + hns3vf_wait_hardware_ready(struct hns3_adapter *hns) + { ++#define HNS3_WAIT_PF_RESET_READY_TIME 5 + struct hns3_hw *hw = &hns->hw; + struct hns3_wait_data *wait_data = hw->reset.wait_data; + struct timeval tv; +@@ -1945,12 +1947,14 @@ hns3vf_wait_hardware_ready(struct hns3_adapter *hns) + return 0; + + wait_data->check_completion = NULL; +- wait_data->interval = 1 * MSEC_PER_SEC * USEC_PER_MSEC; ++ wait_data->interval = HNS3_WAIT_PF_RESET_READY_TIME * ++ MSEC_PER_SEC * USEC_PER_MSEC; + wait_data->count = 1; + wait_data->result = HNS3_WAIT_REQUEST; + rte_eal_alarm_set(wait_data->interval, hns3_wait_callback, + wait_data); +- hns3_warn(hw, "hardware is ready, delay 1 sec for PF reset complete"); ++ hns3_warn(hw, "hardware is ready, delay %d sec for PF reset complete", ++ HNS3_WAIT_PF_RESET_READY_TIME); + return -EAGAIN; + } else if (wait_data->result == HNS3_WAIT_TIMEOUT) { + hns3_clock_gettime(&tv); +@@ -2472,7 +2476,6 @@ hns3vf_dev_uninit(struct rte_eth_dev *eth_dev) + PMD_INIT_FUNC_TRACE(); + + if (rte_eal_process_type() != RTE_PROC_PRIMARY) { +- __atomic_fetch_sub(&hw->secondary_cnt, 1, __ATOMIC_RELAXED); + hns3_mp_uninit(eth_dev); + return 0; + } +diff --git a/dpdk/drivers/net/hns3/hns3_fdir.c b/dpdk/drivers/net/hns3/hns3_fdir.c +index d043f5786d..2426264138 100644 +--- a/dpdk/drivers/net/hns3/hns3_fdir.c ++++ b/dpdk/drivers/net/hns3/hns3_fdir.c +@@ -631,7 +631,7 @@ static bool hns3_fd_convert_tuple(struct hns3_hw *hw, + break; + default: + hns3_warn(hw, "not support tuple of (%u)", tuple); +- break; ++ return false; + } + return true; } -@@ -1487,7 +1486,9 @@ nl_parse_act_ct(struct nlattr *options, struct tc_flower *flower) - if (ipv4_max) { - ovs_be32 addr = nl_attr_get_be32(ipv4_max); +diff --git a/dpdk/drivers/net/hns3/hns3_fdir.h b/dpdk/drivers/net/hns3/hns3_fdir.h +index f9efff3b52..07b393393d 100644 +--- a/dpdk/drivers/net/hns3/hns3_fdir.h ++++ b/dpdk/drivers/net/hns3/hns3_fdir.h +@@ -139,7 +139,7 @@ struct hns3_fdir_rule { + uint32_t flags; + uint32_t fd_id; /* APP marked unique value for this rule. */ + uint8_t action; +- /* VF id, avaiblable when flags with HNS3_RULE_FLAG_VF_ID. */ ++ /* VF id, available when flags with HNS3_RULE_FLAG_VF_ID. */ + uint8_t vf_id; + /* + * equal 0 when action is drop. +diff --git a/dpdk/drivers/net/hns3/hns3_flow.c b/dpdk/drivers/net/hns3/hns3_flow.c +index 9f2f9cb6cd..24caf8e870 100644 +--- a/dpdk/drivers/net/hns3/hns3_flow.c ++++ b/dpdk/drivers/net/hns3/hns3_flow.c +@@ -10,15 +10,6 @@ + #include "hns3_logs.h" + #include "hns3_flow.h" + +-/* Default default keys */ +-static uint8_t hns3_hash_key[] = { +- 0x6D, 0x5A, 0x56, 0xDA, 0x25, 0x5B, 0x0E, 0xC2, +- 0x41, 0x67, 0x25, 0x3D, 0x43, 0xA3, 0x8F, 0xB0, +- 0xD0, 0xCA, 0x2B, 0xCB, 0xAE, 0x7B, 0x30, 0xB4, +- 0x77, 0xCB, 0x2D, 0xA3, 0x80, 0x30, 0xF2, 0x0C, +- 0x6A, 0x42, 0xB7, 0x3B, 0xBE, 0xAC, 0x01, 0xFA +-}; +- + static const uint8_t full_mask[VNI_OR_TNI_LEN] = { 0xFF, 0xFF, 0xFF }; + static const uint8_t zero_mask[VNI_OR_TNI_LEN] = { 0x00, 0x00, 0x00 }; -- action->ct.range.ipv4.max = addr; -+ if (action->ct.range.ipv4.min != addr) { -+ action->ct.range.ipv4.max = addr; -+ } - } - } else if (ipv6_min) { - action->ct.range.ip_family = AF_INET6; -@@ -1496,7 +1497,9 @@ nl_parse_act_ct(struct nlattr *options, struct tc_flower *flower) - if (ipv6_max) { - struct in6_addr addr = nl_attr_get_in6_addr(ipv6_max); +@@ -338,7 +329,7 @@ hns3_handle_action_queue_region(struct rte_eth_dev *dev, + * + * @param actions[in] + * @param rule[out] +- * NIC specfilc actions derived from the actions. ++ * NIC specific actions derived from the actions. + * @param error[out] + */ + static int +@@ -369,7 +360,7 @@ hns3_handle_actions(struct rte_eth_dev *dev, + * Queue region is implemented by FDIR + RSS in hns3 hardware, + * the FDIR's action is one queue region (start_queue_id and + * queue_num), then RSS spread packets to the queue region by +- * RSS algorigthm. ++ * RSS algorithm. + */ + case RTE_FLOW_ACTION_TYPE_RSS: + ret = hns3_handle_action_queue_region(dev, actions, +@@ -940,7 +931,7 @@ hns3_parse_nvgre(const struct rte_flow_item *item, struct hns3_fdir_rule *rule, + if (nvgre_mask->protocol || nvgre_mask->c_k_s_rsvd0_ver) + return rte_flow_error_set(error, EINVAL, + RTE_FLOW_ERROR_TYPE_ITEM_MASK, item, +- "Ver/protocal is not supported in NVGRE"); ++ "Ver/protocol is not supported in NVGRE"); + + /* TNI must be totally masked or not. */ + if (memcmp(nvgre_mask->tni, full_mask, VNI_OR_TNI_LEN) && +@@ -985,7 +976,7 @@ hns3_parse_geneve(const struct rte_flow_item *item, struct hns3_fdir_rule *rule, + if (geneve_mask->ver_opt_len_o_c_rsvd0 || geneve_mask->protocol) + return rte_flow_error_set(error, EINVAL, + RTE_FLOW_ERROR_TYPE_ITEM_MASK, item, +- "Ver/protocal is not supported in GENEVE"); ++ "Ver/protocol is not supported in GENEVE"); + /* VNI must be totally masked or not. */ + if (memcmp(geneve_mask->vni, full_mask, VNI_OR_TNI_LEN) && + memcmp(geneve_mask->vni, zero_mask, VNI_OR_TNI_LEN)) +@@ -1238,6 +1229,7 @@ static bool + hns3_action_rss_same(const struct rte_flow_action_rss *comp, + const struct rte_flow_action_rss *with) + { ++ bool rss_key_is_same; + bool func_is_same; + + /* +@@ -1251,13 +1243,19 @@ hns3_action_rss_same(const struct rte_flow_action_rss *comp, + if (comp->func == RTE_ETH_HASH_FUNCTION_MAX) + func_is_same = false; + else +- func_is_same = with->func ? (comp->func == with->func) : true; ++ func_is_same = (with->func != RTE_ETH_HASH_FUNCTION_DEFAULT) ? ++ (comp->func == with->func) : true; + +- return (func_is_same && ++ if (with->key_len == 0 || with->key == NULL) ++ rss_key_is_same = 1; ++ else ++ rss_key_is_same = comp->key_len == with->key_len && ++ !memcmp(comp->key, with->key, with->key_len); ++ ++ return (func_is_same && rss_key_is_same && + comp->types == (with->types & HNS3_ETH_RSS_SUPPORT) && +- comp->level == with->level && comp->key_len == with->key_len && ++ comp->level == with->level && + comp->queue_num == with->queue_num && +- !memcmp(comp->key, with->key, with->key_len) && + !memcmp(comp->queue, with->queue, + sizeof(*with->queue) * with->queue_num)); + } +@@ -1309,7 +1307,7 @@ hns3_rss_input_tuple_supported(struct hns3_hw *hw, + } -- action->ct.range.ipv6.max = addr; -+ if (!ipv6_addr_equals(&action->ct.range.ipv6.min, &addr)) { -+ action->ct.range.ipv6.max = addr; -+ } - } - } + /* +- * This function is used to parse rss action validatation. ++ * This function is used to parse rss action validation. + */ + static int + hns3_parse_rss_filter(struct rte_eth_dev *dev, +@@ -1391,15 +1389,10 @@ hns3_disable_rss(struct hns3_hw *hw) + { + int ret; -@@ -1504,6 +1507,10 @@ nl_parse_act_ct(struct nlattr *options, struct tc_flower *flower) - action->ct.range.port.min = nl_attr_get_be16(port_min); - if (port_max) { - action->ct.range.port.max = nl_attr_get_be16(port_max); -+ if (action->ct.range.port.min == -+ action->ct.range.port.max) { -+ action->ct.range.port.max = 0; -+ } - } - } - } -@@ -1702,6 +1709,9 @@ static const struct nl_policy stats_policy[] = { - [TCA_STATS_BASIC] = { .type = NL_A_UNSPEC, - .min_len = sizeof(struct gnet_stats_basic), - .optional = false, }, -+ [TCA_STATS_BASIC_HW] = { .type = NL_A_UNSPEC, -+ .min_len = sizeof(struct gnet_stats_basic), -+ .optional = true, }, - }; +- /* Redirected the redirection table to queue 0 */ +- ret = hns3_rss_reset_indir_table(hw); ++ ret = hns3_set_rss_tuple_by_rss_hf(hw, 0); + if (ret) + return ret; +- /* Disable RSS */ +- hw->rss_info.conf.types = 0; +- hw->rss_dis_flag = true; +- + return 0; + } + +@@ -1445,7 +1438,6 @@ hns3_parse_rss_algorithm(struct hns3_hw *hw, enum rte_eth_hash_function *func, static int -@@ -1714,8 +1724,9 @@ nl_parse_single_action(struct nlattr *action, struct tc_flower *flower, - const char *act_kind; - struct nlattr *action_attrs[ARRAY_SIZE(act_policy)]; - struct nlattr *stats_attrs[ARRAY_SIZE(stats_policy)]; -- struct ovs_flow_stats *stats = &flower->stats; -- const struct gnet_stats_basic *bs; -+ struct ovs_flow_stats *stats_sw = &flower->stats_sw; -+ struct ovs_flow_stats *stats_hw = &flower->stats_hw; -+ struct gnet_stats_basic bs_all, bs_hw, bs_sw; - int err = 0; + hns3_hw_rss_hash_set(struct hns3_hw *hw, struct rte_flow_action_rss *rss_config) + { +- struct hns3_rss_tuple_cfg *tuple; + int ret; - if (!nl_parse_nested(action, act_policy, action_attrs, -@@ -1771,10 +1782,30 @@ nl_parse_single_action(struct nlattr *action, struct tc_flower *flower, - return EPROTO; - } + hns3_parse_rss_key(hw, rss_config); +@@ -1461,8 +1453,7 @@ hns3_hw_rss_hash_set(struct hns3_hw *hw, struct rte_flow_action_rss *rss_config) -- bs = nl_attr_get_unspec(stats_attrs[TCA_STATS_BASIC], sizeof *bs); -- if (bs->packets) { -- put_32aligned_u64(&stats->n_packets, bs->packets); -- put_32aligned_u64(&stats->n_bytes, bs->bytes); -+ memcpy(&bs_all, -+ nl_attr_get_unspec(stats_attrs[TCA_STATS_BASIC], sizeof bs_all), -+ sizeof bs_all); -+ if (stats_attrs[TCA_STATS_BASIC_HW]) { -+ memcpy(&bs_hw, nl_attr_get_unspec(stats_attrs[TCA_STATS_BASIC_HW], -+ sizeof bs_hw), -+ sizeof bs_hw); -+ -+ bs_sw.packets = bs_all.packets - bs_hw.packets; -+ bs_sw.bytes = bs_all.bytes - bs_hw.bytes; -+ } else { -+ bs_sw.packets = bs_all.packets; -+ bs_sw.bytes = bs_all.bytes; -+ } -+ -+ if (bs_sw.packets > get_32aligned_u64(&stats_sw->n_packets)) { -+ put_32aligned_u64(&stats_sw->n_packets, bs_sw.packets); -+ put_32aligned_u64(&stats_sw->n_bytes, bs_sw.bytes); -+ } -+ -+ if (stats_attrs[TCA_STATS_BASIC_HW] -+ && bs_hw.packets > get_32aligned_u64(&stats_hw->n_packets)) { -+ put_32aligned_u64(&stats_hw->n_packets, bs_hw.packets); -+ put_32aligned_u64(&stats_hw->n_bytes, bs_hw.bytes); - } + hw->rss_info.conf.func = rss_config->func; - return 0; -@@ -2399,14 +2430,14 @@ nl_msg_put_act_flags(struct ofpbuf *request) { - * first_word_mask/last_word_mask - the mask to use for the first/last read - * (as we read entire words). */ - static void --calc_offsets(struct tc_flower *flower, struct flower_key_to_pedit *m, -+calc_offsets(struct tc_action *action, struct flower_key_to_pedit *m, - int *cur_offset, int *cnt, ovs_be32 *last_word_mask, - ovs_be32 *first_word_mask, ovs_be32 **mask, ovs_be32 **data) - { - int start_offset, max_offset, total_size; - int diff, right_zero_bits, left_zero_bits; -- char *rewrite_key = (void *) &flower->rewrite.key; -- char *rewrite_mask = (void *) &flower->rewrite.mask; -+ char *rewrite_key = (void *) &action->rewrite.key; -+ char *rewrite_mask = (void *) &action->rewrite.mask; +- tuple = &hw->rss_info.rss_tuple_sets; +- ret = hns3_set_rss_tuple_by_rss_hf(hw, tuple, rss_config->types); ++ ret = hns3_set_rss_tuple_by_rss_hf(hw, rss_config->types); + if (ret) + hns3_err(hw, "Update RSS tuples by rss hf failed %d", ret); - max_offset = m->offset + m->size; - start_offset = ROUND_DOWN(m->offset, 4); -@@ -2473,7 +2504,8 @@ csum_update_flag(struct tc_flower *flower, +@@ -1682,7 +1673,7 @@ hns3_flow_args_check(const struct rte_flow_attr *attr, + /* + * Check if the flow rule is supported by hns3. +- * It only checkes the format. Don't guarantee the rule can be programmed into ++ * It only checks the format. Don't guarantee the rule can be programmed into + * the HW. Because there can be no enough room for the rule. + */ static int - nl_msg_put_flower_rewrite_pedits(struct ofpbuf *request, -- struct tc_flower *flower) -+ struct tc_flower *flower, -+ struct tc_action *action) +diff --git a/dpdk/drivers/net/hns3/hns3_mbx.c b/dpdk/drivers/net/hns3/hns3_mbx.c +index b3563d4694..02028dcd9c 100644 +--- a/dpdk/drivers/net/hns3/hns3_mbx.c ++++ b/dpdk/drivers/net/hns3/hns3_mbx.c +@@ -78,14 +78,14 @@ hns3_get_mbx_resp(struct hns3_hw *hw, uint16_t code, uint16_t subcode, + mbx_time_limit = (uint32_t)hns->mbx_time_limit_ms * US_PER_MS; + while (wait_time < mbx_time_limit) { + if (__atomic_load_n(&hw->reset.disable_cmd, __ATOMIC_RELAXED)) { +- hns3_err(hw, "Don't wait for mbx respone because of " ++ hns3_err(hw, "Don't wait for mbx response because of " + "disable_cmd"); + return -EBUSY; + } + + if (is_reset_pending(hns)) { + hw->mbx_resp.req_msg_data = 0; +- hns3_err(hw, "Don't wait for mbx respone because of " ++ hns3_err(hw, "Don't wait for mbx response because of " + "reset pending"); + return -EIO; + } +diff --git a/dpdk/drivers/net/hns3/hns3_mbx.h b/dpdk/drivers/net/hns3/hns3_mbx.h +index d637bd2b23..0172a2e288 100644 +--- a/dpdk/drivers/net/hns3/hns3_mbx.h ++++ b/dpdk/drivers/net/hns3/hns3_mbx.h +@@ -22,7 +22,7 @@ enum HNS3_MBX_OPCODE { + HNS3_MBX_GET_RETA, /* (VF -> PF) get RETA */ + HNS3_MBX_GET_RSS_KEY, /* (VF -> PF) get RSS key */ + HNS3_MBX_GET_MAC_ADDR, /* (VF -> PF) get MAC addr */ +- HNS3_MBX_PF_VF_RESP, /* (PF -> VF) generate respone to VF */ ++ HNS3_MBX_PF_VF_RESP, /* (PF -> VF) generate response to VF */ + HNS3_MBX_GET_BDNUM, /* (VF -> PF) get BD num */ + HNS3_MBX_GET_BUFSIZE, /* (VF -> PF) get buffer size */ + HNS3_MBX_GET_STREAMID, /* (VF -> PF) get stream id */ +diff --git a/dpdk/drivers/net/hns3/hns3_mp.c b/dpdk/drivers/net/hns3/hns3_mp.c +index 999b407f7d..e74ddea195 100644 +--- a/dpdk/drivers/net/hns3/hns3_mp.c ++++ b/dpdk/drivers/net/hns3/hns3_mp.c +@@ -74,7 +74,6 @@ mp_secondary_handle(const struct rte_mp_msg *mp_msg, const void *peer) + struct hns3_mp_param *res = (struct hns3_mp_param *)mp_res.param; + const struct hns3_mp_param *param = + (const struct hns3_mp_param *)mp_msg->param; +- eth_tx_prep_t prep = NULL; + struct rte_eth_dev *dev; + int ret; + +@@ -98,14 +97,12 @@ mp_secondary_handle(const struct rte_mp_msg *mp_msg, const void *peer) + case HNS3_MP_REQ_START_TX: + PMD_INIT_LOG(INFO, "port %u starting Tx datapath", + dev->data->port_id); +- dev->tx_pkt_burst = hns3_get_tx_function(dev, &prep); +- dev->tx_pkt_prepare = prep; ++ hns3_start_tx_datapath(dev); + break; + case HNS3_MP_REQ_STOP_TX: + PMD_INIT_LOG(INFO, "port %u stopping Tx datapath", + dev->data->port_id); +- dev->tx_pkt_burst = hns3_dummy_rxtx_burst; +- dev->tx_pkt_prepare = NULL; ++ hns3_stop_tx_datapath(dev); + break; + default: + rte_errno = EINVAL; +diff --git a/dpdk/drivers/net/hns3/hns3_ptp.c b/dpdk/drivers/net/hns3/hns3_ptp.c +index 9a829d7011..0b0061bba5 100644 +--- a/dpdk/drivers/net/hns3/hns3_ptp.c ++++ b/dpdk/drivers/net/hns3/hns3_ptp.c +@@ -81,7 +81,7 @@ hns3_timesync_configure(struct hns3_adapter *hns, bool en) + struct hns3_hw *hw = &hns->hw; + struct hns3_pf *pf = &hns->pf; + struct hns3_cmd_desc desc; +- int val; ++ uint32_t val; + int ret; + + hns3_cmd_setup_basic_desc(&desc, HNS3_OPC_CFG_PTP_MODE, false); +@@ -125,6 +125,7 @@ hns3_timesync_enable(struct rte_eth_dev *dev) + + if (pf->ptp_enable) + return 0; ++ hns3_warn(hw, "note: please ensure Rx/Tx burst mode is simple or common when enabling PTP!"); + + rte_spinlock_lock(&hw->lock); + ret = hns3_timesync_configure(hns, true); +diff --git a/dpdk/drivers/net/hns3/hns3_rss.c b/dpdk/drivers/net/hns3/hns3_rss.c +index 3a4b699ae2..980fbe74e8 100644 +--- a/dpdk/drivers/net/hns3/hns3_rss.c ++++ b/dpdk/drivers/net/hns3/hns3_rss.c +@@ -9,10 +9,8 @@ + #include "hns3_ethdev.h" + #include "hns3_logs.h" + +-/* +- * The hash key used for rss initialization. +- */ +-static const uint8_t hns3_hash_key[] = { ++/* Default hash keys */ ++const uint8_t hns3_hash_key[] = { + 0x6D, 0x5A, 0x56, 0xDA, 0x25, 0x5B, 0x0E, 0xC2, + 0x41, 0x67, 0x25, 0x3D, 0x43, 0xA3, 0x8F, 0xB0, + 0xD0, 0xCA, 0x2B, 0xCB, 0xAE, 0x7B, 0x30, 0xB4, +@@ -152,10 +150,6 @@ static const struct { + BIT_ULL(HNS3_RSS_FIELD_IPV4_TCP_EN_IP_D) | + BIT_ULL(HNS3_RSS_FIELD_IPV4_TCP_EN_TCP_S) | + BIT_ULL(HNS3_RSS_FIELD_IPV4_TCP_EN_TCP_D) }, +- { RTE_ETH_RSS_NONFRAG_IPV4_TCP, BIT_ULL(HNS3_RSS_FIELD_IPV4_TCP_EN_IP_S) | +- BIT_ULL(HNS3_RSS_FIELD_IPV4_TCP_EN_IP_D) | +- BIT_ULL(HNS3_RSS_FIELD_IPV4_TCP_EN_TCP_S) | +- BIT_ULL(HNS3_RSS_FIELD_IPV4_TCP_EN_TCP_D) }, + { RTE_ETH_RSS_NONFRAG_IPV4_UDP, BIT_ULL(HNS3_RSS_FIELD_IPV4_UDP_EN_IP_S) | + BIT_ULL(HNS3_RSS_FIELD_IPV4_UDP_EN_IP_D) | + BIT_ULL(HNS3_RSS_FIELD_IPV4_UDP_EN_UDP_S) | +@@ -241,31 +235,6 @@ hns3_rss_set_algo_key(struct hns3_hw *hw, const uint8_t *key) + return 0; + } + +-/* +- * Used to configure the tuple selection for RSS hash input. +- */ +-static int +-hns3_rss_set_input_tuple(struct hns3_hw *hw) +-{ +- struct hns3_rss_conf *rss_config = &hw->rss_info; +- struct hns3_rss_input_tuple_cmd *req; +- struct hns3_cmd_desc desc_tuple; +- int ret; +- +- hns3_cmd_setup_basic_desc(&desc_tuple, HNS3_OPC_RSS_INPUT_TUPLE, false); +- +- req = (struct hns3_rss_input_tuple_cmd *)desc_tuple.data; +- +- req->tuple_field = +- rte_cpu_to_le_64(rss_config->rss_tuple_sets.rss_tuple_fields); +- +- ret = hns3_cmd_send(hw, &desc_tuple, 1); +- if (ret) +- hns3_err(hw, "Configure RSS input tuple mode failed %d", ret); +- +- return ret; +-} +- + /* + * rss_indirection_table command function, opcode:0x0D07. + * Used to configure the indirection table of rss. +@@ -339,8 +308,7 @@ hns3_rss_reset_indir_table(struct hns3_hw *hw) + } + + int +-hns3_set_rss_tuple_by_rss_hf(struct hns3_hw *hw, +- struct hns3_rss_tuple_cfg *tuple, uint64_t rss_hf) ++hns3_set_rss_tuple_by_rss_hf(struct hns3_hw *hw, uint64_t rss_hf) { - struct { - struct tc_pedit sel; -@@ -2497,12 +2529,12 @@ nl_msg_put_flower_rewrite_pedits(struct ofpbuf *request, - continue; - } + struct hns3_rss_input_tuple_cmd *req; + struct hns3_cmd_desc desc; +@@ -385,7 +353,8 @@ hns3_set_rss_tuple_by_rss_hf(struct hns3_hw *hw, + return ret; + } -- calc_offsets(flower, m, &cur_offset, &cnt, &last_word_mask, -+ calc_offsets(action, m, &cur_offset, &cnt, &last_word_mask, - &first_word_mask, &mask, &data); +- tuple->rss_tuple_fields = rte_le_to_cpu_64(req->tuple_field); ++ /* Update supported flow types when set tuple success */ ++ hw->rss_info.conf.types = rss_hf; - for (j = 0; j < cnt; j++, mask++, data++, cur_offset += 4) { -- ovs_be32 mask_word = *mask; -- ovs_be32 data_word = *data; -+ ovs_be32 mask_word = get_unaligned_be32(mask); -+ ovs_be32 data_word = get_unaligned_be32(data); + return 0; + } +@@ -403,55 +372,36 @@ int + hns3_dev_rss_hash_update(struct rte_eth_dev *dev, + struct rte_eth_rss_conf *rss_conf) + { +- struct hns3_adapter *hns = dev->data->dev_private; +- struct hns3_hw *hw = &hns->hw; +- struct hns3_rss_tuple_cfg *tuple = &hw->rss_info.rss_tuple_sets; +- struct hns3_rss_conf *rss_cfg = &hw->rss_info; ++ struct hns3_hw *hw = HNS3_DEV_PRIVATE_TO_HW(dev->data->dev_private); ++ uint64_t rss_hf_bk = hw->rss_info.conf.types; + uint8_t key_len = rss_conf->rss_key_len; + uint64_t rss_hf = rss_conf->rss_hf; + uint8_t *key = rss_conf->rss_key; + int ret; + +- if (hw->rss_dis_flag) ++ if (key && key_len != HNS3_RSS_KEY_SIZE) { ++ hns3_err(hw, "the hash key len(%u) is invalid, must be %u", ++ key_len, HNS3_RSS_KEY_SIZE); + return -EINVAL; ++ } + + rte_spinlock_lock(&hw->lock); +- ret = hns3_set_rss_tuple_by_rss_hf(hw, tuple, rss_hf); ++ ret = hns3_set_rss_tuple_by_rss_hf(hw, rss_hf); + if (ret) +- goto conf_err; +- +- if (rss_cfg->conf.types && rss_hf == 0) { +- /* Disable RSS, reset indirection table by local variable */ +- ret = hns3_rss_reset_indir_table(hw); +- if (ret) +- goto conf_err; +- } else if (rss_hf && rss_cfg->conf.types == 0) { +- /* Enable RSS, restore indirection table by hw's config */ +- ret = hns3_set_rss_indir_table(hw, rss_cfg->rss_indirection_tbl, +- hw->rss_ind_tbl_size); +- if (ret) +- goto conf_err; +- } +- +- /* Update supported flow types when set tuple success */ +- rss_cfg->conf.types = rss_hf; ++ goto set_tuple_fail; + + if (key) { +- if (key_len != HNS3_RSS_KEY_SIZE) { +- hns3_err(hw, "The hash key len(%u) is invalid", +- key_len); +- ret = -EINVAL; +- goto conf_err; +- } + ret = hns3_rss_set_algo_key(hw, key); + if (ret) +- goto conf_err; ++ goto set_algo_key_fail; + } + rte_spinlock_unlock(&hw->lock); - if (j == 0) { - mask_word &= first_word_mask; -@@ -2556,6 +2588,29 @@ nl_msg_put_flower_acts_release(struct ofpbuf *request, uint16_t act_index) - nl_msg_end_nested(request, act_offset); + return 0; + +-conf_err: ++set_algo_key_fail: ++ (void)hns3_set_rss_tuple_by_rss_hf(hw, rss_hf_bk); ++set_tuple_fail: + rte_spinlock_unlock(&hw->lock); + return ret; + } +@@ -582,33 +532,59 @@ hns3_dev_rss_reta_query(struct rte_eth_dev *dev, + return 0; } -+/* Aggregates all previous successive pedit actions csum_update_flags -+ * to flower->csum_update_flags. Only append one csum action to the -+ * last pedit action. */ +-/* +- * Used to configure the tc_size and tc_offset. +- */ +static void -+nl_msg_put_csum_act(struct ofpbuf *request, struct tc_flower *flower, -+ uint16_t *act_index) ++hns3_set_rss_tc_mode_entry(struct hns3_hw *hw, uint8_t *tc_valid, ++ uint16_t *tc_size, uint16_t *tc_offset, ++ uint8_t tc_num) +{ -+ size_t act_offset; -+ -+ /* No pedit actions or processed already. */ -+ if (!flower->csum_update_flags) { -+ return; -+ } -+ -+ act_offset = nl_msg_start_nested(request, (*act_index)++); -+ nl_msg_put_act_csum(request, flower->csum_update_flags); -+ nl_msg_put_act_flags(request); -+ nl_msg_end_nested(request, act_offset); -+ -+ /* Clear it. So we can have another series of pedit actions. */ -+ flower->csum_update_flags = 0; ++ struct hns3_adapter *hns = HNS3_DEV_HW_TO_ADAPTER(hw); ++ uint16_t rss_size = hw->alloc_rss_size; ++ uint16_t roundup_size; ++ uint16_t i; ++ ++ roundup_size = roundup_pow_of_two(rss_size); ++ roundup_size = ilog2(roundup_size); ++ ++ for (i = 0; i < tc_num; i++) { ++ if (hns->is_vf) { ++ /* ++ * For packets with VLAN priorities destined for the VF, ++ * hardware still assign Rx queue based on the Up-to-TC ++ * mapping PF configured. But VF has only one TC. If ++ * other TC don't enable, it causes that the priority ++ * packets that aren't destined for TC0 aren't received ++ * by RSS hash but is destined for queue 0. So driver ++ * has to enable the unused TC by using TC0 queue ++ * mapping configuration. ++ */ ++ tc_valid[i] = (hw->hw_tc_map & BIT(i)) ? ++ !!(hw->hw_tc_map & BIT(i)) : 1; ++ tc_size[i] = roundup_size; ++ tc_offset[i] = (hw->hw_tc_map & BIT(i)) ? ++ rss_size * i : 0; ++ } else { ++ tc_valid[i] = !!(hw->hw_tc_map & BIT(i)); ++ tc_size[i] = tc_valid[i] ? roundup_size : 0; ++ tc_offset[i] = tc_valid[i] ? rss_size * i : 0; ++ } ++ } +} + static int - nl_msg_put_flower_acts(struct ofpbuf *request, struct tc_flower *flower) + hns3_set_rss_tc_mode(struct hns3_hw *hw) { -@@ -2572,20 +2627,22 @@ nl_msg_put_flower_acts(struct ofpbuf *request, struct tc_flower *flower) +- uint16_t rss_size = hw->alloc_rss_size; + struct hns3_rss_tc_mode_cmd *req; + uint16_t tc_offset[HNS3_MAX_TC_NUM]; + uint8_t tc_valid[HNS3_MAX_TC_NUM]; + uint16_t tc_size[HNS3_MAX_TC_NUM]; + struct hns3_cmd_desc desc; +- uint16_t roundup_size; + uint16_t i; + int ret; + +- req = (struct hns3_rss_tc_mode_cmd *)desc.data; +- +- roundup_size = roundup_pow_of_two(rss_size); +- roundup_size = ilog2(roundup_size); +- +- for (i = 0; i < HNS3_MAX_TC_NUM; i++) { +- tc_valid[i] = !!(hw->hw_tc_map & BIT(i)); +- tc_size[i] = roundup_size; +- tc_offset[i] = rss_size * i; +- } ++ hns3_set_rss_tc_mode_entry(hw, tc_valid, tc_size, ++ tc_offset, HNS3_MAX_TC_NUM); + ++ req = (struct hns3_rss_tc_mode_cmd *)desc.data; + hns3_cmd_setup_basic_desc(&desc, HNS3_OPC_RSS_TC_MODE, false); + for (i = 0; i < HNS3_MAX_TC_NUM; i++) { + uint16_t mode = 0; +@@ -675,7 +651,8 @@ hns3_config_rss(struct hns3_adapter *hns) + struct hns3_hw *hw = &hns->hw; + struct hns3_rss_conf *rss_cfg = &hw->rss_info; + uint8_t *hash_key = rss_cfg->key; +- int ret, ret1; ++ uint64_t rss_hf; ++ int ret; + + enum rte_eth_rx_mq_mode mq_mode = hw->data->dev_conf.rxmode.mq_mode; + +@@ -691,51 +668,30 @@ hns3_config_rss(struct hns3_adapter *hns) + break; + } - action = flower->actions; - for (i = 0; i < flower->action_count; i++, action++) { -+ if (action->type != TC_ACT_PEDIT) { -+ nl_msg_put_csum_act(request, flower, &act_index); -+ } - switch (action->type) { - case TC_ACT_PEDIT: { - act_offset = nl_msg_start_nested(request, act_index++); -- error = nl_msg_put_flower_rewrite_pedits(request, flower); -+ error = nl_msg_put_flower_rewrite_pedits(request, flower, -+ action); - if (error) { - return error; - } - nl_msg_end_nested(request, act_offset); +- /* When RSS is off, redirect the packet queue 0 */ +- if (((uint32_t)mq_mode & RTE_ETH_MQ_RX_RSS_FLAG) == 0) +- hns3_rss_uninit(hns); +- + /* Configure RSS hash algorithm and hash key offset */ + ret = hns3_rss_set_algo_key(hw, hash_key); + if (ret) + return ret; + +- /* Configure the tuple selection for RSS hash input */ +- ret = hns3_rss_set_input_tuple(hw); ++ ret = hns3_set_rss_indir_table(hw, rss_cfg->rss_indirection_tbl, ++ hw->rss_ind_tbl_size); + if (ret) + return ret; + +- /* +- * When RSS is off, it doesn't need to configure rss redirection table +- * to hardware. +- */ +- if (((uint32_t)mq_mode & RTE_ETH_MQ_RX_RSS_FLAG)) { +- ret = hns3_set_rss_indir_table(hw, rss_cfg->rss_indirection_tbl, +- hw->rss_ind_tbl_size); +- if (ret) +- goto rss_tuple_uninit; +- } +- + ret = hns3_set_rss_tc_mode(hw); + if (ret) +- goto rss_indir_table_uninit; +- +- return ret; +- +-rss_indir_table_uninit: +- if (((uint32_t)mq_mode & RTE_ETH_MQ_RX_RSS_FLAG)) { +- ret1 = hns3_rss_reset_indir_table(hw); +- if (ret1 != 0) +- return ret; +- } +- +-rss_tuple_uninit: +- hns3_rss_tuple_uninit(hw); ++ return ret; + +- /* Disable RSS */ +- hw->rss_info.conf.types = 0; ++ /* ++ * When muli-queue RSS mode flag is not set or unsupported tuples are ++ * set, disable all tuples. ++ */ ++ rss_hf = hw->rss_info.conf.types; ++ if (!((uint32_t)mq_mode & RTE_ETH_MQ_RX_RSS_FLAG) || ++ !(rss_hf & HNS3_ETH_RSS_SUPPORT)) ++ rss_hf = 0; + +- return ret; ++ return hns3_set_rss_tuple_by_rss_hf(hw, rss_hf); + } -- if (flower->csum_update_flags) { -- act_offset = nl_msg_start_nested(request, act_index++); -- nl_msg_put_act_csum(request, flower->csum_update_flags); -- nl_msg_put_act_flags(request); -- nl_msg_end_nested(request, act_offset); -+ if (i == flower->action_count - 1) { -+ /* If this is the last action check csum calc again. */ -+ nl_msg_put_csum_act(request, flower, &act_index); - } - } - break; -@@ -2914,13 +2971,13 @@ nl_msg_put_flower_options(struct ofpbuf *request, struct tc_flower *flower) - FLOWER_PUT_MASKED_VALUE(icmp_code, TCA_FLOWER_KEY_ICMPV6_CODE); - FLOWER_PUT_MASKED_VALUE(icmp_type, TCA_FLOWER_KEY_ICMPV6_TYPE); - } + /* +diff --git a/dpdk/drivers/net/hns3/hns3_rss.h b/dpdk/drivers/net/hns3/hns3_rss.h +index 6f153a1b7b..56627cbd4c 100644 +--- a/dpdk/drivers/net/hns3/hns3_rss.h ++++ b/dpdk/drivers/net/hns3/hns3_rss.h +@@ -41,9 +41,8 @@ struct hns3_rss_tuple_cfg { + struct hns3_rss_conf { + /* RSS parameters :algorithm, flow_types, key, queue */ + struct rte_flow_action_rss conf; +- uint8_t hash_algo; /* hash function type definited by hardware */ ++ uint8_t hash_algo; /* hash function type defined by hardware */ + uint8_t key[HNS3_RSS_KEY_SIZE]; /* Hash key */ +- struct hns3_rss_tuple_cfg rss_tuple_sets; + uint16_t rss_indirection_tbl[HNS3_RSS_IND_TBL_SIZE_MAX]; + uint16_t queue[HNS3_RSS_QUEUES_BUFFER_NUM]; /* Queues indices to use */ + bool valid; /* check if RSS rule is valid */ +@@ -89,6 +88,8 @@ static inline uint32_t roundup_pow_of_two(uint32_t x) + return 1UL << fls(x - 1); + } + ++extern const uint8_t hns3_hash_key[]; ++ + struct hns3_adapter; + + int hns3_dev_rss_hash_update(struct rte_eth_dev *dev, +@@ -107,9 +108,7 @@ int hns3_set_rss_indir_table(struct hns3_hw *hw, uint16_t *indir, + int hns3_rss_reset_indir_table(struct hns3_hw *hw); + int hns3_config_rss(struct hns3_adapter *hns); + void hns3_rss_uninit(struct hns3_adapter *hns); +-int hns3_set_rss_tuple_by_rss_hf(struct hns3_hw *hw, +- struct hns3_rss_tuple_cfg *tuple, +- uint64_t rss_hf); ++int hns3_set_rss_tuple_by_rss_hf(struct hns3_hw *hw, uint64_t rss_hf); + int hns3_rss_set_algo_key(struct hns3_hw *hw, const uint8_t *key); + int hns3_restore_rss_filter(struct rte_eth_dev *dev); + +diff --git a/dpdk/drivers/net/hns3/hns3_rxtx.c b/dpdk/drivers/net/hns3/hns3_rxtx.c +index f365daadf8..403f811a51 100644 +--- a/dpdk/drivers/net/hns3/hns3_rxtx.c ++++ b/dpdk/drivers/net/hns3/hns3_rxtx.c +@@ -776,7 +776,7 @@ hns3vf_reset_all_tqps(struct hns3_hw *hw) + int ret; + int i; + +- memset(msg_data, 0, sizeof(uint16_t)); ++ memset(msg_data, 0, sizeof(msg_data)); + ret = hns3_send_mbx_msg(hw, HNS3_MBX_QUEUE_RESET, 0, msg_data, + sizeof(msg_data), true, &reset_status, + sizeof(reset_status)); +@@ -1763,7 +1763,8 @@ hns3_rxq_conf_runtime_check(struct hns3_hw *hw, uint16_t buf_size, + return -EINVAL; + } + +- if (pkt_burst == hns3_recv_pkts_vec) { ++ if (pkt_burst == hns3_recv_pkts_vec || ++ pkt_burst == hns3_recv_pkts_vec_sve) { + min_vec_bds = HNS3_DEFAULT_RXQ_REARM_THRESH + + HNS3_DEFAULT_RX_BURST; + if (nb_desc < min_vec_bds || +@@ -1903,7 +1904,7 @@ hns3_rx_queue_setup(struct rte_eth_dev *dev, uint16_t idx, uint16_t nb_desc, + * For hns3 VF device, whether it needs to process PVID depends + * on the configuration of PF kernel mode netdevice driver. And the + * related PF configuration is delivered through the mailbox and finally +- * reflectd in port_base_vlan_cfg. ++ * reflected in port_base_vlan_cfg. + */ + if (hns->is_vf || hw->vlan_mode == HNS3_SW_SHIFT_AND_DISCARD_MODE) + rxq->pvid_sw_discard_en = hw->port_base_vlan_cfg.state == +@@ -2388,14 +2389,14 @@ hns3_rx_alloc_buffer(struct hns3_rx_queue *rxq) + return rte_mbuf_raw_alloc(rxq->mb_pool); + } + +-static inline void ++static void + hns3_rx_ptp_timestamp_handle(struct hns3_rx_queue *rxq, struct rte_mbuf *mbuf, +- volatile struct hns3_desc *rxd) ++ uint64_t timestamp) + { + struct hns3_pf *pf = HNS3_DEV_PRIVATE_TO_PF(rxq->hns); +- uint64_t timestamp = rte_le_to_cpu_64(rxd->timestamp); + +- mbuf->ol_flags |= RTE_MBUF_F_RX_IEEE1588_PTP | RTE_MBUF_F_RX_IEEE1588_TMST; ++ mbuf->ol_flags |= RTE_MBUF_F_RX_IEEE1588_PTP | ++ RTE_MBUF_F_RX_IEEE1588_TMST; + if (hns3_timestamp_rx_dynflag > 0) { + *RTE_MBUF_DYNFIELD(mbuf, hns3_timestamp_dynfield_offset, + rte_mbuf_timestamp_t *) = timestamp; +@@ -2469,7 +2470,8 @@ hns3_recv_pkts_simple(void *rx_queue, + rxe->mbuf = nmb; + + if (unlikely(bd_base_info & BIT(HNS3_RXD_TS_VLD_B))) +- hns3_rx_ptp_timestamp_handle(rxq, rxm, rxdp); ++ hns3_rx_ptp_timestamp_handle(rxq, rxm, ++ rte_le_to_cpu_64(rxdp->timestamp)); + + dma_addr = rte_mbuf_data_iova_default(nmb); + rxdp->addr = rte_cpu_to_le_64(dma_addr); +@@ -2540,6 +2542,7 @@ hns3_recv_scattered_pkts(void *rx_queue, + struct rte_mbuf *rxm; + struct rte_eth_dev *dev; + uint32_t bd_base_info; ++ uint64_t timestamp; + uint32_t l234_info; + uint32_t gro_size; + uint32_t ol_info; +@@ -2649,6 +2652,9 @@ hns3_recv_scattered_pkts(void *rx_queue, + rxm = rxe->mbuf; + rxe->mbuf = nmb; + ++ if (unlikely(bd_base_info & BIT(HNS3_RXD_TS_VLD_B))) ++ timestamp = rte_le_to_cpu_64(rxdp->timestamp); ++ + dma_addr = rte_cpu_to_le_64(rte_mbuf_data_iova_default(nmb)); + rxdp->rx.bd_base_info = 0; + rxdp->addr = dma_addr; +@@ -2671,7 +2677,7 @@ hns3_recv_scattered_pkts(void *rx_queue, + } + + if (unlikely(bd_base_info & BIT(HNS3_RXD_TS_VLD_B))) +- hns3_rx_ptp_timestamp_handle(rxq, first_seg, rxdp); ++ hns3_rx_ptp_timestamp_handle(rxq, first_seg, timestamp); + + /* + * The last buffer of the received packet. packet len from +@@ -3043,7 +3049,7 @@ hns3_tx_queue_setup(struct rte_eth_dev *dev, uint16_t idx, uint16_t nb_desc, + * For hns3 VF device, whether it needs to process PVID depends + * on the configuration of PF kernel mode netdev driver. And the + * related PF configuration is delivered through the mailbox and finally +- * reflectd in port_base_vlan_cfg. ++ * reflected in port_base_vlan_cfg. + */ + if (hns->is_vf || hw->vlan_mode == HNS3_SW_SHIFT_AND_DISCARD_MODE) + txq->pvid_sw_shift_en = hw->port_base_vlan_cfg.state == +@@ -3208,7 +3214,7 @@ hns3_fill_first_desc(struct hns3_tx_queue *txq, struct hns3_desc *desc, + * in Tx direction based on hns3 network engine. So when the number of + * VLANs in the packets represented by rxm plus the number of VLAN + * offload by hardware such as PVID etc, exceeds two, the packets will +- * be discarded or the original VLAN of the packets will be overwitted ++ * be discarded or the original VLAN of the packets will be overwritten + * by hardware. When the PF PVID is enabled by calling the API function + * named rte_eth_dev_set_vlan_pvid or the VF PVID is enabled by the hns3 + * PF kernel ether driver, the outer VLAN tag will always be the PVID. +@@ -3393,7 +3399,7 @@ hns3_parse_inner_params(struct rte_mbuf *m, uint32_t *ol_type_vlan_len_msec, + /* + * The inner l2 length of mbuf is the sum of outer l4 length, + * tunneling header length and inner l2 length for a tunnel +- * packect. But in hns3 tx descriptor, the tunneling header ++ * packet. But in hns3 tx descriptor, the tunneling header + * length is contained in the field of outer L4 length. + * Therefore, driver need to calculate the outer L4 length and + * inner L2 length. +@@ -3409,7 +3415,7 @@ hns3_parse_inner_params(struct rte_mbuf *m, uint32_t *ol_type_vlan_len_msec, + tmp_outer |= hns3_gen_field_val(HNS3_TXD_TUNTYPE_M, + HNS3_TXD_TUNTYPE_S, HNS3_TUN_NVGRE); + /* +- * For NVGRE tunnel packect, the outer L4 is empty. So only ++ * For NVGRE tunnel packet, the outer L4 is empty. So only + * fill the NVGRE header length to the outer L4 field. + */ + tmp_outer |= hns3_gen_field_val(HNS3_TXD_L4LEN_M, +@@ -3452,7 +3458,7 @@ hns3_parse_tunneling_params(struct hns3_tx_queue *txq, struct rte_mbuf *m, + * mbuf, but for hns3 descriptor, it is contained in the outer L4. So, + * there is a need that switching between them. To avoid multiple + * calculations, the length of the L2 header include the outer and +- * inner, will be filled during the parsing of tunnel packects. ++ * inner, will be filled during the parsing of tunnel packets. + */ + if (!(ol_flags & RTE_MBUF_F_TX_TUNNEL_MASK)) { + /* +@@ -3632,7 +3638,7 @@ hns3_outer_ipv4_cksum_prepared(struct rte_mbuf *m, uint64_t ol_flags, + if (ol_flags & RTE_MBUF_F_TX_OUTER_UDP_CKSUM) { + struct rte_udp_hdr *udp_hdr; + /* +- * If OUTER_UDP_CKSUM is support, HW can caclulate the pseudo ++ * If OUTER_UDP_CKSUM is support, HW can calculate the pseudo + * header for TSO packets + */ + if (ol_flags & RTE_MBUF_F_TX_TCP_SEG) +@@ -3657,7 +3663,7 @@ hns3_outer_ipv6_cksum_prepared(struct rte_mbuf *m, uint64_t ol_flags, + if (ol_flags & RTE_MBUF_F_TX_OUTER_UDP_CKSUM) { + struct rte_udp_hdr *udp_hdr; + /* +- * If OUTER_UDP_CKSUM is support, HW can caclulate the pseudo ++ * If OUTER_UDP_CKSUM is support, HW can calculate the pseudo + * header for TSO packets + */ + if (ol_flags & RTE_MBUF_F_TX_TCP_SEG) +@@ -4044,7 +4050,7 @@ static inline void + hns3_tx_setup_4bd(struct hns3_desc *txdp, struct rte_mbuf **pkts) + { + #define PER_LOOP_NUM 4 +- const uint16_t bd_flag = BIT(HNS3_TXD_VLD_B) | BIT(HNS3_TXD_FE_B); ++ uint16_t bd_flag = BIT(HNS3_TXD_VLD_B) | BIT(HNS3_TXD_FE_B); + uint64_t dma_addr; + uint32_t i; + +@@ -4055,6 +4061,8 @@ hns3_tx_setup_4bd(struct hns3_desc *txdp, struct rte_mbuf **pkts) + txdp->tx.paylen_fd_dop_ol4cs = 0; + txdp->tx.type_cs_vlan_tso_len = 0; + txdp->tx.ol_type_vlan_len_msec = 0; ++ if (unlikely((*pkts)->ol_flags & RTE_MBUF_F_TX_IEEE1588_TMST)) ++ bd_flag |= BIT(HNS3_TXD_TSYN_B); + txdp->tx.tp_fe_sc_vld_ra_ri = rte_cpu_to_le_16(bd_flag); + } + } +@@ -4062,7 +4070,7 @@ hns3_tx_setup_4bd(struct hns3_desc *txdp, struct rte_mbuf **pkts) + static inline void + hns3_tx_setup_1bd(struct hns3_desc *txdp, struct rte_mbuf **pkts) + { +- const uint16_t bd_flag = BIT(HNS3_TXD_VLD_B) | BIT(HNS3_TXD_FE_B); ++ uint16_t bd_flag = BIT(HNS3_TXD_VLD_B) | BIT(HNS3_TXD_FE_B); + uint64_t dma_addr; + + dma_addr = rte_mbuf_data_iova(*pkts); +@@ -4071,6 +4079,8 @@ hns3_tx_setup_1bd(struct hns3_desc *txdp, struct rte_mbuf **pkts) + txdp->tx.paylen_fd_dop_ol4cs = 0; + txdp->tx.type_cs_vlan_tso_len = 0; + txdp->tx.ol_type_vlan_len_msec = 0; ++ if (unlikely((*pkts)->ol_flags & RTE_MBUF_F_TX_IEEE1588_TMST)) ++ bd_flag |= BIT(HNS3_TXD_TSYN_B); + txdp->tx.tp_fe_sc_vld_ra_ri = rte_cpu_to_le_16(bd_flag); + } + +@@ -4312,10 +4322,6 @@ hns3_tx_check_simple_support(struct rte_eth_dev *dev) + { + uint64_t offloads = dev->data->dev_conf.txmode.offloads; + +- struct hns3_hw *hw = HNS3_DEV_PRIVATE_TO_HW(dev->data->dev_private); +- if (hns3_dev_get_support(hw, PTP)) +- return false; - -- FLOWER_PUT_MASKED_VALUE(ct_state, TCA_FLOWER_KEY_CT_STATE); -- FLOWER_PUT_MASKED_VALUE(ct_zone, TCA_FLOWER_KEY_CT_ZONE); -- FLOWER_PUT_MASKED_VALUE(ct_mark, TCA_FLOWER_KEY_CT_MARK); -- FLOWER_PUT_MASKED_VALUE(ct_label, TCA_FLOWER_KEY_CT_LABELS); - } + return (offloads == (offloads & RTE_ETH_TX_OFFLOAD_MBUF_FAST_FREE)); + } -+ FLOWER_PUT_MASKED_VALUE(ct_state, TCA_FLOWER_KEY_CT_STATE); -+ FLOWER_PUT_MASKED_VALUE(ct_zone, TCA_FLOWER_KEY_CT_ZONE); -+ FLOWER_PUT_MASKED_VALUE(ct_mark, TCA_FLOWER_KEY_CT_MARK); -+ FLOWER_PUT_MASKED_VALUE(ct_label, TCA_FLOWER_KEY_CT_LABELS); -+ - if (host_eth_type == ETH_P_IP) { - FLOWER_PUT_MASKED_VALUE(ipv4.ipv4_src, TCA_FLOWER_KEY_IPV4_SRC); - FLOWER_PUT_MASKED_VALUE(ipv4.ipv4_dst, TCA_FLOWER_KEY_IPV4_DST); -@@ -2993,12 +3050,79 @@ nl_msg_put_flower_options(struct ofpbuf *request, struct tc_flower *flower) - return 0; +@@ -4408,7 +4414,23 @@ hns3_trace_rxtx_function(struct rte_eth_dev *dev) + rx_mode.info, tx_mode.info); } +-void hns3_set_rxtx_function(struct rte_eth_dev *eth_dev) +static void -+log_tc_flower_match(const char *msg, -+ const struct tc_flower *a, -+ const struct tc_flower *b) ++hns3_eth_dev_fp_ops_config(const struct rte_eth_dev *dev) +{ -+ uint8_t key_a[sizeof(struct tc_flower_key)]; -+ uint8_t key_b[sizeof(struct tc_flower_key)]; -+ struct ds s = DS_EMPTY_INITIALIZER; ++ struct rte_eth_fp_ops *fpo = rte_eth_fp_ops; ++ uint16_t port_id = dev->data->port_id; ++ ++ fpo[port_id].rx_pkt_burst = dev->rx_pkt_burst; ++ fpo[port_id].tx_pkt_burst = dev->tx_pkt_burst; ++ fpo[port_id].tx_pkt_prepare = dev->tx_pkt_prepare; ++ fpo[port_id].rx_descriptor_status = dev->rx_descriptor_status; ++ fpo[port_id].tx_descriptor_status = dev->tx_descriptor_status; ++ fpo[port_id].rxq.data = dev->data->rx_queues; ++ fpo[port_id].txq.data = dev->data->tx_queues; ++} + -+ for (int i = 0; i < sizeof a->key; i++) { -+ uint8_t mask_a = ((uint8_t *) &a->mask)[i]; -+ uint8_t mask_b = ((uint8_t *) &b->mask)[i]; ++void ++hns3_set_rxtx_function(struct rte_eth_dev *eth_dev) + { + struct hns3_hw *hw = HNS3_DEV_PRIVATE_TO_HW(eth_dev->data->dev_private); + struct hns3_adapter *hns = eth_dev->data->dev_private; +@@ -4429,6 +4451,8 @@ void hns3_set_rxtx_function(struct rte_eth_dev *eth_dev) + eth_dev->tx_pkt_burst = hns3_dummy_rxtx_burst; + eth_dev->tx_pkt_prepare = NULL; + } + -+ key_a[i] = ((uint8_t *) &a->key)[i] & mask_a; -+ key_b[i] = ((uint8_t *) &b->key)[i] & mask_b; -+ } -+ ds_put_cstr(&s, "\nExpected Mask:\n"); -+ ds_put_hex(&s, &a->mask, sizeof a->mask); -+ ds_put_cstr(&s, "\nReceived Mask:\n"); -+ ds_put_hex(&s, &b->mask, sizeof b->mask); -+ ds_put_cstr(&s, "\nExpected Key:\n"); -+ ds_put_hex(&s, &a->key, sizeof a->key); -+ ds_put_cstr(&s, "\nReceived Key:\n"); -+ ds_put_hex(&s, &b->key, sizeof b->key); -+ ds_put_cstr(&s, "\nExpected Masked Key:\n"); -+ ds_put_hex(&s, key_a, sizeof key_a); -+ ds_put_cstr(&s, "\nReceived Masked Key:\n"); -+ ds_put_hex(&s, key_b, sizeof key_b); ++ hns3_eth_dev_fp_ops_config(eth_dev); + } + + void +@@ -4591,7 +4615,7 @@ hns3_dev_tx_queue_stop(struct rte_eth_dev *dev, uint16_t tx_queue_id) + static int + hns3_tx_done_cleanup_full(struct hns3_tx_queue *txq, uint32_t free_cnt) + { +- uint16_t round_free_cnt; ++ uint16_t round_cnt; + uint32_t idx; + + if (free_cnt == 0 || free_cnt > txq->nb_tx_desc) +@@ -4600,13 +4624,13 @@ hns3_tx_done_cleanup_full(struct hns3_tx_queue *txq, uint32_t free_cnt) + if (txq->tx_rs_thresh == 0) + return 0; + +- round_free_cnt = roundup(free_cnt, txq->tx_rs_thresh); +- for (idx = 0; idx < round_free_cnt; idx += txq->tx_rs_thresh) { ++ round_cnt = rounddown(free_cnt, txq->tx_rs_thresh); ++ for (idx = 0; idx < round_cnt; idx += txq->tx_rs_thresh) { + if (hns3_tx_free_useless_buffer(txq) != 0) + break; + } + +- return RTE_MIN(idx, free_cnt); ++ return idx; + } + + int +@@ -4729,6 +4753,11 @@ hns3_stop_tx_datapath(struct rte_eth_dev *dev) + { + dev->tx_pkt_burst = hns3_dummy_rxtx_burst; + dev->tx_pkt_prepare = NULL; ++ hns3_eth_dev_fp_ops_config(dev); + -+ if (a->action_count != b->action_count) { -+ /* If action count is not equal, we print all actions to see which -+ * ones are missing. */ -+ const struct tc_action *action; -+ int i; ++ if (rte_eal_process_type() == RTE_PROC_SECONDARY) ++ return; + -+ ds_put_cstr(&s, "\nExpected Actions:\n"); -+ for (i = 0, action = a->actions; i < a->action_count; i++, action++) { -+ ds_put_cstr(&s, " - "); -+ ds_put_hex(&s, action, sizeof *action); -+ ds_put_cstr(&s, "\n"); -+ } -+ ds_put_cstr(&s, "Received Actions:\n"); -+ for (i = 0, action = b->actions; i < b->action_count; i++, action++) { -+ ds_put_cstr(&s, " - "); -+ ds_put_hex(&s, action, sizeof *action); -+ ds_put_cstr(&s, "\n"); -+ } -+ } else { -+ /* Only dump the delta in actions. */ -+ const struct tc_action *action_a = a->actions; -+ const struct tc_action *action_b = b->actions; + rte_wmb(); + /* Disable tx datapath on secondary process. */ + hns3_mp_req_stop_tx(dev); +@@ -4743,5 +4772,10 @@ hns3_start_tx_datapath(struct rte_eth_dev *dev) + + dev->tx_pkt_burst = hns3_get_tx_function(dev, &prep); + dev->tx_pkt_prepare = prep; ++ hns3_eth_dev_fp_ops_config(dev); + -+ for (int i = 0; i < a->action_count; i++, action_a++, action_b++) { -+ if (memcmp(action_a, action_b, sizeof *action_a)) { -+ ds_put_format(&s, -+ "\nAction %d mismatch:\n - Expected Action: ", -+ i); -+ ds_put_hex(&s, action_a, sizeof *action_a); -+ ds_put_cstr(&s, "\n - Received Action: "); -+ ds_put_hex(&s, action_b, sizeof *action_b); -+ } -+ } -+ } -+ VLOG_DBG_RL(&error_rl, "%s%s", msg, ds_cstr(&s)); -+ ds_destroy(&s); -+} ++ if (rte_eal_process_type() == RTE_PROC_SECONDARY) ++ return; + - static bool - cmp_tc_flower_match_action(const struct tc_flower *a, - const struct tc_flower *b) + hns3_mp_req_start_tx(dev); + } +diff --git a/dpdk/drivers/net/hns3/hns3_rxtx.h b/dpdk/drivers/net/hns3/hns3_rxtx.h +index 5423568cd0..0e412d07b3 100644 +--- a/dpdk/drivers/net/hns3/hns3_rxtx.h ++++ b/dpdk/drivers/net/hns3/hns3_rxtx.h +@@ -349,7 +349,7 @@ struct hns3_rx_queue { + * The following fields are not accessed in the I/O path, so they are + * placed at the end. + */ +- void *io_base; ++ void *io_base __rte_cache_aligned; + struct hns3_adapter *hns; + uint64_t rx_ring_phys_addr; /* RX ring DMA address */ + const struct rte_memzone *mz; +@@ -523,7 +523,7 @@ struct hns3_tx_queue { + * The following fields are not accessed in the I/O path, so they are + * placed at the end. + */ +- void *io_base; ++ void *io_base __rte_cache_aligned; + struct hns3_adapter *hns; + uint64_t tx_ring_phys_addr; /* TX ring DMA address */ + const struct rte_memzone *mz; +@@ -611,7 +611,7 @@ hns3_handle_bdinfo(struct hns3_rx_queue *rxq, struct rte_mbuf *rxm, + + /* + * If packet len bigger than mtu when recv with no-scattered algorithm, +- * the first n bd will without FE bit, we need process this sisution. ++ * the first n bd will without FE bit, we need process this situation. + * Note: we don't need add statistic counter because latest BD which + * with FE bit will mark HNS3_RXD_L2E_B bit. + */ +diff --git a/dpdk/drivers/net/hns3/hns3_rxtx_vec.c b/dpdk/drivers/net/hns3/hns3_rxtx_vec.c +index 455110361a..73f0ab6bc8 100644 +--- a/dpdk/drivers/net/hns3/hns3_rxtx_vec.c ++++ b/dpdk/drivers/net/hns3/hns3_rxtx_vec.c +@@ -17,15 +17,17 @@ int + hns3_tx_check_vec_support(struct rte_eth_dev *dev) { - if (memcmp(&a->mask, &b->mask, sizeof a->mask)) { -- VLOG_DBG_RL(&error_rl, "tc flower compare failed mask compare"); -+ log_tc_flower_match("tc flower compare failed mask compare:", a, b); - return false; - } + struct rte_eth_txmode *txmode = &dev->data->dev_conf.txmode; +- +- struct hns3_hw *hw = HNS3_DEV_PRIVATE_TO_HW(dev->data->dev_private); +- if (hns3_dev_get_support(hw, PTP)) +- return -ENOTSUP; ++ struct hns3_adapter *hns = dev->data->dev_private; ++ struct hns3_pf *pf = &hns->pf; -@@ -3011,8 +3135,8 @@ cmp_tc_flower_match_action(const struct tc_flower *a, - uint8_t key_b = ((uint8_t *)&b->key)[i] & mask; + /* Only support RTE_ETH_TX_OFFLOAD_MBUF_FAST_FREE */ + if (txmode->offloads != RTE_ETH_TX_OFFLOAD_MBUF_FAST_FREE) + return -ENOTSUP; - if (key_a != key_b) { -- VLOG_DBG_RL(&error_rl, "tc flower compare failed key compare at " -- "%d", i); -+ log_tc_flower_match("tc flower compare failed masked key compare:", -+ a, b); - return false; - } - } -@@ -3022,14 +3146,15 @@ cmp_tc_flower_match_action(const struct tc_flower *a, - const struct tc_action *action_b = b->actions; ++ /* Vec is not supported when PTP enabled */ ++ if (pf->ptp_enable) ++ return -ENOTSUP; ++ + return 0; + } - if (a->action_count != b->action_count) { -- VLOG_DBG_RL(&error_rl, "tc flower compare failed action length check"); -+ log_tc_flower_match("tc flower compare failed action length check", -+ a, b); - return false; - } +@@ -232,10 +234,8 @@ hns3_rx_check_vec_support(struct rte_eth_dev *dev) + struct rte_eth_rxmode *rxmode = &dev->data->dev_conf.rxmode; + uint64_t offloads_mask = RTE_ETH_RX_OFFLOAD_TCP_LRO | + RTE_ETH_RX_OFFLOAD_VLAN; +- +- struct hns3_hw *hw = HNS3_DEV_PRIVATE_TO_HW(dev->data->dev_private); +- if (hns3_dev_get_support(hw, PTP)) +- return -ENOTSUP; ++ struct hns3_adapter *hns = dev->data->dev_private; ++ struct hns3_pf *pf = &hns->pf; + + if (dev->data->scattered_rx) + return -ENOTSUP; +@@ -249,5 +249,9 @@ hns3_rx_check_vec_support(struct rte_eth_dev *dev) + if (hns3_rxq_iterate(dev, hns3_rxq_vec_check, NULL) != 0) + return -ENOTSUP; + ++ /* Vec is not supported when PTP enabled */ ++ if (pf->ptp_enable) ++ return -ENOTSUP; ++ + return 0; + } +diff --git a/dpdk/drivers/net/hns3/hns3_stats.c b/dpdk/drivers/net/hns3/hns3_stats.c +index 0fe853d626..1b0464f3f7 100644 +--- a/dpdk/drivers/net/hns3/hns3_stats.c ++++ b/dpdk/drivers/net/hns3/hns3_stats.c +@@ -307,24 +307,21 @@ static const struct hns3_xstats_name_offset hns3_imissed_stats_strings[] = { + + static void hns3_tqp_stats_clear(struct hns3_hw *hw); + +-/* +- * Query all the MAC statistics data of Network ICL command ,opcode id: 0x0034. +- * This command is used before send 'query_mac_stat command', the descriptor +- * number of 'query_mac_stat command' must match with reg_num in this command. +- * @praram hw +- * Pointer to structure hns3_hw. +- * @return +- * 0 on success. +- */ + static int +-hns3_update_mac_stats(struct hns3_hw *hw, const uint32_t desc_num) ++hns3_update_mac_stats(struct hns3_hw *hw) + { ++#define HNS3_MAC_STATS_REG_NUM_PER_DESC 4 ++ + uint64_t *data = (uint64_t *)(&hw->mac_stats); + struct hns3_cmd_desc *desc; ++ uint32_t stats_iterms; + uint64_t *desc_data; +- uint16_t i, k, n; ++ uint32_t desc_num; ++ uint16_t i; + int ret; + ++ /* The first desc has a 64-bit header, so need to consider it. */ ++ desc_num = hw->mac_stats_reg_num / HNS3_MAC_STATS_REG_NUM_PER_DESC + 1; + desc = rte_malloc("hns3_mac_desc", + desc_num * sizeof(struct hns3_cmd_desc), 0); + if (desc == NULL) { +@@ -340,65 +337,71 @@ hns3_update_mac_stats(struct hns3_hw *hw, const uint32_t desc_num) + return ret; + } - for (int i = 0; i < a->action_count; i++, action_a++, action_b++) { - if (memcmp(action_a, action_b, sizeof *action_a)) { -- VLOG_DBG_RL(&error_rl, "tc flower compare failed action compare " -- "for %d", i); -+ log_tc_flower_match("tc flower compare failed action compare", -+ a, b); - return false; - } - } -diff --git a/lib/tc.h b/lib/tc.h -index a147ca461d..d6cdddd169 100644 ---- a/lib/tc.h -+++ b/lib/tc.h -@@ -256,11 +256,23 @@ struct tc_action { - bool force; - bool commit; - } ct; +- for (i = 0; i < desc_num; i++) { +- /* For special opcode 0034, only the first desc has the head */ +- if (i == 0) { +- desc_data = (uint64_t *)(&desc[i].data[0]); +- n = HNS3_RD_FIRST_STATS_NUM; +- } else { +- desc_data = (uint64_t *)(&desc[i]); +- n = HNS3_RD_OTHER_STATS_NUM; +- } +- +- for (k = 0; k < n; k++) { +- *data += rte_le_to_cpu_64(*desc_data); +- data++; +- desc_data++; +- } ++ stats_iterms = RTE_MIN(sizeof(hw->mac_stats) / sizeof(uint64_t), ++ hw->mac_stats_reg_num); ++ desc_data = (uint64_t *)(&desc[0].data[0]); ++ for (i = 0; i < stats_iterms; i++) { ++ /* ++ * Data memory is continuous and only the first descriptor has a ++ * header in this command. ++ */ ++ *data += rte_le_to_cpu_64(*desc_data); ++ data++; ++ desc_data++; + } + rte_free(desc); + + return 0; + } + +-/* +- * Query Mac stat reg num command ,opcode id: 0x0033. +- * This command is used before send 'query_mac_stat command', the descriptor +- * number of 'query_mac_stat command' must match with reg_num in this command. +- * @praram rte_stats +- * Pointer to structure rte_eth_stats. +- * @return +- * 0 on success. +- */ + static int +-hns3_mac_query_reg_num(struct rte_eth_dev *dev, uint32_t *desc_num) ++hns3_mac_query_reg_num(struct hns3_hw *hw, uint32_t *reg_num) + { +- struct hns3_adapter *hns = dev->data->dev_private; +- struct hns3_hw *hw = &hns->hw; ++#define HNS3_MAC_STATS_RSV_REG_NUM_ON_HIP08_B 3 + struct hns3_cmd_desc desc; +- uint32_t *desc_data; +- uint32_t reg_num; + int ret; + + hns3_cmd_setup_basic_desc(&desc, HNS3_OPC_QUERY_MAC_REG_NUM, true); + ret = hns3_cmd_send(hw, &desc, 1); +- if (ret) ++ if (ret) { ++ hns3_err(hw, "failed to query MAC statistic reg number, ret = %d", ++ ret); + return ret; ++ } + +- /* +- * The num of MAC statistics registers that are provided by IMP in this +- * version. +- */ +- desc_data = (uint32_t *)(&desc.data[0]); +- reg_num = rte_le_to_cpu_32(*desc_data); ++ /* The number of MAC statistics registers are provided by firmware. */ ++ *reg_num = rte_le_to_cpu_32(desc.data[0]); ++ if (*reg_num == 0) { ++ hns3_err(hw, "MAC statistic reg number is invalid!"); ++ return -ENODATA; ++ } + + /* +- * The descriptor number of 'query_additional_mac_stat command' is +- * '1 + (reg_num-3)/4 + ((reg_num-3)%4 !=0)'; +- * This value is 83 in this version ++ * If driver doesn't request the firmware to report more MAC statistics ++ * iterms and the total number of MAC statistics registers by using new ++ * method, firmware will only reports the number of valid statistics ++ * registers. However, structure hns3_mac_stats in driver contains valid ++ * and reserved statistics iterms. In this case, the total register ++ * number must be added to three reserved statistics registers. + */ +- *desc_num = 1 + ((reg_num - 3) >> 2) + +- (uint32_t)(((reg_num - 3) & 0x3) ? 1 : 0); ++ *reg_num += HNS3_MAC_STATS_RSV_REG_NUM_ON_HIP08_B; ++ ++ return 0; ++} + -+ struct { -+ struct tc_flower_key key; -+ struct tc_flower_key mask; -+ } rewrite; - }; ++int ++hns3_query_mac_stats_reg_num(struct hns3_hw *hw) ++{ ++ uint32_t mac_stats_reg_num = 0; ++ int ret; ++ ++ ret = hns3_mac_query_reg_num(hw, &mac_stats_reg_num); ++ if (ret) ++ return ret; ++ ++ hw->mac_stats_reg_num = mac_stats_reg_num; ++ if (hw->mac_stats_reg_num > sizeof(hw->mac_stats) / sizeof(uint64_t)) ++ hns3_warn(hw, "MAC stats reg number from firmware is greater than stats iterms in driver."); - enum tc_action_type type; - }; + return 0; + } +@@ -408,15 +411,8 @@ hns3_query_update_mac_stats(struct rte_eth_dev *dev) + { + struct hns3_adapter *hns = dev->data->dev_private; + struct hns3_hw *hw = &hns->hw; +- uint32_t desc_num; +- int ret; + +- ret = hns3_mac_query_reg_num(dev, &desc_num); +- if (ret == 0) +- ret = hns3_update_mac_stats(hw, desc_num); +- else +- hns3_err(hw, "Query mac reg num fail : %d", ret); +- return ret; ++ return hns3_update_mac_stats(hw); + } -+/* assert that if we overflow with a masked write of uint32_t to the last byte -+ * of action.rewrite we overflow inside struct tc_action. -+ * shouldn't happen unless someone moves rewrite to the end of action */ -+BUILD_ASSERT_DECL(offsetof(struct tc_action, rewrite) -+ + MEMBER_SIZEOF(struct tc_action, rewrite) -+ + sizeof(uint32_t) - 2 < sizeof(struct tc_action)); + static int +@@ -544,7 +540,7 @@ hns3_update_port_tx_ssu_drop_stats(struct hns3_hw *hw) + return 0; + } + +-int ++static int + hns3_update_imissed_stats(struct hns3_hw *hw, bool is_clear) + { + struct hns3_adapter *hns = HNS3_DEV_HW_TO_ADAPTER(hw); +@@ -588,6 +584,28 @@ hns3_update_oerror_stats(struct hns3_hw *hw, bool is_clear) + return 0; + } + ++static void ++hns3_rcb_rx_ring_stats_get(struct hns3_rx_queue *rxq, ++ struct hns3_tqp_stats *stats) ++{ ++ uint32_t cnt; + - enum tc_offloaded_state { - TC_OFFLOADED_STATE_UNDEFINED, - TC_OFFLOADED_STATE_IN_HW, -@@ -330,15 +342,10 @@ struct tc_flower { - int action_count; - struct tc_action actions[TCA_ACT_MAX_NUM]; ++ cnt = hns3_read_dev(rxq, HNS3_RING_RX_PKTNUM_RECORD_REG); ++ stats->rcb_rx_ring_pktnum_rcd += cnt; ++ stats->rcb_rx_ring_pktnum[rxq->queue_id] += cnt; ++} ++ ++static void ++hns3_rcb_tx_ring_stats_get(struct hns3_tx_queue *txq, ++ struct hns3_tqp_stats *stats) ++{ ++ uint32_t cnt; ++ ++ cnt = hns3_read_dev(txq, HNS3_RING_TX_PKTNUM_RECORD_REG); ++ stats->rcb_tx_ring_pktnum_rcd += cnt; ++ stats->rcb_tx_ring_pktnum[txq->queue_id] += cnt; ++} ++ + /* + * Query tqp tx queue statistics ,opcode id: 0x0B03. + * Query tqp rx queue statistics ,opcode id: 0x0B13. +@@ -608,16 +626,15 @@ hns3_stats_get(struct rte_eth_dev *eth_dev, struct rte_eth_stats *rte_stats) + struct hns3_tqp_stats *stats = &hw->tqp_stats; + struct hns3_rx_queue *rxq; + struct hns3_tx_queue *txq; +- uint64_t cnt; + uint16_t i; + int ret; + ++ rte_spinlock_lock(&hw->stats_lock); + /* Update imissed stats */ + ret = hns3_update_imissed_stats(hw, false); + if (ret) { +- hns3_err(hw, "update imissed stats failed, ret = %d", +- ret); +- return ret; ++ hns3_err(hw, "update imissed stats failed, ret = %d", ret); ++ goto out; + } + rte_stats->imissed = imissed_stats->rpu_rx_drop_cnt + + imissed_stats->ssu_rx_drop_cnt; +@@ -628,15 +645,9 @@ hns3_stats_get(struct rte_eth_dev *eth_dev, struct rte_eth_stats *rte_stats) + if (rxq == NULL) + continue; + +- cnt = hns3_read_dev(rxq, HNS3_RING_RX_PKTNUM_RECORD_REG); +- /* +- * Read hardware and software in adjacent positions to minumize +- * the timing variance. +- */ ++ hns3_rcb_rx_ring_stats_get(rxq, stats); + rte_stats->ierrors += rxq->err_stats.l2_errors + + rxq->err_stats.pkt_len_errors; +- stats->rcb_rx_ring_pktnum_rcd += cnt; +- stats->rcb_rx_ring_pktnum[i] += cnt; + rte_stats->ibytes += rxq->basic_stats.bytes; + } -- struct ovs_flow_stats stats; -+ struct ovs_flow_stats stats_sw; -+ struct ovs_flow_stats stats_hw; - uint64_t lastused; +@@ -646,17 +657,14 @@ hns3_stats_get(struct rte_eth_dev *eth_dev, struct rte_eth_stats *rte_stats) + if (txq == NULL) + continue; -- struct { -- bool rewrite; -- struct tc_flower_key key; -- struct tc_flower_key mask; -- } rewrite; -- - uint32_t csum_update_flags; +- cnt = hns3_read_dev(txq, HNS3_RING_TX_PKTNUM_RECORD_REG); +- stats->rcb_tx_ring_pktnum_rcd += cnt; +- stats->rcb_tx_ring_pktnum[i] += cnt; ++ hns3_rcb_tx_ring_stats_get(txq, stats); + rte_stats->obytes += txq->basic_stats.bytes; + } - bool tunnel; -@@ -352,13 +359,6 @@ struct tc_flower { - enum tc_offload_policy tc_policy; - }; + ret = hns3_update_oerror_stats(hw, false); + if (ret) { +- hns3_err(hw, "update oerror stats failed, ret = %d", +- ret); +- return ret; ++ hns3_err(hw, "update oerror stats failed, ret = %d", ret); ++ goto out; + } + rte_stats->oerrors = hw->oerror_stats; --/* assert that if we overflow with a masked write of uint32_t to the last byte -- * of flower.rewrite we overflow inside struct flower. -- * shouldn't happen unless someone moves rewrite to the end of flower */ --BUILD_ASSERT_DECL(offsetof(struct tc_flower, rewrite) -- + MEMBER_SIZEOF(struct tc_flower, rewrite) -- + sizeof(uint32_t) - 2 < sizeof(struct tc_flower)); -- - int tc_replace_flower(struct tcf_id *id, struct tc_flower *flower); - int tc_del_filter(struct tcf_id *id); - int tc_get_flower(struct tcf_id *id, struct tc_flower *flower); -diff --git a/lib/tnl-ports.c b/lib/tnl-ports.c -index 58269d3b16..050eafa6b8 100644 ---- a/lib/tnl-ports.c -+++ b/lib/tnl-ports.c -@@ -71,7 +71,7 @@ tnl_port_cast(const struct cls_rule *cr) - { - BUILD_ASSERT_DECL(offsetof(struct tnl_port_in, cr) == 0); +@@ -672,7 +680,10 @@ hns3_stats_get(struct rte_eth_dev *eth_dev, struct rte_eth_stats *rte_stats) + rte_stats->oerrors; + rte_stats->rx_nombuf = eth_dev->data->rx_mbuf_alloc_failed; -- return CONTAINER_OF(cr, struct tnl_port_in, cr); -+ return cr ? CONTAINER_OF(cr, struct tnl_port_in, cr) : NULL; +- return 0; ++out: ++ rte_spinlock_unlock(&hw->stats_lock); ++ ++ return ret; } - static void -@@ -259,14 +259,14 @@ ipdev_map_delete(struct ip_device *ip_dev, ovs_be16 tp_port, uint8_t nw_proto) - void - tnl_port_map_delete(odp_port_t port, const char type[]) - { -- struct tnl_port *p, *next; -+ struct tnl_port *p; - struct ip_device *ip_dev; - uint8_t nw_proto; + int +@@ -685,6 +696,7 @@ hns3_stats_reset(struct rte_eth_dev *eth_dev) + uint16_t i; + int ret; + ++ rte_spinlock_lock(&hw->stats_lock); + /* + * Note: Reading hardware statistics of imissed registers will + * clear them. +@@ -692,7 +704,7 @@ hns3_stats_reset(struct rte_eth_dev *eth_dev) + ret = hns3_update_imissed_stats(hw, true); + if (ret) { + hns3_err(hw, "clear imissed stats failed, ret = %d", ret); +- return ret; ++ goto out; + } - nw_proto = tnl_type_to_nw_proto(type); + /* +@@ -701,9 +713,8 @@ hns3_stats_reset(struct rte_eth_dev *eth_dev) + */ + ret = hns3_update_oerror_stats(hw, true); + if (ret) { +- hns3_err(hw, "clear oerror stats failed, ret = %d", +- ret); +- return ret; ++ hns3_err(hw, "clear oerror stats failed, ret = %d", ret); ++ goto out; + } - ovs_mutex_lock(&mutex); -- LIST_FOR_EACH_SAFE(p, next, node, &port_list) { -+ LIST_FOR_EACH_SAFE (p, node, &port_list) { - if (p->port == port && p->nw_proto == nw_proto && - ovs_refcount_unref_relaxed(&p->ref_cnt) == 1) { - ovs_list_remove(&p->node); -@@ -444,11 +444,11 @@ delete_ipdev(struct ip_device *ip_dev) - void - tnl_port_map_insert_ipdev(const char dev_name[]) - { -- struct ip_device *ip_dev, *next; -+ struct ip_device *ip_dev; + for (i = 0; i < eth_dev->data->nb_rx_queues; i++) { +@@ -745,7 +756,10 @@ hns3_stats_reset(struct rte_eth_dev *eth_dev) - ovs_mutex_lock(&mutex); + hns3_tqp_stats_clear(hw); -- LIST_FOR_EACH_SAFE(ip_dev, next, node, &addr_list) { -+ LIST_FOR_EACH_SAFE (ip_dev, node, &addr_list) { - if (!strcmp(netdev_get_name(ip_dev->dev), dev_name)) { - if (ip_dev->change_seq == netdev_get_change_seq(ip_dev->dev)) { - goto out; -@@ -466,10 +466,10 @@ out: - void - tnl_port_map_delete_ipdev(const char dev_name[]) - { -- struct ip_device *ip_dev, *next; -+ struct ip_device *ip_dev; +- return 0; ++out: ++ rte_spinlock_unlock(&hw->stats_lock); ++ ++ return ret; + } - ovs_mutex_lock(&mutex); -- LIST_FOR_EACH_SAFE(ip_dev, next, node, &addr_list) { -+ LIST_FOR_EACH_SAFE (ip_dev, node, &addr_list) { - if (!strcmp(netdev_get_name(ip_dev->dev), dev_name)) { - delete_ipdev(ip_dev); - } -@@ -480,10 +480,10 @@ tnl_port_map_delete_ipdev(const char dev_name[]) - void - tnl_port_map_run(void) - { -- struct ip_device *ip_dev, *next; -+ struct ip_device *ip_dev; + static int +@@ -912,7 +926,6 @@ hns3_rxq_basic_stats_get(struct rte_eth_dev *dev, struct rte_eth_xstat *xstats, + struct hns3_rx_basic_stats *rxq_stats; + struct hns3_rx_queue *rxq; + uint16_t i, j; +- uint32_t cnt; + char *val; + + for (i = 0; i < dev->data->nb_rx_queues; i++) { +@@ -920,16 +933,10 @@ hns3_rxq_basic_stats_get(struct rte_eth_dev *dev, struct rte_eth_xstat *xstats, + if (rxq == NULL) + continue; + +- cnt = hns3_read_dev(rxq, HNS3_RING_RX_PKTNUM_RECORD_REG); +- /* +- * Read hardware and software in adjacent positions to minimize +- * the time difference. +- */ ++ hns3_rcb_rx_ring_stats_get(rxq, stats); + rxq_stats = &rxq->basic_stats; + rxq_stats->errors = rxq->err_stats.l2_errors + + rxq->err_stats.pkt_len_errors; +- stats->rcb_rx_ring_pktnum_rcd += cnt; +- stats->rcb_rx_ring_pktnum[i] += cnt; + + /* + * If HW statistics are reset by stats_reset, but a lot of +@@ -959,7 +966,6 @@ hns3_txq_basic_stats_get(struct rte_eth_dev *dev, struct rte_eth_xstat *xstats, + struct hns3_tx_basic_stats *txq_stats; + struct hns3_tx_queue *txq; + uint16_t i, j; +- uint32_t cnt; + char *val; + + for (i = 0; i < dev->data->nb_tx_queues; i++) { +@@ -967,9 +973,7 @@ hns3_txq_basic_stats_get(struct rte_eth_dev *dev, struct rte_eth_xstat *xstats, + if (txq == NULL) + continue; + +- cnt = hns3_read_dev(txq, HNS3_RING_TX_PKTNUM_RECORD_REG); +- stats->rcb_tx_ring_pktnum_rcd += cnt; +- stats->rcb_tx_ring_pktnum[i] += cnt; ++ hns3_rcb_tx_ring_stats_get(txq, stats); + + txq_stats = &txq->basic_stats; + txq_stats->packets = stats->rcb_tx_ring_pktnum[i]; +@@ -1024,9 +1028,13 @@ hns3_imissed_stats_get(struct rte_eth_dev *dev, struct rte_eth_xstat *xstats, + * @praram xstats + * A pointer to a table of structure of type *rte_eth_xstat* + * to be filled with device statistics ids and values. +- * This parameter can be set to NULL if n is 0. ++ * This parameter can be set to NULL if and only if n is 0. + * @param n + * The size of the xstats array (number of elements). ++ * If lower than the required number of elements, the function returns the ++ * required number of elements. ++ * If equal to zero, the xstats parameter must be NULL, the function returns ++ * the required number of elements. + * @return + * 0 on fail, count(The size of the statistics elements) on success. + */ +@@ -1045,15 +1053,13 @@ hns3_dev_xstats_get(struct rte_eth_dev *dev, struct rte_eth_xstat *xstats, + int count; + int ret; - ovs_mutex_lock(&mutex); -- LIST_FOR_EACH_SAFE(ip_dev, next, node, &addr_list) { -+ LIST_FOR_EACH_SAFE (ip_dev, node, &addr_list) { - char dev_name[IFNAMSIZ]; +- if (xstats == NULL) +- return 0; +- + count = hns3_xstats_calc_num(dev); + if ((int)n < count) + return count; + + count = 0; + ++ rte_spinlock_lock(&hw->stats_lock); + hns3_tqp_basic_stats_get(dev, xstats, &count); + + if (!hns->is_vf) { +@@ -1061,6 +1067,7 @@ hns3_dev_xstats_get(struct rte_eth_dev *dev, struct rte_eth_xstat *xstats, + ret = hns3_query_update_mac_stats(dev); + if (ret < 0) { + hns3_err(hw, "Update Mac stats fail : %d", ret); ++ rte_spinlock_unlock(&hw->stats_lock); + return ret; + } + +@@ -1075,8 +1082,8 @@ hns3_dev_xstats_get(struct rte_eth_dev *dev, struct rte_eth_xstat *xstats, + + ret = hns3_update_imissed_stats(hw, false); + if (ret) { +- hns3_err(hw, "update imissed stats failed, ret = %d", +- ret); ++ hns3_err(hw, "update imissed stats failed, ret = %d", ret); ++ rte_spinlock_unlock(&hw->stats_lock); + return ret; + } - if (ip_dev->change_seq == netdev_get_change_seq(ip_dev->dev)) { -diff --git a/lib/unixctl.c b/lib/unixctl.c -index 69aed6722c..103357ee91 100644 ---- a/lib/unixctl.c -+++ b/lib/unixctl.c -@@ -390,8 +390,8 @@ unixctl_server_run(struct unixctl_server *server) - } - } +@@ -1107,6 +1114,7 @@ hns3_dev_xstats_get(struct rte_eth_dev *dev, struct rte_eth_xstat *xstats, -- struct unixctl_conn *conn, *next; -- LIST_FOR_EACH_SAFE (conn, next, node, &server->conns) { -+ struct unixctl_conn *conn; -+ LIST_FOR_EACH_SAFE (conn, node, &server->conns) { - int error = run_connection(conn); - if (error && error != EAGAIN) { - kill_connection(conn); -@@ -422,9 +422,9 @@ void - unixctl_server_destroy(struct unixctl_server *server) - { - if (server) { -- struct unixctl_conn *conn, *next; -+ struct unixctl_conn *conn; + hns3_tqp_dfx_stats_get(dev, xstats, &count); + hns3_queue_stats_get(dev, xstats, &count); ++ rte_spinlock_unlock(&hw->stats_lock); -- LIST_FOR_EACH_SAFE (conn, next, node, &server->conns) { -+ LIST_FOR_EACH_SAFE (conn, node, &server->conns) { - kill_connection(conn); - } + return count; + } +@@ -1289,7 +1297,7 @@ hns3_dev_xstats_get_names(struct rte_eth_dev *dev, + * A pointer to an ids array passed by application. This tells which + * statistics values function should retrieve. This parameter + * can be set to NULL if size is 0. In this case function will retrieve +- * all avalible statistics. ++ * all available statistics. + * @param values + * A pointer to a table to be filled with device statistics values. + * @param size +@@ -1457,6 +1465,7 @@ int + hns3_dev_xstats_reset(struct rte_eth_dev *dev) + { + struct hns3_adapter *hns = dev->data->dev_private; ++ struct hns3_hw *hw = &hns->hw; + int ret; -diff --git a/lib/vconn.c b/lib/vconn.c -index 7415e6291f..b556762277 100644 ---- a/lib/vconn.c -+++ b/lib/vconn.c -@@ -960,8 +960,8 @@ vconn_transact_multipart(struct vconn *vconn, - ovs_list_init(replies); + /* Clear tqp stats */ +@@ -1464,23 +1473,25 @@ hns3_dev_xstats_reset(struct rte_eth_dev *dev) + if (ret) + return ret; - /* Send all the requests. */ -- struct ofpbuf *b, *next; -- LIST_FOR_EACH_SAFE (b, next, list_node, requests) { -+ struct ofpbuf *b; -+ LIST_FOR_EACH_SAFE (b, list_node, requests) { - ovs_list_remove(&b->list_node); - int error = vconn_send_block(vconn, b); - if (error) { -diff --git a/ofproto/bond.c b/ofproto/bond.c -index cdfdf0b9d8..845f69e21d 100644 ---- a/ofproto/bond.c -+++ b/ofproto/bond.c -@@ -338,7 +338,7 @@ static void - update_recirc_rules__(struct bond *bond) - { - struct match match; -- struct bond_pr_rule_op *pr_op, *next_op; -+ struct bond_pr_rule_op *pr_op; - uint64_t ofpacts_stub[128 / 8]; - struct ofpbuf ofpacts; - int i; -@@ -372,7 +372,7 @@ update_recirc_rules__(struct bond *bond) ++ rte_spinlock_lock(&hw->stats_lock); + hns3_tqp_dfx_stats_clear(dev); - ofpbuf_use_stub(&ofpacts, ofpacts_stub, sizeof ofpacts_stub); + /* Clear reset stats */ + memset(&hns->hw.reset.stats, 0, sizeof(struct hns3_reset_stats)); -- HMAP_FOR_EACH_SAFE(pr_op, next_op, hmap_node, &bond->pr_rule_ops) { -+ HMAP_FOR_EACH_SAFE (pr_op, hmap_node, &bond->pr_rule_ops) { - int error; - switch (pr_op->op) { - case ADD: -@@ -1258,7 +1258,7 @@ insert_bal(struct ovs_list *bals, struct bond_member *member) - break; - } - } -- ovs_list_insert(&pos->bal_node, &member->bal_node); -+ ovs_list_insert(pos ? &pos->bal_node : bals, &member->bal_node); - } + if (hns->is_vf) +- return 0; ++ goto out; - /* Removes 'member' from its current list and then inserts it into 'bals' so -diff --git a/ofproto/connmgr.c b/ofproto/connmgr.c -index fa8f6cd0e8..172a58cfb7 100644 ---- a/ofproto/connmgr.c -+++ b/ofproto/connmgr.c -@@ -310,8 +310,8 @@ connmgr_destroy(struct connmgr *mgr) - return; - } + /* HW registers are cleared on read */ + ret = hns3_mac_stats_reset(dev); +- if (ret) +- return ret; -- struct ofservice *ofservice, *next_ofservice; -- HMAP_FOR_EACH_SAFE (ofservice, next_ofservice, hmap_node, &mgr->services) { -+ struct ofservice *ofservice; -+ HMAP_FOR_EACH_SAFE (ofservice, hmap_node, &mgr->services) { - ofservice_destroy(ofservice); - } - hmap_destroy(&mgr->services); -@@ -351,8 +351,8 @@ connmgr_run(struct connmgr *mgr, - } - } +- return 0; ++out: ++ rte_spinlock_unlock(&hw->stats_lock); ++ ++ return ret; + } -- struct ofconn *ofconn, *next_ofconn; -- LIST_FOR_EACH_SAFE (ofconn, next_ofconn, connmgr_node, &mgr->conns) { -+ struct ofconn *ofconn; -+ LIST_FOR_EACH_SAFE (ofconn, connmgr_node, &mgr->conns) { - ofconn_run(ofconn, handle_openflow); - } - ofmonitor_run(mgr); -@@ -592,8 +592,8 @@ connmgr_set_controllers(struct connmgr *mgr, struct shash *controllers) +-int ++static int + hns3_tqp_stats_init(struct hns3_hw *hw) + { + struct hns3_tqp_stats *tqp_stats = &hw->tqp_stats; +@@ -1504,7 +1515,7 @@ hns3_tqp_stats_init(struct hns3_hw *hw) + return 0; + } - /* Delete services that are no longer configured. - * Update configuration of all now-existing services. */ -- struct ofservice *ofservice, *next_ofservice; -- HMAP_FOR_EACH_SAFE (ofservice, next_ofservice, hmap_node, &mgr->services) { -+ struct ofservice *ofservice; -+ HMAP_FOR_EACH_SAFE (ofservice, hmap_node, &mgr->services) { - const char *target = ofservice->target; - struct ofproto_controller *c = shash_find_data(controllers, target); - if (!c) { -@@ -1137,9 +1137,9 @@ ofconn_remove_bundle(struct ofconn *ofconn, struct ofp_bundle *bundle) - static void - bundle_remove_all(struct ofconn *ofconn) +-void ++static void + hns3_tqp_stats_uninit(struct hns3_hw *hw) { -- struct ofp_bundle *b, *next; -+ struct ofp_bundle *b; - -- HMAP_FOR_EACH_SAFE (b, next, node, &ofconn->bundles) { -+ HMAP_FOR_EACH_SAFE (b, node, &ofconn->bundles) { - ofp_bundle_remove__(ofconn, b); - } + struct hns3_tqp_stats *tqp_stats = &hw->tqp_stats; +@@ -1525,3 +1536,64 @@ hns3_tqp_stats_clear(struct hns3_hw *hw) + memset(stats->rcb_rx_ring_pktnum, 0, sizeof(uint64_t) * hw->tqps_num); + memset(stats->rcb_tx_ring_pktnum, 0, sizeof(uint64_t) * hw->tqps_num); } -@@ -1149,8 +1149,8 @@ bundle_remove_expired(struct ofconn *ofconn, long long int now) - { - long long int limit = now - bundle_idle_timeout; ++ ++int ++hns3_stats_init(struct hns3_hw *hw) ++{ ++ int ret; ++ ++ rte_spinlock_init(&hw->stats_lock); ++ /* Hardware statistics of imissed registers cleared. */ ++ ret = hns3_update_imissed_stats(hw, true); ++ if (ret) { ++ hns3_err(hw, "clear imissed stats failed, ret = %d", ret); ++ return ret; ++ } ++ ++ return hns3_tqp_stats_init(hw); ++} ++ ++void ++hns3_stats_uninit(struct hns3_hw *hw) ++{ ++ hns3_tqp_stats_uninit(hw); ++} ++ ++static void ++hns3_update_queues_stats(struct hns3_hw *hw) ++{ ++ struct rte_eth_dev_data *data = hw->data; ++ struct hns3_rx_queue *rxq; ++ struct hns3_tx_queue *txq; ++ uint16_t i; ++ ++ for (i = 0; i < data->nb_rx_queues; i++) { ++ rxq = data->rx_queues[i]; ++ if (rxq != NULL) ++ hns3_rcb_rx_ring_stats_get(rxq, &hw->tqp_stats); ++ } ++ ++ for (i = 0; i < data->nb_tx_queues; i++) { ++ txq = data->tx_queues[i]; ++ if (txq != NULL) ++ hns3_rcb_tx_ring_stats_get(txq, &hw->tqp_stats); ++ } ++} ++ ++/* ++ * Some hardware statistics registers are not 64-bit. If hardware statistics are ++ * not obtained for a long time, these statistics may be reversed. This function ++ * is used to update these hardware statistics in periodic task. ++ */ ++void ++hns3_update_hw_stats(struct hns3_hw *hw) ++{ ++ struct hns3_adapter *hns = HNS3_DEV_HW_TO_ADAPTER(hw); ++ ++ rte_spinlock_lock(&hw->stats_lock); ++ if (!hns->is_vf) ++ hns3_update_mac_stats(hw); ++ ++ hns3_update_queues_stats(hw); ++ rte_spinlock_unlock(&hw->stats_lock); ++} +diff --git a/dpdk/drivers/net/hns3/hns3_stats.h b/dpdk/drivers/net/hns3/hns3_stats.h +index d1230f94cb..b5cd6188b4 100644 +--- a/dpdk/drivers/net/hns3/hns3_stats.h ++++ b/dpdk/drivers/net/hns3/hns3_stats.h +@@ -5,11 +5,6 @@ + #ifndef _HNS3_STATS_H_ + #define _HNS3_STATS_H_ + +-/* stats macro */ +-#define HNS3_MAC_CMD_NUM 21 +-#define HNS3_RD_FIRST_STATS_NUM 2 +-#define HNS3_RD_OTHER_STATS_NUM 4 +- + /* TQP stats */ + struct hns3_tqp_stats { + uint64_t rcb_tx_ring_pktnum_rcd; /* Total num of transmitted packets */ +@@ -22,6 +17,7 @@ struct hns3_tqp_stats { + struct hns3_mac_stats { + uint64_t mac_tx_mac_pause_num; + uint64_t mac_rx_mac_pause_num; ++ uint64_t rsv0; + uint64_t mac_tx_pfc_pri0_pkt_num; + uint64_t mac_tx_pfc_pri1_pkt_num; + uint64_t mac_tx_pfc_pri2_pkt_num; +@@ -58,7 +54,7 @@ struct hns3_mac_stats { + uint64_t mac_tx_1519_2047_oct_pkt_num; + uint64_t mac_tx_2048_4095_oct_pkt_num; + uint64_t mac_tx_4096_8191_oct_pkt_num; +- uint64_t rsv0; ++ uint64_t rsv1; + uint64_t mac_tx_8192_9216_oct_pkt_num; + uint64_t mac_tx_9217_12287_oct_pkt_num; + uint64_t mac_tx_12288_16383_oct_pkt_num; +@@ -85,7 +81,7 @@ struct hns3_mac_stats { + uint64_t mac_rx_1519_2047_oct_pkt_num; + uint64_t mac_rx_2048_4095_oct_pkt_num; + uint64_t mac_rx_4096_8191_oct_pkt_num; +- uint64_t rsv1; ++ uint64_t rsv2; + uint64_t mac_rx_8192_9216_oct_pkt_num; + uint64_t mac_rx_9217_12287_oct_pkt_num; + uint64_t mac_rx_12288_16383_oct_pkt_num; +@@ -165,8 +161,9 @@ int hns3_dev_xstats_get_names_by_id(struct rte_eth_dev *dev, + struct rte_eth_xstat_name *xstats_names, + uint32_t size); + int hns3_stats_reset(struct rte_eth_dev *dev); +-int hns3_tqp_stats_init(struct hns3_hw *hw); +-void hns3_tqp_stats_uninit(struct hns3_hw *hw); +-int hns3_update_imissed_stats(struct hns3_hw *hw, bool is_clear); ++int hns3_stats_init(struct hns3_hw *hw); ++void hns3_stats_uninit(struct hns3_hw *hw); ++int hns3_query_mac_stats_reg_num(struct hns3_hw *hw); ++void hns3_update_hw_stats(struct hns3_hw *hw); + + #endif /* _HNS3_STATS_H_ */ +diff --git a/dpdk/drivers/net/i40e/i40e_ethdev.c b/dpdk/drivers/net/i40e/i40e_ethdev.c +index c0bfff43ee..bad27355fc 100644 +--- a/dpdk/drivers/net/i40e/i40e_ethdev.c ++++ b/dpdk/drivers/net/i40e/i40e_ethdev.c +@@ -386,6 +386,7 @@ static int i40e_set_default_mac_addr(struct rte_eth_dev *dev, + struct rte_ether_addr *mac_addr); + + static int i40e_dev_mtu_set(struct rte_eth_dev *dev, uint16_t mtu); ++static void i40e_set_mac_max_frame(struct rte_eth_dev *dev, uint16_t size); + + static int i40e_ethertype_filter_convert( + const struct rte_eth_ethertype_filter *input, +@@ -1709,11 +1710,6 @@ eth_i40e_dev_init(struct rte_eth_dev *dev, void *init_params __rte_unused) + */ + i40e_add_tx_flow_control_drop_filter(pf); + +- /* Set the max frame size to 0x2600 by default, +- * in case other drivers changed the default value. +- */ +- i40e_aq_set_mac_config(hw, I40E_FRAME_SIZE_MAX, TRUE, false, 0, NULL); +- + /* initialize RSS rule list */ + TAILQ_INIT(&pf->rss_config_list); -- struct ofp_bundle *b, *next; -- HMAP_FOR_EACH_SAFE (b, next, node, &ofconn->bundles) { -+ struct ofp_bundle *b; -+ HMAP_FOR_EACH_SAFE (b, node, &ofconn->bundles) { - if (b->used <= limit) { - ofconn_send_error(ofconn, b->msg, OFPERR_OFPBFC_TIMEOUT); - ofp_bundle_remove__(ofconn, b); -@@ -1247,8 +1247,8 @@ ofconn_destroy(struct ofconn *ofconn) +@@ -2364,6 +2360,7 @@ i40e_dev_start(struct rte_eth_dev *dev) + uint32_t intr_vector = 0; + struct i40e_vsi *vsi; + uint16_t nb_rxq, nb_txq; ++ uint16_t max_frame_size; - free(ofconn->async_cfg); + hw->adapter_stopped = 0; -- struct ofmonitor *monitor, *next_monitor; -- HMAP_FOR_EACH_SAFE (monitor, next_monitor, ofconn_node, -+ struct ofmonitor *monitor; -+ HMAP_FOR_EACH_SAFE (monitor, ofconn_node, - &ofconn->monitors) { - ofmonitor_destroy(monitor); - } -@@ -1953,8 +1953,8 @@ static void - ofservice_close_all(struct ofservice *ofservice) - OVS_REQUIRES(ofproto_mutex) +@@ -2483,7 +2480,7 @@ i40e_dev_start(struct rte_eth_dev *dev) + if (ret != I40E_SUCCESS) + PMD_DRV_LOG(WARNING, "Fail to set phy mask"); + +- /* Call get_link_info aq commond to enable/disable LSE */ ++ /* Call get_link_info aq command to enable/disable LSE */ + i40e_dev_link_update(dev, 0); + } + +@@ -2502,6 +2499,9 @@ i40e_dev_start(struct rte_eth_dev *dev) + "please call hierarchy_commit() " + "before starting the port"); + ++ max_frame_size = dev->data->mtu + I40E_ETH_OVERHEAD; ++ i40e_set_mac_max_frame(dev, max_frame_size); ++ + return I40E_SUCCESS; + + tx_err: +@@ -2848,6 +2848,9 @@ i40e_dev_set_link_down(struct rte_eth_dev *dev) + return i40e_phy_conf_link(hw, abilities, speed, false); + } + ++#define CHECK_INTERVAL 100 /* 100ms */ ++#define MAX_REPEAT_TIME 10 /* 1s (10 * 100ms) in total */ ++ + static __rte_always_inline void + update_link_reg(struct i40e_hw *hw, struct rte_eth_link *link) { -- struct ofconn *ofconn, *next; -- LIST_FOR_EACH_SAFE (ofconn, next, ofservice_node, &ofservice->conns) { -+ struct ofconn *ofconn; -+ LIST_FOR_EACH_SAFE (ofconn, ofservice_node, &ofservice->conns) { - ofconn_destroy(ofconn); - } +@@ -2914,8 +2917,6 @@ static __rte_always_inline void + update_link_aq(struct i40e_hw *hw, struct rte_eth_link *link, + bool enable_lse, int wait_to_complete) + { +-#define CHECK_INTERVAL 100 /* 100ms */ +-#define MAX_REPEAT_TIME 10 /* 1s (10 * 100ms) in total */ + uint32_t rep_cnt = MAX_REPEAT_TIME; + struct i40e_link_status link_status; + int status; +@@ -3555,7 +3556,7 @@ static int i40e_dev_xstats_get_names(__rte_unused struct rte_eth_dev *dev, + count++; + } + +- /* Get individiual stats from i40e_hw_port struct */ ++ /* Get individual stats from i40e_hw_port struct */ + for (i = 0; i < I40E_NB_HW_PORT_XSTATS; i++) { + strlcpy(xstats_names[count].name, + rte_i40e_hw_port_strings[i].name, +@@ -3613,7 +3614,7 @@ i40e_dev_xstats_get(struct rte_eth_dev *dev, struct rte_eth_xstat *xstats, + count++; + } + +- /* Get individiual stats from i40e_hw_port struct */ ++ /* Get individual stats from i40e_hw_port struct */ + for (i = 0; i < I40E_NB_HW_PORT_XSTATS; i++) { + xstats[count].value = *(uint64_t *)(((char *)hw_stats) + + rte_i40e_hw_port_strings[i].offset); +@@ -5544,7 +5545,7 @@ i40e_vsi_get_bw_config(struct i40e_vsi *vsi) + &ets_sla_config, NULL); + if (ret != I40E_SUCCESS) { + PMD_DRV_LOG(ERR, +- "VSI failed to get TC bandwdith configuration %u", ++ "VSI failed to get TC bandwidth configuration %u", + hw->aq.asq_last_status); + return ret; + } +@@ -6719,6 +6720,7 @@ i40e_dev_handle_aq_msg(struct rte_eth_dev *dev) + if (!ret) + rte_eth_dev_callback_process(dev, + RTE_ETH_EVENT_INTR_LSC, NULL); ++ + break; + default: + PMD_DRV_LOG(DEBUG, "Request %u is not supported yet", +@@ -6822,7 +6824,7 @@ i40e_handle_mdd_event(struct rte_eth_dev *dev) + * @param handle + * Pointer to interrupt handle. + * @param param +- * The address of parameter (struct rte_eth_dev *) regsitered before. ++ * The address of parameter (struct rte_eth_dev *) registered before. + * + * @return + * void +@@ -9719,7 +9721,7 @@ i40e_ethertype_filter_convert(const struct rte_eth_ethertype_filter *input, + return 0; } -diff --git a/ofproto/in-band.c b/ofproto/in-band.c -index 82d8dfa147..3992251f5f 100644 ---- a/ofproto/in-band.c -+++ b/ofproto/in-band.c -@@ -377,7 +377,7 @@ in_band_run(struct in_band *ib) - uint64_t ofpacts_stub[128 / 8]; - struct ofpbuf ofpacts; -- struct in_band_rule *rule, *next; -+ struct in_band_rule *rule; +-/* Check if there exists the ehtertype filter */ ++/* Check if there exists the ethertype filter */ + struct i40e_ethertype_filter * + i40e_sw_ethertype_filter_lookup(struct i40e_ethertype_rule *ethertype_rule, + const struct i40e_ethertype_filter_input *input) +@@ -12103,6 +12105,35 @@ i40e_cloud_filter_qinq_create(struct i40e_pf *pf) + return ret; + } - ofpbuf_use_stub(&ofpacts, ofpacts_stub, sizeof ofpacts_stub); ++static void ++i40e_set_mac_max_frame(struct rte_eth_dev *dev, uint16_t size) ++{ ++ struct i40e_hw *hw = I40E_DEV_PRIVATE_TO_HW(dev->data->dev_private); ++ uint32_t rep_cnt = MAX_REPEAT_TIME; ++ struct rte_eth_link link; ++ enum i40e_status_code status; ++ bool can_be_set = true; ++ ++ /* I40E_MEDIA_TYPE_BASET link up can be ignored */ ++ if (hw->phy.media_type != I40E_MEDIA_TYPE_BASET) { ++ do { ++ update_link_reg(hw, &link); ++ if (link.link_status) ++ break; ++ rte_delay_ms(CHECK_INTERVAL); ++ } while (--rep_cnt); ++ can_be_set = !!link.link_status; ++ } ++ ++ if (can_be_set) { ++ status = i40e_aq_set_mac_config(hw, size, TRUE, 0, false, NULL); ++ if (status != I40E_SUCCESS) ++ PMD_DRV_LOG(ERR, "Failed to set max frame size at port level"); ++ } else { ++ PMD_DRV_LOG(ERR, "Set max frame size at port level not applicable on link down"); ++ } ++} ++ + RTE_LOG_REGISTER_SUFFIX(i40e_logtype_init, init, NOTICE); + RTE_LOG_REGISTER_SUFFIX(i40e_logtype_driver, driver, NOTICE); + #ifdef RTE_ETHDEV_DEBUG_RX +diff --git a/dpdk/drivers/net/i40e/i40e_ethdev.h b/dpdk/drivers/net/i40e/i40e_ethdev.h +index 2d182f8000..a1ebdc093c 100644 +--- a/dpdk/drivers/net/i40e/i40e_ethdev.h ++++ b/dpdk/drivers/net/i40e/i40e_ethdev.h +@@ -897,7 +897,7 @@ struct i40e_tunnel_filter { + TAILQ_ENTRY(i40e_tunnel_filter) rules; + struct i40e_tunnel_filter_input input; + uint8_t is_to_vf; /* 0 - to PF, 1 - to VF */ +- uint16_t vf_id; /* VF id, avaiblable when is_to_vf is 1. */ ++ uint16_t vf_id; /* VF id, available when is_to_vf is 1. */ + uint16_t queue; /* Queue assigned to when match */ + }; -@@ -391,7 +391,7 @@ in_band_run(struct in_band *ib) +@@ -966,7 +966,7 @@ struct i40e_tunnel_filter_conf { + uint32_t tenant_id; /**< Tenant ID to match. VNI, GRE key... */ + uint16_t queue_id; /**< Queue assigned to if match. */ + uint8_t is_to_vf; /**< 0 - to PF, 1 - to VF */ +- uint16_t vf_id; /**< VF id, avaiblable when is_to_vf is 1. */ ++ uint16_t vf_id; /**< VF id, available when is_to_vf is 1. */ + }; - update_rules(ib); + TAILQ_HEAD(i40e_flow_list, rte_flow); +@@ -1100,7 +1100,7 @@ struct i40e_vf_msg_cfg { + /* + * If message statistics from a VF exceed the maximal limitation, + * the PF will ignore any new message from that VF for +- * 'ignor_second' time. ++ * 'ignore_second' time. + */ + uint32_t ignore_second; + }; +@@ -1257,7 +1257,7 @@ struct i40e_adapter { + }; -- HMAP_FOR_EACH_SAFE (rule, next, hmap_node, &ib->rules) { -+ HMAP_FOR_EACH_SAFE (rule, hmap_node, &ib->rules) { - switch (rule->op) { - case ADD: - ofproto_add_flow(ib->ofproto, &rule->match, rule->priority, -diff --git a/ofproto/netflow.c b/ofproto/netflow.c -index ed58de17de..aad9f9c77a 100644 ---- a/ofproto/netflow.c -+++ b/ofproto/netflow.c -@@ -299,7 +299,7 @@ static void - netflow_run__(struct netflow *nf) OVS_REQUIRES(mutex) - { - long long int now = time_msec(); -- struct netflow_flow *nf_flow, *next; -+ struct netflow_flow *nf_flow; + /** +- * Strucute to store private data for each VF representor instance ++ * Structure to store private data for each VF representor instance + */ + struct i40e_vf_representor { + uint16_t switch_domain_id; +@@ -1265,7 +1265,7 @@ struct i40e_vf_representor { + uint16_t vf_id; + /**< Virtual Function ID */ + struct i40e_adapter *adapter; +- /**< Private data store of assocaiated physical function */ ++ /**< Private data store of associated physical function */ + struct i40e_eth_stats stats_offset; + /**< Zero-point of VF statistics*/ + }; +diff --git a/dpdk/drivers/net/i40e/i40e_fdir.c b/dpdk/drivers/net/i40e/i40e_fdir.c +index df2a5aaecc..8caedea14e 100644 +--- a/dpdk/drivers/net/i40e/i40e_fdir.c ++++ b/dpdk/drivers/net/i40e/i40e_fdir.c +@@ -142,7 +142,7 @@ i40e_fdir_rx_queue_init(struct i40e_rx_queue *rxq) + I40E_QRX_TAIL(rxq->vsi->base_queue); + + rte_wmb(); +- /* Init the RX tail regieter. */ ++ /* Init the RX tail register. */ + I40E_PCI_REG_WRITE(rxq->qrx_tail, rxq->nb_rx_desc - 1); + + return err; +@@ -430,7 +430,7 @@ i40e_check_fdir_flex_payload(const struct rte_eth_flex_payload_cfg *flex_cfg) + + for (i = 0; i < I40E_FDIR_MAX_FLEX_LEN; i++) { + if (flex_cfg->src_offset[i] >= I40E_MAX_FLX_SOURCE_OFF) { +- PMD_DRV_LOG(ERR, "exceeds maxmial payload limit."); ++ PMD_DRV_LOG(ERR, "exceeds maximal payload limit."); + return -EINVAL; + } + } +@@ -438,7 +438,7 @@ i40e_check_fdir_flex_payload(const struct rte_eth_flex_payload_cfg *flex_cfg) + memset(flex_pit, 0, sizeof(flex_pit)); + num = i40e_srcoff_to_flx_pit(flex_cfg->src_offset, flex_pit); + if (num > I40E_MAX_FLXPLD_FIED) { +- PMD_DRV_LOG(ERR, "exceeds maxmial number of flex fields."); ++ PMD_DRV_LOG(ERR, "exceeds maximal number of flex fields."); + return -EINVAL; + } + for (i = 0; i < num; i++) { +@@ -948,7 +948,7 @@ i40e_flow_fdir_construct_pkt(struct i40e_pf *pf, + uint8_t pctype = fdir_input->pctype; + struct i40e_customized_pctype *cus_pctype; + +- /* raw pcket template - just copy contents of the raw packet */ ++ /* raw packet template - just copy contents of the raw packet */ + if (fdir_input->flow_ext.pkt_template) { + memcpy(raw_pkt, fdir_input->flow.raw_flow.packet, + fdir_input->flow.raw_flow.length); +@@ -1831,7 +1831,7 @@ i40e_flow_add_del_fdir_filter(struct rte_eth_dev *dev, + &check_filter.fdir.input); + if (!node) { + PMD_DRV_LOG(ERR, +- "There's no corresponding flow firector filter!"); ++ "There's no corresponding flow director filter!"); + return -EINVAL; + } + +diff --git a/dpdk/drivers/net/i40e/i40e_flow.c b/dpdk/drivers/net/i40e/i40e_flow.c +index c9676caab5..4f3808cb5f 100644 +--- a/dpdk/drivers/net/i40e/i40e_flow.c ++++ b/dpdk/drivers/net/i40e/i40e_flow.c +@@ -3043,7 +3043,7 @@ i40e_flow_parse_fdir_pattern(struct rte_eth_dev *dev, + rte_flow_error_set(error, EINVAL, + RTE_FLOW_ERROR_TYPE_ITEM, + item, +- "Exceeds maxmial payload limit."); ++ "Exceeds maximal payload limit."); + return -rte_errno; + } + +@@ -3142,8 +3142,11 @@ i40e_flow_parse_fdir_pattern(struct rte_eth_dev *dev, + /* Check if the input set is valid */ + if (i40e_validate_input_set(pctype, RTE_ETH_FILTER_FDIR, + input_set) != 0) { +- PMD_DRV_LOG(ERR, "Invalid input set"); +- return -EINVAL; ++ rte_flow_error_set(error, EINVAL, ++ RTE_FLOW_ERROR_TYPE_ITEM, ++ item, ++ "Invalid input set"); ++ return -rte_errno; + } + + filter->input.flow_ext.input_set = input_set; +diff --git a/dpdk/drivers/net/i40e/i40e_pf.c b/dpdk/drivers/net/i40e/i40e_pf.c +index ccb3924a5f..15d9ff868f 100644 +--- a/dpdk/drivers/net/i40e/i40e_pf.c ++++ b/dpdk/drivers/net/i40e/i40e_pf.c +@@ -343,7 +343,7 @@ i40e_pf_host_process_cmd_get_vf_resource(struct i40e_pf_vf *vf, uint8_t *msg, + vf->request_caps = *(uint32_t *)msg; + + /* enable all RSS by default, +- * doesn't support hena setting by virtchnnl yet. ++ * doesn't support hena setting by virtchnl yet. + */ + if (vf->request_caps & VIRTCHNL_VF_OFFLOAD_RSS_PF) { + I40E_WRITE_REG(hw, I40E_VFQF_HENA1(0, vf->vf_idx), +@@ -597,14 +597,14 @@ i40e_pf_config_irq_link_list(struct i40e_pf_vf *vf, + tempmap = vvm->rxq_map; + for (i = 0; i < sizeof(vvm->rxq_map) * BITS_PER_CHAR; i++) { + if (tempmap & 0x1) +- linklistmap |= (1 << (2 * i)); ++ linklistmap |= RTE_BIT64(2 * i); + tempmap >>= 1; + } - if (nf->packet.size) { - collectors_send(nf->collectors, nf->packet.data, nf->packet.size); -@@ -312,7 +312,7 @@ netflow_run__(struct netflow *nf) OVS_REQUIRES(mutex) + tempmap = vvm->txq_map; + for (i = 0; i < sizeof(vvm->txq_map) * BITS_PER_CHAR; i++) { + if (tempmap & 0x1) +- linklistmap |= (1 << (2 * i + 1)); ++ linklistmap |= RTE_BIT64(2 * i + 1); + tempmap >>= 1; + } - nf->next_timeout = now + 1000; +@@ -725,7 +725,7 @@ i40e_pf_host_process_cmd_config_irq_map(struct i40e_pf_vf *vf, + if ((map->rxq_map < qbit_max) && (map->txq_map < qbit_max)) { + i40e_pf_config_irq_link_list(vf, map); + } else { +- /* configured queue size excceed limit */ ++ /* configured queue size exceed limit */ + ret = I40E_ERR_PARAM; + goto send_msg; + } +diff --git a/dpdk/drivers/net/i40e/i40e_rxtx.c b/dpdk/drivers/net/i40e/i40e_rxtx.c +index e4cb33dc3c..9a00a9b71e 100644 +--- a/dpdk/drivers/net/i40e/i40e_rxtx.c ++++ b/dpdk/drivers/net/i40e/i40e_rxtx.c +@@ -609,7 +609,7 @@ i40e_rx_alloc_bufs(struct i40e_rx_queue *rxq) + rxdp[i].read.pkt_addr = dma_addr; + } -- HMAP_FOR_EACH_SAFE (nf_flow, next, hmap_node, &nf->flows) { -+ HMAP_FOR_EACH_SAFE (nf_flow, hmap_node, &nf->flows) { - if (now > nf_flow->last_expired + nf->active_timeout) { - bool idle = nf_flow->used < nf_flow->last_expired; - netflow_expire__(nf, nf_flow); -@@ -416,8 +416,8 @@ netflow_unref(struct netflow *nf) - collectors_destroy(nf->collectors); - ofpbuf_uninit(&nf->packet); +- /* Update rx tail regsiter */ ++ /* Update rx tail register */ + I40E_PCI_REG_WRITE(rxq->qrx_tail, rxq->rx_free_trigger); + + rxq->rx_free_trigger = +@@ -995,7 +995,7 @@ i40e_recv_scattered_pkts(void *rx_queue, + * threshold of the queue, advance the Receive Descriptor Tail (RDT) + * register. Update the RDT with the value of the last processed RX + * descriptor minus 1, to guarantee that the RDT register is never +- * equal to the RDH register, which creates a "full" ring situtation ++ * equal to the RDH register, which creates a "full" ring situation + * from the hardware point of view. + */ + nb_hold = (uint16_t)(nb_hold + rxq->nb_rx_hold); +@@ -1467,7 +1467,7 @@ tx_xmit_pkts(struct i40e_tx_queue *txq, + i40e_tx_fill_hw_ring(txq, tx_pkts + n, (uint16_t)(nb_pkts - n)); + txq->tx_tail = (uint16_t)(txq->tx_tail + (nb_pkts - n)); + +- /* Determin if RS bit needs to be set */ ++ /* Determine if RS bit needs to be set */ + if (txq->tx_tail > txq->tx_next_rs) { + txr[txq->tx_next_rs].cmd_type_offset_bsz |= + rte_cpu_to_le_64(((uint64_t)I40E_TX_DESC_CMD_RS) << +@@ -1697,7 +1697,7 @@ i40e_dev_rx_queue_start(struct rte_eth_dev *dev, uint16_t rx_queue_id) + } -- struct netflow_flow *nf_flow, *next; -- HMAP_FOR_EACH_SAFE (nf_flow, next, hmap_node, &nf->flows) { -+ struct netflow_flow *nf_flow; -+ HMAP_FOR_EACH_SAFE (nf_flow, hmap_node, &nf->flows) { - hmap_remove(&nf->flows, &nf_flow->hmap_node); - free(nf_flow); - } -diff --git a/ofproto/ofproto-dpif-ipfix.c b/ofproto/ofproto-dpif-ipfix.c -index 9280e008ea..742eed3998 100644 ---- a/ofproto/ofproto-dpif-ipfix.c -+++ b/ofproto/ofproto-dpif-ipfix.c -@@ -926,17 +926,21 @@ dpif_ipfix_bridge_exporter_destroy(struct dpif_ipfix_bridge_exporter *exporter) - static void - dpif_ipfix_bridge_exporter_set_options( - struct dpif_ipfix_bridge_exporter *exporter, -- const struct ofproto_ipfix_bridge_exporter_options *options) -+ const struct ofproto_ipfix_bridge_exporter_options *options, -+ bool *options_changed) - { -- bool options_changed; -- - if (!options || sset_is_empty(&options->targets)) { - /* No point in doing any work if there are no targets. */ -- dpif_ipfix_bridge_exporter_clear(exporter); -+ if (exporter->options) { -+ dpif_ipfix_bridge_exporter_clear(exporter); -+ *options_changed = true; -+ } else { -+ *options_changed = false; -+ } - return; - } + if (rxq->rx_deferred_start) +- PMD_DRV_LOG(WARNING, "RX queue %u is deferrd start", ++ PMD_DRV_LOG(WARNING, "RX queue %u is deferred start", + rx_queue_id); -- options_changed = ( -+ *options_changed = ( - !exporter->options - || !ofproto_ipfix_bridge_exporter_options_equal( - options, exporter->options)); -@@ -945,7 +949,7 @@ dpif_ipfix_bridge_exporter_set_options( - * shortchanged in collectors (which indicates that opening one or - * more of the configured collectors failed, so that we should - * retry). */ -- if (options_changed -+ if (*options_changed - || collectors_count(exporter->exporter.collectors) - < sset_count(&options->targets)) { - if (!dpif_ipfix_exporter_set_options( -@@ -957,7 +961,7 @@ dpif_ipfix_bridge_exporter_set_options( - } + err = i40e_alloc_rx_queue_mbufs(rxq); +@@ -1706,7 +1706,7 @@ i40e_dev_rx_queue_start(struct rte_eth_dev *dev, uint16_t rx_queue_id) + return err; + } - /* Avoid reconfiguring if options didn't change. */ -- if (!options_changed) { -+ if (!*options_changed) { - return; - } +- /* Init the RX tail regieter. */ ++ /* Init the RX tail register. */ + I40E_PCI_REG_WRITE(rxq->qrx_tail, rxq->nb_rx_desc - 1); -@@ -1015,17 +1019,21 @@ dpif_ipfix_flow_exporter_destroy(struct dpif_ipfix_flow_exporter *exporter) - static bool - dpif_ipfix_flow_exporter_set_options( - struct dpif_ipfix_flow_exporter *exporter, -- const struct ofproto_ipfix_flow_exporter_options *options) -+ const struct ofproto_ipfix_flow_exporter_options *options, -+ bool *options_changed) - { -- bool options_changed; -- - if (sset_is_empty(&options->targets)) { - /* No point in doing any work if there are no targets. */ -- dpif_ipfix_flow_exporter_clear(exporter); -+ if (exporter->options) { -+ dpif_ipfix_flow_exporter_clear(exporter); -+ *options_changed = true; -+ } else { -+ *options_changed = false; -+ } - return true; - } + err = i40e_switch_rx_queue(hw, rxq->reg_idx, TRUE); +@@ -1771,7 +1771,7 @@ i40e_dev_tx_queue_start(struct rte_eth_dev *dev, uint16_t tx_queue_id) + } -- options_changed = ( -+ *options_changed = ( - !exporter->options - || !ofproto_ipfix_flow_exporter_options_equal( - options, exporter->options)); -@@ -1034,7 +1042,7 @@ dpif_ipfix_flow_exporter_set_options( - * shortchanged in collectors (which indicates that opening one or - * more of the configured collectors failed, so that we should - * retry). */ -- if (options_changed -+ if (*options_changed - || collectors_count(exporter->exporter.collectors) - < sset_count(&options->targets)) { - if (!dpif_ipfix_exporter_set_options( -@@ -1046,7 +1054,7 @@ dpif_ipfix_flow_exporter_set_options( - } + if (txq->tx_deferred_start) +- PMD_DRV_LOG(WARNING, "TX queue %u is deferrd start", ++ PMD_DRV_LOG(WARNING, "TX queue %u is deferred start", + tx_queue_id); - /* Avoid reconfiguring if options didn't change. */ -- if (!options_changed) { -+ if (!*options_changed) { - return true; - } + /* +@@ -1930,7 +1930,7 @@ i40e_dev_rx_queue_setup_runtime(struct rte_eth_dev *dev, + PMD_DRV_LOG(ERR, "Can't use default burst."); + return -EINVAL; + } +- /* check scatterred conflict */ ++ /* check scattered conflict */ + if (!dev->data->scattered_rx && use_scattered_rx) { + PMD_DRV_LOG(ERR, "Scattered rx is required."); + return -EINVAL; +@@ -2014,7 +2014,7 @@ i40e_dev_rx_queue_setup(struct rte_eth_dev *dev, + rxq->rx_deferred_start = rx_conf->rx_deferred_start; + rxq->offloads = offloads; + +- /* Allocate the maximun number of RX ring hardware descriptor. */ ++ /* Allocate the maximum number of RX ring hardware descriptor. */ + len = I40E_MAX_RING_DESC; + + /** +@@ -2322,7 +2322,7 @@ i40e_dev_tx_queue_setup(struct rte_eth_dev *dev, + */ + tx_free_thresh = (uint16_t)((tx_conf->tx_free_thresh) ? + tx_conf->tx_free_thresh : DEFAULT_TX_FREE_THRESH); +- /* force tx_rs_thresh to adapt an aggresive tx_free_thresh */ ++ /* force tx_rs_thresh to adapt an aggressive tx_free_thresh */ + tx_rs_thresh = (DEFAULT_TX_RS_THRESH + tx_free_thresh > nb_desc) ? + nb_desc - tx_free_thresh : DEFAULT_TX_RS_THRESH; + if (tx_conf->tx_rs_thresh > 0) +@@ -2991,7 +2991,7 @@ i40e_rx_queue_init(struct i40e_rx_queue *rxq) + if (rxq->max_pkt_len > buf_size) + dev_data->scattered_rx = 1; + +- /* Init the RX tail regieter. */ ++ /* Init the RX tail register. */ + I40E_PCI_REG_WRITE(rxq->qrx_tail, rxq->nb_rx_desc - 1); + + return 0; +diff --git a/dpdk/drivers/net/i40e/i40e_rxtx_vec_altivec.c b/dpdk/drivers/net/i40e/i40e_rxtx_vec_altivec.c +index d0bf86dfba..00a015013e 100644 +--- a/dpdk/drivers/net/i40e/i40e_rxtx_vec_altivec.c ++++ b/dpdk/drivers/net/i40e/i40e_rxtx_vec_altivec.c +@@ -27,10 +27,10 @@ i40e_rxq_rearm(struct i40e_rx_queue *rxq) + struct i40e_rx_entry *rxep = &rxq->sw_ring[rxq->rxrearm_start]; + struct rte_mbuf *mb0, *mb1; + +- vector unsigned long hdr_room = (vector unsigned long){ ++ __vector unsigned long hdr_room = (__vector unsigned long){ + RTE_PKTMBUF_HEADROOM, + RTE_PKTMBUF_HEADROOM}; +- vector unsigned long dma_addr0, dma_addr1; ++ __vector unsigned long dma_addr0, dma_addr1; + + rxdp = rxq->rx_ring + rxq->rxrearm_start; + +@@ -40,11 +40,11 @@ i40e_rxq_rearm(struct i40e_rx_queue *rxq) + RTE_I40E_RXQ_REARM_THRESH) < 0) { + if (rxq->rxrearm_nb + RTE_I40E_RXQ_REARM_THRESH >= + rxq->nb_rx_desc) { +- dma_addr0 = (vector unsigned long){}; ++ dma_addr0 = (__vector unsigned long){}; + for (i = 0; i < RTE_I40E_DESCS_PER_LOOP; i++) { + rxep[i].mbuf = &rxq->fake_mbuf; + vec_st(dma_addr0, 0, +- (vector unsigned long *)&rxdp[i].read); ++ (__vector unsigned long *)&rxdp[i].read); + } + } + rte_eth_devices[rxq->port_id].data->rx_mbuf_alloc_failed += +@@ -54,7 +54,7 @@ i40e_rxq_rearm(struct i40e_rx_queue *rxq) + + /* Initialize the mbufs in vector, process 2 mbufs in one loop */ + for (i = 0; i < RTE_I40E_RXQ_REARM_THRESH; i += 2, rxep += 2) { +- vector unsigned long vaddr0, vaddr1; ++ __vector unsigned long vaddr0, vaddr1; + uintptr_t p0, p1; + + mb0 = rxep[0].mbuf; +@@ -72,8 +72,8 @@ i40e_rxq_rearm(struct i40e_rx_queue *rxq) + *(uint64_t *)p1 = rxq->mbuf_initializer; + + /* load buf_addr(lo 64bit) and buf_iova(hi 64bit) */ +- vaddr0 = vec_ld(0, (vector unsigned long *)&mb0->buf_addr); +- vaddr1 = vec_ld(0, (vector unsigned long *)&mb1->buf_addr); ++ vaddr0 = vec_ld(0, (__vector unsigned long *)&mb0->buf_addr); ++ vaddr1 = vec_ld(0, (__vector unsigned long *)&mb1->buf_addr); + + /* convert pa to dma_addr hdr/data */ + dma_addr0 = vec_mergel(vaddr0, vaddr0); +@@ -84,8 +84,8 @@ i40e_rxq_rearm(struct i40e_rx_queue *rxq) + dma_addr1 = vec_add(dma_addr1, hdr_room); + + /* flush desc with pa dma_addr */ +- vec_st(dma_addr0, 0, (vector unsigned long *)&rxdp++->read); +- vec_st(dma_addr1, 0, (vector unsigned long *)&rxdp++->read); ++ vec_st(dma_addr0, 0, (__vector unsigned long *)&rxdp++->read); ++ vec_st(dma_addr1, 0, (__vector unsigned long *)&rxdp++->read); + } -@@ -1069,7 +1077,7 @@ remove_flow_exporter(struct dpif_ipfix *di, - free(node); + rxq->rxrearm_start += RTE_I40E_RXQ_REARM_THRESH; +@@ -102,32 +102,32 @@ i40e_rxq_rearm(struct i40e_rx_queue *rxq) } --void -+bool - dpif_ipfix_set_options( - struct dpif_ipfix *di, - const struct ofproto_ipfix_bridge_exporter_options *bridge_exporter_options, -@@ -1077,16 +1085,19 @@ dpif_ipfix_set_options( - size_t n_flow_exporters_options) OVS_EXCLUDED(mutex) + static inline void +-desc_to_olflags_v(vector unsigned long descs[4], struct rte_mbuf **rx_pkts) ++desc_to_olflags_v(__vector unsigned long descs[4], struct rte_mbuf **rx_pkts) { - int i; -+ bool beo_changed, feo_changed, entry_changed; - struct ofproto_ipfix_flow_exporter_options *options; -- struct dpif_ipfix_flow_exporter_map_node *node, *next; -+ struct dpif_ipfix_flow_exporter_map_node *node; +- vector unsigned int vlan0, vlan1, rss, l3_l4e; ++ __vector unsigned int vlan0, vlan1, rss, l3_l4e; + + /* mask everything except RSS, flow director and VLAN flags + * bit2 is for VLAN tag, bit11 for flow director indication + * bit13:12 for RSS indication. + */ +- const vector unsigned int rss_vlan_msk = (vector unsigned int){ ++ const __vector unsigned int rss_vlan_msk = (__vector unsigned int){ + (int32_t)0x1c03804, (int32_t)0x1c03804, + (int32_t)0x1c03804, (int32_t)0x1c03804}; + + /* map rss and vlan type to rss hash and vlan flag */ +- const vector unsigned char vlan_flags = (vector unsigned char){ ++ const __vector unsigned char vlan_flags = (__vector unsigned char){ + 0, 0, 0, 0, + RTE_MBUF_F_RX_VLAN | RTE_MBUF_F_RX_VLAN_STRIPPED, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0}; + +- const vector unsigned char rss_flags = (vector unsigned char){ ++ const __vector unsigned char rss_flags = (__vector unsigned char){ + 0, RTE_MBUF_F_RX_FDIR, 0, 0, + 0, 0, RTE_MBUF_F_RX_RSS_HASH, RTE_MBUF_F_RX_RSS_HASH | RTE_MBUF_F_RX_FDIR, + 0, 0, 0, 0, + 0, 0, 0, 0}; + +- const vector unsigned char l3_l4e_flags = (vector unsigned char){ ++ const __vector unsigned char l3_l4e_flags = (__vector unsigned char){ + 0, + RTE_MBUF_F_RX_IP_CKSUM_BAD, + RTE_MBUF_F_RX_L4_CKSUM_BAD, +@@ -139,23 +139,23 @@ desc_to_olflags_v(vector unsigned long descs[4], struct rte_mbuf **rx_pkts) + | RTE_MBUF_F_RX_IP_CKSUM_BAD, + 0, 0, 0, 0, 0, 0, 0, 0}; + +- vlan0 = (vector unsigned int)vec_mergel(descs[0], descs[1]); +- vlan1 = (vector unsigned int)vec_mergel(descs[2], descs[3]); +- vlan0 = (vector unsigned int)vec_mergeh(vlan0, vlan1); ++ vlan0 = (__vector unsigned int)vec_mergel(descs[0], descs[1]); ++ vlan1 = (__vector unsigned int)vec_mergel(descs[2], descs[3]); ++ vlan0 = (__vector unsigned int)vec_mergeh(vlan0, vlan1); + + vlan1 = vec_and(vlan0, rss_vlan_msk); +- vlan0 = (vector unsigned int)vec_perm(vlan_flags, +- (vector unsigned char){}, +- *(vector unsigned char *)&vlan1); ++ vlan0 = (__vector unsigned int)vec_perm(vlan_flags, ++ (__vector unsigned char){}, ++ *(__vector unsigned char *)&vlan1); + +- rss = vec_sr(vlan1, (vector unsigned int){11, 11, 11, 11}); +- rss = (vector unsigned int)vec_perm(rss_flags, (vector unsigned char){}, +- *(vector unsigned char *)&rss); ++ rss = vec_sr(vlan1, (__vector unsigned int){11, 11, 11, 11}); ++ rss = (__vector unsigned int)vec_perm(rss_flags, (__vector unsigned char){}, ++ *(__vector unsigned char *)&rss); + +- l3_l4e = vec_sr(vlan1, (vector unsigned int){22, 22, 22, 22}); +- l3_l4e = (vector unsigned int)vec_perm(l3_l4e_flags, +- (vector unsigned char){}, +- *(vector unsigned char *)&l3_l4e); ++ l3_l4e = vec_sr(vlan1, (__vector unsigned int){22, 22, 22, 22}); ++ l3_l4e = (__vector unsigned int)vec_perm(l3_l4e_flags, ++ (__vector unsigned char){}, ++ *(__vector unsigned char *)&l3_l4e); + + vlan0 = vec_or(vlan0, rss); + vlan0 = vec_or(vlan0, l3_l4e); +@@ -169,23 +169,23 @@ desc_to_olflags_v(vector unsigned long descs[4], struct rte_mbuf **rx_pkts) + #define PKTLEN_SHIFT 10 + + static inline void +-desc_to_ptype_v(vector unsigned long descs[4], struct rte_mbuf **rx_pkts, ++desc_to_ptype_v(__vector unsigned long descs[4], struct rte_mbuf **rx_pkts, + uint32_t *ptype_tbl) + { +- vector unsigned long ptype0 = vec_mergel(descs[0], descs[1]); +- vector unsigned long ptype1 = vec_mergel(descs[2], descs[3]); ++ __vector unsigned long ptype0 = vec_mergel(descs[0], descs[1]); ++ __vector unsigned long ptype1 = vec_mergel(descs[2], descs[3]); + +- ptype0 = vec_sr(ptype0, (vector unsigned long){30, 30}); +- ptype1 = vec_sr(ptype1, (vector unsigned long){30, 30}); ++ ptype0 = vec_sr(ptype0, (__vector unsigned long){30, 30}); ++ ptype1 = vec_sr(ptype1, (__vector unsigned long){30, 30}); + + rx_pkts[0]->packet_type = +- ptype_tbl[(*(vector unsigned char *)&ptype0)[0]]; ++ ptype_tbl[(*(__vector unsigned char *)&ptype0)[0]]; + rx_pkts[1]->packet_type = +- ptype_tbl[(*(vector unsigned char *)&ptype0)[8]]; ++ ptype_tbl[(*(__vector unsigned char *)&ptype0)[8]]; + rx_pkts[2]->packet_type = +- ptype_tbl[(*(vector unsigned char *)&ptype1)[0]]; ++ ptype_tbl[(*(__vector unsigned char *)&ptype1)[0]]; + rx_pkts[3]->packet_type = +- ptype_tbl[(*(vector unsigned char *)&ptype1)[8]]; ++ ptype_tbl[(*(__vector unsigned char *)&ptype1)[8]]; + } - ovs_mutex_lock(&mutex); - dpif_ipfix_bridge_exporter_set_options(&di->bridge_exporter, -- bridge_exporter_options); -+ bridge_exporter_options, -+ &beo_changed); + /** +@@ -204,17 +204,17 @@ _recv_raw_pkts_vec(struct i40e_rx_queue *rxq, struct rte_mbuf **rx_pkts, + uint16_t nb_pkts_recd; + int pos; + uint64_t var; +- vector unsigned char shuf_msk; ++ __vector unsigned char shuf_msk; + uint32_t *ptype_tbl = rxq->vsi->adapter->ptype_tbl; + +- vector unsigned short crc_adjust = (vector unsigned short){ ++ __vector unsigned short crc_adjust = (__vector unsigned short){ + 0, 0, /* ignore pkt_type field */ + rxq->crc_len, /* sub crc on pkt_len */ + 0, /* ignore high-16bits of pkt_len */ + rxq->crc_len, /* sub crc on data_len */ + 0, 0, 0 /* ignore non-length fields */ + }; +- vector unsigned long dd_check, eop_check; ++ __vector unsigned long dd_check, eop_check; + + /* nb_pkts has to be floor-aligned to RTE_I40E_DESCS_PER_LOOP */ + nb_pkts = RTE_ALIGN_FLOOR(nb_pkts, RTE_I40E_DESCS_PER_LOOP); +@@ -240,15 +240,15 @@ _recv_raw_pkts_vec(struct i40e_rx_queue *rxq, struct rte_mbuf **rx_pkts, + return 0; - /* Add new flow exporters and update current flow exporters. */ - options = (struct ofproto_ipfix_flow_exporter_options *) - flow_exporters_options; -+ feo_changed = false; - for (i = 0; i < n_flow_exporters_options; i++) { - node = dpif_ipfix_find_flow_exporter_map_node( - di, options->collector_set_id); -@@ -1095,15 +1106,19 @@ dpif_ipfix_set_options( - dpif_ipfix_flow_exporter_init(&node->exporter); - hmap_insert(&di->flow_exporter_map, &node->node, - hash_int(options->collector_set_id, 0)); -+ feo_changed = true; - } -- if (!dpif_ipfix_flow_exporter_set_options(&node->exporter, options)) { -+ if (!dpif_ipfix_flow_exporter_set_options(&node->exporter, -+ options, -+ &entry_changed)) { - remove_flow_exporter(di, node); - } -+ feo_changed = entry_changed ? true : feo_changed; - options++; - } + /* 4 packets DD mask */ +- dd_check = (vector unsigned long){0x0000000100000001ULL, ++ dd_check = (__vector unsigned long){0x0000000100000001ULL, + 0x0000000100000001ULL}; + + /* 4 packets EOP mask */ +- eop_check = (vector unsigned long){0x0000000200000002ULL, ++ eop_check = (__vector unsigned long){0x0000000200000002ULL, + 0x0000000200000002ULL}; + + /* mask to shuffle from desc. to mbuf */ +- shuf_msk = (vector unsigned char){ ++ shuf_msk = (__vector unsigned char){ + 0xFF, 0xFF, /* pkt_type set as unknown */ + 0xFF, 0xFF, /* pkt_type set as unknown */ + 14, 15, /* octet 15~14, low 16 bits pkt_len */ +@@ -274,35 +274,35 @@ _recv_raw_pkts_vec(struct i40e_rx_queue *rxq, struct rte_mbuf **rx_pkts, + for (pos = 0, nb_pkts_recd = 0; pos < nb_pkts; + pos += RTE_I40E_DESCS_PER_LOOP, + rxdp += RTE_I40E_DESCS_PER_LOOP) { +- vector unsigned long descs[RTE_I40E_DESCS_PER_LOOP]; +- vector unsigned char pkt_mb1, pkt_mb2, pkt_mb3, pkt_mb4; +- vector unsigned short staterr, sterr_tmp1, sterr_tmp2; +- vector unsigned long mbp1, mbp2; /* two mbuf pointer ++ __vector unsigned long descs[RTE_I40E_DESCS_PER_LOOP]; ++ __vector unsigned char pkt_mb1, pkt_mb2, pkt_mb3, pkt_mb4; ++ __vector unsigned short staterr, sterr_tmp1, sterr_tmp2; ++ __vector unsigned long mbp1, mbp2; /* two mbuf pointer + * in one XMM reg. + */ + + /* B.1 load 2 mbuf point */ +- mbp1 = *(vector unsigned long *)&sw_ring[pos]; ++ mbp1 = *(__vector unsigned long *)&sw_ring[pos]; + /* Read desc statuses backwards to avoid race condition */ + /* A.1 load desc[3] */ +- descs[3] = *(vector unsigned long *)(rxdp + 3); ++ descs[3] = *(__vector unsigned long *)(rxdp + 3); + rte_compiler_barrier(); + + /* B.2 copy 2 mbuf point into rx_pkts */ +- *(vector unsigned long *)&rx_pkts[pos] = mbp1; ++ *(__vector unsigned long *)&rx_pkts[pos] = mbp1; + + /* B.1 load 2 mbuf point */ +- mbp2 = *(vector unsigned long *)&sw_ring[pos + 2]; ++ mbp2 = *(__vector unsigned long *)&sw_ring[pos + 2]; + + /* A.1 load desc[2-0] */ +- descs[2] = *(vector unsigned long *)(rxdp + 2); ++ descs[2] = *(__vector unsigned long *)(rxdp + 2); + rte_compiler_barrier(); +- descs[1] = *(vector unsigned long *)(rxdp + 1); ++ descs[1] = *(__vector unsigned long *)(rxdp + 1); + rte_compiler_barrier(); +- descs[0] = *(vector unsigned long *)(rxdp); ++ descs[0] = *(__vector unsigned long *)(rxdp); + + /* B.2 copy 2 mbuf point into rx_pkts */ +- *(vector unsigned long *)&rx_pkts[pos + 2] = mbp2; ++ *(__vector unsigned long *)&rx_pkts[pos + 2] = mbp2; + + if (split_packet) { + rte_mbuf_prefetch_part2(rx_pkts[pos]); +@@ -315,78 +315,78 @@ _recv_raw_pkts_vec(struct i40e_rx_queue *rxq, struct rte_mbuf **rx_pkts, + rte_compiler_barrier(); + + /* pkt 3,4 shift the pktlen field to be 16-bit aligned*/ +- const vector unsigned int len3 = vec_sl( +- vec_ld(0, (vector unsigned int *)&descs[3]), +- (vector unsigned int){0, 0, 0, PKTLEN_SHIFT}); ++ const __vector unsigned int len3 = vec_sl( ++ vec_ld(0, (__vector unsigned int *)&descs[3]), ++ (__vector unsigned int){0, 0, 0, PKTLEN_SHIFT}); + +- const vector unsigned int len2 = vec_sl( +- vec_ld(0, (vector unsigned int *)&descs[2]), +- (vector unsigned int){0, 0, 0, PKTLEN_SHIFT}); ++ const __vector unsigned int len2 = vec_sl( ++ vec_ld(0, (__vector unsigned int *)&descs[2]), ++ (__vector unsigned int){0, 0, 0, PKTLEN_SHIFT}); + + /* merge the now-aligned packet length fields back in */ +- descs[3] = (vector unsigned long)len3; +- descs[2] = (vector unsigned long)len2; ++ descs[3] = (__vector unsigned long)len3; ++ descs[2] = (__vector unsigned long)len2; + + /* D.1 pkt 3,4 convert format from desc to pktmbuf */ +- pkt_mb4 = vec_perm((vector unsigned char)descs[3], +- (vector unsigned char){}, shuf_msk); +- pkt_mb3 = vec_perm((vector unsigned char)descs[2], +- (vector unsigned char){}, shuf_msk); ++ pkt_mb4 = vec_perm((__vector unsigned char)descs[3], ++ (__vector unsigned char){}, shuf_msk); ++ pkt_mb3 = vec_perm((__vector unsigned char)descs[2], ++ (__vector unsigned char){}, shuf_msk); + + /* C.1 4=>2 filter staterr info only */ +- sterr_tmp2 = vec_mergel((vector unsigned short)descs[3], +- (vector unsigned short)descs[2]); ++ sterr_tmp2 = vec_mergel((__vector unsigned short)descs[3], ++ (__vector unsigned short)descs[2]); + /* C.1 4=>2 filter staterr info only */ +- sterr_tmp1 = vec_mergel((vector unsigned short)descs[1], +- (vector unsigned short)descs[0]); ++ sterr_tmp1 = vec_mergel((__vector unsigned short)descs[1], ++ (__vector unsigned short)descs[0]); + /* D.2 pkt 3,4 set in_port/nb_seg and remove crc */ +- pkt_mb4 = (vector unsigned char)vec_sub( +- (vector unsigned short)pkt_mb4, crc_adjust); +- pkt_mb3 = (vector unsigned char)vec_sub( +- (vector unsigned short)pkt_mb3, crc_adjust); ++ pkt_mb4 = (__vector unsigned char)vec_sub( ++ (__vector unsigned short)pkt_mb4, crc_adjust); ++ pkt_mb3 = (__vector unsigned char)vec_sub( ++ (__vector unsigned short)pkt_mb3, crc_adjust); + + /* pkt 1,2 shift the pktlen field to be 16-bit aligned*/ +- const vector unsigned int len1 = vec_sl( +- vec_ld(0, (vector unsigned int *)&descs[1]), +- (vector unsigned int){0, 0, 0, PKTLEN_SHIFT}); +- const vector unsigned int len0 = vec_sl( +- vec_ld(0, (vector unsigned int *)&descs[0]), +- (vector unsigned int){0, 0, 0, PKTLEN_SHIFT}); ++ const __vector unsigned int len1 = vec_sl( ++ vec_ld(0, (__vector unsigned int *)&descs[1]), ++ (__vector unsigned int){0, 0, 0, PKTLEN_SHIFT}); ++ const __vector unsigned int len0 = vec_sl( ++ vec_ld(0, (__vector unsigned int *)&descs[0]), ++ (__vector unsigned int){0, 0, 0, PKTLEN_SHIFT}); + + /* merge the now-aligned packet length fields back in */ +- descs[1] = (vector unsigned long)len1; +- descs[0] = (vector unsigned long)len0; ++ descs[1] = (__vector unsigned long)len1; ++ descs[0] = (__vector unsigned long)len0; + + /* D.1 pkt 1,2 convert format from desc to pktmbuf */ +- pkt_mb2 = vec_perm((vector unsigned char)descs[1], +- (vector unsigned char){}, shuf_msk); +- pkt_mb1 = vec_perm((vector unsigned char)descs[0], +- (vector unsigned char){}, shuf_msk); ++ pkt_mb2 = vec_perm((__vector unsigned char)descs[1], ++ (__vector unsigned char){}, shuf_msk); ++ pkt_mb1 = vec_perm((__vector unsigned char)descs[0], ++ (__vector unsigned char){}, shuf_msk); + + /* C.2 get 4 pkts staterr value */ +- staterr = (vector unsigned short)vec_mergeh( ++ staterr = (__vector unsigned short)vec_mergeh( + sterr_tmp1, sterr_tmp2); + + /* D.3 copy final 3,4 data to rx_pkts */ + vec_st(pkt_mb4, 0, +- (vector unsigned char *)&rx_pkts[pos + 3] ++ (__vector unsigned char *)&rx_pkts[pos + 3] + ->rx_descriptor_fields1 + ); + vec_st(pkt_mb3, 0, +- (vector unsigned char *)&rx_pkts[pos + 2] ++ (__vector unsigned char *)&rx_pkts[pos + 2] + ->rx_descriptor_fields1 + ); + + /* D.2 pkt 1,2 set in_port/nb_seg and remove crc */ +- pkt_mb2 = (vector unsigned char)vec_sub( +- (vector unsigned short)pkt_mb2, crc_adjust); +- pkt_mb1 = (vector unsigned char)vec_sub( +- (vector unsigned short)pkt_mb1, crc_adjust); ++ pkt_mb2 = (__vector unsigned char)vec_sub( ++ (__vector unsigned short)pkt_mb2, crc_adjust); ++ pkt_mb1 = (__vector unsigned char)vec_sub( ++ (__vector unsigned short)pkt_mb1, crc_adjust); + + /* C* extract and record EOP bit */ + if (split_packet) { +- vector unsigned char eop_shuf_mask = +- (vector unsigned char){ ++ __vector unsigned char eop_shuf_mask = ++ (__vector unsigned char){ + 0xFF, 0xFF, 0xFF, 0xFF, + 0xFF, 0xFF, 0xFF, 0xFF, + 0xFF, 0xFF, 0xFF, 0xFF, +@@ -394,19 +394,19 @@ _recv_raw_pkts_vec(struct i40e_rx_queue *rxq, struct rte_mbuf **rx_pkts, + }; + + /* and with mask to extract bits, flipping 1-0 */ +- vector unsigned char eop_bits = vec_and( +- (vector unsigned char)vec_nor(staterr, staterr), +- (vector unsigned char)eop_check); ++ __vector unsigned char eop_bits = vec_and( ++ (__vector unsigned char)vec_nor(staterr, staterr), ++ (__vector unsigned char)eop_check); + /* the staterr values are not in order, as the count + * of dd bits doesn't care. However, for end of + * packet tracking, we do care, so shuffle. This also + * compresses the 32-bit values to 8-bit + */ +- eop_bits = vec_perm(eop_bits, (vector unsigned char){}, ++ eop_bits = vec_perm(eop_bits, (__vector unsigned char){}, + eop_shuf_mask); + /* store the resulting 32-bit value */ + *split_packet = (vec_ld(0, +- (vector unsigned int *)&eop_bits))[0]; ++ (__vector unsigned int *)&eop_bits))[0]; + split_packet += RTE_I40E_DESCS_PER_LOOP; + + /* zero-out next pointers */ +@@ -417,22 +417,22 @@ _recv_raw_pkts_vec(struct i40e_rx_queue *rxq, struct rte_mbuf **rx_pkts, + } + + /* C.3 calc available number of desc */ +- staterr = vec_and(staterr, (vector unsigned short)dd_check); ++ staterr = vec_and(staterr, (__vector unsigned short)dd_check); + + /* D.3 copy final 1,2 data to rx_pkts */ + vec_st(pkt_mb2, 0, +- (vector unsigned char *)&rx_pkts[pos + 1] ++ (__vector unsigned char *)&rx_pkts[pos + 1] + ->rx_descriptor_fields1 + ); + vec_st(pkt_mb1, 0, +- (vector unsigned char *)&rx_pkts[pos]->rx_descriptor_fields1 ++ (__vector unsigned char *)&rx_pkts[pos]->rx_descriptor_fields1 + ); + desc_to_ptype_v(descs, &rx_pkts[pos], ptype_tbl); + desc_to_olflags_v(descs, &rx_pkts[pos]); + +- /* C.4 calc avaialbe number of desc */ ++ /* C.4 calc available number of desc */ + var = __builtin_popcountll((vec_ld(0, +- (vector unsigned long *)&staterr)[0])); ++ (__vector unsigned long *)&staterr)[0])); + nb_pkts_recd += var; + if (likely(var != RTE_I40E_DESCS_PER_LOOP)) + break; +@@ -533,9 +533,9 @@ vtx1(volatile struct i40e_tx_desc *txdp, + ((uint64_t)flags << I40E_TXD_QW1_CMD_SHIFT) | + ((uint64_t)pkt->data_len << I40E_TXD_QW1_TX_BUF_SZ_SHIFT)); + +- vector unsigned long descriptor = (vector unsigned long){ ++ __vector unsigned long descriptor = (__vector unsigned long){ + pkt->buf_iova + pkt->data_off, high_qw}; +- *(vector unsigned long *)txdp = descriptor; ++ *(__vector unsigned long *)txdp = descriptor; + } - /* Remove dropped flow exporters, if any needs to be removed. */ -- HMAP_FOR_EACH_SAFE (node, next, node, &di->flow_exporter_map) { -+ HMAP_FOR_EACH_SAFE (node, node, &di->flow_exporter_map) { - /* This is slow but doesn't take any extra memory, and - * this table is not supposed to contain many rows anyway. */ - options = (struct ofproto_ipfix_flow_exporter_options *) -@@ -1117,10 +1132,12 @@ dpif_ipfix_set_options( - } - if (i == n_flow_exporters_options) { /* Not found. */ - remove_flow_exporter(di, node); -+ feo_changed = true; - } - } + static inline void +diff --git a/dpdk/drivers/net/i40e/i40e_rxtx_vec_neon.c b/dpdk/drivers/net/i40e/i40e_rxtx_vec_neon.c +index b951ea2dc3..507468531f 100644 +--- a/dpdk/drivers/net/i40e/i40e_rxtx_vec_neon.c ++++ b/dpdk/drivers/net/i40e/i40e_rxtx_vec_neon.c +@@ -151,7 +151,7 @@ desc_to_olflags_v(struct i40e_rx_queue *rxq, uint64x2_t descs[4], + vreinterpretq_u8_u32(l3_l4e))); + /* then we shift left 1 bit */ + l3_l4e = vshlq_n_u32(l3_l4e, 1); +- /* we need to mask out the reduntant bits */ ++ /* we need to mask out the redundant bits */ + l3_l4e = vandq_u32(l3_l4e, cksum_mask); + + vlan0 = vorrq_u32(vlan0, rss); +@@ -416,7 +416,7 @@ _recv_raw_pkts_vec(struct i40e_rx_queue *__rte_restrict rxq, + I40E_UINT16_BIT - 1)); + stat = ~vgetq_lane_u64(vreinterpretq_u64_u16(staterr), 0); + +- /* C.4 calc avaialbe number of desc */ ++ /* C.4 calc available number of desc */ + if (unlikely(stat == 0)) { + nb_pkts_recd += RTE_I40E_DESCS_PER_LOOP; + } else { +diff --git a/dpdk/drivers/net/i40e/i40e_rxtx_vec_sse.c b/dpdk/drivers/net/i40e/i40e_rxtx_vec_sse.c +index 497b2404c6..3782e8052f 100644 +--- a/dpdk/drivers/net/i40e/i40e_rxtx_vec_sse.c ++++ b/dpdk/drivers/net/i40e/i40e_rxtx_vec_sse.c +@@ -282,7 +282,7 @@ desc_to_olflags_v(struct i40e_rx_queue *rxq, volatile union i40e_rx_desc *rxdp, + l3_l4e = _mm_shuffle_epi8(l3_l4e_flags, l3_l4e); + /* then we shift left 1 bit */ + l3_l4e = _mm_slli_epi32(l3_l4e, 1); +- /* we need to mask out the reduntant bits */ ++ /* we need to mask out the redundant bits */ + l3_l4e = _mm_and_si128(l3_l4e, cksum_mask); + + vlan0 = _mm_or_si128(vlan0, rss); +@@ -297,7 +297,7 @@ desc_to_olflags_v(struct i40e_rx_queue *rxq, volatile union i40e_rx_desc *rxdp, + __m128i v_fdir_ol_flags = descs_to_fdir_16b(desc_fltstat, + descs, rx_pkts); + #endif +- /* OR in ol_flag bits after descriptor speicific extraction */ ++ /* OR in ol_flag bits after descriptor specific extraction */ + vlan0 = _mm_or_si128(vlan0, v_fdir_ol_flags); + } - ovs_mutex_unlock(&mutex); -+ return beo_changed || feo_changed; - } +@@ -577,7 +577,7 @@ _recv_raw_pkts_vec(struct i40e_rx_queue *rxq, struct rte_mbuf **rx_pkts, + _mm_storeu_si128((void *)&rx_pkts[pos]->rx_descriptor_fields1, + pkt_mb1); + desc_to_ptype_v(descs, &rx_pkts[pos], ptype_tbl); +- /* C.4 calc avaialbe number of desc */ ++ /* C.4 calc available number of desc */ + var = __builtin_popcountll(_mm_cvtsi128_si64(staterr)); + nb_pkts_recd += var; + if (likely(var != RTE_I40E_DESCS_PER_LOOP)) +diff --git a/dpdk/drivers/net/i40e/rte_pmd_i40e.c b/dpdk/drivers/net/i40e/rte_pmd_i40e.c +index a492959b75..35829a1eea 100644 +--- a/dpdk/drivers/net/i40e/rte_pmd_i40e.c ++++ b/dpdk/drivers/net/i40e/rte_pmd_i40e.c +@@ -1427,7 +1427,7 @@ rte_pmd_i40e_set_tc_strict_prio(uint16_t port, uint8_t tc_map) + /* Get all TCs' bandwidth. */ + for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) { + if (veb->enabled_tc & BIT_ULL(i)) { +- /* For rubust, if bandwidth is 0, use 1 instead. */ ++ /* For robust, if bandwidth is 0, use 1 instead. */ + if (veb->bw_info.bw_ets_share_credits[i]) + ets_data.tc_bw_share_credits[i] = + veb->bw_info.bw_ets_share_credits[i]; +diff --git a/dpdk/drivers/net/iavf/iavf.h b/dpdk/drivers/net/iavf/iavf.h +index 0bb5698583..29692e3994 100644 +--- a/dpdk/drivers/net/iavf/iavf.h ++++ b/dpdk/drivers/net/iavf/iavf.h +@@ -18,7 +18,7 @@ + + #define IAVF_AQ_LEN 32 + #define IAVF_AQ_BUF_SZ 4096 +-#define IAVF_RESET_WAIT_CNT 50 ++#define IAVF_RESET_WAIT_CNT 500 + #define IAVF_BUF_SIZE_MIN 1024 + #define IAVF_FRAME_SIZE_MAX 9728 + #define IAVF_QUEUE_BASE_ADDR_UNIT 128 +@@ -296,6 +296,7 @@ struct iavf_adapter { + bool tx_vec_allowed; + uint32_t ptype_tbl[IAVF_MAX_PKT_TYPE] __rte_cache_min_aligned; + bool stopped; ++ bool closed; + uint16_t fdir_ref_cnt; + struct iavf_devargs devargs; + }; +diff --git a/dpdk/drivers/net/iavf/iavf_ethdev.c b/dpdk/drivers/net/iavf/iavf_ethdev.c +index 377d7bc7a6..f835457e4f 100644 +--- a/dpdk/drivers/net/iavf/iavf_ethdev.c ++++ b/dpdk/drivers/net/iavf/iavf_ethdev.c +@@ -229,9 +229,15 @@ static const struct eth_dev_ops iavf_eth_dev_ops = { + }; - struct dpif_ipfix * -@@ -1215,7 +1232,7 @@ static void - dpif_ipfix_clear(struct dpif_ipfix *di) OVS_REQUIRES(mutex) + static int +-iavf_tm_ops_get(struct rte_eth_dev *dev __rte_unused, ++iavf_tm_ops_get(struct rte_eth_dev *dev, + void *arg) { - struct dpif_ipfix_flow_exporter_map_node *exp_node; -- struct dpif_ipfix_port *dip, *next; -+ struct dpif_ipfix_port *dip; - - dpif_ipfix_bridge_exporter_clear(&di->bridge_exporter); ++ struct iavf_adapter *adapter = ++ IAVF_DEV_PRIVATE_TO_ADAPTER(dev->data->dev_private); ++ ++ if (adapter->closed) ++ return -EIO; ++ + if (!arg) + return -EINVAL; -@@ -1224,7 +1241,7 @@ dpif_ipfix_clear(struct dpif_ipfix *di) OVS_REQUIRES(mutex) - free(exp_node); - } +@@ -342,6 +348,9 @@ iavf_set_mc_addr_list(struct rte_eth_dev *dev, + return -EINVAL; + } -- HMAP_FOR_EACH_SAFE (dip, next, hmap_node, &di->ports) { -+ HMAP_FOR_EACH_SAFE (dip, hmap_node, &di->ports) { - dpif_ipfix_del_port__(di, dip); - } - } -@@ -2799,7 +2816,7 @@ dpif_ipfix_cache_expire(struct dpif_ipfix_exporter *exporter, - bool forced_end, const uint64_t export_time_usec, - const uint32_t export_time_sec) - { -- struct ipfix_flow_cache_entry *entry, *next_entry; -+ struct ipfix_flow_cache_entry *entry; - uint64_t max_flow_start_timestamp_usec; - bool template_msg_sent = false; - enum ipfix_flow_end_reason flow_end_reason; -@@ -2811,7 +2828,7 @@ dpif_ipfix_cache_expire(struct dpif_ipfix_exporter *exporter, - max_flow_start_timestamp_usec = export_time_usec - - 1000000LL * exporter->cache_active_timeout; ++ if (adapter->closed) ++ return -EIO; ++ + /* flush previous addresses */ + err = iavf_add_del_mc_addr_list(adapter, vf->mc_addrs, vf->mc_addrs_num, + false); +@@ -516,7 +525,7 @@ iavf_init_rss(struct iavf_adapter *adapter) + j = 0; + vf->rss_lut[i] = j; + } +- /* send virtchnnl ops to configure rss*/ ++ /* send virtchnl ops to configure RSS */ + ret = iavf_configure_rss_lut(adapter); + if (ret) + return ret; +@@ -613,6 +622,9 @@ iavf_dev_configure(struct rte_eth_dev *dev) + dev->data->nb_tx_queues); + int ret; + ++ if (ad->closed) ++ return -EIO; ++ + ad->rx_bulk_alloc_allowed = true; + /* Initialize to TRUE. If any of Rx queues doesn't meet the + * vector Rx/Tx preconditions, it will be reset. +@@ -831,7 +843,7 @@ static int iavf_config_rx_queues_irqs(struct rte_eth_dev *dev, + "vector %u are mapping to all Rx queues", + vf->msix_base); + } else { +- /* If Rx interrupt is reuquired, and we can use ++ /* If Rx interrupt is required, and we can use + * multi interrupts, then the vec is from 1 + */ + vf->nb_msix = +@@ -896,28 +908,38 @@ iavf_start_queues(struct rte_eth_dev *dev) + struct iavf_rx_queue *rxq; + struct iavf_tx_queue *txq; + int i; ++ uint16_t nb_txq, nb_rxq; + +- for (i = 0; i < dev->data->nb_tx_queues; i++) { +- txq = dev->data->tx_queues[i]; ++ for (nb_txq = 0; nb_txq < dev->data->nb_tx_queues; nb_txq++) { ++ txq = dev->data->tx_queues[nb_txq]; + if (txq->tx_deferred_start) + continue; +- if (iavf_dev_tx_queue_start(dev, i) != 0) { +- PMD_DRV_LOG(ERR, "Fail to start queue %u", i); +- return -1; ++ if (iavf_dev_tx_queue_start(dev, nb_txq) != 0) { ++ PMD_DRV_LOG(ERR, "Fail to start tx queue %u", nb_txq); ++ goto tx_err; + } + } -- LIST_FOR_EACH_SAFE (entry, next_entry, cache_flow_start_timestamp_list_node, -+ LIST_FOR_EACH_SAFE (entry, cache_flow_start_timestamp_list_node, - &exporter->cache_flow_start_timestamp_list) { - if (forced_end) { - flow_end_reason = FORCED_END; -diff --git a/ofproto/ofproto-dpif-ipfix.h b/ofproto/ofproto-dpif-ipfix.h -index 1f42cd5275..75c0ab81ac 100644 ---- a/ofproto/ofproto-dpif-ipfix.h -+++ b/ofproto/ofproto-dpif-ipfix.h -@@ -48,7 +48,7 @@ bool dpif_ipfix_get_bridge_exporter_output_sampling(const struct dpif_ipfix *); - bool dpif_ipfix_get_flow_exporter_tunnel_sampling(const struct dpif_ipfix *, - const uint32_t); - bool dpif_ipfix_is_tunnel_port(const struct dpif_ipfix *, odp_port_t); --void dpif_ipfix_set_options( -+bool dpif_ipfix_set_options( - struct dpif_ipfix *, - const struct ofproto_ipfix_bridge_exporter_options *, - const struct ofproto_ipfix_flow_exporter_options *, size_t); -diff --git a/ofproto/ofproto-dpif-sflow.c b/ofproto/ofproto-dpif-sflow.c -index 30e7caf54a..e8e1de920b 100644 ---- a/ofproto/ofproto-dpif-sflow.c -+++ b/ofproto/ofproto-dpif-sflow.c -@@ -468,7 +468,8 @@ sflow_choose_agent_address(const char *agent_device, - const char *target; - SSET_FOR_EACH (target, targets) { - struct sockaddr_storage ss; -- if (inet_parse_active(target, SFL_DEFAULT_COLLECTOR_PORT, &ss, true)) { -+ if (inet_parse_active(target, SFL_DEFAULT_COLLECTOR_PORT, -+ &ss, true, NULL)) { - /* sFlow only supports target in default routing table with - * packet mark zero. - */ -@@ -590,10 +591,10 @@ void - dpif_sflow_unref(struct dpif_sflow *ds) OVS_EXCLUDED(mutex) - { - if (ds && ovs_refcount_unref_relaxed(&ds->ref_cnt) == 1) { -- struct dpif_sflow_port *dsp, *next; -+ struct dpif_sflow_port *dsp; +- for (i = 0; i < dev->data->nb_rx_queues; i++) { +- rxq = dev->data->rx_queues[i]; ++ for (nb_rxq = 0; nb_rxq < dev->data->nb_rx_queues; nb_rxq++) { ++ rxq = dev->data->rx_queues[nb_rxq]; + if (rxq->rx_deferred_start) + continue; +- if (iavf_dev_rx_queue_start(dev, i) != 0) { +- PMD_DRV_LOG(ERR, "Fail to start queue %u", i); +- return -1; ++ if (iavf_dev_rx_queue_start(dev, nb_rxq) != 0) { ++ PMD_DRV_LOG(ERR, "Fail to start rx queue %u", nb_rxq); ++ goto rx_err; + } + } - dpif_sflow_clear(ds); -- HMAP_FOR_EACH_SAFE (dsp, next, hmap_node, &ds->ports) { -+ HMAP_FOR_EACH_SAFE (dsp, hmap_node, &ds->ports) { - dpif_sflow_del_port__(ds, dsp); - } - hmap_destroy(&ds->ports); -diff --git a/ofproto/ofproto-dpif-trace.c b/ofproto/ofproto-dpif-trace.c -index 78a54c715d..109940ad2a 100644 ---- a/ofproto/ofproto-dpif-trace.c -+++ b/ofproto/ofproto-dpif-trace.c -@@ -65,8 +65,8 @@ static void - oftrace_node_list_destroy(struct ovs_list *nodes) - { - if (nodes) { -- struct oftrace_node *node, *next; -- LIST_FOR_EACH_SAFE (node, next, node, nodes) { -+ struct oftrace_node *node; -+ LIST_FOR_EACH_SAFE (node, node, nodes) { - ovs_list_remove(&node->node); - oftrace_node_destroy(node); - } -diff --git a/ofproto/ofproto-dpif-xlate-cache.c b/ofproto/ofproto-dpif-xlate-cache.c -index dcc91cb380..9224ee2e6d 100644 ---- a/ofproto/ofproto-dpif-xlate-cache.c -+++ b/ofproto/ofproto-dpif-xlate-cache.c -@@ -209,6 +209,7 @@ xlate_cache_clear_entry(struct xc_entry *entry) - { - switch (entry->type) { - case XC_TABLE: -+ ofproto_unref(&(entry->table.ofproto->up)); - break; - case XC_RULE: - ofproto_rule_unref(&entry->rule->up); -@@ -231,6 +232,7 @@ xlate_cache_clear_entry(struct xc_entry *entry) - free(entry->learn.ofm); - break; - case XC_NORMAL: -+ ofproto_unref(&(entry->normal.ofproto->up)); - break; - case XC_FIN_TIMEOUT: - /* 'u.fin.rule' is always already held as a XC_RULE, which -diff --git a/ofproto/ofproto-dpif-xlate-cache.h b/ofproto/ofproto-dpif-xlate-cache.h -index 114aff8ea3..0fc6d2ea60 100644 ---- a/ofproto/ofproto-dpif-xlate-cache.h -+++ b/ofproto/ofproto-dpif-xlate-cache.h -@@ -61,9 +61,8 @@ enum xc_type { - * that a flow relates to, although they may be used for other effects as well - * (for instance, refreshing hard timeouts for learned flows). - * -- * An explicit reference is taken to all pointers other than the ones for -- * struct ofproto_dpif. ofproto_dpif pointers are explicitly protected by -- * destroying all xlate caches before the ofproto is destroyed. */ -+ * An explicit reference is taken to all pointers. -+ */ - struct xc_entry { - enum xc_type type; - union { -diff --git a/ofproto/ofproto-dpif-xlate.c b/ofproto/ofproto-dpif-xlate.c -index 578cbfe581..7716c22f49 100644 ---- a/ofproto/ofproto-dpif-xlate.c -+++ b/ofproto/ofproto-dpif-xlate.c -@@ -865,7 +865,7 @@ xlate_xbridge_init(struct xlate_cfg *xcfg, struct xbridge *xbridge) - ovs_list_init(&xbridge->xbundles); - hmap_init(&xbridge->xports); - hmap_insert(&xcfg->xbridges, &xbridge->hmap_node, -- hash_pointer(xbridge->ofproto, 0)); -+ uuid_hash(&xbridge->ofproto->uuid)); + return 0; ++ ++rx_err: ++ for (i = 0; i < nb_rxq; i++) ++ iavf_dev_rx_queue_stop(dev, i); ++tx_err: ++ for (i = 0; i < nb_txq; i++) ++ iavf_dev_tx_queue_stop(dev, i); ++ ++ return -1; } - static void -@@ -1222,13 +1222,13 @@ xlate_txn_start(void) - static void - xlate_xcfg_free(struct xlate_cfg *xcfg) - { -- struct xbridge *xbridge, *next_xbridge; -+ struct xbridge *xbridge; + static int +@@ -932,6 +954,9 @@ iavf_dev_start(struct rte_eth_dev *dev) - if (!xcfg) { - return; - } + PMD_INIT_FUNC_TRACE(); -- HMAP_FOR_EACH_SAFE (xbridge, next_xbridge, hmap_node, &xcfg->xbridges) { -+ HMAP_FOR_EACH_SAFE (xbridge, hmap_node, &xcfg->xbridges) { - xlate_xbridge_remove(xcfg, xbridge); - } ++ if (adapter->closed) ++ return -1; ++ + adapter->stopped = 0; -@@ -1282,18 +1282,18 @@ xlate_ofproto_set(struct ofproto_dpif *ofproto, const char *name, - static void - xlate_xbridge_remove(struct xlate_cfg *xcfg, struct xbridge *xbridge) - { -- struct xbundle *xbundle, *next_xbundle; -- struct xport *xport, *next_xport; -+ struct xbundle *xbundle; -+ struct xport *xport; + vf->max_pkt_len = dev->data->mtu + IAVF_ETH_OVERHEAD; +@@ -1009,6 +1034,9 @@ iavf_dev_stop(struct rte_eth_dev *dev) - if (!xbridge) { - return; - } + PMD_INIT_FUNC_TRACE(); -- HMAP_FOR_EACH_SAFE (xport, next_xport, ofp_node, &xbridge->xports) { -+ HMAP_FOR_EACH_SAFE (xport, ofp_node, &xbridge->xports) { - xlate_xport_remove(xcfg, xport); - } ++ if (adapter->closed) ++ return -1; ++ + if (!(vf->vf_res->vf_cap_flags & VIRTCHNL_VF_OFFLOAD_WB_ON_ITR) && + dev->data->dev_conf.intr_conf.rxq != 0) + rte_intr_disable(intr_handle); +@@ -1030,9 +1058,6 @@ iavf_dev_stop(struct rte_eth_dev *dev) + iavf_add_del_mc_addr_list(adapter, vf->mc_addrs, vf->mc_addrs_num, + false); -- LIST_FOR_EACH_SAFE (xbundle, next_xbundle, list_node, &xbridge->xbundles) { -+ LIST_FOR_EACH_SAFE (xbundle, list_node, &xbridge->xbundles) { - xlate_xbundle_remove(xcfg, xbundle); - } +- /* free iAVF security device context all related resources */ +- iavf_security_ctx_destroy(adapter); +- + adapter->stopped = 1; + dev->data->dev_started = 0; -@@ -1639,7 +1639,7 @@ xbridge_lookup(struct xlate_cfg *xcfg, const struct ofproto_dpif *ofproto) +@@ -1046,6 +1071,9 @@ iavf_dev_info_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info) + IAVF_DEV_PRIVATE_TO_ADAPTER(dev->data->dev_private); + struct iavf_info *vf = &adapter->vf; - xbridges = &xcfg->xbridges; ++ if (adapter->closed) ++ return -EIO; ++ + dev_info->max_rx_queues = IAVF_MAX_NUM_QUEUES_LV; + dev_info->max_tx_queues = IAVF_MAX_NUM_QUEUES_LV; + dev_info->min_rx_bufsize = IAVF_BUF_SIZE_MIN; +@@ -1286,6 +1314,9 @@ iavf_dev_vlan_filter_set(struct rte_eth_dev *dev, uint16_t vlan_id, int on) + struct iavf_info *vf = IAVF_DEV_PRIVATE_TO_VF(adapter); + int err; -- HMAP_FOR_EACH_IN_BUCKET (xbridge, hmap_node, hash_pointer(ofproto, 0), -+ HMAP_FOR_EACH_IN_BUCKET (xbridge, hmap_node, uuid_hash(&ofproto->uuid), - xbridges) { - if (xbridge->ofproto == ofproto) { - return xbridge; -@@ -1661,6 +1661,23 @@ xbridge_lookup_by_uuid(struct xlate_cfg *xcfg, const struct uuid *uuid) - return NULL; - } ++ if (adapter->closed) ++ return -EIO; ++ + if (vf->vf_res->vf_cap_flags & VIRTCHNL_VF_OFFLOAD_VLAN_V2) { + err = iavf_add_del_vlan_v2(adapter, vlan_id, on); + if (err) +@@ -1362,6 +1393,9 @@ iavf_dev_vlan_offload_set(struct rte_eth_dev *dev, int mask) + struct rte_eth_conf *dev_conf = &dev->data->dev_conf; + int err; -+struct ofproto_dpif * -+xlate_ofproto_lookup(const struct uuid *uuid) -+{ -+ struct xlate_cfg *xcfg = ovsrcu_get(struct xlate_cfg *, &xcfgp); -+ struct xbridge *xbridge; ++ if (adapter->closed) ++ return -EIO; + -+ if (!xcfg) { -+ return NULL; -+ } + if (vf->vf_res->vf_cap_flags & VIRTCHNL_VF_OFFLOAD_VLAN_V2) + return iavf_dev_vlan_offload_set_v2(dev, mask); + +@@ -1394,6 +1428,9 @@ iavf_dev_rss_reta_update(struct rte_eth_dev *dev, + uint16_t i, idx, shift; + int ret; + ++ if (adapter->closed) ++ return -EIO; + -+ xbridge = xbridge_lookup_by_uuid(xcfg, uuid); -+ if (xbridge != NULL) { -+ return xbridge->ofproto; -+ } -+ return NULL; -+} -+ - static struct xbundle * - xbundle_lookup(struct xlate_cfg *xcfg, const struct ofbundle *ofbundle) - { -@@ -2125,9 +2142,14 @@ mirror_packet(struct xlate_ctx *ctx, struct xbundle *xbundle, - int snaplen; - - /* Get the details of the mirror represented by the rightmost 1-bit. */ -- ovs_assert(mirror_get(xbridge->mbridge, raw_ctz(mirrors), -- &vlans, &dup_mirrors, -- &out, &snaplen, &out_vlan)); -+ if (OVS_UNLIKELY(!mirror_get(xbridge->mbridge, raw_ctz(mirrors), -+ &vlans, &dup_mirrors, -+ &out, &snaplen, &out_vlan))) { -+ /* The mirror got reconfigured before we got to read it's -+ * configuration. */ -+ mirrors = zero_rightmost_1bit(mirrors); -+ continue; -+ } + if (!(vf->vf_res->vf_cap_flags & VIRTCHNL_VF_OFFLOAD_RSS_PF)) + return -ENOTSUP; +@@ -1420,7 +1457,7 @@ iavf_dev_rss_reta_update(struct rte_eth_dev *dev, + } - /* If this mirror selects on the basis of VLAN, and it does not select -@@ -3015,7 +3037,7 @@ xlate_normal(struct xlate_ctx *ctx) - bool is_grat_arp = is_gratuitous_arp(flow, wc); - if (ctx->xin->allow_side_effects - && flow->packet_type == htonl(PT_ETH) -- && in_port->pt_mode != NETDEV_PT_LEGACY_L3 -+ && in_port && in_port->pt_mode != NETDEV_PT_LEGACY_L3 - ) { - update_learning_table(ctx, in_xbundle, flow->dl_src, vlan, - is_grat_arp); -@@ -3024,12 +3046,14 @@ xlate_normal(struct xlate_ctx *ctx) - struct xc_entry *entry; + rte_memcpy(vf->rss_lut, lut, reta_size); +- /* send virtchnnl ops to configure rss*/ ++ /* send virtchnl ops to configure RSS */ + ret = iavf_configure_rss_lut(adapter); + if (ret) /* revert back */ + rte_memcpy(vf->rss_lut, lut, reta_size); +@@ -1439,6 +1476,9 @@ iavf_dev_rss_reta_query(struct rte_eth_dev *dev, + struct iavf_info *vf = IAVF_DEV_PRIVATE_TO_VF(adapter); + uint16_t i, idx, shift; - /* Save just enough info to update mac learning table later. */ -- entry = xlate_cache_add_entry(ctx->xin->xcache, XC_NORMAL); -- entry->normal.ofproto = ctx->xbridge->ofproto; -- entry->normal.in_port = flow->in_port.ofp_port; -- entry->normal.dl_src = flow->dl_src; -- entry->normal.vlan = vlan; -- entry->normal.is_gratuitous_arp = is_grat_arp; -+ if (ofproto_try_ref(&ctx->xbridge->ofproto->up)) { -+ entry = xlate_cache_add_entry(ctx->xin->xcache, XC_NORMAL); -+ entry->normal.ofproto = ctx->xbridge->ofproto; -+ entry->normal.in_port = flow->in_port.ofp_port; -+ entry->normal.dl_src = flow->dl_src; -+ entry->normal.vlan = vlan; -+ entry->normal.is_gratuitous_arp = is_grat_arp; -+ } - } ++ if (adapter->closed) ++ return -EIO; ++ + if (!(vf->vf_res->vf_cap_flags & VIRTCHNL_VF_OFFLOAD_RSS_PF)) + return -ENOTSUP; - /* Determine output bundle. */ -@@ -3048,7 +3072,6 @@ xlate_normal(struct xlate_ctx *ctx) - */ - ctx->xout->slow |= SLOW_ACTION; +@@ -1492,6 +1532,9 @@ iavf_dev_rss_hash_update(struct rte_eth_dev *dev, -- memset(&wc->masks.tp_src, 0xff, sizeof wc->masks.tp_src); - if (mcast_snooping_is_membership(flow->tp_src) || - mcast_snooping_is_query(flow->tp_src)) { - if (ctx->xin->allow_side_effects && ctx->xin->packet) { -@@ -3523,6 +3546,9 @@ propagate_tunnel_data_to_flow__(struct flow *dst_flow, - dst_flow->dl_dst = dmac; - dst_flow->dl_src = smac; + adapter->dev_data->dev_conf.rx_adv_conf.rss_conf = *rss_conf; -+ /* Clear VLAN entries which do not apply for tunnel flows. */ -+ memset(dst_flow->vlans, 0, sizeof dst_flow->vlans); ++ if (adapter->closed) ++ return -EIO; + - dst_flow->packet_type = htonl(PT_ETH); - dst_flow->nw_dst = src_flow->tunnel.ip_dst; - dst_flow->nw_src = src_flow->tunnel.ip_src; -@@ -3654,14 +3680,27 @@ native_tunnel_output(struct xlate_ctx *ctx, const struct xport *xport, + if (!(vf->vf_res->vf_cap_flags & VIRTCHNL_VF_OFFLOAD_RSS_PF)) + return -ENOTSUP; - err = tnl_neigh_lookup(out_dev->xbridge->name, &d_ip6, &dmac); - if (err) { -+ struct in6_addr nh_s_ip6 = in6addr_any; -+ - xlate_report(ctx, OFT_DETAIL, - "neighbor cache miss for %s on bridge %s, " - "sending %s request", - buf_dip6, out_dev->xbridge->name, d_ip ? "ARP" : "ND"); -+ -+ err = ovs_router_get_netdev_source_address(&d_ip6, -+ out_dev->xbridge->name, -+ &nh_s_ip6); -+ if (err) { -+ nh_s_ip6 = s_ip6; -+ } -+ - if (d_ip) { -- tnl_send_arp_request(ctx, out_dev, smac, s_ip, d_ip); -+ ovs_be32 nh_s_ip; +@@ -1545,6 +1588,9 @@ iavf_dev_rss_hash_conf_get(struct rte_eth_dev *dev, + IAVF_DEV_PRIVATE_TO_ADAPTER(dev->data->dev_private); + struct iavf_info *vf = IAVF_DEV_PRIVATE_TO_VF(adapter); + ++ if (adapter->closed) ++ return -EIO; + -+ nh_s_ip = in6_addr_get_mapped_ipv4(&nh_s_ip6); -+ tnl_send_arp_request(ctx, out_dev, smac, nh_s_ip, d_ip); - } else { -- tnl_send_nd_request(ctx, out_dev, smac, &s_ip6, &d_ip6); -+ tnl_send_nd_request(ctx, out_dev, smac, &nh_s_ip6, &d_ip6); - } - return err; - } -@@ -4176,6 +4215,10 @@ compose_output_action__(struct xlate_ctx *ctx, ofp_port_t ofp_port, - if (xport->pt_mode == NETDEV_PT_LEGACY_L3) { - flow->packet_type = PACKET_TYPE_BE(OFPHTN_ETHERTYPE, - ntohs(flow->dl_type)); -+ if (ctx->pending_encap) { -+ /* The Ethernet header was not actually added yet. */ -+ ctx->pending_encap = false; -+ } - } - } + if (!(vf->vf_res->vf_cap_flags & VIRTCHNL_VF_OFFLOAD_RSS_PF)) + return -ENOTSUP; -@@ -5622,7 +5665,8 @@ xlate_sample_action(struct xlate_ctx *ctx, +@@ -1792,6 +1838,9 @@ iavf_dev_rx_queue_intr_enable(struct rte_eth_dev *dev, uint16_t queue_id) + struct iavf_info *vf = IAVF_DEV_PRIVATE_TO_VF(adapter); + uint16_t msix_intr; - /* Scale the probability from 16-bit to 32-bit while representing - * the same percentage. */ -- uint32_t probability = (os->probability << 16) | os->probability; -+ uint32_t probability = -+ ((uint32_t) os->probability << 16) | os->probability; ++ if (adapter->closed) ++ return -EIO; ++ + msix_intr = rte_intr_vec_list_index_get(pci_dev->intr_handle, + queue_id); + if (msix_intr == IAVF_MISC_VEC_ID) { +@@ -1833,7 +1882,7 @@ iavf_dev_rx_queue_intr_disable(struct rte_eth_dev *dev, uint16_t queue_id) - /* If ofp_port in flow sample action is equel to ofp_port, - * this sample action is a input port action. */ -@@ -7784,6 +7828,12 @@ xlate_actions(struct xlate_in *xin, struct xlate_out *xout) - goto exit; - } + IAVF_WRITE_REG(hw, + IAVF_VFINT_DYN_CTLN1(msix_intr - IAVF_RX_VEC_START), +- 0); ++ IAVF_VFINT_DYN_CTLN1_WB_ON_ITR_MASK); -+ if (!xin->frozen_state -+ && xin->flow.ct_state -+ && xin->flow.ct_state & CS_TRACKED) { -+ ctx.conntracked = true; -+ } + IAVF_WRITE_FLUSH(hw); + return 0; +@@ -2412,8 +2461,11 @@ static int + iavf_dev_flow_ops_get(struct rte_eth_dev *dev, + const struct rte_flow_ops **ops) + { +- if (!dev) +- return -EINVAL; ++ struct iavf_adapter *adapter = ++ IAVF_DEV_PRIVATE_TO_ADAPTER(dev->data->dev_private); + - /* Tunnel metadata in udpif format must be normalized before translation. */ - if (flow->tunnel.flags & FLOW_TNL_F_UDPIF) { - const struct tun_table *tun_tab = ofproto_get_tun_tab( -diff --git a/ofproto/ofproto-dpif-xlate.h b/ofproto/ofproto-dpif-xlate.h -index 851088d794..2ba90e999c 100644 ---- a/ofproto/ofproto-dpif-xlate.h -+++ b/ofproto/ofproto-dpif-xlate.h -@@ -176,6 +176,7 @@ void xlate_ofproto_set(struct ofproto_dpif *, const char *name, struct dpif *, - bool forward_bpdu, bool has_in_band, - const struct dpif_backer_support *support); - void xlate_remove_ofproto(struct ofproto_dpif *); -+struct ofproto_dpif *xlate_ofproto_lookup(const struct uuid *uuid); ++ if (adapter->closed) ++ return -EIO; - void xlate_bundle_set(struct ofproto_dpif *, struct ofbundle *, - const char *name, enum port_vlan_mode, -diff --git a/ofproto/ofproto-dpif.c b/ofproto/ofproto-dpif.c -index 8143dd965f..f9562dee87 100644 ---- a/ofproto/ofproto-dpif.c -+++ b/ofproto/ofproto-dpif.c -@@ -215,10 +215,6 @@ struct shash all_dpif_backers = SHASH_INITIALIZER(&all_dpif_backers); - static struct hmap all_ofproto_dpifs_by_name = - HMAP_INITIALIZER(&all_ofproto_dpifs_by_name); + *ops = &iavf_flow_ops; + return 0; +@@ -2554,7 +2606,7 @@ iavf_dev_init(struct rte_eth_dev *eth_dev) --/* All existing ofproto_dpif instances, indexed by ->uuid. */ --static struct hmap all_ofproto_dpifs_by_uuid = -- HMAP_INITIALIZER(&all_ofproto_dpifs_by_uuid); + /* Start device watchdog */ + iavf_dev_watchdog_enable(adapter); - - static bool ofproto_use_tnl_push_pop = true; - static void ofproto_unixctl_init(void); - static void ct_zone_config_init(struct dpif_backer *backer); -@@ -1663,7 +1659,7 @@ static int - construct(struct ofproto *ofproto_) - { - struct ofproto_dpif *ofproto = ofproto_dpif_cast(ofproto_); -- struct shash_node *node, *next; -+ struct shash_node *node; - int error; - - /* Tunnel module can get used right after the udpif threads are running. */ -@@ -1701,7 +1697,7 @@ construct(struct ofproto *ofproto_) - ofproto->ams_seqno = seq_read(ofproto->ams_seq); ++ adapter->closed = false; + return 0; -- SHASH_FOR_EACH_SAFE (node, next, &init_ofp_ports) { -+ SHASH_FOR_EACH_SAFE (node, &init_ofp_ports) { - struct iface_hint *iface_hint = node->data; +@@ -2582,7 +2634,16 @@ iavf_dev_close(struct rte_eth_dev *dev) + if (rte_eal_process_type() != RTE_PROC_PRIMARY) + return 0; - if (!strcmp(iface_hint->br_name, ofproto->up.name)) { -@@ -1720,9 +1716,6 @@ construct(struct ofproto *ofproto_) - hmap_insert(&all_ofproto_dpifs_by_name, - &ofproto->all_ofproto_dpifs_by_name_node, - hash_string(ofproto->up.name, 0)); -- hmap_insert(&all_ofproto_dpifs_by_uuid, -- &ofproto->all_ofproto_dpifs_by_uuid_node, -- uuid_hash(&ofproto->uuid)); - memset(&ofproto->stats, 0, sizeof ofproto->stats); ++ if (adapter->closed) { ++ ret = 0; ++ goto out; ++ } ++ + ret = iavf_dev_stop(dev); ++ adapter->closed = true; ++ ++ /* free iAVF security device context all related resources */ ++ iavf_security_ctx_destroy(adapter); + + iavf_flow_flush(dev, NULL); + iavf_flow_uninit(adapter); +@@ -2636,6 +2697,7 @@ iavf_dev_close(struct rte_eth_dev *dev) + * the bus master bit will not be disabled, and this call will have no + * effect. + */ ++out: + if (vf->vf_reset && !rte_pci_set_bus_master(pci_dev, true)) + vf->vf_reset = false; + +diff --git a/dpdk/drivers/net/iavf/iavf_fdir.c b/dpdk/drivers/net/iavf/iavf_fdir.c +index b63aaca91d..6b847894d8 100644 +--- a/dpdk/drivers/net/iavf/iavf_fdir.c ++++ b/dpdk/drivers/net/iavf/iavf_fdir.c +@@ -1185,8 +1185,22 @@ iavf_fdir_parse_pattern(__rte_unused struct iavf_adapter *ad, + GTPU_DWN, QFI); + } + +- rte_memcpy(hdr->buffer, gtp_psc_spec, +- sizeof(*gtp_psc_spec)); ++ /* ++ * New structure to fix gap between kernel driver and ++ * rte_gtp_psc_generic_hdr. ++ */ ++ struct iavf_gtp_psc_spec_hdr { ++ uint8_t len; ++ uint8_t qfi:6; ++ uint8_t type:4; ++ uint8_t next; ++ } psc; ++ psc.len = gtp_psc_spec->hdr.ext_hdr_len; ++ psc.qfi = gtp_psc_spec->hdr.qfi; ++ psc.type = gtp_psc_spec->hdr.type; ++ psc.next = 0; ++ rte_memcpy(hdr->buffer, &psc, ++ sizeof(struct iavf_gtp_psc_spec_hdr)); + } + + hdrs->count = ++layer; +diff --git a/dpdk/drivers/net/iavf/iavf_ipsec_crypto.c b/dpdk/drivers/net/iavf/iavf_ipsec_crypto.c +index 884169e061..75f05ee558 100644 +--- a/dpdk/drivers/net/iavf/iavf_ipsec_crypto.c ++++ b/dpdk/drivers/net/iavf/iavf_ipsec_crypto.c +@@ -69,7 +69,7 @@ struct iavf_security_session { + * 16B - 3 + * + * but we also need the IV Length for TSO to correctly calculate the total +- * header length so placing it in the upper 6-bits here for easier reterival. ++ * header length so placing it in the upper 6-bits here for easier retrieval. + */ + static inline uint8_t + calc_ipsec_desc_iv_len_field(uint16_t iv_sz) +@@ -448,7 +448,7 @@ sa_add_set_auth_params(struct virtchnl_ipsec_crypto_cfg_item *cfg, + /** + * Send SA add virtual channel request to Inline IPsec driver. + * +- * Inline IPsec driver expects SPI and destination IP adderss to be in host ++ * Inline IPsec driver expects SPI and destination IP address to be in host + * order, but DPDK APIs are network order, therefore we need to do a htonl + * conversion of these parameters. + */ +@@ -614,7 +614,7 @@ set_session_parameter(struct iavf_security_ctx *iavf_sctx, + if (conf->crypto_xform->auth.algo == RTE_CRYPTO_AUTH_AES_GMAC) { + sess->block_sz = get_auth_blocksize(iavf_sctx, + conf->crypto_xform->auth.algo); +- sess->iv_sz = conf->crypto_xform->auth.iv.length; ++ sess->iv_sz = sizeof(uint64_t); /* iv len inc. salt */ + sess->icv_sz = conf->crypto_xform->auth.digest_length; + } else { + sess->block_sz = get_cipher_blocksize(iavf_sctx, +@@ -726,7 +726,7 @@ iavf_ipsec_crypto_action_valid(struct rte_eth_dev *ethdev, + /** + * Send virtual channel security policy add request to IES driver. + * +- * IES driver expects SPI and destination IP adderss to be in host ++ * IES driver expects SPI and destination IP address to be in host + * order, but DPDK APIs are network order, therefore we need to do a htonl + * conversion of these parameters. + */ +@@ -736,7 +736,9 @@ iavf_ipsec_crypto_inbound_security_policy_add(struct iavf_adapter *adapter, + uint8_t is_v4, + rte_be32_t v4_dst_addr, + uint8_t *v6_dst_addr, +- uint8_t drop) ++ uint8_t drop, ++ bool is_udp, ++ uint16_t udp_port) + { + struct inline_ipsec_msg *request = NULL, *response = NULL; + size_t request_len, response_len; +@@ -781,6 +783,8 @@ iavf_ipsec_crypto_inbound_security_policy_add(struct iavf_adapter *adapter, + /** Traffic Class/Congestion Domain currently not support */ + request->ipsec_data.sp_cfg->set_tc = 0; + request->ipsec_data.sp_cfg->cgd = 0; ++ request->ipsec_data.sp_cfg->is_udp = is_udp; ++ request->ipsec_data.sp_cfg->udp_port = htons(udp_port); + + response_len = sizeof(struct inline_ipsec_msg) + + sizeof(struct virtchnl_ipsec_sp_cfg_resp); +@@ -994,7 +998,7 @@ iavf_ipsec_crypto_sa_del(struct iavf_adapter *adapter, + request->req_id = (uint16_t)0xDEADBEEF; + + /** +- * SA delete supports deletetion of 1-8 specified SA's or if the flag ++ * SA delete supports deletion of 1-8 specified SA's or if the flag + * field is zero, all SA's associated with VF will be deleted. + */ + if (sess) { +@@ -1114,11 +1118,14 @@ iavf_ipsec_crypto_compute_l4_payload_length(struct rte_mbuf *m, + * ipv4/6 hdr + ext hdrs + */ + +- if (s->udp_encap.enabled) ++ if (s->udp_encap.enabled) { + ol4_len = sizeof(struct rte_udp_hdr); +- +- l3_len = m->l3_len; +- l4_len = m->l4_len; ++ l3_len = m->l3_len - ol4_len; ++ l4_len = l3_len; ++ } else { ++ l3_len = m->l3_len; ++ l4_len = m->l4_len; ++ } + + return rte_pktmbuf_pkt_len(m) - (ol2_len + ol3_len + ol4_len + + esp_hlen + l3_len + l4_len + esp_tlen); +@@ -1147,7 +1154,7 @@ iavf_ipsec_crypto_pkt_metadata_set(void *device, + md = RTE_MBUF_DYNFIELD(m, iavf_sctx->pkt_md_offset, + struct iavf_ipsec_crypto_pkt_metadata *); + +- /* Set immutatable metadata values from session template */ ++ /* Set immutable metadata values from session template */ + memcpy(md, &iavf_sess->pkt_metadata_template, + sizeof(struct iavf_ipsec_crypto_pkt_metadata)); + +@@ -1352,10 +1359,12 @@ iavf_ipsec_crypto_set_security_capabililites(struct iavf_security_ctx + capabilities = rte_zmalloc("crypto_cap", + sizeof(struct rte_cryptodev_capabilities) * + (number_of_capabilities + 1), 0); ++ if (!capabilities) ++ return -ENOMEM; + capabilities[number_of_capabilities].op = RTE_CRYPTO_OP_TYPE_UNDEFINED; + + /** +- * Iterate over each virtchl crypto capability by crypto type and ++ * Iterate over each virtchnl crypto capability by crypto type and + * algorithm. + */ + for (i = 0; i < VIRTCHNL_IPSEC_MAX_CRYPTO_CAP_NUM; i++) { +@@ -1454,7 +1463,7 @@ iavf_ipsec_crypto_capabilities_get(void *device) + /** + * Update the security capabilities struct with the runtime discovered + * crypto capabilities, except for last element of the array which is +- * the null terminatation ++ * the null termination + */ + for (i = 0; i < ((sizeof(iavf_security_capabilities) / + sizeof(iavf_security_capabilities[0])) - 1); i++) { +@@ -1545,29 +1554,90 @@ iavf_security_ctx_destroy(struct iavf_adapter *adapter) + if (iavf_sctx == NULL) + return -ENODEV; + +- /* TODO: Add resources cleanup */ +- + /* free and reset security data structures */ + rte_free(iavf_sctx); + rte_free(sctx); - ofproto_init_tables(ofproto_, N_TABLES); -@@ -1820,8 +1813,6 @@ destruct(struct ofproto *ofproto_, bool del) +- iavf_sctx = NULL; +- sctx = NULL; ++ adapter->security_ctx = NULL; ++ adapter->vf.eth_dev->security_ctx = NULL; - hmap_remove(&all_ofproto_dpifs_by_name, - &ofproto->all_ofproto_dpifs_by_name_node); -- hmap_remove(&all_ofproto_dpifs_by_uuid, -- &ofproto->all_ofproto_dpifs_by_uuid_node); + return 0; + } - OFPROTO_FOR_EACH_TABLE (table, &ofproto->up) { - CLS_FOR_EACH (rule, up.cr, &table->cls) { -@@ -1857,6 +1848,8 @@ destruct(struct ofproto *ofproto_, bool del) ++static int ++iavf_ipsec_crypto_status_get(struct iavf_adapter *adapter, ++ struct virtchnl_ipsec_status *status) ++{ ++ /* Perform pf-vf comms */ ++ struct inline_ipsec_msg *request = NULL, *response = NULL; ++ size_t request_len, response_len; ++ int rc; ++ ++ request_len = sizeof(struct inline_ipsec_msg); ++ ++ request = rte_malloc("iavf-device-status-request", request_len, 0); ++ if (request == NULL) { ++ rc = -ENOMEM; ++ goto update_cleanup; ++ } ++ ++ response_len = sizeof(struct inline_ipsec_msg) + ++ sizeof(struct virtchnl_ipsec_cap); ++ response = rte_malloc("iavf-device-status-response", ++ response_len, 0); ++ if (response == NULL) { ++ rc = -ENOMEM; ++ goto update_cleanup; ++ } ++ ++ /* set msg header params */ ++ request->ipsec_opcode = INLINE_IPSEC_OP_GET_STATUS; ++ request->req_id = (uint16_t)0xDEADBEEF; ++ ++ /* send virtual channel request to add SA to hardware database */ ++ rc = iavf_ipsec_crypto_request(adapter, ++ (uint8_t *)request, request_len, ++ (uint8_t *)response, response_len); ++ if (rc) ++ goto update_cleanup; ++ ++ /* verify response id */ ++ if (response->ipsec_opcode != request->ipsec_opcode || ++ response->req_id != request->req_id){ ++ rc = -EFAULT; ++ goto update_cleanup; ++ } ++ memcpy(status, response->ipsec_data.ipsec_status, sizeof(*status)); ++ ++update_cleanup: ++ rte_free(response); ++ rte_free(request); ++ ++ return rc; ++} ++ ++ + int + iavf_ipsec_crypto_supported(struct iavf_adapter *adapter) + { + struct virtchnl_vf_resource *resources = adapter->vf.vf_res; ++ int crypto_supported = false; + + /** Capability check for IPsec Crypto */ + if (resources && (resources->vf_cap_flags & +- VIRTCHNL_VF_OFFLOAD_INLINE_IPSEC_CRYPTO)) +- return true; ++ VIRTCHNL_VF_OFFLOAD_INLINE_IPSEC_CRYPTO)) { ++ struct virtchnl_ipsec_status status; ++ int rc = iavf_ipsec_crypto_status_get(adapter, &status); ++ if (rc == 0 && status.status == INLINE_IPSEC_STATUS_AVAILABLE) ++ crypto_supported = true; ++ } + +- return false; ++ /* Clear the VF flag to return faster next call */ ++ if (resources && !crypto_supported) ++ resources->vf_cap_flags &= ++ ~(VIRTCHNL_VF_OFFLOAD_INLINE_IPSEC_CRYPTO); ++ ++ return crypto_supported; + } - seq_destroy(ofproto->ams_seq); + #define IAVF_IPSEC_INSET_ESP (\ +@@ -1623,6 +1693,7 @@ struct iavf_ipsec_flow_item { + struct rte_ipv6_hdr ipv6_hdr; + }; + struct rte_udp_hdr udp_hdr; ++ uint8_t is_udp; + }; -+ /* Wait for all the meter destroy work to finish. */ -+ ovsrcu_barrier(); - close_dpif_backer(ofproto->backer, del); - } + static void +@@ -1735,6 +1806,7 @@ iavf_ipsec_flow_item_parse(struct rte_eth_dev *ethdev, + parse_udp_item((const struct rte_flow_item_udp *) + pattern[2].spec, + &ipsec_flow->udp_hdr); ++ ipsec_flow->is_udp = true; + ipsec_flow->spi = + ((const struct rte_flow_item_esp *) + pattern[3].spec)->hdr.spi; +@@ -1804,7 +1876,9 @@ iavf_ipsec_flow_create(struct iavf_adapter *ad, + 1, + ipsec_flow->ipv4_hdr.dst_addr, + NULL, +- 0); ++ 0, ++ ipsec_flow->is_udp, ++ ipsec_flow->udp_hdr.dst_port); + } else { + ipsec_flow->id = + iavf_ipsec_crypto_inbound_security_policy_add(ad, +@@ -1812,7 +1886,9 @@ iavf_ipsec_flow_create(struct iavf_adapter *ad, + 0, + 0, + ipsec_flow->ipv6_hdr.dst_addr, +- 0); ++ 0, ++ ipsec_flow->is_udp, ++ ipsec_flow->udp_hdr.dst_port); + } -@@ -1945,7 +1938,7 @@ run(struct ofproto *ofproto_) + if (ipsec_flow->id < 1) { +diff --git a/dpdk/drivers/net/iavf/iavf_ipsec_crypto.h b/dpdk/drivers/net/iavf/iavf_ipsec_crypto.h +index 4e4c8798ec..8ea0f9540e 100644 +--- a/dpdk/drivers/net/iavf/iavf_ipsec_crypto.h ++++ b/dpdk/drivers/net/iavf/iavf_ipsec_crypto.h +@@ -73,7 +73,7 @@ enum iavf_ipsec_iv_len { + }; - new_dump_seq = seq_read(udpif_dump_seq(ofproto->backer->udpif)); - if (ofproto->dump_seq != new_dump_seq) { -- struct rule *rule, *next_rule; -+ struct rule *rule; - long long now = time_msec(); - /* We know stats are relatively fresh, so now is a good time to do some -@@ -1955,7 +1948,7 @@ run(struct ofproto *ofproto_) - /* Expire OpenFlow flows whose idle_timeout or hard_timeout - * has passed. */ - ovs_mutex_lock(&ofproto_mutex); -- LIST_FOR_EACH_SAFE (rule, next_rule, expirable, -+ LIST_FOR_EACH_SAFE (rule, expirable, - &ofproto->up.expirable) { - rule_expire(rule_dpif_cast(rule), now); - } -@@ -2346,6 +2339,7 @@ set_ipfix( - struct dpif_ipfix *di = ofproto->ipfix; - bool has_options = bridge_exporter_options || flow_exporters_options; - bool new_di = false; -+ bool options_changed = false; +-/* IPsec Crypto Packet Metaday offload flags */ ++/* IPsec Crypto Packet Metadata offload flags */ + #define IAVF_IPSEC_CRYPTO_OL_FLAGS_IS_TUN (0x1 << 0) + #define IAVF_IPSEC_CRYPTO_OL_FLAGS_ESN (0x1 << 1) + #define IAVF_IPSEC_CRYPTO_OL_FLAGS_IPV6_EXT_HDRS (0x1 << 2) +@@ -145,7 +145,9 @@ iavf_ipsec_crypto_inbound_security_policy_add(struct iavf_adapter *adapter, + uint8_t is_v4, + rte_be32_t v4_dst_addr, + uint8_t *v6_dst_addr, +- uint8_t drop); ++ uint8_t drop, ++ bool is_udp, ++ uint16_t udp_port); + + /** + * Delete inbound security policy rule from hardware +diff --git a/dpdk/drivers/net/iavf/iavf_rxtx.c b/dpdk/drivers/net/iavf/iavf_rxtx.c +index 154472c50f..3a0dfca2a7 100644 +--- a/dpdk/drivers/net/iavf/iavf_rxtx.c ++++ b/dpdk/drivers/net/iavf/iavf_rxtx.c +@@ -363,12 +363,24 @@ release_txq_mbufs(struct iavf_tx_queue *txq) + } + } - if (has_options && !di) { - di = ofproto->ipfix = dpif_ipfix_create(); -@@ -2355,7 +2349,7 @@ set_ipfix( - if (di) { - /* Call set_options in any case to cleanly flush the flow - * caches in the last exporters that are to be destroyed. */ -- dpif_ipfix_set_options( -+ options_changed = dpif_ipfix_set_options( - di, bridge_exporter_options, flow_exporters_options, - n_flow_exporters_options); +-static const struct iavf_rxq_ops def_rxq_ops = { +- .release_mbufs = release_rxq_mbufs, ++static const ++struct iavf_rxq_ops iavf_rxq_release_mbufs_ops[] = { ++ [IAVF_REL_MBUFS_DEFAULT].release_mbufs = release_rxq_mbufs, ++#ifdef RTE_ARCH_X86 ++ [IAVF_REL_MBUFS_SSE_VEC].release_mbufs = iavf_rx_queue_release_mbufs_sse, ++#endif + }; -@@ -2371,6 +2365,10 @@ set_ipfix( - dpif_ipfix_unref(di); - ofproto->ipfix = NULL; - } +-static const struct iavf_txq_ops def_txq_ops = { +- .release_mbufs = release_txq_mbufs, ++static const ++struct iavf_txq_ops iavf_txq_release_mbufs_ops[] = { ++ [IAVF_REL_MBUFS_DEFAULT].release_mbufs = release_txq_mbufs, ++#ifdef RTE_ARCH_X86 ++ [IAVF_REL_MBUFS_SSE_VEC].release_mbufs = iavf_tx_queue_release_mbufs_sse, ++#ifdef CC_AVX512_SUPPORT ++ [IAVF_REL_MBUFS_AVX512_VEC].release_mbufs = iavf_tx_queue_release_mbufs_avx512, ++#endif ++#endif + -+ if (new_di || options_changed) { -+ ofproto->backer->need_revalidate = REV_RECONFIGURE; -+ } - } + }; - return 0; -@@ -2493,11 +2491,11 @@ set_lldp(struct ofport *ofport_, + static inline void +@@ -476,54 +488,56 @@ iavf_rxd_to_pkt_fields_by_comms_aux_v2(struct iavf_rx_queue *rxq, + #endif + } + ++static const ++iavf_rxd_to_pkt_fields_t rxd_to_pkt_fields_ops[IAVF_RXDID_LAST + 1] = { ++ [IAVF_RXDID_COMMS_AUX_VLAN] = iavf_rxd_to_pkt_fields_by_comms_aux_v1, ++ [IAVF_RXDID_COMMS_AUX_IPV4] = iavf_rxd_to_pkt_fields_by_comms_aux_v1, ++ [IAVF_RXDID_COMMS_AUX_IPV6] = iavf_rxd_to_pkt_fields_by_comms_aux_v1, ++ [IAVF_RXDID_COMMS_AUX_IPV6_FLOW] = ++ iavf_rxd_to_pkt_fields_by_comms_aux_v1, ++ [IAVF_RXDID_COMMS_AUX_TCP] = iavf_rxd_to_pkt_fields_by_comms_aux_v1, ++ [IAVF_RXDID_COMMS_AUX_IP_OFFSET] = ++ iavf_rxd_to_pkt_fields_by_comms_aux_v2, ++ [IAVF_RXDID_COMMS_IPSEC_CRYPTO] = ++ iavf_rxd_to_pkt_fields_by_comms_aux_v2, ++ [IAVF_RXDID_COMMS_OVS_1] = iavf_rxd_to_pkt_fields_by_comms_ovs, ++}; ++ + static void + iavf_select_rxd_to_pkt_fields_handler(struct iavf_rx_queue *rxq, uint32_t rxdid) { - struct ofport_dpif *ofport = ofport_dpif_cast(ofport_); - struct ofproto_dpif *ofproto = ofproto_dpif_cast(ofport->up.ofproto); -+ bool old_enable = lldp_is_enabled(ofport->lldp); - int error = 0; ++ rxq->rxdid = rxdid; ++ + switch (rxdid) { + case IAVF_RXDID_COMMS_AUX_VLAN: + rxq->xtr_ol_flag = rte_pmd_ifd_dynflag_proto_xtr_vlan_mask; +- rxq->rxd_to_pkt_fields = +- iavf_rxd_to_pkt_fields_by_comms_aux_v1; + break; + case IAVF_RXDID_COMMS_AUX_IPV4: + rxq->xtr_ol_flag = rte_pmd_ifd_dynflag_proto_xtr_ipv4_mask; +- rxq->rxd_to_pkt_fields = +- iavf_rxd_to_pkt_fields_by_comms_aux_v1; + break; + case IAVF_RXDID_COMMS_AUX_IPV6: + rxq->xtr_ol_flag = rte_pmd_ifd_dynflag_proto_xtr_ipv6_mask; +- rxq->rxd_to_pkt_fields = +- iavf_rxd_to_pkt_fields_by_comms_aux_v1; + break; + case IAVF_RXDID_COMMS_AUX_IPV6_FLOW: + rxq->xtr_ol_flag = + rte_pmd_ifd_dynflag_proto_xtr_ipv6_flow_mask; +- rxq->rxd_to_pkt_fields = +- iavf_rxd_to_pkt_fields_by_comms_aux_v1; + break; + case IAVF_RXDID_COMMS_AUX_TCP: + rxq->xtr_ol_flag = rte_pmd_ifd_dynflag_proto_xtr_tcp_mask; +- rxq->rxd_to_pkt_fields = +- iavf_rxd_to_pkt_fields_by_comms_aux_v1; + break; + case IAVF_RXDID_COMMS_AUX_IP_OFFSET: + rxq->xtr_ol_flag = + rte_pmd_ifd_dynflag_proto_xtr_ip_offset_mask; +- rxq->rxd_to_pkt_fields = +- iavf_rxd_to_pkt_fields_by_comms_aux_v2; + break; + case IAVF_RXDID_COMMS_IPSEC_CRYPTO: + rxq->xtr_ol_flag = + rte_pmd_ifd_dynflag_proto_xtr_ipsec_crypto_said_mask; +- rxq->rxd_to_pkt_fields = +- iavf_rxd_to_pkt_fields_by_comms_aux_v2; + break; + case IAVF_RXDID_COMMS_OVS_1: +- rxq->rxd_to_pkt_fields = iavf_rxd_to_pkt_fields_by_comms_ovs; + break; + default: + /* update this according to the RXDID for FLEX_DESC_NONE */ +- rxq->rxd_to_pkt_fields = iavf_rxd_to_pkt_fields_by_comms_ovs; ++ rxq->rxdid = IAVF_RXDID_COMMS_OVS_1; + break; + } -- if (cfg) { -+ if (cfg && !smap_is_empty(cfg)) { - if (!ofport->lldp) { -- ofproto->backer->need_revalidate = REV_RECONFIGURE; - ofport->lldp = lldp_create(ofport->up.netdev, ofport_->mtu, cfg); - } +@@ -553,6 +567,9 @@ iavf_dev_rx_queue_setup(struct rte_eth_dev *dev, uint16_t queue_idx, -@@ -2509,6 +2507,9 @@ set_lldp(struct ofport *ofport_, - } else if (ofport->lldp) { - lldp_unref(ofport->lldp); - ofport->lldp = NULL; -+ } + PMD_INIT_FUNC_TRACE(); + ++ if (ad->closed) ++ return -EIO; + -+ if (lldp_is_enabled(ofport->lldp) != old_enable) { - ofproto->backer->need_revalidate = REV_RECONFIGURE; - } + offloads = rx_conf->offloads | dev->data->dev_conf.rxmode.offloads; -@@ -3106,11 +3107,11 @@ bundle_flush_macs(struct ofbundle *bundle, bool all_ofprotos) - { - struct ofproto_dpif *ofproto = bundle->ofproto; - struct mac_learning *ml = ofproto->ml; -- struct mac_entry *mac, *next_mac; -+ struct mac_entry *mac; + if (nb_desc % IAVF_ALIGN_RING_DESC != 0 || +@@ -648,8 +665,8 @@ iavf_dev_rx_queue_setup(struct rte_eth_dev *dev, uint16_t queue_idx, + return -ENOMEM; + } - ofproto->backer->need_revalidate = REV_RECONFIGURE; - ovs_rwlock_wrlock(&ml->rwlock); -- LIST_FOR_EACH_SAFE (mac, next_mac, lru_node, &ml->lrus) { -+ LIST_FOR_EACH_SAFE (mac, lru_node, &ml->lrus) { - if (mac_entry_get_port(ml, mac) == bundle) { - if (all_ofprotos) { - struct ofproto_dpif *o; -@@ -3141,13 +3142,13 @@ bundle_move(struct ofbundle *old, struct ofbundle *new) - { - struct ofproto_dpif *ofproto = old->ofproto; - struct mac_learning *ml = ofproto->ml; -- struct mac_entry *mac, *next_mac; -+ struct mac_entry *mac; +- /* Allocate the maximun number of RX ring hardware descriptor with +- * a liitle more to support bulk allocate. ++ /* Allocate the maximum number of RX ring hardware descriptor with ++ * a little more to support bulk allocate. + */ + len = IAVF_MAX_RING_DESC + IAVF_RX_MAX_BURST; + ring_size = RTE_ALIGN(len * sizeof(union iavf_rx_desc), +@@ -673,7 +690,7 @@ iavf_dev_rx_queue_setup(struct rte_eth_dev *dev, uint16_t queue_idx, + rxq->q_set = true; + dev->data->rx_queues[queue_idx] = rxq; + rxq->qrx_tail = hw->hw_addr + IAVF_QRX_TAIL1(rxq->queue_id); +- rxq->ops = &def_rxq_ops; ++ rxq->rel_mbufs_type = IAVF_REL_MBUFS_DEFAULT; + + if (check_rx_bulk_allow(rxq) == true) { + PMD_INIT_LOG(DEBUG, "Rx Burst Bulk Alloc Preconditions are " +@@ -714,6 +731,9 @@ iavf_dev_tx_queue_setup(struct rte_eth_dev *dev, + + PMD_INIT_FUNC_TRACE(); + ++ if (adapter->closed) ++ return -EIO; ++ + offloads = tx_conf->offloads | dev->data->dev_conf.txmode.offloads; + + if (nb_desc % IAVF_ALIGN_RING_DESC != 0 || +@@ -810,7 +830,7 @@ iavf_dev_tx_queue_setup(struct rte_eth_dev *dev, + txq->q_set = true; + dev->data->tx_queues[queue_idx] = txq; + txq->qtx_tail = hw->hw_addr + IAVF_QTX_TAIL1(queue_idx); +- txq->ops = &def_txq_ops; ++ txq->rel_mbufs_type = IAVF_REL_MBUFS_DEFAULT; + + if (check_tx_vec_allow(txq) == false) { + struct iavf_adapter *ad = +@@ -942,7 +962,7 @@ iavf_dev_rx_queue_stop(struct rte_eth_dev *dev, uint16_t rx_queue_id) + } - ovs_assert(new->ofproto == old->ofproto); + rxq = dev->data->rx_queues[rx_queue_id]; +- rxq->ops->release_mbufs(rxq); ++ iavf_rxq_release_mbufs_ops[rxq->rel_mbufs_type].release_mbufs(rxq); + reset_rx_queue(rxq); + dev->data->rx_queue_state[rx_queue_id] = RTE_ETH_QUEUE_STATE_STOPPED; - ofproto->backer->need_revalidate = REV_RECONFIGURE; - ovs_rwlock_wrlock(&ml->rwlock); -- LIST_FOR_EACH_SAFE (mac, next_mac, lru_node, &ml->lrus) { -+ LIST_FOR_EACH_SAFE (mac, lru_node, &ml->lrus) { - if (mac_entry_get_port(ml, mac) == old) { - mac_entry_set_port(ml, mac, new); - } -@@ -3244,7 +3245,7 @@ static void - bundle_destroy(struct ofbundle *bundle) - { - struct ofproto_dpif *ofproto; -- struct ofport_dpif *port, *next_port; -+ struct ofport_dpif *port; +@@ -970,7 +990,7 @@ iavf_dev_tx_queue_stop(struct rte_eth_dev *dev, uint16_t tx_queue_id) + } - if (!bundle) { - return; -@@ -3257,7 +3258,7 @@ bundle_destroy(struct ofbundle *bundle) - xlate_bundle_remove(bundle); - xlate_txn_commit(); + txq = dev->data->tx_queues[tx_queue_id]; +- txq->ops->release_mbufs(txq); ++ iavf_txq_release_mbufs_ops[txq->rel_mbufs_type].release_mbufs(txq); + reset_tx_queue(txq); + dev->data->tx_queue_state[tx_queue_id] = RTE_ETH_QUEUE_STATE_STOPPED; + +@@ -985,7 +1005,7 @@ iavf_dev_rx_queue_release(struct rte_eth_dev *dev, uint16_t qid) + if (!q) + return; + +- q->ops->release_mbufs(q); ++ iavf_rxq_release_mbufs_ops[q->rel_mbufs_type].release_mbufs(q); + rte_free(q->sw_ring); + rte_memzone_free(q->mz); + rte_free(q); +@@ -999,7 +1019,7 @@ iavf_dev_tx_queue_release(struct rte_eth_dev *dev, uint16_t qid) + if (!q) + return; + +- q->ops->release_mbufs(q); ++ iavf_txq_release_mbufs_ops[q->rel_mbufs_type].release_mbufs(q); + rte_free(q->sw_ring); + rte_memzone_free(q->mz); + rte_free(q); +@@ -1033,7 +1053,7 @@ iavf_stop_queues(struct rte_eth_dev *dev) + txq = dev->data->tx_queues[i]; + if (!txq) + continue; +- txq->ops->release_mbufs(txq); ++ iavf_txq_release_mbufs_ops[txq->rel_mbufs_type].release_mbufs(txq); + reset_tx_queue(txq); + dev->data->tx_queue_state[i] = RTE_ETH_QUEUE_STATE_STOPPED; + } +@@ -1041,7 +1061,7 @@ iavf_stop_queues(struct rte_eth_dev *dev) + rxq = dev->data->rx_queues[i]; + if (!rxq) + continue; +- rxq->ops->release_mbufs(rxq); ++ iavf_rxq_release_mbufs_ops[rxq->rel_mbufs_type].release_mbufs(rxq); + reset_rx_queue(rxq); + dev->data->rx_queue_state[i] = RTE_ETH_QUEUE_STATE_STOPPED; + } +@@ -1484,7 +1504,7 @@ iavf_recv_pkts_flex_rxd(void *rx_queue, + iavf_flex_rxd_to_vlan_tci(rxm, &rxd); + iavf_flex_rxd_to_ipsec_crypto_status(rxm, &rxd, + &rxq->stats.ipsec_crypto); +- rxq->rxd_to_pkt_fields(rxq, rxm, &rxd); ++ rxd_to_pkt_fields_ops[rxq->rxdid](rxq, rxm, &rxd); + pkt_flags = iavf_flex_rxd_error_to_pkt_flags(rx_stat_err0); + rxm->ol_flags |= pkt_flags; + +@@ -1628,7 +1648,7 @@ iavf_recv_scattered_pkts_flex_rxd(void *rx_queue, struct rte_mbuf **rx_pkts, + iavf_flex_rxd_to_vlan_tci(first_seg, &rxd); + iavf_flex_rxd_to_ipsec_crypto_status(first_seg, &rxd, + &rxq->stats.ipsec_crypto); +- rxq->rxd_to_pkt_fields(rxq, first_seg, &rxd); ++ rxd_to_pkt_fields_ops[rxq->rxdid](rxq, first_seg, &rxd); + pkt_flags = iavf_flex_rxd_error_to_pkt_flags(rx_stat_err0); + + first_seg->ol_flags |= pkt_flags; +@@ -1819,7 +1839,7 @@ iavf_rx_scan_hw_ring_flex_rxd(struct iavf_rx_queue *rxq) + struct rte_mbuf *mb; + uint16_t stat_err0; + uint16_t pkt_len; +- int32_t s[IAVF_LOOK_AHEAD], nb_dd; ++ int32_t s[IAVF_LOOK_AHEAD], var, nb_dd; + int32_t i, j, nb_rx = 0; + uint64_t pkt_flags; + const uint32_t *ptype_tbl = rxq->vsi->adapter->ptype_tbl; +@@ -1844,9 +1864,27 @@ iavf_rx_scan_hw_ring_flex_rxd(struct iavf_rx_queue *rxq) + + rte_smp_rmb(); + +- /* Compute how many status bits were set */ +- for (j = 0, nb_dd = 0; j < IAVF_LOOK_AHEAD; j++) +- nb_dd += s[j] & (1 << IAVF_RX_FLEX_DESC_STATUS0_DD_S); ++ /* Compute how many contiguous DD bits were set */ ++ for (j = 0, nb_dd = 0; j < IAVF_LOOK_AHEAD; j++) { ++ var = s[j] & (1 << IAVF_RX_FLEX_DESC_STATUS0_DD_S); ++#ifdef RTE_ARCH_ARM ++ /* For Arm platforms, count only contiguous descriptors ++ * whose DD bit is set to 1. On Arm platforms, reads of ++ * descriptors can be reordered. Since the CPU may ++ * be reading the descriptors as the NIC updates them ++ * in memory, it is possbile that the DD bit for a ++ * descriptor earlier in the queue is read as not set ++ * while the DD bit for a descriptor later in the queue ++ * is read as set. ++ */ ++ if (var) ++ nb_dd += 1; ++ else ++ break; ++#else ++ nb_dd += var; ++#endif ++ } + + nb_rx += nb_dd; + +@@ -1868,7 +1906,7 @@ iavf_rx_scan_hw_ring_flex_rxd(struct iavf_rx_queue *rxq) + iavf_flex_rxd_to_vlan_tci(mb, &rxdp[j]); + iavf_flex_rxd_to_ipsec_crypto_status(mb, &rxdp[j], + &rxq->stats.ipsec_crypto); +- rxq->rxd_to_pkt_fields(rxq, mb, &rxdp[j]); ++ rxd_to_pkt_fields_ops[rxq->rxdid](rxq, mb, &rxdp[j]); + stat_err0 = rte_le_to_cpu_16(rxdp[j].wb.status_error0); + pkt_flags = iavf_flex_rxd_error_to_pkt_flags(stat_err0); + +@@ -1898,7 +1936,7 @@ iavf_rx_scan_hw_ring(struct iavf_rx_queue *rxq) + uint16_t pkt_len; + uint64_t qword1; + uint32_t rx_status; +- int32_t s[IAVF_LOOK_AHEAD], nb_dd; ++ int32_t s[IAVF_LOOK_AHEAD], var, nb_dd; + int32_t i, j, nb_rx = 0; + uint64_t pkt_flags; + const uint32_t *ptype_tbl = rxq->vsi->adapter->ptype_tbl; +@@ -1929,9 +1967,27 @@ iavf_rx_scan_hw_ring(struct iavf_rx_queue *rxq) + + rte_smp_rmb(); + +- /* Compute how many status bits were set */ +- for (j = 0, nb_dd = 0; j < IAVF_LOOK_AHEAD; j++) +- nb_dd += s[j] & (1 << IAVF_RX_DESC_STATUS_DD_SHIFT); ++ /* Compute how many contiguous DD bits were set */ ++ for (j = 0, nb_dd = 0; j < IAVF_LOOK_AHEAD; j++) { ++ var = s[j] & (1 << IAVF_RX_DESC_STATUS_DD_SHIFT); ++#ifdef RTE_ARCH_ARM ++ /* For Arm platforms, count only contiguous descriptors ++ * whose DD bit is set to 1. On Arm platforms, reads of ++ * descriptors can be reordered. Since the CPU may ++ * be reading the descriptors as the NIC updates them ++ * in memory, it is possbile that the DD bit for a ++ * descriptor earlier in the queue is read as not set ++ * while the DD bit for a descriptor later in the queue ++ * is read as set. ++ */ ++ if (var) ++ nb_dd += 1; ++ else ++ break; ++#else ++ nb_dd += var; ++#endif ++ } + + nb_rx += nb_dd; + +@@ -2439,6 +2495,14 @@ iavf_fill_data_desc(volatile struct iavf_tx_desc *desc, + if (m->ol_flags & RTE_MBUF_F_TX_SEC_OFFLOAD) + hdrlen += ipseclen; + bufsz = hdrlen + tlen; ++ } else if ((m->ol_flags & RTE_MBUF_F_TX_SEC_OFFLOAD) && ++ (m->ol_flags & (RTE_MBUF_F_TX_TCP_SEG | ++ RTE_MBUF_F_TX_UDP_SEG))) { ++ hdrlen += m->outer_l3_len + m->l3_len + ipseclen; ++ if (m->ol_flags & RTE_MBUF_F_TX_L4_MASK) ++ hdrlen += m->l4_len; ++ bufsz = hdrlen + tlen; ++ + } else { + bufsz = m->data_len; + } +@@ -2484,12 +2548,6 @@ iavf_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts, uint16_t nb_pkts) + desc_idx = txq->tx_tail; + txe = &txe_ring[desc_idx]; -- LIST_FOR_EACH_SAFE (port, next_port, bundle_node, &bundle->ports) { -+ LIST_FOR_EACH_SAFE (port, bundle_node, &bundle->ports) { - bundle_del_port(port); - } +-#ifdef RTE_LIBRTE_IAVF_DEBUG_TX_DESC_RING +- iavf_dump_tx_entry_ring(txq); +- iavf_dump_tx_desc_ring(txq); +-#endif +- +- + for (idx = 0; idx < nb_pkts; idx++) { + volatile struct iavf_tx_desc *ddesc; + struct iavf_ipsec_crypto_pkt_metadata *ipsec_md; +@@ -2694,6 +2752,10 @@ iavf_prep_pkts(__rte_unused void *tx_queue, struct rte_mbuf **tx_pkts, + struct iavf_tx_queue *txq = tx_queue; + struct rte_eth_dev *dev = &rte_eth_devices[txq->port_id]; + struct iavf_info *vf = IAVF_DEV_PRIVATE_TO_VF(dev->data->dev_private); ++ struct iavf_adapter *adapter = IAVF_DEV_PRIVATE_TO_ADAPTER(dev->data->dev_private); ++ ++ if (adapter->closed) ++ return 0; + + for (i = 0; i < nb_pkts; i++) { + m = tx_pkts[i]; +@@ -2750,14 +2812,27 @@ iavf_set_rx_function(struct rte_eth_dev *dev) + struct iavf_adapter *adapter = + IAVF_DEV_PRIVATE_TO_ADAPTER(dev->data->dev_private); + struct iavf_info *vf = IAVF_DEV_PRIVATE_TO_VF(dev->data->dev_private); ++ int i; ++ struct iavf_rx_queue *rxq; ++ bool use_flex = true; ++ ++ for (i = 0; i < dev->data->nb_rx_queues; i++) { ++ rxq = dev->data->rx_queues[i]; ++ if (rxq->rxdid <= IAVF_RXDID_LEGACY_1) { ++ PMD_DRV_LOG(NOTICE, "request RXDID[%d] in Queue[%d] is legacy, " ++ "set rx_pkt_burst as legacy for all queues", rxq->rxdid, i); ++ use_flex = false; ++ } else if (!(vf->supported_rxdid & BIT(rxq->rxdid))) { ++ PMD_DRV_LOG(NOTICE, "request RXDID[%d] in Queue[%d] is not supported, " ++ "set rx_pkt_burst as legacy for all queues", rxq->rxdid, i); ++ use_flex = false; ++ } ++ } + + #ifdef RTE_ARCH_X86 +- struct iavf_rx_queue *rxq; +- int i; + int check_ret; + bool use_avx2 = false; + bool use_avx512 = false; +- bool use_flex = false; + + check_ret = iavf_rx_vec_dev_check(dev); + if (check_ret >= 0 && +@@ -2774,10 +2849,6 @@ iavf_set_rx_function(struct rte_eth_dev *dev) + use_avx512 = true; + #endif -@@ -3347,9 +3348,7 @@ bundle_set(struct ofproto *ofproto_, void *aux, - } - } - if (!ok || ovs_list_size(&bundle->ports) != s->n_members) { -- struct ofport_dpif *next_port; +- if (vf->vf_res->vf_cap_flags & +- VIRTCHNL_VF_OFFLOAD_RX_FLEX_DESC) +- use_flex = true; - -- LIST_FOR_EACH_SAFE (port, next_port, bundle_node, &bundle->ports) { -+ LIST_FOR_EACH_SAFE (port, bundle_node, &bundle->ports) { - for (i = 0; i < s->n_members; i++) { - if (s->members[i] == port->up.ofp_port) { - goto found; -@@ -3963,6 +3962,10 @@ port_add(struct ofproto *ofproto_, struct netdev *netdev) - simap_put(&ofproto->backer->tnl_backers, - dp_port_name, odp_to_u32(port_no)); - } -+ } else { -+ struct dpif *dpif = ofproto->backer->dpif; -+ const char *dpif_type_str = dpif_normalize_type(dpif_type(dpif)); -+ netdev_set_dpif_type(netdev, dpif_type_str); - } + for (i = 0; i < dev->data->nb_rx_queues; i++) { + rxq = dev->data->rx_queues[i]; + (void)iavf_rxq_vec_setup(rxq); +@@ -2881,7 +2952,7 @@ iavf_set_rx_function(struct rte_eth_dev *dev) + if (dev->data->scattered_rx) { + PMD_DRV_LOG(DEBUG, "Using a Scattered Rx callback (port=%d).", + dev->data->port_id); +- if (vf->vf_res->vf_cap_flags & VIRTCHNL_VF_OFFLOAD_RX_FLEX_DESC) ++ if (use_flex) + dev->rx_pkt_burst = iavf_recv_scattered_pkts_flex_rxd; + else + dev->rx_pkt_burst = iavf_recv_scattered_pkts; +@@ -2892,7 +2963,7 @@ iavf_set_rx_function(struct rte_eth_dev *dev) + } else { + PMD_DRV_LOG(DEBUG, "Using Basic Rx callback (port=%d).", + dev->data->port_id); +- if (vf->vf_res->vf_cap_flags & VIRTCHNL_VF_OFFLOAD_RX_FLEX_DESC) ++ if (use_flex) + dev->rx_pkt_burst = iavf_recv_pkts_flex_rxd; + else + dev->rx_pkt_burst = iavf_recv_pkts; +diff --git a/dpdk/drivers/net/iavf/iavf_rxtx.h b/dpdk/drivers/net/iavf/iavf_rxtx.h +index b610176b30..48cc0da6f5 100644 +--- a/dpdk/drivers/net/iavf/iavf_rxtx.h ++++ b/dpdk/drivers/net/iavf/iavf_rxtx.h +@@ -187,6 +187,7 @@ struct iavf_rx_queue { + struct rte_mbuf *pkt_last_seg; /* last segment of current packet */ + struct rte_mbuf fake_mbuf; /* dummy mbuf */ + uint8_t rxdid; ++ uint8_t rel_mbufs_type; + + /* used for VPMD */ + uint16_t rxrearm_nb; /* number of remaining to be re-armed */ +@@ -217,8 +218,6 @@ struct iavf_rx_queue { + uint8_t proto_xtr; /* protocol extraction type */ + uint64_t xtr_ol_flag; + /* flexible descriptor metadata extraction offload flag */ +- iavf_rxd_to_pkt_fields_t rxd_to_pkt_fields; +- /* handle flexible descriptor by RXDID */ + struct iavf_rx_queue_stats stats; + uint64_t offloads; + }; +@@ -248,6 +247,7 @@ struct iavf_tx_queue { + uint16_t last_desc_cleaned; /* last desc have been cleaned*/ + uint16_t free_thresh; + uint16_t rs_thresh; ++ uint8_t rel_mbufs_type; + + uint16_t port_id; + uint16_t queue_id; +@@ -391,6 +391,12 @@ struct iavf_32b_rx_flex_desc_comms_ipsec { + __le32 ipsec_said; + }; - if (netdev_get_tunnel_config(netdev)) { -@@ -4471,12 +4474,14 @@ rule_dpif_lookup_from_table(struct ofproto_dpif *ofproto, - atomic_add_relaxed(&tbl->n_matched, stats->n_packets, &orig); - } - if (xcache) { -- struct xc_entry *entry; -+ if (ofproto_try_ref(&ofproto->up)) { -+ struct xc_entry *entry; ++enum iavf_rxtx_rel_mbufs_type { ++ IAVF_REL_MBUFS_DEFAULT = 0, ++ IAVF_REL_MBUFS_SSE_VEC = 1, ++ IAVF_REL_MBUFS_AVX512_VEC = 2, ++}; ++ + /* Receive Flex Descriptor profile IDs: There are a total + * of 64 profiles where profile IDs 0/1 are for legacy; and + * profiles 2-63 are flex profiles that can be programmed +@@ -694,6 +700,9 @@ int iavf_txq_vec_setup_avx512(struct iavf_tx_queue *txq); + uint8_t iavf_proto_xtr_type_to_rxdid(uint8_t xtr_type); + + void iavf_set_default_ptype_table(struct rte_eth_dev *dev); ++void iavf_tx_queue_release_mbufs_avx512(struct iavf_tx_queue *txq); ++void iavf_rx_queue_release_mbufs_sse(struct iavf_rx_queue *rxq); ++void iavf_tx_queue_release_mbufs_sse(struct iavf_tx_queue *txq); + + static inline + void iavf_dump_rx_descriptor(struct iavf_rx_queue *rxq, +diff --git a/dpdk/drivers/net/iavf/iavf_rxtx_vec_avx512.c b/dpdk/drivers/net/iavf/iavf_rxtx_vec_avx512.c +index 6ff38ac368..c975a5e7d7 100644 +--- a/dpdk/drivers/net/iavf/iavf_rxtx_vec_avx512.c ++++ b/dpdk/drivers/net/iavf/iavf_rxtx_vec_avx512.c +@@ -1994,7 +1994,7 @@ iavf_xmit_pkts_vec_avx512(void *tx_queue, struct rte_mbuf **tx_pkts, + return iavf_xmit_pkts_vec_avx512_cmn(tx_queue, tx_pkts, nb_pkts, false); + } -- entry = xlate_cache_add_entry(xcache, XC_TABLE); -- entry->table.ofproto = ofproto; -- entry->table.id = *table_id; -- entry->table.match = true; -+ entry = xlate_cache_add_entry(xcache, XC_TABLE); -+ entry->table.ofproto = ofproto; -+ entry->table.id = *table_id; -+ entry->table.match = true; -+ } - } - return rule; - } -@@ -4507,12 +4512,14 @@ rule_dpif_lookup_from_table(struct ofproto_dpif *ofproto, - stats->n_packets, &orig); - } - if (xcache) { -- struct xc_entry *entry; -+ if (ofproto_try_ref(&ofproto->up)) { -+ struct xc_entry *entry; +-static inline void ++void __rte_cold + iavf_tx_queue_release_mbufs_avx512(struct iavf_tx_queue *txq) + { + unsigned int i; +@@ -2014,14 +2014,10 @@ iavf_tx_queue_release_mbufs_avx512(struct iavf_tx_queue *txq) + } + } -- entry = xlate_cache_add_entry(xcache, XC_TABLE); -- entry->table.ofproto = ofproto; -- entry->table.id = next_id; -- entry->table.match = (rule != NULL); -+ entry = xlate_cache_add_entry(xcache, XC_TABLE); -+ entry->table.ofproto = ofproto; -+ entry->table.id = next_id; -+ entry->table.match = (rule != NULL); -+ } - } - if (rule) { - goto out; /* Match. */ -@@ -5550,9 +5557,9 @@ ct_zone_timeout_policy_sweep(struct dpif_backer *backer) +-static const struct iavf_txq_ops avx512_vec_txq_ops = { +- .release_mbufs = iavf_tx_queue_release_mbufs_avx512, +-}; +- + int __rte_cold + iavf_txq_vec_setup_avx512(struct iavf_tx_queue *txq) { - if (!ovs_list_is_empty(&backer->ct_tp_kill_list) - && time_msec() >= timeout_policy_cleanup_timer) { -- struct ct_timeout_policy *ct_tp, *next; -+ struct ct_timeout_policy *ct_tp; +- txq->ops = &avx512_vec_txq_ops; ++ txq->rel_mbufs_type = IAVF_REL_MBUFS_AVX512_VEC; + return 0; + } -- LIST_FOR_EACH_SAFE (ct_tp, next, list_node, &backer->ct_tp_kill_list) { -+ LIST_FOR_EACH_SAFE (ct_tp, list_node, &backer->ct_tp_kill_list) { - if (!ct_dpif_del_timeout_policy(backer->dpif, ct_tp->tp_id)) { - ovs_list_remove(&ct_tp->list_node); - ct_timeout_policy_destroy(ct_tp, backer->tp_ids); -@@ -5594,6 +5601,7 @@ ct_set_zone_timeout_policy(const char *datapath_type, uint16_t zone_id, - ct_timeout_policy_unref(backer, ct_zone->ct_tp); - ct_zone->ct_tp = ct_tp; - ct_tp->ref_count++; -+ backer->need_revalidate = REV_RECONFIGURE; - } - } else { - struct ct_zone *new_ct_zone = ct_zone_alloc(zone_id); -@@ -5601,6 +5609,7 @@ ct_set_zone_timeout_policy(const char *datapath_type, uint16_t zone_id, - cmap_insert(&backer->ct_zones, &new_ct_zone->node, - hash_int(zone_id, 0)); - ct_tp->ref_count++; -+ backer->need_revalidate = REV_RECONFIGURE; - } +diff --git a/dpdk/drivers/net/iavf/iavf_rxtx_vec_sse.c b/dpdk/drivers/net/iavf/iavf_rxtx_vec_sse.c +index 1bac59bf0e..4b23ca8d82 100644 +--- a/dpdk/drivers/net/iavf/iavf_rxtx_vec_sse.c ++++ b/dpdk/drivers/net/iavf/iavf_rxtx_vec_sse.c +@@ -159,7 +159,7 @@ desc_to_olflags_v(struct iavf_rx_queue *rxq, __m128i descs[4], + l3_l4e = _mm_shuffle_epi8(l3_l4e_flags, l3_l4e); + /* then we shift left 1 bit */ + l3_l4e = _mm_slli_epi32(l3_l4e, 1); +- /* we need to mask out the reduntant bits */ ++ /* we need to mask out the redundant bits */ + l3_l4e = _mm_and_si128(l3_l4e, cksum_mask); + + vlan0 = _mm_or_si128(vlan0, rss); +@@ -613,7 +613,7 @@ _recv_raw_pkts_vec(struct iavf_rx_queue *rxq, struct rte_mbuf **rx_pkts, + _mm_storeu_si128((void *)&rx_pkts[pos]->rx_descriptor_fields1, + pkt_mb1); + desc_to_ptype_v(descs, &rx_pkts[pos], ptype_tbl); +- /* C.4 calc avaialbe number of desc */ ++ /* C.4 calc available number of desc */ + var = __builtin_popcountll(_mm_cvtsi128_si64(staterr)); + nb_pkts_recd += var; + if (likely(var != IAVF_VPMD_DESCS_PER_LOOP)) +@@ -1200,37 +1200,29 @@ iavf_xmit_pkts_vec(void *tx_queue, struct rte_mbuf **tx_pkts, + return nb_tx; } -@@ -5617,6 +5626,7 @@ ct_del_zone_timeout_policy(const char *datapath_type, uint16_t zone_id) - if (ct_zone) { - ct_timeout_policy_unref(backer, ct_zone->ct_tp); - ct_zone_remove_and_destroy(backer, ct_zone); -+ backer->need_revalidate = REV_RECONFIGURE; - } +-static void __rte_cold ++void __rte_cold + iavf_rx_queue_release_mbufs_sse(struct iavf_rx_queue *rxq) + { + _iavf_rx_queue_release_mbufs_vec(rxq); } -@@ -5818,15 +5828,7 @@ ofproto_dpif_lookup_by_name(const char *name) - struct ofproto_dpif * - ofproto_dpif_lookup_by_uuid(const struct uuid *uuid) +-static void __rte_cold ++void __rte_cold + iavf_tx_queue_release_mbufs_sse(struct iavf_tx_queue *txq) { -- struct ofproto_dpif *ofproto; + _iavf_tx_queue_release_mbufs_vec(txq); + } + +-static const struct iavf_rxq_ops sse_vec_rxq_ops = { +- .release_mbufs = iavf_rx_queue_release_mbufs_sse, +-}; - -- HMAP_FOR_EACH_WITH_HASH (ofproto, all_ofproto_dpifs_by_uuid_node, -- uuid_hash(uuid), &all_ofproto_dpifs_by_uuid) { -- if (uuid_equals(&ofproto->uuid, uuid)) { -- return ofproto; -- } -- } -- return NULL; -+ return xlate_ofproto_lookup(uuid); +-static const struct iavf_txq_ops sse_vec_txq_ops = { +- .release_mbufs = iavf_tx_queue_release_mbufs_sse, +-}; +- + int __rte_cold + iavf_txq_vec_setup(struct iavf_tx_queue *txq) + { +- txq->ops = &sse_vec_txq_ops; ++ txq->rel_mbufs_type = IAVF_REL_MBUFS_SSE_VEC; + return 0; } - static void -diff --git a/ofproto/ofproto-provider.h b/ofproto/ofproto-provider.h -index 14b909973d..47e96e62e1 100644 ---- a/ofproto/ofproto-provider.h -+++ b/ofproto/ofproto-provider.h -@@ -143,6 +143,8 @@ struct ofproto { - /* Variable length mf_field mapping. Stores all configured variable length - * meta-flow fields (struct mf_field) in a switch. */ - struct vl_mff_map vl_mff_map; -+ /* refcount to this ofproto, held by rule/group/xlate_caches */ -+ struct ovs_refcount refcount; - }; + int __rte_cold + iavf_rxq_vec_setup(struct iavf_rx_queue *rxq) + { +- rxq->ops = &sse_vec_rxq_ops; ++ rxq->rel_mbufs_type = IAVF_REL_MBUFS_SSE_VEC; + return iavf_rxq_vec_setup_default(rxq); + } - void ofproto_init_tables(struct ofproto *, int n_tables); -diff --git a/ofproto/ofproto.c b/ofproto/ofproto.c -index 56aeac7209..933f7de2dc 100644 ---- a/ofproto/ofproto.c -+++ b/ofproto/ofproto.c -@@ -549,6 +549,7 @@ ofproto_create(const char *datapath_name, const char *datapath_type, +diff --git a/dpdk/drivers/net/iavf/iavf_vchnl.c b/dpdk/drivers/net/iavf/iavf_vchnl.c +index 145b059837..1bd3559ec2 100644 +--- a/dpdk/drivers/net/iavf/iavf_vchnl.c ++++ b/dpdk/drivers/net/iavf/iavf_vchnl.c +@@ -265,6 +265,11 @@ iavf_handle_pf_event_msg(struct rte_eth_dev *dev, uint8_t *msg, + struct virtchnl_pf_event *pf_msg = + (struct virtchnl_pf_event *)msg; + ++ if (adapter->closed) { ++ PMD_DRV_LOG(DEBUG, "Port closed"); ++ return; ++ } ++ + if (msglen < sizeof(struct virtchnl_pf_event)) { + PMD_DRV_LOG(DEBUG, "Error event"); + return; +@@ -461,7 +466,7 @@ iavf_check_api_version(struct iavf_adapter *adapter) + (vf->virtchnl_version.major == VIRTCHNL_VERSION_MAJOR_START && + vf->virtchnl_version.minor < VIRTCHNL_VERSION_MINOR_START)) { + PMD_INIT_LOG(ERR, "VIRTCHNL API version should not be lower" +- " than (%u.%u) to support Adapative VF", ++ " than (%u.%u) to support Adaptive VF", + VIRTCHNL_VERSION_MAJOR_START, + VIRTCHNL_VERSION_MAJOR_START); + return -1; +@@ -502,7 +507,7 @@ iavf_get_vf_resource(struct iavf_adapter *adapter) + VIRTCHNL_VF_OFFLOAD_VLAN_V2 | + VIRTCHNL_VF_LARGE_NUM_QPAIRS | + VIRTCHNL_VF_OFFLOAD_QOS | +-+ VIRTCHNL_VF_OFFLOAD_INLINE_IPSEC_CRYPTO; ++ VIRTCHNL_VF_OFFLOAD_INLINE_IPSEC_CRYPTO; + + args.in_args = (uint8_t *)∩︀ + args.in_args_size = sizeof(caps); +@@ -777,6 +782,9 @@ iavf_switch_queue(struct iavf_adapter *adapter, uint16_t qid, + struct iavf_cmd_info args; + int err; + ++ if (adapter->closed) ++ return -EIO; ++ + memset(&queue_select, 0, sizeof(queue_select)); + queue_select.vsi_id = vf->vsi_res->vsi_id; + if (rx) +@@ -1241,6 +1249,9 @@ iavf_query_stats(struct iavf_adapter *adapter, + struct iavf_cmd_info args; + int err; + ++ if (adapter->closed) ++ return -EIO; ++ + memset(&q_stats, 0, sizeof(q_stats)); + q_stats.vsi_id = vf->vsi_res->vsi_id; + args.ops = VIRTCHNL_OP_GET_STATS; +@@ -1269,6 +1280,9 @@ iavf_config_promisc(struct iavf_adapter *adapter, + struct iavf_cmd_info args; + int err; + ++ if (adapter->closed) ++ return -EIO; ++ + promisc.flags = 0; + promisc.vsi_id = vf->vsi_res->vsi_id; + +@@ -1312,6 +1326,9 @@ iavf_add_del_eth_addr(struct iavf_adapter *adapter, struct rte_ether_addr *addr, + struct iavf_cmd_info args; + int err; + ++ if (adapter->closed) ++ return -EIO; ++ + list = (struct virtchnl_ether_addr_list *)cmd_buffer; + list->vsi_id = vf->vsi_res->vsi_id; + list->num_elements = 1; +@@ -1487,7 +1504,7 @@ iavf_fdir_check(struct iavf_adapter *adapter, + + err = iavf_execute_vf_cmd(adapter, &args, 0); + if (err) { +- PMD_DRV_LOG(ERR, "fail to check flow direcotor rule"); ++ PMD_DRV_LOG(ERR, "fail to check flow director rule"); + return err; + } - ovs_mutex_init(&ofproto->vl_mff_map.mutex); - cmap_init(&ofproto->vl_mff_map.cmap); -+ ovs_refcount_init(&ofproto->refcount); +diff --git a/dpdk/drivers/net/ice/base/ice_flex_pipe.c b/dpdk/drivers/net/ice/base/ice_flex_pipe.c +index 395787806b..3918169001 100644 +--- a/dpdk/drivers/net/ice/base/ice_flex_pipe.c ++++ b/dpdk/drivers/net/ice/base/ice_flex_pipe.c +@@ -1785,8 +1785,12 @@ static enum ice_prof_type + ice_get_sw_prof_type(struct ice_hw *hw, struct ice_fv *fv) + { + u16 i; ++ bool valid_prof = false; + + for (i = 0; i < hw->blk[ICE_BLK_SW].es.fvw; i++) { ++ if (fv->ew[i].off != ICE_NAN_OFFSET) ++ valid_prof = true; ++ + /* UDP tunnel will have UDP_OF protocol ID and VNI offset */ + if (fv->ew[i].prot_id == (u8)ICE_PROT_UDP_OF && + fv->ew[i].off == ICE_VNI_OFFSET) +@@ -1801,7 +1805,7 @@ ice_get_sw_prof_type(struct ice_hw *hw, struct ice_fv *fv) + return ICE_PROF_TUN_PPPOE; + } - error = ofproto->ofproto_class->construct(ofproto); - if (error) { -@@ -1695,9 +1696,33 @@ ofproto_destroy__(struct ofproto *ofproto) - ofproto->ofproto_class->dealloc(ofproto); +- return ICE_PROF_NON_TUN; ++ return valid_prof ? ICE_PROF_NON_TUN : ICE_PROF_INVALID; } --/* Destroying rules is doubly deferred, must have 'ofproto' around for them. -- * - 1st we defer the removal of the rules from the classifier -- * - 2nd we defer the actual destruction of the rules. */ -+/* -+ * Rule destruction requires ofproto to remain accessible. -+ * Depending on the rule destruction call (shown in below), it can take several -+ * RCU grace periods before the ofproto reference is not needed anymore. -+ * The ofproto destruction callback is thus protected by a refcount, -+ * and such destruction is itself deferred. -+ * -+ * remove_rules_postponed (one grace period) -+ * -> remove_rule_rcu -+ * -> remove_rule_rcu__ -+ * -> ofproto_rule_unref -> ref count != 1 -+ * -> ... more grace periods. -+ * -> rule_destroy_cb (> 2 grace periods) -+ * -> free -+ * -+ * NOTE: The original ofproto destruction is only deferred by two grace -+ * periods to keep ofproto accessible. By using refcount together the -+ * destruction can be deferred for longer time. Now ofproto has 3 states: -+ * -+ * state 1: alive, with refcount >= 1 -+ * state 2: dying, with refcount == 0, however pointer is valid -+ * state 3: died, memory freed, pointer might be dangling. -+ * -+ * We only need to add refcount to certain objects whose destruction can -+ * take several RCU grace periods (rule, group, xlate_cache). Other -+ * references to ofproto must be cleared before the 2 RCU grace periods. -+ */ - static void - ofproto_destroy_defer__(struct ofproto *ofproto) - OVS_EXCLUDED(ofproto_mutex) -@@ -1705,11 +1730,31 @@ ofproto_destroy_defer__(struct ofproto *ofproto) - ovsrcu_postpone(ofproto_destroy__, ofproto); + /** +@@ -1818,11 +1822,6 @@ ice_get_sw_fv_bitmap(struct ice_hw *hw, enum ice_prof_type req_profs, + struct ice_seg *ice_seg; + struct ice_fv *fv; + +- if (req_profs == ICE_PROF_ALL) { +- ice_bitmap_set(bm, 0, ICE_MAX_NUM_PROFILES); +- return; +- } +- + ice_memset(&state, 0, sizeof(state), ICE_NONDMA_MEM); + ice_zero_bitmap(bm, ICE_MAX_NUM_PROFILES); + ice_seg = hw->seg; +@@ -2565,7 +2564,7 @@ enum ice_status ice_destroy_tunnel(struct ice_hw *hw, u16 port, bool all) + * @off: variable to receive the protocol offset + */ + enum ice_status +-ice_find_prot_off(struct ice_hw *hw, enum ice_block blk, u8 prof, u16 fv_idx, ++ice_find_prot_off(struct ice_hw *hw, enum ice_block blk, u8 prof, u8 fv_idx, + u8 *prot, u16 *off) + { + struct ice_fv_word *fv_ext; +diff --git a/dpdk/drivers/net/ice/base/ice_flex_pipe.h b/dpdk/drivers/net/ice/base/ice_flex_pipe.h +index 23ba45564a..ab897de4f3 100644 +--- a/dpdk/drivers/net/ice/base/ice_flex_pipe.h ++++ b/dpdk/drivers/net/ice/base/ice_flex_pipe.h +@@ -25,7 +25,7 @@ enum ice_status + ice_acquire_change_lock(struct ice_hw *hw, enum ice_aq_res_access_type access); + void ice_release_change_lock(struct ice_hw *hw); + enum ice_status +-ice_find_prot_off(struct ice_hw *hw, enum ice_block blk, u8 prof, u16 fv_idx, ++ice_find_prot_off(struct ice_hw *hw, enum ice_block blk, u8 prof, u8 fv_idx, + u8 *prot, u16 *off); + enum ice_status + ice_find_label_value(struct ice_seg *ice_seg, char const *name, u32 type, +diff --git a/dpdk/drivers/net/ice/base/ice_flex_type.h b/dpdk/drivers/net/ice/base/ice_flex_type.h +index 59eeca0a30..09a02fe9ac 100644 +--- a/dpdk/drivers/net/ice/base/ice_flex_type.h ++++ b/dpdk/drivers/net/ice/base/ice_flex_type.h +@@ -1003,6 +1003,7 @@ struct ice_chs_chg { + #define ICE_FLOW_PTYPE_MAX ICE_XLT1_CNT + + enum ice_prof_type { ++ ICE_PROF_INVALID = 0x0, + ICE_PROF_NON_TUN = 0x1, + ICE_PROF_TUN_UDP = 0x2, + ICE_PROF_TUN_GRE = 0x4, +diff --git a/dpdk/drivers/net/ice/base/ice_protocol_type.h b/dpdk/drivers/net/ice/base/ice_protocol_type.h +index cef8354f77..d27ef46713 100644 +--- a/dpdk/drivers/net/ice/base/ice_protocol_type.h ++++ b/dpdk/drivers/net/ice/base/ice_protocol_type.h +@@ -54,6 +54,7 @@ enum ice_protocol_type { + ICE_GTP_NO_PAY, + ICE_VLAN_EX, + ICE_VLAN_IN, ++ ICE_FLG_DIR, + ICE_PROTOCOL_LAST + }; + +@@ -191,6 +192,7 @@ enum ice_prot_id { + + #define ICE_VNI_OFFSET 12 /* offset of VNI from ICE_PROT_UDP_OF */ + ++#define ICE_NAN_OFFSET 511 + #define ICE_MAC_OFOS_HW 1 + #define ICE_MAC_IL_HW 4 + #define ICE_ETYPE_OL_HW 9 +@@ -217,9 +219,10 @@ enum ice_prot_id { + #define ICE_META_DATA_ID_HW 255 /* this is used for tunnel type */ + + #define ICE_MDID_SIZE 2 +-#define ICE_TUN_FLAG_MDID 21 +-#define ICE_TUN_FLAG_MDID_OFF (ICE_MDID_SIZE * ICE_TUN_FLAG_MDID) ++#define ICE_TUN_FLAG_MDID 20 ++#define ICE_TUN_FLAG_MDID_OFF(word) (ICE_MDID_SIZE * (ICE_TUN_FLAG_MDID + (word))) + #define ICE_TUN_FLAG_MASK 0xFF ++#define ICE_DIR_FLAG_MASK 0x10 + #define ICE_TUN_FLAG_VLAN_MASK 0x01 + #define ICE_TUN_FLAG_FV_IND 2 + +@@ -420,7 +423,7 @@ struct ice_recp_grp_entry { + #define ICE_INVAL_CHAIN_IND 0xFF + u16 rid; + u8 chain_idx; +- u16 fv_idx[ICE_NUM_WORDS_RECIPE]; ++ u8 fv_idx[ICE_NUM_WORDS_RECIPE]; + u16 fv_mask[ICE_NUM_WORDS_RECIPE]; + struct ice_pref_recipe_group r_group; + }; +diff --git a/dpdk/drivers/net/ice/base/ice_sched.c b/dpdk/drivers/net/ice/base/ice_sched.c +index 2620892c9e..e697c579be 100644 +--- a/dpdk/drivers/net/ice/base/ice_sched.c ++++ b/dpdk/drivers/net/ice/base/ice_sched.c +@@ -4774,12 +4774,12 @@ ice_sched_get_node_by_id_type(struct ice_port_info *pi, u32 id, + + case ICE_AGG_TYPE_Q: + /* The current implementation allows single queue to modify */ +- node = ice_sched_get_node(pi, id); ++ node = ice_sched_find_node_by_teid(pi->root, id); + break; + + case ICE_AGG_TYPE_QG: + /* The current implementation allows single qg to modify */ +- child_node = ice_sched_get_node(pi, id); ++ child_node = ice_sched_find_node_by_teid(pi->root, id); + if (!child_node) + break; + node = child_node->parent; +diff --git a/dpdk/drivers/net/ice/base/ice_switch.c b/dpdk/drivers/net/ice/base/ice_switch.c +index 1fee790c25..c0df3a1815 100644 +--- a/dpdk/drivers/net/ice/base/ice_switch.c ++++ b/dpdk/drivers/net/ice/base/ice_switch.c +@@ -2303,7 +2303,7 @@ ice_get_recp_frm_fw(struct ice_hw *hw, struct ice_sw_recipe *recps, u8 rid, + lkup_exts->field_mask[fv_word_idx] = + rg_entry->fv_mask[i]; + if (prot == ICE_META_DATA_ID_HW && +- off == ICE_TUN_FLAG_MDID_OFF) ++ off == ICE_TUN_FLAG_MDID_OFF(1)) + vlan = true; + fv_word_idx++; + } +@@ -6770,6 +6770,7 @@ static struct ice_protocol_entry ice_prot_id_tbl[ICE_PROTOCOL_LAST] = { + { ICE_GTP_NO_PAY, ICE_UDP_ILOS_HW }, + { ICE_VLAN_EX, ICE_VLAN_OF_HW }, + { ICE_VLAN_IN, ICE_VLAN_OL_HW }, ++ { ICE_FLG_DIR, ICE_META_DATA_ID_HW}, + }; + + /** +@@ -7488,9 +7489,10 @@ ice_get_fv(struct ice_hw *hw, struct ice_adv_lkup_elem *lkups, u16 lkups_cnt, + /** + * ice_tun_type_match_word - determine if tun type needs a match mask + * @tun_type: tunnel type ++ * @off: offset of packet flag + * @mask: mask to be used for the tunnel + */ +-static bool ice_tun_type_match_word(enum ice_sw_tunnel_type tun_type, u16 *mask) ++static bool ice_tun_type_match_word(enum ice_sw_tunnel_type tun_type, u16 *off, u16 *mask) + { + switch (tun_type) { + case ICE_SW_TUN_VXLAN_GPE: +@@ -7506,15 +7508,23 @@ static bool ice_tun_type_match_word(enum ice_sw_tunnel_type tun_type, u16 *mask) + case ICE_SW_TUN_PPPOE_IPV4_QINQ: + case ICE_SW_TUN_PPPOE_IPV6_QINQ: + *mask = ICE_TUN_FLAG_MASK; ++ *off = ICE_TUN_FLAG_MDID_OFF(1); ++ return true; ++ ++ case ICE_SW_TUN_AND_NON_TUN: ++ *mask = ICE_DIR_FLAG_MASK; ++ *off = ICE_TUN_FLAG_MDID_OFF(0); + return true; + + case ICE_SW_TUN_GENEVE_VLAN: + case ICE_SW_TUN_VXLAN_VLAN: + *mask = ICE_TUN_FLAG_MASK & ~ICE_TUN_FLAG_VLAN_MASK; ++ *off = ICE_TUN_FLAG_MDID_OFF(1); + return true; + + default: + *mask = 0; ++ *off = 0; + return false; + } + } +@@ -7529,16 +7539,18 @@ ice_add_special_words(struct ice_adv_rule_info *rinfo, + struct ice_prot_lkup_ext *lkup_exts) + { + u16 mask; ++ u16 off; + + /* If this is a tunneled packet, then add recipe index to match the +- * tunnel bit in the packet metadata flags. ++ * tunnel bit in the packet metadata flags. If this is a tun_and_non_tun ++ * packet, then add recipe index to match the direction bit in the flag. + */ +- if (ice_tun_type_match_word(rinfo->tun_type, &mask)) { ++ if (ice_tun_type_match_word(rinfo->tun_type, &off, &mask)) { + if (lkup_exts->n_val_words < ICE_MAX_CHAIN_WORDS) { + u8 word = lkup_exts->n_val_words++; + + lkup_exts->fv_words[word].prot_id = ICE_META_DATA_ID_HW; +- lkup_exts->fv_words[word].off = ICE_TUN_FLAG_MDID_OFF; ++ lkup_exts->fv_words[word].off = off; + lkup_exts->field_mask[word] = mask; + } else { + return ICE_ERR_MAX_LIMIT; +@@ -7779,6 +7791,7 @@ ice_get_compat_fv_bitmap(struct ice_hw *hw, struct ice_adv_rule_info *rinfo, + bool ice_is_prof_rule(enum ice_sw_tunnel_type type) + { + switch (type) { ++ case ICE_SW_TUN_AND_NON_TUN: + case ICE_SW_TUN_PROFID_IPV6_ESP: + case ICE_SW_TUN_PROFID_IPV6_AH: + case ICE_SW_TUN_PROFID_MAC_IPV6_L2TPV3: +@@ -7863,6 +7876,15 @@ ice_add_adv_recipe(struct ice_hw *hw, struct ice_adv_lkup_elem *lkups, + */ + ice_get_compat_fv_bitmap(hw, rinfo, fv_bitmap); + ++ /* If it is a packet to match any, add a lookup element to match direction ++ * flag of source interface. ++ */ ++ if (rinfo->tun_type == ICE_SW_TUN_AND_NON_TUN && ++ lkups_cnt < ICE_MAX_CHAIN_WORDS) { ++ lkups[lkups_cnt].type = ICE_FLG_DIR; ++ lkups_cnt++; ++ } ++ + status = ice_get_fv(hw, lkups, lkups_cnt, fv_bitmap, &rm->fv_list); + if (status) + goto err_unroll; +diff --git a/dpdk/drivers/net/ice/base/ice_switch.h b/dpdk/drivers/net/ice/base/ice_switch.h +index a2b3c80107..c67cd09d21 100644 +--- a/dpdk/drivers/net/ice/base/ice_switch.h ++++ b/dpdk/drivers/net/ice/base/ice_switch.h +@@ -203,7 +203,7 @@ struct ice_fltr_info { + + struct ice_update_recipe_lkup_idx_params { + u16 rid; +- u16 fv_idx; ++ u8 fv_idx; + bool ignore_valid; + u16 mask; + bool mask_valid; +diff --git a/dpdk/drivers/net/ice/ice_dcf.c b/dpdk/drivers/net/ice/ice_dcf.c +index cca1d7bf46..7f0c074b01 100644 +--- a/dpdk/drivers/net/ice/ice_dcf.c ++++ b/dpdk/drivers/net/ice/ice_dcf.c +@@ -864,7 +864,7 @@ ice_dcf_init_rss(struct ice_dcf_hw *hw) + j = 0; + hw->rss_lut[i] = j; + } +- /* send virtchnnl ops to configure rss*/ ++ /* send virtchnl ops to configure RSS */ + ret = ice_dcf_configure_rss_lut(hw); + if (ret) + return ret; +diff --git a/dpdk/drivers/net/ice/ice_dcf_ethdev.c b/dpdk/drivers/net/ice/ice_dcf_ethdev.c +index 28f7f7fb72..6e9e80c1df 100644 +--- a/dpdk/drivers/net/ice/ice_dcf_ethdev.c ++++ b/dpdk/drivers/net/ice/ice_dcf_ethdev.c +@@ -203,7 +203,7 @@ ice_dcf_config_rx_queues_irqs(struct rte_eth_dev *dev, + "vector %u are mapping to all Rx queues", + hw->msix_base); + } else { +- /* If Rx interrupt is reuquired, and we can use ++ /* If Rx interrupt is required, and we can use + * multi interrupts, then the vec is from 1 + */ + hw->nb_msix = RTE_MIN(hw->vf_res->max_vectors, +@@ -664,6 +664,8 @@ ice_dcf_dev_info_get(struct rte_eth_dev *dev, + dev_info->reta_size = hw->vf_res->rss_lut_size; + dev_info->flow_type_rss_offloads = ICE_RSS_OFFLOAD_ALL; + dev_info->dev_capa &= ~RTE_ETH_DEV_CAPA_FLOW_RULE_KEEP; ++ dev_info->max_mtu = dev_info->max_rx_pktlen - ICE_ETH_OVERHEAD; ++ dev_info->min_mtu = RTE_ETHER_MIN_MTU; + + dev_info->rx_offload_capa = + RTE_ETH_RX_OFFLOAD_VLAN_STRIP | +@@ -681,6 +683,7 @@ ice_dcf_dev_info_get(struct rte_eth_dev *dev, + RTE_ETH_TX_OFFLOAD_TCP_CKSUM | + RTE_ETH_TX_OFFLOAD_SCTP_CKSUM | + RTE_ETH_TX_OFFLOAD_OUTER_IPV4_CKSUM | ++ RTE_ETH_TX_OFFLOAD_OUTER_UDP_CKSUM | + RTE_ETH_TX_OFFLOAD_TCP_TSO | + RTE_ETH_TX_OFFLOAD_VXLAN_TNL_TSO | + RTE_ETH_TX_OFFLOAD_GRE_TNL_TSO | +@@ -956,6 +959,13 @@ ice_dcf_link_update(struct rte_eth_dev *dev, + return rte_eth_linkstatus_set(dev, &new_link); } -+void -+ofproto_ref(struct ofproto *ofproto) -+{ -+ ovs_refcount_ref(&ofproto->refcount); -+} -+ +bool -+ofproto_try_ref(struct ofproto *ofproto) -+{ -+ return ovs_refcount_try_ref_rcu(&ofproto->refcount); -+} -+ -+void -+ofproto_unref(struct ofproto *ofproto) ++ice_dcf_adminq_need_retry(struct ice_adapter *ad) +{ -+ if (ofproto && ovs_refcount_unref(&ofproto->refcount) == 1) { -+ ovsrcu_postpone(ofproto_destroy_defer__, ofproto); -+ } ++ return ad->hw.dcf_enabled && ++ !__atomic_load_n(&ad->dcf_state_on, __ATOMIC_RELAXED); +} + - void - ofproto_destroy(struct ofproto *p, bool del) - OVS_EXCLUDED(ofproto_mutex) + /* Add UDP tunneling port */ + static int + ice_dcf_dev_udp_tunnel_port_add(struct rte_eth_dev *dev, +@@ -1105,6 +1115,7 @@ static int + ice_dcf_dev_init(struct rte_eth_dev *eth_dev) { -- struct ofport *ofport, *next_ofport; -+ struct ofport *ofport; - struct ofport_usage *usage; + struct ice_dcf_adapter *adapter = eth_dev->data->dev_private; ++ struct ice_adapter *parent_adapter = &adapter->parent; + + eth_dev->dev_ops = &ice_dcf_eth_dev_ops; + eth_dev->rx_pkt_burst = ice_dcf_recv_pkts; +@@ -1116,9 +1127,13 @@ ice_dcf_dev_init(struct rte_eth_dev *eth_dev) + adapter->real_hw.vc_event_msg_cb = ice_dcf_handle_pf_event_msg; + if (ice_dcf_init_hw(eth_dev, &adapter->real_hw) != 0) { + PMD_INIT_LOG(ERR, "Failed to init DCF hardware"); ++ __atomic_store_n(&parent_adapter->dcf_state_on, false, ++ __ATOMIC_RELAXED); + return -1; + } - if (!p) { -@@ -1717,7 +1762,7 @@ ofproto_destroy(struct ofproto *p, bool del) - } ++ __atomic_store_n(&parent_adapter->dcf_state_on, true, __ATOMIC_RELAXED); ++ + if (ice_dcf_init_parent_adapter(eth_dev) != 0) { + PMD_INIT_LOG(ERR, "Failed to init DCF parent adapter"); + ice_dcf_uninit_hw(eth_dev, &adapter->real_hw); +diff --git a/dpdk/drivers/net/ice/ice_dcf_ethdev.h b/dpdk/drivers/net/ice/ice_dcf_ethdev.h +index 8510e37119..11a1305038 100644 +--- a/dpdk/drivers/net/ice/ice_dcf_ethdev.h ++++ b/dpdk/drivers/net/ice/ice_dcf_ethdev.h +@@ -64,5 +64,6 @@ int ice_dcf_vf_repr_init(struct rte_eth_dev *vf_rep_eth_dev, void *init_param); + int ice_dcf_vf_repr_uninit(struct rte_eth_dev *vf_rep_eth_dev); + int ice_dcf_vf_repr_init_vlan(struct rte_eth_dev *vf_rep_eth_dev); + void ice_dcf_vf_repr_stop_all(struct ice_dcf_adapter *dcf_adapter); ++bool ice_dcf_adminq_need_retry(struct ice_adapter *ad); + + #endif /* _ICE_DCF_ETHDEV_H_ */ +diff --git a/dpdk/drivers/net/ice/ice_dcf_parent.c b/dpdk/drivers/net/ice/ice_dcf_parent.c +index 1ff2c47172..2f96dedcce 100644 +--- a/dpdk/drivers/net/ice/ice_dcf_parent.c ++++ b/dpdk/drivers/net/ice/ice_dcf_parent.c +@@ -119,7 +119,9 @@ ice_dcf_vsi_update_service_handler(void *param) + { + struct ice_dcf_reset_event_param *reset_param = param; + struct ice_dcf_hw *hw = reset_param->dcf_hw; +- struct ice_dcf_adapter *adapter; ++ struct ice_dcf_adapter *adapter = ++ container_of(hw, struct ice_dcf_adapter, real_hw); ++ struct ice_adapter *parent_adapter = &adapter->parent; - ofproto_flush__(p, del); -- HMAP_FOR_EACH_SAFE (ofport, next_ofport, hmap_node, &p->ports) { -+ HMAP_FOR_EACH_SAFE (ofport, hmap_node, &p->ports) { - ofport_destroy(ofport, del); - } + pthread_detach(pthread_self()); -@@ -1736,8 +1781,7 @@ ofproto_destroy(struct ofproto *p, bool del) - p->connmgr = NULL; - ovs_mutex_unlock(&ofproto_mutex); +@@ -127,11 +129,12 @@ ice_dcf_vsi_update_service_handler(void *param) -- /* Destroying rules is deferred, must have 'ofproto' around for them. */ -- ovsrcu_postpone(ofproto_destroy_defer__, p); -+ ofproto_unref(p); - } + rte_spinlock_lock(&vsi_update_lock); - /* Destroys the datapath with the respective 'name' and 'type'. With the Linux -@@ -2782,7 +2826,7 @@ init_ports(struct ofproto *p) +- adapter = container_of(hw, struct ice_dcf_adapter, real_hw); +- +- if (!ice_dcf_handle_vsi_update_event(hw)) ++ if (!ice_dcf_handle_vsi_update_event(hw)) { ++ __atomic_store_n(&parent_adapter->dcf_state_on, true, ++ __ATOMIC_RELAXED); + ice_dcf_update_vf_vsi_map(&adapter->parent.hw, + hw->num_vfs, hw->vf_vsi_map); ++ } + + if (reset_param->vfr && adapter->repr_infos) { + struct rte_eth_dev *vf_rep_eth_dev = +@@ -224,6 +227,9 @@ ice_dcf_handle_pf_event_msg(struct ice_dcf_hw *dcf_hw, + uint8_t *msg, uint16_t msglen) { - struct ofproto_port_dump dump; - struct ofproto_port ofproto_port; -- struct shash_node *node, *next; -+ struct shash_node *node; + struct virtchnl_pf_event *pf_msg = (struct virtchnl_pf_event *)msg; ++ struct ice_dcf_adapter *adapter = ++ container_of(dcf_hw, struct ice_dcf_adapter, real_hw); ++ struct ice_adapter *parent_adapter = &adapter->parent; + + if (msglen < sizeof(struct virtchnl_pf_event)) { + PMD_DRV_LOG(DEBUG, "Invalid event message length : %u", msglen); +@@ -258,6 +264,8 @@ ice_dcf_handle_pf_event_msg(struct ice_dcf_hw *dcf_hw, + PMD_DRV_LOG(DEBUG, "VIRTCHNL_EVENT_DCF_VSI_MAP_UPDATE event : VF%u with VSI num %u", + pf_msg->event_data.vf_vsi_map.vf_id, + pf_msg->event_data.vf_vsi_map.vsi_id); ++ __atomic_store_n(&parent_adapter->dcf_state_on, false, ++ __ATOMIC_RELAXED); + start_vsi_reset_thread(dcf_hw, true, + pf_msg->event_data.vf_vsi_map.vf_id); + break; +@@ -332,7 +340,7 @@ ice_dcf_init_parent_hw(struct ice_hw *hw) + goto err_unroll_alloc; + + /* Initialize port_info struct with link information */ +- status = ice_aq_get_link_info(hw->port_info, false, NULL, NULL); ++ status = ice_aq_get_link_info(hw->port_info, true, NULL, NULL); + if (status) + goto err_unroll_alloc; + +diff --git a/dpdk/drivers/net/ice/ice_ethdev.c b/dpdk/drivers/net/ice/ice_ethdev.c +index 13a7a9702a..7df1b4ec19 100644 +--- a/dpdk/drivers/net/ice/ice_ethdev.c ++++ b/dpdk/drivers/net/ice/ice_ethdev.c +@@ -1264,7 +1264,7 @@ ice_handle_aq_msg(struct rte_eth_dev *dev) + * @param handle + * Pointer to interrupt handle. + * @param param +- * The address of parameter (struct rte_eth_dev *) regsitered before. ++ * The address of parameter (struct rte_eth_dev *) registered before. + * + * @return + * void +@@ -1627,7 +1627,7 @@ ice_setup_vsi(struct ice_pf *pf, enum ice_vsi_type type) + } - OFPROTO_PORT_FOR_EACH (&ofproto_port, &dump, p) { - const char *name = ofproto_port.name; -@@ -2813,7 +2857,7 @@ init_ports(struct ofproto *p) - } - } + /* At the beginning, only TC0. */ +- /* What we need here is the maximam number of the TX queues. ++ /* What we need here is the maximum number of the TX queues. + * Currently vsi->nb_qps means it. + * Correct it if any change. + */ +@@ -3235,7 +3235,8 @@ static int ice_init_rss(struct ice_pf *pf) + RTE_MIN(rss_conf->rss_key_len, + vsi->rss_key_size)); + +- rte_memcpy(key.standard_rss_key, vsi->rss_key, vsi->rss_key_size); ++ rte_memcpy(key.standard_rss_key, vsi->rss_key, ++ RTE_MIN(sizeof(key.standard_rss_key), vsi->rss_key_size)); + ret = ice_aq_set_rss_key(hw, vsi->idx, &key); + if (ret) + goto out; +@@ -3576,7 +3577,7 @@ ice_dev_start(struct rte_eth_dev *dev) + goto rx_err; + } -- SHASH_FOR_EACH_SAFE(node, next, &init_ofp_ports) { -+ SHASH_FOR_EACH_SAFE (node, &init_ofp_ports) { - struct iface_hint *iface_hint = node->data; +- /* enable Rx interrput and mapping Rx queue to interrupt vector */ ++ /* enable Rx interrupt and mapping Rx queue to interrupt vector */ + if (ice_rxq_intr_setup(dev)) + return -EIO; - if (!strcmp(iface_hint->br_name, p->name)) { -@@ -2929,6 +2973,9 @@ ofproto_rule_destroy__(struct rule *rule) - cls_rule_destroy(CONST_CAST(struct cls_rule *, &rule->cr)); - rule_actions_destroy(rule_get_actions(rule)); - ovs_mutex_destroy(&rule->mutex); -+ /* ofproto_unref() must be called first. It is possible because ofproto -+ * destruction is deferred by an RCU grace period. */ -+ ofproto_unref(rule->ofproto); - rule->ofproto->ofproto_class->rule_dealloc(rule); +@@ -3603,8 +3604,8 @@ ice_dev_start(struct rte_eth_dev *dev) + + ice_dev_set_link_up(dev); + +- /* Call get_link_info aq commond to enable/disable LSE */ +- ice_link_update(dev, 0); ++ /* Call get_link_info aq command to enable/disable LSE */ ++ ice_link_update(dev, 1); + + pf->adapter_stopped = false; + +@@ -5395,7 +5396,7 @@ ice_xstats_get(struct rte_eth_dev *dev, struct rte_eth_xstat *xstats, + count++; + } + +- /* Get individiual stats from ice_hw_port struct */ ++ /* Get individual stats from ice_hw_port struct */ + for (i = 0; i < ICE_NB_HW_PORT_XSTATS; i++) { + xstats[count].value = + *(uint64_t *)((char *)hw_stats + +@@ -5426,7 +5427,7 @@ static int ice_xstats_get_names(__rte_unused struct rte_eth_dev *dev, + count++; + } + +- /* Get individiual stats from ice_hw_port struct */ ++ /* Get individual stats from ice_hw_port struct */ + for (i = 0; i < ICE_NB_HW_PORT_XSTATS; i++) { + strlcpy(xstats_names[count].name, ice_hw_port_strings[i].name, + sizeof(xstats_names[count].name)); +@@ -5454,6 +5455,8 @@ ice_dev_udp_tunnel_port_add(struct rte_eth_dev *dev, + { + int ret = 0; + struct ice_hw *hw = ICE_DEV_PRIVATE_TO_HW(dev->data->dev_private); ++ struct ice_adapter *ad = ++ ICE_DEV_PRIVATE_TO_ADAPTER(dev->data->dev_private); + + if (udp_tunnel == NULL) + return -EINVAL; +@@ -5461,6 +5464,9 @@ ice_dev_udp_tunnel_port_add(struct rte_eth_dev *dev, + switch (udp_tunnel->prot_type) { + case RTE_ETH_TUNNEL_TYPE_VXLAN: + ret = ice_create_tunnel(hw, TNL_VXLAN, udp_tunnel->udp_port); ++ if (!ret && ad->psr != NULL) ++ ice_parser_vxlan_tunnel_set(ad->psr, ++ udp_tunnel->udp_port, true); + break; + default: + PMD_DRV_LOG(ERR, "Invalid tunnel type"); +@@ -5478,6 +5484,8 @@ ice_dev_udp_tunnel_port_del(struct rte_eth_dev *dev, + { + int ret = 0; + struct ice_hw *hw = ICE_DEV_PRIVATE_TO_HW(dev->data->dev_private); ++ struct ice_adapter *ad = ++ ICE_DEV_PRIVATE_TO_ADAPTER(dev->data->dev_private); + + if (udp_tunnel == NULL) + return -EINVAL; +@@ -5485,6 +5493,9 @@ ice_dev_udp_tunnel_port_del(struct rte_eth_dev *dev, + switch (udp_tunnel->prot_type) { + case RTE_ETH_TUNNEL_TYPE_VXLAN: + ret = ice_destroy_tunnel(hw, udp_tunnel->udp_port, 0); ++ if (!ret && ad->psr != NULL) ++ ice_parser_vxlan_tunnel_set(ad->psr, ++ udp_tunnel->udp_port, false); + break; + default: + PMD_DRV_LOG(ERR, "Invalid tunnel type"); +diff --git a/dpdk/drivers/net/ice/ice_ethdev.h b/dpdk/drivers/net/ice/ice_ethdev.h +index 2e3e45f3d7..ac56c3cc60 100644 +--- a/dpdk/drivers/net/ice/ice_ethdev.h ++++ b/dpdk/drivers/net/ice/ice_ethdev.h +@@ -531,6 +531,9 @@ struct ice_adapter { + uint64_t time_hw; + struct ice_fdir_prof_info fdir_prof_info[ICE_MAX_PTGS]; + struct ice_rss_prof_info rss_prof_info[ICE_MAX_PTGS]; ++ /* True if DCF state of the associated PF is on */ ++ bool dcf_state_on; ++ struct ice_parser *psr; + #ifdef RTE_ARCH_X86 + bool rx_use_avx2; + bool rx_use_avx512; +diff --git a/dpdk/drivers/net/ice/ice_fdir_filter.c b/dpdk/drivers/net/ice/ice_fdir_filter.c +index 13a2ac42df..72c8bd8f02 100644 +--- a/dpdk/drivers/net/ice/ice_fdir_filter.c ++++ b/dpdk/drivers/net/ice/ice_fdir_filter.c +@@ -1828,7 +1828,6 @@ ice_fdir_parse_pattern(__rte_unused struct ice_adapter *ad, + struct ice_fdir_v4 *p_v4 = NULL; + struct ice_fdir_v6 *p_v6 = NULL; + struct ice_parser_result rslt; +- struct ice_parser *psr; + uint8_t item_num = 0; + + for (item = pattern; item->type != RTE_FLOW_ITEM_TYPE_END; item++) { +@@ -1863,6 +1862,9 @@ ice_fdir_parse_pattern(__rte_unused struct ice_adapter *ad, + + switch (item_type) { + case RTE_FLOW_ITEM_TYPE_RAW: { ++ if (ad->psr == NULL) ++ return -rte_errno; ++ + raw_spec = item->spec; + raw_mask = item->mask; + +@@ -1870,11 +1872,11 @@ ice_fdir_parse_pattern(__rte_unused struct ice_adapter *ad, + break; + + /* convert raw spec & mask from byte string to int */ +- unsigned char *tmp_spec = ++ unsigned char *spec_pattern = + (uint8_t *)(uintptr_t)raw_spec->pattern; +- unsigned char *tmp_mask = ++ unsigned char *mask_pattern = + (uint8_t *)(uintptr_t)raw_mask->pattern; +- uint16_t udp_port = 0; ++ uint8_t *tmp_spec, *tmp_mask; + uint16_t tmp_val = 0; + uint8_t pkt_len = 0; + uint8_t tmp = 0; +@@ -1885,8 +1887,18 @@ ice_fdir_parse_pattern(__rte_unused struct ice_adapter *ad, + pkt_len) + return -rte_errno; + ++ tmp_spec = rte_zmalloc(NULL, pkt_len / 2, 0); ++ if (!tmp_spec) ++ return -rte_errno; ++ ++ tmp_mask = rte_zmalloc(NULL, pkt_len / 2, 0); ++ if (!tmp_mask) { ++ rte_free(tmp_spec); ++ return -rte_errno; ++ } ++ + for (i = 0, j = 0; i < pkt_len; i += 2, j++) { +- tmp = tmp_spec[i]; ++ tmp = spec_pattern[i]; + if (tmp >= 'a' && tmp <= 'f') + tmp_val = tmp - 'a' + 10; + if (tmp >= 'A' && tmp <= 'F') +@@ -1895,7 +1907,7 @@ ice_fdir_parse_pattern(__rte_unused struct ice_adapter *ad, + tmp_val = tmp - '0'; + + tmp_val *= 16; +- tmp = tmp_spec[i + 1]; ++ tmp = spec_pattern[i + 1]; + if (tmp >= 'a' && tmp <= 'f') + tmp_spec[j] = tmp_val + tmp - 'a' + 10; + if (tmp >= 'A' && tmp <= 'F') +@@ -1903,7 +1915,7 @@ ice_fdir_parse_pattern(__rte_unused struct ice_adapter *ad, + if (tmp >= '0' && tmp <= '9') + tmp_spec[j] = tmp_val + tmp - '0'; + +- tmp = tmp_mask[i]; ++ tmp = mask_pattern[i]; + if (tmp >= 'a' && tmp <= 'f') + tmp_val = tmp - 'a' + 10; + if (tmp >= 'A' && tmp <= 'F') +@@ -1912,7 +1924,7 @@ ice_fdir_parse_pattern(__rte_unused struct ice_adapter *ad, + tmp_val = tmp - '0'; + + tmp_val *= 16; +- tmp = tmp_mask[i + 1]; ++ tmp = mask_pattern[i + 1]; + if (tmp >= 'a' && tmp <= 'f') + tmp_mask[j] = tmp_val + tmp - 'a' + 10; + if (tmp >= 'A' && tmp <= 'F') +@@ -1923,15 +1935,8 @@ ice_fdir_parse_pattern(__rte_unused struct ice_adapter *ad, + + pkt_len /= 2; + +- if (ice_parser_create(&ad->hw, &psr)) +- return -rte_errno; +- if (ice_get_open_tunnel_port(&ad->hw, TNL_VXLAN, +- &udp_port)) +- ice_parser_vxlan_tunnel_set(psr, udp_port, +- true); +- if (ice_parser_run(psr, tmp_spec, pkt_len, &rslt)) ++ if (ice_parser_run(ad->psr, tmp_spec, pkt_len, &rslt)) + return -rte_errno; +- ice_parser_destroy(psr); + + if (!tmp_mask) + return -rte_errno; +@@ -1955,6 +1960,8 @@ ice_fdir_parse_pattern(__rte_unused struct ice_adapter *ad, + + filter->parser_ena = true; + ++ rte_free(tmp_spec); ++ rte_free(tmp_mask); + break; + } + +@@ -2037,6 +2044,14 @@ ice_fdir_parse_pattern(__rte_unused struct ice_adapter *ad, + return -rte_errno; + } + ++ /* Mask for IPv4 src/dst addrs not supported */ ++ if (ipv4_mask->hdr.src_addr && ++ ipv4_mask->hdr.src_addr != UINT32_MAX) ++ return -rte_errno; ++ if (ipv4_mask->hdr.dst_addr && ++ ipv4_mask->hdr.dst_addr != UINT32_MAX) ++ return -rte_errno; ++ + if (ipv4_mask->hdr.dst_addr == UINT32_MAX) + *input_set |= ICE_INSET_IPV4_DST; + if (ipv4_mask->hdr.src_addr == UINT32_MAX) +@@ -2178,6 +2193,14 @@ ice_fdir_parse_pattern(__rte_unused struct ice_adapter *ad, + return -rte_errno; + } + ++ /* Mask for TCP src/dst ports not supported */ ++ if (tcp_mask->hdr.src_port && ++ tcp_mask->hdr.src_port != UINT16_MAX) ++ return -rte_errno; ++ if (tcp_mask->hdr.dst_port && ++ tcp_mask->hdr.dst_port != UINT16_MAX) ++ return -rte_errno; ++ + if (tcp_mask->hdr.src_port == UINT16_MAX) + *input_set |= ICE_INSET_TCP_SRC_PORT; + if (tcp_mask->hdr.dst_port == UINT16_MAX) +@@ -2217,6 +2240,14 @@ ice_fdir_parse_pattern(__rte_unused struct ice_adapter *ad, + return -rte_errno; + } + ++ /* Mask for UDP src/dst ports not supported */ ++ if (udp_mask->hdr.src_port && ++ udp_mask->hdr.src_port != UINT16_MAX) ++ return -rte_errno; ++ if (udp_mask->hdr.dst_port && ++ udp_mask->hdr.dst_port != UINT16_MAX) ++ return -rte_errno; ++ + if (udp_mask->hdr.src_port == UINT16_MAX) + *input_set |= ICE_INSET_UDP_SRC_PORT; + if (udp_mask->hdr.dst_port == UINT16_MAX) +@@ -2254,6 +2285,14 @@ ice_fdir_parse_pattern(__rte_unused struct ice_adapter *ad, + return -rte_errno; + } + ++ /* Mask for SCTP src/dst ports not supported */ ++ if (sctp_mask->hdr.src_port && ++ sctp_mask->hdr.src_port != UINT16_MAX) ++ return -rte_errno; ++ if (sctp_mask->hdr.dst_port && ++ sctp_mask->hdr.dst_port != UINT16_MAX) ++ return -rte_errno; ++ + if (sctp_mask->hdr.src_port == UINT16_MAX) + *input_set |= ICE_INSET_SCTP_SRC_PORT; + if (sctp_mask->hdr.dst_port == UINT16_MAX) +diff --git a/dpdk/drivers/net/ice/ice_generic_flow.c b/dpdk/drivers/net/ice/ice_generic_flow.c +index c673feb7a6..f9be3a5c94 100644 +--- a/dpdk/drivers/net/ice/ice_generic_flow.c ++++ b/dpdk/drivers/net/ice/ice_generic_flow.c +@@ -1826,6 +1826,9 @@ ice_flow_init(struct ice_adapter *ad) + TAILQ_INIT(&pf->dist_parser_list); + rte_spinlock_init(&pf->flow_ops_lock); + ++ if (ice_parser_create(&ad->hw, &ad->psr) != ICE_SUCCESS) ++ PMD_INIT_LOG(WARNING, "Failed to initialize DDP parser, raw packet filter will not be supported"); ++ + RTE_TAILQ_FOREACH_SAFE(engine, &engine_list, node, temp) { + if (engine->init == NULL) { + PMD_INIT_LOG(ERR, "Invalid engine type (%d)", +@@ -1880,6 +1883,11 @@ ice_flow_uninit(struct ice_adapter *ad) + TAILQ_REMOVE(&pf->dist_parser_list, p_parser, node); + rte_free(p_parser); + } ++ ++ if (ad->psr != NULL) { ++ ice_parser_destroy(ad->psr); ++ ad->psr = NULL; ++ } } -@@ -3069,6 +3116,9 @@ group_destroy_cb(struct ofgroup *group) - &group->props)); - ofputil_bucket_list_destroy(CONST_CAST(struct ovs_list *, - &group->buckets)); -+ /* ofproto_unref() must be called first. It is possible because ofproto -+ * destruction is deferred by an RCU grace period. */ -+ ofproto_unref(group->ofproto); - group->ofproto->ofproto_class->group_dealloc(group); + static struct ice_parser_list * +@@ -2515,7 +2523,9 @@ ice_flow_flush(struct rte_eth_dev *dev, + ret = ice_flow_destroy(dev, p_flow, error); + if (ret) { + PMD_DRV_LOG(ERR, "Failed to flush flows"); +- return -EINVAL; ++ if (ret != -EAGAIN) ++ ret = -EINVAL; ++ return ret; + } + } + +diff --git a/dpdk/drivers/net/ice/ice_hash.c b/dpdk/drivers/net/ice/ice_hash.c +index afbb357fa3..f35727856e 100644 +--- a/dpdk/drivers/net/ice/ice_hash.c ++++ b/dpdk/drivers/net/ice/ice_hash.c +@@ -653,13 +653,15 @@ ice_hash_parse_raw_pattern(struct ice_adapter *ad, + const struct rte_flow_item_raw *raw_spec, *raw_mask; + struct ice_parser_profile prof; + struct ice_parser_result rslt; +- struct ice_parser *psr; + uint8_t *pkt_buf, *msk_buf; + uint8_t spec_len, pkt_len; + uint8_t tmp_val = 0; + uint8_t tmp_c = 0; + int i, j; + ++ if (ad->psr == NULL) ++ return -rte_errno; ++ + raw_spec = item->spec; + raw_mask = item->mask; + +@@ -713,11 +715,8 @@ ice_hash_parse_raw_pattern(struct ice_adapter *ad, + msk_buf[j] = tmp_val * 16 + tmp_c - '0'; + } + +- if (ice_parser_create(&ad->hw, &psr)) +- return -rte_errno; +- if (ice_parser_run(psr, pkt_buf, pkt_len, &rslt)) ++ if (ice_parser_run(ad->psr, pkt_buf, pkt_len, &rslt)) + return -rte_errno; +- ice_parser_destroy(psr); + + if (ice_parser_profile_init(&rslt, pkt_buf, msk_buf, + pkt_len, ICE_BLK_RSS, true, &prof)) +diff --git a/dpdk/drivers/net/ice/ice_rxtx.c b/dpdk/drivers/net/ice/ice_rxtx.c +index f6d8564ab8..71e5c6f5d6 100644 +--- a/dpdk/drivers/net/ice/ice_rxtx.c ++++ b/dpdk/drivers/net/ice/ice_rxtx.c +@@ -163,6 +163,8 @@ ice_rxd_to_pkt_fields_by_comms_aux_v1(struct ice_rx_queue *rxq, + *RTE_NET_ICE_DYNF_PROTO_XTR_METADATA(mb) = metadata; + } + } ++#else ++ RTE_SET_USED(rxq); + #endif } -@@ -5271,10 +5321,15 @@ ofproto_rule_create(struct ofproto *ofproto, struct cls_rule *cr, - struct rule *rule; - enum ofperr error; +@@ -201,6 +203,8 @@ ice_rxd_to_pkt_fields_by_comms_aux_v2(struct ice_rx_queue *rxq, + *RTE_NET_ICE_DYNF_PROTO_XTR_METADATA(mb) = metadata; + } + } ++#else ++ RTE_SET_USED(rxq); + #endif + } -+ if (!ofproto_try_ref(ofproto)) { -+ return OFPERR_OFPFMFC_UNKNOWN; -+ } +@@ -1118,7 +1122,7 @@ ice_rx_queue_setup(struct rte_eth_dev *dev, + rxq->proto_xtr = pf->proto_xtr != NULL ? + pf->proto_xtr[queue_idx] : PROTO_XTR_NONE; + +- /* Allocate the maximun number of RX ring hardware descriptor. */ ++ /* Allocate the maximum number of RX ring hardware descriptor. */ + len = ICE_MAX_RING_DESC; + + /** +@@ -1248,7 +1252,7 @@ ice_tx_queue_setup(struct rte_eth_dev *dev, + tx_free_thresh = (uint16_t)(tx_conf->tx_free_thresh ? + tx_conf->tx_free_thresh : + ICE_DEFAULT_TX_FREE_THRESH); +- /* force tx_rs_thresh to adapt an aggresive tx_free_thresh */ ++ /* force tx_rs_thresh to adapt an aggressive tx_free_thresh */ + tx_rs_thresh = + (ICE_DEFAULT_TX_RSBIT_THRESH + tx_free_thresh > nb_desc) ? + nb_desc - tx_free_thresh : ICE_DEFAULT_TX_RSBIT_THRESH; +@@ -1554,6 +1558,9 @@ ice_rxd_to_vlan_tci(struct rte_mbuf *mb, volatile union ice_rx_flex_desc *rxdp) + #if (ICE_LOOK_AHEAD != 8) + #error "PMD ICE: ICE_LOOK_AHEAD must be 8\n" + #endif + - /* Allocate new rule. */ - rule = ofproto->ofproto_class->rule_alloc(); - if (!rule) { - cls_rule_destroy(cr); -+ ofproto_unref(ofproto); - VLOG_WARN_RL(&rl, "%s: failed to allocate a rule.", ofproto->name); - return OFPERR_OFPFMFC_UNKNOWN; - } -@@ -6797,9 +6852,9 @@ static void - meter_delete_all(struct ofproto *ofproto) - OVS_REQUIRES(ofproto_mutex) ++#define ICE_PTP_TS_VALID 0x1 ++ + static inline int + ice_rx_scan_hw_ring(struct ice_rx_queue *rxq) { -- struct meter *meter, *next; -+ struct meter *meter; +@@ -1567,9 +1574,10 @@ ice_rx_scan_hw_ring(struct ice_rx_queue *rxq) + uint64_t pkt_flags = 0; + uint32_t *ptype_tbl = rxq->vsi->adapter->ptype_tbl; + #ifndef RTE_LIBRTE_ICE_16BYTE_RX_DESC ++ bool is_tsinit = false; ++ uint64_t ts_ns; + struct ice_vsi *vsi = rxq->vsi; + struct ice_hw *hw = ICE_VSI_TO_HW(vsi); +- uint64_t ts_ns; + struct ice_adapter *ad = rxq->vsi->adapter; + #endif + rxdp = &rxq->rx_ring[rxq->rx_tail]; +@@ -1581,8 +1589,14 @@ ice_rx_scan_hw_ring(struct ice_rx_queue *rxq) + if (!(stat_err0 & (1 << ICE_RX_FLEX_DESC_STATUS0_DD_S))) + return 0; -- HMAP_FOR_EACH_SAFE (meter, next, node, &ofproto->meters) { -+ HMAP_FOR_EACH_SAFE (meter, node, &ofproto->meters) { - hmap_remove(&ofproto->meters, &meter->node); - meter_destroy(ofproto, meter); - } -@@ -7339,8 +7394,13 @@ init_group(struct ofproto *ofproto, const struct ofputil_group_mod *gm, - return OFPERR_OFPGMFC_BAD_TYPE; - } +- if (rxq->offloads & RTE_ETH_RX_OFFLOAD_TIMESTAMP) +- rxq->hw_register_set = 1; ++#ifndef RTE_LIBRTE_ICE_16BYTE_RX_DESC ++ if (rxq->offloads & RTE_ETH_RX_OFFLOAD_TIMESTAMP) { ++ uint64_t sw_cur_time = rte_get_timer_cycles() / (rte_get_timer_hz() / 1000); ++ ++ if (unlikely(sw_cur_time - rxq->hw_time_update > 4)) ++ is_tsinit = 1; ++ } ++#endif -+ if (!ofproto_try_ref(ofproto)) { -+ return OFPERR_OFPFMFC_UNKNOWN; -+ } + /** + * Scan LOOK_AHEAD descriptors at a time to determine which +@@ -1618,14 +1632,26 @@ ice_rx_scan_hw_ring(struct ice_rx_queue *rxq) + rxd_to_pkt_fields_ops[rxq->rxdid](rxq, mb, &rxdp[j]); + #ifndef RTE_LIBRTE_ICE_16BYTE_RX_DESC + if (ice_timestamp_dynflag > 0) { +- ts_ns = ice_tstamp_convert_32b_64b(hw, ad, +- rxq->hw_register_set, +- rte_le_to_cpu_32(rxdp[j].wb.flex_ts.ts_high)); +- rxq->hw_register_set = 0; ++ rxq->time_high = ++ rte_le_to_cpu_32(rxdp[j].wb.flex_ts.ts_high); ++ if (unlikely(is_tsinit)) { ++ ts_ns = ice_tstamp_convert_32b_64b(hw, ad, 1, ++ rxq->time_high); ++ rxq->hw_time_low = (uint32_t)ts_ns; ++ rxq->hw_time_high = (uint32_t)(ts_ns >> 32); ++ is_tsinit = false; ++ } else { ++ if (rxq->time_high < rxq->hw_time_low) ++ rxq->hw_time_high += 1; ++ ts_ns = (uint64_t)rxq->hw_time_high << 32 | rxq->time_high; ++ rxq->hw_time_low = rxq->time_high; ++ } ++ rxq->hw_time_update = rte_get_timer_cycles() / ++ (rte_get_timer_hz() / 1000); + *RTE_MBUF_DYNFIELD(mb, +- ice_timestamp_dynfield_offset, +- rte_mbuf_timestamp_t *) = ts_ns; +- mb->ol_flags |= ice_timestamp_dynflag; ++ ice_timestamp_dynfield_offset, ++ rte_mbuf_timestamp_t *) = ts_ns; ++ pkt_flags |= ice_timestamp_dynflag; + } + + if (ad->ptp_ena && ((mb->packet_type & +@@ -1634,6 +1660,10 @@ ice_rx_scan_hw_ring(struct ice_rx_queue *rxq) + rte_le_to_cpu_32(rxdp[j].wb.flex_ts.ts_high); + mb->timesync = rxq->queue_id; + pkt_flags |= RTE_MBUF_F_RX_IEEE1588_PTP; ++ if (rxdp[j].wb.time_stamp_low & ++ ICE_PTP_TS_VALID) ++ pkt_flags |= ++ RTE_MBUF_F_RX_IEEE1588_TMST; + } + #endif + mb->ol_flags |= pkt_flags; +@@ -1714,7 +1744,7 @@ ice_rx_alloc_bufs(struct ice_rx_queue *rxq) + rxdp[i].read.pkt_addr = dma_addr; + } + +- /* Update rx tail regsiter */ ++ /* Update Rx tail register */ + ICE_PCI_REG_WRITE(rxq->qrx_tail, rxq->rx_free_trigger); + + rxq->rx_free_trigger = +@@ -1820,14 +1850,19 @@ ice_recv_scattered_pkts(void *rx_queue, + uint64_t pkt_flags; + uint32_t *ptype_tbl = rxq->vsi->adapter->ptype_tbl; + #ifndef RTE_LIBRTE_ICE_16BYTE_RX_DESC ++ bool is_tsinit = false; ++ uint64_t ts_ns; + struct ice_vsi *vsi = rxq->vsi; + struct ice_hw *hw = ICE_VSI_TO_HW(vsi); +- uint64_t ts_ns; + struct ice_adapter *ad = rxq->vsi->adapter; +-#endif + +- if (rxq->offloads & RTE_ETH_RX_OFFLOAD_TIMESTAMP) +- rxq->hw_register_set = 1; ++ if (rxq->offloads & RTE_ETH_RX_OFFLOAD_TIMESTAMP) { ++ uint64_t sw_cur_time = rte_get_timer_cycles() / (rte_get_timer_hz() / 1000); + - *ofgroup = ofproto->ofproto_class->group_alloc(); - if (!*ofgroup) { -+ ofproto_unref(ofproto); - VLOG_WARN_RL(&rl, "%s: failed to allocate group", ofproto->name); - return OFPERR_OFPGMFC_OUT_OF_GROUPS; - } -@@ -7377,6 +7437,7 @@ init_group(struct ofproto *ofproto, const struct ofputil_group_mod *gm, - &(*ofgroup)->props)); - ofputil_bucket_list_destroy(CONST_CAST(struct ovs_list *, - &(*ofgroup)->buckets)); -+ ofproto_unref(ofproto); - ofproto->ofproto_class->group_dealloc(*ofgroup); - } - return error; -@@ -8902,7 +8963,7 @@ eviction_group_hash_rule(struct rule *rule) - hash = table->eviction_group_id_basis; - miniflow_expand(rule->cr.match.flow, &flow); - for (sf = table->eviction_fields; -- sf < &table->eviction_fields[table->n_eviction_fields]; -+ sf && sf < &table->eviction_fields[table->n_eviction_fields]; - sf++) - { - if (mf_are_prereqs_ok(sf->field, &flow, NULL)) { -@@ -9138,8 +9199,8 @@ oftable_configure_eviction(struct oftable *table, unsigned int eviction, ++ if (unlikely(sw_cur_time - rxq->hw_time_update > 4)) ++ is_tsinit = true; ++ } ++#endif - /* Destroy existing eviction groups, then destroy and recreate data - * structures to recover memory. */ -- struct eviction_group *evg, *next; -- HMAP_FOR_EACH_SAFE (evg, next, id_node, &table->eviction_groups_by_id) { -+ struct eviction_group *evg; -+ HMAP_FOR_EACH_SAFE (evg, id_node, &table->eviction_groups_by_id) { - eviction_group_destroy(table, evg); - } - hmap_destroy(&table->eviction_groups_by_id); -diff --git a/ofproto/ofproto.h b/ofproto/ofproto.h -index b0262da2df..4e15167ab7 100644 ---- a/ofproto/ofproto.h -+++ b/ofproto/ofproto.h -@@ -563,6 +563,10 @@ int ofproto_port_get_cfm_status(const struct ofproto *, - enum ofputil_table_miss ofproto_table_get_miss_config(const struct ofproto *, - uint8_t table_id); + while (nb_rx < nb_pkts) { + rxdp = &rx_ring[rx_id]; +@@ -1940,14 +1975,25 @@ ice_recv_scattered_pkts(void *rx_queue, + pkt_flags = ice_rxd_error_to_pkt_flags(rx_stat_err0); + #ifndef RTE_LIBRTE_ICE_16BYTE_RX_DESC + if (ice_timestamp_dynflag > 0) { +- ts_ns = ice_tstamp_convert_32b_64b(hw, ad, +- rxq->hw_register_set, +- rte_le_to_cpu_32(rxd.wb.flex_ts.ts_high)); +- rxq->hw_register_set = 0; +- *RTE_MBUF_DYNFIELD(first_seg, +- ice_timestamp_dynfield_offset, +- rte_mbuf_timestamp_t *) = ts_ns; +- first_seg->ol_flags |= ice_timestamp_dynflag; ++ rxq->time_high = ++ rte_le_to_cpu_32(rxd.wb.flex_ts.ts_high); ++ if (unlikely(is_tsinit)) { ++ ts_ns = ice_tstamp_convert_32b_64b(hw, ad, 1, rxq->time_high); ++ rxq->hw_time_low = (uint32_t)ts_ns; ++ rxq->hw_time_high = (uint32_t)(ts_ns >> 32); ++ is_tsinit = false; ++ } else { ++ if (rxq->time_high < rxq->hw_time_low) ++ rxq->hw_time_high += 1; ++ ts_ns = (uint64_t)rxq->hw_time_high << 32 | rxq->time_high; ++ rxq->hw_time_low = rxq->time_high; ++ } ++ rxq->hw_time_update = rte_get_timer_cycles() / ++ (rte_get_timer_hz() / 1000); ++ *RTE_MBUF_DYNFIELD(rxm, ++ (ice_timestamp_dynfield_offset), ++ rte_mbuf_timestamp_t *) = ts_ns; ++ pkt_flags |= ice_timestamp_dynflag; + } + + if (ad->ptp_ena && ((first_seg->packet_type & RTE_PTYPE_L2_MASK) +@@ -1976,7 +2022,7 @@ ice_recv_scattered_pkts(void *rx_queue, + * threshold of the queue, advance the Receive Descriptor Tail (RDT) + * register. Update the RDT with the value of the last processed RX + * descriptor minus 1, to guarantee that the RDT register is never +- * equal to the RDH register, which creates a "full" ring situtation ++ * equal to the RDH register, which creates a "full" ring situation + * from the hardware point of view. + */ + nb_hold = (uint16_t)(nb_hold + rxq->nb_rx_hold); +@@ -2314,14 +2360,19 @@ ice_recv_pkts(void *rx_queue, + uint64_t pkt_flags; + uint32_t *ptype_tbl = rxq->vsi->adapter->ptype_tbl; + #ifndef RTE_LIBRTE_ICE_16BYTE_RX_DESC ++ bool is_tsinit = false; ++ uint64_t ts_ns; + struct ice_vsi *vsi = rxq->vsi; + struct ice_hw *hw = ICE_VSI_TO_HW(vsi); +- uint64_t ts_ns; + struct ice_adapter *ad = rxq->vsi->adapter; +-#endif -+void ofproto_ref(struct ofproto *); -+void ofproto_unref(struct ofproto *); -+bool ofproto_try_ref(struct ofproto *); +- if (rxq->offloads & RTE_ETH_RX_OFFLOAD_TIMESTAMP) +- rxq->hw_register_set = 1; ++ if (rxq->offloads & RTE_ETH_RX_OFFLOAD_TIMESTAMP) { ++ uint64_t sw_cur_time = rte_get_timer_cycles() / (rte_get_timer_hz() / 1000); + - #ifdef __cplusplus - } - #endif -diff --git a/ovsdb/.gitignore b/ovsdb/.gitignore -index fbcefafc6e..a4f9d38f11 100644 ---- a/ovsdb/.gitignore -+++ b/ovsdb/.gitignore -@@ -1,5 +1,7 @@ - /_server.ovsschema.inc - /_server.ovsschema.stamp -+/local-config.ovsschema.stamp -+/ovsdb.local-config.5 - /ovsdb-client - /ovsdb-client.1 - /ovsdb-doc -diff --git a/ovsdb/automake.mk b/ovsdb/automake.mk -index 62cc02686f..3b3140102b 100644 ---- a/ovsdb/automake.mk -+++ b/ovsdb/automake.mk -@@ -148,4 +148,25 @@ ovsdb/ovsdb-server.5: \ - $(srcdir)/ovsdb/_server.xml > $@.tmp && \ - mv $@.tmp $@ ++ if (unlikely(sw_cur_time - rxq->hw_time_update > 4)) ++ is_tsinit = 1; ++ } ++#endif -+EXTRA_DIST += ovsdb/local-config.ovsschema -+pkgdata_DATA += ovsdb/local-config.ovsschema + while (nb_rx < nb_pkts) { + rxdp = &rx_ring[rx_id]; +@@ -2375,14 +2426,25 @@ ice_recv_pkts(void *rx_queue, + pkt_flags = ice_rxd_error_to_pkt_flags(rx_stat_err0); + #ifndef RTE_LIBRTE_ICE_16BYTE_RX_DESC + if (ice_timestamp_dynflag > 0) { +- ts_ns = ice_tstamp_convert_32b_64b(hw, ad, +- rxq->hw_register_set, +- rte_le_to_cpu_32(rxd.wb.flex_ts.ts_high)); +- rxq->hw_register_set = 0; ++ rxq->time_high = ++ rte_le_to_cpu_32(rxd.wb.flex_ts.ts_high); ++ if (unlikely(is_tsinit)) { ++ ts_ns = ice_tstamp_convert_32b_64b(hw, ad, 1, rxq->time_high); ++ rxq->hw_time_low = (uint32_t)ts_ns; ++ rxq->hw_time_high = (uint32_t)(ts_ns >> 32); ++ is_tsinit = false; ++ } else { ++ if (rxq->time_high < rxq->hw_time_low) ++ rxq->hw_time_high += 1; ++ ts_ns = (uint64_t)rxq->hw_time_high << 32 | rxq->time_high; ++ rxq->hw_time_low = rxq->time_high; ++ } ++ rxq->hw_time_update = rte_get_timer_cycles() / ++ (rte_get_timer_hz() / 1000); + *RTE_MBUF_DYNFIELD(rxm, +- ice_timestamp_dynfield_offset, +- rte_mbuf_timestamp_t *) = ts_ns; +- rxm->ol_flags |= ice_timestamp_dynflag; ++ (ice_timestamp_dynfield_offset), ++ rte_mbuf_timestamp_t *) = ts_ns; ++ pkt_flags |= ice_timestamp_dynflag; + } + + if (ad->ptp_ena && ((rxm->packet_type & RTE_PTYPE_L2_MASK) == +@@ -2397,6 +2459,7 @@ ice_recv_pkts(void *rx_queue, + /* copy old mbuf to rx_pkts */ + rx_pkts[nb_rx++] = rxm; + } + -+# Version checking for local-config.ovsschema. -+ALL_LOCAL += ovsdb/local-config.ovsschema.stamp -+ovsdb/local-config.ovsschema.stamp: ovsdb/local-config.ovsschema -+ $(srcdir)/build-aux/cksum-schema-check $? $@ -+CLEANFILES += ovsdb/local-config.ovsschema.stamp + rxq->rx_tail = rx_id; + /** + * If the number of free RX descriptors is greater than the RX free +@@ -2493,15 +2556,15 @@ ice_txd_enable_checksum(uint64_t ol_flags, + if (ol_flags & RTE_MBUF_F_TX_IP_CKSUM) { + *td_cmd |= ICE_TX_DESC_CMD_IIPT_IPV4_CSUM; + *td_offset |= (tx_offload.l3_len >> 2) << +- ICE_TX_DESC_LEN_IPLEN_S; ++ ICE_TX_DESC_LEN_IPLEN_S; + } else if (ol_flags & RTE_MBUF_F_TX_IPV4) { + *td_cmd |= ICE_TX_DESC_CMD_IIPT_IPV4; + *td_offset |= (tx_offload.l3_len >> 2) << +- ICE_TX_DESC_LEN_IPLEN_S; ++ ICE_TX_DESC_LEN_IPLEN_S; + } else if (ol_flags & RTE_MBUF_F_TX_IPV6) { + *td_cmd |= ICE_TX_DESC_CMD_IIPT_IPV6; + *td_offset |= (tx_offload.l3_len >> 2) << +- ICE_TX_DESC_LEN_IPLEN_S; ++ ICE_TX_DESC_LEN_IPLEN_S; + } + + if (ol_flags & RTE_MBUF_F_TX_TCP_SEG) { +@@ -3117,7 +3180,7 @@ tx_xmit_pkts(struct ice_tx_queue *txq, + ice_tx_fill_hw_ring(txq, tx_pkts + n, (uint16_t)(nb_pkts - n)); + txq->tx_tail = (uint16_t)(txq->tx_tail + (nb_pkts - n)); + +- /* Determin if RS bit needs to be set */ ++ /* Determine if RS bit needs to be set */ + if (txq->tx_tail > txq->tx_next_rs) { + txr[txq->tx_next_rs].cmd_type_offset_bsz |= + rte_cpu_to_le_64(((uint64_t)ICE_TX_DESC_CMD_RS) << +@@ -3541,8 +3604,9 @@ static const struct { + { ice_xmit_pkts_vec_avx512, "Vector AVX512" }, + { ice_xmit_pkts_vec_avx512_offload, "Offload Vector AVX512" }, + #endif +- { ice_xmit_pkts_vec_avx2, "Vector AVX2" }, +- { ice_xmit_pkts_vec, "Vector SSE" }, ++ { ice_xmit_pkts_vec_avx2, "Vector AVX2" }, ++ { ice_xmit_pkts_vec_avx2_offload, "Offload Vector AVX2" }, ++ { ice_xmit_pkts_vec, "Vector SSE" }, + #endif + }; + +diff --git a/dpdk/drivers/net/ice/ice_rxtx.h b/dpdk/drivers/net/ice/ice_rxtx.h +index bb18a01951..f5337d5284 100644 +--- a/dpdk/drivers/net/ice/ice_rxtx.h ++++ b/dpdk/drivers/net/ice/ice_rxtx.h +@@ -95,6 +95,9 @@ struct ice_rx_queue { + uint32_t time_high; + uint32_t hw_register_set; + const struct rte_memzone *mz; ++ uint32_t hw_time_high; /* high 32 bits of timestamp */ ++ uint32_t hw_time_low; /* low 32 bits of timestamp */ ++ uint64_t hw_time_update; /* SW time of HW record updating */ + }; + + struct ice_tx_entry { +diff --git a/dpdk/drivers/net/ice/ice_rxtx_vec_common.h b/dpdk/drivers/net/ice/ice_rxtx_vec_common.h +index dfe60c81d9..2dd2d83650 100644 +--- a/dpdk/drivers/net/ice/ice_rxtx_vec_common.h ++++ b/dpdk/drivers/net/ice/ice_rxtx_vec_common.h +@@ -250,7 +250,8 @@ ice_rxq_vec_setup_default(struct ice_rx_queue *rxq) + #define ICE_TX_NO_VECTOR_FLAGS ( \ + RTE_ETH_TX_OFFLOAD_MULTI_SEGS | \ + RTE_ETH_TX_OFFLOAD_OUTER_IPV4_CKSUM | \ +- RTE_ETH_TX_OFFLOAD_TCP_TSO) ++ RTE_ETH_TX_OFFLOAD_TCP_TSO | \ ++ RTE_ETH_TX_OFFLOAD_OUTER_UDP_CKSUM) + + #define ICE_TX_VECTOR_OFFLOAD ( \ + RTE_ETH_TX_OFFLOAD_VLAN_INSERT | \ +@@ -366,7 +367,7 @@ ice_txd_enable_offload(struct rte_mbuf *tx_pkt, + /* Tx Checksum Offload */ + /* SET MACLEN */ + td_offset |= (tx_pkt->l2_len >> 1) << +- ICE_TX_DESC_LEN_MACLEN_S; ++ ICE_TX_DESC_LEN_MACLEN_S; + + /* Enable L3 checksum offload */ + if (ol_flags & RTE_MBUF_F_TX_IP_CKSUM) { +diff --git a/dpdk/drivers/net/ice/ice_rxtx_vec_sse.c b/dpdk/drivers/net/ice/ice_rxtx_vec_sse.c +index 6cd44c5847..fd94cedde3 100644 +--- a/dpdk/drivers/net/ice/ice_rxtx_vec_sse.c ++++ b/dpdk/drivers/net/ice/ice_rxtx_vec_sse.c +@@ -202,7 +202,7 @@ ice_rx_desc_to_olflags_v(struct ice_rx_queue *rxq, __m128i descs[4], + __m128i l3_l4_mask = _mm_set_epi32(~0x6, ~0x6, ~0x6, ~0x6); + __m128i l3_l4_flags = _mm_and_si128(flags, l3_l4_mask); + flags = _mm_or_si128(l3_l4_flags, l4_outer_flags); +- /* we need to mask out the reduntant bits introduced by RSS or ++ /* we need to mask out the redundant bits introduced by RSS or + * VLAN fields. + */ + flags = _mm_and_si128(flags, cksum_mask); +@@ -566,7 +566,7 @@ _ice_recv_raw_pkts_vec(struct ice_rx_queue *rxq, struct rte_mbuf **rx_pkts, + _mm_storeu_si128((void *)&rx_pkts[pos]->rx_descriptor_fields1, + pkt_mb0); + ice_rx_desc_to_ptype_v(descs, &rx_pkts[pos], ptype_tbl); +- /* C.4 calc avaialbe number of desc */ ++ /* C.4 calc available number of desc */ + var = __builtin_popcountll(_mm_cvtsi128_si64(staterr)); + nb_pkts_recd += var; + if (likely(var != ICE_DESCS_PER_LOOP)) +diff --git a/dpdk/drivers/net/ice/ice_switch_filter.c b/dpdk/drivers/net/ice/ice_switch_filter.c +index ed29c00d77..bd805d9606 100644 +--- a/dpdk/drivers/net/ice/ice_switch_filter.c ++++ b/dpdk/drivers/net/ice/ice_switch_filter.c +@@ -400,6 +400,14 @@ ice_switch_create(struct ice_adapter *ad, + "lookup list should not be NULL"); + goto error; + } + -+# Local_Config schema documentation -+EXTRA_DIST += ovsdb/local-config.xml -+CLEANFILES += ovsdb/ovsdb.local-config.5 -+man_MANS += ovsdb/ovsdb.local-config.5 -+ovsdb/ovsdb.local-config.5: \ -+ ovsdb/ovsdb-doc ovsdb/ ovsdb/local-config.xml ovsdb/local-config.ovsschema -+ $(AM_V_GEN)$(OVSDB_DOC) \ -+ --version=$(VERSION) \ -+ $(srcdir)/ovsdb/local-config.ovsschema \ -+ $(srcdir)/ovsdb/local-config.xml > $@.tmp && \ -+ mv $@.tmp $@ ++ if (ice_dcf_adminq_need_retry(ad)) { ++ rte_flow_error_set(error, EAGAIN, ++ RTE_FLOW_ERROR_TYPE_ITEM, NULL, ++ "DCF is not on"); ++ goto error; ++ } ++ + ret = ice_add_adv_rule(hw, list, lkups_cnt, rule_info, &rule_added); + if (!ret) { + filter_conf_ptr = rte_zmalloc("ice_switch_filter", +@@ -423,7 +431,12 @@ ice_switch_create(struct ice_adapter *ad, + + flow->rule = filter_conf_ptr; + } else { +- rte_flow_error_set(error, EINVAL, ++ if (ice_dcf_adminq_need_retry(ad)) ++ ret = -EAGAIN; ++ else ++ ret = -EINVAL; ++ ++ rte_flow_error_set(error, -ret, + RTE_FLOW_ERROR_TYPE_HANDLE, NULL, + "switch filter create flow fail"); + goto error; +@@ -475,9 +488,21 @@ ice_switch_destroy(struct ice_adapter *ad, + return -rte_errno; + } + ++ if (ice_dcf_adminq_need_retry(ad)) { ++ rte_flow_error_set(error, EAGAIN, ++ RTE_FLOW_ERROR_TYPE_ITEM, NULL, ++ "DCF is not on"); ++ return -rte_errno; ++ } ++ + ret = ice_rem_adv_rule_by_id(hw, &filter_conf_ptr->sw_query_data); + if (ret) { +- rte_flow_error_set(error, EINVAL, ++ if (ice_dcf_adminq_need_retry(ad)) ++ ret = -EAGAIN; ++ else ++ ret = -EINVAL; ++ ++ rte_flow_error_set(error, -ret, + RTE_FLOW_ERROR_TYPE_HANDLE, NULL, + "fail to destroy switch filter rule"); + return -rte_errno; +@@ -2016,6 +2041,12 @@ ice_switch_redirect(struct ice_adapter *ad, + } + + rmv_rule: ++ if (ice_dcf_adminq_need_retry(ad)) { ++ PMD_DRV_LOG(WARNING, "DCF is not on"); ++ ret = -EAGAIN; ++ goto out; ++ } ++ + /* Remove the old rule */ + ret = ice_rem_adv_rule(hw, lkups_ref, lkups_cnt, &rinfo); + if (ret) { +@@ -2028,6 +2059,12 @@ ice_switch_redirect(struct ice_adapter *ad, + } + + add_rule: ++ if (ice_dcf_adminq_need_retry(ad)) { ++ PMD_DRV_LOG(WARNING, "DCF is not on"); ++ ret = -EAGAIN; ++ goto out; ++ } + - EXTRA_DIST += ovsdb/TODO.rst -diff --git a/ovsdb/condition.c b/ovsdb/condition.c -index 388dd54a16..9aa3788dbb 100644 ---- a/ovsdb/condition.c -+++ b/ovsdb/condition.c -@@ -220,13 +220,13 @@ ovsdb_condition_optimize(struct ovsdb_condition *cnd) - static void - ovsdb_condition_optimize_destroy(struct ovsdb_condition *cnd) - { -- struct shash_node *node, *next; -+ struct shash_node *node; + /* Update VSI context */ + hw->vsi_ctx[rd->vsi_handle]->vsi_num = rd->new_vsi_num; -- SHASH_FOR_EACH_SAFE (node, next, &cnd->o_columns) { -+ SHASH_FOR_EACH_SAFE (node, &cnd->o_columns) { - struct ovsdb_o_column *o_column = node->data; -- struct ovsdb_o_clause *c, *c_next; -+ struct ovsdb_o_clause *c; +@@ -2047,6 +2084,10 @@ ice_switch_redirect(struct ice_adapter *ad, + } -- HMAP_FOR_EACH_SAFE(c, c_next, hmap_node, &o_column->o_clauses) { -+ HMAP_FOR_EACH_SAFE (c, hmap_node, &o_column->o_clauses) { - hmap_remove(&o_column->o_clauses, &c->hmap_node); - free(c); - } -diff --git a/ovsdb/file.c b/ovsdb/file.c -index 9f44007d97..ca80c28235 100644 ---- a/ovsdb/file.c -+++ b/ovsdb/file.c -@@ -524,6 +524,7 @@ ovsdb_file_read__(const char *filename, bool rw, + out: ++ if (ret == -EINVAL) ++ if (ice_dcf_adminq_need_retry(ad)) ++ ret = -EAGAIN; ++ + ice_free(hw, lkups_dp); + return ret; + } +diff --git a/dpdk/drivers/net/igc/igc_ethdev.c b/dpdk/drivers/net/igc/igc_ethdev.c +index a1f1a9772b..8d8a0da424 100644 +--- a/dpdk/drivers/net/igc/igc_ethdev.c ++++ b/dpdk/drivers/net/igc/igc_ethdev.c +@@ -1234,8 +1234,15 @@ eth_igc_dev_init(struct rte_eth_dev *dev) + * has already done this work. Only check we don't need a different + * RX function. + */ +- if (rte_eal_process_type() != RTE_PROC_PRIMARY) ++ if (rte_eal_process_type() != RTE_PROC_PRIMARY) { ++ dev->rx_pkt_burst = igc_recv_pkts; ++ if (dev->data->scattered_rx) ++ dev->rx_pkt_burst = igc_recv_scattered_pkts; ++ ++ dev->tx_pkt_burst = igc_xmit_pkts; ++ dev->tx_pkt_prepare = eth_igc_prep_pkts; + return 0; ++ } + + rte_eth_copy_pci_info(dev, pci_dev); + dev->data->dev_flags |= RTE_ETH_DEV_AUTOFILL_QUEUE_XSTATS; +diff --git a/dpdk/drivers/net/igc/igc_filter.c b/dpdk/drivers/net/igc/igc_filter.c +index 51fcabfb59..bff98df200 100644 +--- a/dpdk/drivers/net/igc/igc_filter.c ++++ b/dpdk/drivers/net/igc/igc_filter.c +@@ -167,7 +167,7 @@ igc_tuple_filter_lookup(const struct igc_adapter *igc, + /* search the filter array */ + for (; i < IGC_MAX_NTUPLE_FILTERS; i++) { + if (igc->ntuple_filters[i].hash_val) { +- /* compare the hase value */ ++ /* compare the hash value */ + if (ntuple->hash_val == + igc->ntuple_filters[i].hash_val) + /* filter be found, return index */ +diff --git a/dpdk/drivers/net/igc/igc_txrx.c b/dpdk/drivers/net/igc/igc_txrx.c +index 339b0c9aa1..ffd219b0df 100644 +--- a/dpdk/drivers/net/igc/igc_txrx.c ++++ b/dpdk/drivers/net/igc/igc_txrx.c +@@ -345,7 +345,7 @@ rx_desc_get_pkt_info(struct igc_rx_queue *rxq, struct rte_mbuf *rxm, + rxm->packet_type = rx_desc_pkt_info_to_pkt_type(pkt_info); + } - error = ovsdb_txn_replay_commit(txn); - if (error) { -+ ovsdb_error_destroy(error); - ovsdb_storage_unread(storage); - break; - } -diff --git a/ovsdb/jsonrpc-server.c b/ovsdb/jsonrpc-server.c -index 351c39d8aa..916a1f414e 100644 ---- a/ovsdb/jsonrpc-server.c -+++ b/ovsdb/jsonrpc-server.c -@@ -197,9 +197,9 @@ ovsdb_jsonrpc_server_remove_db(struct ovsdb_jsonrpc_server *svr, - void - ovsdb_jsonrpc_server_destroy(struct ovsdb_jsonrpc_server *svr) +-static uint16_t ++uint16_t + igc_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t nb_pkts) { -- struct shash_node *node, *next; -+ struct shash_node *node; + struct igc_rx_queue * const rxq = rx_queue; +@@ -488,7 +488,7 @@ igc_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t nb_pkts) + return nb_rx; + } -- SHASH_FOR_EACH_SAFE (node, next, &svr->remotes) { -+ SHASH_FOR_EACH_SAFE (node, &svr->remotes) { - ovsdb_jsonrpc_server_del_remote(node); - } - shash_destroy(&svr->remotes); -@@ -227,9 +227,9 @@ void - ovsdb_jsonrpc_server_set_remotes(struct ovsdb_jsonrpc_server *svr, - const struct shash *new_remotes) +-static uint16_t ++uint16_t + igc_recv_scattered_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, + uint16_t nb_pkts) { -- struct shash_node *node, *next; -+ struct shash_node *node; +@@ -1397,7 +1397,7 @@ eth_igc_rx_queue_setup(struct rte_eth_dev *dev, + } -- SHASH_FOR_EACH_SAFE (node, next, &svr->remotes) { -+ SHASH_FOR_EACH_SAFE (node, &svr->remotes) { - struct ovsdb_jsonrpc_remote *remote = node->data; - struct ovsdb_jsonrpc_options *options - = shash_find_data(new_remotes, node->name); -@@ -585,9 +585,9 @@ ovsdb_jsonrpc_session_set_options(struct ovsdb_jsonrpc_session *session, - static void - ovsdb_jsonrpc_session_run_all(struct ovsdb_jsonrpc_remote *remote) + /* prepare packets for transmit */ +-static uint16_t ++uint16_t + eth_igc_prep_pkts(__rte_unused void *tx_queue, struct rte_mbuf **tx_pkts, + uint16_t nb_pkts) { -- struct ovsdb_jsonrpc_session *s, *next; -+ struct ovsdb_jsonrpc_session *s; +@@ -1604,7 +1604,7 @@ tx_desc_cksum_flags_to_olinfo(uint64_t ol_flags) + return tmp; + } -- LIST_FOR_EACH_SAFE (s, next, node, &remote->sessions) { -+ LIST_FOR_EACH_SAFE (s, node, &remote->sessions) { - int error = ovsdb_jsonrpc_session_run(s); - if (error) { - ovsdb_jsonrpc_session_close(s); -@@ -642,9 +642,9 @@ ovsdb_jsonrpc_session_get_memory_usage_all( - static void - ovsdb_jsonrpc_session_close_all(struct ovsdb_jsonrpc_remote *remote) +-static uint16_t ++uint16_t + igc_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts, uint16_t nb_pkts) { -- struct ovsdb_jsonrpc_session *s, *next; -+ struct ovsdb_jsonrpc_session *s; + struct igc_tx_queue * const txq = tx_queue; +@@ -2099,7 +2099,7 @@ eth_igc_tx_done_cleanup(void *txqueue, uint32_t free_cnt) + sw_ring[tx_id].mbuf = NULL; + sw_ring[tx_id].last_id = tx_id; + +- /* Move to next segemnt. */ ++ /* Move to next segment. */ + tx_id = sw_ring[tx_id].next_id; + } while (tx_id != tx_next); + +@@ -2133,7 +2133,7 @@ eth_igc_tx_done_cleanup(void *txqueue, uint32_t free_cnt) + * Walk the list and find the next mbuf, if any. + */ + do { +- /* Move to next segemnt. */ ++ /* Move to next segment. */ + tx_id = sw_ring[tx_id].next_id; + + if (sw_ring[tx_id].mbuf) +diff --git a/dpdk/drivers/net/igc/igc_txrx.h b/dpdk/drivers/net/igc/igc_txrx.h +index 535108a868..02a0a051bb 100644 +--- a/dpdk/drivers/net/igc/igc_txrx.h ++++ b/dpdk/drivers/net/igc/igc_txrx.h +@@ -49,6 +49,12 @@ void eth_igc_txq_info_get(struct rte_eth_dev *dev, uint16_t queue_id, + struct rte_eth_txq_info *qinfo); + void eth_igc_vlan_strip_queue_set(struct rte_eth_dev *dev, + uint16_t rx_queue_id, int on); ++uint16_t igc_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t nb_pkts); ++uint16_t igc_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts, uint16_t nb_pkts); ++uint16_t eth_igc_prep_pkts(__rte_unused void *tx_queue, struct rte_mbuf **tx_pkts, ++ uint16_t nb_pkts); ++uint16_t igc_recv_scattered_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, ++ uint16_t nb_pkts); + #ifdef __cplusplus + } + #endif +diff --git a/dpdk/drivers/net/ionic/ionic_if.h b/dpdk/drivers/net/ionic/ionic_if.h +index 693b44d764..45bad9b040 100644 +--- a/dpdk/drivers/net/ionic/ionic_if.h ++++ b/dpdk/drivers/net/ionic/ionic_if.h +@@ -2068,7 +2068,7 @@ typedef struct ionic_admin_comp ionic_fw_download_comp; + * enum ionic_fw_control_oper - FW control operations + * @IONIC_FW_RESET: Reset firmware + * @IONIC_FW_INSTALL: Install firmware +- * @IONIC_FW_ACTIVATE: Acticate firmware ++ * @IONIC_FW_ACTIVATE: Activate firmware + */ + enum ionic_fw_control_oper { + IONIC_FW_RESET = 0, +@@ -2091,7 +2091,7 @@ struct ionic_fw_control_cmd { + }; -- LIST_FOR_EACH_SAFE (s, next, node, &remote->sessions) { -+ LIST_FOR_EACH_SAFE (s, node, &remote->sessions) { - ovsdb_jsonrpc_session_close(s); - } + /** +- * struct ionic_fw_control_comp - Firmware control copletion ++ * struct ionic_fw_control_comp - Firmware control completion + * @status: Status of the command (enum ionic_status_code) + * @comp_index: Index in the descriptor ring for which this is the completion + * @slot: Slot where the firmware was installed +@@ -2878,7 +2878,7 @@ struct ionic_doorbell { + * and @identity->intr_coal_div to convert from + * usecs to device units: + * +- * coal_init = coal_usecs * coal_mutl / coal_div ++ * coal_init = coal_usecs * coal_mult / coal_div + * + * When an interrupt is sent the interrupt + * coalescing timer current value +diff --git a/dpdk/drivers/net/ipn3ke/ipn3ke_ethdev.c b/dpdk/drivers/net/ipn3ke/ipn3ke_ethdev.c +index 964506c6db..014e438dd5 100644 +--- a/dpdk/drivers/net/ipn3ke/ipn3ke_ethdev.c ++++ b/dpdk/drivers/net/ipn3ke/ipn3ke_ethdev.c +@@ -483,7 +483,7 @@ static int ipn3ke_vswitch_probe(struct rte_afu_device *afu_dev) + RTE_CACHE_LINE_SIZE, + afu_dev->device.numa_node); + if (!hw) { +- IPN3KE_AFU_PMD_ERR("failed to allocate hardwart data"); ++ IPN3KE_AFU_PMD_ERR("failed to allocate hardware data"); + retval = -ENOMEM; + return -ENOMEM; + } +diff --git a/dpdk/drivers/net/ipn3ke/ipn3ke_ethdev.h b/dpdk/drivers/net/ipn3ke/ipn3ke_ethdev.h +index 041f13d9c3..58fcc50c57 100644 +--- a/dpdk/drivers/net/ipn3ke/ipn3ke_ethdev.h ++++ b/dpdk/drivers/net/ipn3ke/ipn3ke_ethdev.h +@@ -223,7 +223,7 @@ struct ipn3ke_hw_cap { + }; + + /** +- * Strucute to store private data for each representor instance ++ * Structure to store private data for each representor instance + */ + struct ipn3ke_rpst { + TAILQ_ENTRY(ipn3ke_rpst) next; /**< Next in device list. */ +@@ -237,7 +237,7 @@ struct ipn3ke_rpst { + uint16_t i40e_pf_eth_port_id; + struct rte_eth_link ori_linfo; + struct ipn3ke_tm_internals tm; +- /**< Private data store of assocaiated physical function */ ++ /**< Private data store of associated physical function */ + struct rte_ether_addr mac_addr; + }; + +diff --git a/dpdk/drivers/net/ipn3ke/ipn3ke_flow.c b/dpdk/drivers/net/ipn3ke/ipn3ke_flow.c +index f5867ca055..66ae31a5a9 100644 +--- a/dpdk/drivers/net/ipn3ke/ipn3ke_flow.c ++++ b/dpdk/drivers/net/ipn3ke/ipn3ke_flow.c +@@ -1299,7 +1299,7 @@ int ipn3ke_flow_init(void *dev) + IPN3KE_AFU_PMD_DEBUG("IPN3KE_CLF_LKUP_ENABLE: %x\n", data); + + +- /* configure rx parse config, settings associatied with VxLAN */ ++ /* configure rx parse config, settings associated with VxLAN */ + IPN3KE_MASK_WRITE_REG(hw, + IPN3KE_CLF_RX_PARSE_CFG, + 0, +diff --git a/dpdk/drivers/net/ipn3ke/ipn3ke_representor.c b/dpdk/drivers/net/ipn3ke/ipn3ke_representor.c +index de325c7d29..abbecfdf2e 100644 +--- a/dpdk/drivers/net/ipn3ke/ipn3ke_representor.c ++++ b/dpdk/drivers/net/ipn3ke/ipn3ke_representor.c +@@ -2218,9 +2218,6 @@ ipn3ke_rpst_xstats_get + struct ipn3ke_rpst_hw_port_stats hw_stats; + struct rte_eth_stats stats; + +- if (!xstats) +- return 0; +- + if (!ethdev) { + IPN3KE_AFU_PMD_ERR("ethernet device to get statistics is NULL"); + return -EINVAL; +@@ -2282,7 +2279,7 @@ ipn3ke_rpst_xstats_get + count++; + } + +- /* Get individiual stats from ipn3ke_rpst_hw_port */ ++ /* Get individual stats from ipn3ke_rpst_hw_port */ + for (i = 0; i < IPN3KE_RPST_HW_PORT_XSTATS_CNT; i++) { + xstats[count].value = *(uint64_t *)(((char *)(&hw_stats)) + + ipn3ke_rpst_hw_port_strings[i].offset); +@@ -2290,7 +2287,7 @@ ipn3ke_rpst_xstats_get + count++; + } + +- /* Get individiual stats from ipn3ke_rpst_rxq_pri */ ++ /* Get individual stats from ipn3ke_rpst_rxq_pri */ + for (i = 0; i < IPN3KE_RPST_RXQ_PRIO_XSTATS_CNT; i++) { + for (prio = 0; prio < IPN3KE_RPST_PRIO_XSTATS_CNT; prio++) { + xstats[count].value = +@@ -2302,7 +2299,7 @@ ipn3ke_rpst_xstats_get + } + } + +- /* Get individiual stats from ipn3ke_rpst_txq_prio */ ++ /* Get individual stats from ipn3ke_rpst_txq_prio */ + for (i = 0; i < IPN3KE_RPST_TXQ_PRIO_XSTATS_CNT; i++) { + for (prio = 0; prio < IPN3KE_RPST_PRIO_XSTATS_CNT; prio++) { + xstats[count].value = +@@ -2340,7 +2337,7 @@ __rte_unused unsigned int limit) + count++; + } + +- /* Get individiual stats from ipn3ke_rpst_hw_port */ ++ /* Get individual stats from ipn3ke_rpst_hw_port */ + for (i = 0; i < IPN3KE_RPST_HW_PORT_XSTATS_CNT; i++) { + snprintf(xstats_names[count].name, + sizeof(xstats_names[count].name), +@@ -2349,7 +2346,7 @@ __rte_unused unsigned int limit) + count++; + } + +- /* Get individiual stats from ipn3ke_rpst_rxq_pri */ ++ /* Get individual stats from ipn3ke_rpst_rxq_pri */ + for (i = 0; i < IPN3KE_RPST_RXQ_PRIO_XSTATS_CNT; i++) { + for (prio = 0; prio < 8; prio++) { + snprintf(xstats_names[count].name, +@@ -2361,7 +2358,7 @@ __rte_unused unsigned int limit) + } + } + +- /* Get individiual stats from ipn3ke_rpst_txq_prio */ ++ /* Get individual stats from ipn3ke_rpst_txq_prio */ + for (i = 0; i < IPN3KE_RPST_TXQ_PRIO_XSTATS_CNT; i++) { + for (prio = 0; prio < 8; prio++) { + snprintf(xstats_names[count].name, +diff --git a/dpdk/drivers/net/ipn3ke/meson.build b/dpdk/drivers/net/ipn3ke/meson.build +index 4bf739809e..104d2f58e5 100644 +--- a/dpdk/drivers/net/ipn3ke/meson.build ++++ b/dpdk/drivers/net/ipn3ke/meson.build +@@ -8,7 +8,7 @@ if is_windows + endif + + # +-# Add the experimenatal APIs called from this PMD ++# Add the experimental APIs called from this PMD + # rte_eth_switch_domain_alloc() + # rte_eth_dev_create() + # rte_eth_dev_destroy() +diff --git a/dpdk/drivers/net/ixgbe/ixgbe_bypass.c b/dpdk/drivers/net/ixgbe/ixgbe_bypass.c +index 67ced6c723..94f34a2996 100644 +--- a/dpdk/drivers/net/ixgbe/ixgbe_bypass.c ++++ b/dpdk/drivers/net/ixgbe/ixgbe_bypass.c +@@ -11,7 +11,7 @@ + + #define BYPASS_STATUS_OFF_MASK 3 + +-/* Macros to check for invlaid function pointers. */ ++/* Macros to check for invalid function pointers. */ + #define FUNC_PTR_OR_ERR_RET(func, retval) do { \ + if ((func) == NULL) { \ + PMD_DRV_LOG(ERR, "%s:%d function not supported", \ +diff --git a/dpdk/drivers/net/ixgbe/ixgbe_bypass_api.h b/dpdk/drivers/net/ixgbe/ixgbe_bypass_api.h +index 8eb773391b..6ef965dbb6 100644 +--- a/dpdk/drivers/net/ixgbe/ixgbe_bypass_api.h ++++ b/dpdk/drivers/net/ixgbe/ixgbe_bypass_api.h +@@ -135,7 +135,7 @@ static s32 ixgbe_bypass_rw_generic(struct ixgbe_hw *hw, u32 cmd, u32 *status) + * ixgbe_bypass_valid_rd_generic - Verify valid return from bit-bang. + * + * If we send a write we can't be sure it took until we can read back +- * that same register. It can be a problem as some of the feilds may ++ * that same register. It can be a problem as some of the fields may + * for valid reasons change between the time wrote the register and + * we read it again to verify. So this function check everything we + * can check and then assumes it worked. +@@ -189,7 +189,7 @@ static bool ixgbe_bypass_valid_rd_generic(u32 in_reg, u32 out_reg) } -@@ -660,9 +660,9 @@ static void - ovsdb_jsonrpc_session_reconnect_all(struct ovsdb_jsonrpc_remote *remote, - bool force, const char *comment) - { -- struct ovsdb_jsonrpc_session *s, *next; -+ struct ovsdb_jsonrpc_session *s; -- LIST_FOR_EACH_SAFE (s, next, node, &remote->sessions) { -+ LIST_FOR_EACH_SAFE (s, node, &remote->sessions) { - if (force || !s->db_change_aware) { - jsonrpc_session_force_reconnect(s->js); - if (comment && jsonrpc_session_is_connected(s->js)) { -@@ -909,9 +909,9 @@ error: - static void - ovsdb_jsonrpc_session_unlock_all(struct ovsdb_jsonrpc_session *s) - { -- struct ovsdb_lock_waiter *waiter, *next; -+ struct ovsdb_lock_waiter *waiter; + /** +- * ixgbe_bypass_set_generic - Set a bypass field in the FW CTRL Regiter. ++ * ixgbe_bypass_set_generic - Set a bypass field in the FW CTRL Register. + * + * @hw: pointer to hardware structure + * @cmd: The control word we are setting. +diff --git a/dpdk/drivers/net/ixgbe/ixgbe_ethdev.c b/dpdk/drivers/net/ixgbe/ixgbe_ethdev.c +index fe61dba81d..31d06b6110 100644 +--- a/dpdk/drivers/net/ixgbe/ixgbe_ethdev.c ++++ b/dpdk/drivers/net/ixgbe/ixgbe_ethdev.c +@@ -128,6 +128,13 @@ + #define IXGBE_EXVET_VET_EXT_SHIFT 16 + #define IXGBE_DMATXCTL_VT_MASK 0xFFFF0000 + ++#define IXGBE_DEVARG_FIBER_SDP3_NOT_TX_DISABLE "fiber_sdp3_no_tx_disable" ++ ++static const char * const ixgbe_valid_arguments[] = { ++ IXGBE_DEVARG_FIBER_SDP3_NOT_TX_DISABLE, ++ NULL ++}; ++ + #define IXGBEVF_DEVARG_PFLINK_FULLCHK "pflink_fullchk" + + static const char * const ixgbevf_valid_arguments[] = { +@@ -348,6 +355,8 @@ static int ixgbe_dev_udp_tunnel_port_del(struct rte_eth_dev *dev, + static int ixgbe_filter_restore(struct rte_eth_dev *dev); + static void ixgbe_l2_tunnel_conf(struct rte_eth_dev *dev); + static int ixgbe_wait_for_link_up(struct ixgbe_hw *hw); ++static int devarg_handle_int(__rte_unused const char *key, const char *value, ++ void *extra_args); + + /* + * Define VF Stats MACRO for Non "cleared on read" register +@@ -781,6 +790,20 @@ ixgbe_is_sfp(struct ixgbe_hw *hw) + case ixgbe_phy_sfp_passive_unknown: + return 1; + default: ++ /* x550em devices may be SFP, check media type */ ++ switch (hw->mac.type) { ++ case ixgbe_mac_X550EM_x: ++ case ixgbe_mac_X550EM_a: ++ switch (ixgbe_get_media_type(hw)) { ++ case ixgbe_media_type_fiber: ++ case ixgbe_media_type_fiber_qsfp: ++ return 1; ++ default: ++ break; ++ } ++ default: ++ break; ++ } + return 0; + } + } +@@ -1018,6 +1041,29 @@ ixgbe_swfw_lock_reset(struct ixgbe_hw *hw) + ixgbe_release_swfw_semaphore(hw, mask); + } -- HMAP_FOR_EACH_SAFE (waiter, next, session_node, &s->up.waiters) { -+ HMAP_FOR_EACH_SAFE (waiter, session_node, &s->up.waiters) { - ovsdb_jsonrpc_session_unlock__(waiter); - } ++static void ++ixgbe_parse_devargs(struct ixgbe_adapter *adapter, ++ struct rte_devargs *devargs) ++{ ++ struct rte_kvargs *kvlist; ++ uint16_t sdp3_no_tx_disable; ++ ++ if (devargs == NULL) ++ return; ++ ++ kvlist = rte_kvargs_parse(devargs->args, ixgbe_valid_arguments); ++ if (kvlist == NULL) ++ return; ++ ++ if (rte_kvargs_count(kvlist, IXGBE_DEVARG_FIBER_SDP3_NOT_TX_DISABLE) == 1 && ++ rte_kvargs_process(kvlist, IXGBE_DEVARG_FIBER_SDP3_NOT_TX_DISABLE, ++ devarg_handle_int, &sdp3_no_tx_disable) == 0 && ++ sdp3_no_tx_disable == 1) ++ adapter->sdp3_no_tx_disable = 1; ++ ++ rte_kvargs_free(kvlist); ++} ++ + /* + * This function is based on code in ixgbe_attach() in base/ixgbe.c. + * It returns 0 on success. +@@ -1081,6 +1127,8 @@ eth_ixgbe_dev_init(struct rte_eth_dev *eth_dev, void *init_params __rte_unused) + } + + rte_atomic32_clear(&ad->link_thread_running); ++ ixgbe_parse_devargs(eth_dev->data->dev_private, ++ pci_dev->device.devargs); + rte_eth_copy_pci_info(eth_dev, pci_dev); + eth_dev->data->dev_flags |= RTE_ETH_DEV_AUTOFILL_QUEUE_XSTATS; + +@@ -1223,13 +1271,8 @@ eth_ixgbe_dev_init(struct rte_eth_dev *eth_dev, void *init_params __rte_unused) + + /* initialize PF if max_vfs not zero */ + ret = ixgbe_pf_host_init(eth_dev); +- if (ret) { +- rte_free(eth_dev->data->mac_addrs); +- eth_dev->data->mac_addrs = NULL; +- rte_free(eth_dev->data->hash_mac_addrs); +- eth_dev->data->hash_mac_addrs = NULL; +- return ret; +- } ++ if (ret) ++ goto err_pf_host_init; + + ctrl_ext = IXGBE_READ_REG(hw, IXGBE_CTRL_EXT); + /* let hardware know driver is loaded */ +@@ -1268,10 +1311,14 @@ eth_ixgbe_dev_init(struct rte_eth_dev *eth_dev, void *init_params __rte_unused) + TAILQ_INIT(&filter_info->fivetuple_list); + + /* initialize flow director filter list & hash */ +- ixgbe_fdir_filter_init(eth_dev); ++ ret = ixgbe_fdir_filter_init(eth_dev); ++ if (ret) ++ goto err_fdir_filter_init; + + /* initialize l2 tunnel filter list & hash */ +- ixgbe_l2_tn_filter_init(eth_dev); ++ ret = ixgbe_l2_tn_filter_init(eth_dev); ++ if (ret) ++ goto err_l2_tn_filter_init; + + /* initialize flow filter lists */ + ixgbe_filterlist_init(); +@@ -1283,6 +1330,21 @@ eth_ixgbe_dev_init(struct rte_eth_dev *eth_dev, void *init_params __rte_unused) + ixgbe_tm_conf_init(eth_dev); + + return 0; ++ ++err_l2_tn_filter_init: ++ ixgbe_fdir_filter_uninit(eth_dev); ++err_fdir_filter_init: ++ ixgbe_disable_intr(hw); ++ rte_intr_disable(intr_handle); ++ rte_intr_callback_unregister(intr_handle, ++ ixgbe_dev_interrupt_handler, eth_dev); ++ ixgbe_pf_host_uninit(eth_dev); ++err_pf_host_init: ++ rte_free(eth_dev->data->mac_addrs); ++ eth_dev->data->mac_addrs = NULL; ++ rte_free(eth_dev->data->hash_mac_addrs); ++ eth_dev->data->hash_mac_addrs = NULL; ++ return ret; } -@@ -1198,8 +1198,8 @@ static void - ovsdb_jsonrpc_trigger_remove__(struct ovsdb_jsonrpc_session *s, - struct ovsdb *db) - { -- struct ovsdb_jsonrpc_trigger *t, *next; -- HMAP_FOR_EACH_SAFE (t, next, hmap_node, &s->triggers) { -+ struct ovsdb_jsonrpc_trigger *t; -+ HMAP_FOR_EACH_SAFE (t, hmap_node, &s->triggers) { - if (!db || t->trigger.db == db) { - ovsdb_jsonrpc_trigger_complete(t); - } -@@ -1226,8 +1226,8 @@ ovsdb_jsonrpc_trigger_complete_all(struct ovsdb_jsonrpc_session *s) - static void - ovsdb_jsonrpc_trigger_complete_done(struct ovsdb_jsonrpc_session *s) - { -- struct ovsdb_jsonrpc_trigger *trigger, *next; -- LIST_FOR_EACH_SAFE (trigger, next, trigger.node, &s->up.completions) { -+ struct ovsdb_jsonrpc_trigger *trigger; -+ LIST_FOR_EACH_SAFE (trigger, trigger.node, &s->up.completions) { - ovsdb_jsonrpc_trigger_complete(trigger); - } + + static int +@@ -2375,7 +2437,7 @@ ixgbe_dev_configure(struct rte_eth_dev *dev) + if (dev->data->dev_conf.rxmode.mq_mode & RTE_ETH_MQ_RX_RSS_FLAG) + dev->data->dev_conf.rxmode.offloads |= RTE_ETH_RX_OFFLOAD_RSS_HASH; + +- /* multipe queue mode checking */ ++ /* multiple queue mode checking */ + ret = ixgbe_check_mq_mode(dev); + if (ret != 0) { + PMD_DRV_LOG(ERR, "ixgbe_check_mq_mode fails with %d.", +@@ -2603,7 +2665,7 @@ ixgbe_dev_start(struct rte_eth_dev *dev) + } + } + +- /* confiugre msix for sleep until rx interrupt */ ++ /* configure MSI-X for sleep until Rx interrupt */ + ixgbe_configure_msix(dev); + + /* initialize transmission unit */ +@@ -2907,7 +2969,7 @@ ixgbe_dev_set_link_up(struct rte_eth_dev *dev) + if (hw->mac.type == ixgbe_mac_82599EB) { + #ifdef RTE_LIBRTE_IXGBE_BYPASS + if (hw->device_id == IXGBE_DEV_ID_82599_BYPASS) { +- /* Not suported in bypass mode */ ++ /* Not supported in bypass mode */ + PMD_INIT_LOG(ERR, "Set link up is not supported " + "by device id 0x%x", hw->device_id); + return -ENOTSUP; +@@ -2938,7 +3000,7 @@ ixgbe_dev_set_link_down(struct rte_eth_dev *dev) + if (hw->mac.type == ixgbe_mac_82599EB) { + #ifdef RTE_LIBRTE_IXGBE_BYPASS + if (hw->device_id == IXGBE_DEV_ID_82599_BYPASS) { +- /* Not suported in bypass mode */ ++ /* Not supported in bypass mode */ + PMD_INIT_LOG(ERR, "Set link down is not supported " + "by device id 0x%x", hw->device_id); + return -ENOTSUP; +@@ -3028,6 +3090,7 @@ ixgbe_dev_close(struct rte_eth_dev *dev) + + #ifdef RTE_LIB_SECURITY + rte_free(dev->security_ctx); ++ dev->security_ctx = NULL; + #endif + + return ret; +@@ -4236,7 +4299,8 @@ ixgbe_dev_link_update_share(struct rte_eth_dev *dev, + return rte_eth_linkstatus_set(dev, &link); + } + +- if (ixgbe_get_media_type(hw) == ixgbe_media_type_fiber) { ++ if (ixgbe_get_media_type(hw) == ixgbe_media_type_fiber && ++ !ad->sdp3_no_tx_disable) { + esdp_reg = IXGBE_READ_REG(hw, IXGBE_ESDP); + if ((esdp_reg & IXGBE_ESDP_SDP3)) + link_up = 0; +@@ -4603,7 +4667,7 @@ ixgbe_dev_interrupt_action(struct rte_eth_dev *dev) + * @param handle + * Pointer to interrupt handle. + * @param param +- * The address of parameter (struct rte_eth_dev *) regsitered before. ++ * The address of parameter (struct rte_eth_dev *) registered before. + * + * @return + * void +@@ -4659,7 +4723,7 @@ ixgbe_dev_interrupt_delayed_handler(void *param) + * @param handle + * Pointer to interrupt handle. + * @param param +- * The address of parameter (struct rte_eth_dev *) regsitered before. ++ * The address of parameter (struct rte_eth_dev *) registered before. + * + * @return + * void +@@ -5921,7 +5985,7 @@ ixgbevf_configure_msix(struct rte_eth_dev *dev) + /* Configure all RX queues of VF */ + for (q_idx = 0; q_idx < dev->data->nb_rx_queues; q_idx++) { + /* Force all queue use vector 0, +- * as IXGBE_VF_MAXMSIVECOTR = 1 ++ * as IXGBE_VF_MAXMSIVECTOR = 1 + */ + ixgbevf_set_ivar_map(hw, 0, q_idx, vector_idx); + rte_intr_vec_list_index_set(intr_handle, q_idx, +@@ -6256,7 +6320,7 @@ ixgbe_inject_5tuple_filter(struct rte_eth_dev *dev, + * @param + * dev: Pointer to struct rte_eth_dev. + * index: the index the filter allocates. +- * filter: ponter to the filter that will be added. ++ * filter: pointer to the filter that will be added. + * rx_queue: the queue id the filter assigned to. + * + * @return +@@ -6872,7 +6936,7 @@ ixgbe_timesync_disable(struct rte_eth_dev *dev) + /* Disable L2 filtering of IEEE1588/802.1AS Ethernet frame types. */ + IXGBE_WRITE_REG(hw, IXGBE_ETQF(IXGBE_ETQF_FILTER_1588), 0); + +- /* Stop incrementating the System Time registers. */ ++ /* Stop incrementing the System Time registers. */ + IXGBE_WRITE_REG(hw, IXGBE_TIMINCA, 0); + + return 0; +@@ -8225,6 +8289,8 @@ ixgbe_dev_macsec_register_disable(struct rte_eth_dev *dev) + RTE_PMD_REGISTER_PCI(net_ixgbe, rte_ixgbe_pmd); + RTE_PMD_REGISTER_PCI_TABLE(net_ixgbe, pci_id_ixgbe_map); + RTE_PMD_REGISTER_KMOD_DEP(net_ixgbe, "* igb_uio | uio_pci_generic | vfio-pci"); ++RTE_PMD_REGISTER_PARAM_STRING(net_ixgbe, ++ IXGBE_DEVARG_FIBER_SDP3_NOT_TX_DISABLE "=<0|1>"); + RTE_PMD_REGISTER_PCI(net_ixgbe_vf, rte_ixgbevf_pmd); + RTE_PMD_REGISTER_PCI_TABLE(net_ixgbe_vf, pci_id_ixgbevf_map); + RTE_PMD_REGISTER_KMOD_DEP(net_ixgbe_vf, "* igb_uio | vfio-pci"); +diff --git a/dpdk/drivers/net/ixgbe/ixgbe_ethdev.h b/dpdk/drivers/net/ixgbe/ixgbe_ethdev.h +index 83e8b5e56a..cc6049a66a 100644 +--- a/dpdk/drivers/net/ixgbe/ixgbe_ethdev.h ++++ b/dpdk/drivers/net/ixgbe/ixgbe_ethdev.h +@@ -68,7 +68,7 @@ + #define IXGBE_LPBK_NONE 0x0 /* Default value. Loopback is disabled. */ + #define IXGBE_LPBK_TX_RX 0x1 /* Tx->Rx loopback operation is enabled. */ + /* X540-X550 specific loopback operations */ +-#define IXGBE_MII_AUTONEG_ENABLE 0x1000 /* Auto-negociation enable (default = 1) */ ++#define IXGBE_MII_AUTONEG_ENABLE 0x1000 /* Auto-negotiation enable (default = 1) */ + + #define IXGBE_MAX_JUMBO_FRAME_SIZE 0x2600 /* Maximum Jumbo frame size. */ + +@@ -501,6 +501,9 @@ struct ixgbe_adapter { + /* For RSS reta table update */ + uint8_t rss_reta_updated; + ++ /* Used for limiting SDP3 TX_DISABLE checks */ ++ uint8_t sdp3_no_tx_disable; ++ + /* Used for VF link sync with PF's physical and logical (by checking + * mailbox status) link status. + */ +diff --git a/dpdk/drivers/net/ixgbe/ixgbe_fdir.c b/dpdk/drivers/net/ixgbe/ixgbe_fdir.c +index 7894047829..834c1b3f51 100644 +--- a/dpdk/drivers/net/ixgbe/ixgbe_fdir.c ++++ b/dpdk/drivers/net/ixgbe/ixgbe_fdir.c +@@ -390,7 +390,7 @@ fdir_set_input_mask_x550(struct rte_eth_dev *dev) + + switch (info->mask.tunnel_type_mask) { + case 0: +- /* Mask turnnel type */ ++ /* Mask tunnel type */ + fdiripv6m |= IXGBE_FDIRIP6M_TUNNEL_TYPE; + break; + case 1: +diff --git a/dpdk/drivers/net/ixgbe/ixgbe_flow.c b/dpdk/drivers/net/ixgbe/ixgbe_flow.c +index bdc9d4796c..368342872a 100644 +--- a/dpdk/drivers/net/ixgbe/ixgbe_flow.c ++++ b/dpdk/drivers/net/ixgbe/ixgbe_flow.c +@@ -135,7 +135,7 @@ const struct rte_flow_action *next_no_void_action( } -@@ -1688,8 +1688,8 @@ ovsdb_jsonrpc_monitor_preremove_db(struct ovsdb_jsonrpc_session *s, - { - ovs_assert(db); -- struct ovsdb_jsonrpc_monitor *m, *next; -- HMAP_FOR_EACH_SAFE (m, next, node, &s->monitors) { -+ struct ovsdb_jsonrpc_monitor *m; -+ HMAP_FOR_EACH_SAFE (m, node, &s->monitors) { - if (m->db == db) { - ovsdb_jsonrpc_monitor_destroy(m, true); - } -@@ -1700,9 +1700,9 @@ ovsdb_jsonrpc_monitor_preremove_db(struct ovsdb_jsonrpc_session *s, - static void - ovsdb_jsonrpc_monitor_remove_all(struct ovsdb_jsonrpc_session *s) + /** +- * Please aware there's an asumption for all the parsers. ++ * Please be aware there's an assumption for all the parsers. + * rte_flow_item is using big endian, rte_flow_attr and + * rte_flow_action are using CPU order. + * Because the pattern is used to describe the packets, +@@ -3261,7 +3261,7 @@ ixgbe_flow_create(struct rte_eth_dev *dev, + + /** + * Check if the flow rule is supported by ixgbe. +- * It only checkes the format. Don't guarantee the rule can be programmed into ++ * It only checks the format. Don't guarantee the rule can be programmed into + * the HW. Because there can be no enough room for the rule. + */ + static int +diff --git a/dpdk/drivers/net/ixgbe/ixgbe_ipsec.c b/dpdk/drivers/net/ixgbe/ixgbe_ipsec.c +index 944c9f2380..c353ae33b4 100644 +--- a/dpdk/drivers/net/ixgbe/ixgbe_ipsec.c ++++ b/dpdk/drivers/net/ixgbe/ixgbe_ipsec.c +@@ -310,7 +310,7 @@ ixgbe_crypto_remove_sa(struct rte_eth_dev *dev, + return -1; + } + +- /* Disable and clear Rx SPI and key table table entryes*/ ++ /* Disable and clear Rx SPI and key table table entries*/ + reg_val = IPSRXIDX_WRITE | IPSRXIDX_TABLE_SPI | (sa_index << 3); + IXGBE_WRITE_REG(hw, IXGBE_IPSRXSPI, 0); + IXGBE_WRITE_REG(hw, IXGBE_IPSRXIPIDX, 0); +diff --git a/dpdk/drivers/net/ixgbe/ixgbe_pf.c b/dpdk/drivers/net/ixgbe/ixgbe_pf.c +index 9f1bd0a62b..c73833b7ae 100644 +--- a/dpdk/drivers/net/ixgbe/ixgbe_pf.c ++++ b/dpdk/drivers/net/ixgbe/ixgbe_pf.c +@@ -242,7 +242,7 @@ int ixgbe_pf_host_configure(struct rte_eth_dev *eth_dev) + /* PFDMA Tx General Switch Control Enables VMDQ loopback */ + IXGBE_WRITE_REG(hw, IXGBE_PFDTXGSWC, IXGBE_PFDTXGSWC_VT_LBEN); + +- /* clear VMDq map to perment rar 0 */ ++ /* clear VMDq map to permanent rar 0 */ + hw->mac.ops.clear_vmdq(hw, 0, IXGBE_CLEAR_VMDQ_ALL); + + /* clear VMDq map to scan rar 127 */ +diff --git a/dpdk/drivers/net/ixgbe/ixgbe_rxtx.c b/dpdk/drivers/net/ixgbe/ixgbe_rxtx.c +index d7c80d4242..99e928a2a9 100644 +--- a/dpdk/drivers/net/ixgbe/ixgbe_rxtx.c ++++ b/dpdk/drivers/net/ixgbe/ixgbe_rxtx.c +@@ -1954,7 +1954,7 @@ ixgbe_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, + * register. + * Update the RDT with the value of the last processed RX descriptor + * minus 1, to guarantee that the RDT register is never equal to the +- * RDH register, which creates a "full" ring situtation from the ++ * RDH register, which creates a "full" ring situation from the + * hardware point of view... + */ + nb_hold = (uint16_t) (nb_hold + rxq->nb_rx_hold); +@@ -2303,7 +2303,7 @@ ixgbe_recv_pkts_lro(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t nb_pkts, + * register. + * Update the RDT with the value of the last processed RX descriptor + * minus 1, to guarantee that the RDT register is never equal to the +- * RDH register, which creates a "full" ring situtation from the ++ * RDH register, which creates a "full" ring situation from the + * hardware point of view... + */ + if (!bulk_alloc && nb_hold > rxq->rx_free_thresh) { +@@ -2666,7 +2666,7 @@ ixgbe_dev_tx_queue_setup(struct rte_eth_dev *dev, + */ + tx_free_thresh = (uint16_t)((tx_conf->tx_free_thresh) ? + tx_conf->tx_free_thresh : DEFAULT_TX_FREE_THRESH); +- /* force tx_rs_thresh to adapt an aggresive tx_free_thresh */ ++ /* force tx_rs_thresh to adapt an aggressive tx_free_thresh */ + tx_rs_thresh = (DEFAULT_TX_RS_THRESH + tx_free_thresh > nb_desc) ? + nb_desc - tx_free_thresh : DEFAULT_TX_RS_THRESH; + if (tx_conf->tx_rs_thresh > 0) +@@ -4831,7 +4831,7 @@ ixgbe_set_rx_function(struct rte_eth_dev *dev) + dev->data->port_id); + dev->rx_pkt_burst = ixgbe_recv_pkts_lro_bulk_alloc; + } else { +- PMD_INIT_LOG(DEBUG, "Using Regualr (non-vector, " ++ PMD_INIT_LOG(DEBUG, "Using Regular (non-vector, " + "single allocation) " + "Scattered Rx callback " + "(port=%d).", +@@ -5170,7 +5170,7 @@ ixgbe_dev_rx_init(struct rte_eth_dev *dev) + /* + * Setup the Checksum Register. + * Disable Full-Packet Checksum which is mutually exclusive with RSS. +- * Enable IP/L4 checkum computation by hardware if requested to do so. ++ * Enable IP/L4 checksum computation by hardware if requested to do so. + */ + rxcsum = IXGBE_READ_REG(hw, IXGBE_RXCSUM); + rxcsum |= IXGBE_RXCSUM_PCSD; +diff --git a/dpdk/drivers/net/ixgbe/ixgbe_rxtx_vec_sse.c b/dpdk/drivers/net/ixgbe/ixgbe_rxtx_vec_sse.c +index 1eed949495..bb34b27168 100644 +--- a/dpdk/drivers/net/ixgbe/ixgbe_rxtx_vec_sse.c ++++ b/dpdk/drivers/net/ixgbe/ixgbe_rxtx_vec_sse.c +@@ -364,6 +364,17 @@ _recv_raw_pkts_vec(struct ixgbe_rx_queue *rxq, struct rte_mbuf **rx_pkts, + uint8_t vlan_flags; + uint16_t udp_p_flag = 0; /* Rx Descriptor UDP header present */ + ++ /* ++ * Under the circumstance that `rx_tail` wrap back to zero ++ * and the advance speed of `rx_tail` is greater than `rxrearm_start`, ++ * `rx_tail` will catch up with `rxrearm_start` and surpass it. ++ * This may cause some mbufs be reused by application. ++ * ++ * So we need to make some restrictions to ensure that ++ * `rx_tail` will not exceed `rxrearm_start`. ++ */ ++ nb_pkts = RTE_MIN(nb_pkts, RTE_IXGBE_RXQ_REARM_THRESH); ++ + /* nb_pkts has to be floor-aligned to RTE_IXGBE_DESCS_PER_LOOP */ + nb_pkts = RTE_ALIGN_FLOOR(nb_pkts, RTE_IXGBE_DESCS_PER_LOOP); + +@@ -562,7 +573,7 @@ _recv_raw_pkts_vec(struct ixgbe_rx_queue *rxq, struct rte_mbuf **rx_pkts, + + desc_to_ptype_v(descs, rxq->pkt_type_mask, &rx_pkts[pos]); + +- /* C.4 calc avaialbe number of desc */ ++ /* C.4 calc available number of desc */ + var = __builtin_popcountll(_mm_cvtsi128_si64(staterr)); + nb_pkts_recd += var; + if (likely(var != RTE_IXGBE_DESCS_PER_LOOP)) +diff --git a/dpdk/drivers/net/kni/rte_eth_kni.c b/dpdk/drivers/net/kni/rte_eth_kni.c +index c428caf441..0532de5315 100644 +--- a/dpdk/drivers/net/kni/rte_eth_kni.c ++++ b/dpdk/drivers/net/kni/rte_eth_kni.c +@@ -124,7 +124,7 @@ eth_kni_start(struct rte_eth_dev *dev) + struct pmd_internals *internals = dev->data->dev_private; + uint16_t port_id = dev->data->port_id; + struct rte_mempool *mb_pool; +- struct rte_kni_conf conf; ++ struct rte_kni_conf conf = {{0}}; + const char *name = dev->device->name + 4; /* remove net_ */ + + mb_pool = internals->rx_queues[0].mb_pool; +diff --git a/dpdk/drivers/net/memif/memif_socket.c b/dpdk/drivers/net/memif/memif_socket.c +index 079cf01269..7886644412 100644 +--- a/dpdk/drivers/net/memif/memif_socket.c ++++ b/dpdk/drivers/net/memif/memif_socket.c +@@ -402,11 +402,10 @@ memif_msg_enq_init(struct rte_eth_dev *dev) { -- struct ovsdb_jsonrpc_monitor *m, *next; -+ struct ovsdb_jsonrpc_monitor *m; + struct pmd_internals *pmd = dev->data->dev_private; + struct memif_msg_queue_elt *e = memif_msg_enq(pmd->cc); +- memif_msg_init_t *i = &e->msg.init; ++ memif_msg_init_t *i; -- HMAP_FOR_EACH_SAFE (m, next, node, &s->monitors) { -+ HMAP_FOR_EACH_SAFE (m, node, &s->monitors) { - ovsdb_jsonrpc_monitor_destroy(m, false); - } + if (e == NULL) + return -1; +- + i = &e->msg.init; + e->msg.type = MEMIF_MSG_TYPE_INIT; + i->version = MEMIF_VERSION; +@@ -726,7 +725,7 @@ memif_msg_receive(struct memif_control_channel *cc) + break; + case MEMIF_MSG_TYPE_INIT: + /* +- * This cc does not have an interface asociated with it. ++ * This cc does not have an interface associated with it. + * If suitable interface is found it will be assigned here. + */ + ret = memif_msg_receive_init(cc, &msg); +diff --git a/dpdk/drivers/net/memif/rte_eth_memif.c b/dpdk/drivers/net/memif/rte_eth_memif.c +index e3d523af57..205d08b028 100644 +--- a/dpdk/drivers/net/memif/rte_eth_memif.c ++++ b/dpdk/drivers/net/memif/rte_eth_memif.c +@@ -351,13 +351,13 @@ eth_memif_rx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts) + goto no_free_bufs; + mbuf = mbuf_head; + mbuf->port = mq->in_port; ++ dst_off = 0; + + next_slot: + s0 = cur_slot & mask; + d0 = &ring->desc[s0]; + + src_len = d0->length; +- dst_off = 0; + src_off = 0; + + do { +@@ -1026,7 +1026,7 @@ memif_regions_init(struct rte_eth_dev *dev) + if (ret < 0) + return ret; + } else { +- /* create one memory region contaning rings and buffers */ ++ /* create one memory region containing rings and buffers */ + ret = memif_region_init_shm(dev, /* has buffers */ 1); + if (ret < 0) + return ret; +@@ -1500,23 +1500,6 @@ memif_stats_reset(struct rte_eth_dev *dev) + return 0; } -diff --git a/ovsdb/local-config.ovsschema b/ovsdb/local-config.ovsschema -new file mode 100644 -index 0000000000..bd86d0f4f6 ---- /dev/null -+++ b/ovsdb/local-config.ovsschema -@@ -0,0 +1,43 @@ -+{ -+ "name": "Local_Config", -+ "version": "1.0.0", -+ "cksum": "2048726482 1858", -+ "tables": { -+ "Config": { -+ "columns": { -+ "connections": { -+ "type": {"key": {"type": "uuid", -+ "refTable": "Connection"}, -+ "min": 0, -+ "max": "unlimited"}}}, -+ "maxRows": 1, -+ "isRoot": true}, -+ "Connection": { -+ "columns": { -+ "target": {"type": "string"}, -+ "max_backoff": {"type": {"key": {"type": "integer", -+ "minInteger": 1000}, -+ "min": 0, -+ "max": 1}}, -+ "inactivity_probe": {"type": {"key": "integer", -+ "min": 0, -+ "max": 1}}, -+ "read_only": {"type": "boolean"}, -+ "role": {"type": "string"}, -+ "other_config": {"type": {"key": "string", -+ "value": "string", -+ "min": 0, -+ "max": "unlimited"}}, -+ "external_ids": {"type": {"key": "string", -+ "value": "string", -+ "min": 0, -+ "max": "unlimited"}}, -+ "is_connected": {"type": "boolean", "ephemeral": true}, -+ "status": {"type": {"key": "string", -+ "value": "string", -+ "min": 0, -+ "max": "unlimited"}, -+ "ephemeral": true}}, -+ "indexes": [["target"]]} -+ } + +-static int +-memif_rx_queue_intr_enable(struct rte_eth_dev *dev __rte_unused, +- uint16_t qid __rte_unused) +-{ +- MIF_LOG(WARNING, "Interrupt mode not supported."); +- +- return -1; +-} +- +-static int +-memif_rx_queue_intr_disable(struct rte_eth_dev *dev, uint16_t qid __rte_unused) +-{ +- struct pmd_internals *pmd __rte_unused = dev->data->dev_private; +- +- return 0; +-} +- + static const struct eth_dev_ops ops = { + .dev_start = memif_dev_start, + .dev_stop = memif_dev_stop, +@@ -1527,8 +1510,6 @@ static const struct eth_dev_ops ops = { + .rx_queue_setup = memif_rx_queue_setup, + .rx_queue_release = memif_rx_queue_release, + .tx_queue_release = memif_tx_queue_release, +- .rx_queue_intr_enable = memif_rx_queue_intr_enable, +- .rx_queue_intr_disable = memif_rx_queue_intr_disable, + .link_update = memif_link_update, + .stats_get = memif_stats_get, + .stats_reset = memif_stats_reset, +diff --git a/dpdk/drivers/net/mlx4/meson.build b/dpdk/drivers/net/mlx4/meson.build +index 99a30eab8f..a038c1ec1b 100644 +--- a/dpdk/drivers/net/mlx4/meson.build ++++ b/dpdk/drivers/net/mlx4/meson.build +@@ -42,7 +42,7 @@ foreach libname:libnames + endforeach + if static_ibverbs or dlopen_ibverbs + # Build without adding shared libs to Requires.private +- ibv_cflags = run_command(pkgconf, '--cflags', 'libibverbs').stdout() ++ ibv_cflags = run_command(pkgconf, '--cflags', 'libibverbs', check:true).stdout() + ext_deps += declare_dependency(compile_args: ibv_cflags.split()) + endif + if static_ibverbs +diff --git a/dpdk/drivers/net/mlx4/mlx4.h b/dpdk/drivers/net/mlx4/mlx4.h +index 2d0c512f79..4023a47602 100644 +--- a/dpdk/drivers/net/mlx4/mlx4.h ++++ b/dpdk/drivers/net/mlx4/mlx4.h +@@ -74,7 +74,7 @@ enum mlx4_mp_req_type { + MLX4_MP_REQ_STOP_RXTX, + }; + +-/* Pameters for IPC. */ ++/* Parameters for IPC. */ + struct mlx4_mp_param { + enum mlx4_mp_req_type type; + int port_id; +diff --git a/dpdk/drivers/net/mlx4/mlx4_ethdev.c b/dpdk/drivers/net/mlx4/mlx4_ethdev.c +index d606ec8ca7..ce74c51ce2 100644 +--- a/dpdk/drivers/net/mlx4/mlx4_ethdev.c ++++ b/dpdk/drivers/net/mlx4/mlx4_ethdev.c +@@ -752,7 +752,7 @@ mlx4_stats_get(struct rte_eth_dev *dev, struct rte_eth_stats *stats) + * Pointer to Ethernet device structure. + * + * @return +- * alwasy 0 on success ++ * always 0 on success + */ + int + mlx4_stats_reset(struct rte_eth_dev *dev) +diff --git a/dpdk/drivers/net/mlx5/linux/mlx5_ethdev_os.c b/dpdk/drivers/net/mlx5/linux/mlx5_ethdev_os.c +index c19825ee52..fadcbd7ef7 100644 +--- a/dpdk/drivers/net/mlx5/linux/mlx5_ethdev_os.c ++++ b/dpdk/drivers/net/mlx5/linux/mlx5_ethdev_os.c +@@ -38,6 +38,7 @@ + #include + #include + #include ++#include + + #include "mlx5.h" + #include "mlx5_rxtx.h" +@@ -760,6 +761,56 @@ mlx5_dev_interrupt_device_fatal(struct mlx5_dev_ctx_shared *sh) + } + } + ++static void ++mlx5_dev_interrupt_nl_cb(struct nlmsghdr *hdr, void *cb_arg) ++{ ++ struct mlx5_dev_ctx_shared *sh = cb_arg; ++ uint32_t i; ++ uint32_t if_index; ++ ++ if (mlx5_nl_parse_link_status_update(hdr, &if_index) < 0) ++ return; ++ for (i = 0; i < sh->max_port; i++) { ++ struct mlx5_dev_shared_port *port = &sh->port[i]; ++ struct rte_eth_dev *dev; ++ struct mlx5_priv *priv; ++ ++ if (port->nl_ih_port_id >= RTE_MAX_ETHPORTS) ++ continue; ++ dev = &rte_eth_devices[port->nl_ih_port_id]; ++ /* Probing may initiate an LSC before configuration is done. */ ++ if (dev->data->dev_configured && ++ !dev->data->dev_conf.intr_conf.lsc) ++ break; ++ priv = dev->data->dev_private; ++ if (priv->if_index == if_index) { ++ /* Block logical LSC events. */ ++ uint16_t prev_status = dev->data->dev_link.link_status; ++ ++ if (mlx5_link_update(dev, 0) < 0) ++ DRV_LOG(ERR, "Failed to update link status: %s", ++ rte_strerror(rte_errno)); ++ else if (prev_status != dev->data->dev_link.link_status) ++ rte_eth_dev_callback_process ++ (dev, RTE_ETH_EVENT_INTR_LSC, NULL); ++ break; ++ } ++ } +} -diff --git a/ovsdb/local-config.xml b/ovsdb/local-config.xml -new file mode 100644 -index 0000000000..b502aea4cf ---- /dev/null -+++ b/ovsdb/local-config.xml -@@ -0,0 +1,296 @@ -+ -+ -+

-+ This database is for local configuration of an ovsdb-server. The -+ database is meant to be unique, even among multiple clustered db -+ servers, so that configuration that is local to that server can -+ be configured separately (e.g. Connection information). -+

+ -+ -+

-+ The root local configuration table for an ovsdb-server. This table -+ must have exactly one row. -+

-+ -+ -+ Database clients to which the Open vSwitch database server should -+ connect or on which it should listen, along with options for how these -+ connections should be configured. See the -+ table for more information. -+ -+ -+
++void ++mlx5_dev_interrupt_handler_nl(void *arg) ++{ ++ struct mlx5_dev_ctx_shared *sh = arg; ++ int nlsk_fd = rte_intr_fd_get(sh->intr_handle_nl); ++ ++ if (nlsk_fd < 0) ++ return; ++ if (mlx5_nl_read_events(nlsk_fd, mlx5_dev_interrupt_nl_cb, sh) < 0) ++ DRV_LOG(ERR, "Failed to process Netlink events: %s", ++ rte_strerror(rte_errno)); ++} + -+ -+

-+ Configuration for a database connection to an Open vSwitch database -+ (OVSDB) client. -+

+ /** + * Handle shared asynchronous events the NIC (removal event + * and link status change). Supports multiport IB device. +@@ -823,18 +874,6 @@ mlx5_dev_interrupt_handler(void *cb_arg) + tmp = sh->port[tmp - 1].ih_port_id; + dev = &rte_eth_devices[tmp]; + MLX5_ASSERT(dev); +- if ((event.event_type == IBV_EVENT_PORT_ACTIVE || +- event.event_type == IBV_EVENT_PORT_ERR) && +- dev->data->dev_conf.intr_conf.lsc) { +- mlx5_glue->ack_async_event(&event); +- if (mlx5_link_update(dev, 0) == -EAGAIN) { +- usleep(0); +- continue; +- } +- rte_eth_dev_callback_process +- (dev, RTE_ETH_EVENT_INTR_LSC, NULL); +- continue; +- } + DRV_LOG(DEBUG, + "port %u cannot handle an unknown event (type %d)", + dev->data->port_id, event.event_type); +@@ -1079,7 +1118,6 @@ mlx5_sysfs_switch_info(unsigned int ifindex, struct mlx5_switch_info *info) + bool port_switch_id_set = false; + bool device_dir = false; + char c; +- int ret; + + if (!if_indextoname(ifindex, ifname)) { + rte_errno = errno; +@@ -1095,10 +1133,9 @@ mlx5_sysfs_switch_info(unsigned int ifindex, struct mlx5_switch_info *info) + + file = fopen(phys_port_name, "rb"); + if (file != NULL) { +- ret = fscanf(file, "%" RTE_STR(IF_NAMESIZE) "s", port_name); +- fclose(file); +- if (ret == 1) ++ if (fgets(port_name, IF_NAMESIZE, file) != NULL) + mlx5_translate_port_name(port_name, &data); ++ fclose(file); + } + file = fopen(phys_switch_id, "rb"); + if (file == NULL) { +@@ -1347,15 +1384,16 @@ mlx5_os_read_dev_counters(struct rte_eth_dev *dev, uint64_t *stats) + } + } else { + ret = _mlx5_os_read_dev_counters(dev, -1, stats); ++ if (ret) ++ return ret; + } + /* Read IB counters. */ + for (i = 0; i != xstats_ctrl->mlx5_stats_n; ++i) { + if (!xstats_ctrl->info[i].dev) + continue; +- ret = mlx5_os_read_dev_stat(priv, xstats_ctrl->info[i].ctr_name, +- &stats[i]); + /* return last xstats counter if fail to read. */ +- if (ret != 0) ++ if (mlx5_os_read_dev_stat(priv, xstats_ctrl->info[i].ctr_name, ++ &stats[i]) == 0) + xstats_ctrl->xstats[i] = stats[i]; + else + stats[i] = xstats_ctrl->xstats[i]; +diff --git a/dpdk/drivers/net/mlx5/linux/mlx5_flow_os.c b/dpdk/drivers/net/mlx5/linux/mlx5_flow_os.c +index 893f00b824..a5956c255a 100644 +--- a/dpdk/drivers/net/mlx5/linux/mlx5_flow_os.c ++++ b/dpdk/drivers/net/mlx5/linux/mlx5_flow_os.c +@@ -14,7 +14,8 @@ mlx5_flow_os_init_workspace_once(void) + { + if (rte_thread_key_create(&key_workspace, flow_release_workspace)) { + DRV_LOG(ERR, "Can't create flow workspace data thread key."); +- return -ENOMEM; ++ rte_errno = ENOMEM; ++ return -rte_errno; + } + return 0; + } +diff --git a/dpdk/drivers/net/mlx5/linux/mlx5_os.c b/dpdk/drivers/net/mlx5/linux/mlx5_os.c +index c29fe3d92b..792dd2cb22 100644 +--- a/dpdk/drivers/net/mlx5/linux/mlx5_os.c ++++ b/dpdk/drivers/net/mlx5/linux/mlx5_os.c +@@ -112,7 +112,7 @@ static struct mlx5_indexed_pool_config icfg[] = { + * Pointer to RQ channel object, which includes the channel fd + * + * @param[out] fd +- * The file descriptor (representing the intetrrupt) used in this channel. ++ * The file descriptor (representing the interrupt) used in this channel. + * + * @return + * 0 on successfully setting the fd to non-blocking, non-zero otherwise. +@@ -138,7 +138,7 @@ mlx5_os_set_nonblock_channel_fd(int fd) + * Pointer to mlx5 device attributes. + * + * @return +- * 0 on success, non zero error number otherwise ++ * 0 on success, a negative errno value otherwise and rte_errno is set. + */ + int + mlx5_os_get_dev_attr(struct mlx5_common_device *cdev, +@@ -150,8 +150,10 @@ mlx5_os_get_dev_attr(struct mlx5_common_device *cdev, + + memset(device_attr, 0, sizeof(*device_attr)); + err = mlx5_glue->query_device_ex(ctx, NULL, &attr_ex); +- if (err) +- return err; ++ if (err) { ++ rte_errno = errno; ++ return -rte_errno; ++ } + device_attr->device_cap_flags_ex = attr_ex.device_cap_flags_ex; + device_attr->max_qp_wr = attr_ex.orig_attr.max_qp_wr; + device_attr->max_sge = attr_ex.orig_attr.max_sge; +@@ -170,8 +172,10 @@ mlx5_os_get_dev_attr(struct mlx5_common_device *cdev, + + struct mlx5dv_context dv_attr = { .comp_mask = 0 }; + err = mlx5_glue->dv_query_device(ctx, &dv_attr); +- if (err) +- return err; ++ if (err) { ++ rte_errno = errno; ++ return -rte_errno; ++ } + + device_attr->flags = dv_attr.flags; + device_attr->comp_mask = dv_attr.comp_mask; +@@ -195,7 +199,7 @@ mlx5_os_get_dev_attr(struct mlx5_common_device *cdev, + strlcpy(device_attr->fw_ver, attr_ex.orig_attr.fw_ver, + sizeof(device_attr->fw_ver)); + +- return err; ++ return 0; + } + + /** +@@ -881,10 +885,6 @@ mlx5_dev_spawn(struct rte_device *dpdk_dev, + unsigned int mpls_en = 0; + unsigned int swp = 0; + unsigned int mprq = 0; +- unsigned int mprq_min_stride_size_n = 0; +- unsigned int mprq_max_stride_size_n = 0; +- unsigned int mprq_min_stride_num_n = 0; +- unsigned int mprq_max_stride_num_n = 0; + struct rte_ether_addr mac; + char name[RTE_ETH_NAME_MAX_LEN]; + int own_domain_id = 0; +@@ -981,10 +981,6 @@ mlx5_dev_spawn(struct rte_device *dpdk_dev, + strerror(rte_errno)); + goto error; + } +- if (config->dv_miss_info) { +- if (switch_info->master || switch_info->representor) +- config->dv_xmeta_en = MLX5_XMETA_MODE_META16; +- } + sh = mlx5_alloc_shared_dev_ctx(spawn, config); + if (!sh) + return NULL; +@@ -1039,15 +1035,17 @@ mlx5_dev_spawn(struct rte_device *dpdk_dev, + mprq_caps.max_single_wqe_log_num_of_strides); + DRV_LOG(DEBUG, "\tsupported_qpts: %d", + mprq_caps.supported_qpts); ++ DRV_LOG(DEBUG, "\tmin_stride_wqe_log_size: %d", ++ config->mprq.log_min_stride_wqe_size); + DRV_LOG(DEBUG, "device supports Multi-Packet RQ"); + mprq = 1; +- mprq_min_stride_size_n = ++ config->mprq.log_min_stride_size = + mprq_caps.min_single_stride_log_num_of_bytes; +- mprq_max_stride_size_n = ++ config->mprq.log_max_stride_size = + mprq_caps.max_single_stride_log_num_of_bytes; +- mprq_min_stride_num_n = ++ config->mprq.log_min_stride_num = + mprq_caps.min_single_wqe_log_num_of_strides; +- mprq_max_stride_num_n = ++ config->mprq.log_max_stride_num = + mprq_caps.max_single_wqe_log_num_of_strides; + } + #endif +@@ -1088,7 +1086,7 @@ mlx5_dev_spawn(struct rte_device *dpdk_dev, + " old OFED/rdma-core version or firmware configuration"); + #endif + config->mpls_en = mpls_en; +- nl_rdma = mlx5_nl_init(NETLINK_RDMA); ++ nl_rdma = mlx5_nl_init(NETLINK_RDMA, 0); + /* Check port status. */ + if (spawn->phys_port <= UINT8_MAX) { + /* Legacy Verbs api only support u8 port number. */ +@@ -1135,7 +1133,7 @@ mlx5_dev_spawn(struct rte_device *dpdk_dev, + priv->mtu = RTE_ETHER_MTU; + /* Some internal functions rely on Netlink sockets, open them now. */ + priv->nl_socket_rdma = nl_rdma; +- priv->nl_socket_route = mlx5_nl_init(NETLINK_ROUTE); ++ priv->nl_socket_route = mlx5_nl_init(NETLINK_ROUTE, 0); + priv->representor = !!switch_info->representor; + priv->master = !!switch_info->master; + priv->domain_id = RTE_ETH_DEV_SWITCH_DOMAIN_ID_INVALID; +@@ -1243,6 +1241,32 @@ mlx5_dev_spawn(struct rte_device *dpdk_dev, + } + /* Override some values set by hardware configuration. */ + mlx5_args(config, dpdk_dev->devargs); ++ /* Update final values for devargs before check sibling config. */ ++ if (config->dv_miss_info) { ++ if (switch_info->master || switch_info->representor) ++ config->dv_xmeta_en = MLX5_XMETA_MODE_META16; ++ } ++#if !defined(HAVE_IBV_FLOW_DV_SUPPORT) || !defined(HAVE_MLX5DV_DR) ++ if (config->dv_flow_en) { ++ DRV_LOG(WARNING, "DV flow is not supported."); ++ config->dv_flow_en = 0; ++ } ++#endif ++#ifdef HAVE_MLX5DV_DR_ESWITCH ++ if (!(sh->cdev->config.hca_attr.eswitch_manager && config->dv_flow_en && ++ (switch_info->representor || switch_info->master))) ++ config->dv_esw_en = 0; ++#else ++ config->dv_esw_en = 0; ++#endif ++ if (!priv->config.dv_esw_en && ++ priv->config.dv_xmeta_en != MLX5_XMETA_MODE_LEGACY) { ++ DRV_LOG(WARNING, ++ "Metadata mode %u is not supported (no E-Switch).", ++ priv->config.dv_xmeta_en); ++ priv->config.dv_xmeta_en = MLX5_XMETA_MODE_LEGACY; ++ } ++ /* Check sibling device configurations. */ + err = mlx5_dev_check_sibling_config(priv, config, dpdk_dev); + if (err) + goto error; +@@ -1253,12 +1277,6 @@ mlx5_dev_spawn(struct rte_device *dpdk_dev, + #if !defined(HAVE_IBV_DEVICE_COUNTERS_SET_V42) && \ + !defined(HAVE_IBV_DEVICE_COUNTERS_SET_V45) + DRV_LOG(DEBUG, "counters are not supported"); +-#endif +-#if !defined(HAVE_IBV_FLOW_DV_SUPPORT) || !defined(HAVE_MLX5DV_DR) +- if (config->dv_flow_en) { +- DRV_LOG(WARNING, "DV flow is not supported"); +- config->dv_flow_en = 0; +- } + #endif + config->ind_table_max_size = + sh->device_attr.max_rwq_indirection_table_size; +@@ -1548,36 +1566,7 @@ mlx5_dev_spawn(struct rte_device *dpdk_dev, + config->hw_fcs_strip = 0; + DRV_LOG(DEBUG, "FCS stripping configuration is %ssupported", + (config->hw_fcs_strip ? "" : "not ")); +- if (config->mprq.enabled && mprq) { +- if (config->mprq.stride_num_n && +- (config->mprq.stride_num_n > mprq_max_stride_num_n || +- config->mprq.stride_num_n < mprq_min_stride_num_n)) { +- config->mprq.stride_num_n = +- RTE_MIN(RTE_MAX(MLX5_MPRQ_STRIDE_NUM_N, +- mprq_min_stride_num_n), +- mprq_max_stride_num_n); +- DRV_LOG(WARNING, +- "the number of strides" +- " for Multi-Packet RQ is out of range," +- " setting default value (%u)", +- 1 << config->mprq.stride_num_n); +- } +- if (config->mprq.stride_size_n && +- (config->mprq.stride_size_n > mprq_max_stride_size_n || +- config->mprq.stride_size_n < mprq_min_stride_size_n)) { +- config->mprq.stride_size_n = +- RTE_MIN(RTE_MAX(MLX5_MPRQ_STRIDE_SIZE_N, +- mprq_min_stride_size_n), +- mprq_max_stride_size_n); +- DRV_LOG(WARNING, +- "the size of a stride" +- " for Multi-Packet RQ is out of range," +- " setting default value (%u)", +- 1 << config->mprq.stride_size_n); +- } +- config->mprq.min_stride_size_n = mprq_min_stride_size_n; +- config->mprq.max_stride_size_n = mprq_max_stride_size_n; +- } else if (config->mprq.enabled && !mprq) { ++ if (config->mprq.enabled && !mprq) { + DRV_LOG(WARNING, "Multi-Packet RQ isn't supported"); + config->mprq.enabled = 0; + } +@@ -1676,20 +1665,12 @@ mlx5_dev_spawn(struct rte_device *dpdk_dev, + /* Bring Ethernet device up. */ + DRV_LOG(DEBUG, "port %u forcing Ethernet interface up", + eth_dev->data->port_id); +- mlx5_set_link_up(eth_dev); +- /* +- * Even though the interrupt handler is not installed yet, +- * interrupts will still trigger on the async_fd from +- * Verbs context returned by ibv_open_device(). +- */ ++ /* Read link status in case it is up and there will be no event. */ + mlx5_link_update(eth_dev, 0); +-#ifdef HAVE_MLX5DV_DR_ESWITCH +- if (!(config->hca_attr.eswitch_manager && config->dv_flow_en && +- (switch_info->representor || switch_info->master))) +- config->dv_esw_en = 0; +-#else +- config->dv_esw_en = 0; +-#endif ++ /* Watch LSC interrupts between port probe and port start. */ ++ priv->sh->port[priv->dev_port - 1].nl_ih_port_id = ++ eth_dev->data->port_id; ++ mlx5_set_link_up(eth_dev); + /* Detect minimal data bytes to inline. */ + mlx5_set_min_inline(spawn, config); + /* Store device configuration on private structure. */ +@@ -1743,7 +1724,7 @@ mlx5_dev_spawn(struct rte_device *dpdk_dev, + priv->drop_queue.hrxq = mlx5_drop_action_create(eth_dev); + if (!priv->drop_queue.hrxq) + goto error; +- /* Port representor shares the same max prioirity with pf port. */ ++ /* Port representor shares the same max priority with pf port. */ + if (!priv->sh->flow_priority_check_flag) { + /* Supported Verbs flow priority number detection. */ + err = mlx5_flow_discover_priorities(eth_dev); +@@ -1756,12 +1737,6 @@ mlx5_dev_spawn(struct rte_device *dpdk_dev, + err = -err; + goto error; + } +- if (!priv->config.dv_esw_en && +- priv->config.dv_xmeta_en != MLX5_XMETA_MODE_LEGACY) { +- DRV_LOG(WARNING, "metadata mode %u is not supported " +- "(no E-Switch)", priv->config.dv_xmeta_en); +- priv->config.dv_xmeta_en = MLX5_XMETA_MODE_LEGACY; +- } + mlx5_set_metadata_mask(eth_dev); + if (priv->config.dv_xmeta_en != MLX5_XMETA_MODE_LEGACY && + !priv->sh->dv_regc0_mask) { +@@ -2068,7 +2043,8 @@ mlx5_device_bond_pci_match(const char *ibdev_name, + } + + static void +-mlx5_os_config_default(struct mlx5_dev_config *config) ++mlx5_os_config_default(struct mlx5_dev_config *config, ++ struct mlx5_common_dev_config *cconf) + { + memset(config, 0, sizeof(*config)); + config->mps = MLX5_ARG_UNSET; +@@ -2080,6 +2056,10 @@ mlx5_os_config_default(struct mlx5_dev_config *config) + config->vf_nl_en = 1; + config->mprq.max_memcpy_len = MLX5_MPRQ_MEMCPY_DEFAULT_LEN; + config->mprq.min_rxqs_num = MLX5_MPRQ_MIN_RXQS; ++ config->mprq.log_min_stride_wqe_size = cconf->devx ? ++ cconf->hca_attr.log_min_stride_wqe_sz : ++ MLX5_MPRQ_LOG_MIN_STRIDE_WQE_SIZE; ++ config->mprq.log_stride_num = MLX5_MPRQ_DEFAULT_LOG_STRIDE_NUM; + config->dv_esw_en = 1; + config->dv_flow_en = 1; + config->decap_en = 1; +@@ -2156,8 +2136,8 @@ mlx5_os_pci_probe_pf(struct mlx5_common_device *cdev, + * matching ones, gathering into the list. + */ + struct ibv_device *ibv_match[ret + 1]; +- int nl_route = mlx5_nl_init(NETLINK_ROUTE); +- int nl_rdma = mlx5_nl_init(NETLINK_RDMA); ++ int nl_route = mlx5_nl_init(NETLINK_ROUTE, 0); ++ int nl_rdma = mlx5_nl_init(NETLINK_RDMA, 0); + unsigned int i; + + while (ret-- > 0) { +@@ -2209,9 +2189,9 @@ mlx5_os_pci_probe_pf(struct mlx5_common_device *cdev, + if (!nd) { + /* No device matches, just complain and bail out. */ + DRV_LOG(WARNING, +- "No Verbs device matches PCI device " PCI_PRI_FMT "," ++ "PF %u doesn't have Verbs device matches PCI device " PCI_PRI_FMT "," + " are kernel drivers loaded?", +- owner_pci.domain, owner_pci.bus, ++ owner_id, owner_pci.domain, owner_pci.bus, + owner_pci.devid, owner_pci.function); + rte_errno = ENOENT; + ret = -rte_errno; +@@ -2300,7 +2280,7 @@ mlx5_os_pci_probe_pf(struct mlx5_common_device *cdev, + /* + * Force standalone bonding + * device for ROCE LAG +- * confgiurations. ++ * configurations. + */ + list[ns].info.master = 0; + list[ns].info.representor = 0; +@@ -2496,7 +2476,7 @@ mlx5_os_pci_probe_pf(struct mlx5_common_device *cdev, + uint32_t restore; + + /* Default configuration. */ +- mlx5_os_config_default(&dev_config); ++ mlx5_os_config_default(&dev_config, &cdev->config); + dev_config.vf = dev_config_vf; + list[i].eth_dev = mlx5_dev_spawn(cdev->dev, &list[i], + &dev_config, ð_da); +@@ -2632,16 +2612,16 @@ mlx5_os_pci_probe(struct mlx5_common_device *cdev) + for (p = 0; p < eth_da.nb_ports; p++) { + ret = mlx5_os_pci_probe_pf(cdev, ð_da, + eth_da.ports[p]); +- if (ret) +- break; +- } +- if (ret) { +- DRV_LOG(ERR, "Probe of PCI device " PCI_PRI_FMT " " +- "aborted due to proding failure of PF %u", +- pci_dev->addr.domain, pci_dev->addr.bus, +- pci_dev->addr.devid, pci_dev->addr.function, +- eth_da.ports[p]); +- mlx5_net_remove(cdev); ++ if (ret) { ++ DRV_LOG(INFO, "Probe of PCI device " PCI_PRI_FMT " " ++ "aborted due to proding failure of PF %u", ++ pci_dev->addr.domain, pci_dev->addr.bus, ++ pci_dev->addr.devid, pci_dev->addr.function, ++ eth_da.ports[p]); ++ mlx5_net_remove(cdev); ++ if (p != 0) ++ break; ++ } + } + } else { + ret = mlx5_os_pci_probe_pf(cdev, ð_da, 0); +@@ -2666,7 +2646,7 @@ mlx5_os_auxiliary_probe(struct mlx5_common_device *cdev) + if (ret != 0) + return ret; + /* Set default config data. */ +- mlx5_os_config_default(&config); ++ mlx5_os_config_default(&config, &cdev->config); + config.sf = 1; + /* Init spawn data. */ + spawn.max_port = 1; +@@ -2733,6 +2713,40 @@ mlx5_os_net_cleanup(void) + mlx5_pmd_socket_uninit(); + } + ++static int ++mlx5_os_dev_shared_handler_install_lsc(struct mlx5_dev_ctx_shared *sh) ++{ ++ int nlsk_fd, flags, ret; ++ ++ nlsk_fd = mlx5_nl_init(NETLINK_ROUTE, RTMGRP_LINK); ++ if (nlsk_fd < 0) { ++ DRV_LOG(ERR, "Failed to create a socket for Netlink events: %s", ++ rte_strerror(rte_errno)); ++ return -1; ++ } ++ flags = fcntl(nlsk_fd, F_GETFL); ++ ret = fcntl(nlsk_fd, F_SETFL, flags | O_NONBLOCK); ++ if (ret != 0) { ++ DRV_LOG(ERR, "Failed to make Netlink event socket non-blocking: %s", ++ strerror(errno)); ++ rte_errno = errno; ++ goto error; ++ } ++ rte_intr_type_set(sh->intr_handle_nl, RTE_INTR_HANDLE_EXT); ++ rte_intr_fd_set(sh->intr_handle_nl, nlsk_fd); ++ if (rte_intr_callback_register(sh->intr_handle_nl, ++ mlx5_dev_interrupt_handler_nl, ++ sh) != 0) { ++ DRV_LOG(ERR, "Failed to register Netlink events interrupt"); ++ rte_intr_fd_set(sh->intr_handle_nl, -1); ++ goto error; ++ } ++ return 0; ++error: ++ close(nlsk_fd); ++ return -1; ++} + -+

-+ This table primarily configures the Open vSwitch database server -+ (ovsdb-server). -+

+ /** + * Install shared asynchronous device events handler. + * This function is implemented to support event sharing +@@ -2770,6 +2784,18 @@ mlx5_os_dev_shared_handler_install(struct mlx5_dev_ctx_shared *sh) + rte_intr_fd_set(sh->intr_handle, -1); + } + } ++ sh->intr_handle_nl = rte_intr_instance_alloc ++ (RTE_INTR_INSTANCE_F_SHARED); ++ if (sh->intr_handle_nl == NULL) { ++ DRV_LOG(ERR, "Fail to allocate intr_handle"); ++ rte_errno = ENOMEM; ++ return; ++ } ++ rte_intr_fd_set(sh->intr_handle_nl, -1); ++ if (mlx5_os_dev_shared_handler_install_lsc(sh) < 0) { ++ DRV_LOG(INFO, "Fail to install the shared Netlink event handler."); ++ rte_intr_fd_set(sh->intr_handle_nl, -1); ++ } + if (sh->devx) { + #ifdef HAVE_IBV_DEVX_ASYNC + sh->intr_handle_devx = +@@ -2817,10 +2843,19 @@ mlx5_os_dev_shared_handler_install(struct mlx5_dev_ctx_shared *sh) + void + mlx5_os_dev_shared_handler_uninstall(struct mlx5_dev_ctx_shared *sh) + { ++ int nlsk_fd; ++ + if (rte_intr_fd_get(sh->intr_handle) >= 0) + mlx5_intr_callback_unregister(sh->intr_handle, + mlx5_dev_interrupt_handler, sh); + rte_intr_instance_free(sh->intr_handle); ++ nlsk_fd = rte_intr_fd_get(sh->intr_handle_nl); ++ if (nlsk_fd >= 0) { ++ mlx5_intr_callback_unregister ++ (sh->intr_handle_nl, mlx5_dev_interrupt_handler_nl, sh); ++ close(nlsk_fd); ++ } ++ rte_intr_instance_free(sh->intr_handle_nl); + #ifdef HAVE_IBV_DEVX_ASYNC + if (rte_intr_fd_get(sh->intr_handle_devx) >= 0) + rte_intr_callback_unregister(sh->intr_handle_devx, +diff --git a/dpdk/drivers/net/mlx5/linux/mlx5_verbs.c b/dpdk/drivers/net/mlx5/linux/mlx5_verbs.c +index 58556d2bf0..b113731097 100644 +--- a/dpdk/drivers/net/mlx5/linux/mlx5_verbs.c ++++ b/dpdk/drivers/net/mlx5/linux/mlx5_verbs.c +@@ -94,7 +94,6 @@ mlx5_ibv_modify_qp(struct mlx5_txq_obj *obj, enum mlx5_txq_modify_type type, + .qp_state = IBV_QPS_RESET, + .port_num = dev_port, + }; +- int attr_mask = (IBV_QP_STATE | IBV_QP_PORT); + int ret; + + if (type != MLX5_TXQ_MOD_RST2RDY) { +@@ -108,10 +107,8 @@ mlx5_ibv_modify_qp(struct mlx5_txq_obj *obj, enum mlx5_txq_modify_type type, + if (type == MLX5_TXQ_MOD_RDY2RST) + return 0; + } +- if (type == MLX5_TXQ_MOD_ERR2RDY) +- attr_mask = IBV_QP_STATE; + mod.qp_state = IBV_QPS_INIT; +- ret = mlx5_glue->modify_qp(obj->qp, &mod, attr_mask); ++ ret = mlx5_glue->modify_qp(obj->qp, &mod, IBV_QP_STATE | IBV_QP_PORT); + if (ret) { + DRV_LOG(ERR, "Cannot change Tx QP state to INIT %s", + strerror(errno)); +@@ -272,8 +269,8 @@ mlx5_rxq_ibv_wq_create(struct mlx5_rxq_priv *rxq) + + wq_attr.mlx5.comp_mask |= MLX5DV_WQ_INIT_ATTR_MASK_STRIDING_RQ; + *mprq_attr = (struct mlx5dv_striding_rq_init_attr){ +- .single_stride_log_num_of_bytes = rxq_data->strd_sz_n, +- .single_wqe_log_num_of_strides = rxq_data->strd_num_n, ++ .single_stride_log_num_of_bytes = rxq_data->log_strd_sz, ++ .single_wqe_log_num_of_strides = rxq_data->log_strd_num, + .two_byte_shift_en = MLX5_MPRQ_TWO_BYTE_SHIFT, + }; + } +diff --git a/dpdk/drivers/net/mlx5/linux/mlx5_vlan_os.c b/dpdk/drivers/net/mlx5/linux/mlx5_vlan_os.c +index 005904bdfe..7ee2460a23 100644 +--- a/dpdk/drivers/net/mlx5/linux/mlx5_vlan_os.c ++++ b/dpdk/drivers/net/mlx5/linux/mlx5_vlan_os.c +@@ -136,7 +136,7 @@ mlx5_vlan_vmwa_init(struct rte_eth_dev *dev, uint32_t ifindex) + return NULL; + } + rte_spinlock_init(&vmwa->sl); +- vmwa->nl_socket = mlx5_nl_init(NETLINK_ROUTE); ++ vmwa->nl_socket = mlx5_nl_init(NETLINK_ROUTE, 0); + if (vmwa->nl_socket < 0) { + DRV_LOG(WARNING, + "Can not create Netlink socket" +diff --git a/dpdk/drivers/net/mlx5/mlx5.c b/dpdk/drivers/net/mlx5/mlx5.c +index aa5f313c1a..2234dc7563 100644 +--- a/dpdk/drivers/net/mlx5/mlx5.c ++++ b/dpdk/drivers/net/mlx5/mlx5.c +@@ -1172,12 +1172,11 @@ mlx5_alloc_shared_dev_ctx(const struct mlx5_dev_spawn_data *spawn, + MLX5_ASSERT(spawn->max_port); + sh = mlx5_malloc(MLX5_MEM_ZERO | MLX5_MEM_RTE, + sizeof(struct mlx5_dev_ctx_shared) + +- spawn->max_port * +- sizeof(struct mlx5_dev_shared_port), ++ spawn->max_port * sizeof(struct mlx5_dev_shared_port), + RTE_CACHE_LINE_SIZE, SOCKET_ID_ANY); + if (!sh) { +- DRV_LOG(ERR, "shared context allocation failure"); +- rte_errno = ENOMEM; ++ DRV_LOG(ERR, "Shared context allocation failure."); ++ rte_errno = ENOMEM; + goto exit; + } + pthread_mutex_init(&sh->txpp.mutex, NULL); +@@ -1199,24 +1198,24 @@ mlx5_alloc_shared_dev_ctx(const struct mlx5_dev_spawn_data *spawn, + strncpy(sh->ibdev_path, mlx5_os_get_ctx_device_path(sh->cdev->ctx), + sizeof(sh->ibdev_path) - 1); + /* +- * Setting port_id to max unallowed value means +- * there is no interrupt subhandler installed for +- * the given port index i. ++ * Setting port_id to max unallowed value means there is no interrupt ++ * subhandler installed for the given port index i. + */ + for (i = 0; i < sh->max_port; i++) { + sh->port[i].ih_port_id = RTE_MAX_ETHPORTS; + sh->port[i].devx_ih_port_id = RTE_MAX_ETHPORTS; ++ sh->port[i].nl_ih_port_id = RTE_MAX_ETHPORTS; + } + if (sh->devx) { + sh->td = mlx5_devx_cmd_create_td(sh->cdev->ctx); + if (!sh->td) { + DRV_LOG(ERR, "TD allocation failure"); +- err = ENOMEM; ++ rte_errno = ENOMEM; + goto error; + } + if (mlx5_setup_tis(sh)) { + DRV_LOG(ERR, "TIS allocation failure"); +- err = ENOMEM; ++ rte_errno = ENOMEM; + goto error; + } + err = mlx5_rxtx_uars_prepare(sh); +@@ -1246,19 +1245,19 @@ mlx5_alloc_shared_dev_ctx(const struct mlx5_dev_spawn_data *spawn, + pthread_mutex_unlock(&mlx5_dev_ctx_list_mutex); + return sh; + error: ++ err = rte_errno; + pthread_mutex_destroy(&sh->txpp.mutex); + pthread_mutex_unlock(&mlx5_dev_ctx_list_mutex); + MLX5_ASSERT(sh); +- if (sh->td) +- claim_zero(mlx5_devx_cmd_destroy(sh->td)); ++ mlx5_rxtx_uars_release(sh); + i = 0; + do { + if (sh->tis[i]) + claim_zero(mlx5_devx_cmd_destroy(sh->tis[i])); + } while (++i < (uint32_t)sh->bond.n_port); +- mlx5_rxtx_uars_release(sh); ++ if (sh->td) ++ claim_zero(mlx5_devx_cmd_destroy(sh->td)); + mlx5_free(sh); +- MLX5_ASSERT(err > 0); + rte_errno = err; + return NULL; + } +@@ -1321,6 +1320,8 @@ mlx5_free_shared_dev_ctx(struct mlx5_dev_ctx_shared *sh) + * Only primary process handles async device events. + **/ + mlx5_flow_counters_mng_close(sh); ++ if (sh->ct_mng) ++ mlx5_flow_aso_ct_mng_close(sh); + if (sh->aso_age_mng) { + mlx5_flow_aso_age_mng_close(sh); + sh->aso_age_mng = NULL; +@@ -1594,8 +1595,6 @@ mlx5_dev_close(struct rte_eth_dev *dev) + if (priv->mreg_cp_tbl) + mlx5_hlist_destroy(priv->mreg_cp_tbl); + mlx5_mprq_free_mp(dev); +- if (priv->sh->ct_mng) +- mlx5_flow_aso_ct_mng_close(priv->sh); + mlx5_os_free_shared_dr(priv); + if (priv->rss_conf.rss_key != NULL) + mlx5_free(priv->rss_conf.rss_key); +@@ -1642,7 +1641,7 @@ mlx5_dev_close(struct rte_eth_dev *dev) + /* + * Free the shared context in last turn, because the cleanup + * routines above may use some shared fields, like +- * mlx5_os_mac_addr_flush() uses ibdev_path for retrieveing ++ * mlx5_os_mac_addr_flush() uses ibdev_path for retrieving + * ifindex if Netlink fails. + */ + mlx5_free_shared_dev_ctx(priv->sh); +@@ -1884,9 +1883,9 @@ mlx5_args_check(const char *key, const char *val, void *opaque) + } else if (strcmp(MLX5_RX_MPRQ_EN, key) == 0) { + config->mprq.enabled = !!tmp; + } else if (strcmp(MLX5_RX_MPRQ_LOG_STRIDE_NUM, key) == 0) { +- config->mprq.stride_num_n = tmp; ++ config->mprq.log_stride_num = tmp; + } else if (strcmp(MLX5_RX_MPRQ_LOG_STRIDE_SIZE, key) == 0) { +- config->mprq.stride_size_n = tmp; ++ config->mprq.log_stride_size = tmp; + } else if (strcmp(MLX5_RX_MPRQ_MAX_MEMCPY_LEN, key) == 0) { + config->mprq.max_memcpy_len = tmp; + } else if (strcmp(MLX5_RXQS_MIN_MPRQ, key) == 0) { +@@ -1962,7 +1961,7 @@ mlx5_args_check(const char *key, const char *val, void *opaque) + if (tmp != MLX5_RCM_NONE && + tmp != MLX5_RCM_LIGHT && + tmp != MLX5_RCM_AGGR) { +- DRV_LOG(ERR, "Unrecognize %s: \"%s\"", key, val); ++ DRV_LOG(ERR, "Unrecognized %s: \"%s\"", key, val); + rte_errno = EINVAL; + return -rte_errno; + } +@@ -2177,17 +2176,17 @@ mlx5_set_metadata_mask(struct rte_eth_dev *dev) + break; + } + if (sh->dv_mark_mask && sh->dv_mark_mask != mark) +- DRV_LOG(WARNING, "metadata MARK mask mismatche %08X:%08X", ++ DRV_LOG(WARNING, "metadata MARK mask mismatch %08X:%08X", + sh->dv_mark_mask, mark); + else + sh->dv_mark_mask = mark; + if (sh->dv_meta_mask && sh->dv_meta_mask != meta) +- DRV_LOG(WARNING, "metadata META mask mismatche %08X:%08X", ++ DRV_LOG(WARNING, "metadata META mask mismatch %08X:%08X", + sh->dv_meta_mask, meta); + else + sh->dv_meta_mask = meta; + if (sh->dv_regc0_mask && sh->dv_regc0_mask != reg_c0) +- DRV_LOG(WARNING, "metadata reg_c0 mask mismatche %08X:%08X", ++ DRV_LOG(WARNING, "metadata reg_c0 mask mismatch %08X:%08X", + sh->dv_meta_mask, reg_c0); + else + sh->dv_regc0_mask = reg_c0; +diff --git a/dpdk/drivers/net/mlx5/mlx5.h b/dpdk/drivers/net/mlx5/mlx5.h +index 8466531060..128ebd6937 100644 +--- a/dpdk/drivers/net/mlx5/mlx5.h ++++ b/dpdk/drivers/net/mlx5/mlx5.h +@@ -275,10 +275,14 @@ struct mlx5_dev_config { + unsigned int hp_delay_drop:1; /* Enable hairpin Rxq delay drop. */ + struct { + unsigned int enabled:1; /* Whether MPRQ is enabled. */ +- unsigned int stride_num_n; /* Number of strides. */ +- unsigned int stride_size_n; /* Size of a stride. */ +- unsigned int min_stride_size_n; /* Min size of a stride. */ +- unsigned int max_stride_size_n; /* Max size of a stride. */ ++ unsigned int log_stride_num; /* Log number of strides. */ ++ unsigned int log_stride_size; /* Log size of a stride. */ ++ unsigned int log_min_stride_size; /* Log min size of a stride.*/ ++ unsigned int log_max_stride_size; /* Log max size of a stride.*/ ++ unsigned int log_min_stride_num; /* Log min num of strides. */ ++ unsigned int log_max_stride_num; /* Log max num of strides. */ ++ unsigned int log_min_stride_wqe_size; ++ /* Log min WQE size, (size of single stride)*(num of strides).*/ + unsigned int max_memcpy_len; + /* Maximum packet size to memcpy Rx packets. */ + unsigned int min_rxqs_num; +@@ -601,6 +605,7 @@ struct mlx5_age_info { + struct mlx5_dev_shared_port { + uint32_t ih_port_id; + uint32_t devx_ih_port_id; ++ uint32_t nl_ih_port_id; + /* + * Interrupt handler port_id. Used by shared interrupt + * handler to find the corresponding rte_eth device +@@ -742,6 +747,8 @@ struct mlx5_flow_meter_policy { + /* If yellow color policy is skipped. */ + uint32_t skip_g:1; + /* If green color policy is skipped. */ ++ uint32_t mark:1; ++ /* If policy contains mark action. */ + rte_spinlock_t sl; + uint32_t ref_cnt; + /* Use count. */ +@@ -956,7 +963,6 @@ union mlx5_flow_tbl_key { + /* Table structure. */ + struct mlx5_flow_tbl_resource { + void *obj; /**< Pointer to DR table object. */ +- uint32_t refcnt; /**< Reference counter. */ + }; + + #define MLX5_MAX_TABLES UINT16_MAX +@@ -977,7 +983,7 @@ struct mlx5_flow_id_pool { + uint32_t base_index; + /**< The next index that can be used without any free elements. */ + uint32_t *curr; /**< Pointer to the index to pop. */ +- uint32_t *last; /**< Pointer to the last element in the empty arrray. */ ++ uint32_t *last; /**< Pointer to the last element in the empty array. */ + uint32_t max_id; /**< Maximum id can be allocated from the pool. */ + }; + +@@ -1014,7 +1020,7 @@ struct mlx5_dev_txpp { + void *pp; /* Packet pacing context. */ + uint16_t pp_id; /* Packet pacing context index. */ + uint16_t ts_n; /* Number of captured timestamps. */ +- uint16_t ts_p; /* Pointer to statisticks timestamp. */ ++ uint16_t ts_p; /* Pointer to statistics timestamp. */ + struct mlx5_txpp_ts *tsa; /* Timestamps sliding window stats. */ + struct mlx5_txpp_ts ts; /* Cached completion id/timestamp. */ + uint32_t sync_lost:1; /* ci/timestamp synchronization lost. */ +@@ -1118,7 +1124,7 @@ struct mlx5_flex_parser_devx { + uint32_t sample_ids[MLX5_GRAPH_NODE_SAMPLE_NUM]; + }; + +-/* Pattern field dscriptor - how to translate flex pattern into samples. */ ++/* Pattern field descriptor - how to translate flex pattern into samples. */ + __extension__ + struct mlx5_flex_pattern_field { + uint16_t width:6; +@@ -1169,7 +1175,7 @@ struct mlx5_dev_ctx_shared { + /* Shared DV/DR flow data section. */ + uint32_t dv_meta_mask; /* flow META metadata supported mask. */ + uint32_t dv_mark_mask; /* flow MARK metadata supported mask. */ +- uint32_t dv_regc0_mask; /* available bits of metatada reg_c[0]. */ ++ uint32_t dv_regc0_mask; /* available bits of metadata reg_c[0]. */ + void *fdb_domain; /* FDB Direct Rules name space handle. */ + void *rx_domain; /* RX Direct Rules name space handle. */ + void *tx_domain; /* TX Direct Rules name space handle. */ +@@ -1199,6 +1205,7 @@ struct mlx5_dev_ctx_shared { + /* Shared interrupt handler section. */ + struct rte_intr_handle *intr_handle; /* Interrupt handler for device. */ + struct rte_intr_handle *intr_handle_devx; /* DEVX interrupt handler. */ ++ struct rte_intr_handle *intr_handle_nl; /* Netlink interrupt handler. */ + void *devx_comp; /* DEVX async comp obj. */ + struct mlx5_devx_obj *tis[16]; /* TIS object. */ + struct mlx5_devx_obj *td; /* Transport domain. */ +@@ -1409,6 +1416,7 @@ struct mlx5_priv { + unsigned int mtr_en:1; /* Whether support meter. */ + unsigned int mtr_reg_share:1; /* Whether support meter REG_C share. */ + unsigned int lb_used:1; /* Loopback queue is referred to. */ ++ uint32_t mark_enabled:1; /* If mark action is enabled on rxqs. */ + uint16_t domain_id; /* Switch domain identifier. */ + uint16_t vport_id; /* Associated VF vport index (if any). */ + uint32_t vport_meta_tag; /* Used for vport index match ove VF LAG. */ +@@ -1580,6 +1588,7 @@ int mlx5_dev_set_flow_ctrl(struct rte_eth_dev *dev, + struct rte_eth_fc_conf *fc_conf); + void mlx5_dev_interrupt_handler(void *arg); + void mlx5_dev_interrupt_handler_devx(void *arg); ++void mlx5_dev_interrupt_handler_nl(void *arg); + int mlx5_set_link_down(struct rte_eth_dev *dev); + int mlx5_set_link_up(struct rte_eth_dev *dev); + int mlx5_is_removed(struct rte_eth_dev *dev); +diff --git a/dpdk/drivers/net/mlx5/mlx5_defs.h b/dpdk/drivers/net/mlx5/mlx5_defs.h +index 258475ed2c..2d48fde010 100644 +--- a/dpdk/drivers/net/mlx5/mlx5_defs.h ++++ b/dpdk/drivers/net/mlx5/mlx5_defs.h +@@ -50,7 +50,7 @@ + #define MLX5_MAX_XSTATS 32 + + /* Maximum Packet headers size (L2+L3+L4) for TSO. */ +-#define MLX5_MAX_TSO_HEADER (128u + 34u) ++#define MLX5_MAX_TSO_HEADER 192U + + /* Inline data size required by NICs. */ + #define MLX5_INLINE_HSIZE_NONE 0 +@@ -113,10 +113,10 @@ + #define MLX5_UAR_PAGE_NUM_MASK ((MLX5_UAR_PAGE_NUM_MAX) - 1) + + /* Log 2 of the default number of strides per WQE for Multi-Packet RQ. */ +-#define MLX5_MPRQ_STRIDE_NUM_N 6U ++#define MLX5_MPRQ_DEFAULT_LOG_STRIDE_NUM 6U + + /* Log 2 of the default size of a stride per WQE for Multi-Packet RQ. */ +-#define MLX5_MPRQ_STRIDE_SIZE_N 11U ++#define MLX5_MPRQ_DEFAULT_LOG_STRIDE_SIZE 11U + + /* Two-byte shift is disabled for Multi-Packet RQ. */ + #define MLX5_MPRQ_TWO_BYTE_SHIFT 0 +diff --git a/dpdk/drivers/net/mlx5/mlx5_devx.c b/dpdk/drivers/net/mlx5/mlx5_devx.c +index 105c3d67f0..44c439bb55 100644 +--- a/dpdk/drivers/net/mlx5/mlx5_devx.c ++++ b/dpdk/drivers/net/mlx5/mlx5_devx.c +@@ -257,11 +257,11 @@ mlx5_rxq_create_devx_rq_resources(struct mlx5_rxq_priv *rxq) + * 512*2^single_wqe_log_num_of_strides. + */ + rq_attr.wq_attr.single_wqe_log_num_of_strides = +- rxq_data->strd_num_n - ++ rxq_data->log_strd_num - + MLX5_MIN_SINGLE_WQE_LOG_NUM_STRIDES; + /* Stride size = (2^single_stride_log_num_of_bytes)*64B. */ + rq_attr.wq_attr.single_stride_log_num_of_bytes = +- rxq_data->strd_sz_n - ++ rxq_data->log_strd_sz - + MLX5_MIN_SINGLE_STRIDE_LOG_NUM_BYTES; + wqe_size = sizeof(struct mlx5_wqe_mprq); + } else { +@@ -706,7 +706,7 @@ mlx5_devx_tir_attr_set(struct rte_eth_dev *dev, const uint8_t *rss_key, + { + struct mlx5_priv *priv = dev->data->dev_private; + enum mlx5_rxq_type rxq_obj_type; +- bool lro = true; ++ bool lro = false; + uint32_t i; + + /* NULL queues designate drop queue. */ +@@ -715,6 +715,7 @@ mlx5_devx_tir_attr_set(struct rte_eth_dev *dev, const uint8_t *rss_key, + mlx5_rxq_ctrl_get(dev, ind_tbl->queues[0]); + rxq_obj_type = rxq_ctrl != NULL ? rxq_ctrl->type : + MLX5_RXQ_TYPE_STANDARD; ++ lro = true; + + /* Enable TIR LRO only if all the queues were configured for. */ + for (i = 0; i < ind_tbl->queues_n; ++i) { +@@ -768,6 +769,7 @@ mlx5_devx_tir_attr_set(struct rte_eth_dev *dev, const uint8_t *rss_key, + tir_attr->self_lb_block = + MLX5_TIRC_SELF_LB_BLOCK_BLOCK_UNICAST; + if (lro) { ++ MLX5_ASSERT(priv->config.lro.supported); + tir_attr->lro_timeout_period_usecs = priv->config.lro.timeout; + tir_attr->lro_max_msg_sz = priv->max_lro_msg_size; + tir_attr->lro_enable_mask = +@@ -931,6 +933,8 @@ mlx5_rxq_devx_obj_drop_create(struct rte_eth_dev *dev) + rte_errno = ENOMEM; + goto error; + } ++ /* set the CPU socket ID where the rxq_ctrl was allocated */ ++ rxq_ctrl->socket = socket_id; + rxq_obj->rxq_ctrl = rxq_ctrl; + rxq_ctrl->type = MLX5_RXQ_TYPE_STANDARD; + rxq_ctrl->sh = priv->sh; +diff --git a/dpdk/drivers/net/mlx5/mlx5_ethdev.c b/dpdk/drivers/net/mlx5/mlx5_ethdev.c +index dc647d5580..9c44471c42 100644 +--- a/dpdk/drivers/net/mlx5/mlx5_ethdev.c ++++ b/dpdk/drivers/net/mlx5/mlx5_ethdev.c +@@ -109,7 +109,7 @@ mlx5_dev_configure(struct rte_eth_dev *dev) + MLX5_MEM_RTE | MLX5_MEM_ZERO, + sizeof(void *) * rxqs_n, 0, + SOCKET_ID_ANY); +- if (priv->rxq_privs == NULL) { ++ if (rxqs_n && priv->rxq_privs == NULL) { + DRV_LOG(ERR, "port %u cannot allocate rxq private data", + dev->data->port_id); + rte_errno = ENOMEM; +diff --git a/dpdk/drivers/net/mlx5/mlx5_flow.c b/dpdk/drivers/net/mlx5/mlx5_flow.c +index f34e4b88aa..42de516bfd 100644 +--- a/dpdk/drivers/net/mlx5/mlx5_flow.c ++++ b/dpdk/drivers/net/mlx5/mlx5_flow.c +@@ -18,6 +18,7 @@ + #include + #include + #include ++#include + + #include + #include +@@ -148,6 +149,8 @@ mlx5_flow_is_rss_expandable_item(const struct rte_flow_item *item) + case RTE_FLOW_ITEM_TYPE_IPV6: + case RTE_FLOW_ITEM_TYPE_UDP: + case RTE_FLOW_ITEM_TYPE_TCP: ++ case RTE_FLOW_ITEM_TYPE_ICMP: ++ case RTE_FLOW_ITEM_TYPE_ICMP6: + case RTE_FLOW_ITEM_TYPE_VXLAN: + case RTE_FLOW_ITEM_TYPE_NVGRE: + case RTE_FLOW_ITEM_TYPE_GRE: +@@ -164,128 +167,152 @@ mlx5_flow_is_rss_expandable_item(const struct rte_flow_item *item) + return false; + } + ++/** ++ * Network Service Header (NSH) and its next protocol values ++ * are described in RFC-8393. ++ */ ++static enum rte_flow_item_type ++mlx5_nsh_proto_to_item_type(uint8_t proto_spec, uint8_t proto_mask) ++{ ++ enum rte_flow_item_type type; + -+

-+ The Open vSwitch database server can initiate and maintain active -+ connections to remote clients. It can also listen for database -+ connections. -+

++ switch (proto_mask & proto_spec) { ++ case 0: ++ type = RTE_FLOW_ITEM_TYPE_VOID; ++ break; ++ case RTE_VXLAN_GPE_TYPE_IPV4: ++ type = RTE_FLOW_ITEM_TYPE_IPV4; ++ break; ++ case RTE_VXLAN_GPE_TYPE_IPV6: ++ type = RTE_VXLAN_GPE_TYPE_IPV6; ++ break; ++ case RTE_VXLAN_GPE_TYPE_ETH: ++ type = RTE_FLOW_ITEM_TYPE_ETH; ++ break; ++ default: ++ type = RTE_FLOW_ITEM_TYPE_END; ++ } ++ return type; ++} + -+ -+ -+

Connection methods for clients.

-+

-+ The following connection methods are currently supported: -+

-+
-+
-+ ssl:host[:port] -+
-+
-+

-+ The specified SSL port on the host at the given -+ host, which can either be a DNS name (if built with -+ unbound library) or an IP address. A valid SSL configuration must -+ be provided when this form is used, this configuration can be -+ specified via command-line options or the -+ table. -+

-+

-+ If port is not specified, it defaults to 6640. -+

-+

-+ SSL support is an optional feature that is not always -+ built as part of Open vSwitch. -+

-+
++static enum rte_flow_item_type ++mlx5_inet_proto_to_item_type(uint8_t proto_spec, uint8_t proto_mask) ++{ ++ enum rte_flow_item_type type; + -+
-+ tcp:host[:port] -+
-+
-+

-+ The specified TCP port on the host at the given -+ host, which can either be a DNS name (if built with -+ unbound library) or an IP address. If host is an IPv6 -+ address, wrap it in square brackets, e.g. -+ tcp:[::1]:6640. -+

-+

-+ If port is not specified, it defaults to 6640. -+

-+
-+
-+ pssl:[port][:host] -+
-+
-+

-+ Listens for SSL connections on the specified TCP port. -+ Specify 0 for port to have the kernel automatically -+ choose an available port. If host, which can either -+ be a DNS name (if built with unbound library) or an IP address, -+ is specified, then connections are restricted to the resolved or -+ specified local IPaddress (either IPv4 or IPv6 address). If -+ host is an IPv6 address, wrap in square brackets, -+ e.g. pssl:6640:[::1]. If host is not -+ specified then it listens only on IPv4 (but not IPv6) addresses. -+ A valid SSL configuration must be provided when this form is -+ used, this can be specified either via command-line options or -+ the table. -+

-+

-+ If port is not specified, it defaults to 6640. -+

-+

-+ SSL support is an optional feature that is not always built as -+ part of Open vSwitch. -+

-+
-+
-+ ptcp:[port][:host] -+
-+
-+

-+ Listens for connections on the specified TCP port. -+ Specify 0 for port to have the kernel automatically -+ choose an available port. If host, which can either -+ be a DNS name (if built with unbound library) or an IP address, -+ is specified, then connections are restricted to the resolved or -+ specified local IP address (either IPv4 or IPv6 address). If -+ host is an IPv6 address, wrap it in square brackets, -+ e.g. ptcp:6640:[::1]. If host is not -+ specified then it listens only on IPv4 addresses. -+

-+

-+ If port is not specified, it defaults to 6640. -+

-+
-+
-+

When multiple clients are configured, the -+ values must be unique. Duplicate values yield -+ unspecified results.

-+ ++ switch (proto_mask & proto_spec) { ++ case 0: ++ type = RTE_FLOW_ITEM_TYPE_VOID; ++ break; ++ case IPPROTO_UDP: ++ type = RTE_FLOW_ITEM_TYPE_UDP; ++ break; ++ case IPPROTO_TCP: ++ type = RTE_FLOW_ITEM_TYPE_TCP; ++ break; ++ case IPPROTO_IPIP: ++ type = RTE_FLOW_ITEM_TYPE_IPV4; ++ break; ++ case IPPROTO_IPV6: ++ type = RTE_FLOW_ITEM_TYPE_IPV6; ++ break; ++ default: ++ type = RTE_FLOW_ITEM_TYPE_END; ++ } ++ return type; ++} + -+ -+ true to restrict these connections to read-only -+ transactions, false to allow them to modify the database. -+ -+ -+ -+ String containing role name for this connection entry. -+ -+ -+ -+ -+ -+ Maximum number of milliseconds to wait between connection attempts. -+ Default is implementation-specific. -+ -+ -+ -+ Maximum number of milliseconds of idle time on connection to the client -+ before sending an inactivity probe message. If Open vSwitch does not -+ communicate with the client for the specified number of seconds, it -+ will send a probe. If a response is not received for the same -+ additional amount of time, Open vSwitch assumes the connection has been -+ broken and attempts to reconnect. Default is implementation-specific. -+ A value of 0 disables inactivity probes. -+ -+ -+ -+ -+

-+ Key-value pair of is always updated. -+ Other key-value pairs in the status columns may be updated depends -+ on the type. -+

-+ -+

-+ When specifies a connection method that -+ listens for inbound connections (e.g. ptcp: or -+ punix:), both and -+ may also be updated while the -+ remaining key-value pairs are omitted. -+

-+ -+

-+ On the other hand, when specifies an -+ outbound connection, all key-value pairs may be updated, except -+ the above-mentioned two key-value pairs associated with inbound -+ connection targets. They are omitted. -+

-+ -+ -+ true if currently connected to this client, -+ false otherwise. -+ -+ -+ -+ A human-readable description of the last error on the connection -+ to the manager; i.e. strerror(errno). This key -+ will exist only if an error has occurred. -+ -+ -+ -+

-+ The state of the connection to the manager: -+

-+
-+
VOID
-+
Connection is disabled.
-+ -+
BACKOFF
-+
Attempting to reconnect at an increasing period.
-+ -+
CONNECTING
-+
Attempting to connect.
-+ -+
ACTIVE
-+
Connected, remote host responsive.
-+ -+
IDLE
-+
Connection is idle. Waiting for response to keep-alive.
-+
-+

-+ These values may change in the future. They are provided only for -+ human consumption. -+

-+ -+ -+ -+ The amount of time since this client last successfully connected -+ to the database (in seconds). Value is empty if client has never -+ successfully been connected. -+ -+ -+ -+ The amount of time since this client last disconnected from the -+ database (in seconds). Value is empty if client has never -+ disconnected. -+ -+ -+ -+ Space-separated list of the names of OVSDB locks that the connection -+ holds. Omitted if the connection does not hold any locks. -+ -+ -+ -+ Space-separated list of the names of OVSDB locks that the connection is -+ currently waiting to acquire. Omitted if the connection is not waiting -+ for any locks. -+ -+ -+ -+ Space-separated list of the names of OVSDB locks that the connection -+ has had stolen by another OVSDB client. Omitted if no locks have been -+ stolen from this connection. -+ -+ -+ -+ When specifies a connection method that -+ listens for inbound connections (e.g. ptcp: or -+ pssl:) and more than one connection is actually active, -+ the value is the number of active connections. Otherwise, this -+ key-value pair is omitted. -+ -+ -+ -+ When is ptcp: or -+ pssl:, this is the TCP port on which the OVSDB server is -+ listening. (This is particularly useful when specifies a port of 0, allowing the kernel to -+ choose any available port.) -+ -+ -+ -+ -+ -+ The Differentiated Service Code Point (DSCP) is specified using 6 bits -+ in the Type of Service (TOS) field in the IP header. DSCP provides a -+ mechanism to classify the network traffic and provide Quality of -+ Service (QoS) on IP networks. -+ -+ The DSCP value specified here is used when establishing the connection -+ between the manager and the Open vSwitch. If no value is specified, a -+ default value of 48 is chosen. Valid DSCP values must be in the range -+ 0 to 63. -+ -+ ++static enum rte_flow_item_type ++mlx5_ethertype_to_item_type(rte_be16_t type_spec, ++ rte_be16_t type_mask, bool is_tunnel) ++{ ++ enum rte_flow_item_type type; + -+ -+ External configuration options ++ switch (rte_be_to_cpu_16(type_spec & type_mask)) { ++ case 0: ++ type = RTE_FLOW_ITEM_TYPE_VOID; ++ break; ++ case RTE_ETHER_TYPE_TEB: ++ type = is_tunnel ? ++ RTE_FLOW_ITEM_TYPE_ETH : RTE_FLOW_ITEM_TYPE_END; ++ break; ++ case RTE_ETHER_TYPE_VLAN: ++ type = !is_tunnel ? ++ RTE_FLOW_ITEM_TYPE_VLAN : RTE_FLOW_ITEM_TYPE_END; ++ break; ++ case RTE_ETHER_TYPE_IPV4: ++ type = RTE_FLOW_ITEM_TYPE_IPV4; ++ break; ++ case RTE_ETHER_TYPE_IPV6: ++ type = RTE_FLOW_ITEM_TYPE_IPV6; ++ break; ++ default: ++ type = RTE_FLOW_ITEM_TYPE_END; ++ } ++ return type; ++} + -+ -+ External client-defined key-value pairs -+ -+ -+
-+ -diff --git a/ovsdb/monitor.c b/ovsdb/monitor.c -index 0f222cc992..952fa902e4 100644 ---- a/ovsdb/monitor.c -+++ b/ovsdb/monitor.c -@@ -638,14 +638,14 @@ ovsdb_monitor_change_set_destroy(struct ovsdb_monitor_change_set *mcs) + static enum rte_flow_item_type + mlx5_flow_expand_rss_item_complete(const struct rte_flow_item *item) { - ovs_list_remove(&mcs->list_node); +- enum rte_flow_item_type ret = RTE_FLOW_ITEM_TYPE_VOID; +- uint16_t ether_type = 0; +- uint16_t ether_type_m; +- uint8_t ip_next_proto = 0; +- uint8_t ip_next_proto_m; ++#define MLX5_XSET_ITEM_MASK_SPEC(type, fld) \ ++ do { \ ++ const void *m = item->mask; \ ++ const void *s = item->spec; \ ++ mask = m ? \ ++ ((const struct rte_flow_item_##type *)m)->fld : \ ++ rte_flow_item_##type##_mask.fld; \ ++ spec = ((const struct rte_flow_item_##type *)s)->fld; \ ++ } while (0) ++ ++ enum rte_flow_item_type ret; ++ uint16_t spec, mask; + + if (item == NULL || item->spec == NULL) +- return ret; ++ return RTE_FLOW_ITEM_TYPE_VOID; + switch (item->type) { + case RTE_FLOW_ITEM_TYPE_ETH: +- if (item->mask) +- ether_type_m = ((const struct rte_flow_item_eth *) +- (item->mask))->type; +- else +- ether_type_m = rte_flow_item_eth_mask.type; +- if (ether_type_m != RTE_BE16(0xFFFF)) +- break; +- ether_type = ((const struct rte_flow_item_eth *) +- (item->spec))->type; +- if (rte_be_to_cpu_16(ether_type) == RTE_ETHER_TYPE_IPV4) +- ret = RTE_FLOW_ITEM_TYPE_IPV4; +- else if (rte_be_to_cpu_16(ether_type) == RTE_ETHER_TYPE_IPV6) +- ret = RTE_FLOW_ITEM_TYPE_IPV6; +- else if (rte_be_to_cpu_16(ether_type) == RTE_ETHER_TYPE_VLAN) +- ret = RTE_FLOW_ITEM_TYPE_VLAN; +- else +- ret = RTE_FLOW_ITEM_TYPE_END; ++ MLX5_XSET_ITEM_MASK_SPEC(eth, type); ++ if (!mask) ++ return RTE_FLOW_ITEM_TYPE_VOID; ++ ret = mlx5_ethertype_to_item_type(spec, mask, false); + break; + case RTE_FLOW_ITEM_TYPE_VLAN: +- if (item->mask) +- ether_type_m = ((const struct rte_flow_item_vlan *) +- (item->mask))->inner_type; +- else +- ether_type_m = rte_flow_item_vlan_mask.inner_type; +- if (ether_type_m != RTE_BE16(0xFFFF)) +- break; +- ether_type = ((const struct rte_flow_item_vlan *) +- (item->spec))->inner_type; +- if (rte_be_to_cpu_16(ether_type) == RTE_ETHER_TYPE_IPV4) +- ret = RTE_FLOW_ITEM_TYPE_IPV4; +- else if (rte_be_to_cpu_16(ether_type) == RTE_ETHER_TYPE_IPV6) +- ret = RTE_FLOW_ITEM_TYPE_IPV6; +- else if (rte_be_to_cpu_16(ether_type) == RTE_ETHER_TYPE_VLAN) +- ret = RTE_FLOW_ITEM_TYPE_VLAN; +- else +- ret = RTE_FLOW_ITEM_TYPE_END; ++ MLX5_XSET_ITEM_MASK_SPEC(vlan, inner_type); ++ if (!mask) ++ return RTE_FLOW_ITEM_TYPE_VOID; ++ ret = mlx5_ethertype_to_item_type(spec, mask, false); + break; + case RTE_FLOW_ITEM_TYPE_IPV4: +- if (item->mask) +- ip_next_proto_m = ((const struct rte_flow_item_ipv4 *) +- (item->mask))->hdr.next_proto_id; +- else +- ip_next_proto_m = +- rte_flow_item_ipv4_mask.hdr.next_proto_id; +- if (ip_next_proto_m != 0xFF) +- break; +- ip_next_proto = ((const struct rte_flow_item_ipv4 *) +- (item->spec))->hdr.next_proto_id; +- if (ip_next_proto == IPPROTO_UDP) +- ret = RTE_FLOW_ITEM_TYPE_UDP; +- else if (ip_next_proto == IPPROTO_TCP) +- ret = RTE_FLOW_ITEM_TYPE_TCP; +- else if (ip_next_proto == IPPROTO_IP) +- ret = RTE_FLOW_ITEM_TYPE_IPV4; +- else if (ip_next_proto == IPPROTO_IPV6) +- ret = RTE_FLOW_ITEM_TYPE_IPV6; +- else +- ret = RTE_FLOW_ITEM_TYPE_END; ++ MLX5_XSET_ITEM_MASK_SPEC(ipv4, hdr.next_proto_id); ++ if (!mask) ++ return RTE_FLOW_ITEM_TYPE_VOID; ++ ret = mlx5_inet_proto_to_item_type(spec, mask); + break; + case RTE_FLOW_ITEM_TYPE_IPV6: +- if (item->mask) +- ip_next_proto_m = ((const struct rte_flow_item_ipv6 *) +- (item->mask))->hdr.proto; +- else +- ip_next_proto_m = +- rte_flow_item_ipv6_mask.hdr.proto; +- if (ip_next_proto_m != 0xFF) +- break; +- ip_next_proto = ((const struct rte_flow_item_ipv6 *) +- (item->spec))->hdr.proto; +- if (ip_next_proto == IPPROTO_UDP) +- ret = RTE_FLOW_ITEM_TYPE_UDP; +- else if (ip_next_proto == IPPROTO_TCP) +- ret = RTE_FLOW_ITEM_TYPE_TCP; +- else if (ip_next_proto == IPPROTO_IP) +- ret = RTE_FLOW_ITEM_TYPE_IPV4; +- else if (ip_next_proto == IPPROTO_IPV6) +- ret = RTE_FLOW_ITEM_TYPE_IPV6; +- else +- ret = RTE_FLOW_ITEM_TYPE_END; ++ MLX5_XSET_ITEM_MASK_SPEC(ipv6, hdr.proto); ++ if (!mask) ++ return RTE_FLOW_ITEM_TYPE_VOID; ++ ret = mlx5_inet_proto_to_item_type(spec, mask); + break; + case RTE_FLOW_ITEM_TYPE_GENEVE: +- ether_type_m = item->mask ? +- ((const struct rte_flow_item_geneve *) +- (item->mask))->protocol : +- rte_flow_item_geneve_mask.protocol; +- ether_type = ((const struct rte_flow_item_geneve *) +- (item->spec))->protocol; +- ether_type_m = rte_be_to_cpu_16(ether_type_m); +- ether_type = rte_be_to_cpu_16(ether_type); +- switch (ether_type_m & ether_type) { +- case RTE_ETHER_TYPE_TEB: +- ret = RTE_FLOW_ITEM_TYPE_ETH; +- break; +- case RTE_ETHER_TYPE_IPV4: +- ret = RTE_FLOW_ITEM_TYPE_IPV4; +- break; +- case RTE_ETHER_TYPE_IPV6: +- ret = RTE_FLOW_ITEM_TYPE_IPV6; +- break; +- default: +- ret = RTE_FLOW_ITEM_TYPE_END; +- } ++ MLX5_XSET_ITEM_MASK_SPEC(geneve, protocol); ++ ret = mlx5_ethertype_to_item_type(spec, mask, true); ++ break; ++ case RTE_FLOW_ITEM_TYPE_GRE: ++ MLX5_XSET_ITEM_MASK_SPEC(gre, protocol); ++ ret = mlx5_ethertype_to_item_type(spec, mask, true); ++ break; ++ case RTE_FLOW_ITEM_TYPE_VXLAN_GPE: ++ MLX5_XSET_ITEM_MASK_SPEC(vxlan_gpe, protocol); ++ ret = mlx5_nsh_proto_to_item_type(spec, mask); + break; + default: + ret = RTE_FLOW_ITEM_TYPE_VOID; + break; + } + return ret; ++#undef MLX5_XSET_ITEM_MASK_SPEC + } -- struct ovsdb_monitor_change_set_for_table *mcst, *next_mcst; -- LIST_FOR_EACH_SAFE (mcst, next_mcst, list_in_change_set, -+ struct ovsdb_monitor_change_set_for_table *mcst; -+ LIST_FOR_EACH_SAFE (mcst, list_in_change_set, - &mcs->change_set_for_tables) { - ovs_list_remove(&mcst->list_in_change_set); - ovs_list_remove(&mcst->list_in_mt); + static const int * +@@ -533,9 +560,11 @@ enum mlx5_expansion { + MLX5_EXPANSION_OUTER_IPV4, + MLX5_EXPANSION_OUTER_IPV4_UDP, + MLX5_EXPANSION_OUTER_IPV4_TCP, ++ MLX5_EXPANSION_OUTER_IPV4_ICMP, + MLX5_EXPANSION_OUTER_IPV6, + MLX5_EXPANSION_OUTER_IPV6_UDP, + MLX5_EXPANSION_OUTER_IPV6_TCP, ++ MLX5_EXPANSION_OUTER_IPV6_ICMP6, + MLX5_EXPANSION_VXLAN, + MLX5_EXPANSION_STD_VXLAN, + MLX5_EXPANSION_L3_VXLAN, +@@ -549,9 +578,11 @@ enum mlx5_expansion { + MLX5_EXPANSION_IPV4, + MLX5_EXPANSION_IPV4_UDP, + MLX5_EXPANSION_IPV4_TCP, ++ MLX5_EXPANSION_IPV4_ICMP, + MLX5_EXPANSION_IPV6, + MLX5_EXPANSION_IPV6_UDP, + MLX5_EXPANSION_IPV6_TCP, ++ MLX5_EXPANSION_IPV6_ICMP6, + MLX5_EXPANSION_IPV6_FRAG_EXT, + MLX5_EXPANSION_GTP, + MLX5_EXPANSION_GENEVE, +@@ -586,6 +617,7 @@ static const struct mlx5_flow_expand_node mlx5_support_expansion[] = { + .next = MLX5_FLOW_EXPAND_RSS_NEXT + (MLX5_EXPANSION_OUTER_IPV4_UDP, + MLX5_EXPANSION_OUTER_IPV4_TCP, ++ MLX5_EXPANSION_OUTER_IPV4_ICMP, + MLX5_EXPANSION_GRE, + MLX5_EXPANSION_NVGRE, + MLX5_EXPANSION_IPV4, +@@ -607,10 +639,14 @@ static const struct mlx5_flow_expand_node mlx5_support_expansion[] = { + .type = RTE_FLOW_ITEM_TYPE_TCP, + .rss_types = RTE_ETH_RSS_NONFRAG_IPV4_TCP, + }, ++ [MLX5_EXPANSION_OUTER_IPV4_ICMP] = { ++ .type = RTE_FLOW_ITEM_TYPE_ICMP, ++ }, + [MLX5_EXPANSION_OUTER_IPV6] = { + .next = MLX5_FLOW_EXPAND_RSS_NEXT + (MLX5_EXPANSION_OUTER_IPV6_UDP, + MLX5_EXPANSION_OUTER_IPV6_TCP, ++ MLX5_EXPANSION_OUTER_IPV6_ICMP6, + MLX5_EXPANSION_IPV4, + MLX5_EXPANSION_IPV6, + MLX5_EXPANSION_GRE, +@@ -632,6 +668,9 @@ static const struct mlx5_flow_expand_node mlx5_support_expansion[] = { + .type = RTE_FLOW_ITEM_TYPE_TCP, + .rss_types = RTE_ETH_RSS_NONFRAG_IPV6_TCP, + }, ++ [MLX5_EXPANSION_OUTER_IPV6_ICMP6] = { ++ .type = RTE_FLOW_ITEM_TYPE_ICMP6, ++ }, + [MLX5_EXPANSION_VXLAN] = { + .next = MLX5_FLOW_EXPAND_RSS_NEXT(MLX5_EXPANSION_ETH, + MLX5_EXPANSION_IPV4, +@@ -691,7 +730,8 @@ static const struct mlx5_flow_expand_node mlx5_support_expansion[] = { + }, + [MLX5_EXPANSION_IPV4] = { + .next = MLX5_FLOW_EXPAND_RSS_NEXT(MLX5_EXPANSION_IPV4_UDP, +- MLX5_EXPANSION_IPV4_TCP), ++ MLX5_EXPANSION_IPV4_TCP, ++ MLX5_EXPANSION_IPV4_ICMP), + .type = RTE_FLOW_ITEM_TYPE_IPV4, + .rss_types = RTE_ETH_RSS_IPV4 | RTE_ETH_RSS_FRAG_IPV4 | + RTE_ETH_RSS_NONFRAG_IPV4_OTHER, +@@ -704,9 +744,13 @@ static const struct mlx5_flow_expand_node mlx5_support_expansion[] = { + .type = RTE_FLOW_ITEM_TYPE_TCP, + .rss_types = RTE_ETH_RSS_NONFRAG_IPV4_TCP, + }, ++ [MLX5_EXPANSION_IPV4_ICMP] = { ++ .type = RTE_FLOW_ITEM_TYPE_ICMP, ++ }, + [MLX5_EXPANSION_IPV6] = { + .next = MLX5_FLOW_EXPAND_RSS_NEXT(MLX5_EXPANSION_IPV6_UDP, + MLX5_EXPANSION_IPV6_TCP, ++ MLX5_EXPANSION_IPV6_ICMP6, + MLX5_EXPANSION_IPV6_FRAG_EXT), + .type = RTE_FLOW_ITEM_TYPE_IPV6, + .rss_types = RTE_ETH_RSS_IPV6 | RTE_ETH_RSS_FRAG_IPV6 | +@@ -723,6 +767,9 @@ static const struct mlx5_flow_expand_node mlx5_support_expansion[] = { + [MLX5_EXPANSION_IPV6_FRAG_EXT] = { + .type = RTE_FLOW_ITEM_TYPE_IPV6_FRAG_EXT, + }, ++ [MLX5_EXPANSION_IPV6_ICMP6] = { ++ .type = RTE_FLOW_ITEM_TYPE_ICMP6, ++ }, + [MLX5_EXPANSION_GTP] = { + .next = MLX5_FLOW_EXPAND_RSS_NEXT(MLX5_EXPANSION_IPV4, + MLX5_EXPANSION_IPV6), +@@ -1206,7 +1253,7 @@ flow_rxq_tunnel_ptype_update(struct mlx5_rxq_ctrl *rxq_ctrl) + } -- struct ovsdb_monitor_row *row, *next; -- HMAP_FOR_EACH_SAFE (row, next, hmap_node, &mcst->rows) { -+ struct ovsdb_monitor_row *row; -+ HMAP_FOR_EACH_SAFE (row, hmap_node, &mcst->rows) { - hmap_remove(&mcst->rows, &row->hmap_node); - ovsdb_monitor_row_destroy(mcst->mt, row, mcst->n_columns); - } -@@ -700,13 +700,13 @@ void - ovsdb_monitor_session_condition_destroy( - struct ovsdb_monitor_session_condition *condition) + /** +- * Set the Rx queue flags (Mark/Flag and Tunnel Ptypes) according to the devive ++ * Set the Rx queue flags (Mark/Flag and Tunnel Ptypes) according to the device + * flow. + * + * @param[in] dev +@@ -1219,7 +1266,6 @@ flow_drv_rxq_flags_set(struct rte_eth_dev *dev, + struct mlx5_flow_handle *dev_handle) { -- struct shash_node *node, *next; -+ struct shash_node *node; - - if (!condition) { - return; - } - -- SHASH_FOR_EACH_SAFE (node, next, &condition->tables) { -+ SHASH_FOR_EACH_SAFE (node, &condition->tables) { - struct ovsdb_monitor_table_condition *mtc = node->data; + struct mlx5_priv *priv = dev->data->dev_private; +- const int mark = dev_handle->mark; + const int tunnel = !!(dev_handle->layers & MLX5_FLOW_LAYER_TUNNEL); + struct mlx5_ind_table_obj *ind_tbl = NULL; + unsigned int i; +@@ -1254,15 +1300,6 @@ flow_drv_rxq_flags_set(struct rte_eth_dev *dev, + * this must be always enabled (metadata may arive + * from other port - not from local flows only. + */ +- if (priv->config.dv_flow_en && +- priv->config.dv_xmeta_en != MLX5_XMETA_MODE_LEGACY && +- mlx5_flow_ext_mreg_supported(dev)) { +- rxq_ctrl->rxq.mark = 1; +- rxq_ctrl->flow_mark_n = 1; +- } else if (mark) { +- rxq_ctrl->rxq.mark = 1; +- rxq_ctrl->flow_mark_n++; +- } + if (tunnel) { + unsigned int j; + +@@ -1280,6 +1317,20 @@ flow_drv_rxq_flags_set(struct rte_eth_dev *dev, + } + } - ovsdb_condition_destroy(&mtc->new_condition); -@@ -1122,11 +1122,11 @@ ovsdb_monitor_compose_update( - json = NULL; - struct ovsdb_monitor_change_set_for_table *mcst; - LIST_FOR_EACH (mcst, list_in_change_set, &mcs->change_set_for_tables) { -- struct ovsdb_monitor_row *row, *next; -+ struct ovsdb_monitor_row *row; - struct json *table_json = NULL; - struct ovsdb_monitor_table *mt = mcst->mt; ++static void ++flow_rxq_mark_flag_set(struct rte_eth_dev *dev) ++{ ++ struct mlx5_priv *priv = dev->data->dev_private; ++ struct mlx5_rxq_ctrl *rxq_ctrl; ++ ++ if (priv->mark_enabled) ++ return; ++ LIST_FOREACH(rxq_ctrl, &priv->rxqsctrl, next) { ++ rxq_ctrl->rxq.mark = 1; ++ } ++ priv->mark_enabled = 1; ++} ++ + /** + * Set the Rx queue flags (Mark/Flag and Tunnel Ptypes) for a flow + * +@@ -1294,7 +1345,11 @@ flow_rxq_flags_set(struct rte_eth_dev *dev, struct rte_flow *flow) + struct mlx5_priv *priv = dev->data->dev_private; + uint32_t handle_idx; + struct mlx5_flow_handle *dev_handle; ++ struct mlx5_flow_workspace *wks = mlx5_flow_get_thread_workspace(); + ++ MLX5_ASSERT(wks); ++ if (wks->mark) ++ flow_rxq_mark_flag_set(dev); + SILIST_FOREACH(priv->sh->ipool[MLX5_IPOOL_MLX5_FLOW], flow->dev_handles, + handle_idx, dev_handle, next) + flow_drv_rxq_flags_set(dev, dev_handle); +@@ -1314,7 +1369,6 @@ flow_drv_rxq_flags_trim(struct rte_eth_dev *dev, + struct mlx5_flow_handle *dev_handle) + { + struct mlx5_priv *priv = dev->data->dev_private; +- const int mark = dev_handle->mark; + const int tunnel = !!(dev_handle->layers & MLX5_FLOW_LAYER_TUNNEL); + struct mlx5_ind_table_obj *ind_tbl = NULL; + unsigned int i; +@@ -1345,15 +1399,6 @@ flow_drv_rxq_flags_trim(struct rte_eth_dev *dev, + MLX5_ASSERT(rxq_ctrl != NULL); + if (rxq_ctrl == NULL) + continue; +- if (priv->config.dv_flow_en && +- priv->config.dv_xmeta_en != MLX5_XMETA_MODE_LEGACY && +- mlx5_flow_ext_mreg_supported(dev)) { +- rxq_ctrl->rxq.mark = 1; +- rxq_ctrl->flow_mark_n = 1; +- } else if (mark) { +- rxq_ctrl->flow_mark_n--; +- rxq_ctrl->rxq.mark = !!rxq_ctrl->flow_mark_n; +- } + if (tunnel) { + unsigned int j; + +@@ -1410,12 +1455,12 @@ flow_rxq_flags_clear(struct rte_eth_dev *dev) + + if (rxq == NULL || rxq->ctrl == NULL) + continue; +- rxq->ctrl->flow_mark_n = 0; + rxq->ctrl->rxq.mark = 0; + for (j = 0; j != MLX5_FLOW_TUNNEL; ++j) + rxq->ctrl->flow_tunnels_n[j] = 0; + rxq->ctrl->rxq.tunnel = 0; + } ++ priv->mark_enabled = 0; + } -- HMAP_FOR_EACH_SAFE (row, next, hmap_node, &mcst->rows) { -+ HMAP_FOR_EACH_SAFE (row, hmap_node, &mcst->rows) { - struct json *row_json; - row_json = (*row_update)(mt, condition, OVSDB_MONITOR_ROW, row, - initial, changed, mcst->n_columns); -@@ -1711,8 +1711,8 @@ ovsdb_monitor_destroy(struct ovsdb_monitor *dbmon) - ovsdb_monitor_json_cache_flush(dbmon); - hmap_destroy(&dbmon->json_cache); + /** +@@ -3008,7 +3053,7 @@ mlx5_flow_validate_item_geneve_opt(const struct rte_flow_item *item, + if ((uint32_t)spec->option_len > MLX5_GENEVE_OPTLEN_MASK) + return rte_flow_error_set + (error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item, +- "Geneve TLV opt length exceeeds the limit (31)"); ++ "Geneve TLV opt length exceeds the limit (31)"); + /* Check if class type and length masks are full. */ + if (full_mask.option_class != mask->option_class || + full_mask.option_type != mask->option_type || +@@ -3957,7 +4002,7 @@ find_graph_root(uint32_t rss_level) + * subflow. + * + * @param[in] dev_flow +- * Pointer the created preifx subflow. ++ * Pointer the created prefix subflow. + * + * @return + * The layers get from prefix subflow. +@@ -4284,7 +4329,7 @@ flow_dv_mreg_create_cb(void *tool_ctx, void *cb_ctx) + [3] = { .type = RTE_FLOW_ACTION_TYPE_END, }, + }; + +- /* Fill the register fileds in the flow. */ ++ /* Fill the register fields in the flow. */ + ret = mlx5_flow_get_reg_id(dev, MLX5_FLOW_MARK, 0, error); + if (ret < 0) + return NULL; +@@ -4353,7 +4398,7 @@ flow_dv_mreg_create_cb(void *tool_ctx, void *cb_ctx) + /* + * The copy Flows are not included in any list. There + * ones are referenced from other Flows and can not +- * be applied, removed, deleted in ardbitrary order ++ * be applied, removed, deleted in arbitrary order + * by list traversing. + */ + mcp_res->rix_flow = flow_list_create(dev, MLX5_FLOW_TYPE_MCP, +@@ -4796,6 +4841,7 @@ flow_create_split_inner(struct rte_eth_dev *dev, + struct rte_flow_error *error) + { + struct mlx5_flow *dev_flow; ++ struct mlx5_flow_workspace *wks = mlx5_flow_get_thread_workspace(); + + dev_flow = flow_drv_prepare(dev, flow, attr, items, actions, + flow_split_info->flow_idx, error); +@@ -4810,12 +4856,14 @@ flow_create_split_inner(struct rte_eth_dev *dev, + /* + * If dev_flow is as one of the suffix flow, some actions in suffix + * flow may need some user defined item layer flags, and pass the +- * Metadate rxq mark flag to suffix flow as well. ++ * Metadata rxq mark flag to suffix flow as well. + */ + if (flow_split_info->prefix_layers) + dev_flow->handle->layers = flow_split_info->prefix_layers; +- if (flow_split_info->prefix_mark) +- dev_flow->handle->mark = 1; ++ if (flow_split_info->prefix_mark) { ++ MLX5_ASSERT(wks); ++ wks->mark = 1; ++ } + if (sub_flow) + *sub_flow = dev_flow; + #ifdef HAVE_IBV_FLOW_DV_SUPPORT +@@ -5006,9 +5054,10 @@ flow_meter_split_prep(struct rte_eth_dev *dev, + uint32_t tag_id = 0; + struct rte_flow_item *vlan_item_dst = NULL; + const struct rte_flow_item *vlan_item_src = NULL; ++ const struct rte_flow_item *orig_items = items; + struct rte_flow_action *hw_mtr_action; + struct rte_flow_action *action_pre_head = NULL; +- int32_t flow_src_port = priv->representor_id; ++ uint16_t flow_src_port = priv->representor_id; + bool mtr_first; + uint8_t mtr_id_offset = priv->mtr_reg_share ? MLX5_MTR_COLOR_BITS : 0; + uint8_t mtr_reg_bits = priv->mtr_reg_share ? +@@ -5016,27 +5065,18 @@ flow_meter_split_prep(struct rte_eth_dev *dev, + uint32_t flow_id = 0; + uint32_t flow_id_reversed = 0; + uint8_t flow_id_bits = 0; ++ bool after_meter = false; + int shift; + + /* Prepare the suffix subflow items. */ + tag_item = sfx_items++; + for (; items->type != RTE_FLOW_ITEM_TYPE_END; items++) { +- struct mlx5_priv *port_priv; +- const struct rte_flow_item_port_id *pid_v; + int item_type = items->type; + + switch (item_type) { + case RTE_FLOW_ITEM_TYPE_PORT_ID: +- pid_v = items->spec; +- MLX5_ASSERT(pid_v); +- port_priv = mlx5_port_to_eswitch_info(pid_v->id, false); +- if (!port_priv) +- return rte_flow_error_set(error, +- rte_errno, +- RTE_FLOW_ERROR_TYPE_ITEM_SPEC, +- pid_v, +- "Failed to get port info."); +- flow_src_port = port_priv->representor_id; ++ if (mlx5_flow_get_item_vport_id(dev, items, &flow_src_port, error)) ++ return -rte_errno; + if (!fm->def_policy && wks->policy->is_hierarchy && + flow_src_port != priv->representor_id) { + if (flow_drv_mtr_hierarchy_rule_create(dev, +@@ -5082,6 +5122,7 @@ flow_meter_split_prep(struct rte_eth_dev *dev, + tag_action = actions_pre++; + action_cur = actions_pre++; + } ++ after_meter = true; + break; + case RTE_FLOW_ACTION_TYPE_VXLAN_DECAP: + case RTE_FLOW_ACTION_TYPE_NVGRE_DECAP: +@@ -5110,6 +5151,11 @@ flow_meter_split_prep(struct rte_eth_dev *dev, + MLX5_RTE_FLOW_ITEM_TYPE_VLAN; + } + break; ++ case RTE_FLOW_ACTION_TYPE_COUNT: ++ if (fm->def_policy) ++ action_cur = after_meter ? ++ actions_sfx++ : actions_pre++; ++ break; + default: + break; + } +@@ -5130,7 +5176,8 @@ flow_meter_split_prep(struct rte_eth_dev *dev, + + if (!fm->def_policy) { + sub_policy = get_meter_sub_policy(dev, flow, wks, +- attr, items, error); ++ attr, orig_items, ++ error); + if (!sub_policy) + return -rte_errno; + } else { +@@ -5359,7 +5406,7 @@ flow_mreg_split_qrss_prep(struct rte_eth_dev *dev, + * @param[out] error + * Perform verbose error reporting if not NULL. + * @param[in] encap_idx +- * The encap action inndex. ++ * The encap action index. + * + * @return + * 0 on success, negative value otherwise +@@ -5527,7 +5574,7 @@ flow_check_match_action(const struct rte_flow_action actions[], + return flag ? actions_n + 1 : 0; + } -- struct ovsdb_monitor_change_set *cs, *cs_next; -- LIST_FOR_EACH_SAFE (cs, cs_next, list_node, &dbmon->change_sets) { -+ struct ovsdb_monitor_change_set *cs; -+ LIST_FOR_EACH_SAFE (cs, list_node, &dbmon->change_sets) { - ovsdb_monitor_change_set_destroy(cs); - } +-#define SAMPLE_SUFFIX_ITEM 2 ++#define SAMPLE_SUFFIX_ITEM 3 -@@ -1760,14 +1760,14 @@ ovsdb_monitors_commit(struct ovsdb *db, const struct ovsdb_txn *txn) + /** + * Split the sample flow. +@@ -5568,6 +5615,7 @@ flow_check_match_action(const struct rte_flow_action actions[], + static int + flow_sample_split_prep(struct rte_eth_dev *dev, + int add_tag, ++ const struct rte_flow_item items[], + struct rte_flow_item sfx_items[], + const struct rte_flow_action actions[], + struct rte_flow_action actions_sfx[], +@@ -5584,8 +5632,9 @@ flow_sample_split_prep(struct rte_eth_dev *dev, + struct mlx5_rte_flow_item_tag *tag_mask; + struct rte_flow_action_jump *jump_action; + uint32_t tag_id = 0; +- int index; + int append_index = 0; ++ int set_tag_idx = -1; ++ int index; + int ret; + + if (sample_action_pos < 0) +@@ -5594,6 +5643,52 @@ flow_sample_split_prep(struct rte_eth_dev *dev, + NULL, "invalid position of sample " + "action in list"); + /* Prepare the actions for prefix and suffix flow. */ ++ if (add_tag) { ++ /* Update the new added tag action index preceding ++ * the PUSH_VLAN or ENCAP action. ++ */ ++ const struct rte_flow_action_raw_encap *raw_encap; ++ const struct rte_flow_action *action = actions; ++ int encap_idx; ++ int action_idx = 0; ++ int raw_decap_idx = -1; ++ int push_vlan_idx = -1; ++ for (; action->type != RTE_FLOW_ACTION_TYPE_END; action++) { ++ switch (action->type) { ++ case RTE_FLOW_ACTION_TYPE_RAW_DECAP: ++ raw_decap_idx = action_idx; ++ break; ++ case RTE_FLOW_ACTION_TYPE_RAW_ENCAP: ++ raw_encap = action->conf; ++ if (raw_encap->size > ++ MLX5_ENCAPSULATION_DECISION_SIZE) { ++ encap_idx = raw_decap_idx != -1 ? ++ raw_decap_idx : action_idx; ++ if (encap_idx < sample_action_pos && ++ push_vlan_idx == -1) ++ set_tag_idx = encap_idx; ++ } ++ break; ++ case RTE_FLOW_ACTION_TYPE_VXLAN_ENCAP: ++ case RTE_FLOW_ACTION_TYPE_NVGRE_ENCAP: ++ encap_idx = action_idx; ++ if (encap_idx < sample_action_pos && ++ push_vlan_idx == -1) ++ set_tag_idx = encap_idx; ++ break; ++ case RTE_FLOW_ACTION_TYPE_OF_PUSH_VLAN: ++ case RTE_FLOW_ACTION_TYPE_OF_SET_VLAN_VID: ++ push_vlan_idx = action_idx; ++ if (push_vlan_idx < sample_action_pos) ++ set_tag_idx = action_idx; ++ break; ++ default: ++ break; ++ } ++ action_idx++; ++ } ++ } ++ /* Prepare the actions for prefix and suffix flow. */ + if (qrss_action_pos >= 0 && qrss_action_pos < sample_action_pos) { + index = qrss_action_pos; + /* Put the preceding the Queue/RSS action into prefix flow. */ +@@ -5610,6 +5705,14 @@ flow_sample_split_prep(struct rte_eth_dev *dev, + memcpy(actions_sfx, actions + qrss_action_pos, + sizeof(struct rte_flow_action)); + actions_sfx++; ++ } else if (add_tag && set_tag_idx >= 0) { ++ if (set_tag_idx > 0) ++ memcpy(actions_pre, actions, ++ sizeof(struct rte_flow_action) * set_tag_idx); ++ memcpy(actions_pre + set_tag_idx + 1, actions + set_tag_idx, ++ sizeof(struct rte_flow_action) * ++ (sample_action_pos - set_tag_idx)); ++ index = sample_action_pos; + } else { + index = sample_action_pos; + if (index != 0) +@@ -5625,6 +5728,12 @@ flow_sample_split_prep(struct rte_eth_dev *dev, + append_index++; + set_tag = (void *)(actions_pre + actions_n + append_index); + ret = mlx5_flow_get_reg_id(dev, MLX5_SAMPLE_ID, 0, error); ++ /* Trust VF/SF on CX5 not supported meter so that the reserved ++ * metadata regC is REG_NON, back to use application tag ++ * index 0. ++ */ ++ if (unlikely(ret == REG_NON)) ++ ret = mlx5_flow_get_reg_id(dev, MLX5_APP_TAG, 0, error); + if (ret < 0) + return ret; + mlx5_ipool_malloc(priv->sh->ipool +@@ -5634,6 +5743,12 @@ flow_sample_split_prep(struct rte_eth_dev *dev, + .data = tag_id, + }; + /* Prepare the suffix subflow items. */ ++ for (; items->type != RTE_FLOW_ITEM_TYPE_END; items++) { ++ if (items->type == RTE_FLOW_ITEM_TYPE_PORT_ID) { ++ memcpy(sfx_items, items, sizeof(*sfx_items)); ++ sfx_items++; ++ } ++ } + tag_spec = (void *)(sfx_items + SAMPLE_SUFFIX_ITEM); + tag_spec->data = tag_id; + tag_spec->id = set_tag->id; +@@ -5651,13 +5766,17 @@ flow_sample_split_prep(struct rte_eth_dev *dev, + RTE_FLOW_ITEM_TYPE_END, + }; + /* Prepare the tag action in prefix subflow. */ +- actions_pre[index++] = ++ set_tag_idx = (set_tag_idx == -1) ? index : set_tag_idx; ++ actions_pre[set_tag_idx] = + (struct rte_flow_action){ + .type = (enum rte_flow_action_type) + MLX5_RTE_FLOW_ACTION_TYPE_TAG, + .conf = set_tag, + }; ++ /* Update next sample position due to add one tag action */ ++ index += 1; + } ++ /* Copy the sample action into prefix flow. */ + memcpy(actions_pre + index, actions + sample_action_pos, + sizeof(struct rte_flow_action)); + index += 1; +@@ -6042,6 +6161,8 @@ flow_create_split_meter(struct rte_eth_dev *dev, + fm->policy_id, + NULL); + MLX5_ASSERT(wks->policy); ++ if (wks->policy->mark) ++ wks->mark = 1; + if (wks->policy->is_hierarchy) { + wks->final_policy = + mlx5_flow_meter_hierarchy_get_final_policy(dev, +@@ -6065,8 +6186,10 @@ flow_create_split_meter(struct rte_eth_dev *dev, + if (!fm->def_policy && !is_mtr_hierarchy && + (!has_modify || !fm->drop_cnt)) + set_mtr_reg = false; +- /* Prefix actions: meter, decap, encap, tag, jump, end. */ +- act_size = sizeof(struct rte_flow_action) * (actions_n + 6) + ++ /* Prefix actions: meter, decap, encap, tag, jump, end, cnt. */ ++#define METER_PREFIX_ACTION 7 ++ act_size = (sizeof(struct rte_flow_action) * ++ (actions_n + METER_PREFIX_ACTION)) + + sizeof(struct mlx5_rte_flow_action_set_tag); + /* Suffix items: tag, vlan, port id, end. */ + #define METER_SUFFIX_ITEM 4 +@@ -6128,7 +6251,7 @@ flow_create_split_meter(struct rte_eth_dev *dev, + MLX5_FLOW_TABLE_LEVEL_METER; + flow_split_info->prefix_layers = + flow_get_prefix_layer_flags(dev_flow); +- flow_split_info->prefix_mark |= dev_flow->handle->mark; ++ flow_split_info->prefix_mark |= wks->mark; + flow_split_info->table_id = MLX5_MTR_TABLE_ID_SUFFIX; + } + /* Add the prefix subflow. */ +@@ -6194,6 +6317,7 @@ flow_create_split_sample(struct rte_eth_dev *dev, + struct mlx5_flow_dv_sample_resource *sample_res; + struct mlx5_flow_tbl_data_entry *sfx_tbl_data; + struct mlx5_flow_tbl_resource *sfx_tbl; ++ struct mlx5_flow_workspace *wks = mlx5_flow_get_thread_workspace(); + #endif + size_t act_size; + size_t item_size; +@@ -6240,7 +6364,7 @@ flow_create_split_sample(struct rte_eth_dev *dev, + jump_table = attr->group * MLX5_FLOW_TABLE_FACTOR + + next_ft_step; + pre_actions = sfx_actions + actions_n; +- tag_id = flow_sample_split_prep(dev, add_tag, sfx_items, ++ tag_id = flow_sample_split_prep(dev, add_tag, items, sfx_items, + actions, sfx_actions, + pre_actions, actions_n, + sample_action_pos, +@@ -6280,7 +6404,8 @@ flow_create_split_sample(struct rte_eth_dev *dev, + } + flow_split_info->prefix_layers = + flow_get_prefix_layer_flags(dev_flow); +- flow_split_info->prefix_mark |= dev_flow->handle->mark; ++ MLX5_ASSERT(wks); ++ flow_split_info->prefix_mark |= wks->mark; + /* Suffix group level already be scaled with factor, set + * MLX5_SCALE_FLOW_GROUP_BIT of skip_scale to 1 to avoid scale + * again in translation. +@@ -6884,7 +7009,7 @@ flow_list_destroy(struct rte_eth_dev *dev, enum mlx5_flow_type type, + * @param type + * Flow type to be flushed. + * @param active +- * If flushing is called avtively. ++ * If flushing is called actively. + */ void - ovsdb_monitors_remove(struct ovsdb *db) + mlx5_flow_list_flush(struct rte_eth_dev *dev, enum mlx5_flow_type type, +@@ -8531,7 +8656,7 @@ mlx5_flow_dev_dump_sh_all(struct rte_eth_dev *dev, + * Perform verbose error reporting if not NULL. PMDs initialize this + * structure in case of error only. + * @return +- * 0 on success, a nagative value otherwise. ++ * 0 on success, a negative value otherwise. + */ + int + mlx5_flow_dev_dump(struct rte_eth_dev *dev, struct rte_flow *flow_idx, +@@ -9009,7 +9134,7 @@ mlx5_get_tof(const struct rte_flow_item *item, + } + + /** +- * tunnel offload functionalilty is defined for DV environment only ++ * tunnel offload functionality is defined for DV environment only + */ + #ifdef HAVE_IBV_FLOW_DV_SUPPORT + __extension__ +@@ -9822,10 +9947,27 @@ mlx5_flow_flex_item_create(struct rte_eth_dev *dev, + struct rte_flow_error *error) { -- struct ovsdb_monitor *m, *next_m; -+ struct ovsdb_monitor *m; + static const char err_msg[] = "flex item creation unsupported"; ++ struct mlx5_priv *priv = dev->data->dev_private; + struct rte_flow_attr attr = { .transfer = 0 }; + const struct mlx5_flow_driver_ops *fops = + flow_get_drv_ops(flow_get_drv_type(dev, &attr)); + ++ if (!priv->pci_dev) { ++ rte_flow_error_set(error, ENOTSUP, ++ RTE_FLOW_ERROR_TYPE_UNSPECIFIED, NULL, ++ "create flex item on PF only"); ++ return NULL; ++ } ++ switch (priv->pci_dev->id.device_id) { ++ case PCI_DEVICE_ID_MELLANOX_CONNECTX6DXBF: ++ case PCI_DEVICE_ID_MELLANOX_CONNECTX7BF: ++ break; ++ default: ++ rte_flow_error_set(error, ENOTSUP, ++ RTE_FLOW_ERROR_TYPE_UNSPECIFIED, NULL, ++ "flex item available on BlueField ports only"); ++ return NULL; ++ } + if (!fops->item_create) { + DRV_LOG(ERR, "port %u %s.", dev->data->port_id, err_msg); + rte_flow_error_set(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ACTION, +@@ -10012,3 +10154,80 @@ mlx5_flow_adjust_priority(struct rte_eth_dev *dev, int32_t priority, + } + return res; + } ++ ++/** ++ * Get the E-Switch Manager vport id. ++ * ++ * @param[in] dev ++ * Pointer to the Ethernet device structure. ++ * ++ * @return ++ * The vport id. ++ */ ++int16_t mlx5_flow_get_esw_manager_vport_id(struct rte_eth_dev *dev) ++{ ++ struct mlx5_priv *priv = dev->data->dev_private; ++ struct mlx5_common_device *cdev = priv->sh->cdev; ++ ++ /* New FW exposes E-Switch Manager vport ID, can use it directly. */ ++ if (cdev->config.hca_attr.esw_mgr_vport_id_valid) ++ return (int16_t)cdev->config.hca_attr.esw_mgr_vport_id; ++ ++ if (priv->pci_dev == NULL) ++ return 0; ++ switch (priv->pci_dev->id.device_id) { ++ case PCI_DEVICE_ID_MELLANOX_CONNECTX5BF: ++ case PCI_DEVICE_ID_MELLANOX_CONNECTX6DXBF: ++ case PCI_DEVICE_ID_MELLANOX_CONNECTX7BF: ++ /* ++ * In old FW which doesn't expose the E-Switch Manager vport ID in the capability, ++ * only the BF embedded CPUs control the E-Switch Manager port. Hence, ++ * ECPF vport ID is selected and not the host port (0) in any BF case. ++ */ ++ return (int16_t)MLX5_ECPF_VPORT_ID; ++ default: ++ return MLX5_PF_VPORT_ID; ++ } ++} ++ ++/** ++ * Parse item to get the vport id. ++ * ++ * @param[in] dev ++ * Pointer to the Ethernet device structure. ++ * @param[in] item ++ * The src port id match item. ++ * @param[out] vport_id ++ * Pointer to put the vport id. ++ * @param[out] error ++ * Pointer to error structure. ++ * ++ * @return ++ * 0 on success, a negative errno value otherwise and rte_errno is set. ++ */ ++int mlx5_flow_get_item_vport_id(struct rte_eth_dev *dev, ++ const struct rte_flow_item *item, ++ uint16_t *vport_id, ++ struct rte_flow_error *error) ++{ ++ struct mlx5_priv *port_priv; ++ const struct rte_flow_item_port_id *pid_v; ++ ++ if (item->type != RTE_FLOW_ITEM_TYPE_PORT_ID) ++ return rte_flow_error_set(error, EINVAL, RTE_FLOW_ERROR_TYPE_ITEM_SPEC, ++ NULL, "Incorrect item type."); ++ pid_v = item->spec; ++ if (!pid_v) ++ return 0; ++ if (pid_v->id == MLX5_PORT_ESW_MGR) { ++ *vport_id = mlx5_flow_get_esw_manager_vport_id(dev); ++ } else { ++ port_priv = mlx5_port_to_eswitch_info(pid_v->id, false); ++ if (!port_priv) ++ return rte_flow_error_set(error, EINVAL, RTE_FLOW_ERROR_TYPE_ITEM_SPEC, ++ NULL, "Failed to get port info."); ++ *vport_id = port_priv->representor_id; ++ } ++ ++ return 0; ++} +diff --git a/dpdk/drivers/net/mlx5/mlx5_flow.h b/dpdk/drivers/net/mlx5/mlx5_flow.h +index 1f54649c69..29ccb98351 100644 +--- a/dpdk/drivers/net/mlx5/mlx5_flow.h ++++ b/dpdk/drivers/net/mlx5/mlx5_flow.h +@@ -426,7 +426,7 @@ enum mlx5_feature_name { + #define MLX5_ACT_NUM_MDF_IPV6 4 + #define MLX5_ACT_NUM_MDF_MAC 2 + #define MLX5_ACT_NUM_MDF_VID 1 +-#define MLX5_ACT_NUM_MDF_PORT 2 ++#define MLX5_ACT_NUM_MDF_PORT 1 + #define MLX5_ACT_NUM_MDF_TTL 1 + #define MLX5_ACT_NUM_DEC_TTL MLX5_ACT_NUM_MDF_TTL + #define MLX5_ACT_NUM_MDF_TCPSEQ 1 +@@ -598,7 +598,7 @@ struct mlx5_flow_tbl_data_entry { + const struct mlx5_flow_tunnel *tunnel; + uint32_t group_id; + uint32_t external:1; +- uint32_t tunnel_offload:1; /* Tunnel offlod table or not. */ ++ uint32_t tunnel_offload:1; /* Tunnel offload table or not. */ + uint32_t is_egress:1; /**< Egress table. */ + uint32_t is_transfer:1; /**< Transfer table. */ + uint32_t dummy:1; /**< DR table. */ +@@ -696,10 +696,8 @@ struct mlx5_flow_handle { + /**< Bit-fields of present layers, see MLX5_FLOW_LAYER_*. */ + void *drv_flow; /**< pointer to driver flow object. */ + uint32_t split_flow_id:27; /**< Sub flow unique match flow id. */ +- uint32_t is_meter_flow_id:1; /**< Indate if flow_id is for meter. */ +- uint32_t mark:1; /**< Metadate rxq mark flag. */ ++ uint32_t is_meter_flow_id:1; /**< Indicate if flow_id is for meter. */ + uint32_t fate_action:3; /**< Fate action type. */ +- uint32_t flex_item; /**< referenced Flex Item bitmask. */ + union { + uint32_t rix_hrxq; /**< Hash Rx queue object index. */ + uint32_t rix_jump; /**< Index to the jump action resource. */ +@@ -715,6 +713,7 @@ struct mlx5_flow_handle { + #if defined(HAVE_IBV_FLOW_DV_SUPPORT) || !defined(HAVE_INFINIBAND_VERBS_H) + struct mlx5_flow_handle_dv dvh; + #endif ++ uint8_t flex_item; /**< referenced Flex Item bitmask. */ + } __rte_packed; + + /* +@@ -1108,6 +1107,7 @@ struct mlx5_flow_workspace { + /* The final policy when meter policy is hierarchy. */ + uint32_t skip_matcher_reg:1; + /* Indicates if need to skip matcher register in translate. */ ++ uint32_t mark:1; /* Indicates if flow contains mark action. */ + }; -- LIST_FOR_EACH_SAFE (m, next_m, list_node, &db->monitors) { -- struct jsonrpc_monitor_node *jm, *next_jm; -+ LIST_FOR_EACH_SAFE (m, list_node, &db->monitors) { -+ struct jsonrpc_monitor_node *jm; + struct mlx5_flow_split_info { +@@ -1450,6 +1450,20 @@ flow_aso_ct_get_by_idx(struct rte_eth_dev *dev, uint32_t own_idx) + return ct; + } - /* Delete all front-end monitors. Removing the last front-end monitor - * will also destroy the corresponding ovsdb_monitor. */ -- LIST_FOR_EACH_SAFE (jm, next_jm, node, &m->jsonrpc_monitors) { -+ LIST_FOR_EACH_SAFE (jm, node, &m->jsonrpc_monitors) { - ovsdb_jsonrpc_monitor_destroy(jm->jsonrpc_monitor, false); - } - } -@@ -1789,14 +1789,14 @@ ovsdb_monitor_get_memory_usage(struct simap *usage) - void - ovsdb_monitor_prereplace_db(struct ovsdb *db) ++static inline uint16_t ++mlx5_translate_tunnel_etypes(uint64_t pattern_flags) ++{ ++ if (pattern_flags & MLX5_FLOW_LAYER_INNER_L2) ++ return RTE_ETHER_TYPE_TEB; ++ else if (pattern_flags & MLX5_FLOW_LAYER_INNER_L3_IPV4) ++ return RTE_ETHER_TYPE_IPV4; ++ else if (pattern_flags & MLX5_FLOW_LAYER_INNER_L3_IPV6) ++ return RTE_ETHER_TYPE_IPV6; ++ else if (pattern_flags & MLX5_FLOW_LAYER_MPLS) ++ return RTE_ETHER_TYPE_MPLS; ++ return 0; ++} ++ + int mlx5_flow_group_to_table(struct rte_eth_dev *dev, + const struct mlx5_flow_tunnel *tunnel, + uint32_t group, uint32_t *table, +@@ -1752,4 +1766,14 @@ const struct mlx5_flow_tunnel * + mlx5_get_tof(const struct rte_flow_item *items, + const struct rte_flow_action *actions, + enum mlx5_tof_rule_type *rule_type); ++ ++#define MLX5_PF_VPORT_ID 0 ++#define MLX5_ECPF_VPORT_ID 0xFFFE ++ ++int16_t mlx5_flow_get_esw_manager_vport_id(struct rte_eth_dev *dev); ++int mlx5_flow_get_item_vport_id(struct rte_eth_dev *dev, ++ const struct rte_flow_item *item, ++ uint16_t *vport_id, ++ struct rte_flow_error *error); ++ + #endif /* RTE_PMD_MLX5_FLOW_H_ */ +diff --git a/dpdk/drivers/net/mlx5/mlx5_flow_aso.c b/dpdk/drivers/net/mlx5/mlx5_flow_aso.c +index ddf4328dec..eb7fc43da3 100644 +--- a/dpdk/drivers/net/mlx5/mlx5_flow_aso.c ++++ b/dpdk/drivers/net/mlx5/mlx5_flow_aso.c +@@ -13,50 +13,6 @@ + #include "mlx5.h" + #include "mlx5_flow.h" + +-/** +- * Destroy Completion Queue used for ASO access. +- * +- * @param[in] cq +- * ASO CQ to destroy. +- */ +-static void +-mlx5_aso_cq_destroy(struct mlx5_aso_cq *cq) +-{ +- if (cq->cq_obj.cq) +- mlx5_devx_cq_destroy(&cq->cq_obj); +- memset(cq, 0, sizeof(*cq)); +-} +- +-/** +- * Create Completion Queue used for ASO access. +- * +- * @param[in] ctx +- * Context returned from mlx5 open_device() glue function. +- * @param[in/out] cq +- * Pointer to CQ to create. +- * @param[in] log_desc_n +- * Log of number of descriptors in queue. +- * @param[in] socket +- * Socket to use for allocation. +- * @param[in] uar_page_id +- * UAR page ID to use. +- * +- * @return +- * 0 on success, a negative errno value otherwise and rte_errno is set. +- */ +-static int +-mlx5_aso_cq_create(void *ctx, struct mlx5_aso_cq *cq, uint16_t log_desc_n, +- int socket, int uar_page_id) +-{ +- struct mlx5_devx_cq_attr attr = { +- .uar_page_id = uar_page_id, +- }; +- +- cq->log_desc_n = log_desc_n; +- cq->cq_ci = 0; +- return mlx5_devx_cq_create(ctx, &cq->cq_obj, log_desc_n, &attr, socket); +-} +- + /** + * Free MR resources. + * +@@ -84,21 +40,18 @@ mlx5_aso_dereg_mr(struct mlx5_common_device *cdev, struct mlx5_pmd_mr *mr) + * Size of MR buffer. + * @param[in/out] mr + * Pointer to MR to create. +- * @param[in] socket +- * Socket to use for allocation. + * + * @return + * 0 on success, a negative errno value otherwise and rte_errno is set. + */ + static int + mlx5_aso_reg_mr(struct mlx5_common_device *cdev, size_t length, +- struct mlx5_pmd_mr *mr, int socket) ++ struct mlx5_pmd_mr *mr) { -- struct ovsdb_monitor *m, *next_m; -+ struct ovsdb_monitor *m; +- + int ret; + + mr->addr = mlx5_malloc(MLX5_MEM_RTE | MLX5_MEM_ZERO, length, 4096, +- socket); ++ SOCKET_ID_ANY); + if (!mr->addr) { + DRV_LOG(ERR, "Failed to create ASO bits mem for MR."); + return -1; +@@ -122,7 +75,7 @@ static void + mlx5_aso_destroy_sq(struct mlx5_aso_sq *sq) + { + mlx5_devx_sq_destroy(&sq->sq_obj); +- mlx5_aso_cq_destroy(&sq->cq); ++ mlx5_devx_cq_destroy(&sq->cq.cq_obj); + memset(sq, 0, sizeof(*sq)); + } -- LIST_FOR_EACH_SAFE (m, next_m, list_node, &db->monitors) { -- struct jsonrpc_monitor_node *jm, *next_jm; -+ LIST_FOR_EACH_SAFE (m, list_node, &db->monitors) { -+ struct jsonrpc_monitor_node *jm; +@@ -226,35 +179,31 @@ mlx5_aso_ct_init_sq(struct mlx5_aso_sq *sq) + /** + * Create Send Queue used for ASO access. + * +- * @param[in] ctx +- * Context returned from mlx5 open_device() glue function. ++ * @param[in] cdev ++ * Pointer to the mlx5 common device. + * @param[in/out] sq + * Pointer to SQ to create. +- * @param[in] socket +- * Socket to use for allocation. + * @param[in] uar + * User Access Region object. +- * @param[in] pdn +- * Protection Domain number to use. +- * @param[in] log_desc_n +- * Log of number of descriptors in queue. +- * @param[in] ts_format +- * timestamp format supported by the queue. + * + * @return + * 0 on success, a negative errno value otherwise and rte_errno is set. + */ + static int +-mlx5_aso_sq_create(void *ctx, struct mlx5_aso_sq *sq, int socket, void *uar, +- uint32_t pdn, uint16_t log_desc_n, uint32_t ts_format) ++mlx5_aso_sq_create(struct mlx5_common_device *cdev, struct mlx5_aso_sq *sq, ++ void *uar) + { +- struct mlx5_devx_create_sq_attr attr = { ++ struct mlx5_devx_cq_attr cq_attr = { ++ .uar_page_id = mlx5_os_get_devx_uar_page_id(uar), ++ }; ++ struct mlx5_devx_create_sq_attr sq_attr = { + .user_index = 0xFFFF, + .wq_attr = (struct mlx5_devx_wq_attr){ +- .pd = pdn, ++ .pd = cdev->pdn, + .uar_page = mlx5_os_get_devx_uar_page_id(uar), + }, +- .ts_format = mlx5_ts_format_conv(ts_format), ++ .ts_format = ++ mlx5_ts_format_conv(cdev->config.hca_attr.sq_ts_format), + }; + struct mlx5_devx_modify_sq_attr modify_attr = { + .state = MLX5_SQC_STATE_RDY, +@@ -262,14 +211,18 @@ mlx5_aso_sq_create(void *ctx, struct mlx5_aso_sq *sq, int socket, void *uar, + uint16_t log_wqbb_n; + int ret; + +- if (mlx5_aso_cq_create(ctx, &sq->cq, log_desc_n, socket, +- mlx5_os_get_devx_uar_page_id(uar))) ++ if (mlx5_devx_cq_create(cdev->ctx, &sq->cq.cq_obj, ++ MLX5_ASO_QUEUE_LOG_DESC, &cq_attr, ++ SOCKET_ID_ANY)) + goto error; +- sq->log_desc_n = log_desc_n; +- attr.cqn = sq->cq.cq_obj.cq->id; ++ sq->cq.cq_ci = 0; ++ sq->cq.log_desc_n = MLX5_ASO_QUEUE_LOG_DESC; ++ sq->log_desc_n = MLX5_ASO_QUEUE_LOG_DESC; ++ sq_attr.cqn = sq->cq.cq_obj.cq->id; + /* for mlx5_aso_wqe that is twice the size of mlx5_wqe */ +- log_wqbb_n = log_desc_n + 1; +- ret = mlx5_devx_sq_create(ctx, &sq->sq_obj, log_wqbb_n, &attr, socket); ++ log_wqbb_n = sq->log_desc_n + 1; ++ ret = mlx5_devx_sq_create(cdev->ctx, &sq->sq_obj, log_wqbb_n, &sq_attr, ++ SOCKET_ID_ANY); + if (ret) { + DRV_LOG(ERR, "Can't create SQ object."); + rte_errno = ENOMEM; +@@ -313,34 +266,28 @@ mlx5_aso_queue_init(struct mlx5_dev_ctx_shared *sh, + switch (aso_opc_mod) { + case ASO_OPC_MOD_FLOW_HIT: + if (mlx5_aso_reg_mr(cdev, (MLX5_ASO_AGE_ACTIONS_PER_POOL / 8) * +- sq_desc_n, &sh->aso_age_mng->aso_sq.mr, 0)) ++ sq_desc_n, &sh->aso_age_mng->aso_sq.mr)) + return -1; +- if (mlx5_aso_sq_create(cdev->ctx, &sh->aso_age_mng->aso_sq, 0, +- sh->tx_uar.obj, cdev->pdn, +- MLX5_ASO_QUEUE_LOG_DESC, +- cdev->config.hca_attr.sq_ts_format)) { ++ if (mlx5_aso_sq_create(cdev, &sh->aso_age_mng->aso_sq, ++ sh->tx_uar.obj)) { + mlx5_aso_dereg_mr(cdev, &sh->aso_age_mng->aso_sq.mr); + return -1; + } + mlx5_aso_age_init_sq(&sh->aso_age_mng->aso_sq); + break; + case ASO_OPC_MOD_POLICER: +- if (mlx5_aso_sq_create(cdev->ctx, &sh->mtrmng->pools_mng.sq, 0, +- sh->tx_uar.obj, cdev->pdn, +- MLX5_ASO_QUEUE_LOG_DESC, +- cdev->config.hca_attr.sq_ts_format)) ++ if (mlx5_aso_sq_create(cdev, &sh->mtrmng->pools_mng.sq, ++ sh->tx_uar.obj)) + return -1; + mlx5_aso_mtr_init_sq(&sh->mtrmng->pools_mng.sq); + break; + case ASO_OPC_MOD_CONNECTION_TRACKING: + /* 64B per object for query. */ + if (mlx5_aso_reg_mr(cdev, 64 * sq_desc_n, +- &sh->ct_mng->aso_sq.mr, 0)) ++ &sh->ct_mng->aso_sq.mr)) + return -1; +- if (mlx5_aso_sq_create(cdev->ctx, &sh->ct_mng->aso_sq, 0, +- sh->tx_uar.obj, cdev->pdn, +- MLX5_ASO_QUEUE_LOG_DESC, +- cdev->config.hca_attr.sq_ts_format)) { ++ if (mlx5_aso_sq_create(cdev, &sh->ct_mng->aso_sq, ++ sh->tx_uar.obj)) { + mlx5_aso_dereg_mr(cdev, &sh->ct_mng->aso_sq.mr); + return -1; + } +diff --git a/dpdk/drivers/net/mlx5/mlx5_flow_dv.c b/dpdk/drivers/net/mlx5/mlx5_flow_dv.c +index 3da122cbb9..70031d3dc9 100644 +--- a/dpdk/drivers/net/mlx5/mlx5_flow_dv.c ++++ b/dpdk/drivers/net/mlx5/mlx5_flow_dv.c +@@ -93,37 +93,6 @@ static int + flow_dv_jump_tbl_resource_release(struct rte_eth_dev *dev, + uint32_t rix_jump); + +-static inline uint16_t +-mlx5_translate_tunnel_etypes(uint64_t pattern_flags) +-{ +- if (pattern_flags & MLX5_FLOW_LAYER_INNER_L2) +- return RTE_ETHER_TYPE_TEB; +- else if (pattern_flags & MLX5_FLOW_LAYER_INNER_L3_IPV4) +- return RTE_ETHER_TYPE_IPV4; +- else if (pattern_flags & MLX5_FLOW_LAYER_INNER_L3_IPV6) +- return RTE_ETHER_TYPE_IPV6; +- else if (pattern_flags & MLX5_FLOW_LAYER_MPLS) +- return RTE_ETHER_TYPE_MPLS; +- return 0; +-} +- +-static int16_t +-flow_dv_get_esw_manager_vport_id(struct rte_eth_dev *dev) +-{ +- struct mlx5_priv *priv = dev->data->dev_private; +- +- if (priv->pci_dev == NULL) +- return 0; +- switch (priv->pci_dev->id.device_id) { +- case PCI_DEVICE_ID_MELLANOX_CONNECTX5BF: +- case PCI_DEVICE_ID_MELLANOX_CONNECTX6DXBF: +- case PCI_DEVICE_ID_MELLANOX_CONNECTX7BF: +- return (int16_t)0xfffe; +- default: +- return 0; +- } +-} +- + /** + * Initialize flow attributes structure according to flow items' types. + * +@@ -172,6 +141,7 @@ flow_dv_attr_init(const struct rte_flow_item *item, union flow_dv_attr *attr, + case RTE_FLOW_ITEM_TYPE_VXLAN_GPE: + case RTE_FLOW_ITEM_TYPE_GENEVE: + case RTE_FLOW_ITEM_TYPE_MPLS: ++ case RTE_FLOW_ITEM_TYPE_GTP: + if (tunnel_decap) + attr->attr = 0; + break; +@@ -326,7 +296,8 @@ flow_dv_hlist_prepare(struct mlx5_dev_ctx_shared *sh, struct mlx5_hlist **phl, + mlx5_list_match_cb cb_match, + mlx5_list_remove_cb cb_remove, + mlx5_list_clone_cb cb_clone, +- mlx5_list_clone_free_cb cb_clone_free) ++ mlx5_list_clone_free_cb cb_clone_free, ++ struct rte_flow_error *error) + { + struct mlx5_hlist *hl; + struct mlx5_hlist *expected = NULL; +@@ -341,7 +312,9 @@ flow_dv_hlist_prepare(struct mlx5_dev_ctx_shared *sh, struct mlx5_hlist **phl, + cb_clone_free); + if (!hl) { + DRV_LOG(ERR, "%s hash creation failed", name); +- rte_errno = ENOMEM; ++ rte_flow_error_set(error, ENOMEM, ++ RTE_FLOW_ERROR_TYPE_UNSPECIFIED, NULL, ++ "cannot allocate resource memory"); + return NULL; + } + if (!__atomic_compare_exchange_n(phl, &expected, hl, false, +@@ -1503,7 +1476,7 @@ mlx5_flow_field_id_to_modify_info + if (data->offset < 16) + info[idx++] = (struct field_modify_info){2, 0, + MLX5_MODI_OUT_DMAC_15_0}; +- info[idx] = (struct field_modify_info){4, 0, ++ info[idx] = (struct field_modify_info){4, off, + MLX5_MODI_OUT_DMAC_47_16}; + } + break; +@@ -1533,7 +1506,7 @@ mlx5_flow_field_id_to_modify_info + if (data->offset < 16) + info[idx++] = (struct field_modify_info){2, 0, + MLX5_MODI_OUT_SMAC_15_0}; +- info[idx] = (struct field_modify_info){4, 0, ++ info[idx] = (struct field_modify_info){4, off, + MLX5_MODI_OUT_SMAC_47_16}; + } + break; +@@ -1881,7 +1854,7 @@ flow_dv_convert_action_modify_field + struct field_modify_info dcopy[MLX5_ACT_MAX_MOD_FIELDS] = { + {0, 0, 0} }; + uint32_t mask[MLX5_ACT_MAX_MOD_FIELDS] = {0, 0, 0, 0, 0}; +- uint32_t type; ++ uint32_t type, meta = 0; + uint32_t shift = 0; + + if (conf->src.field == RTE_FLOW_FIELD_POINTER || +@@ -1894,6 +1867,11 @@ flow_dv_convert_action_modify_field + item.spec = conf->src.field == RTE_FLOW_FIELD_POINTER ? + (void *)(uintptr_t)conf->src.pvalue : + (void *)(uintptr_t)&conf->src.value; ++ if (conf->dst.field == RTE_FLOW_FIELD_META) { ++ meta = *(const unaligned_uint32_t *)item.spec; ++ meta = rte_cpu_to_be_32(meta); ++ item.spec = &meta; ++ } + } else { + type = MLX5_MODIFICATION_TYPE_COPY; + /** For COPY fill the destination field (dcopy) without mask. */ +@@ -2032,7 +2010,7 @@ flow_dv_validate_item_meta(struct rte_eth_dev *dev __rte_unused, + if (reg == REG_NON) + return rte_flow_error_set(error, ENOTSUP, + RTE_FLOW_ERROR_TYPE_ITEM, item, +- "unavalable extended metadata register"); ++ "unavailable extended metadata register"); + if (reg == REG_B) + return rte_flow_error_set(error, ENOTSUP, + RTE_FLOW_ERROR_TYPE_ITEM, item, +@@ -2879,8 +2857,6 @@ flow_dv_validate_action_push_vlan(struct rte_eth_dev *dev, + { + const struct rte_flow_action_of_push_vlan *push_vlan = action->conf; + const struct mlx5_priv *priv = dev->data->dev_private; +- struct mlx5_dev_ctx_shared *sh = priv->sh; +- bool direction_error = false; + + if (push_vlan->ethertype != RTE_BE16(RTE_ETHER_TYPE_VLAN) && + push_vlan->ethertype != RTE_BE16(RTE_ETHER_TYPE_QINQ)) +@@ -2892,22 +2868,6 @@ flow_dv_validate_action_push_vlan(struct rte_eth_dev *dev, + RTE_FLOW_ERROR_TYPE_ACTION, action, + "wrong action order, port_id should " + "be after push VLAN"); +- /* Push VLAN is not supported in ingress except for CX6 FDB mode. */ +- if (attr->transfer) { +- bool fdb_tx = priv->representor_id != UINT16_MAX; +- bool is_cx5 = sh->steering_format_version == +- MLX5_STEERING_LOGIC_FORMAT_CONNECTX_5; +- +- if (!fdb_tx && is_cx5) +- direction_error = true; +- } else if (attr->ingress) { +- direction_error = true; +- } +- if (direction_error) +- return rte_flow_error_set(error, ENOTSUP, +- RTE_FLOW_ERROR_TYPE_ATTR_INGRESS, +- NULL, +- "push vlan action not supported for ingress"); + if (!attr->transfer && priv->representor) + return rte_flow_error_set(error, ENOTSUP, + RTE_FLOW_ERROR_TYPE_UNSPECIFIED, NULL, +@@ -3205,7 +3165,7 @@ flow_dv_validate_action_set_meta(struct rte_eth_dev *dev, + if (reg == REG_NON) + return rte_flow_error_set(error, ENOTSUP, + RTE_FLOW_ERROR_TYPE_ACTION, action, +- "unavalable extended metadata register"); ++ "unavailable extended metadata register"); + if (reg != REG_A && reg != REG_B) { + struct mlx5_priv *priv = dev->data->dev_private; + +@@ -3283,6 +3243,25 @@ flow_dv_validate_action_set_tag(struct rte_eth_dev *dev, + return 0; + } - /* Delete all front-end monitors. Removing the last front-end monitor - * will also destroy the corresponding ovsdb_monitor. */ -- LIST_FOR_EACH_SAFE (jm, next_jm, node, &m->jsonrpc_monitors) { -+ LIST_FOR_EACH_SAFE (jm, node, &m->jsonrpc_monitors) { - ovsdb_jsonrpc_monitor_destroy(jm->jsonrpc_monitor, true); - } - } -diff --git a/ovsdb/ovsdb-idlc.in b/ovsdb/ovsdb-idlc.in -index 10a70ae26f..13c5359395 100755 ---- a/ovsdb/ovsdb-idlc.in -+++ b/ovsdb/ovsdb-idlc.in -@@ -251,10 +251,18 @@ const struct %(s)s *%(s)s_table_first(const struct %(s)s_table *); - for ((ROW) = %(s)s_table_first(TABLE); \\ - (ROW); \\ - (ROW) = %(s)s_next(ROW)) --#define %(S)s_TABLE_FOR_EACH_SAFE(ROW, NEXT, TABLE) \\ -+#define %(S)s_TABLE_FOR_EACH_SAFE_LONG(ROW, NEXT, TABLE) \\ - for ((ROW) = %(s)s_table_first(TABLE); \\ - (ROW) ? ((NEXT) = %(s)s_next(ROW), 1) : 0; \\ - (ROW) = (NEXT)) -+#define %(S)s_TABLE_FOR_EACH_SAFE_SHORT(ROW, TABLE) \\ -+ for (const struct %(s)s * ROW__next = ((ROW) = %(s)s_table_first(TABLE), NULL); \\ -+ (ROW) ? (ROW__next = %(s)s_next(ROW), 1) : (ROW__next = NULL, 0); \\ -+ (ROW) = ROW__next) -+#define %(S)s_TABLE_FOR_EACH_SAFE(...) \\ -+ OVERLOAD_SAFE_MACRO(%(S)s_TABLE_FOR_EACH_SAFE_LONG, \\ -+ %(S)s_TABLE_FOR_EACH_SAFE_SHORT, 3, __VA_ARGS__) ++/** ++ * Indicates whether ASO aging is supported. ++ * ++ * @param[in] sh ++ * Pointer to shared device context structure. ++ * @param[in] attr ++ * Attributes of flow that includes AGE action. ++ * ++ * @return ++ * True when ASO aging is supported, false otherwise. ++ */ ++static inline bool ++flow_hit_aso_supported(const struct mlx5_dev_ctx_shared *sh, ++ const struct rte_flow_attr *attr) ++{ ++ MLX5_ASSERT(sh && attr); ++ return (sh->flow_hit_aso_en && (attr->transfer || attr->group)); ++} + + /** + * Validate count action. + * +@@ -3292,6 +3271,8 @@ flow_dv_validate_action_set_tag(struct rte_eth_dev *dev, + * Indicator if action is shared. + * @param[in] action_flags + * Holds the actions detected until now. ++ * @param[in] attr ++ * Attributes of flow that includes this action. + * @param[out] error + * Pointer to error structure. + * +@@ -3301,6 +3282,7 @@ flow_dv_validate_action_set_tag(struct rte_eth_dev *dev, + static int + flow_dv_validate_action_count(struct rte_eth_dev *dev, bool shared, + uint64_t action_flags, ++ const struct rte_flow_attr *attr, + struct rte_flow_error *error) + { + struct mlx5_priv *priv = dev->data->dev_private; +@@ -3312,10 +3294,10 @@ flow_dv_validate_action_count(struct rte_eth_dev *dev, bool shared, + RTE_FLOW_ERROR_TYPE_ACTION, NULL, + "duplicate count actions set"); + if (shared && (action_flags & MLX5_FLOW_ACTION_AGE) && +- !priv->sh->flow_hit_aso_en) ++ !flow_hit_aso_supported(priv->sh, attr)) + return rte_flow_error_set(error, EINVAL, + RTE_FLOW_ERROR_TYPE_ACTION, NULL, +- "old age and shared count combination is not supported"); ++ "old age and indirect count combination is not supported"); + #ifdef HAVE_IBV_FLOW_DEVX_COUNTERS + return 0; + #endif +@@ -3740,7 +3722,8 @@ flow_dv_encap_decap_resource_register + flow_dv_encap_decap_match_cb, + flow_dv_encap_decap_remove_cb, + flow_dv_encap_decap_clone_cb, +- flow_dv_encap_decap_clone_free_cb); ++ flow_dv_encap_decap_clone_free_cb, ++ error); + if (unlikely(!encaps_decaps)) + return -rte_errno; + resource->flags = dev_flow->dv.group ? 0 : 1; +@@ -4982,7 +4965,7 @@ flow_dv_validate_action_jump(struct rte_eth_dev *dev, + const struct rte_flow_attr *attributes, + bool external, struct rte_flow_error *error) + { +- uint32_t target_group, table; ++ uint32_t target_group, table = 0; + int ret = 0; + struct flow_grp_info grp_info = { + .external = !!external, +@@ -5013,6 +4996,10 @@ flow_dv_validate_action_jump(struct rte_eth_dev *dev, + RTE_FLOW_ERROR_TYPE_ACTION, NULL, + "target group must be other than" + " the current flow group"); ++ if (table == 0) ++ return rte_flow_error_set(error, EINVAL, ++ RTE_FLOW_ERROR_TYPE_ACTION_CONF, ++ NULL, "root table shouldn't be destination"); + return 0; + } - const struct %(s)s *%(s)s_get_for_uuid(const struct ovsdb_idl *, const struct uuid *); - const struct %(s)s *%(s)s_table_get_for_uuid(const struct %(s)s_table *, const struct uuid *); -@@ -264,10 +272,17 @@ const struct %(s)s *%(s)s_next(const struct %(s)s *); - for ((ROW) = %(s)s_first(IDL); \\ - (ROW); \\ - (ROW) = %(s)s_next(ROW)) --#define %(S)s_FOR_EACH_SAFE(ROW, NEXT, IDL) \\ -+#define %(S)s_FOR_EACH_SAFE_LONG(ROW, NEXT, IDL) \\ - for ((ROW) = %(s)s_first(IDL); \\ - (ROW) ? ((NEXT) = %(s)s_next(ROW), 1) : 0; \\ - (ROW) = (NEXT)) -+#define %(S)s_FOR_EACH_SAFE_SHORT(ROW, IDL) \\ -+ for (const struct %(s)s * ROW__next = ((ROW) = %(s)s_first(IDL), NULL); \\ -+ (ROW) ? (ROW__next = %(s)s_next(ROW), 1) : (ROW__next = NULL, 0); \\ -+ (ROW) = ROW__next) -+#define %(S)s_FOR_EACH_SAFE(...) \\ -+ OVERLOAD_SAFE_MACRO(%(S)s_FOR_EACH_SAFE_LONG, \\ -+ %(S)s_FOR_EACH_SAFE_SHORT, 3, __VA_ARGS__) +@@ -5145,7 +5132,7 @@ flow_dv_modify_hdr_action_max(struct rte_eth_dev *dev __rte_unused, + * Pointer to error structure. + * + * @return +- * 0 on success, a negative errno value otherwise and rte_ernno is set. ++ * 0 on success, a negative errno value otherwise and rte_errno is set. + */ + static int + mlx5_flow_validate_action_meter(struct rte_eth_dev *dev, +@@ -5230,21 +5217,12 @@ mlx5_flow_validate_action_meter(struct rte_eth_dev *dev, + */ + struct mlx5_priv *policy_port_priv = + mtr_policy->dev->data->dev_private; +- int32_t flow_src_port = priv->representor_id; ++ uint16_t flow_src_port = priv->representor_id; + + if (port_id_item) { +- const struct rte_flow_item_port_id *spec = +- port_id_item->spec; +- struct mlx5_priv *port_priv = +- mlx5_port_to_eswitch_info(spec->id, +- false); +- if (!port_priv) +- return rte_flow_error_set(error, +- rte_errno, +- RTE_FLOW_ERROR_TYPE_ITEM_SPEC, +- spec, +- "Failed to get port info."); +- flow_src_port = port_priv->representor_id; ++ if (mlx5_flow_get_item_vport_id(dev, port_id_item, ++ &flow_src_port, error)) ++ return -rte_errno; + } + if (flow_src_port != policy_port_priv->representor_id) + return rte_flow_error_set(error, +@@ -5678,7 +5656,7 @@ flow_dv_validate_action_sample(uint64_t *action_flags, + case RTE_FLOW_ACTION_TYPE_COUNT: + ret = flow_dv_validate_action_count + (dev, false, *action_flags | sub_action_flags, +- error); ++ attr, error); + if (ret < 0) + return ret; + *count = act->conf; +@@ -5832,7 +5810,8 @@ flow_dv_modify_hdr_resource_register + flow_dv_modify_match_cb, + flow_dv_modify_remove_cb, + flow_dv_modify_clone_cb, +- flow_dv_modify_clone_free_cb); ++ flow_dv_modify_clone_free_cb, ++ error); + if (unlikely(!modify_cmds)) + return -rte_errno; + resource->root = !dev_flow->dv.group; +@@ -6714,6 +6693,12 @@ flow_dv_validate_item_integrity(struct rte_eth_dev *dev, + RTE_FLOW_ERROR_TYPE_ITEM, + integrity_item, + "unsupported integrity filter"); ++ if ((mask->l3_ok & !spec->l3_ok) || (mask->l4_ok & !spec->l4_ok) || ++ (mask->ipv4_csum_ok & !spec->ipv4_csum_ok) || ++ (mask->l4_csum_ok & !spec->l4_csum_ok)) ++ return rte_flow_error_set(error, EINVAL, ++ RTE_FLOW_ERROR_TYPE_ITEM, ++ NULL, "negative integrity flow is not supported"); + if (spec->level > 1) { + if (pattern_flags & MLX5_FLOW_ITEM_INNER_INTEGRITY) + return rte_flow_error_set +@@ -6844,7 +6829,7 @@ flow_dv_validate(struct rte_eth_dev *dev, const struct rte_flow_attr *attr, + bool external, int hairpin, struct rte_flow_error *error) + { + int ret; +- uint64_t action_flags = 0; ++ uint64_t aso_mask, action_flags = 0; + uint64_t item_flags = 0; + uint64_t last_item = 0; + uint8_t next_protocol = 0xff; +@@ -6911,7 +6896,11 @@ flow_dv_validate(struct rte_eth_dev *dev, const struct rte_flow_attr *attr, + const struct rte_flow_item *integrity_items[2] = {NULL, NULL}; + const struct rte_flow_item *port_id_item = NULL; + bool def_policy = false; ++ bool shared_count = false; + uint16_t udp_dport = 0; ++ uint32_t tag_id = 0; ++ const struct rte_flow_action_age *non_shared_age = NULL; ++ const struct rte_flow_action_count *count = NULL; + + if (items == NULL) + return -1; +@@ -7209,8 +7198,10 @@ flow_dv_validate(struct rte_eth_dev *dev, const struct rte_flow_attr *attr, + return ret; + last_item = MLX5_FLOW_ITEM_TAG; + break; +- case MLX5_RTE_FLOW_ITEM_TYPE_TAG: + case MLX5_RTE_FLOW_ITEM_TYPE_TX_QUEUE: ++ last_item = MLX5_FLOW_ITEM_TX_QUEUE; ++ break; ++ case MLX5_RTE_FLOW_ITEM_TYPE_TAG: + break; + case RTE_FLOW_ITEM_TYPE_GTP: + ret = flow_dv_validate_item_gtp(dev, items, item_flags, +@@ -7281,7 +7272,6 @@ flow_dv_validate(struct rte_eth_dev *dev, const struct rte_flow_attr *attr, + } + for (; actions->type != RTE_FLOW_ACTION_TYPE_END; actions++) { + int type = actions->type; +- bool shared_count = false; + + if (!mlx5_flow_os_action_supported(type)) + return rte_flow_error_set(error, ENOTSUP, +@@ -7380,6 +7370,8 @@ flow_dv_validate(struct rte_eth_dev *dev, const struct rte_flow_attr *attr, + ++actions_n; + if (action_flags & MLX5_FLOW_ACTION_SAMPLE) + modify_after_mirror = 1; ++ tag_id = ((const struct rte_flow_action_set_tag *) ++ actions->conf)->index; + action_flags |= MLX5_FLOW_ACTION_SET_TAG; + rw_act_num += MLX5_ACT_NUM_SET_TAG; + break; +@@ -7438,9 +7430,10 @@ flow_dv_validate(struct rte_eth_dev *dev, const struct rte_flow_attr *attr, + case RTE_FLOW_ACTION_TYPE_COUNT: + ret = flow_dv_validate_action_count(dev, shared_count, + action_flags, +- error); ++ attr, error); + if (ret < 0) + return ret; ++ count = actions->conf; + action_flags |= MLX5_FLOW_ACTION_COUNT; + ++actions_n; + break; +@@ -7746,6 +7739,7 @@ flow_dv_validate(struct rte_eth_dev *dev, const struct rte_flow_attr *attr, + ++actions_n; + break; + case RTE_FLOW_ACTION_TYPE_AGE: ++ non_shared_age = actions->conf; + ret = flow_dv_validate_action_age(action_flags, + actions, dev, + error); +@@ -7753,15 +7747,15 @@ flow_dv_validate(struct rte_eth_dev *dev, const struct rte_flow_attr *attr, + return ret; + /* + * Validate the regular AGE action (using counter) +- * mutual exclusion with share counter actions. ++ * mutual exclusion with indirect counter actions. + */ +- if (!priv->sh->flow_hit_aso_en) { ++ if (!flow_hit_aso_supported(priv->sh, attr)) { + if (shared_count) + return rte_flow_error_set + (error, EINVAL, + RTE_FLOW_ERROR_TYPE_ACTION, + NULL, +- "old age and shared count combination is not supported"); ++ "old age and indirect count combination is not supported"); + if (sample_count) + return rte_flow_error_set + (error, EINVAL, +@@ -7814,6 +7808,11 @@ flow_dv_validate(struct rte_eth_dev *dev, const struct rte_flow_attr *attr, + error); + if (ret < 0) + return ret; ++ if ((action_flags & MLX5_FLOW_ACTION_SET_TAG) && ++ tag_id == 0 && priv->mtr_color_reg == REG_NON) ++ return rte_flow_error_set(error, EINVAL, ++ RTE_FLOW_ERROR_TYPE_ACTION, NULL, ++ "sample after tag action causes metadata tag index 0 corruption"); + action_flags |= MLX5_FLOW_ACTION_SAMPLE; + ++actions_n; + break; +@@ -7858,7 +7857,7 @@ flow_dv_validate(struct rte_eth_dev *dev, const struct rte_flow_attr *attr, + * - Explicit decap action is prohibited by the tunnel offload API. + * - Drop action in tunnel steer rule is prohibited by the API. + * - Application cannot use MARK action because it's value can mask +- * tunnel default miss nitification. ++ * tunnel default miss notification. + * - JUMP in tunnel match rule has no support in current PMD + * implementation. + * - TAG & META are reserved for future uses. +@@ -7970,6 +7969,28 @@ flow_dv_validate(struct rte_eth_dev *dev, const struct rte_flow_attr *attr, + RTE_FLOW_ERROR_TYPE_ACTION, + NULL, "encap and decap " + "combination aren't supported"); ++ /* Push VLAN is not supported in ingress except for NICs newer than CX5. */ ++ if (action_flags & MLX5_FLOW_ACTION_OF_PUSH_VLAN) { ++ struct mlx5_dev_ctx_shared *sh = priv->sh; ++ bool direction_error = false; ++ ++ if (attr->transfer) { ++ bool fdb_tx = priv->representor_id != UINT16_MAX; ++ bool is_cx5 = sh->steering_format_version == ++ MLX5_STEERING_LOGIC_FORMAT_CONNECTX_5; ++ ++ if (!fdb_tx && is_cx5) ++ direction_error = true; ++ } else if (attr->ingress) { ++ direction_error = true; ++ } ++ if (direction_error) ++ return rte_flow_error_set(error, ENOTSUP, ++ RTE_FLOW_ERROR_TYPE_ATTR_INGRESS, ++ NULL, ++ "push VLAN action not supported " ++ "for ingress"); ++ } + if (!attr->transfer && attr->ingress) { + if (action_flags & MLX5_FLOW_ACTION_ENCAP) + return rte_flow_error_set +@@ -7977,12 +7998,6 @@ flow_dv_validate(struct rte_eth_dev *dev, const struct rte_flow_attr *attr, + RTE_FLOW_ERROR_TYPE_ACTION, + NULL, "encap is not supported" + " for ingress traffic"); +- else if (action_flags & MLX5_FLOW_ACTION_OF_PUSH_VLAN) +- return rte_flow_error_set +- (error, ENOTSUP, +- RTE_FLOW_ERROR_TYPE_ACTION, +- NULL, "push VLAN action not " +- "supported for ingress"); + else if ((action_flags & MLX5_FLOW_VLAN_ACTIONS) == + MLX5_FLOW_VLAN_ACTIONS) + return rte_flow_error_set +@@ -8022,6 +8037,20 @@ flow_dv_validate(struct rte_eth_dev *dev, const struct rte_flow_attr *attr, + "cannot be done before meter action"); + } + } ++ /* ++ * Only support one ASO action in a single flow rule. ++ * non-shared AGE + counter will fallback to use HW counter, no ASO hit object. ++ * Group 0 uses HW counter for AGE too even if no counter action. ++ */ ++ aso_mask = (action_flags & MLX5_FLOW_ACTION_METER && priv->sh->meter_aso_en) << 2 | ++ (action_flags & MLX5_FLOW_ACTION_CT && priv->sh->ct_aso_en) << 1 | ++ (action_flags & MLX5_FLOW_ACTION_AGE && ++ !(non_shared_age && count) && ++ (attr->group || (attr->transfer && priv->fdb_def_rule)) && ++ priv->sh->flow_hit_aso_en); ++ if (__builtin_popcountl(aso_mask) > 1) ++ return rte_flow_error_set(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ACTION, ++ NULL, "unsupported combining AGE, METER, CT ASO actions in a single rule"); + /* + * Hairpin flow will add one more TAG action in TX implicit mode. + * In TX explicit mode, there will be no hairpin flow ID. +@@ -8045,6 +8074,18 @@ flow_dv_validate(struct rte_eth_dev *dev, const struct rte_flow_attr *attr, + return rte_flow_error_set(error, EINVAL, + RTE_FLOW_ERROR_TYPE_ACTION, NULL, + "sample before modify action is not supported"); ++ /* ++ * Validation the NIC Egress flow on representor, except implicit ++ * hairpin default egress flow with TX_QUEUE item, other flows not ++ * work due to metadata regC0 mismatch. ++ */ ++ if ((!attr->transfer && attr->egress) && priv->representor && ++ !(item_flags & MLX5_FLOW_ITEM_TX_QUEUE)) ++ return rte_flow_error_set(error, EINVAL, ++ RTE_FLOW_ERROR_TYPE_ITEM, ++ NULL, ++ "NIC egress rules on representors" ++ " is not supported"); + return 0; + } - unsigned int %(s)s_get_seqno(const struct ovsdb_idl *); - unsigned int %(s)s_row_get_seqno(const struct %(s)s *row, enum ovsdb_idl_change change); -diff --git a/ovsdb/ovsdb-server.c b/ovsdb/ovsdb-server.c -index 9fe90592ea..774416fc7a 100644 ---- a/ovsdb/ovsdb-server.c -+++ b/ovsdb/ovsdb-server.c -@@ -26,6 +26,7 @@ - #include "command-line.h" - #include "daemon.h" - #include "dirs.h" -+#include "dns-resolve.h" - #include "openvswitch/dynamic-string.h" - #include "fatal-signal.h" - #include "file.h" -@@ -228,8 +229,7 @@ main_loop(struct server_config *config, +@@ -9184,7 +9225,7 @@ flow_dev_geneve_tlv_option_resource_register(struct rte_eth_dev *dev, + geneve_opt_v->option_type && + geneve_opt_resource->length == + geneve_opt_v->option_len) { +- /* We already have GENVE TLV option obj allocated. */ ++ /* We already have GENEVE TLV option obj allocated. */ + __atomic_fetch_add(&geneve_opt_resource->refcnt, 1, + __ATOMIC_RELAXED); + } else { +@@ -9713,7 +9754,7 @@ flow_dv_translate_item_port_id(struct rte_eth_dev *dev, void *matcher, + + if (pid_v && pid_v->id == MLX5_PORT_ESW_MGR) { + flow_dv_translate_item_source_vport(matcher, key, +- flow_dv_get_esw_manager_vport_id(dev), 0xffff); ++ mlx5_flow_get_esw_manager_vport_id(dev), 0xffff); + return 0; + } + mask = pid_m ? pid_m->id : 0xffff; +@@ -10170,7 +10211,7 @@ flow_dv_translate_item_flex(struct rte_eth_dev *dev, void *matcher, void *key, + /* Don't count both inner and outer flex items in one rule. */ + if (mlx5_flex_acquire_index(dev, spec->handle, true) != index) + MLX5_ASSERT(false); +- dev_flow->handle->flex_item |= RTE_BIT32(index); ++ dev_flow->handle->flex_item |= (uint8_t)RTE_BIT32(index); + } + mlx5_flex_flow_translate_item(dev, matcher, key, item, is_inner); + } +@@ -10226,7 +10267,7 @@ __flow_dv_adjust_buf_size(size_t *size, uint8_t match_criteria) + * Check flow matching criteria first, subtract misc5/4 length if flow + * doesn't own misc5/4 parameters. In some old rdma-core releases, + * misc5/4 are not supported, and matcher creation failure is expected +- * w/o subtration. If misc5 is provided, misc4 must be counted in since ++ * w/o subtraction. If misc5 is provided, misc4 must be counted in since + * misc5 is right after misc4. + */ + if (!(match_criteria & (1 << MLX5_MATCH_CRITERIA_ENABLE_MISC5_BIT))) { +@@ -10514,7 +10555,8 @@ flow_dv_tbl_remove_cb(void *tool_ctx, struct mlx5_list_entry *entry) + tbl_data->tunnel->tunnel_id : 0, + tbl_data->group_id); + } +- mlx5_list_destroy(tbl_data->matchers); ++ if (tbl_data->matchers) ++ mlx5_list_destroy(tbl_data->matchers); + mlx5_ipool_free(sh->ipool[MLX5_IPOOL_JUMP], tbl_data->idx); + } - ovsdb_relay_run(); +@@ -10769,7 +10811,8 @@ flow_dv_tag_resource_register + flow_dv_tag_match_cb, + flow_dv_tag_remove_cb, + flow_dv_tag_clone_cb, +- flow_dv_tag_clone_free_cb); ++ flow_dv_tag_clone_free_cb, ++ error); + if (unlikely(!tag_table)) + return -rte_errno; + entry = mlx5_hlist_register(tag_table, tag_be24, &ctx); +@@ -11074,6 +11117,7 @@ flow_dv_hrxq_prepare(struct rte_eth_dev *dev, + { + struct mlx5_priv *priv = dev->data->dev_private; + struct mlx5_flow_handle *dh = dev_flow->handle; ++ uint32_t shared_rss = rss_desc->shared_rss; + struct mlx5_hrxq *hrxq; + + MLX5_ASSERT(rss_desc->queue_num); +@@ -11088,6 +11132,7 @@ flow_dv_hrxq_prepare(struct rte_eth_dev *dev, + return NULL; + hrxq = mlx5_ipool_get(priv->sh->ipool[MLX5_IPOOL_HRXQ], + *hrxq_idx); ++ rss_desc->shared_rss = shared_rss; + return hrxq; + } -- struct shash_node *next; -- SHASH_FOR_EACH_SAFE (node, next, all_dbs) { -+ SHASH_FOR_EACH_SAFE (node, all_dbs) { - struct db *db = node->data; - ovsdb_txn_history_run(db->db); - ovsdb_storage_run(db->db->storage); -@@ -321,7 +321,7 @@ main(int argc, char *argv[]) - FILE *config_tmpfile; - struct server_config server_config; - struct shash all_dbs; -- struct shash_node *node, *next; -+ struct shash_node *node; - int replication_probe_interval = REPLICATION_DEFAULT_PROBE_INTERVAL; +@@ -11425,7 +11470,7 @@ flow_dv_dest_array_create_cb(void *tool_ctx __rte_unused, void *cb_ctx) + goto error; + } + } +- /* create a dest array actioin */ ++ /* create a dest array action */ + ret = mlx5_os_flow_dr_create_flow_action_dest_array + (domain, + resource->num_of_dest, +@@ -11660,7 +11705,7 @@ flow_dv_translate_action_sample(struct rte_eth_dev *dev, + (((const struct rte_flow_action_mark *) + (sub_actions->conf))->id); + +- dev_flow->handle->mark = 1; ++ wks->mark = 1; + pre_rix = dev_flow->handle->dvh.rix_tag; + /* Save the mark resource before sample */ + pre_r = dev_flow->dv.tag_resource; +@@ -12820,7 +12865,7 @@ flow_dv_translate(struct rte_eth_dev *dev, + break; + case RTE_FLOW_ACTION_TYPE_FLAG: + action_flags |= MLX5_FLOW_ACTION_FLAG; +- dev_flow->handle->mark = 1; ++ wks->mark = 1; + if (dev_conf->dv_xmeta_en != MLX5_XMETA_MODE_LEGACY) { + struct rte_flow_action_mark mark = { + .id = MLX5_FLOW_MARK_DEFAULT, +@@ -12849,7 +12894,7 @@ flow_dv_translate(struct rte_eth_dev *dev, + break; + case RTE_FLOW_ACTION_TYPE_MARK: + action_flags |= MLX5_FLOW_ACTION_MARK; +- dev_flow->handle->mark = 1; ++ wks->mark = 1; + if (dev_conf->dv_xmeta_en != MLX5_XMETA_MODE_LEGACY) { + const struct rte_flow_action_mark *mark = + (const struct rte_flow_action_mark *) +@@ -13306,8 +13351,7 @@ flow_dv_translate(struct rte_eth_dev *dev, + */ + if (action_flags & MLX5_FLOW_ACTION_AGE) { + if ((non_shared_age && count) || +- !(priv->sh->flow_hit_aso_en && +- (attr->group || attr->transfer))) { ++ !flow_hit_aso_supported(priv->sh, attr)) { + /* Creates age by counters. */ + cnt_act = flow_dv_prepare_counter + (dev, dev_flow, +@@ -13538,11 +13582,13 @@ flow_dv_translate(struct rte_eth_dev *dev, + case RTE_FLOW_ITEM_TYPE_ICMP: + flow_dv_translate_item_icmp(match_mask, match_value, + items, tunnel); ++ matcher.priority = MLX5_PRIORITY_MAP_L4; + last_item = MLX5_FLOW_LAYER_ICMP; + break; + case RTE_FLOW_ITEM_TYPE_ICMP6: + flow_dv_translate_item_icmp6(match_mask, match_value, + items, tunnel); ++ matcher.priority = MLX5_PRIORITY_MAP_L4; + last_item = MLX5_FLOW_LAYER_ICMP6; + break; + case RTE_FLOW_ITEM_TYPE_TAG: +@@ -13617,12 +13663,14 @@ flow_dv_translate(struct rte_eth_dev *dev, + /* + * When E-Switch mode is enabled, we have two cases where we need to + * set the source port manually. +- * The first one, is in case of Nic steering rule, and the second is +- * E-Switch rule where no port_id item was found. In both cases +- * the source port is set according the current port in use. ++ * The first one, is in case of NIC ingress steering rule, and the ++ * second is E-Switch rule where no port_id item was found. ++ * In both cases the source port is set according the current port ++ * in use. + */ + if (!(item_flags & MLX5_FLOW_ITEM_PORT_ID) && +- (priv->representor || priv->master)) { ++ (priv->representor || priv->master) && ++ !(attr->egress && !attr->transfer)) { + if (flow_dv_translate_item_port_id(dev, match_mask, + match_value, NULL, attr)) + return -rte_errno; +@@ -14508,7 +14556,7 @@ flow_dv_destroy(struct rte_eth_dev *dev, struct rte_flow *flow) + int index = rte_bsf32(dev_handle->flex_item); + + mlx5_flex_release_index(dev, index); +- dev_handle->flex_item &= ~RTE_BIT32(index); ++ dev_handle->flex_item &= ~(uint8_t)RTE_BIT32(index); + } + if (dev_handle->dvh.matcher) + flow_dv_matcher_release(dev, dev_handle); +@@ -14607,8 +14655,8 @@ __flow_dv_action_rss_hrxqs_release(struct rte_eth_dev *dev, + * MLX5_RSS_HASH_IPV4_DST_ONLY are mutually exclusive so they can share + * same slot in mlx5_rss_hash_fields. + * +- * @param[in] rss +- * Pointer to the shared action RSS conf. ++ * @param[in] orig_rss_types ++ * RSS type as provided in shared RSS action. + * @param[in, out] hash_field + * hash_field variable needed to be adjusted. + * +@@ -14616,10 +14664,10 @@ __flow_dv_action_rss_hrxqs_release(struct rte_eth_dev *dev, + * void + */ + static void +-__flow_dv_action_rss_l34_hash_adjust(struct mlx5_shared_action_rss *rss, ++__flow_dv_action_rss_l34_hash_adjust(uint64_t orig_rss_types, + uint64_t *hash_field) + { +- uint64_t rss_types = rss->origin.types; ++ uint64_t rss_types = rte_eth_rss_hf_refine(orig_rss_types); + + switch (*hash_field & ~IBV_RX_HASH_INNER) { + case MLX5_RSS_HASH_IPV4: +@@ -14721,7 +14769,8 @@ __flow_dv_action_rss_setup(struct rte_eth_dev *dev, + uint64_t hash_fields = mlx5_rss_hash_fields[i]; + int tunnel = 0; + +- __flow_dv_action_rss_l34_hash_adjust(shared_rss, &hash_fields); ++ __flow_dv_action_rss_l34_hash_adjust(shared_rss->origin.types, ++ &hash_fields); + if (shared_rss->origin.level > 1) { + hash_fields |= IBV_RX_HASH_INNER; + tunnel = 1; +@@ -15455,7 +15504,6 @@ __flow_dv_create_domain_policy_acts(struct rte_eth_dev *dev, + NULL, + "cannot create policy " + "mark action for this color"); +- dev_flow.handle->mark = 1; + if (flow_dv_tag_resource_register(dev, tag_be, + &dev_flow, &flow_err)) + return -rte_mtr_error_set(error, +@@ -15467,6 +15515,7 @@ __flow_dv_create_domain_policy_acts(struct rte_eth_dev *dev, + act_cnt->rix_mark = + dev_flow.handle->dvh.rix_tag; + action_flags |= MLX5_FLOW_ACTION_MARK; ++ mtr_policy->mark = 1; + break; + } + case RTE_FLOW_ACTION_TYPE_SET_TAG: +@@ -15750,6 +15799,8 @@ __flow_dv_create_domain_policy_acts(struct rte_eth_dev *dev, + act_cnt->next_sub_policy = NULL; + mtr_policy->is_hierarchy = 1; + mtr_policy->dev = next_policy->dev; ++ if (next_policy->mark) ++ mtr_policy->mark = 1; + action_flags |= + MLX5_FLOW_ACTION_METER_WITH_TERMINATED_POLICY; + break; +@@ -16880,7 +16931,9 @@ __flow_dv_meter_get_rss_sub_policy(struct rte_eth_dev *dev, + struct mlx5_meter_policy_action_container *act_cnt; + uint32_t domain = MLX5_MTR_DOMAIN_INGRESS; + uint16_t sub_policy_num; ++ struct mlx5_flow_workspace *wks = mlx5_flow_get_thread_workspace(); + ++ MLX5_ASSERT(wks); + rte_spinlock_lock(&mtr_policy->sl); + for (i = 0; i < MLX5_MTR_RTE_COLORS; i++) { + if (!rss_desc[i]) +@@ -16914,7 +16967,8 @@ __flow_dv_meter_get_rss_sub_policy(struct rte_eth_dev *dev, + } + } + /* Create sub policy. */ +- if (!mtr_policy->sub_policys[domain][0]->rix_hrxq[0]) { ++ if (!mtr_policy->sub_policys[domain][0]->rix_hrxq[RTE_COLOR_GREEN] && ++ !mtr_policy->sub_policys[domain][0]->rix_hrxq[RTE_COLOR_YELLOW]) { + /* Reuse the first pre-allocated sub_policy. */ + sub_policy = mtr_policy->sub_policys[domain][0]; + sub_policy_idx = sub_policy->idx; +@@ -16954,7 +17008,7 @@ __flow_dv_meter_get_rss_sub_policy(struct rte_eth_dev *dev, + if (act_cnt->rix_mark || act_cnt->modify_hdr) { + memset(&dh, 0, sizeof(struct mlx5_flow_handle)); + if (act_cnt->rix_mark) +- dh.mark = 1; ++ wks->mark = 1; + dh.fate_action = MLX5_FLOW_FATE_QUEUE; + dh.rix_hrxq = hrxq_idx[i]; + flow_drv_rxq_flags_set(dev, &dh); +@@ -17635,7 +17689,7 @@ flow_dv_action_validate(struct rte_eth_dev *dev, + "Indirect age action not supported"); + return flow_dv_validate_action_age(0, action, dev, err); + case RTE_FLOW_ACTION_TYPE_COUNT: +- return flow_dv_validate_action_count(dev, true, 0, err); ++ return flow_dv_validate_action_count(dev, true, 0, NULL, err); + case RTE_FLOW_ACTION_TYPE_CONNTRACK: + if (!priv->sh->ct_aso_en) + return rte_flow_error_set(err, ENOTSUP, +@@ -18291,4 +18345,3 @@ const struct mlx5_flow_driver_ops mlx5_flow_dv_drv_ops = { + }; - ovs_cmdl_proctitle_init(argc, argv); -@@ -329,6 +329,7 @@ main(int argc, char *argv[]) - service_start(&argc, &argv); - fatal_ignore_sigpipe(); - process_init(); -+ dns_resolve_init(true); + #endif /* HAVE_IBV_FLOW_DV_SUPPORT */ +- +diff --git a/dpdk/drivers/net/mlx5/mlx5_flow_flex.c b/dpdk/drivers/net/mlx5/mlx5_flow_flex.c +index 64867dc9e2..3ef46db1f6 100644 +--- a/dpdk/drivers/net/mlx5/mlx5_flow_flex.c ++++ b/dpdk/drivers/net/mlx5/mlx5_flow_flex.c +@@ -205,7 +205,7 @@ mlx5_flex_set_match_sample(void *misc4_m, void *misc4_v, + * @param dev + * Ethernet device to translate flex item on. + * @param[in, out] matcher +- * Flow matcher to confgiure ++ * Flow matcher to configure + * @param[in, out] key + * Flow matcher value. + * @param[in] item +@@ -382,15 +382,11 @@ mlx5_flex_translate_length(struct mlx5_hca_flex_attr *attr, + return rte_flow_error_set + (error, EINVAL, RTE_FLOW_ERROR_TYPE_ITEM, NULL, + "unsupported header length field mode (FIXED)"); +- if (attr->header_length_mask_width < field->field_size) +- return rte_flow_error_set +- (error, EINVAL, RTE_FLOW_ERROR_TYPE_ITEM, NULL, +- "header length field width exceeds limit"); +- if (field->offset_shift < 0 || +- field->offset_shift > attr->header_length_mask_width) ++ if (field->field_size || ++ field->offset_mask || field->offset_shift) + return rte_flow_error_set + (error, EINVAL, RTE_FLOW_ERROR_TYPE_ITEM, NULL, +- "invalid header length field shift (FIXED"); ++ "invalid fields for fixed mode"); + if (field->field_base < 0) + return rte_flow_error_set + (error, EINVAL, RTE_FLOW_ERROR_TYPE_ITEM, NULL, +@@ -457,7 +453,7 @@ mlx5_flex_translate_length(struct mlx5_hca_flex_attr *attr, + if (field->offset_shift > 15 || field->offset_shift < 0) + return rte_flow_error_set + (error, EINVAL, RTE_FLOW_ERROR_TYPE_ITEM, NULL, +- "header length field shift exceeeds limit"); ++ "header length field shift exceeds limit"); + node->header_length_field_shift = field->offset_shift; + node->header_length_field_offset = field->offset_base; + } +diff --git a/dpdk/drivers/net/mlx5/mlx5_flow_meter.c b/dpdk/drivers/net/mlx5/mlx5_flow_meter.c +index f4a7b697e6..a58e30dc83 100644 +--- a/dpdk/drivers/net/mlx5/mlx5_flow_meter.c ++++ b/dpdk/drivers/net/mlx5/mlx5_flow_meter.c +@@ -15,6 +15,9 @@ + #include "mlx5.h" + #include "mlx5_flow.h" + ++static int mlx5_flow_meter_disable(struct rte_eth_dev *dev, ++ uint32_t meter_id, struct rte_mtr_error *error); ++ + /** + * Create the meter action. + * +@@ -251,7 +254,7 @@ mlx5_flow_meter_xir_man_exp_calc(int64_t xir, uint8_t *man, uint8_t *exp) + uint8_t _exp = 0; + uint64_t m, e; + +- /* Special case xir == 0 ? both exp and matissa are 0. */ ++ /* Special case xir == 0 ? both exp and mantissa are 0. */ + if (xir == 0) { + *man = 0; + *exp = 0; +@@ -287,7 +290,7 @@ mlx5_flow_meter_xbs_man_exp_calc(uint64_t xbs, uint8_t *man, uint8_t *exp) + int _exp; + double _man; + +- /* Special case xbs == 0 ? both exp and matissa are 0. */ ++ /* Special case xbs == 0 ? both exp and mantissa are 0. */ + if (xbs == 0) { + *man = 0; + *exp = 0; +@@ -295,8 +298,10 @@ mlx5_flow_meter_xbs_man_exp_calc(uint64_t xbs, uint8_t *man, uint8_t *exp) + } + /* xbs = xbs_mantissa * 2^xbs_exponent */ + _man = frexp(xbs, &_exp); +- _man = _man * pow(2, MLX5_MAN_WIDTH); +- _exp = _exp - MLX5_MAN_WIDTH; ++ if (_exp >= MLX5_MAN_WIDTH) { ++ _man = _man * pow(2, MLX5_MAN_WIDTH); ++ _exp = _exp - MLX5_MAN_WIDTH; ++ } + *man = (uint8_t)ceil(_man); + *exp = _exp; + } +@@ -305,7 +310,7 @@ mlx5_flow_meter_xbs_man_exp_calc(uint64_t xbs, uint8_t *man, uint8_t *exp) + * Fill the prm meter parameter. + * + * @param[in,out] fmp +- * Pointer to meter profie to be converted. ++ * Pointer to meter profile to be converted. + * @param[out] error + * Pointer to the error structure. + * +@@ -437,11 +442,14 @@ mlx5_flow_mtr_cap_get(struct rte_eth_dev *dev, + /* 2 meters per one ASO cache line. */ + cap->n_max = 1 << (qattr->log_max_num_meter_aso + 1); + cap->srtcm_rfc2697_packet_mode_supported = 1; ++ cap->trtcm_rfc2698_packet_mode_supported = 1; ++ cap->trtcm_rfc4115_packet_mode_supported = 1; + } else { + cap->n_max = 1 << qattr->log_max_flow_meter; +- cap->srtcm_rfc2697_packet_mode_supported = 0; + } + cap->srtcm_rfc2697_byte_mode_supported = 1; ++ cap->trtcm_rfc2698_byte_mode_supported = 1; ++ cap->trtcm_rfc4115_byte_mode_supported = 1; + cap->n_shared_max = cap->n_max; + cap->identical = 1; + cap->shared_identical = 1; +@@ -449,7 +457,10 @@ mlx5_flow_mtr_cap_get(struct rte_eth_dev *dev, + /* 2M flows can share the same meter. */ + cap->chaining_n_mtrs_per_flow_max = 1; /* Chaining is not supported. */ + cap->meter_srtcm_rfc2697_n_max = qattr->flow_meter_old ? cap->n_max : 0; ++ cap->meter_trtcm_rfc2698_n_max = qattr->flow_meter_old ? cap->n_max : 0; ++ cap->meter_trtcm_rfc4115_n_max = qattr->flow_meter_old ? cap->n_max : 0; + cap->meter_rate_max = 1ULL << 40; /* 1 Tera tokens per sec. */ ++ cap->meter_policy_n_max = cap->n_max; + cap->stats_mask = RTE_MTR_STATS_N_BYTES_DROPPED | + RTE_MTR_STATS_N_PKTS_DROPPED; + return 0; +@@ -1101,7 +1112,7 @@ mlx5_flow_meter_action_modify(struct mlx5_priv *priv, + if (ret) + return ret; + } +- /* Update succeedded modify meter parameters. */ ++ /* Update succeeded modify meter parameters. */ + if (modify_bits & MLX5_FLOW_METER_OBJ_MODIFY_FIELD_ACTIVE) + fm->active_state = !!active_state; + } +@@ -1167,7 +1178,8 @@ mlx5_flow_meter_create(struct rte_eth_dev *dev, uint32_t meter_id, + struct mlx5_legacy_flow_meters *fms = &priv->flow_meters; + struct mlx5_flow_meter_profile *fmp; + struct mlx5_flow_meter_info *fm; +- struct mlx5_legacy_flow_meter *legacy_fm; ++ /* GCC fails to infer legacy_fm is set when !priv->sh->meter_aso_en. */ ++ struct mlx5_legacy_flow_meter *legacy_fm = NULL; + struct mlx5_flow_meter_policy *mtr_policy = NULL; + struct mlx5_indexed_pool_config flow_ipool_cfg = { + .size = 0, +@@ -1273,11 +1285,13 @@ mlx5_flow_meter_create(struct rte_eth_dev *dev, uint32_t meter_id, + if (mlx5_flow_create_mtr_tbls(dev, fm, mtr_idx, domain_bitmap)) + goto error; + /* Add to the flow meter list. */ +- if (!priv->sh->meter_aso_en) ++ if (!priv->sh->meter_aso_en) { ++ MLX5_ASSERT(legacy_fm != NULL); + TAILQ_INSERT_TAIL(fms, legacy_fm, next); ++ } + /* Add to the flow meter list. */ + fm->active_state = 1; /* Config meter starts as active. */ +- fm->is_enable = 1; ++ fm->is_enable = params->meter_enable; + fm->shared = !!shared; + __atomic_add_fetch(&fm->profile->ref_cnt, 1, __ATOMIC_RELAXED); + if (params->meter_policy_id == priv->sh->mtrmng->def_policy_id) { +@@ -1302,7 +1316,10 @@ mlx5_flow_meter_create(struct rte_eth_dev *dev, uint32_t meter_id, + data.dword = mtr_idx; + if (mlx5_l3t_set_entry(priv->mtr_idx_tbl, meter_id, &data)) + goto error; ++ } else if (!params->meter_enable && mlx5_flow_meter_disable(dev, meter_id, error)) { ++ goto error; + } ++ fm->active_state = params->meter_enable; + if (mtr_policy) + __atomic_add_fetch(&mtr_policy->ref_cnt, 1, __ATOMIC_RELAXED); + return 0; +@@ -1615,7 +1632,7 @@ mlx5_flow_meter_profile_update(struct rte_eth_dev *dev, + return -rte_mtr_error_set(error, -ret, + RTE_MTR_ERROR_TYPE_MTR_PARAMS, + NULL, "Failed to update meter" +- " parmeters in hardware."); ++ " parameters in hardware."); + } + old_fmp->ref_cnt--; + fmp->ref_cnt++; +diff --git a/dpdk/drivers/net/mlx5/mlx5_flow_verbs.c b/dpdk/drivers/net/mlx5/mlx5_flow_verbs.c +index 29cd694752..165786f864 100644 +--- a/dpdk/drivers/net/mlx5/mlx5_flow_verbs.c ++++ b/dpdk/drivers/net/mlx5/mlx5_flow_verbs.c +@@ -882,13 +882,48 @@ flow_verbs_item_gre_ip_protocol_update(struct ibv_flow_attr *attr, + } + } - bool active = false; - parse_options(argc, argv, &db_filenames, &remotes, &unixctl_path, -@@ -490,7 +491,7 @@ main(int argc, char *argv[]) - main_loop(&server_config, jsonrpc, &all_dbs, unixctl, &remotes, - run_process, &exiting, &is_backup); ++/** ++ * Reserve space for GRE spec in spec buffer. ++ * ++ * @param[in,out] dev_flow ++ * Pointer to dev_flow structure. ++ * ++ * @return ++ * Pointer to reserved space in spec buffer. ++ */ ++static uint8_t * ++flow_verbs_reserve_gre(struct mlx5_flow *dev_flow) ++{ ++ uint8_t *buffer; ++ struct mlx5_flow_verbs_workspace *verbs = &dev_flow->verbs; ++#ifndef HAVE_IBV_DEVICE_MPLS_SUPPORT ++ unsigned int size = sizeof(struct ibv_flow_spec_tunnel); ++ struct ibv_flow_spec_tunnel tunnel = { ++ .type = IBV_FLOW_SPEC_VXLAN_TUNNEL, ++ .size = size, ++ }; ++#else ++ unsigned int size = sizeof(struct ibv_flow_spec_gre); ++ struct ibv_flow_spec_gre tunnel = { ++ .type = IBV_FLOW_SPEC_GRE, ++ .size = size, ++ }; ++#endif ++ ++ buffer = verbs->specs + verbs->size; ++ flow_verbs_spec_add(verbs, &tunnel, size); ++ return buffer; ++} ++ + /** + * Convert the @p item into a Verbs specification. This function assumes that +- * the input is valid and that there is space to insert the requested item +- * into the flow. ++ * the input is valid and that Verbs specification will be placed in ++ * the pre-reserved space. + * + * @param[in, out] dev_flow + * Pointer to dev_flow structure. ++ * @param[in, out] gre_spec ++ * Pointer to space reserved for GRE spec. + * @param[in] item + * Item specification. + * @param[in] item_flags +@@ -896,6 +931,7 @@ flow_verbs_item_gre_ip_protocol_update(struct ibv_flow_attr *attr, + */ + static void + flow_verbs_translate_item_gre(struct mlx5_flow *dev_flow, ++ uint8_t *gre_spec, + const struct rte_flow_item *item __rte_unused, + uint64_t item_flags) + { +@@ -907,6 +943,7 @@ flow_verbs_translate_item_gre(struct mlx5_flow *dev_flow, + .size = size, + }; + #else ++ static const struct rte_flow_item_gre empty_gre = {0,}; + const struct rte_flow_item_gre *spec = item->spec; + const struct rte_flow_item_gre *mask = item->mask; + unsigned int size = sizeof(struct ibv_flow_spec_gre); +@@ -915,17 +952,29 @@ flow_verbs_translate_item_gre(struct mlx5_flow *dev_flow, + .size = size, + }; + +- if (!mask) +- mask = &rte_flow_item_gre_mask; +- if (spec) { +- tunnel.val.c_ks_res0_ver = spec->c_rsvd0_ver; +- tunnel.val.protocol = spec->protocol; +- tunnel.mask.c_ks_res0_ver = mask->c_rsvd0_ver; +- tunnel.mask.protocol = mask->protocol; +- /* Remove unwanted bits from values. */ +- tunnel.val.c_ks_res0_ver &= tunnel.mask.c_ks_res0_ver; ++ if (!spec) { ++ spec = &empty_gre; ++ mask = &empty_gre; ++ } else { ++ if (!mask) ++ mask = &rte_flow_item_gre_mask; ++ } ++ tunnel.val.c_ks_res0_ver = spec->c_rsvd0_ver; ++ tunnel.val.protocol = spec->protocol; ++ tunnel.mask.c_ks_res0_ver = mask->c_rsvd0_ver; ++ tunnel.mask.protocol = mask->protocol; ++ /* Remove unwanted bits from values. */ ++ tunnel.val.c_ks_res0_ver &= tunnel.mask.c_ks_res0_ver; ++ tunnel.val.key &= tunnel.mask.key; ++ if (tunnel.mask.protocol) { + tunnel.val.protocol &= tunnel.mask.protocol; +- tunnel.val.key &= tunnel.mask.key; ++ } else { ++ tunnel.val.protocol = mlx5_translate_tunnel_etypes(item_flags); ++ if (tunnel.val.protocol) { ++ tunnel.mask.protocol = 0xFFFF; ++ tunnel.val.protocol = ++ rte_cpu_to_be_16(tunnel.val.protocol); ++ } + } + #endif + if (item_flags & MLX5_FLOW_LAYER_OUTER_L3_IPV4) +@@ -936,7 +985,8 @@ flow_verbs_translate_item_gre(struct mlx5_flow *dev_flow, + flow_verbs_item_gre_ip_protocol_update(&verbs->attr, + IBV_FLOW_SPEC_IPV6, + IPPROTO_GRE); +- flow_verbs_spec_add(verbs, &tunnel, size); ++ MLX5_ASSERT(gre_spec); ++ memcpy(gre_spec, &tunnel, size); + } -- SHASH_FOR_EACH_SAFE(node, next, &all_dbs) { -+ SHASH_FOR_EACH_SAFE (node, &all_dbs) { - struct db *db = node->data; - close_db(&server_config, db, NULL); - shash_delete(&all_dbs, node); -@@ -511,6 +512,7 @@ main(int argc, char *argv[]) - run_command, process_status_msg(status)); - } - } -+ dns_resolve_destroy(); - perf_counters_destroy(); - service_stop(); - return 0; -@@ -1240,8 +1242,8 @@ update_server_status(struct shash *all_dbs) + /** +@@ -1666,6 +1716,8 @@ flow_verbs_translate(struct rte_eth_dev *dev, + struct mlx5_priv *priv = dev->data->dev_private; + struct mlx5_flow_workspace *wks = mlx5_flow_get_thread_workspace(); + struct mlx5_flow_rss_desc *rss_desc; ++ const struct rte_flow_item *tunnel_item = NULL; ++ uint8_t *gre_spec = NULL; + + MLX5_ASSERT(wks); + rss_desc = &wks->rss_desc; +@@ -1680,12 +1732,12 @@ flow_verbs_translate(struct rte_eth_dev *dev, + case RTE_FLOW_ACTION_TYPE_FLAG: + flow_verbs_translate_action_flag(dev_flow, actions); + action_flags |= MLX5_FLOW_ACTION_FLAG; +- dev_flow->handle->mark = 1; ++ wks->mark = 1; + break; + case RTE_FLOW_ACTION_TYPE_MARK: + flow_verbs_translate_action_mark(dev_flow, actions); + action_flags |= MLX5_FLOW_ACTION_MARK; +- dev_flow->handle->mark = 1; ++ wks->mark = 1; + break; + case RTE_FLOW_ACTION_TYPE_DROP: + flow_verbs_translate_action_drop(dev_flow, actions); +@@ -1803,10 +1855,10 @@ flow_verbs_translate(struct rte_eth_dev *dev, + item_flags |= MLX5_FLOW_LAYER_VXLAN_GPE; + break; + case RTE_FLOW_ITEM_TYPE_GRE: +- flow_verbs_translate_item_gre(dev_flow, items, +- item_flags); ++ gre_spec = flow_verbs_reserve_gre(dev_flow); + subpriority = MLX5_TUNNEL_PRIO_GET(rss_desc); + item_flags |= MLX5_FLOW_LAYER_GRE; ++ tunnel_item = items; + break; + case RTE_FLOW_ITEM_TYPE_MPLS: + flow_verbs_translate_item_mpls(dev_flow, items, +@@ -1820,6 +1872,9 @@ flow_verbs_translate(struct rte_eth_dev *dev, + NULL, "item not supported"); + } + } ++ if (item_flags & MLX5_FLOW_LAYER_GRE) ++ flow_verbs_translate_item_gre(dev_flow, gre_spec, ++ tunnel_item, item_flags); + dev_flow->handle->layers = item_flags; + /* Other members of attr will be ignored. */ + dev_flow->verbs.attr.priority = +diff --git a/dpdk/drivers/net/mlx5/mlx5_rx.c b/dpdk/drivers/net/mlx5/mlx5_rx.c +index e8215f7381..9fcd039c22 100644 +--- a/dpdk/drivers/net/mlx5/mlx5_rx.c ++++ b/dpdk/drivers/net/mlx5/mlx5_rx.c +@@ -73,7 +73,7 @@ rx_queue_count(struct mlx5_rxq_data *rxq) + const unsigned int cqe_n = (1 << rxq->cqe_n); + const unsigned int sges_n = (1 << rxq->sges_n); + const unsigned int elts_n = (1 << rxq->elts_n); +- const unsigned int strd_n = (1 << rxq->strd_num_n); ++ const unsigned int strd_n = RTE_BIT32(rxq->log_strd_num); + const unsigned int cqe_cnt = cqe_n - 1; + unsigned int cq_ci, used; + +@@ -167,8 +167,8 @@ mlx5_rxq_info_get(struct rte_eth_dev *dev, uint16_t rx_queue_id, + qinfo->conf.offloads = dev->data->dev_conf.rxmode.offloads; + qinfo->scattered_rx = dev->data->scattered_rx; + qinfo->nb_desc = mlx5_rxq_mprq_enabled(rxq) ? +- (1 << rxq->elts_n) * (1 << rxq->strd_num_n) : +- (1 << rxq->elts_n); ++ RTE_BIT32(rxq->elts_n) * RTE_BIT32(rxq->log_strd_num) : ++ RTE_BIT32(rxq->elts_n); + } - /* Update rows for databases that still exist. - * Delete rows for databases that no longer exist. */ -- const struct ovsdb_row *row, *next_row; -- HMAP_FOR_EACH_SAFE (row, next_row, hmap_node, &database_table->rows) { -+ const struct ovsdb_row *row; -+ HMAP_FOR_EACH_SAFE (row, hmap_node, &database_table->rows) { - const char *name; - ovsdb_util_read_string_column(row, "name", &name); - struct db *db = shash_find_data(all_dbs, name); -diff --git a/ovsdb/ovsdb-tool.c b/ovsdb/ovsdb-tool.c -index d4a9e34cc4..df2e373c3c 100644 ---- a/ovsdb/ovsdb-tool.c -+++ b/ovsdb/ovsdb-tool.c -@@ -1579,15 +1579,14 @@ do_check_cluster(struct ovs_cmdl_context *ctx) - } - free(c.servers); - -- struct commit *next_commit; -- HMAP_FOR_EACH_SAFE (commit, next_commit, hmap_node, &c.commits) { -+ HMAP_FOR_EACH_SAFE (commit, hmap_node, &c.commits) { - hmap_remove(&c.commits, &commit->hmap_node); - free(commit); - } - hmap_destroy(&c.commits); - -- struct leader *leader, *next_leader; -- HMAP_FOR_EACH_SAFE (leader, next_leader, hmap_node, &c.leaders) { -+ struct leader *leader; -+ HMAP_FOR_EACH_SAFE (leader, hmap_node, &c.leaders) { - hmap_remove(&c.leaders, &leader->hmap_node); - free(leader); - } -diff --git a/ovsdb/ovsdb.c b/ovsdb/ovsdb.c -index e6d866182c..91b4a01af8 100644 ---- a/ovsdb/ovsdb.c -+++ b/ovsdb/ovsdb.c -@@ -571,8 +571,8 @@ ovsdb_replace(struct ovsdb *dst, struct ovsdb *src) - ovsdb_monitor_prereplace_db(dst); - - /* Cancel triggers. */ -- struct ovsdb_trigger *trigger, *next; -- LIST_FOR_EACH_SAFE (trigger, next, node, &dst->triggers) { -+ struct ovsdb_trigger *trigger; -+ LIST_FOR_EACH_SAFE (trigger, node, &dst->triggers) { - ovsdb_trigger_prereplace_db(trigger); - } - -diff --git a/ovsdb/query.c b/ovsdb/query.c -index de74519989..eebe564127 100644 ---- a/ovsdb/query.c -+++ b/ovsdb/query.c -@@ -40,9 +40,9 @@ ovsdb_query(struct ovsdb_table *table, const struct ovsdb_condition *cnd, - } - } else { - /* Linear scan. */ -- const struct ovsdb_row *row, *next; -+ const struct ovsdb_row *row; + /** +@@ -178,7 +178,7 @@ mlx5_rxq_info_get(struct rte_eth_dev *dev, uint16_t rx_queue_id, + * Pointer to the device structure. + * + * @param rx_queue_id +- * Rx queue identificatior. ++ * Rx queue identification. + * + * @param mode + * Pointer to the burts mode information. +@@ -354,10 +354,10 @@ mlx5_rxq_initialize(struct mlx5_rxq_data *rxq) + + scat = &((volatile struct mlx5_wqe_mprq *) + rxq->wqes)[i].dseg; +- addr = (uintptr_t)mlx5_mprq_buf_addr(buf, +- 1 << rxq->strd_num_n); +- byte_count = (1 << rxq->strd_sz_n) * +- (1 << rxq->strd_num_n); ++ addr = (uintptr_t)mlx5_mprq_buf_addr ++ (buf, RTE_BIT32(rxq->log_strd_num)); ++ byte_count = RTE_BIT32(rxq->log_strd_sz) * ++ RTE_BIT32(rxq->log_strd_num); + lkey = mlx5_rx_addr2mr(rxq, addr); + } else { + struct rte_mbuf *buf = (*rxq->elts)[i]; +@@ -383,13 +383,18 @@ mlx5_rxq_initialize(struct mlx5_rxq_data *rxq) + .ai = 0, + }; + rxq->elts_ci = mlx5_rxq_mprq_enabled(rxq) ? +- (wqe_n >> rxq->sges_n) * (1 << rxq->strd_num_n) : 0; ++ (wqe_n >> rxq->sges_n) * RTE_BIT32(rxq->log_strd_num) : 0; + /* Update doorbell counter. */ + rxq->rq_ci = wqe_n >> rxq->sges_n; + rte_io_wmb(); + *rxq->rq_db = rte_cpu_to_be_32(rxq->rq_ci); + } -- HMAP_FOR_EACH_SAFE (row, next, hmap_node, &table->rows) { -+ HMAP_FOR_EACH_SAFE (row, hmap_node, &table->rows) { - if (ovsdb_condition_match_every_clause(row, cnd) && - !output_row(row, aux)) { - break; -diff --git a/ovsdb/raft-private.c b/ovsdb/raft-private.c -index 30760233ee..e685c8103b 100644 ---- a/ovsdb/raft-private.c -+++ b/ovsdb/raft-private.c -@@ -36,7 +36,10 @@ raft_address_validate(const char *address) - return NULL; - } else if (!strncmp(address, "ssl:", 4) || !strncmp(address, "tcp:", 4)) { - struct sockaddr_storage ss; -- if (!inet_parse_active(address + 4, -1, &ss, true)) { -+ bool dns_failure = false; ++/* Must be negative. */ ++#define MLX5_ERROR_CQE_RET (-1) ++/* Must not be negative. */ ++#define MLX5_RECOVERY_ERROR_RET 0 + -+ if (!inet_parse_active(address + 4, -1, &ss, true, &dns_failure) -+ && !dns_failure) { - return ovsdb_error(NULL, "%s: syntax error in address", address); - } - return NULL; -@@ -147,8 +150,8 @@ raft_server_destroy(struct raft_server *s) - void - raft_servers_destroy(struct hmap *servers) - { -- struct raft_server *s, *next; -- HMAP_FOR_EACH_SAFE (s, next, hmap_node, servers) { -+ struct raft_server *s; -+ HMAP_FOR_EACH_SAFE (s, hmap_node, servers) { - hmap_remove(servers, &s->hmap_node); - raft_server_destroy(s); - } -diff --git a/ovsdb/raft.c b/ovsdb/raft.c -index 1a3447a8dd..856d083f21 100644 ---- a/ovsdb/raft.c -+++ b/ovsdb/raft.c -@@ -74,9 +74,12 @@ enum raft_failure_test { - FT_CRASH_BEFORE_SEND_EXEC_REQ, - FT_CRASH_AFTER_SEND_EXEC_REQ, - FT_CRASH_AFTER_RECV_APPEND_REQ_UPDATE, -+ FT_CRASH_BEFORE_SEND_SNAPSHOT_REP, - FT_DELAY_ELECTION, - FT_DONT_SEND_VOTE_REQUEST, - FT_STOP_RAFT_RPC, -+ FT_TRANSFER_LEADERSHIP, -+ FT_TRANSFER_LEADERSHIP_AFTER_SEND_APPEND_REQ, - }; - static enum raft_failure_test failure_test; + /** + * Handle a Rx error. + * The function inserts the RQ state to reset when the first error CQE is +@@ -404,7 +409,7 @@ mlx5_rxq_initialize(struct mlx5_rxq_data *rxq) + * 0 when called from non-vectorized Rx burst. + * + * @return +- * -1 in case of recovery error, otherwise the CQE status. ++ * MLX5_RECOVERY_ERROR_RET in case of recovery error, otherwise the CQE status. + */ + int + mlx5_rx_err_handle(struct mlx5_rxq_data *rxq, uint8_t vec) +@@ -412,7 +417,7 @@ mlx5_rx_err_handle(struct mlx5_rxq_data *rxq, uint8_t vec) + const uint16_t cqe_n = 1 << rxq->cqe_n; + const uint16_t cqe_mask = cqe_n - 1; + const uint16_t wqe_n = 1 << rxq->elts_n; +- const uint16_t strd_n = 1 << rxq->strd_num_n; ++ const uint16_t strd_n = RTE_BIT32(rxq->log_strd_num); + struct mlx5_rxq_ctrl *rxq_ctrl = + container_of(rxq, struct mlx5_rxq_ctrl, rxq); + union { +@@ -433,7 +438,7 @@ mlx5_rx_err_handle(struct mlx5_rxq_data *rxq, uint8_t vec) + sm.queue_id = rxq->idx; + sm.state = IBV_WQS_RESET; + if (mlx5_queue_state_modify(RXQ_DEV(rxq_ctrl), &sm)) +- return -1; ++ return MLX5_RECOVERY_ERROR_RET; + if (rxq_ctrl->dump_file_n < + RXQ_PORT(rxq_ctrl)->config.max_dump_files_num) { + MKSTR(err_str, "Unexpected CQE error syndrome " +@@ -473,7 +478,7 @@ mlx5_rx_err_handle(struct mlx5_rxq_data *rxq, uint8_t vec) + sm.queue_id = rxq->idx; + sm.state = IBV_WQS_RDY; + if (mlx5_queue_state_modify(RXQ_DEV(rxq_ctrl), &sm)) +- return -1; ++ return MLX5_RECOVERY_ERROR_RET; + if (vec) { + const uint32_t elts_n = + mlx5_rxq_mprq_enabled(rxq) ? +@@ -501,7 +506,7 @@ mlx5_rx_err_handle(struct mlx5_rxq_data *rxq, uint8_t vec) + rte_pktmbuf_free_seg + (*elt); + } +- return -1; ++ return MLX5_RECOVERY_ERROR_RET; + } + } + for (i = 0; i < (int)elts_n; ++i) { +@@ -520,7 +525,7 @@ mlx5_rx_err_handle(struct mlx5_rxq_data *rxq, uint8_t vec) + } + return ret; + default: +- return -1; ++ return MLX5_RECOVERY_ERROR_RET; + } + } -@@ -379,12 +382,19 @@ static bool raft_handle_write_error(struct raft *, struct ovsdb_error *); - static void raft_run_reconfigure(struct raft *); +@@ -538,7 +543,9 @@ mlx5_rx_err_handle(struct mlx5_rxq_data *rxq, uint8_t vec) + * written. + * + * @return +- * 0 in case of empty CQE, otherwise the packet size in bytes. ++ * 0 in case of empty CQE, MLX5_ERROR_CQE_RET in case of error CQE, ++ * otherwise the packet size in regular RxQ, and striding byte ++ * count format in mprq case. + */ + static inline int + mlx5_rx_poll_len(struct mlx5_rxq_data *rxq, volatile struct mlx5_cqe *cqe, +@@ -605,8 +612,8 @@ mlx5_rx_poll_len(struct mlx5_rxq_data *rxq, volatile struct mlx5_cqe *cqe, + rxq->err_state)) { + ret = mlx5_rx_err_handle(rxq, 0); + if (ret == MLX5_CQE_STATUS_HW_OWN || +- ret == -1) +- return 0; ++ ret == MLX5_RECOVERY_ERROR_RET) ++ return MLX5_ERROR_CQE_RET; + } else { + return 0; + } +@@ -851,8 +858,10 @@ mlx5_rx_burst(void *dpdk_rxq, struct rte_mbuf **pkts, uint16_t pkts_n) + if (!pkt) { + cqe = &(*rxq->cqes)[rxq->cq_ci & cqe_cnt]; + len = mlx5_rx_poll_len(rxq, cqe, cqe_cnt, &mcqe); +- if (!len) { ++ if (len <= 0) { + rte_mbuf_raw_free(rep); ++ if (unlikely(len == MLX5_ERROR_CQE_RET)) ++ rq_ci = rxq->rq_ci << sges_n; + break; + } + pkt = seg; +@@ -1045,8 +1054,8 @@ uint16_t + mlx5_rx_burst_mprq(void *dpdk_rxq, struct rte_mbuf **pkts, uint16_t pkts_n) + { + struct mlx5_rxq_data *rxq = dpdk_rxq; +- const uint32_t strd_n = 1 << rxq->strd_num_n; +- const uint32_t strd_sz = 1 << rxq->strd_sz_n; ++ const uint32_t strd_n = RTE_BIT32(rxq->log_strd_num); ++ const uint32_t strd_sz = RTE_BIT32(rxq->log_strd_sz); + const uint32_t cq_mask = (1 << rxq->cqe_n) - 1; + const uint32_t wq_mask = (1 << rxq->elts_n) - 1; + volatile struct mlx5_cqe *cqe = &(*rxq->cqes)[rxq->cq_ci & cq_mask]; +@@ -1075,8 +1084,13 @@ mlx5_rx_burst_mprq(void *dpdk_rxq, struct rte_mbuf **pkts, uint16_t pkts_n) + } + cqe = &(*rxq->cqes)[rxq->cq_ci & cq_mask]; + ret = mlx5_rx_poll_len(rxq, cqe, cq_mask, &mcqe); +- if (!ret) ++ if (ret == 0) ++ break; ++ if (unlikely(ret == MLX5_ERROR_CQE_RET)) { ++ rq_ci = rxq->rq_ci; ++ consumed_strd = rxq->consumed_strd; + break; ++ } + byte_cnt = ret; + len = (byte_cnt & MLX5_MPRQ_LEN_MASK) >> MLX5_MPRQ_LEN_SHIFT; + MLX5_ASSERT((int)len >= (rxq->crc_present << 2)); +diff --git a/dpdk/drivers/net/mlx5/mlx5_rx.h b/dpdk/drivers/net/mlx5/mlx5_rx.h +index f808bf288f..423d80e4a7 100644 +--- a/dpdk/drivers/net/mlx5/mlx5_rx.h ++++ b/dpdk/drivers/net/mlx5/mlx5_rx.h +@@ -88,8 +88,8 @@ struct mlx5_rxq_data { + unsigned int elts_n:4; /* Log 2 of Mbufs. */ + unsigned int rss_hash:1; /* RSS hash result is enabled. */ + unsigned int mark:1; /* Marked flow available on the queue. */ +- unsigned int strd_num_n:5; /* Log 2 of the number of stride. */ +- unsigned int strd_sz_n:4; /* Log 2 of stride size. */ ++ unsigned int log_strd_num:5; /* Log 2 of the number of stride. */ ++ unsigned int log_strd_sz:4; /* Log 2 of stride size. */ + unsigned int strd_shift_en:1; /* Enable 2bytes shift on a stride. */ + unsigned int err_state:2; /* enum mlx5_rxq_err_state. */ + unsigned int strd_scatter_en:1; /* Scattered packets from a stride. */ +@@ -125,6 +125,7 @@ struct mlx5_rxq_data { + struct mlx5_dev_ctx_shared *sh; /* Shared context. */ + uint16_t idx; /* Queue index. */ + struct mlx5_rxq_stats stats; ++ struct mlx5_rxq_stats stats_reset; /* stats on last reset. */ + rte_xmm_t mbuf_initializer; /* Default rearm/flags for vectorized Rx. */ + struct rte_mbuf fake_mbuf; /* elts padding for vectorized Rx. */ + struct mlx5_uar_data uar_data; /* CQ doorbell. */ +@@ -161,7 +162,6 @@ struct mlx5_rxq_ctrl { + uint16_t share_qid; /* Shared RxQ ID in group. */ + unsigned int started:1; /* Whether (shared) RXQ has been started. */ + unsigned int irq:1; /* Whether IRQ is enabled. */ +- uint32_t flow_mark_n; /* Number of Mark/Flag flows using this Queue. */ + uint32_t flow_tunnels_n[MLX5_FLOW_TUNNEL]; /* Tunnels counters. */ + uint32_t wqn; /* WQ number. */ + uint32_t rxseg_n; /* Number of split segment descriptions. */ +@@ -401,7 +401,7 @@ mlx5_timestamp_set(struct rte_mbuf *mbuf, int offset, + static __rte_always_inline void + mprq_buf_replace(struct mlx5_rxq_data *rxq, uint16_t rq_idx) + { +- const uint32_t strd_n = 1 << rxq->strd_num_n; ++ const uint32_t strd_n = RTE_BIT32(rxq->log_strd_num); + struct mlx5_mprq_buf *rep = rxq->mprq_repl; + volatile struct mlx5_wqe_data_seg *wqe = + &((volatile struct mlx5_wqe_mprq *)rxq->wqes)[rq_idx].dseg; +@@ -459,8 +459,8 @@ static __rte_always_inline enum mlx5_rqx_code + mprq_buf_to_pkt(struct mlx5_rxq_data *rxq, struct rte_mbuf *pkt, uint32_t len, + struct mlx5_mprq_buf *buf, uint16_t strd_idx, uint16_t strd_cnt) + { +- const uint32_t strd_n = 1 << rxq->strd_num_n; +- const uint16_t strd_sz = 1 << rxq->strd_sz_n; ++ const uint32_t strd_n = RTE_BIT32(rxq->log_strd_num); ++ const uint16_t strd_sz = RTE_BIT32(rxq->log_strd_sz); + const uint16_t strd_shift = + MLX5_MPRQ_STRIDE_SHIFT_BYTE * rxq->strd_shift_en; + const int32_t hdrm_overlap = +@@ -543,7 +543,7 @@ mprq_buf_to_pkt(struct mlx5_rxq_data *rxq, struct rte_mbuf *pkt, uint32_t len, + buf_len, shinfo); + /* Set mbuf head-room. */ + SET_DATA_OFF(pkt, RTE_PKTMBUF_HEADROOM); +- MLX5_ASSERT(pkt->ol_flags == RTE_MBUF_F_EXTERNAL); ++ MLX5_ASSERT(pkt->ol_flags & RTE_MBUF_F_EXTERNAL); + MLX5_ASSERT(rte_pktmbuf_tailroom(pkt) >= + len - (hdrm_overlap > 0 ? hdrm_overlap : 0)); + DATA_LEN(pkt) = len; +@@ -605,7 +605,7 @@ mlx5_check_mprq_support(struct rte_eth_dev *dev) + static __rte_always_inline int + mlx5_rxq_mprq_enabled(struct mlx5_rxq_data *rxq) + { +- return rxq->strd_num_n > 0; ++ return rxq->log_strd_num > 0; + } - static void raft_set_leader(struct raft *, const struct uuid *sid); -+ - static struct raft_server * - raft_find_server(const struct raft *raft, const struct uuid *sid) + /** +diff --git a/dpdk/drivers/net/mlx5/mlx5_rxq.c b/dpdk/drivers/net/mlx5/mlx5_rxq.c +index f77d42dedf..807aaf2fc9 100644 +--- a/dpdk/drivers/net/mlx5/mlx5_rxq.c ++++ b/dpdk/drivers/net/mlx5/mlx5_rxq.c +@@ -67,7 +67,7 @@ mlx5_rxq_cqe_num(struct mlx5_rxq_data *rxq_data) + unsigned int wqe_n = 1 << rxq_data->elts_n; + + if (mlx5_rxq_mprq_enabled(rxq_data)) +- cqe_n = wqe_n * (1 << rxq_data->strd_num_n) - 1; ++ cqe_n = wqe_n * RTE_BIT32(rxq_data->log_strd_num) - 1; + else + cqe_n = wqe_n - 1; + return cqe_n; +@@ -137,8 +137,10 @@ rxq_alloc_elts_sprq(struct mlx5_rxq_ctrl *rxq_ctrl) { - return raft_server_find(&raft->servers, sid); + const unsigned int sges_n = 1 << rxq_ctrl->rxq.sges_n; + unsigned int elts_n = mlx5_rxq_mprq_enabled(&rxq_ctrl->rxq) ? +- (1 << rxq_ctrl->rxq.elts_n) * (1 << rxq_ctrl->rxq.strd_num_n) : +- (1 << rxq_ctrl->rxq.elts_n); ++ RTE_BIT32(rxq_ctrl->rxq.elts_n) * ++ RTE_BIT32(rxq_ctrl->rxq.log_strd_num) : ++ RTE_BIT32(rxq_ctrl->rxq.elts_n); ++ bool has_vec_support = mlx5_rxq_check_vec_support(&rxq_ctrl->rxq) > 0; + unsigned int i; + int err; + +@@ -160,8 +162,9 @@ rxq_alloc_elts_sprq(struct mlx5_rxq_ctrl *rxq_ctrl) + rte_errno = ENOMEM; + goto error; + } +- /* Headroom is reserved by rte_pktmbuf_alloc(). */ +- MLX5_ASSERT(DATA_OFF(buf) == RTE_PKTMBUF_HEADROOM); ++ /* Only vectored Rx routines rely on headroom size. */ ++ MLX5_ASSERT(!has_vec_support || ++ DATA_OFF(buf) >= RTE_PKTMBUF_HEADROOM); + /* Buffer is supposed to be empty. */ + MLX5_ASSERT(rte_pktmbuf_data_len(buf) == 0); + MLX5_ASSERT(rte_pktmbuf_pkt_len(buf) == 0); +@@ -174,7 +177,7 @@ rxq_alloc_elts_sprq(struct mlx5_rxq_ctrl *rxq_ctrl) + (*rxq_ctrl->rxq.elts)[i] = buf; + } + /* If Rx vector is activated. */ +- if (mlx5_rxq_check_vec_support(&rxq_ctrl->rxq) > 0) { ++ if (has_vec_support) { + struct mlx5_rxq_data *rxq = &rxq_ctrl->rxq; + struct rte_mbuf *mbuf_init = &rxq->fake_mbuf; + struct rte_pktmbuf_pool_private *priv = +@@ -293,8 +296,8 @@ rxq_free_elts_sprq(struct mlx5_rxq_ctrl *rxq_ctrl) + { + struct mlx5_rxq_data *rxq = &rxq_ctrl->rxq; + const uint16_t q_n = mlx5_rxq_mprq_enabled(&rxq_ctrl->rxq) ? +- (1 << rxq->elts_n) * (1 << rxq->strd_num_n) : +- (1 << rxq->elts_n); ++ RTE_BIT32(rxq->elts_n) * RTE_BIT32(rxq->log_strd_num) : ++ RTE_BIT32(rxq->elts_n); + const uint16_t q_mask = q_n - 1; + uint16_t elts_ci = mlx5_rxq_mprq_enabled(&rxq_ctrl->rxq) ? + rxq->elts_ci : rxq->rq_ci; +@@ -838,6 +841,14 @@ mlx5_rx_queue_setup(struct rte_eth_dev *dev, uint16_t idx, uint16_t desc, + uint64_t offloads = conf->offloads | + dev->data->dev_conf.rxmode.offloads; + ++ if ((offloads & RTE_ETH_RX_OFFLOAD_TCP_LRO) && ++ !priv->config.lro.supported) { ++ DRV_LOG(ERR, ++ "Port %u queue %u LRO is configured but not supported.", ++ dev->data->port_id, idx); ++ rte_errno = EINVAL; ++ return -rte_errno; ++ } + if (mp) { + /* + * The parameters should be checked on rte_eth_dev layer. +@@ -1378,8 +1389,8 @@ mlx5_mprq_alloc_mp(struct rte_eth_dev *dev) + unsigned int buf_len; + unsigned int obj_num; + unsigned int obj_size; +- unsigned int strd_num_n = 0; +- unsigned int strd_sz_n = 0; ++ unsigned int log_strd_num = 0; ++ unsigned int log_strd_sz = 0; + unsigned int i; + unsigned int n_ibv = 0; + int ret; +@@ -1398,16 +1409,18 @@ mlx5_mprq_alloc_mp(struct rte_eth_dev *dev) + n_ibv++; + desc += 1 << rxq->elts_n; + /* Get the max number of strides. */ +- if (strd_num_n < rxq->strd_num_n) +- strd_num_n = rxq->strd_num_n; ++ if (log_strd_num < rxq->log_strd_num) ++ log_strd_num = rxq->log_strd_num; + /* Get the max size of a stride. */ +- if (strd_sz_n < rxq->strd_sz_n) +- strd_sz_n = rxq->strd_sz_n; +- } +- MLX5_ASSERT(strd_num_n && strd_sz_n); +- buf_len = (1 << strd_num_n) * (1 << strd_sz_n); +- obj_size = sizeof(struct mlx5_mprq_buf) + buf_len + (1 << strd_num_n) * +- sizeof(struct rte_mbuf_ext_shared_info) + RTE_PKTMBUF_HEADROOM; ++ if (log_strd_sz < rxq->log_strd_sz) ++ log_strd_sz = rxq->log_strd_sz; ++ } ++ MLX5_ASSERT(log_strd_num && log_strd_sz); ++ buf_len = RTE_BIT32(log_strd_num) * RTE_BIT32(log_strd_sz); ++ obj_size = sizeof(struct mlx5_mprq_buf) + buf_len + ++ RTE_BIT32(log_strd_num) * ++ sizeof(struct rte_mbuf_ext_shared_info) + ++ RTE_PKTMBUF_HEADROOM; + /* + * Received packets can be either memcpy'd or externally referenced. In + * case that the packet is attached to an mbuf as an external buffer, as +@@ -1453,7 +1466,7 @@ mlx5_mprq_alloc_mp(struct rte_eth_dev *dev) + snprintf(name, sizeof(name), "port-%u-mprq", dev->data->port_id); + mp = rte_mempool_create(name, obj_num, obj_size, MLX5_MPRQ_MP_CACHE_SZ, + 0, NULL, NULL, mlx5_mprq_buf_init, +- (void *)((uintptr_t)1 << strd_num_n), ++ (void *)((uintptr_t)1 << log_strd_num), + dev->device->numa_node, 0); + if (mp == NULL) { + DRV_LOG(ERR, +@@ -1530,6 +1543,126 @@ mlx5_max_lro_msg_size_adjust(struct rte_eth_dev *dev, uint16_t idx, + priv->max_lro_msg_size * MLX5_LRO_SEG_CHUNK_SIZE); } -+static struct raft_server * -+raft_find_new_server(struct raft *raft, const struct uuid *uuid) ++/** ++ * Prepare both size and number of stride for Multi-Packet RQ. ++ * ++ * @param dev ++ * Pointer to Ethernet device. ++ * @param idx ++ * RX queue index. ++ * @param desc ++ * Number of descriptors to configure in queue. ++ * @param rx_seg_en ++ * Indicator if Rx segment enables, if so Multi-Packet RQ doesn't enable. ++ * @param min_mbuf_size ++ * Non scatter min mbuf size, max_rx_pktlen plus overhead. ++ * @param actual_log_stride_num ++ * Log number of strides to configure for this queue. ++ * @param actual_log_stride_size ++ * Log stride size to configure for this queue. ++ * ++ * @return ++ * 0 if Multi-Packet RQ is supported, otherwise -1. ++ */ ++static int ++mlx5_mprq_prepare(struct rte_eth_dev *dev, uint16_t idx, uint16_t desc, ++ bool rx_seg_en, uint32_t min_mbuf_size, ++ uint32_t *actual_log_stride_num, ++ uint32_t *actual_log_stride_size) +{ -+ return raft_server_find(&raft->add_servers, uuid); ++ struct mlx5_priv *priv = dev->data->dev_private; ++ struct mlx5_dev_config *config = &priv->config; ++ uint32_t log_min_stride_num = config->mprq.log_min_stride_num; ++ uint32_t log_max_stride_num = config->mprq.log_max_stride_num; ++ uint32_t log_def_stride_num = ++ RTE_MIN(RTE_MAX(MLX5_MPRQ_DEFAULT_LOG_STRIDE_NUM, ++ log_min_stride_num), ++ log_max_stride_num); ++ uint32_t log_min_stride_size = config->mprq.log_min_stride_size; ++ uint32_t log_max_stride_size = config->mprq.log_max_stride_size; ++ uint32_t log_def_stride_size = ++ RTE_MIN(RTE_MAX(MLX5_MPRQ_DEFAULT_LOG_STRIDE_SIZE, ++ log_min_stride_size), ++ log_max_stride_size); ++ uint32_t log_stride_wqe_size; ++ ++ if (mlx5_check_mprq_support(dev) != 1 || rx_seg_en) ++ goto unsupport; ++ /* Checks if chosen number of strides is in supported range. */ ++ if (config->mprq.log_stride_num > log_max_stride_num || ++ config->mprq.log_stride_num < log_min_stride_num) { ++ *actual_log_stride_num = log_def_stride_num; ++ DRV_LOG(WARNING, ++ "Port %u Rx queue %u number of strides for Multi-Packet RQ is out of range, setting default value (%u)", ++ dev->data->port_id, idx, RTE_BIT32(log_def_stride_num)); ++ } else { ++ *actual_log_stride_num = config->mprq.log_stride_num; ++ } ++ if (config->mprq.log_stride_size) { ++ /* Checks if chosen size of stride is in supported range. */ ++ if (config->mprq.log_stride_size > log_max_stride_size || ++ config->mprq.log_stride_size < log_min_stride_size) { ++ *actual_log_stride_size = log_def_stride_size; ++ DRV_LOG(WARNING, ++ "Port %u Rx queue %u size of a stride for Multi-Packet RQ is out of range, setting default value (%u)", ++ dev->data->port_id, idx, ++ RTE_BIT32(log_def_stride_size)); ++ } else { ++ *actual_log_stride_size = config->mprq.log_stride_size; ++ } ++ } else { ++ if (min_mbuf_size <= RTE_BIT32(log_max_stride_size)) ++ *actual_log_stride_size = log2above(min_mbuf_size); ++ else ++ goto unsupport; ++ } ++ log_stride_wqe_size = *actual_log_stride_num + *actual_log_stride_size; ++ /* Check if WQE buffer size is supported by hardware. */ ++ if (log_stride_wqe_size < config->mprq.log_min_stride_wqe_size) { ++ *actual_log_stride_num = log_def_stride_num; ++ *actual_log_stride_size = log_def_stride_size; ++ DRV_LOG(WARNING, ++ "Port %u Rx queue %u size of WQE buffer for Multi-Packet RQ is too small, setting default values (stride_num_n=%u, stride_size_n=%u)", ++ dev->data->port_id, idx, RTE_BIT32(log_def_stride_num), ++ RTE_BIT32(log_def_stride_size)); ++ log_stride_wqe_size = log_def_stride_num + log_def_stride_size; ++ } ++ MLX5_ASSERT(log_stride_wqe_size >= config->mprq.log_min_stride_wqe_size); ++ if (desc <= RTE_BIT32(*actual_log_stride_num)) ++ goto unsupport; ++ if (min_mbuf_size > RTE_BIT32(log_stride_wqe_size)) { ++ DRV_LOG(WARNING, "Port %u Rx queue %u " ++ "Multi-Packet RQ is unsupported, WQE buffer size (%u) " ++ "is smaller than min mbuf size (%u)", ++ dev->data->port_id, idx, RTE_BIT32(log_stride_wqe_size), ++ min_mbuf_size); ++ goto unsupport; ++ } ++ DRV_LOG(DEBUG, "Port %u Rx queue %u " ++ "Multi-Packet RQ is enabled strd_num_n = %u, strd_sz_n = %u", ++ dev->data->port_id, idx, RTE_BIT32(*actual_log_stride_num), ++ RTE_BIT32(*actual_log_stride_size)); ++ return 0; ++unsupport: ++ if (config->mprq.enabled) ++ DRV_LOG(WARNING, ++ "Port %u MPRQ is requested but cannot be enabled\n" ++ " (requested: pkt_sz = %u, desc_num = %u," ++ " rxq_num = %u, stride_sz = %u, stride_num = %u\n" ++ " supported: min_rxqs_num = %u, min_buf_wqe_sz = %u" ++ " min_stride_sz = %u, max_stride_sz = %u).\n" ++ "Rx segment is %senable.", ++ dev->data->port_id, min_mbuf_size, desc, priv->rxqs_n, ++ RTE_BIT32(config->mprq.log_stride_size), ++ RTE_BIT32(config->mprq.log_stride_num), ++ config->mprq.min_rxqs_num, ++ RTE_BIT32(config->mprq.log_min_stride_wqe_size), ++ RTE_BIT32(config->mprq.log_min_stride_size), ++ RTE_BIT32(config->mprq.log_max_stride_size), ++ rx_seg_en ? "" : "not "); ++ return -1; +} + - static char * - raft_make_address_passive(const char *address_) + /** + * Create a DPDK Rx queue. + * +@@ -1567,41 +1700,33 @@ mlx5_rxq_new(struct rte_eth_dev *dev, struct mlx5_rxq_priv *rxq, + RTE_PKTMBUF_HEADROOM; + unsigned int max_lro_size = 0; + unsigned int first_mb_free_size = mb_len - RTE_PKTMBUF_HEADROOM; +- const int mprq_en = mlx5_check_mprq_support(dev) > 0 && n_seg == 1 && +- !rx_seg[0].offset && !rx_seg[0].length; +- unsigned int mprq_stride_nums = config->mprq.stride_num_n ? +- config->mprq.stride_num_n : MLX5_MPRQ_STRIDE_NUM_N; +- unsigned int mprq_stride_size = non_scatter_min_mbuf_size <= +- (1U << config->mprq.max_stride_size_n) ? +- log2above(non_scatter_min_mbuf_size) : MLX5_MPRQ_STRIDE_SIZE_N; +- unsigned int mprq_stride_cap = (config->mprq.stride_num_n ? +- (1U << config->mprq.stride_num_n) : (1U << mprq_stride_nums)) * +- (config->mprq.stride_size_n ? +- (1U << config->mprq.stride_size_n) : (1U << mprq_stride_size)); ++ uint32_t mprq_log_actual_stride_num = 0; ++ uint32_t mprq_log_actual_stride_size = 0; ++ bool rx_seg_en = n_seg != 1 || rx_seg[0].offset || rx_seg[0].length; ++ const int mprq_en = !mlx5_mprq_prepare(dev, idx, desc, rx_seg_en, ++ non_scatter_min_mbuf_size, ++ &mprq_log_actual_stride_num, ++ &mprq_log_actual_stride_size); + /* + * Always allocate extra slots, even if eventually + * the vector Rx will not be used. + */ + uint16_t desc_n = desc + config->rx_vec_en * MLX5_VPMD_DESCS_PER_LOOP; ++ size_t alloc_size = sizeof(*tmpl) + desc_n * sizeof(struct rte_mbuf *); + const struct rte_eth_rxseg_split *qs_seg = rx_seg; + unsigned int tail_len; + +- tmpl = mlx5_malloc(MLX5_MEM_RTE | MLX5_MEM_ZERO, +- sizeof(*tmpl) + desc_n * sizeof(struct rte_mbuf *) + +- (!!mprq_en) * +- (desc >> mprq_stride_nums) * sizeof(struct mlx5_mprq_buf *), +- 0, socket); ++ if (mprq_en) { ++ /* Trim the number of descs needed. */ ++ desc >>= mprq_log_actual_stride_num; ++ alloc_size += desc * sizeof(struct mlx5_mprq_buf *); ++ } ++ tmpl = mlx5_malloc(MLX5_MEM_RTE | MLX5_MEM_ZERO, alloc_size, 0, socket); + if (!tmpl) { + rte_errno = ENOMEM; + return NULL; + } + LIST_INIT(&tmpl->owners); +- if (conf->share_group > 0) { +- tmpl->rxq.shared = 1; +- tmpl->share_group = conf->share_group; +- tmpl->share_qid = conf->share_qid; +- LIST_INSERT_HEAD(&priv->sh->shared_rxqs, tmpl, share_entry); +- } + rxq->ctrl = tmpl; + LIST_INSERT_HEAD(&tmpl->owners, rxq, owner_entry); + MLX5_ASSERT(n_seg && n_seg <= MLX5_MAX_RXQ_NSEG); +@@ -1695,43 +1820,19 @@ mlx5_rxq_new(struct rte_eth_dev *dev, struct mlx5_rxq_priv *rxq, + tmpl->socket = socket; + if (dev->data->dev_conf.intr_conf.rxq) + tmpl->irq = 1; +- /* +- * This Rx queue can be configured as a Multi-Packet RQ if all of the +- * following conditions are met: +- * - MPRQ is enabled. +- * - The number of descs is more than the number of strides. +- * - max_rx_pktlen plus overhead is less than the max size +- * of a stride or mprq_stride_size is specified by a user. +- * Need to make sure that there are enough strides to encap +- * the maximum packet size in case mprq_stride_size is set. +- * Otherwise, enable Rx scatter if necessary. +- */ +- if (mprq_en && desc > (1U << mprq_stride_nums) && +- (non_scatter_min_mbuf_size <= +- (1U << config->mprq.max_stride_size_n) || +- (config->mprq.stride_size_n && +- non_scatter_min_mbuf_size <= mprq_stride_cap))) { ++ if (mprq_en) { + /* TODO: Rx scatter isn't supported yet. */ + tmpl->rxq.sges_n = 0; +- /* Trim the number of descs needed. */ +- desc >>= mprq_stride_nums; +- tmpl->rxq.strd_num_n = config->mprq.stride_num_n ? +- config->mprq.stride_num_n : mprq_stride_nums; +- tmpl->rxq.strd_sz_n = config->mprq.stride_size_n ? +- config->mprq.stride_size_n : mprq_stride_size; ++ tmpl->rxq.log_strd_num = mprq_log_actual_stride_num; ++ tmpl->rxq.log_strd_sz = mprq_log_actual_stride_size; + tmpl->rxq.strd_shift_en = MLX5_MPRQ_TWO_BYTE_SHIFT; + tmpl->rxq.strd_scatter_en = + !!(offloads & RTE_ETH_RX_OFFLOAD_SCATTER); + tmpl->rxq.mprq_max_memcpy_len = RTE_MIN(first_mb_free_size, + config->mprq.max_memcpy_len); + max_lro_size = RTE_MIN(max_rx_pktlen, +- (1u << tmpl->rxq.strd_num_n) * +- (1u << tmpl->rxq.strd_sz_n)); +- DRV_LOG(DEBUG, +- "port %u Rx queue %u: Multi-Packet RQ is enabled" +- " strd_num_n = %u, strd_sz_n = %u", +- dev->data->port_id, idx, +- tmpl->rxq.strd_num_n, tmpl->rxq.strd_sz_n); ++ RTE_BIT32(tmpl->rxq.log_strd_num) * ++ RTE_BIT32(tmpl->rxq.log_strd_sz)); + } else if (tmpl->rxq.rxseg_n == 1) { + MLX5_ASSERT(max_rx_pktlen <= first_mb_free_size); + tmpl->rxq.sges_n = 0; +@@ -1765,24 +1866,6 @@ mlx5_rxq_new(struct rte_eth_dev *dev, struct mlx5_rxq_priv *rxq, + tmpl->rxq.sges_n = sges_n; + max_lro_size = max_rx_pktlen; + } +- if (config->mprq.enabled && !mlx5_rxq_mprq_enabled(&tmpl->rxq)) +- DRV_LOG(WARNING, +- "port %u MPRQ is requested but cannot be enabled\n" +- " (requested: pkt_sz = %u, desc_num = %u," +- " rxq_num = %u, stride_sz = %u, stride_num = %u\n" +- " supported: min_rxqs_num = %u," +- " min_stride_sz = %u, max_stride_sz = %u).", +- dev->data->port_id, non_scatter_min_mbuf_size, +- desc, priv->rxqs_n, +- config->mprq.stride_size_n ? +- (1U << config->mprq.stride_size_n) : +- (1U << mprq_stride_size), +- config->mprq.stride_num_n ? +- (1U << config->mprq.stride_num_n) : +- (1U << mprq_stride_nums), +- config->mprq.min_rxqs_num, +- (1U << config->mprq.min_stride_size_n), +- (1U << config->mprq.max_stride_size_n)); + DRV_LOG(DEBUG, "port %u maximum number of segments per packet: %u", + dev->data->port_id, 1 << tmpl->rxq.sges_n); + if (desc % (1 << tmpl->rxq.sges_n)) { +@@ -1840,20 +1923,24 @@ mlx5_rxq_new(struct rte_eth_dev *dev, struct mlx5_rxq_priv *rxq, + dev->data->port_id, + tmpl->rxq.crc_present ? "disabled" : "enabled", + tmpl->rxq.crc_present << 2); +- /* Save port ID. */ + tmpl->rxq.rss_hash = !!priv->rss_conf.rss_hf && + (!!(dev->data->dev_conf.rxmode.mq_mode & RTE_ETH_MQ_RX_RSS)); ++ /* Save port ID. */ + tmpl->rxq.port_id = dev->data->port_id; + tmpl->sh = priv->sh; + tmpl->rxq.mp = rx_seg[0].mp; + tmpl->rxq.elts_n = log2above(desc); +- tmpl->rxq.rq_repl_thresh = +- MLX5_VPMD_RXQ_RPLNSH_THRESH(desc_n); +- tmpl->rxq.elts = +- (struct rte_mbuf *(*)[desc_n])(tmpl + 1); ++ tmpl->rxq.rq_repl_thresh = MLX5_VPMD_RXQ_RPLNSH_THRESH(desc_n); ++ tmpl->rxq.elts = (struct rte_mbuf *(*)[desc_n])(tmpl + 1); + tmpl->rxq.mprq_bufs = + (struct mlx5_mprq_buf *(*)[desc])(*tmpl->rxq.elts + desc_n); + tmpl->rxq.idx = idx; ++ if (conf->share_group > 0) { ++ tmpl->rxq.shared = 1; ++ tmpl->share_group = conf->share_group; ++ tmpl->share_qid = conf->share_qid; ++ LIST_INSERT_HEAD(&priv->sh->shared_rxqs, tmpl, share_entry); ++ } + LIST_INSERT_HEAD(&priv->rxqsctrl, tmpl, next); + return tmpl; + error: +@@ -1969,6 +2056,8 @@ mlx5_rxq_get(struct rte_eth_dev *dev, uint16_t idx) { -@@ -692,8 +702,8 @@ static void - raft_set_servers(struct raft *raft, const struct hmap *new_servers, - enum vlog_level level) + struct mlx5_priv *priv = dev->data->dev_private; + ++ if (idx >= priv->rxqs_n) ++ return NULL; + MLX5_ASSERT(priv->rxq_privs != NULL); + return (*priv->rxq_privs)[idx]; + } +@@ -2152,7 +2241,7 @@ mlx5_rxq_get_hairpin_conf(struct rte_eth_dev *dev, uint16_t idx) + * Number of queues in the array. + * + * @return +- * 1 if all queues in indirection table match 0 othrwise. ++ * 1 if all queues in indirection table match 0 otherwise. + */ + static int + mlx5_ind_table_obj_match_queues(const struct mlx5_ind_table_obj *ind_tbl, +@@ -2586,7 +2675,7 @@ mlx5_hrxq_modify(struct rte_eth_dev *dev, uint32_t hrxq_idx, + if (hrxq->standalone) { + /* + * Replacement of indirection table unsupported for +- * stanalone hrxq objects (used by shared RSS). ++ * standalone hrxq objects (used by shared RSS). + */ + rte_errno = ENOTSUP; + return -rte_errno; +@@ -2828,7 +2917,7 @@ mlx5_drop_action_create(struct rte_eth_dev *dev) + + if (priv->drop_queue.hrxq) + return priv->drop_queue.hrxq; +- hrxq = mlx5_malloc(MLX5_MEM_ZERO, sizeof(*hrxq), 0, SOCKET_ID_ANY); ++ hrxq = mlx5_malloc(MLX5_MEM_ZERO, sizeof(*hrxq) + MLX5_RSS_HASH_KEY_LEN, 0, SOCKET_ID_ANY); + if (!hrxq) { + DRV_LOG(WARNING, + "Port %u cannot allocate memory for drop queue.", +diff --git a/dpdk/drivers/net/mlx5/mlx5_rxtx_vec.c b/dpdk/drivers/net/mlx5/mlx5_rxtx_vec.c +index 6212ce8247..0e2eab068a 100644 +--- a/dpdk/drivers/net/mlx5/mlx5_rxtx_vec.c ++++ b/dpdk/drivers/net/mlx5/mlx5_rxtx_vec.c +@@ -148,7 +148,7 @@ static inline void + mlx5_rx_mprq_replenish_bulk_mbuf(struct mlx5_rxq_data *rxq) { -- struct raft_server *s, *next; -- HMAP_FOR_EACH_SAFE (s, next, hmap_node, &raft->servers) { -+ struct raft_server *s; -+ HMAP_FOR_EACH_SAFE (s, hmap_node, &raft->servers) { - if (!raft_server_find(new_servers, &s->sid)) { - ovs_assert(s != raft->remove_server); + const uint16_t wqe_n = 1 << rxq->elts_n; +- const uint32_t strd_n = 1 << rxq->strd_num_n; ++ const uint32_t strd_n = RTE_BIT32(rxq->log_strd_num); + const uint32_t elts_n = wqe_n * strd_n; + const uint32_t wqe_mask = elts_n - 1; + uint32_t n = elts_n - (rxq->elts_ci - rxq->rq_pi); +@@ -197,8 +197,8 @@ rxq_copy_mprq_mbuf_v(struct mlx5_rxq_data *rxq, + { + const uint16_t wqe_n = 1 << rxq->elts_n; + const uint16_t wqe_mask = wqe_n - 1; +- const uint16_t strd_sz = 1 << rxq->strd_sz_n; +- const uint32_t strd_n = 1 << rxq->strd_num_n; ++ const uint16_t strd_sz = RTE_BIT32(rxq->log_strd_sz); ++ const uint32_t strd_n = RTE_BIT32(rxq->log_strd_num); + const uint32_t elts_n = wqe_n * strd_n; + const uint32_t elts_mask = elts_n - 1; + uint32_t elts_idx = rxq->rq_pi & elts_mask; +@@ -428,7 +428,7 @@ rxq_burst_mprq_v(struct mlx5_rxq_data *rxq, struct rte_mbuf **pkts, + const uint16_t q_n = 1 << rxq->cqe_n; + const uint16_t q_mask = q_n - 1; + const uint16_t wqe_n = 1 << rxq->elts_n; +- const uint32_t strd_n = 1 << rxq->strd_num_n; ++ const uint32_t strd_n = RTE_BIT32(rxq->log_strd_num); + const uint32_t elts_n = wqe_n * strd_n; + const uint32_t elts_mask = elts_n - 1; + volatile struct mlx5_cqe *cq; +diff --git a/dpdk/drivers/net/mlx5/mlx5_rxtx_vec_altivec.h b/dpdk/drivers/net/mlx5/mlx5_rxtx_vec_altivec.h +index 423e229508..683a8f9a6c 100644 +--- a/dpdk/drivers/net/mlx5/mlx5_rxtx_vec_altivec.h ++++ b/dpdk/drivers/net/mlx5/mlx5_rxtx_vec_altivec.h +@@ -47,11 +47,11 @@ rxq_copy_mbuf_v(struct rte_mbuf **elts, struct rte_mbuf **pkts, uint16_t n) + uint16_t p = n & -2; + + for (pos = 0; pos < p; pos += 2) { +- vector unsigned char mbp; ++ __vector unsigned char mbp; + +- mbp = (vector unsigned char)vec_vsx_ld(0, ++ mbp = (__vector unsigned char)vec_vsx_ld(0, + (signed int const *)&elts[pos]); +- *(vector unsigned char *)&pkts[pos] = mbp; ++ *(__vector unsigned char *)&pkts[pos] = mbp; + } + if (n & 1) + pkts[pos] = elts[pos]; +@@ -78,15 +78,15 @@ rxq_cq_decompress_v(struct mlx5_rxq_data *rxq, volatile struct mlx5_cqe *cq, + { + volatile struct mlx5_mini_cqe8 *mcq = (void *)&(cq + 1)->pkt_info; + struct rte_mbuf *t_pkt = elts[0]; /* Title packet is pre-built. */ +- const vector unsigned char zero = (vector unsigned char){0}; ++ const __vector unsigned char zero = (__vector unsigned char){0}; + /* Mask to shuffle from extracted mini CQE to mbuf. */ +- const vector unsigned char shuf_mask1 = (vector unsigned char){ ++ const __vector unsigned char shuf_mask1 = (__vector unsigned char){ + -1, -1, -1, -1, /* skip packet_type */ + 7, 6, -1, -1, /* bswap16, pkt_len */ + 7, 6, /* bswap16, data_len */ + -1, -1, /* skip vlan_tci */ + 3, 2, 1, 0}; /* bswap32, rss */ +- const vector unsigned char shuf_mask2 = (vector unsigned char){ ++ const __vector unsigned char shuf_mask2 = (__vector unsigned char){ + -1, -1, -1, -1, /* skip packet_type */ + 15, 14, -1, -1, /* bswap16, pkt_len */ + 15, 14, /* data_len, bswap16 */ +@@ -95,30 +95,30 @@ rxq_cq_decompress_v(struct mlx5_rxq_data *rxq, volatile struct mlx5_cqe *cq, + /* Restore the compressed count. Must be 16 bits. */ + const uint16_t mcqe_n = t_pkt->data_len + + (rxq->crc_present * RTE_ETHER_CRC_LEN); +- const vector unsigned char rearm = +- (vector unsigned char)vec_vsx_ld(0, ++ const __vector unsigned char rearm = ++ (__vector unsigned char)vec_vsx_ld(0, + (signed int const *)&t_pkt->rearm_data); +- const vector unsigned char rxdf = +- (vector unsigned char)vec_vsx_ld(0, ++ const __vector unsigned char rxdf = ++ (__vector unsigned char)vec_vsx_ld(0, + (signed int const *)&t_pkt->rx_descriptor_fields1); +- const vector unsigned char crc_adj = +- (vector unsigned char)(vector unsigned short){ ++ const __vector unsigned char crc_adj = ++ (__vector unsigned char)(__vector unsigned short){ + 0, 0, rxq->crc_present * RTE_ETHER_CRC_LEN, 0, + rxq->crc_present * RTE_ETHER_CRC_LEN, 0, 0, 0}; +- const vector unsigned short rxdf_sel_mask = +- (vector unsigned short){ ++ const __vector unsigned short rxdf_sel_mask = ++ (__vector unsigned short){ + 0xffff, 0xffff, 0, 0, 0, 0xffff, 0, 0}; +- vector unsigned char ol_flags = (vector unsigned char){0}; +- vector unsigned char ol_flags_mask = (vector unsigned char){0}; ++ __vector unsigned char ol_flags = (__vector unsigned char){0}; ++ __vector unsigned char ol_flags_mask = (__vector unsigned char){0}; + unsigned int pos; + unsigned int i; + unsigned int inv = 0; + + #ifdef MLX5_PMD_SOFT_COUNTERS +- const vector unsigned char ones = vec_splat_u8(-1); ++ const __vector unsigned char ones = vec_splat_u8(-1); + uint32_t rcvd_byte = 0; + /* Mask to shuffle byte_cnt to add up stats. Do bswap16 for all. */ +- const vector unsigned char len_shuf_mask = (vector unsigned char){ ++ const __vector unsigned char len_shuf_mask = (__vector unsigned char){ + 3, 2, 11, 10, + 7, 6, 15, 14, + -1, -1, -1, -1, +@@ -133,125 +133,125 @@ rxq_cq_decompress_v(struct mlx5_rxq_data *rxq, volatile struct mlx5_cqe *cq, + * E. store flow tag (rte_flow mark). + */ + for (pos = 0; pos < mcqe_n; ) { +- vector unsigned char mcqe1, mcqe2; +- vector unsigned char rxdf1, rxdf2; ++ __vector unsigned char mcqe1, mcqe2; ++ __vector unsigned char rxdf1, rxdf2; + #ifdef MLX5_PMD_SOFT_COUNTERS +- const vector unsigned short mcqe_sel_mask = +- (vector unsigned short){0, 0, 0xffff, 0xffff, ++ const __vector unsigned short mcqe_sel_mask = ++ (__vector unsigned short){0, 0, 0xffff, 0xffff, + 0, 0, 0xfff, 0xffff}; +- const vector unsigned char lower_half = { ++ const __vector unsigned char lower_half = { + 0, 1, 4, 5, 8, 9, 12, 13, 16, + 17, 20, 21, 24, 25, 28, 29}; +- const vector unsigned char upper_half = { ++ const __vector unsigned char upper_half = { + 2, 3, 6, 7, 10, 11, 14, 15, + 18, 19, 22, 23, 26, 27, 30, 31}; +- vector unsigned short left, right; +- vector unsigned char byte_cnt, invalid_mask; +- vector unsigned long lshift; ++ __vector unsigned short left, right; ++ __vector unsigned char byte_cnt, invalid_mask; ++ __vector unsigned long lshift; + __attribute__((altivec(vector__))) + __attribute__((altivec(bool__))) + unsigned long long shmask; +- const vector unsigned long shmax = {64, 64}; ++ const __vector unsigned long shmax = {64, 64}; + #endif -@@ -703,7 +713,7 @@ raft_set_servers(struct raft *raft, const struct hmap *new_servers, - } - } + for (i = 0; i < MLX5_VPMD_DESCS_PER_LOOP; ++i) + if (likely(pos + i < mcqe_n)) + rte_prefetch0((void *)(cq + pos + i)); + /* A.1 load mCQEs into a 128bit register. */ +- mcqe1 = (vector unsigned char)vec_vsx_ld(0, ++ mcqe1 = (__vector unsigned char)vec_vsx_ld(0, + (signed int const *)&mcq[pos % 8]); +- mcqe2 = (vector unsigned char)vec_vsx_ld(0, ++ mcqe2 = (__vector unsigned char)vec_vsx_ld(0, + (signed int const *)&mcq[pos % 8 + 2]); + + /* B.1 store rearm data to mbuf. */ +- *(vector unsigned char *) ++ *(__vector unsigned char *) + &elts[pos]->rearm_data = rearm; +- *(vector unsigned char *) ++ *(__vector unsigned char *) + &elts[pos + 1]->rearm_data = rearm; + + /* C.1 combine data from mCQEs with rx_descriptor_fields1. */ + rxdf1 = vec_perm(mcqe1, zero, shuf_mask1); + rxdf2 = vec_perm(mcqe1, zero, shuf_mask2); +- rxdf1 = (vector unsigned char) +- ((vector unsigned short)rxdf1 - +- (vector unsigned short)crc_adj); +- rxdf2 = (vector unsigned char) +- ((vector unsigned short)rxdf2 - +- (vector unsigned short)crc_adj); +- rxdf1 = (vector unsigned char) +- vec_sel((vector unsigned short)rxdf1, +- (vector unsigned short)rxdf, rxdf_sel_mask); +- rxdf2 = (vector unsigned char) +- vec_sel((vector unsigned short)rxdf2, +- (vector unsigned short)rxdf, rxdf_sel_mask); ++ rxdf1 = (__vector unsigned char) ++ ((__vector unsigned short)rxdf1 - ++ (__vector unsigned short)crc_adj); ++ rxdf2 = (__vector unsigned char) ++ ((__vector unsigned short)rxdf2 - ++ (__vector unsigned short)crc_adj); ++ rxdf1 = (__vector unsigned char) ++ vec_sel((__vector unsigned short)rxdf1, ++ (__vector unsigned short)rxdf, rxdf_sel_mask); ++ rxdf2 = (__vector unsigned char) ++ vec_sel((__vector unsigned short)rxdf2, ++ (__vector unsigned short)rxdf, rxdf_sel_mask); + + /* D.1 store rx_descriptor_fields1. */ +- *(vector unsigned char *) ++ *(__vector unsigned char *) + &elts[pos]->rx_descriptor_fields1 = rxdf1; +- *(vector unsigned char *) ++ *(__vector unsigned char *) + &elts[pos + 1]->rx_descriptor_fields1 = rxdf2; + + /* B.1 store rearm data to mbuf. */ +- *(vector unsigned char *) ++ *(__vector unsigned char *) + &elts[pos + 2]->rearm_data = rearm; +- *(vector unsigned char *) ++ *(__vector unsigned char *) + &elts[pos + 3]->rearm_data = rearm; + + /* C.1 combine data from mCQEs with rx_descriptor_fields1. */ + rxdf1 = vec_perm(mcqe2, zero, shuf_mask1); + rxdf2 = vec_perm(mcqe2, zero, shuf_mask2); +- rxdf1 = (vector unsigned char) +- ((vector unsigned short)rxdf1 - +- (vector unsigned short)crc_adj); +- rxdf2 = (vector unsigned char) +- ((vector unsigned short)rxdf2 - +- (vector unsigned short)crc_adj); +- rxdf1 = (vector unsigned char) +- vec_sel((vector unsigned short)rxdf1, +- (vector unsigned short)rxdf, rxdf_sel_mask); +- rxdf2 = (vector unsigned char) +- vec_sel((vector unsigned short)rxdf2, +- (vector unsigned short)rxdf, rxdf_sel_mask); ++ rxdf1 = (__vector unsigned char) ++ ((__vector unsigned short)rxdf1 - ++ (__vector unsigned short)crc_adj); ++ rxdf2 = (__vector unsigned char) ++ ((__vector unsigned short)rxdf2 - ++ (__vector unsigned short)crc_adj); ++ rxdf1 = (__vector unsigned char) ++ vec_sel((__vector unsigned short)rxdf1, ++ (__vector unsigned short)rxdf, rxdf_sel_mask); ++ rxdf2 = (__vector unsigned char) ++ vec_sel((__vector unsigned short)rxdf2, ++ (__vector unsigned short)rxdf, rxdf_sel_mask); + + /* D.1 store rx_descriptor_fields1. */ +- *(vector unsigned char *) ++ *(__vector unsigned char *) + &elts[pos + 2]->rx_descriptor_fields1 = rxdf1; +- *(vector unsigned char *) ++ *(__vector unsigned char *) + &elts[pos + 3]->rx_descriptor_fields1 = rxdf2; + + #ifdef MLX5_PMD_SOFT_COUNTERS +- invalid_mask = (vector unsigned char)(vector unsigned long){ ++ invalid_mask = (__vector unsigned char)(__vector unsigned long){ + (mcqe_n - pos) * sizeof(uint16_t) * 8, 0}; + + lshift = +- vec_splat((vector unsigned long)invalid_mask, 0); ++ vec_splat((__vector unsigned long)invalid_mask, 0); + shmask = vec_cmpgt(shmax, lshift); +- invalid_mask = (vector unsigned char) +- vec_sl((vector unsigned long)ones, lshift); +- invalid_mask = (vector unsigned char) +- vec_sel((vector unsigned long)shmask, +- (vector unsigned long)invalid_mask, shmask); +- +- byte_cnt = (vector unsigned char) +- vec_sel((vector unsigned short) +- vec_sro((vector unsigned short)mcqe1, +- (vector unsigned char){32}), +- (vector unsigned short)mcqe2, mcqe_sel_mask); ++ invalid_mask = (__vector unsigned char) ++ vec_sl((__vector unsigned long)ones, lshift); ++ invalid_mask = (__vector unsigned char) ++ vec_sel((__vector unsigned long)shmask, ++ (__vector unsigned long)invalid_mask, shmask); ++ ++ byte_cnt = (__vector unsigned char) ++ vec_sel((__vector unsigned short) ++ vec_sro((__vector unsigned short)mcqe1, ++ (__vector unsigned char){32}), ++ (__vector unsigned short)mcqe2, mcqe_sel_mask); + byte_cnt = vec_perm(byte_cnt, zero, len_shuf_mask); +- byte_cnt = (vector unsigned char) +- vec_andc((vector unsigned long)byte_cnt, +- (vector unsigned long)invalid_mask); +- left = vec_perm((vector unsigned short)byte_cnt, +- (vector unsigned short)zero, lower_half); +- right = vec_perm((vector unsigned short)byte_cnt, +- (vector unsigned short)zero, upper_half); +- byte_cnt = (vector unsigned char)vec_add(left, right); +- left = vec_perm((vector unsigned short)byte_cnt, +- (vector unsigned short)zero, lower_half); +- right = vec_perm((vector unsigned short)byte_cnt, +- (vector unsigned short)zero, upper_half); +- byte_cnt = (vector unsigned char)vec_add(left, right); +- rcvd_byte += ((vector unsigned long)byte_cnt)[0]; ++ byte_cnt = (__vector unsigned char) ++ vec_andc((__vector unsigned long)byte_cnt, ++ (__vector unsigned long)invalid_mask); ++ left = vec_perm((__vector unsigned short)byte_cnt, ++ (__vector unsigned short)zero, lower_half); ++ right = vec_perm((__vector unsigned short)byte_cnt, ++ (__vector unsigned short)zero, upper_half); ++ byte_cnt = (__vector unsigned char)vec_add(left, right); ++ left = vec_perm((__vector unsigned short)byte_cnt, ++ (__vector unsigned short)zero, lower_half); ++ right = vec_perm((__vector unsigned short)byte_cnt, ++ (__vector unsigned short)zero, upper_half); ++ byte_cnt = (__vector unsigned char)vec_add(left, right); ++ rcvd_byte += ((__vector unsigned long)byte_cnt)[0]; + #endif -- HMAP_FOR_EACH_SAFE (s, next, hmap_node, new_servers) { -+ HMAP_FOR_EACH_SAFE (s, hmap_node, new_servers) { - if (!raft_find_server(raft, &s->sid)) { - VLOG(level, "server %s added to configuration", s->nickname); + if (rxq->mark) { +@@ -265,99 +265,99 @@ rxq_cq_decompress_v(struct mlx5_rxq_data *rxq, volatile struct mlx5_cqe *cq, + elts[pos + 2]->hash.fdir.hi = flow_tag; + elts[pos + 3]->hash.fdir.hi = flow_tag; + } else { +- const vector unsigned char flow_mark_adj = +- (vector unsigned char) +- (vector unsigned int){ ++ const __vector unsigned char flow_mark_adj = ++ (__vector unsigned char) ++ (__vector unsigned int){ + -1, -1, -1, -1}; +- const vector unsigned char flow_mark_shuf = +- (vector unsigned char){ ++ const __vector unsigned char flow_mark_shuf = ++ (__vector unsigned char){ + -1, -1, -1, -1, + -1, -1, -1, -1, + 12, 8, 9, -1, + 4, 0, 1, -1}; +- const vector unsigned char ft_mask = +- (vector unsigned char) +- (vector unsigned int){ ++ const __vector unsigned char ft_mask = ++ (__vector unsigned char) ++ (__vector unsigned int){ + 0xffffff00, 0xffffff00, + 0xffffff00, 0xffffff00}; +- const vector unsigned char fdir_flags = +- (vector unsigned char) +- (vector unsigned int){ ++ const __vector unsigned char fdir_flags = ++ (__vector unsigned char) ++ (__vector unsigned int){ + RTE_MBUF_F_RX_FDIR, RTE_MBUF_F_RX_FDIR, + RTE_MBUF_F_RX_FDIR, RTE_MBUF_F_RX_FDIR}; +- const vector unsigned char fdir_all_flags = +- (vector unsigned char) +- (vector unsigned int){ ++ const __vector unsigned char fdir_all_flags = ++ (__vector unsigned char) ++ (__vector unsigned int){ + RTE_MBUF_F_RX_FDIR | RTE_MBUF_F_RX_FDIR_ID, + RTE_MBUF_F_RX_FDIR | RTE_MBUF_F_RX_FDIR_ID, + RTE_MBUF_F_RX_FDIR | RTE_MBUF_F_RX_FDIR_ID, + RTE_MBUF_F_RX_FDIR | RTE_MBUF_F_RX_FDIR_ID}; +- vector unsigned char fdir_id_flags = +- (vector unsigned char) +- (vector unsigned int){ ++ __vector unsigned char fdir_id_flags = ++ (__vector unsigned char) ++ (__vector unsigned int){ + RTE_MBUF_F_RX_FDIR_ID, RTE_MBUF_F_RX_FDIR_ID, + RTE_MBUF_F_RX_FDIR_ID, RTE_MBUF_F_RX_FDIR_ID}; + /* Extract flow_tag field. */ +- vector unsigned char ftag0 = vec_perm(mcqe1, ++ __vector unsigned char ftag0 = vec_perm(mcqe1, + zero, flow_mark_shuf); +- vector unsigned char ftag1 = vec_perm(mcqe2, ++ __vector unsigned char ftag1 = vec_perm(mcqe2, + zero, flow_mark_shuf); +- vector unsigned char ftag = +- (vector unsigned char) +- vec_mergel((vector unsigned int)ftag0, +- (vector unsigned int)ftag1); +- vector unsigned char invalid_mask = +- (vector unsigned char) +- vec_cmpeq((vector unsigned int)ftag, +- (vector unsigned int)zero); +- +- ol_flags_mask = (vector unsigned char) +- vec_or((vector unsigned long) ++ __vector unsigned char ftag = ++ (__vector unsigned char) ++ vec_mergel((__vector unsigned int)ftag0, ++ (__vector unsigned int)ftag1); ++ __vector unsigned char invalid_mask = ++ (__vector unsigned char) ++ vec_cmpeq((__vector unsigned int)ftag, ++ (__vector unsigned int)zero); ++ ++ ol_flags_mask = (__vector unsigned char) ++ vec_or((__vector unsigned long) + ol_flags_mask, +- (vector unsigned long)fdir_all_flags); ++ (__vector unsigned long)fdir_all_flags); + + /* Set RTE_MBUF_F_RX_FDIR if flow tag is non-zero. */ +- invalid_mask = (vector unsigned char) +- vec_cmpeq((vector unsigned int)ftag, +- (vector unsigned int)zero); +- ol_flags = (vector unsigned char) +- vec_or((vector unsigned long)ol_flags, +- (vector unsigned long) +- vec_andc((vector unsigned long) ++ invalid_mask = (__vector unsigned char) ++ vec_cmpeq((__vector unsigned int)ftag, ++ (__vector unsigned int)zero); ++ ol_flags = (__vector unsigned char) ++ vec_or((__vector unsigned long)ol_flags, ++ (__vector unsigned long) ++ vec_andc((__vector unsigned long) + fdir_flags, +- (vector unsigned long)invalid_mask)); +- ol_flags_mask = (vector unsigned char) +- vec_or((vector unsigned long) ++ (__vector unsigned long)invalid_mask)); ++ ol_flags_mask = (__vector unsigned char) ++ vec_or((__vector unsigned long) + ol_flags_mask, +- (vector unsigned long)fdir_flags); ++ (__vector unsigned long)fdir_flags); + + /* Mask out invalid entries. */ +- fdir_id_flags = (vector unsigned char) +- vec_andc((vector unsigned long) ++ fdir_id_flags = (__vector unsigned char) ++ vec_andc((__vector unsigned long) + fdir_id_flags, +- (vector unsigned long)invalid_mask); ++ (__vector unsigned long)invalid_mask); + + /* Check if flow tag MLX5_FLOW_MARK_DEFAULT. */ +- ol_flags = (vector unsigned char) +- vec_or((vector unsigned long)ol_flags, +- (vector unsigned long) +- vec_andc((vector unsigned long) ++ ol_flags = (__vector unsigned char) ++ vec_or((__vector unsigned long)ol_flags, ++ (__vector unsigned long) ++ vec_andc((__vector unsigned long) + fdir_id_flags, +- (vector unsigned long) +- vec_cmpeq((vector unsigned int)ftag, +- (vector unsigned int)ft_mask))); ++ (__vector unsigned long) ++ vec_cmpeq((__vector unsigned int)ftag, ++ (__vector unsigned int)ft_mask))); + +- ftag = (vector unsigned char) +- ((vector unsigned int)ftag + +- (vector unsigned int)flow_mark_adj); ++ ftag = (__vector unsigned char) ++ ((__vector unsigned int)ftag + ++ (__vector unsigned int)flow_mark_adj); + elts[pos]->hash.fdir.hi = +- ((vector unsigned int)ftag)[0]; ++ ((__vector unsigned int)ftag)[0]; + elts[pos + 1]->hash.fdir.hi = +- ((vector unsigned int)ftag)[1]; ++ ((__vector unsigned int)ftag)[1]; + elts[pos + 2]->hash.fdir.hi = +- ((vector unsigned int)ftag)[2]; ++ ((__vector unsigned int)ftag)[2]; + elts[pos + 3]->hash.fdir.hi = +- ((vector unsigned int)ftag)[3]; ++ ((__vector unsigned int)ftag)[3]; + } + } + if (unlikely(rxq->mcqe_format != MLX5_CQE_RESP_FORMAT_HASH)) { +@@ -373,37 +373,37 @@ rxq_cq_decompress_v(struct mlx5_rxq_data *rxq, volatile struct mlx5_cqe *cq, + mcq[pos % 8 + 2].hdr_type; + const uint8_t pkt_hdr3 = + mcq[pos % 8 + 3].hdr_type; +- const vector unsigned char vlan_mask = +- (vector unsigned char) +- (vector unsigned int) { ++ const __vector unsigned char vlan_mask = ++ (__vector unsigned char) ++ (__vector unsigned int) { + (RTE_MBUF_F_RX_VLAN | RTE_MBUF_F_RX_VLAN_STRIPPED), + (RTE_MBUF_F_RX_VLAN | RTE_MBUF_F_RX_VLAN_STRIPPED), + (RTE_MBUF_F_RX_VLAN | RTE_MBUF_F_RX_VLAN_STRIPPED), + (RTE_MBUF_F_RX_VLAN | RTE_MBUF_F_RX_VLAN_STRIPPED)}; +- const vector unsigned char cv_mask = +- (vector unsigned char) +- (vector unsigned int) { ++ const __vector unsigned char cv_mask = ++ (__vector unsigned char) ++ (__vector unsigned int) { + MLX5_CQE_VLAN_STRIPPED, + MLX5_CQE_VLAN_STRIPPED, + MLX5_CQE_VLAN_STRIPPED, + MLX5_CQE_VLAN_STRIPPED}; +- vector unsigned char pkt_cv = +- (vector unsigned char) +- (vector unsigned int) { ++ __vector unsigned char pkt_cv = ++ (__vector unsigned char) ++ (__vector unsigned int) { + pkt_hdr0 & 0x1, pkt_hdr1 & 0x1, + pkt_hdr2 & 0x1, pkt_hdr3 & 0x1}; + +- ol_flags_mask = (vector unsigned char) +- vec_or((vector unsigned long) ++ ol_flags_mask = (__vector unsigned char) ++ vec_or((__vector unsigned long) + ol_flags_mask, +- (vector unsigned long)vlan_mask); +- ol_flags = (vector unsigned char) +- vec_or((vector unsigned long)ol_flags, +- (vector unsigned long) +- vec_and((vector unsigned long)vlan_mask, +- (vector unsigned long) +- vec_cmpeq((vector unsigned int)pkt_cv, +- (vector unsigned int)cv_mask))); ++ (__vector unsigned long)vlan_mask); ++ ol_flags = (__vector unsigned char) ++ vec_or((__vector unsigned long)ol_flags, ++ (__vector unsigned long) ++ vec_and((__vector unsigned long)vlan_mask, ++ (__vector unsigned long) ++ vec_cmpeq((__vector unsigned int)pkt_cv, ++ (__vector unsigned int)cv_mask))); + elts[pos]->packet_type = + mlx5_ptype_table[(pkt_hdr0 >> 2) | + pkt_info]; +@@ -431,36 +431,36 @@ rxq_cq_decompress_v(struct mlx5_rxq_data *rxq, volatile struct mlx5_cqe *cq, + pkt_info) & (1 << 6)); + } + } +- const vector unsigned char hash_mask = +- (vector unsigned char)(vector unsigned int) { ++ const __vector unsigned char hash_mask = ++ (__vector unsigned char)(__vector unsigned int) { + RTE_MBUF_F_RX_RSS_HASH, + RTE_MBUF_F_RX_RSS_HASH, + RTE_MBUF_F_RX_RSS_HASH, + RTE_MBUF_F_RX_RSS_HASH}; +- const vector unsigned char rearm_flags = +- (vector unsigned char)(vector unsigned int) { ++ const __vector unsigned char rearm_flags = ++ (__vector unsigned char)(__vector unsigned int) { + (uint32_t)t_pkt->ol_flags, + (uint32_t)t_pkt->ol_flags, + (uint32_t)t_pkt->ol_flags, + (uint32_t)t_pkt->ol_flags}; + +- ol_flags_mask = (vector unsigned char) +- vec_or((vector unsigned long)ol_flags_mask, +- (vector unsigned long)hash_mask); +- ol_flags = (vector unsigned char) +- vec_or((vector unsigned long)ol_flags, +- (vector unsigned long) +- vec_andc((vector unsigned long)rearm_flags, +- (vector unsigned long)ol_flags_mask)); ++ ol_flags_mask = (__vector unsigned char) ++ vec_or((__vector unsigned long)ol_flags_mask, ++ (__vector unsigned long)hash_mask); ++ ol_flags = (__vector unsigned char) ++ vec_or((__vector unsigned long)ol_flags, ++ (__vector unsigned long) ++ vec_andc((__vector unsigned long)rearm_flags, ++ (__vector unsigned long)ol_flags_mask)); + + elts[pos]->ol_flags = +- ((vector unsigned int)ol_flags)[0]; ++ ((__vector unsigned int)ol_flags)[0]; + elts[pos + 1]->ol_flags = +- ((vector unsigned int)ol_flags)[1]; ++ ((__vector unsigned int)ol_flags)[1]; + elts[pos + 2]->ol_flags = +- ((vector unsigned int)ol_flags)[2]; ++ ((__vector unsigned int)ol_flags)[2]; + elts[pos + 3]->ol_flags = +- ((vector unsigned int)ol_flags)[3]; ++ ((__vector unsigned int)ol_flags)[3]; + elts[pos]->hash.rss = 0; + elts[pos + 1]->hash.rss = 0; + elts[pos + 2]->hash.rss = 0; +@@ -524,13 +524,13 @@ rxq_cq_decompress_v(struct mlx5_rxq_data *rxq, volatile struct mlx5_cqe *cq, + */ + static inline void + rxq_cq_to_ptype_oflags_v(struct mlx5_rxq_data *rxq, +- vector unsigned char cqes[4], vector unsigned char op_err, ++ __vector unsigned char cqes[4], __vector unsigned char op_err, + struct rte_mbuf **pkts) + { +- vector unsigned char pinfo0, pinfo1; +- vector unsigned char pinfo, ptype; +- vector unsigned char ol_flags = (vector unsigned char) +- (vector unsigned int){ ++ __vector unsigned char pinfo0, pinfo1; ++ __vector unsigned char pinfo, ptype; ++ __vector unsigned char ol_flags = (__vector unsigned char) ++ (__vector unsigned int){ + rxq->rss_hash * RTE_MBUF_F_RX_RSS_HASH | + rxq->hw_timestamp * rxq->timestamp_rx_flag, + rxq->rss_hash * RTE_MBUF_F_RX_RSS_HASH | +@@ -539,25 +539,25 @@ rxq_cq_to_ptype_oflags_v(struct mlx5_rxq_data *rxq, + rxq->hw_timestamp * rxq->timestamp_rx_flag, + rxq->rss_hash * RTE_MBUF_F_RX_RSS_HASH | + rxq->hw_timestamp * rxq->timestamp_rx_flag}; +- vector unsigned char cv_flags; +- const vector unsigned char zero = (vector unsigned char){0}; +- const vector unsigned char ptype_mask = +- (vector unsigned char)(vector unsigned int){ ++ __vector unsigned char cv_flags; ++ const __vector unsigned char zero = (__vector unsigned char){0}; ++ const __vector unsigned char ptype_mask = ++ (__vector unsigned char)(__vector unsigned int){ + 0x0000fd06, 0x0000fd06, 0x0000fd06, 0x0000fd06}; +- const vector unsigned char ptype_ol_mask = +- (vector unsigned char)(vector unsigned int){ ++ const __vector unsigned char ptype_ol_mask = ++ (__vector unsigned char)(__vector unsigned int){ + 0x00000106, 0x00000106, 0x00000106, 0x00000106}; +- const vector unsigned char pinfo_mask = +- (vector unsigned char)(vector unsigned int){ ++ const __vector unsigned char pinfo_mask = ++ (__vector unsigned char)(__vector unsigned int){ + 0x00000003, 0x00000003, 0x00000003, 0x00000003}; +- const vector unsigned char cv_flag_sel = (vector unsigned char){ ++ const __vector unsigned char cv_flag_sel = (__vector unsigned char){ + 0, (uint8_t)(RTE_MBUF_F_RX_VLAN | RTE_MBUF_F_RX_VLAN_STRIPPED), + (uint8_t)(RTE_MBUF_F_RX_IP_CKSUM_GOOD >> 1), 0, + (uint8_t)(RTE_MBUF_F_RX_L4_CKSUM_GOOD >> 1), 0, + (uint8_t)((RTE_MBUF_F_RX_IP_CKSUM_GOOD | RTE_MBUF_F_RX_L4_CKSUM_GOOD) >> 1), + 0, 0, 0, 0, 0, 0, 0, 0, 0}; +- const vector unsigned char cv_mask = +- (vector unsigned char)(vector unsigned int){ ++ const __vector unsigned char cv_mask = ++ (__vector unsigned char)(__vector unsigned int){ + RTE_MBUF_F_RX_IP_CKSUM_GOOD | RTE_MBUF_F_RX_L4_CKSUM_GOOD | + RTE_MBUF_F_RX_VLAN | RTE_MBUF_F_RX_VLAN_STRIPPED, + RTE_MBUF_F_RX_IP_CKSUM_GOOD | RTE_MBUF_F_RX_L4_CKSUM_GOOD | +@@ -566,77 +566,77 @@ rxq_cq_to_ptype_oflags_v(struct mlx5_rxq_data *rxq, + RTE_MBUF_F_RX_VLAN | RTE_MBUF_F_RX_VLAN_STRIPPED, + RTE_MBUF_F_RX_IP_CKSUM_GOOD | RTE_MBUF_F_RX_L4_CKSUM_GOOD | + RTE_MBUF_F_RX_VLAN | RTE_MBUF_F_RX_VLAN_STRIPPED}; +- const vector unsigned char mbuf_init = +- (vector unsigned char)vec_vsx_ld +- (0, (vector unsigned char *)&rxq->mbuf_initializer); +- const vector unsigned short rearm_sel_mask = +- (vector unsigned short){0, 0, 0, 0, 0xffff, 0xffff, 0, 0}; +- vector unsigned char rearm0, rearm1, rearm2, rearm3; ++ const __vector unsigned char mbuf_init = ++ (__vector unsigned char)vec_vsx_ld ++ (0, (__vector unsigned char *)&rxq->mbuf_initializer); ++ const __vector unsigned short rearm_sel_mask = ++ (__vector unsigned short){0, 0, 0, 0, 0xffff, 0xffff, 0, 0}; ++ __vector unsigned char rearm0, rearm1, rearm2, rearm3; + uint8_t pt_idx0, pt_idx1, pt_idx2, pt_idx3; + + /* Extract pkt_info field. */ +- pinfo0 = (vector unsigned char) +- vec_mergeh((vector unsigned int)cqes[0], +- (vector unsigned int)cqes[1]); +- pinfo1 = (vector unsigned char) +- vec_mergeh((vector unsigned int)cqes[2], +- (vector unsigned int)cqes[3]); +- pinfo = (vector unsigned char) +- vec_mergeh((vector unsigned long)pinfo0, +- (vector unsigned long)pinfo1); ++ pinfo0 = (__vector unsigned char) ++ vec_mergeh((__vector unsigned int)cqes[0], ++ (__vector unsigned int)cqes[1]); ++ pinfo1 = (__vector unsigned char) ++ vec_mergeh((__vector unsigned int)cqes[2], ++ (__vector unsigned int)cqes[3]); ++ pinfo = (__vector unsigned char) ++ vec_mergeh((__vector unsigned long)pinfo0, ++ (__vector unsigned long)pinfo1); + + /* Extract hdr_type_etc field. */ +- pinfo0 = (vector unsigned char) +- vec_mergel((vector unsigned int)cqes[0], +- (vector unsigned int)cqes[1]); +- pinfo1 = (vector unsigned char) +- vec_mergel((vector unsigned int)cqes[2], +- (vector unsigned int)cqes[3]); +- ptype = (vector unsigned char) +- vec_mergeh((vector unsigned long)pinfo0, +- (vector unsigned long)pinfo1); ++ pinfo0 = (__vector unsigned char) ++ vec_mergel((__vector unsigned int)cqes[0], ++ (__vector unsigned int)cqes[1]); ++ pinfo1 = (__vector unsigned char) ++ vec_mergel((__vector unsigned int)cqes[2], ++ (__vector unsigned int)cqes[3]); ++ ptype = (__vector unsigned char) ++ vec_mergeh((__vector unsigned long)pinfo0, ++ (__vector unsigned long)pinfo1); + + if (rxq->mark) { +- const vector unsigned char pinfo_ft_mask = +- (vector unsigned char)(vector unsigned int){ ++ const __vector unsigned char pinfo_ft_mask = ++ (__vector unsigned char)(__vector unsigned int){ + 0xffffff00, 0xffffff00, 0xffffff00, 0xffffff00}; +- const vector unsigned char fdir_flags = +- (vector unsigned char)(vector unsigned int){ ++ const __vector unsigned char fdir_flags = ++ (__vector unsigned char)(__vector unsigned int){ + RTE_MBUF_F_RX_FDIR, RTE_MBUF_F_RX_FDIR, + RTE_MBUF_F_RX_FDIR, RTE_MBUF_F_RX_FDIR}; +- vector unsigned char fdir_id_flags = +- (vector unsigned char)(vector unsigned int){ ++ __vector unsigned char fdir_id_flags = ++ (__vector unsigned char)(__vector unsigned int){ + RTE_MBUF_F_RX_FDIR_ID, RTE_MBUF_F_RX_FDIR_ID, + RTE_MBUF_F_RX_FDIR_ID, RTE_MBUF_F_RX_FDIR_ID}; +- vector unsigned char flow_tag, invalid_mask; ++ __vector unsigned char flow_tag, invalid_mask; + +- flow_tag = (vector unsigned char) +- vec_and((vector unsigned long)pinfo, +- (vector unsigned long)pinfo_ft_mask); ++ flow_tag = (__vector unsigned char) ++ vec_and((__vector unsigned long)pinfo, ++ (__vector unsigned long)pinfo_ft_mask); + + /* Check if flow tag is non-zero then set RTE_MBUF_F_RX_FDIR. */ +- invalid_mask = (vector unsigned char) +- vec_cmpeq((vector unsigned int)flow_tag, +- (vector unsigned int)zero); +- ol_flags = (vector unsigned char) +- vec_or((vector unsigned long)ol_flags, +- (vector unsigned long) +- vec_andc((vector unsigned long)fdir_flags, +- (vector unsigned long)invalid_mask)); ++ invalid_mask = (__vector unsigned char) ++ vec_cmpeq((__vector unsigned int)flow_tag, ++ (__vector unsigned int)zero); ++ ol_flags = (__vector unsigned char) ++ vec_or((__vector unsigned long)ol_flags, ++ (__vector unsigned long) ++ vec_andc((__vector unsigned long)fdir_flags, ++ (__vector unsigned long)invalid_mask)); + + /* Mask out invalid entries. */ +- fdir_id_flags = (vector unsigned char) +- vec_andc((vector unsigned long)fdir_id_flags, +- (vector unsigned long)invalid_mask); ++ fdir_id_flags = (__vector unsigned char) ++ vec_andc((__vector unsigned long)fdir_id_flags, ++ (__vector unsigned long)invalid_mask); + + /* Check if flow tag MLX5_FLOW_MARK_DEFAULT. */ +- ol_flags = (vector unsigned char) +- vec_or((vector unsigned long)ol_flags, +- (vector unsigned long) +- vec_andc((vector unsigned long)fdir_id_flags, +- (vector unsigned long) +- vec_cmpeq((vector unsigned int)flow_tag, +- (vector unsigned int)pinfo_ft_mask))); ++ ol_flags = (__vector unsigned char) ++ vec_or((__vector unsigned long)ol_flags, ++ (__vector unsigned long) ++ vec_andc((__vector unsigned long)fdir_id_flags, ++ (__vector unsigned long) ++ vec_cmpeq((__vector unsigned int)flow_tag, ++ (__vector unsigned int)pinfo_ft_mask))); + } + /* + * Merge the two fields to generate the following: +@@ -649,39 +649,39 @@ rxq_cq_to_ptype_oflags_v(struct mlx5_rxq_data *rxq, + * bit[16] = tunneled + * bit[17] = outer_l3_type + */ +- ptype = (vector unsigned char) +- vec_and((vector unsigned long)ptype, +- (vector unsigned long)ptype_mask); +- pinfo = (vector unsigned char) +- vec_and((vector unsigned long)pinfo, +- (vector unsigned long)pinfo_mask); +- pinfo = (vector unsigned char) +- vec_sl((vector unsigned int)pinfo, +- (vector unsigned int){16, 16, 16, 16}); ++ ptype = (__vector unsigned char) ++ vec_and((__vector unsigned long)ptype, ++ (__vector unsigned long)ptype_mask); ++ pinfo = (__vector unsigned char) ++ vec_and((__vector unsigned long)pinfo, ++ (__vector unsigned long)pinfo_mask); ++ pinfo = (__vector unsigned char) ++ vec_sl((__vector unsigned int)pinfo, ++ (__vector unsigned int){16, 16, 16, 16}); + + /* Make pinfo has merged fields for ol_flags calculation. */ +- pinfo = (vector unsigned char) +- vec_or((vector unsigned long)ptype, +- (vector unsigned long)pinfo); +- ptype = (vector unsigned char) +- vec_sr((vector unsigned int)pinfo, +- (vector unsigned int){10, 10, 10, 10}); +- ptype = (vector unsigned char) +- vec_packs((vector unsigned int)ptype, +- (vector unsigned int)zero); ++ pinfo = (__vector unsigned char) ++ vec_or((__vector unsigned long)ptype, ++ (__vector unsigned long)pinfo); ++ ptype = (__vector unsigned char) ++ vec_sr((__vector unsigned int)pinfo, ++ (__vector unsigned int){10, 10, 10, 10}); ++ ptype = (__vector unsigned char) ++ vec_packs((__vector unsigned int)ptype, ++ (__vector unsigned int)zero); + + /* Errored packets will have RTE_PTYPE_ALL_MASK. */ +- op_err = (vector unsigned char) +- vec_sr((vector unsigned short)op_err, +- (vector unsigned short){8, 8, 8, 8, 8, 8, 8, 8}); +- ptype = (vector unsigned char) +- vec_or((vector unsigned long)ptype, +- (vector unsigned long)op_err); +- +- pt_idx0 = (uint8_t)((vector unsigned char)ptype)[0]; +- pt_idx1 = (uint8_t)((vector unsigned char)ptype)[2]; +- pt_idx2 = (uint8_t)((vector unsigned char)ptype)[4]; +- pt_idx3 = (uint8_t)((vector unsigned char)ptype)[6]; ++ op_err = (__vector unsigned char) ++ vec_sr((__vector unsigned short)op_err, ++ (__vector unsigned short){8, 8, 8, 8, 8, 8, 8, 8}); ++ ptype = (__vector unsigned char) ++ vec_or((__vector unsigned long)ptype, ++ (__vector unsigned long)op_err); ++ ++ pt_idx0 = (uint8_t)((__vector unsigned char)ptype)[0]; ++ pt_idx1 = (uint8_t)((__vector unsigned char)ptype)[2]; ++ pt_idx2 = (uint8_t)((__vector unsigned char)ptype)[4]; ++ pt_idx3 = (uint8_t)((__vector unsigned char)ptype)[6]; + + pkts[0]->packet_type = mlx5_ptype_table[pt_idx0] | + !!(pt_idx0 & (1 << 6)) * rxq->tunnel; +@@ -693,63 +693,63 @@ rxq_cq_to_ptype_oflags_v(struct mlx5_rxq_data *rxq, + !!(pt_idx3 & (1 << 6)) * rxq->tunnel; + + /* Fill flags for checksum and VLAN. */ +- pinfo = (vector unsigned char) +- vec_and((vector unsigned long)pinfo, +- (vector unsigned long)ptype_ol_mask); ++ pinfo = (__vector unsigned char) ++ vec_and((__vector unsigned long)pinfo, ++ (__vector unsigned long)ptype_ol_mask); + pinfo = vec_perm(cv_flag_sel, zero, pinfo); + + /* Locate checksum flags at byte[2:1] and merge with VLAN flags. */ +- cv_flags = (vector unsigned char) +- vec_sl((vector unsigned int)pinfo, +- (vector unsigned int){9, 9, 9, 9}); +- cv_flags = (vector unsigned char) +- vec_or((vector unsigned long)pinfo, +- (vector unsigned long)cv_flags); ++ cv_flags = (__vector unsigned char) ++ vec_sl((__vector unsigned int)pinfo, ++ (__vector unsigned int){9, 9, 9, 9}); ++ cv_flags = (__vector unsigned char) ++ vec_or((__vector unsigned long)pinfo, ++ (__vector unsigned long)cv_flags); + + /* Move back flags to start from byte[0]. */ +- cv_flags = (vector unsigned char) +- vec_sr((vector unsigned int)cv_flags, +- (vector unsigned int){8, 8, 8, 8}); ++ cv_flags = (__vector unsigned char) ++ vec_sr((__vector unsigned int)cv_flags, ++ (__vector unsigned int){8, 8, 8, 8}); + + /* Mask out garbage bits. */ +- cv_flags = (vector unsigned char) +- vec_and((vector unsigned long)cv_flags, +- (vector unsigned long)cv_mask); ++ cv_flags = (__vector unsigned char) ++ vec_and((__vector unsigned long)cv_flags, ++ (__vector unsigned long)cv_mask); + + /* Merge to ol_flags. */ +- ol_flags = (vector unsigned char) +- vec_or((vector unsigned long)ol_flags, +- (vector unsigned long)cv_flags); ++ ol_flags = (__vector unsigned char) ++ vec_or((__vector unsigned long)ol_flags, ++ (__vector unsigned long)cv_flags); + + /* Merge mbuf_init and ol_flags. */ +- rearm0 = (vector unsigned char) +- vec_sel((vector unsigned short)mbuf_init, +- (vector unsigned short) +- vec_slo((vector unsigned short)ol_flags, +- (vector unsigned char){64}), rearm_sel_mask); +- rearm1 = (vector unsigned char) +- vec_sel((vector unsigned short)mbuf_init, +- (vector unsigned short) +- vec_slo((vector unsigned short)ol_flags, +- (vector unsigned char){32}), rearm_sel_mask); +- rearm2 = (vector unsigned char) +- vec_sel((vector unsigned short)mbuf_init, +- (vector unsigned short)ol_flags, rearm_sel_mask); +- rearm3 = (vector unsigned char) +- vec_sel((vector unsigned short)mbuf_init, +- (vector unsigned short) +- vec_sro((vector unsigned short)ol_flags, +- (vector unsigned char){32}), rearm_sel_mask); ++ rearm0 = (__vector unsigned char) ++ vec_sel((__vector unsigned short)mbuf_init, ++ (__vector unsigned short) ++ vec_slo((__vector unsigned short)ol_flags, ++ (__vector unsigned char){64}), rearm_sel_mask); ++ rearm1 = (__vector unsigned char) ++ vec_sel((__vector unsigned short)mbuf_init, ++ (__vector unsigned short) ++ vec_slo((__vector unsigned short)ol_flags, ++ (__vector unsigned char){32}), rearm_sel_mask); ++ rearm2 = (__vector unsigned char) ++ vec_sel((__vector unsigned short)mbuf_init, ++ (__vector unsigned short)ol_flags, rearm_sel_mask); ++ rearm3 = (__vector unsigned char) ++ vec_sel((__vector unsigned short)mbuf_init, ++ (__vector unsigned short) ++ vec_sro((__vector unsigned short)ol_flags, ++ (__vector unsigned char){32}), rearm_sel_mask); + + /* Write 8B rearm_data and 8B ol_flags. */ + vec_vsx_st(rearm0, 0, +- (vector unsigned char *)&pkts[0]->rearm_data); ++ (__vector unsigned char *)&pkts[0]->rearm_data); + vec_vsx_st(rearm1, 0, +- (vector unsigned char *)&pkts[1]->rearm_data); ++ (__vector unsigned char *)&pkts[1]->rearm_data); + vec_vsx_st(rearm2, 0, +- (vector unsigned char *)&pkts[2]->rearm_data); ++ (__vector unsigned char *)&pkts[2]->rearm_data); + vec_vsx_st(rearm3, 0, +- (vector unsigned char *)&pkts[3]->rearm_data); ++ (__vector unsigned char *)&pkts[3]->rearm_data); + } -@@ -1376,8 +1386,8 @@ raft_close__(struct raft *raft) - raft->remove_server = NULL; - } + /** +@@ -788,31 +788,31 @@ rxq_cq_process_v(struct mlx5_rxq_data *rxq, volatile struct mlx5_cqe *cq, + uint64_t comp_idx = MLX5_VPMD_DESCS_PER_LOOP; + uint16_t nocmp_n = 0; + unsigned int ownership = !!(rxq->cq_ci & (q_mask + 1)); +- const vector unsigned char zero = (vector unsigned char){0}; +- const vector unsigned char ones = vec_splat_u8(-1); +- const vector unsigned char owner_check = +- (vector unsigned char)(vector unsigned long){ ++ const __vector unsigned char zero = (__vector unsigned char){0}; ++ const __vector unsigned char ones = vec_splat_u8(-1); ++ const __vector unsigned char owner_check = ++ (__vector unsigned char)(__vector unsigned long){ + 0x0100000001000000LL, 0x0100000001000000LL}; +- const vector unsigned char opcode_check = +- (vector unsigned char)(vector unsigned long){ ++ const __vector unsigned char opcode_check = ++ (__vector unsigned char)(__vector unsigned long){ + 0xf0000000f0000000LL, 0xf0000000f0000000LL}; +- const vector unsigned char format_check = +- (vector unsigned char)(vector unsigned long){ ++ const __vector unsigned char format_check = ++ (__vector unsigned char)(__vector unsigned long){ + 0x0c0000000c000000LL, 0x0c0000000c000000LL}; +- const vector unsigned char resp_err_check = +- (vector unsigned char)(vector unsigned long){ ++ const __vector unsigned char resp_err_check = ++ (__vector unsigned char)(__vector unsigned long){ + 0xe0000000e0000000LL, 0xe0000000e0000000LL}; + #ifdef MLX5_PMD_SOFT_COUNTERS + uint32_t rcvd_byte = 0; + /* Mask to shuffle byte_cnt to add up stats. Do bswap16 for all. */ +- const vector unsigned char len_shuf_mask = (vector unsigned char){ ++ const __vector unsigned char len_shuf_mask = (__vector unsigned char){ + 1, 0, 5, 4, + 9, 8, 13, 12, + -1, -1, -1, -1, + -1, -1, -1, -1}; + #endif + /* Mask to shuffle from extracted CQE to mbuf. */ +- const vector unsigned char shuf_mask = (vector unsigned char){ ++ const __vector unsigned char shuf_mask = (__vector unsigned char){ + 5, 4, /* bswap16, pkt_len */ + -1, -1, /* zero out 2nd half of pkt_len */ + 5, 4, /* bswap16, data_len */ +@@ -821,22 +821,22 @@ rxq_cq_process_v(struct mlx5_rxq_data *rxq, volatile struct mlx5_cqe *cq, + 1, 2, 3, -1}; /* fdir.hi */ + /* Mask to blend from the last Qword to the first DQword. */ + /* Mask to blend from the last Qword to the first DQword. */ +- const vector unsigned char blend_mask = (vector unsigned char){ ++ const __vector unsigned char blend_mask = (__vector unsigned char){ + -1, 0, 0, 0, + 0, 0, 0, 0, + -1, -1, -1, -1, + -1, -1, -1, -1}; +- const vector unsigned char crc_adj = +- (vector unsigned char)(vector unsigned short){ ++ const __vector unsigned char crc_adj = ++ (__vector unsigned char)(__vector unsigned short){ + rxq->crc_present * RTE_ETHER_CRC_LEN, 0, + rxq->crc_present * RTE_ETHER_CRC_LEN, 0, 0, 0, 0, 0}; +- const vector unsigned char flow_mark_adj = +- (vector unsigned char)(vector unsigned int){ ++ const __vector unsigned char flow_mark_adj = ++ (__vector unsigned char)(__vector unsigned int){ + 0, 0, 0, rxq->mark * (-1)}; +- const vector unsigned short cqe_sel_mask1 = +- (vector unsigned short){0, 0, 0, 0, 0xffff, 0xffff, 0, 0}; +- const vector unsigned short cqe_sel_mask2 = +- (vector unsigned short){0, 0, 0xffff, 0, 0, 0, 0, 0}; ++ const __vector unsigned short cqe_sel_mask1 = ++ (__vector unsigned short){0, 0, 0, 0, 0xffff, 0xffff, 0, 0}; ++ const __vector unsigned short cqe_sel_mask2 = ++ (__vector unsigned short){0, 0, 0xffff, 0, 0, 0, 0, 0}; + + /* + * A. load first Qword (8bytes) in one loop. +@@ -861,30 +861,30 @@ rxq_cq_process_v(struct mlx5_rxq_data *rxq, volatile struct mlx5_cqe *cq, + for (pos = 0; + pos < pkts_n; + pos += MLX5_VPMD_DESCS_PER_LOOP) { +- vector unsigned char cqes[MLX5_VPMD_DESCS_PER_LOOP]; +- vector unsigned char cqe_tmp1, cqe_tmp2; +- vector unsigned char pkt_mb0, pkt_mb1, pkt_mb2, pkt_mb3; +- vector unsigned char op_own, op_own_tmp1, op_own_tmp2; +- vector unsigned char opcode, owner_mask, invalid_mask; +- vector unsigned char comp_mask; +- vector unsigned char mask; ++ __vector unsigned char cqes[MLX5_VPMD_DESCS_PER_LOOP]; ++ __vector unsigned char cqe_tmp1, cqe_tmp2; ++ __vector unsigned char pkt_mb0, pkt_mb1, pkt_mb2, pkt_mb3; ++ __vector unsigned char op_own, op_own_tmp1, op_own_tmp2; ++ __vector unsigned char opcode, owner_mask, invalid_mask; ++ __vector unsigned char comp_mask; ++ __vector unsigned char mask; + #ifdef MLX5_PMD_SOFT_COUNTERS +- const vector unsigned char lower_half = { ++ const __vector unsigned char lower_half = { + 0, 1, 4, 5, 8, 9, 12, 13, + 16, 17, 20, 21, 24, 25, 28, 29}; +- const vector unsigned char upper_half = { ++ const __vector unsigned char upper_half = { + 2, 3, 6, 7, 10, 11, 14, 15, + 18, 19, 22, 23, 26, 27, 30, 31}; +- const vector unsigned long shmax = {64, 64}; +- vector unsigned char byte_cnt; +- vector unsigned short left, right; +- vector unsigned long lshift; +- vector __attribute__((altivec(bool__))) ++ const __vector unsigned long shmax = {64, 64}; ++ __vector unsigned char byte_cnt; ++ __vector unsigned short left, right; ++ __vector unsigned long lshift; ++ __vector __attribute__((altivec(bool__))) + unsigned long shmask; + #endif +- vector unsigned char mbp1, mbp2; +- vector unsigned char p = +- (vector unsigned char)(vector unsigned short){ ++ __vector unsigned char mbp1, mbp2; ++ __vector unsigned char p = ++ (__vector unsigned char)(__vector unsigned short){ + 0, 1, 2, 3, 0, 0, 0, 0}; + unsigned int p1, p2, p3; + +@@ -897,295 +897,295 @@ rxq_cq_process_v(struct mlx5_rxq_data *rxq, volatile struct mlx5_cqe *cq, + } + + /* A.0 do not cross the end of CQ. */ +- mask = (vector unsigned char)(vector unsigned long){ ++ mask = (__vector unsigned char)(__vector unsigned long){ + (pkts_n - pos) * sizeof(uint16_t) * 8, 0}; + + { +- vector unsigned long lshift; +- vector __attribute__((altivec(bool__))) ++ __vector unsigned long lshift; ++ __vector __attribute__((altivec(bool__))) + unsigned long shmask; +- const vector unsigned long shmax = {64, 64}; ++ const __vector unsigned long shmax = {64, 64}; + +- lshift = vec_splat((vector unsigned long)mask, 0); ++ lshift = vec_splat((__vector unsigned long)mask, 0); + shmask = vec_cmpgt(shmax, lshift); +- mask = (vector unsigned char) +- vec_sl((vector unsigned long)ones, lshift); +- mask = (vector unsigned char) +- vec_sel((vector unsigned long)shmask, +- (vector unsigned long)mask, shmask); ++ mask = (__vector unsigned char) ++ vec_sl((__vector unsigned long)ones, lshift); ++ mask = (__vector unsigned char) ++ vec_sel((__vector unsigned long)shmask, ++ (__vector unsigned long)mask, shmask); + } + +- p = (vector unsigned char) +- vec_andc((vector unsigned long)p, +- (vector unsigned long)mask); ++ p = (__vector unsigned char) ++ vec_andc((__vector unsigned long)p, ++ (__vector unsigned long)mask); + + /* A.1 load cqes. */ +- p3 = (unsigned int)((vector unsigned short)p)[3]; +- cqes[3] = (vector unsigned char)(vector unsigned long){ ++ p3 = (unsigned int)((__vector unsigned short)p)[3]; ++ cqes[3] = (__vector unsigned char)(__vector unsigned long){ + *(__rte_aligned(8) unsigned long *) + &cq[pos + p3].sop_drop_qpn, 0LL}; + rte_compiler_barrier(); + +- p2 = (unsigned int)((vector unsigned short)p)[2]; +- cqes[2] = (vector unsigned char)(vector unsigned long){ ++ p2 = (unsigned int)((__vector unsigned short)p)[2]; ++ cqes[2] = (__vector unsigned char)(__vector unsigned long){ + *(__rte_aligned(8) unsigned long *) + &cq[pos + p2].sop_drop_qpn, 0LL}; + rte_compiler_barrier(); + + /* B.1 load mbuf pointers. */ +- mbp1 = (vector unsigned char)vec_vsx_ld(0, ++ mbp1 = (__vector unsigned char)vec_vsx_ld(0, + (signed int const *)&elts[pos]); +- mbp2 = (vector unsigned char)vec_vsx_ld(0, ++ mbp2 = (__vector unsigned char)vec_vsx_ld(0, + (signed int const *)&elts[pos + 2]); + + /* A.1 load a block having op_own. */ +- p1 = (unsigned int)((vector unsigned short)p)[1]; +- cqes[1] = (vector unsigned char)(vector unsigned long){ ++ p1 = (unsigned int)((__vector unsigned short)p)[1]; ++ cqes[1] = (__vector unsigned char)(__vector unsigned long){ + *(__rte_aligned(8) unsigned long *) + &cq[pos + p1].sop_drop_qpn, 0LL}; + rte_compiler_barrier(); + +- cqes[0] = (vector unsigned char)(vector unsigned long){ ++ cqes[0] = (__vector unsigned char)(__vector unsigned long){ + *(__rte_aligned(8) unsigned long *) + &cq[pos].sop_drop_qpn, 0LL}; + rte_compiler_barrier(); + + /* B.2 copy mbuf pointers. */ +- *(vector unsigned char *)&pkts[pos] = mbp1; +- *(vector unsigned char *)&pkts[pos + 2] = mbp2; ++ *(__vector unsigned char *)&pkts[pos] = mbp1; ++ *(__vector unsigned char *)&pkts[pos + 2] = mbp2; + rte_io_rmb(); + + /* C.1 load remaining CQE data and extract necessary fields. */ +- cqe_tmp2 = *(vector unsigned char *) ++ cqe_tmp2 = *(__vector unsigned char *) + &cq[pos + p3].pkt_info; +- cqe_tmp1 = *(vector unsigned char *) ++ cqe_tmp1 = *(__vector unsigned char *) + &cq[pos + p2].pkt_info; + cqes[3] = vec_sel(cqes[3], cqe_tmp2, blend_mask); + cqes[2] = vec_sel(cqes[2], cqe_tmp1, blend_mask); +- cqe_tmp2 = (vector unsigned char)vec_vsx_ld(0, ++ cqe_tmp2 = (__vector unsigned char)vec_vsx_ld(0, + (signed int const *)&cq[pos + p3].csum); +- cqe_tmp1 = (vector unsigned char)vec_vsx_ld(0, ++ cqe_tmp1 = (__vector unsigned char)vec_vsx_ld(0, + (signed int const *)&cq[pos + p2].csum); +- cqes[3] = (vector unsigned char) +- vec_sel((vector unsigned short)cqes[3], +- (vector unsigned short)cqe_tmp2, cqe_sel_mask1); +- cqes[2] = (vector unsigned char) +- vec_sel((vector unsigned short)cqes[2], +- (vector unsigned short)cqe_tmp1, cqe_sel_mask1); +- cqe_tmp2 = (vector unsigned char)(vector unsigned long){ ++ cqes[3] = (__vector unsigned char) ++ vec_sel((__vector unsigned short)cqes[3], ++ (__vector unsigned short)cqe_tmp2, cqe_sel_mask1); ++ cqes[2] = (__vector unsigned char) ++ vec_sel((__vector unsigned short)cqes[2], ++ (__vector unsigned short)cqe_tmp1, cqe_sel_mask1); ++ cqe_tmp2 = (__vector unsigned char)(__vector unsigned long){ + *(__rte_aligned(8) unsigned long *) + &cq[pos + p3].rsvd4[2], 0LL}; +- cqe_tmp1 = (vector unsigned char)(vector unsigned long){ ++ cqe_tmp1 = (__vector unsigned char)(__vector unsigned long){ + *(__rte_aligned(8) unsigned long *) + &cq[pos + p2].rsvd4[2], 0LL}; +- cqes[3] = (vector unsigned char) +- vec_sel((vector unsigned short)cqes[3], +- (vector unsigned short)cqe_tmp2, +- (vector unsigned short)cqe_sel_mask2); +- cqes[2] = (vector unsigned char) +- vec_sel((vector unsigned short)cqes[2], +- (vector unsigned short)cqe_tmp1, +- (vector unsigned short)cqe_sel_mask2); ++ cqes[3] = (__vector unsigned char) ++ vec_sel((__vector unsigned short)cqes[3], ++ (__vector unsigned short)cqe_tmp2, ++ (__vector unsigned short)cqe_sel_mask2); ++ cqes[2] = (__vector unsigned char) ++ vec_sel((__vector unsigned short)cqes[2], ++ (__vector unsigned short)cqe_tmp1, ++ (__vector unsigned short)cqe_sel_mask2); + + /* C.2 generate final structure for mbuf with swapping bytes. */ + pkt_mb3 = vec_perm(cqes[3], zero, shuf_mask); + pkt_mb2 = vec_perm(cqes[2], zero, shuf_mask); + + /* C.3 adjust CRC length. */ +- pkt_mb3 = (vector unsigned char) +- ((vector unsigned short)pkt_mb3 - +- (vector unsigned short)crc_adj); +- pkt_mb2 = (vector unsigned char) +- ((vector unsigned short)pkt_mb2 - +- (vector unsigned short)crc_adj); ++ pkt_mb3 = (__vector unsigned char) ++ ((__vector unsigned short)pkt_mb3 - ++ (__vector unsigned short)crc_adj); ++ pkt_mb2 = (__vector unsigned char) ++ ((__vector unsigned short)pkt_mb2 - ++ (__vector unsigned short)crc_adj); + + /* C.4 adjust flow mark. */ +- pkt_mb3 = (vector unsigned char) +- ((vector unsigned int)pkt_mb3 + +- (vector unsigned int)flow_mark_adj); +- pkt_mb2 = (vector unsigned char) +- ((vector unsigned int)pkt_mb2 + +- (vector unsigned int)flow_mark_adj); ++ pkt_mb3 = (__vector unsigned char) ++ ((__vector unsigned int)pkt_mb3 + ++ (__vector unsigned int)flow_mark_adj); ++ pkt_mb2 = (__vector unsigned char) ++ ((__vector unsigned int)pkt_mb2 + ++ (__vector unsigned int)flow_mark_adj); + + /* D.1 fill in mbuf - rx_descriptor_fields1. */ +- *(vector unsigned char *) ++ *(__vector unsigned char *) + &pkts[pos + 3]->pkt_len = pkt_mb3; +- *(vector unsigned char *) ++ *(__vector unsigned char *) + &pkts[pos + 2]->pkt_len = pkt_mb2; + + /* E.1 extract op_own field. */ +- op_own_tmp2 = (vector unsigned char) +- vec_mergeh((vector unsigned int)cqes[2], +- (vector unsigned int)cqes[3]); ++ op_own_tmp2 = (__vector unsigned char) ++ vec_mergeh((__vector unsigned int)cqes[2], ++ (__vector unsigned int)cqes[3]); + + /* C.1 load remaining CQE data and extract necessary fields. */ +- cqe_tmp2 = *(vector unsigned char *) ++ cqe_tmp2 = *(__vector unsigned char *) + &cq[pos + p1].pkt_info; +- cqe_tmp1 = *(vector unsigned char *) ++ cqe_tmp1 = *(__vector unsigned char *) + &cq[pos].pkt_info; + cqes[1] = vec_sel(cqes[1], cqe_tmp2, blend_mask); + cqes[0] = vec_sel(cqes[0], cqe_tmp2, blend_mask); +- cqe_tmp2 = (vector unsigned char)vec_vsx_ld(0, ++ cqe_tmp2 = (__vector unsigned char)vec_vsx_ld(0, + (signed int const *)&cq[pos + p1].csum); +- cqe_tmp1 = (vector unsigned char)vec_vsx_ld(0, ++ cqe_tmp1 = (__vector unsigned char)vec_vsx_ld(0, + (signed int const *)&cq[pos].csum); +- cqes[1] = (vector unsigned char) +- vec_sel((vector unsigned short)cqes[1], +- (vector unsigned short)cqe_tmp2, cqe_sel_mask1); +- cqes[0] = (vector unsigned char) +- vec_sel((vector unsigned short)cqes[0], +- (vector unsigned short)cqe_tmp1, cqe_sel_mask1); +- cqe_tmp2 = (vector unsigned char)(vector unsigned long){ ++ cqes[1] = (__vector unsigned char) ++ vec_sel((__vector unsigned short)cqes[1], ++ (__vector unsigned short)cqe_tmp2, cqe_sel_mask1); ++ cqes[0] = (__vector unsigned char) ++ vec_sel((__vector unsigned short)cqes[0], ++ (__vector unsigned short)cqe_tmp1, cqe_sel_mask1); ++ cqe_tmp2 = (__vector unsigned char)(__vector unsigned long){ + *(__rte_aligned(8) unsigned long *) + &cq[pos + p1].rsvd4[2], 0LL}; +- cqe_tmp1 = (vector unsigned char)(vector unsigned long){ ++ cqe_tmp1 = (__vector unsigned char)(__vector unsigned long){ + *(__rte_aligned(8) unsigned long *) + &cq[pos].rsvd4[2], 0LL}; +- cqes[1] = (vector unsigned char) +- vec_sel((vector unsigned short)cqes[1], +- (vector unsigned short)cqe_tmp2, cqe_sel_mask2); +- cqes[0] = (vector unsigned char) +- vec_sel((vector unsigned short)cqes[0], +- (vector unsigned short)cqe_tmp1, cqe_sel_mask2); ++ cqes[1] = (__vector unsigned char) ++ vec_sel((__vector unsigned short)cqes[1], ++ (__vector unsigned short)cqe_tmp2, cqe_sel_mask2); ++ cqes[0] = (__vector unsigned char) ++ vec_sel((__vector unsigned short)cqes[0], ++ (__vector unsigned short)cqe_tmp1, cqe_sel_mask2); + + /* C.2 generate final structure for mbuf with swapping bytes. */ + pkt_mb1 = vec_perm(cqes[1], zero, shuf_mask); + pkt_mb0 = vec_perm(cqes[0], zero, shuf_mask); + + /* C.3 adjust CRC length. */ +- pkt_mb1 = (vector unsigned char) +- ((vector unsigned short)pkt_mb1 - +- (vector unsigned short)crc_adj); +- pkt_mb0 = (vector unsigned char) +- ((vector unsigned short)pkt_mb0 - +- (vector unsigned short)crc_adj); ++ pkt_mb1 = (__vector unsigned char) ++ ((__vector unsigned short)pkt_mb1 - ++ (__vector unsigned short)crc_adj); ++ pkt_mb0 = (__vector unsigned char) ++ ((__vector unsigned short)pkt_mb0 - ++ (__vector unsigned short)crc_adj); + + /* C.4 adjust flow mark. */ +- pkt_mb1 = (vector unsigned char) +- ((vector unsigned int)pkt_mb1 + +- (vector unsigned int)flow_mark_adj); +- pkt_mb0 = (vector unsigned char) +- ((vector unsigned int)pkt_mb0 + +- (vector unsigned int)flow_mark_adj); ++ pkt_mb1 = (__vector unsigned char) ++ ((__vector unsigned int)pkt_mb1 + ++ (__vector unsigned int)flow_mark_adj); ++ pkt_mb0 = (__vector unsigned char) ++ ((__vector unsigned int)pkt_mb0 + ++ (__vector unsigned int)flow_mark_adj); + + /* E.1 extract op_own byte. */ +- op_own_tmp1 = (vector unsigned char) +- vec_mergeh((vector unsigned int)cqes[0], +- (vector unsigned int)cqes[1]); +- op_own = (vector unsigned char) +- vec_mergel((vector unsigned long)op_own_tmp1, +- (vector unsigned long)op_own_tmp2); ++ op_own_tmp1 = (__vector unsigned char) ++ vec_mergeh((__vector unsigned int)cqes[0], ++ (__vector unsigned int)cqes[1]); ++ op_own = (__vector unsigned char) ++ vec_mergel((__vector unsigned long)op_own_tmp1, ++ (__vector unsigned long)op_own_tmp2); + + /* D.1 fill in mbuf - rx_descriptor_fields1. */ +- *(vector unsigned char *) ++ *(__vector unsigned char *) + &pkts[pos + 1]->pkt_len = pkt_mb1; +- *(vector unsigned char *) ++ *(__vector unsigned char *) + &pkts[pos]->pkt_len = pkt_mb0; + + /* E.2 flip owner bit to mark CQEs from last round. */ +- owner_mask = (vector unsigned char) +- vec_and((vector unsigned long)op_own, +- (vector unsigned long)owner_check); ++ owner_mask = (__vector unsigned char) ++ vec_and((__vector unsigned long)op_own, ++ (__vector unsigned long)owner_check); + if (ownership) +- owner_mask = (vector unsigned char) +- vec_xor((vector unsigned long)owner_mask, +- (vector unsigned long)owner_check); +- owner_mask = (vector unsigned char) +- vec_cmpeq((vector unsigned int)owner_mask, +- (vector unsigned int)owner_check); +- owner_mask = (vector unsigned char) +- vec_packs((vector unsigned int)owner_mask, +- (vector unsigned int)zero); ++ owner_mask = (__vector unsigned char) ++ vec_xor((__vector unsigned long)owner_mask, ++ (__vector unsigned long)owner_check); ++ owner_mask = (__vector unsigned char) ++ vec_cmpeq((__vector unsigned int)owner_mask, ++ (__vector unsigned int)owner_check); ++ owner_mask = (__vector unsigned char) ++ vec_packs((__vector unsigned int)owner_mask, ++ (__vector unsigned int)zero); + + /* E.3 get mask for invalidated CQEs. */ +- opcode = (vector unsigned char) +- vec_and((vector unsigned long)op_own, +- (vector unsigned long)opcode_check); +- invalid_mask = (vector unsigned char) +- vec_cmpeq((vector unsigned int)opcode_check, +- (vector unsigned int)opcode); +- invalid_mask = (vector unsigned char) +- vec_packs((vector unsigned int)invalid_mask, +- (vector unsigned int)zero); ++ opcode = (__vector unsigned char) ++ vec_and((__vector unsigned long)op_own, ++ (__vector unsigned long)opcode_check); ++ invalid_mask = (__vector unsigned char) ++ vec_cmpeq((__vector unsigned int)opcode_check, ++ (__vector unsigned int)opcode); ++ invalid_mask = (__vector unsigned char) ++ vec_packs((__vector unsigned int)invalid_mask, ++ (__vector unsigned int)zero); + + /* E.4 mask out beyond boundary. */ +- invalid_mask = (vector unsigned char) +- vec_or((vector unsigned long)invalid_mask, +- (vector unsigned long)mask); ++ invalid_mask = (__vector unsigned char) ++ vec_or((__vector unsigned long)invalid_mask, ++ (__vector unsigned long)mask); + + /* E.5 merge invalid_mask with invalid owner. */ +- invalid_mask = (vector unsigned char) +- vec_or((vector unsigned long)invalid_mask, +- (vector unsigned long)owner_mask); ++ invalid_mask = (__vector unsigned char) ++ vec_or((__vector unsigned long)invalid_mask, ++ (__vector unsigned long)owner_mask); + + /* F.1 find compressed CQE format. */ +- comp_mask = (vector unsigned char) +- vec_and((vector unsigned long)op_own, +- (vector unsigned long)format_check); +- comp_mask = (vector unsigned char) +- vec_cmpeq((vector unsigned int)comp_mask, +- (vector unsigned int)format_check); +- comp_mask = (vector unsigned char) +- vec_packs((vector unsigned int)comp_mask, +- (vector unsigned int)zero); ++ comp_mask = (__vector unsigned char) ++ vec_and((__vector unsigned long)op_own, ++ (__vector unsigned long)format_check); ++ comp_mask = (__vector unsigned char) ++ vec_cmpeq((__vector unsigned int)comp_mask, ++ (__vector unsigned int)format_check); ++ comp_mask = (__vector unsigned char) ++ vec_packs((__vector unsigned int)comp_mask, ++ (__vector unsigned int)zero); + + /* F.2 mask out invalid entries. */ +- comp_mask = (vector unsigned char) +- vec_andc((vector unsigned long)comp_mask, +- (vector unsigned long)invalid_mask); +- comp_idx = ((vector unsigned long)comp_mask)[0]; ++ comp_mask = (__vector unsigned char) ++ vec_andc((__vector unsigned long)comp_mask, ++ (__vector unsigned long)invalid_mask); ++ comp_idx = ((__vector unsigned long)comp_mask)[0]; + + /* F.3 get the first compressed CQE. */ + comp_idx = comp_idx ? __builtin_ctzll(comp_idx) / + (sizeof(uint16_t) * 8) : MLX5_VPMD_DESCS_PER_LOOP; + + /* E.6 mask out entries after the compressed CQE. */ +- mask = (vector unsigned char)(vector unsigned long){ ++ mask = (__vector unsigned char)(__vector unsigned long){ + (comp_idx * sizeof(uint16_t) * 8), 0}; +- lshift = vec_splat((vector unsigned long)mask, 0); ++ lshift = vec_splat((__vector unsigned long)mask, 0); + shmask = vec_cmpgt(shmax, lshift); +- mask = (vector unsigned char) +- vec_sl((vector unsigned long)ones, lshift); +- mask = (vector unsigned char) +- vec_sel((vector unsigned long)shmask, +- (vector unsigned long)mask, shmask); +- invalid_mask = (vector unsigned char) +- vec_or((vector unsigned long)invalid_mask, +- (vector unsigned long)mask); ++ mask = (__vector unsigned char) ++ vec_sl((__vector unsigned long)ones, lshift); ++ mask = (__vector unsigned char) ++ vec_sel((__vector unsigned long)shmask, ++ (__vector unsigned long)mask, shmask); ++ invalid_mask = (__vector unsigned char) ++ vec_or((__vector unsigned long)invalid_mask, ++ (__vector unsigned long)mask); + + /* E.7 count non-compressed valid CQEs. */ +- n = ((vector unsigned long)invalid_mask)[0]; ++ n = ((__vector unsigned long)invalid_mask)[0]; + n = n ? __builtin_ctzll(n) / (sizeof(uint16_t) * 8) : + MLX5_VPMD_DESCS_PER_LOOP; + nocmp_n += n; + + /* D.2 get the final invalid mask. */ +- mask = (vector unsigned char)(vector unsigned long){ ++ mask = (__vector unsigned char)(__vector unsigned long){ + (n * sizeof(uint16_t) * 8), 0}; +- lshift = vec_splat((vector unsigned long)mask, 0); ++ lshift = vec_splat((__vector unsigned long)mask, 0); + shmask = vec_cmpgt(shmax, lshift); +- mask = (vector unsigned char) +- vec_sl((vector unsigned long)ones, lshift); +- mask = (vector unsigned char) +- vec_sel((vector unsigned long)shmask, +- (vector unsigned long)mask, shmask); +- invalid_mask = (vector unsigned char) +- vec_or((vector unsigned long)invalid_mask, +- (vector unsigned long)mask); ++ mask = (__vector unsigned char) ++ vec_sl((__vector unsigned long)ones, lshift); ++ mask = (__vector unsigned char) ++ vec_sel((__vector unsigned long)shmask, ++ (__vector unsigned long)mask, shmask); ++ invalid_mask = (__vector unsigned char) ++ vec_or((__vector unsigned long)invalid_mask, ++ (__vector unsigned long)mask); + + /* D.3 check error in opcode. */ +- opcode = (vector unsigned char) +- vec_cmpeq((vector unsigned int)resp_err_check, +- (vector unsigned int)opcode); +- opcode = (vector unsigned char) +- vec_packs((vector unsigned int)opcode, +- (vector unsigned int)zero); +- opcode = (vector unsigned char) +- vec_andc((vector unsigned long)opcode, +- (vector unsigned long)invalid_mask); ++ opcode = (__vector unsigned char) ++ vec_cmpeq((__vector unsigned int)resp_err_check, ++ (__vector unsigned int)opcode); ++ opcode = (__vector unsigned char) ++ vec_packs((__vector unsigned int)opcode, ++ (__vector unsigned int)zero); ++ opcode = (__vector unsigned char) ++ vec_andc((__vector unsigned long)opcode, ++ (__vector unsigned long)invalid_mask); + + /* D.4 mark if any error is set */ +- *err |= ((vector unsigned long)opcode)[0]; ++ *err |= ((__vector unsigned long)opcode)[0]; + + /* D.5 fill in mbuf - rearm_data and packet_type. */ + rxq_cq_to_ptype_oflags_v(rxq, cqes, opcode, &pkts[pos]); +@@ -1230,7 +1230,7 @@ rxq_cq_process_v(struct mlx5_rxq_data *rxq, volatile struct mlx5_cqe *cq, + uint32_t mask = rxq->flow_meta_port_mask; + uint32_t metadata; + +- /* This code is subject for futher optimization. */ ++ /* This code is subject for further optimization. */ + metadata = rte_be_to_cpu_32 + (cq[pos].flow_table_metadata) & mask; + *RTE_MBUF_DYNFIELD(pkts[pos], offs, uint32_t *) = +@@ -1255,20 +1255,20 @@ rxq_cq_process_v(struct mlx5_rxq_data *rxq, volatile struct mlx5_cqe *cq, + #ifdef MLX5_PMD_SOFT_COUNTERS + /* Add up received bytes count. */ + byte_cnt = vec_perm(op_own, zero, len_shuf_mask); +- byte_cnt = (vector unsigned char) +- vec_andc((vector unsigned long)byte_cnt, +- (vector unsigned long)invalid_mask); +- left = vec_perm((vector unsigned short)byte_cnt, +- (vector unsigned short)zero, lower_half); +- right = vec_perm((vector unsigned short)byte_cnt, +- (vector unsigned short)zero, upper_half); +- byte_cnt = (vector unsigned char)vec_add(left, right); +- left = vec_perm((vector unsigned short)byte_cnt, +- (vector unsigned short)zero, lower_half); +- right = vec_perm((vector unsigned short)byte_cnt, +- (vector unsigned short)zero, upper_half); +- byte_cnt = (vector unsigned char)vec_add(left, right); +- rcvd_byte += ((vector unsigned long)byte_cnt)[0]; ++ byte_cnt = (__vector unsigned char) ++ vec_andc((__vector unsigned long)byte_cnt, ++ (__vector unsigned long)invalid_mask); ++ left = vec_perm((__vector unsigned short)byte_cnt, ++ (__vector unsigned short)zero, lower_half); ++ right = vec_perm((__vector unsigned short)byte_cnt, ++ (__vector unsigned short)zero, upper_half); ++ byte_cnt = (__vector unsigned char)vec_add(left, right); ++ left = vec_perm((__vector unsigned short)byte_cnt, ++ (__vector unsigned short)zero, lower_half); ++ right = vec_perm((__vector unsigned short)byte_cnt, ++ (__vector unsigned short)zero, upper_half); ++ byte_cnt = (__vector unsigned char)vec_add(left, right); ++ rcvd_byte += ((__vector unsigned long)byte_cnt)[0]; + #endif -- struct raft_conn *conn, *next; -- LIST_FOR_EACH_SAFE (conn, next, list_node, &raft->conns) { -+ struct raft_conn *conn; -+ LIST_FOR_EACH_SAFE (conn, list_node, &raft->conns) { - raft_conn_close(conn); - } + /* +diff --git a/dpdk/drivers/net/mlx5/mlx5_rxtx_vec_neon.h b/dpdk/drivers/net/mlx5/mlx5_rxtx_vec_neon.h +index b1d16baa61..f7bbde4e0e 100644 +--- a/dpdk/drivers/net/mlx5/mlx5_rxtx_vec_neon.h ++++ b/dpdk/drivers/net/mlx5/mlx5_rxtx_vec_neon.h +@@ -839,7 +839,7 @@ rxq_cq_process_v(struct mlx5_rxq_data *rxq, volatile struct mlx5_cqe *cq, + } + } + if (rxq->dynf_meta) { +- /* This code is subject for futher optimization. */ ++ /* This code is subject for further optimization. */ + int32_t offs = rxq->flow_meta_offset; + uint32_t mask = rxq->flow_meta_port_mask; + +diff --git a/dpdk/drivers/net/mlx5/mlx5_rxtx_vec_sse.h b/dpdk/drivers/net/mlx5/mlx5_rxtx_vec_sse.h +index f3d838389e..185d2695db 100644 +--- a/dpdk/drivers/net/mlx5/mlx5_rxtx_vec_sse.h ++++ b/dpdk/drivers/net/mlx5/mlx5_rxtx_vec_sse.h +@@ -772,7 +772,7 @@ rxq_cq_process_v(struct mlx5_rxq_data *rxq, volatile struct mlx5_cqe *cq, + } + } + if (rxq->dynf_meta) { +- /* This code is subject for futher optimization. */ ++ /* This code is subject for further optimization. */ + int32_t offs = rxq->flow_meta_offset; + uint32_t mask = rxq->flow_meta_port_mask; + +diff --git a/dpdk/drivers/net/mlx5/mlx5_stats.c b/dpdk/drivers/net/mlx5/mlx5_stats.c +index 732775954a..f64fa3587b 100644 +--- a/dpdk/drivers/net/mlx5/mlx5_stats.c ++++ b/dpdk/drivers/net/mlx5/mlx5_stats.c +@@ -114,18 +114,23 @@ mlx5_stats_get(struct rte_eth_dev *dev, struct rte_eth_stats *stats) + idx = rxq->idx; + if (idx < RTE_ETHDEV_QUEUE_STAT_CNTRS) { + #ifdef MLX5_PMD_SOFT_COUNTERS +- tmp.q_ipackets[idx] += rxq->stats.ipackets; +- tmp.q_ibytes[idx] += rxq->stats.ibytes; ++ tmp.q_ipackets[idx] += rxq->stats.ipackets - ++ rxq->stats_reset.ipackets; ++ tmp.q_ibytes[idx] += rxq->stats.ibytes - ++ rxq->stats_reset.ibytes; + #endif + tmp.q_errors[idx] += (rxq->stats.idropped + +- rxq->stats.rx_nombuf); ++ rxq->stats.rx_nombuf) - ++ (rxq->stats_reset.idropped + ++ rxq->stats_reset.rx_nombuf); + } + #ifdef MLX5_PMD_SOFT_COUNTERS +- tmp.ipackets += rxq->stats.ipackets; +- tmp.ibytes += rxq->stats.ibytes; ++ tmp.ipackets += rxq->stats.ipackets - rxq->stats_reset.ipackets; ++ tmp.ibytes += rxq->stats.ibytes - rxq->stats_reset.ibytes; + #endif +- tmp.ierrors += rxq->stats.idropped; +- tmp.rx_nombuf += rxq->stats.rx_nombuf; ++ tmp.ierrors += rxq->stats.idropped - rxq->stats_reset.idropped; ++ tmp.rx_nombuf += rxq->stats.rx_nombuf - ++ rxq->stats_reset.rx_nombuf; + } + for (i = 0; (i != priv->txqs_n); ++i) { + struct mlx5_txq_data *txq = (*priv->txqs)[i]; +@@ -135,15 +140,17 @@ mlx5_stats_get(struct rte_eth_dev *dev, struct rte_eth_stats *stats) + idx = txq->idx; + if (idx < RTE_ETHDEV_QUEUE_STAT_CNTRS) { + #ifdef MLX5_PMD_SOFT_COUNTERS +- tmp.q_opackets[idx] += txq->stats.opackets; +- tmp.q_obytes[idx] += txq->stats.obytes; ++ tmp.q_opackets[idx] += txq->stats.opackets - ++ txq->stats_reset.opackets; ++ tmp.q_obytes[idx] += txq->stats.obytes - ++ txq->stats_reset.obytes; + #endif + } + #ifdef MLX5_PMD_SOFT_COUNTERS +- tmp.opackets += txq->stats.opackets; +- tmp.obytes += txq->stats.obytes; ++ tmp.opackets += txq->stats.opackets - txq->stats_reset.opackets; ++ tmp.obytes += txq->stats.obytes - txq->stats_reset.obytes; + #endif +- tmp.oerrors += txq->stats.oerrors; ++ tmp.oerrors += txq->stats.oerrors - txq->stats_reset.oerrors; + } + ret = mlx5_os_read_dev_stat(priv, "out_of_buffer", &tmp.imissed); + if (ret == 0) { +@@ -185,13 +192,14 @@ mlx5_stats_reset(struct rte_eth_dev *dev) + + if (rxq_data == NULL) + continue; +- memset(&rxq_data->stats, 0, sizeof(struct mlx5_rxq_stats)); ++ rxq_data->stats_reset = rxq_data->stats; + } + for (i = 0; (i != priv->txqs_n); ++i) { +- if ((*priv->txqs)[i] == NULL) ++ struct mlx5_txq_data *txq_data = (*priv->txqs)[i]; ++ ++ if (txq_data == NULL) + continue; +- memset(&(*priv->txqs)[i]->stats, 0, +- sizeof(struct mlx5_txq_stats)); ++ txq_data->stats_reset = txq_data->stats; + } + mlx5_os_read_dev_stat(priv, "out_of_buffer", &stats_ctrl->imissed_base); + stats_ctrl->imissed = 0; +diff --git a/dpdk/drivers/net/mlx5/mlx5_trigger.c b/dpdk/drivers/net/mlx5/mlx5_trigger.c +index 74c9c0a4ff..2ba456ad7a 100644 +--- a/dpdk/drivers/net/mlx5/mlx5_trigger.c ++++ b/dpdk/drivers/net/mlx5/mlx5_trigger.c +@@ -1205,11 +1205,18 @@ mlx5_dev_start(struct rte_eth_dev *dev) + priv->sh->port[priv->dev_port - 1].ih_port_id = + (uint32_t)dev->data->port_id; + } else { +- DRV_LOG(INFO, "port %u starts without LSC and RMV interrupts.", ++ DRV_LOG(INFO, "port %u starts without RMV interrupts.", + dev->data->port_id); +- dev->data->dev_conf.intr_conf.lsc = 0; + dev->data->dev_conf.intr_conf.rmv = 0; + } ++ if (rte_intr_fd_get(priv->sh->intr_handle_nl) >= 0) { ++ priv->sh->port[priv->dev_port - 1].nl_ih_port_id = ++ (uint32_t)dev->data->port_id; ++ } else { ++ DRV_LOG(INFO, "port %u starts without LSC interrupts.", ++ dev->data->port_id); ++ dev->data->dev_conf.intr_conf.lsc = 0; ++ } + if (rte_intr_fd_get(priv->sh->intr_handle_devx) >= 0) + priv->sh->port[priv->dev_port - 1].devx_ih_port_id = + (uint32_t)dev->data->port_id; +@@ -1261,6 +1268,7 @@ mlx5_dev_stop(struct rte_eth_dev *dev) + mlx5_rx_intr_vec_disable(dev); + priv->sh->port[priv->dev_port - 1].ih_port_id = RTE_MAX_ETHPORTS; + priv->sh->port[priv->dev_port - 1].devx_ih_port_id = RTE_MAX_ETHPORTS; ++ priv->sh->port[priv->dev_port - 1].nl_ih_port_id = RTE_MAX_ETHPORTS; + mlx5_txq_stop(dev); + mlx5_rxq_stop(dev); + if (priv->obj_ops.lb_dummy_queue_release) +diff --git a/dpdk/drivers/net/mlx5/mlx5_tx.c b/dpdk/drivers/net/mlx5/mlx5_tx.c +index 5492d64cae..fd2cf20967 100644 +--- a/dpdk/drivers/net/mlx5/mlx5_tx.c ++++ b/dpdk/drivers/net/mlx5/mlx5_tx.c +@@ -728,7 +728,7 @@ mlx5_txq_info_get(struct rte_eth_dev *dev, uint16_t tx_queue_id, + * Pointer to the device structure. + * + * @param tx_queue_id +- * Tx queue identificatior. ++ * Tx queue identification. + * + * @param mode + * Pointer to the burts mode information. +diff --git a/dpdk/drivers/net/mlx5/mlx5_tx.h b/dpdk/drivers/net/mlx5/mlx5_tx.h +index 099e72935a..6ed00f722e 100644 +--- a/dpdk/drivers/net/mlx5/mlx5_tx.h ++++ b/dpdk/drivers/net/mlx5/mlx5_tx.h +@@ -161,6 +161,7 @@ struct mlx5_txq_data { + int32_t ts_offset; /* Timestamp field dynamic offset. */ + struct mlx5_dev_ctx_shared *sh; /* Shared context. */ + struct mlx5_txq_stats stats; /* TX queue counters. */ ++ struct mlx5_txq_stats stats_reset; /* stats on last reset. */ + struct mlx5_uar_data uar_data; + struct rte_mbuf *elts[0]; + /* Storage for queued packets, must be the last field. */ +@@ -1710,7 +1711,6 @@ mlx5_tx_packet_multi_tso(struct mlx5_txq_data *__rte_restrict txq, + inlen <= MLX5_ESEG_MIN_INLINE_SIZE || + inlen > (dlen + vlan))) + return MLX5_TXCMP_CODE_ERROR; +- MLX5_ASSERT(inlen >= txq->inlen_mode); + /* + * Check whether there are enough free WQEBBs: + * - Control Segment +@@ -2019,7 +2019,7 @@ mlx5_tx_packet_multi_inline(struct mlx5_txq_data *__rte_restrict txq, + if (unlikely(loc->wqe_free < ((ds + 3) / 4))) + return MLX5_TXCMP_CODE_EXIT; + /* Check for maximal WQE size. */ +- if (unlikely((MLX5_WQE_SIZE_MAX / MLX5_WSEG_SIZE) < ((ds + 3) / 4))) ++ if (unlikely((MLX5_WQE_SIZE_MAX / MLX5_WSEG_SIZE) < ds)) + return MLX5_TXCMP_CODE_ERROR; + #ifdef MLX5_PMD_SOFT_COUNTERS + /* Update sent data bytes/packets counters. */ +diff --git a/dpdk/drivers/net/mlx5/mlx5_utils.c b/dpdk/drivers/net/mlx5/mlx5_utils.c +index e4e66ae4c5..4115a2ad77 100644 +--- a/dpdk/drivers/net/mlx5/mlx5_utils.c ++++ b/dpdk/drivers/net/mlx5/mlx5_utils.c +@@ -340,6 +340,8 @@ mlx5_ipool_allocate_from_global(struct mlx5_indexed_pool *pool, int cidx) + /* Enqueue half of the index to global. */ + ts_idx = mlx5_trunk_idx_offset_get(pool, trunk_idx) + 1; + fetch_size = trunk->free >> 1; ++ if (fetch_size > pool->cfg.per_core_cache) ++ fetch_size = trunk->free - pool->cfg.per_core_cache; + for (i = 0; i < fetch_size; i++) + lc->idx[i] = ts_idx + i; + lc->len = fetch_size; +@@ -1184,44 +1186,3 @@ mlx5_l3t_set_entry(struct mlx5_l3t_tbl *tbl, uint32_t idx, + rte_spinlock_unlock(&tbl->sl); + return ret; + } +- +-int32_t +-mlx5_l3t_prepare_entry(struct mlx5_l3t_tbl *tbl, uint32_t idx, +- union mlx5_l3t_data *data, +- mlx5_l3t_alloc_callback_fn cb, void *ctx) +-{ +- int32_t ret; +- +- rte_spinlock_lock(&tbl->sl); +- /* Check if entry data is ready. */ +- ret = __l3t_get_entry(tbl, idx, data); +- if (!ret) { +- switch (tbl->type) { +- case MLX5_L3T_TYPE_WORD: +- if (data->word) +- goto out; +- break; +- case MLX5_L3T_TYPE_DWORD: +- if (data->dword) +- goto out; +- break; +- case MLX5_L3T_TYPE_QWORD: +- if (data->qword) +- goto out; +- break; +- default: +- if (data->ptr) +- goto out; +- break; +- } +- } +- /* Entry data is not ready, use user callback to create it. */ +- ret = cb(ctx, data); +- if (ret) +- goto out; +- /* Save the new allocated data to entry. */ +- ret = __l3t_set_entry(tbl, idx, data); +-out: +- rte_spinlock_unlock(&tbl->sl); +- return ret; +-} +diff --git a/dpdk/drivers/net/mlx5/mlx5_utils.h b/dpdk/drivers/net/mlx5/mlx5_utils.h +index cf3db89403..254c879d1a 100644 +--- a/dpdk/drivers/net/mlx5/mlx5_utils.h ++++ b/dpdk/drivers/net/mlx5/mlx5_utils.h +@@ -55,7 +55,7 @@ extern int mlx5_logtype; + + /* + * For the case which data is linked with sequence increased index, the +- * array table will be more efficiect than hash table once need to serarch ++ * array table will be more efficient than hash table once need to search + * one data entry in large numbers of entries. Since the traditional hash + * tables has fixed table size, when huge numbers of data saved to the hash + * table, it also comes lots of hash conflict. +@@ -459,34 +459,6 @@ void mlx5_l3t_destroy(struct mlx5_l3t_tbl *tbl); + int32_t mlx5_l3t_get_entry(struct mlx5_l3t_tbl *tbl, uint32_t idx, + union mlx5_l3t_data *data); + +-/** +- * This function gets the index entry from Three-level table. +- * +- * If the index entry is not available, allocate new one by callback +- * function and fill in the entry. +- * +- * @param tbl +- * Pointer to the l3t. +- * @param idx +- * Index to the entry. +- * @param data +- * Pointer to the memory which saves the entry data. +- * When function call returns 0, data contains the entry data get from +- * l3t. +- * When function call returns -1, data is not modified. +- * @param cb +- * Callback function to allocate new data. +- * @param ctx +- * Context for callback function. +- * +- * @return +- * 0 if success, -1 on error. +- */ +- +-int32_t mlx5_l3t_prepare_entry(struct mlx5_l3t_tbl *tbl, uint32_t idx, +- union mlx5_l3t_data *data, +- mlx5_l3t_alloc_callback_fn cb, void *ctx); +- + /** + * This function decreases and clear index entry if reference + * counter is 0 from Three-level table. +diff --git a/dpdk/drivers/net/mlx5/windows/mlx5_flow_os.c b/dpdk/drivers/net/mlx5/windows/mlx5_flow_os.c +index c4d5790726..f5e3893ed4 100644 +--- a/dpdk/drivers/net/mlx5/windows/mlx5_flow_os.c ++++ b/dpdk/drivers/net/mlx5/windows/mlx5_flow_os.c +@@ -372,7 +372,7 @@ mlx5_flow_os_init_workspace_once(void) + + if (err) { + DRV_LOG(ERR, "Can't create flow workspace data thread key."); +- return err; ++ return -rte_errno; + } + pthread_mutex_init(&lock_thread_list, NULL); + return 0; +@@ -400,7 +400,7 @@ mlx5_flow_os_set_specific_workspace(struct mlx5_flow_workspace *data) + /* + * set_specific_workspace when current value is NULL + * can happen only once per thread, mark this thread in +- * linked list to be able to release reasorces later on. ++ * linked list to be able to release resources later on. + */ + err = mlx5_add_workspace_to_list(data); + if (err) { +diff --git a/dpdk/drivers/net/mlx5/windows/mlx5_os.c b/dpdk/drivers/net/mlx5/windows/mlx5_os.c +index dec4b923d0..ba99901c5c 100644 +--- a/dpdk/drivers/net/mlx5/windows/mlx5_os.c ++++ b/dpdk/drivers/net/mlx5/windows/mlx5_os.c +@@ -136,7 +136,7 @@ mlx5_init_once(void) + * Pointer to mlx5 device attributes. + * + * @return +- * 0 on success, non zero error number otherwise. ++ * 0 on success, a negative errno value otherwise and rte_errno is set. + */ + int + mlx5_os_get_dev_attr(struct mlx5_common_device *cdev, +@@ -145,10 +145,11 @@ mlx5_os_get_dev_attr(struct mlx5_common_device *cdev, + struct mlx5_context *mlx5_ctx; + void *pv_iseg = NULL; + u32 cb_iseg = 0; +- int err = 0; + +- if (!cdev || !cdev->ctx) +- return -EINVAL; ++ if (!cdev || !cdev->ctx) { ++ rte_errno = EINVAL; ++ return -rte_errno; ++ } + mlx5_ctx = (struct mlx5_context *)cdev->ctx; + memset(device_attr, 0, sizeof(*device_attr)); + device_attr->max_cq = 1 << cdev->config.hca_attr.log_max_cq; +@@ -171,15 +172,14 @@ mlx5_os_get_dev_attr(struct mlx5_common_device *cdev, + pv_iseg = mlx5_glue->query_hca_iseg(mlx5_ctx, &cb_iseg); + if (pv_iseg == NULL) { + DRV_LOG(ERR, "Failed to get device hca_iseg"); +- return errno; +- } +- if (!err) { +- snprintf(device_attr->fw_ver, 64, "%x.%x.%04x", +- MLX5_GET(initial_seg, pv_iseg, fw_rev_major), +- MLX5_GET(initial_seg, pv_iseg, fw_rev_minor), +- MLX5_GET(initial_seg, pv_iseg, fw_rev_subminor)); ++ rte_errno = errno; ++ return -rte_errno; + } +- return err; ++ snprintf(device_attr->fw_ver, 64, "%x.%x.%04x", ++ MLX5_GET(initial_seg, pv_iseg, fw_rev_major), ++ MLX5_GET(initial_seg, pv_iseg, fw_rev_minor), ++ MLX5_GET(initial_seg, pv_iseg, fw_rev_subminor)); ++ return 0; } -@@ -1713,8 +1723,8 @@ raft_waiters_run(struct raft *raft) - } - uint64_t cur = ovsdb_log_commit_progress(raft->log); -- struct raft_waiter *w, *next; -- LIST_FOR_EACH_SAFE (w, next, list_node, &raft->waiters) { -+ struct raft_waiter *w; -+ LIST_FOR_EACH_SAFE (w, list_node, &raft->waiters) { - if (cur < w->commit_ticket) { - break; - } -@@ -1736,8 +1746,8 @@ raft_waiters_wait(struct raft *raft) - static void - raft_waiters_destroy(struct raft *raft) + /** +@@ -226,7 +226,7 @@ mlx5_os_free_shared_dr(struct mlx5_priv *priv) + * Pointer to RQ channel object, which includes the channel fd + * + * @param[out] fd +- * The file descriptor (representing the intetrrupt) used in this channel. ++ * The file descriptor (representing the interrupt) used in this channel. + * + * @return + * 0 on successfully setting the fd to non-blocking, non-zero otherwise. +@@ -423,6 +423,21 @@ mlx5_dev_spawn(struct rte_device *dpdk_dev, + } + /* Override some values set by hardware configuration. */ + mlx5_args(config, dpdk_dev->devargs); ++ /* Update final values for devargs before check sibling config. */ ++ config->dv_esw_en = 0; ++ if (!config->dv_flow_en) { ++ DRV_LOG(ERR, "Windows flow mode must be DV flow enable."); ++ err = ENOTSUP; ++ goto error; ++ } ++ if (!priv->config.dv_esw_en && ++ priv->config.dv_xmeta_en != MLX5_XMETA_MODE_LEGACY) { ++ DRV_LOG(WARNING, ++ "Metadata mode %u is not supported (no E-Switch).", ++ priv->config.dv_xmeta_en); ++ priv->config.dv_xmeta_en = MLX5_XMETA_MODE_LEGACY; ++ } ++ /* Check sibling device configurations. */ + err = mlx5_dev_check_sibling_config(priv, config, dpdk_dev); + if (err) + goto error; +@@ -584,7 +599,6 @@ mlx5_dev_spawn(struct rte_device *dpdk_dev, + * Verbs context returned by ibv_open_device(). + */ + mlx5_link_update(eth_dev, 0); +- config->dv_esw_en = 0; + /* Detect minimal data bytes to inline. */ + mlx5_set_min_inline(spawn, config); + /* Store device configuration on private structure. */ +@@ -606,12 +620,6 @@ mlx5_dev_spawn(struct rte_device *dpdk_dev, + } + /* No supported flow priority number detection. */ + priv->sh->flow_max_priority = -1; +- if (!priv->config.dv_esw_en && +- priv->config.dv_xmeta_en != MLX5_XMETA_MODE_LEGACY) { +- DRV_LOG(WARNING, "metadata mode %u is not supported " +- "(no E-Switch)", priv->config.dv_xmeta_en); +- priv->config.dv_xmeta_en = MLX5_XMETA_MODE_LEGACY; +- } + mlx5_set_metadata_mask(eth_dev); + if (priv->config.dv_xmeta_en != MLX5_XMETA_MODE_LEGACY && + !priv->sh->dv_regc0_mask) { +@@ -645,12 +653,10 @@ mlx5_dev_spawn(struct rte_device *dpdk_dev, + goto error; + } + } +- if (sh->devx && config->dv_flow_en) { ++ if (sh->devx) { + priv->obj_ops = devx_obj_ops; + } else { +- DRV_LOG(ERR, "Flow mode %u is not supported " +- "(Windows flow must be DevX with DV flow enabled).", +- priv->config.dv_flow_en); ++ DRV_LOG(ERR, "Windows flow must be DevX."); + err = ENOTSUP; + goto error; + } +diff --git a/dpdk/drivers/net/mvneta/mvneta_ethdev.c b/dpdk/drivers/net/mvneta/mvneta_ethdev.c +index 10fe6d828c..eef016aa0b 100644 +--- a/dpdk/drivers/net/mvneta/mvneta_ethdev.c ++++ b/dpdk/drivers/net/mvneta/mvneta_ethdev.c +@@ -247,7 +247,7 @@ mvneta_mtu_set(struct rte_eth_dev *dev, uint16_t mtu) + (mru + MRVL_NETA_PKT_OFFS > mbuf_data_size)) { + mru = mbuf_data_size - MRVL_NETA_PKT_OFFS; + mtu = MRVL_NETA_MRU_TO_MTU(mru); +- MVNETA_LOG(WARNING, "MTU too big, max MTU possible limitted by" ++ MVNETA_LOG(WARNING, "MTU too big, max MTU possible limited by" + " current mbuf size: %u. Set MTU to %u, MRU to %u", + mbuf_data_size, mtu, mru); + } +diff --git a/dpdk/drivers/net/mvpp2/mrvl_ethdev.c b/dpdk/drivers/net/mvpp2/mrvl_ethdev.c +index 9c7fe13f7f..735efb6cfc 100644 +--- a/dpdk/drivers/net/mvpp2/mrvl_ethdev.c ++++ b/dpdk/drivers/net/mvpp2/mrvl_ethdev.c +@@ -579,7 +579,7 @@ mrvl_mtu_set(struct rte_eth_dev *dev, uint16_t mtu) + if (mru - RTE_ETHER_CRC_LEN + MRVL_PKT_OFFS > mbuf_data_size) { + mru = mbuf_data_size + RTE_ETHER_CRC_LEN - MRVL_PKT_OFFS; + mtu = MRVL_PP2_MRU_TO_MTU(mru); +- MRVL_LOG(WARNING, "MTU too big, max MTU possible limitted " ++ MRVL_LOG(WARNING, "MTU too big, max MTU possible limited " + "by current mbuf size: %u. Set MTU to %u, MRU to %u", + mbuf_data_size, mtu, mru); + } +@@ -1626,13 +1626,14 @@ mrvl_xstats_get(struct rte_eth_dev *dev, { -- struct raft_waiter *w, *next; -- LIST_FOR_EACH_SAFE (w, next, list_node, &raft->waiters) { -+ struct raft_waiter *w; -+ LIST_FOR_EACH_SAFE (w, list_node, &raft->waiters) { - raft_waiter_destroy(w); - } + struct mrvl_priv *priv = dev->data->dev_private; + struct pp2_ppio_statistics ppio_stats; +- unsigned int i; ++ unsigned int i, count; + +- if (!stats) +- return 0; ++ count = RTE_DIM(mrvl_xstats_tbl); ++ if (n < count) ++ return count; + + pp2_ppio_get_statistics(priv->ppio, &ppio_stats, 0); +- for (i = 0; i < n && i < RTE_DIM(mrvl_xstats_tbl); i++) { ++ for (i = 0; i < count; i++) { + uint64_t val; + + if (mrvl_xstats_tbl[i].size == sizeof(uint32_t)) +@@ -1648,7 +1649,7 @@ mrvl_xstats_get(struct rte_eth_dev *dev, + stats[i].value = val; + } + +- return n; ++ return count; } -@@ -1867,6 +1877,8 @@ raft_open_conn(struct raft *raft, const char *address, const struct uuid *sid) + + /** +diff --git a/dpdk/drivers/net/mvpp2/mrvl_qos.c b/dpdk/drivers/net/mvpp2/mrvl_qos.c +index dbfc3b5d20..99f0ee56d1 100644 +--- a/dpdk/drivers/net/mvpp2/mrvl_qos.c ++++ b/dpdk/drivers/net/mvpp2/mrvl_qos.c +@@ -301,7 +301,7 @@ get_entry_values(const char *entry, uint8_t *tab, + } + + /** +- * Parse Traffic Class'es mapping configuration. ++ * Parse Traffic Classes mapping configuration. + * + * @param file Config file handle. + * @param port Which port to look for. +@@ -736,7 +736,7 @@ mrvl_get_cfg(const char *key __rte_unused, const char *path, void *extra_args) + + /* MRVL_TOK_START_HDR replaces MRVL_TOK_DSA_MODE parameter. + * MRVL_TOK_DSA_MODE will be supported for backward +- * compatibillity. ++ * compatibility. + */ + entry = rte_cfgfile_get_entry(file, sec_name, + MRVL_TOK_START_HDR); +diff --git a/dpdk/drivers/net/netvsc/hn_ethdev.c b/dpdk/drivers/net/netvsc/hn_ethdev.c +index 8a950403ac..787139c0b2 100644 +--- a/dpdk/drivers/net/netvsc/hn_ethdev.c ++++ b/dpdk/drivers/net/netvsc/hn_ethdev.c +@@ -554,9 +554,10 @@ static int hn_subchan_configure(struct hn_data *hv, + static void netvsc_hotplug_retry(void *args) + { + int ret; +- struct hn_data *hv = args; ++ struct hv_hotadd_context *hot_ctx = args; ++ struct hn_data *hv = hot_ctx->hv; + struct rte_eth_dev *dev = &rte_eth_devices[hv->port_id]; +- struct rte_devargs *d = &hv->devargs; ++ struct rte_devargs *d = &hot_ctx->da; + char buf[256]; + + DIR *di; +@@ -566,10 +567,13 @@ static void netvsc_hotplug_retry(void *args) + int s; + + PMD_DRV_LOG(DEBUG, "%s: retry count %d", +- __func__, hv->eal_hot_plug_retry); ++ __func__, hot_ctx->eal_hot_plug_retry); + +- if (hv->eal_hot_plug_retry++ > NETVSC_MAX_HOTADD_RETRY) +- return; ++ if (hot_ctx->eal_hot_plug_retry++ > NETVSC_MAX_HOTADD_RETRY) { ++ PMD_DRV_LOG(NOTICE, "Failed to parse PCI device retry=%d", ++ hot_ctx->eal_hot_plug_retry); ++ goto free_hotadd_ctx; ++ } + + snprintf(buf, sizeof(buf), "/sys/bus/pci/devices/%s/net", d->name); + di = opendir(buf); +@@ -602,7 +606,7 @@ static void netvsc_hotplug_retry(void *args) + } + if (req.ifr_hwaddr.sa_family != ARPHRD_ETHER) { + closedir(di); +- return; ++ goto free_hotadd_ctx; + } + memcpy(eth_addr.addr_bytes, req.ifr_hwaddr.sa_data, + RTE_DIM(eth_addr.addr_bytes)); +@@ -611,8 +615,13 @@ static void netvsc_hotplug_retry(void *args) + PMD_DRV_LOG(NOTICE, + "Found matching MAC address, adding device %s network name %s", + d->name, dir->d_name); ++ ++ /* If this device has been hot removed from this ++ * parent device, restore its args. ++ */ + ret = rte_eal_hotplug_add(d->bus->name, d->name, +- d->args); ++ hv->vf_devargs ? ++ hv->vf_devargs : ""); + if (ret) { + PMD_DRV_LOG(ERR, + "Failed to add PCI device %s", +@@ -624,12 +633,20 @@ static void netvsc_hotplug_retry(void *args) + * the device, or its MAC address did not match. + */ + closedir(di); +- return; ++ goto free_hotadd_ctx; + } + closedir(di); + retry: + /* The device is still being initialized, retry after 1 second */ +- rte_eal_alarm_set(1000000, netvsc_hotplug_retry, hv); ++ rte_eal_alarm_set(1000000, netvsc_hotplug_retry, hot_ctx); ++ return; ++ ++free_hotadd_ctx: ++ rte_spinlock_lock(&hv->hotadd_lock); ++ LIST_REMOVE(hot_ctx, list); ++ rte_spinlock_unlock(&hv->hotadd_lock); ++ ++ rte_free(hot_ctx); + } + static void - raft_conn_close(struct raft_conn *conn) +@@ -637,7 +654,8 @@ netvsc_hotadd_callback(const char *device_name, enum rte_dev_event_type type, + void *arg) { -+ VLOG_DBG("closing connection to server %s (%s)", -+ conn->nickname, jsonrpc_session_get_name(conn->js)); - jsonrpc_session_close(conn->js); - ovs_list_remove(&conn->list_node); - free(conn->nickname); -@@ -1921,6 +1933,13 @@ raft_run(struct raft *raft) - return; - } - -+ if (failure_test == FT_TRANSFER_LEADERSHIP) { -+ /* Using this function as it conveniently implements all we need and -+ * snapshotting is the main test scenario for leadership transfer. */ -+ raft_notify_snapshot_recommended(raft); -+ failure_test = FT_NO_TEST; -+ } + struct hn_data *hv = arg; +- struct rte_devargs *d = &hv->devargs; ++ struct hv_hotadd_context *hot_ctx; ++ struct rte_devargs *d; + int ret; + + PMD_DRV_LOG(INFO, "Device notification type=%d device_name=%s", +@@ -649,26 +667,42 @@ netvsc_hotadd_callback(const char *device_name, enum rte_dev_event_type type, + if (hv->vf_ctx.vf_state > vf_removed) + break; + ++ hot_ctx = rte_zmalloc("NETVSC-HOTADD", sizeof(*hot_ctx), ++ rte_mem_page_size()); ++ ++ if (!hot_ctx) { ++ PMD_DRV_LOG(ERR, "Failed to allocate hotadd context"); ++ return; ++ } ++ ++ hot_ctx->hv = hv; ++ d = &hot_ctx->da; ++ + ret = rte_devargs_parse(d, device_name); + if (ret) { + PMD_DRV_LOG(ERR, + "devargs parsing failed ret=%d", ret); +- return; ++ goto free_ctx; + } + + if (!strcmp(d->bus->name, "pci")) { + /* Start the process of figuring out if this + * PCI device is a VF device + */ +- hv->eal_hot_plug_retry = 0; +- rte_eal_alarm_set(1000000, netvsc_hotplug_retry, hv); ++ rte_spinlock_lock(&hv->hotadd_lock); ++ LIST_INSERT_HEAD(&hv->hotadd_list, hot_ctx, list); ++ rte_spinlock_unlock(&hv->hotadd_lock); ++ rte_eal_alarm_set(1000000, netvsc_hotplug_retry, hot_ctx); ++ return; + } + + /* We will switch to VF on RDNIS configure message + * sent from VSP + */ +- ++free_ctx: ++ rte_free(hot_ctx); + break; + - raft_waiters_run(raft); - - if (!raft->listener && time_msec() >= raft->listen_backoff) { -@@ -1957,16 +1976,29 @@ raft_run(struct raft *raft) - } + default: + break; + } +@@ -1003,12 +1037,20 @@ hn_dev_close(struct rte_eth_dev *dev) + { + int ret; + struct hn_data *hv = dev->data->dev_private; ++ struct hv_hotadd_context *hot_ctx; - /* Close unneeded sessions. */ -- struct raft_conn *next; -- LIST_FOR_EACH_SAFE (conn, next, list_node, &raft->conns) { -+ struct raft_server *server; -+ LIST_FOR_EACH_SAFE (conn, list_node, &raft->conns) { - if (!raft_conn_should_stay_open(raft, conn)) { -+ server = raft_find_new_server(raft, &conn->sid); -+ if (server) { -+ /* We only have one incoming connection from joining servers, -+ * so if it's closed, we need to destroy the record about the -+ * server. This way the process can be started over on the -+ * next join request. */ -+ static struct vlog_rate_limit rl = VLOG_RATE_LIMIT_INIT(5, 5); -+ VLOG_INFO_RL(&rl, "cluster "CID_FMT": server %s (%s) " -+ "disconnected while joining", -+ CID_ARGS(&raft->cid), -+ server->nickname, server->address); -+ hmap_remove(&raft->add_servers, &server->hmap_node); -+ raft_server_destroy(server); -+ } - raft->n_disconnections++; - raft_conn_close(conn); - } - } + PMD_INIT_FUNC_TRACE(); + if (rte_eal_process_type() != RTE_PROC_PRIMARY) + return 0; - /* Open needed sessions. */ -- struct raft_server *server; - HMAP_FOR_EACH (server, hmap_node, &raft->servers) { - raft_open_conn(raft, server->address, &server->sid); - } -@@ -2039,11 +2071,18 @@ raft_run(struct raft *raft) - * commands becomes new leader: the pending commands can still complete - * if the crashed leader has replicated the transactions to majority of - * followers before it crashed. */ -- struct raft_command *cmd, *next_cmd; -- HMAP_FOR_EACH_SAFE (cmd, next_cmd, hmap_node, &raft->commands) { -+ struct raft_command *cmd; -+ HMAP_FOR_EACH_SAFE (cmd, hmap_node, &raft->commands) { - if (cmd->timestamp - && now - cmd->timestamp > raft->election_timer * 2) { -- raft_command_complete(raft, cmd, RAFT_CMD_TIMEOUT); -+ if (cmd->index && raft->role != RAFT_LEADER) { -+ /* This server lost leadership and command didn't complete -+ * in time. Likely, it wasn't replicated to the majority -+ * of servers before losing the leadership. */ -+ raft_command_complete(raft, cmd, RAFT_CMD_LOST_LEADERSHIP); -+ } else { -+ raft_command_complete(raft, cmd, RAFT_CMD_TIMEOUT); -+ } - } - } - raft_reset_ping_timer(raft); -@@ -2235,6 +2274,9 @@ raft_command_initiate(struct raft *raft, - if (failure_test == FT_CRASH_AFTER_SEND_APPEND_REQ) { - ovs_fatal(0, "Raft test: crash after sending append_request."); - } -+ if (failure_test == FT_TRANSFER_LEADERSHIP_AFTER_SEND_APPEND_REQ) { -+ failure_test = FT_TRANSFER_LEADERSHIP; -+ } - raft_reset_ping_timer(raft); +- rte_eal_alarm_cancel(netvsc_hotplug_retry, &hv->devargs); ++ rte_spinlock_lock(&hv->hotadd_lock); ++ while (!LIST_EMPTY(&hv->hotadd_list)) { ++ hot_ctx = LIST_FIRST(&hv->hotadd_list); ++ rte_eal_alarm_cancel(netvsc_hotplug_retry, hot_ctx); ++ LIST_REMOVE(hot_ctx, list); ++ rte_free(hot_ctx); ++ } ++ rte_spinlock_unlock(&hv->hotadd_lock); + + ret = hn_vf_close(dev); + hn_dev_free_queues(dev); +@@ -1097,6 +1139,9 @@ eth_hn_dev_init(struct rte_eth_dev *eth_dev) + + PMD_INIT_FUNC_TRACE(); + ++ rte_spinlock_init(&hv->hotadd_lock); ++ LIST_INIT(&hv->hotadd_list); ++ + vmbus = container_of(device, struct rte_vmbus_device, device); + eth_dev->dev_ops = &hn_eth_dev_ops; + eth_dev->rx_queue_count = hn_dev_rx_queue_count; +@@ -1124,8 +1169,8 @@ eth_hn_dev_init(struct rte_eth_dev *eth_dev) + } - return cmd; -@@ -2243,8 +2285,12 @@ raft_command_initiate(struct raft *raft, - static void - log_all_commands(struct raft *raft) + hv->vmbus = vmbus; +- hv->rxbuf_res = &vmbus->resource[HV_RECV_BUF_MAP]; +- hv->chim_res = &vmbus->resource[HV_SEND_BUF_MAP]; ++ hv->rxbuf_res = vmbus->resource[HV_RECV_BUF_MAP]; ++ hv->chim_res = vmbus->resource[HV_SEND_BUF_MAP]; + hv->port_id = eth_dev->data->port_id; + hv->latency = HN_CHAN_LATENCY_NS; + hv->rx_copybreak = HN_RXCOPY_THRESHOLD; +@@ -1221,6 +1266,9 @@ eth_hn_dev_uninit(struct rte_eth_dev *eth_dev) + ret_stop = hn_dev_stop(eth_dev); + hn_dev_close(eth_dev); + ++ free(hv->vf_devargs); ++ hv->vf_devargs = NULL; ++ + hn_detach(hv); + hn_chim_uninit(eth_dev); + rte_vmbus_chan_close(hv->primary->chan); +diff --git a/dpdk/drivers/net/netvsc/hn_nvs.c b/dpdk/drivers/net/netvsc/hn_nvs.c +index 89dbba6cd9..b90280c9ff 100644 +--- a/dpdk/drivers/net/netvsc/hn_nvs.c ++++ b/dpdk/drivers/net/netvsc/hn_nvs.c +@@ -193,11 +193,11 @@ hn_nvs_conn_rxbuf(struct hn_data *hv) + * Connect RXBUF to NVS. + */ + conn.type = NVS_TYPE_RXBUF_CONN; +- conn.gpadl = hv->rxbuf_res->phys_addr; ++ conn.gpadl = hv->rxbuf_res.phys_addr; + conn.sig = NVS_RXBUF_SIG; + PMD_DRV_LOG(DEBUG, "connect rxbuff va=%p gpad=%#" PRIx64, +- hv->rxbuf_res->addr, +- hv->rxbuf_res->phys_addr); ++ hv->rxbuf_res.addr, ++ hv->rxbuf_res.phys_addr); + + error = hn_nvs_execute(hv, &conn, sizeof(conn), + &resp, sizeof(resp), +@@ -229,7 +229,7 @@ hn_nvs_conn_rxbuf(struct hn_data *hv) + hv->rxbuf_section_cnt = resp.nvs_sect[0].slotcnt; + + /* +- * Pimary queue's rxbuf_info is not allocated at creation time. ++ * Primary queue's rxbuf_info is not allocated at creation time. + * Now we can allocate it after we figure out the slotcnt. + */ + hv->primary->rxbuf_info = rte_calloc("HN_RXBUF_INFO", +@@ -308,17 +308,17 @@ hn_nvs_conn_chim(struct hn_data *hv) + struct hn_nvs_chim_conn chim; + struct hn_nvs_chim_connresp resp; + uint32_t sectsz; +- unsigned long len = hv->chim_res->len; ++ unsigned long len = hv->chim_res.len; + int error; + + /* Connect chimney sending buffer to NVS */ + memset(&chim, 0, sizeof(chim)); + chim.type = NVS_TYPE_CHIM_CONN; +- chim.gpadl = hv->chim_res->phys_addr; ++ chim.gpadl = hv->chim_res.phys_addr; + chim.sig = NVS_CHIM_SIG; + PMD_DRV_LOG(DEBUG, "connect send buf va=%p gpad=%#" PRIx64, +- hv->chim_res->addr, +- hv->chim_res->phys_addr); ++ hv->chim_res.addr, ++ hv->chim_res.phys_addr); + + error = hn_nvs_execute(hv, &chim, sizeof(chim), + &resp, sizeof(resp), +diff --git a/dpdk/drivers/net/netvsc/hn_rxtx.c b/dpdk/drivers/net/netvsc/hn_rxtx.c +index 028f176c7e..7a3bd523a5 100644 +--- a/dpdk/drivers/net/netvsc/hn_rxtx.c ++++ b/dpdk/drivers/net/netvsc/hn_rxtx.c +@@ -578,11 +578,11 @@ static void hn_rxpkt(struct hn_rx_queue *rxq, struct hn_rx_bufinfo *rxb, + rte_iova_t iova; + + /* +- * Build an external mbuf that points to recveive area. ++ * Build an external mbuf that points to receive area. + * Use refcount to handle multiple packets in same + * receive buffer section. + */ +- rxbuf = hv->rxbuf_res->addr; ++ rxbuf = hv->rxbuf_res.addr; + iova = rte_mem_virt2iova(rxbuf) + RTE_PTR_DIFF(data, rxbuf); + shinfo = &rxb->shinfo; + +@@ -765,8 +765,8 @@ hn_nvs_handle_rxbuf(struct rte_eth_dev *dev, { -- struct raft_command *cmd, *next; -- HMAP_FOR_EACH_SAFE (cmd, next, hmap_node, &raft->commands) { -+ if (!VLOG_IS_DBG_ENABLED()) { -+ return; -+ } + const struct vmbus_chanpkt_rxbuf *pkt; + const struct hn_nvs_hdr *nvs_hdr = buf; +- uint32_t rxbuf_sz = hv->rxbuf_res->len; +- char *rxbuf = hv->rxbuf_res->addr; ++ uint32_t rxbuf_sz = hv->rxbuf_res.len; ++ char *rxbuf = hv->rxbuf_res.addr; + unsigned int i, hlen, count; + struct hn_rx_bufinfo *rxb; + +@@ -1031,7 +1031,7 @@ hn_dev_rx_queue_count(void *rx_queue) + * returns: + * - -EINVAL - offset outside of ring + * - RTE_ETH_RX_DESC_AVAIL - no data available yet +- * - RTE_ETH_RX_DESC_DONE - data is waiting in stagin ring ++ * - RTE_ETH_RX_DESC_DONE - data is waiting in staging ring + */ + int hn_dev_rx_queue_status(void *arg, uint16_t offset) + { +@@ -1266,7 +1266,7 @@ hn_try_txagg(struct hn_data *hv, struct hn_tx_queue *txq, + if (txd->chim_index == NVS_CHIM_IDX_INVALID) + return NULL; + +- chim = (uint8_t *)hv->chim_res->addr ++ chim = (uint8_t *)hv->chim_res.addr + + txd->chim_index * hv->chim_szmax; + + txq->agg_txd = txd; +@@ -1348,8 +1348,11 @@ static void hn_encap(struct rndis_packet_msg *pkt, + *pi_data = NDIS_LSO2_INFO_MAKEIPV4(hlen, + m->tso_segsz); + } +- } else if (m->ol_flags & +- (RTE_MBUF_F_TX_TCP_CKSUM | RTE_MBUF_F_TX_UDP_CKSUM | RTE_MBUF_F_TX_IP_CKSUM)) { ++ } else if ((m->ol_flags & RTE_MBUF_F_TX_L4_MASK) == ++ RTE_MBUF_F_TX_TCP_CKSUM || ++ (m->ol_flags & RTE_MBUF_F_TX_L4_MASK) == ++ RTE_MBUF_F_TX_UDP_CKSUM || ++ (m->ol_flags & RTE_MBUF_F_TX_IP_CKSUM)) { + pi_data = hn_rndis_pktinfo_append(pkt, NDIS_TXCSUM_INFO_SIZE, + NDIS_PKTINFO_TYPE_CSUM); + *pi_data = 0; +@@ -1363,9 +1366,11 @@ static void hn_encap(struct rndis_packet_msg *pkt, + *pi_data |= NDIS_TXCSUM_INFO_IPCS; + } + +- if (m->ol_flags & RTE_MBUF_F_TX_TCP_CKSUM) ++ if ((m->ol_flags & RTE_MBUF_F_TX_L4_MASK) == ++ RTE_MBUF_F_TX_TCP_CKSUM) + *pi_data |= NDIS_TXCSUM_INFO_MKTCPCS(hlen); +- else if (m->ol_flags & RTE_MBUF_F_TX_UDP_CKSUM) ++ else if ((m->ol_flags & RTE_MBUF_F_TX_L4_MASK) == ++ RTE_MBUF_F_TX_UDP_CKSUM) + *pi_data |= NDIS_TXCSUM_INFO_MKUDPCS(hlen); + } + +diff --git a/dpdk/drivers/net/netvsc/hn_var.h b/dpdk/drivers/net/netvsc/hn_var.h +index fbb3995507..98a3b83033 100644 +--- a/dpdk/drivers/net/netvsc/hn_var.h ++++ b/dpdk/drivers/net/netvsc/hn_var.h +@@ -126,6 +126,13 @@ struct hn_vf_ctx { + enum vf_device_state vf_state; + }; + ++struct hv_hotadd_context { ++ LIST_ENTRY(hv_hotadd_context) list; ++ struct hn_data *hv; ++ struct rte_devargs da; ++ int eal_hot_plug_retry; ++}; + -+ struct raft_command *cmd; -+ HMAP_FOR_EACH (cmd, hmap_node, &raft->commands) { - VLOG_DBG("raft command eid: "UUID_FMT, UUID_ARGS(&cmd->eid)); - } - } -@@ -2398,8 +2444,8 @@ raft_command_complete(struct raft *raft, - static void - raft_complete_all_commands(struct raft *raft, enum raft_command_status status) + struct hn_data { + struct rte_vmbus_device *vmbus; + struct hn_rx_queue *primary; +@@ -140,7 +147,7 @@ struct hn_data { + uint32_t link_status; + uint32_t link_speed; + +- struct rte_mem_resource *rxbuf_res; /* UIO resource for Rx */ ++ struct rte_mem_resource rxbuf_res; /* UIO resource for Rx */ + uint32_t rxbuf_section_cnt; /* # of Rx sections */ + uint32_t rx_copybreak; + uint32_t rx_extmbuf_enable; +@@ -149,7 +156,7 @@ struct hn_data { + uint64_t rss_offloads; + + rte_spinlock_t chim_lock; +- struct rte_mem_resource *chim_res; /* UIO resource for Tx */ ++ struct rte_mem_resource chim_res; /* UIO resource for Tx */ + struct rte_bitmap *chim_bmap; /* Send buffer map */ + void *chim_bmem; + uint32_t tx_copybreak; +@@ -175,8 +182,9 @@ struct hn_data { + + struct vmbus_channel *channels[HN_MAX_CHANNELS]; + +- struct rte_devargs devargs; +- int eal_hot_plug_retry; ++ rte_spinlock_t hotadd_lock; ++ LIST_HEAD(hotadd_list, hv_hotadd_context) hotadd_list; ++ char *vf_devargs; + }; + + static inline struct vmbus_channel * +diff --git a/dpdk/drivers/net/netvsc/hn_vf.c b/dpdk/drivers/net/netvsc/hn_vf.c +index fead8eba5d..62948bf889 100644 +--- a/dpdk/drivers/net/netvsc/hn_vf.c ++++ b/dpdk/drivers/net/netvsc/hn_vf.c +@@ -103,7 +103,7 @@ static void hn_remove_delayed(void *args) + struct rte_device *dev = rte_eth_devices[port_id].device; + int ret; + +- /* Tell VSP to switch data path to synthentic */ ++ /* Tell VSP to switch data path to synthetic */ + hn_vf_remove(hv); + + PMD_DRV_LOG(NOTICE, "Start to remove port %d", port_id); +@@ -129,6 +129,10 @@ static void hn_remove_delayed(void *args) + PMD_DRV_LOG(ERR, "rte_eth_dev_stop failed port_id=%u ret=%d", + port_id, ret); + ++ /* Record the device parameters for possible hotplug events */ ++ if (dev->devargs && dev->devargs->args) ++ hv->vf_devargs = strdup(dev->devargs->args); ++ + ret = rte_eth_dev_close(port_id); + if (ret) + PMD_DRV_LOG(ERR, "rte_eth_dev_close failed port_id=%u ret=%d", +diff --git a/dpdk/drivers/net/nfb/nfb.h b/dpdk/drivers/net/nfb/nfb.h +index 59d3ab4986..96c44c3a45 100644 +--- a/dpdk/drivers/net/nfb/nfb.h ++++ b/dpdk/drivers/net/nfb/nfb.h +@@ -48,10 +48,6 @@ struct pmd_internals { + + char nfb_dev[PATH_MAX]; + struct nfb_device *nfb; +- /* Place to remember if filter was promiscuous or filtering by table, +- * when disabling allmulticast +- */ +- enum nc_rxmac_mac_filter rx_filter_original; + }; + + #endif /* _NFB_H_ */ +diff --git a/dpdk/drivers/net/nfb/nfb_ethdev.c b/dpdk/drivers/net/nfb/nfb_ethdev.c +index 3c39937816..d9e43bc027 100644 +--- a/dpdk/drivers/net/nfb/nfb_ethdev.c ++++ b/dpdk/drivers/net/nfb/nfb_ethdev.c +@@ -77,9 +77,10 @@ static void + nfb_nc_rxmac_deinit(struct nc_rxmac *rxmac[RTE_MAX_NC_RXMAC], + uint16_t max_rxmac) { -- struct raft_command *cmd, *next; -- HMAP_FOR_EACH_SAFE (cmd, next, hmap_node, &raft->commands) { -+ struct raft_command *cmd; -+ HMAP_FOR_EACH_SAFE (cmd, hmap_node, &raft->commands) { - raft_command_complete(raft, cmd, status); - } +- for (; max_rxmac > 0; --max_rxmac) { +- nc_rxmac_close(rxmac[max_rxmac]); +- rxmac[max_rxmac] = NULL; ++ uint16_t i; ++ for (i = 0; i < max_rxmac; i++) { ++ nc_rxmac_close(rxmac[i]); ++ rxmac[i] = NULL; + } } -@@ -2601,7 +2647,13 @@ raft_become_follower(struct raft *raft) - * configuration is already part of the log. Possibly the configuration - * log entry will not be committed, but until we know that we must use the - * new configuration. Our AppendEntries processing will properly update -- * the server configuration later, if necessary. */ -+ * the server configuration later, if necessary. -+ * -+ * Also we do not complete commands here, as they can still be completed -+ * if their log entries have already been replicated to other servers. -+ * If the entries were actually committed according to the new leader, our -+ * AppendEntries processing will complete the corresponding commands. -+ */ - struct raft_server *s; - HMAP_FOR_EACH (s, hmap_node, &raft->add_servers) { - raft_send_add_server_reply__(raft, &s->sid, s->address, false, -@@ -2615,8 +2667,6 @@ raft_become_follower(struct raft *raft) - raft_server_destroy(raft->remove_server); - raft->remove_server = NULL; - } -- -- raft_complete_all_commands(raft, RAFT_CMD_LOST_LEADERSHIP); + +@@ -95,9 +96,10 @@ static void + nfb_nc_txmac_deinit(struct nc_txmac *txmac[RTE_MAX_NC_TXMAC], + uint16_t max_txmac) + { +- for (; max_txmac > 0; --max_txmac) { +- nc_txmac_close(txmac[max_txmac]); +- txmac[max_txmac] = NULL; ++ uint16_t i; ++ for (i = 0; i < max_txmac; i++) { ++ nc_txmac_close(txmac[i]); ++ txmac[i] = NULL; + } } - static void -@@ -2865,61 +2915,56 @@ raft_update_commit_index(struct raft *raft, uint64_t new_commit_index) - return false; - } +@@ -514,7 +516,6 @@ nfb_eth_dev_init(struct rte_eth_dev *dev) -- if (raft->role == RAFT_LEADER) { -- while (raft->commit_index < new_commit_index) { -- uint64_t index = ++raft->commit_index; -- const struct raft_entry *e = raft_get_entry(raft, index); -- if (raft_entry_has_data(e)) { -- struct raft_command *cmd -- = raft_find_command_by_eid(raft, &e->eid); -- if (cmd) { -- if (!cmd->index) { -- VLOG_DBG("Command completed after role change from" -- " follower to leader "UUID_FMT, -- UUID_ARGS(&e->eid)); -- cmd->index = index; -- } -- raft_command_complete(raft, cmd, RAFT_CMD_SUCCESS); -+ while (raft->commit_index < new_commit_index) { -+ uint64_t index = ++raft->commit_index; -+ const struct raft_entry *e = raft_get_entry(raft, index); -+ -+ if (raft_entry_has_data(e)) { -+ struct raft_command *cmd = raft_find_command_by_eid(raft, &e->eid); -+ static struct vlog_rate_limit rl = VLOG_RATE_LIMIT_INIT(5, 5); -+ -+ if (cmd) { -+ if (!cmd->index && raft->role == RAFT_LEADER) { -+ VLOG_INFO_RL(&rl, -+ "command completed after role change from " -+ "follower to leader (eid: "UUID_FMT", " -+ "commit index: %"PRIu64")", UUID_ARGS(&e->eid), index); -+ } else if (!cmd->index && raft->role != RAFT_LEADER) { -+ /* This can happen when leader fail-over before sending -+ * execute_command_reply. */ -+ VLOG_INFO_RL(&rl, -+ "command completed without reply (eid: "UUID_FMT", " -+ "commit index: %"PRIu64")", UUID_ARGS(&e->eid), index); -+ } else if (cmd->index && raft->role != RAFT_LEADER) { -+ /* This can happen if current server lost leadership after -+ * sending append requests to the majority of servers, but -+ * before receiving majority of append replies. */ -+ VLOG_INFO_RL(&rl, -+ "command completed after role change from " -+ "leader to follower (eid: "UUID_FMT", " -+ "commit index: %"PRIu64")", UUID_ARGS(&e->eid), index); -+ /* Clearing 'sid' to avoid sending cmd execution reply. */ -+ cmd->sid = UUID_ZERO; -+ } else { -+ /* (cmd->index && raft->role == RAFT_LEADER) -+ * Normal command completion on a leader. */ - } -- } -- if (e->election_timer) { -- VLOG_INFO("Election timer changed from %"PRIu64" to %"PRIu64, -- raft->election_timer, e->election_timer); -- raft->election_timer = e->election_timer; -- raft->election_timer_new = 0; -- raft_update_probe_intervals(raft); -- } -- if (e->servers) { -- /* raft_run_reconfigure() can write a new Raft entry, which can -- * reallocate raft->entries, which would invalidate 'e', so -- * this case must be last, after the one for 'e->data'. */ -- raft_run_reconfigure(raft); -+ cmd->index = index; -+ raft_command_complete(raft, cmd, RAFT_CMD_SUCCESS); - } - } -- } else { -- while (raft->commit_index < new_commit_index) { -- uint64_t index = ++raft->commit_index; -- const struct raft_entry *e = raft_get_entry(raft, index); -- if (e->election_timer) { -- VLOG_INFO("Election timer changed from %"PRIu64" to %"PRIu64, -- raft->election_timer, e->election_timer); -- raft->election_timer = e->election_timer; -- raft_update_probe_intervals(raft); -- } -+ if (e->election_timer) { -+ VLOG_INFO("Election timer changed from %"PRIu64" to %"PRIu64, -+ raft->election_timer, e->election_timer); -+ raft->election_timer = e->election_timer; -+ raft->election_timer_new = 0; -+ raft_update_probe_intervals(raft); - } -- /* Check if any pending command can be completed, and complete it. -- * This can happen when leader fail-over before sending -- * execute_command_reply. */ -- const struct uuid *eid = raft_get_eid(raft, new_commit_index); -- struct raft_command *cmd = raft_find_command_by_eid(raft, eid); -- if (cmd) { -- static struct vlog_rate_limit rl = VLOG_RATE_LIMIT_INIT(5, 5); -- VLOG_INFO_RL(&rl, -- "Command completed without reply (eid: "UUID_FMT", " -- "commit index: %"PRIu64")", -- UUID_ARGS(eid), new_commit_index); -- cmd->index = new_commit_index; -- raft_command_complete(raft, cmd, RAFT_CMD_SUCCESS); -+ if (e->servers && raft->role == RAFT_LEADER) { -+ /* raft_run_reconfigure() can write a new Raft entry, which can -+ * reallocate raft->entries, which would invalidate 'e', so -+ * this case must be last, after the one for 'e->data'. */ -+ raft_run_reconfigure(raft); - } - } + data->promiscuous = nfb_eth_promiscuous_get(dev); + data->all_multicast = nfb_eth_allmulticast_get(dev); +- internals->rx_filter_original = data->promiscuous; -@@ -3354,12 +3399,6 @@ raft_find_peer(struct raft *raft, const struct uuid *uuid) - return s && !uuid_equals(&raft->sid, &s->sid) ? s : NULL; - } + dev->data->dev_flags |= RTE_ETH_DEV_AUTOFILL_QUEUE_XSTATS; --static struct raft_server * --raft_find_new_server(struct raft *raft, const struct uuid *uuid) --{ -- return raft_server_find(&raft->add_servers, uuid); --} +diff --git a/dpdk/drivers/net/nfb/nfb_rxmode.c b/dpdk/drivers/net/nfb/nfb_rxmode.c +index 2d0b613d21..ca6e4d5578 100644 +--- a/dpdk/drivers/net/nfb/nfb_rxmode.c ++++ b/dpdk/drivers/net/nfb/nfb_rxmode.c +@@ -14,8 +14,6 @@ nfb_eth_promiscuous_enable(struct rte_eth_dev *dev) + dev->data->dev_private; + uint16_t i; + +- internals->rx_filter_original = RXMAC_MAC_FILTER_PROMISCUOUS; - - /* Figure 3.1: "If there exists an N such that N > commitIndex, a - * majority of matchIndex[i] >= N, and log[N].term == currentTerm, set - * commitIndex = N (sections 3.5 and 3.6)." */ -@@ -4142,6 +4181,10 @@ static void - raft_handle_install_snapshot_request( - struct raft *raft, const struct raft_install_snapshot_request *rq) - { -+ if (failure_test == FT_CRASH_BEFORE_SEND_SNAPSHOT_REP) { -+ ovs_fatal(0, "Raft test: crash before sending install_snapshot_reply"); -+ } + for (i = 0; i < internals->max_rxmac; ++i) { + nc_rxmac_mac_filter_enable(internals->rxmac[i], + RXMAC_MAC_FILTER_PROMISCUOUS); +@@ -30,16 +28,13 @@ nfb_eth_promiscuous_disable(struct rte_eth_dev *dev) + struct pmd_internals *internals = (struct pmd_internals *) + dev->data->dev_private; + uint16_t i; ++ enum nc_rxmac_mac_filter filter = RXMAC_MAC_FILTER_TABLE_BCAST; + +- internals->rx_filter_original = RXMAC_MAC_FILTER_TABLE; +- +- /* if promisc is not enabled, do nothing */ +- if (!nfb_eth_promiscuous_get(dev)) +- return 0; ++ if (dev->data->all_multicast) ++ filter = RXMAC_MAC_FILTER_TABLE_BCAST_MCAST; + + for (i = 0; i < internals->max_rxmac; ++i) { +- nc_rxmac_mac_filter_enable(internals->rxmac[i], +- RXMAC_MAC_FILTER_TABLE); ++ nc_rxmac_mac_filter_enable(internals->rxmac[i], filter); + } + + return 0; +@@ -67,6 +62,8 @@ nfb_eth_allmulticast_enable(struct rte_eth_dev *dev) + dev->data->dev_private; + + uint16_t i; ++ if (dev->data->promiscuous) ++ return 0; + for (i = 0; i < internals->max_rxmac; ++i) { + nc_rxmac_mac_filter_enable(internals->rxmac[i], + RXMAC_MAC_FILTER_TABLE_BCAST_MCAST); +@@ -83,13 +80,12 @@ nfb_eth_allmulticast_disable(struct rte_eth_dev *dev) + + uint16_t i; + +- /* if multicast is not enabled do nothing */ +- if (!nfb_eth_allmulticast_get(dev)) ++ if (dev->data->promiscuous) + return 0; + + for (i = 0; i < internals->max_rxmac; ++i) { + nc_rxmac_mac_filter_enable(internals->rxmac[i], +- internals->rx_filter_original); ++ RXMAC_MAC_FILTER_TABLE_BCAST); + } + + return 0; +diff --git a/dpdk/drivers/net/nfp/nfp_common.c b/dpdk/drivers/net/nfp/nfp_common.c +index f8978e803a..34e3a03edd 100644 +--- a/dpdk/drivers/net/nfp/nfp_common.c ++++ b/dpdk/drivers/net/nfp/nfp_common.c +@@ -176,6 +176,13 @@ nfp_net_configure(struct rte_eth_dev *dev) + return -EINVAL; + } + ++ /* Checking MTU set */ ++ if (rxmode->mtu > hw->flbufsz) { ++ PMD_INIT_LOG(INFO, "MTU (%u) larger then current mbufsize (%u) not supported", ++ rxmode->mtu, hw->flbufsz); ++ return -ERANGE; ++ } + - if (raft_handle_install_snapshot_request__(raft, rq)) { - union raft_rpc rpy = { - .install_snapshot_reply = { -@@ -4940,6 +4983,8 @@ raft_unixctl_failure_test(struct unixctl_conn *conn OVS_UNUSED, - failure_test = FT_CRASH_AFTER_SEND_EXEC_REQ; - } else if (!strcmp(test, "crash-after-receiving-append-request-update")) { - failure_test = FT_CRASH_AFTER_RECV_APPEND_REQ_UPDATE; -+ } else if (!strcmp(test, "crash-before-sending-install-snapshot-reply")) { -+ failure_test = FT_CRASH_BEFORE_SEND_SNAPSHOT_REP; - } else if (!strcmp(test, "delay-election")) { - failure_test = FT_DELAY_ELECTION; - struct raft *raft; -@@ -4952,6 +4997,11 @@ raft_unixctl_failure_test(struct unixctl_conn *conn OVS_UNUSED, - failure_test = FT_DONT_SEND_VOTE_REQUEST; - } else if (!strcmp(test, "stop-raft-rpc")) { - failure_test = FT_STOP_RAFT_RPC; -+ } else if (!strcmp(test, -+ "transfer-leadership-after-sending-append-request")) { -+ failure_test = FT_TRANSFER_LEADERSHIP_AFTER_SEND_APPEND_REQ; -+ } else if (!strcmp(test, "transfer-leadership")) { -+ failure_test = FT_TRANSFER_LEADERSHIP; - } else if (!strcmp(test, "clear")) { - failure_test = FT_NO_TEST; - unixctl_command_reply(conn, "test dismissed"); -diff --git a/ovsdb/relay.c b/ovsdb/relay.c -index ef0e44d340..122ee8c52f 100644 ---- a/ovsdb/relay.c -+++ b/ovsdb/relay.c -@@ -269,9 +269,9 @@ ovsdb_relay_clear(struct ovsdb *db) + return 0; + } - SHASH_FOR_EACH (table_node, &db->tables) { - struct ovsdb_table *table = table_node->data; -- struct ovsdb_row *row, *next; -+ struct ovsdb_row *row; +@@ -280,10 +287,6 @@ nfp_set_mac_addr(struct rte_eth_dev *dev, struct rte_ether_addr *mac_addr) + return -EBUSY; + } -- HMAP_FOR_EACH_SAFE (row, next, hmap_node, &table->rows) { -+ HMAP_FOR_EACH_SAFE (row, hmap_node, &table->rows) { - ovsdb_txn_row_delete(txn, row); - } - } -diff --git a/ovsdb/replication.c b/ovsdb/replication.c -index d8b56d8131..477c69d701 100644 ---- a/ovsdb/replication.c -+++ b/ovsdb/replication.c -@@ -549,8 +549,8 @@ reset_database(struct ovsdb *db) - /* Delete all rows if the table is not excluded. */ - if (!excluded_tables_find(db->schema->name, table_node->name)) { - struct ovsdb_table *table = table_node->data; -- struct ovsdb_row *row, *next; -- HMAP_FOR_EACH_SAFE (row, next, hmap_node, &table->rows) { -+ struct ovsdb_row *row; -+ HMAP_FOR_EACH_SAFE (row, hmap_node, &table->rows) { - ovsdb_txn_row_delete(txn, row); - } - } -@@ -769,9 +769,9 @@ replication_dbs_destroy(void) - return; - } +- if ((hw->ctrl & NFP_NET_CFG_CTRL_ENABLE) && +- !(hw->cap & NFP_NET_CFG_CTRL_LIVE_ADDR)) +- return -EBUSY; +- + /* Writing new MAC to the specific port BAR address */ + nfp_net_write_mac(hw, (uint8_t *)mac_addr); + +@@ -696,7 +699,17 @@ nfp_net_infos_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info) + dev_info->max_rx_queues = (uint16_t)hw->max_rx_queues; + dev_info->max_tx_queues = (uint16_t)hw->max_tx_queues; + dev_info->min_rx_bufsize = RTE_ETHER_MIN_MTU; +- dev_info->max_rx_pktlen = hw->max_mtu; ++ /* ++ * The maximum rx packet length (max_rx_pktlen) is set to the ++ * maximum supported frame size that the NFP can handle. This ++ * includes layer 2 headers, CRC and other metadata that can ++ * optionally be used. ++ * The maximum layer 3 MTU (max_mtu) is read from hardware, ++ * which was set by the firmware loaded onto the card. ++ */ ++ dev_info->max_rx_pktlen = NFP_FRAME_SIZE_MAX; ++ dev_info->max_mtu = hw->max_mtu; ++ dev_info->min_mtu = RTE_ETHER_MIN_MTU; + /* Next should change when PF support is implemented */ + dev_info->max_mac_addrs = 1; + +@@ -956,6 +969,13 @@ nfp_net_dev_mtu_set(struct rte_eth_dev *dev, uint16_t mtu) + return -EBUSY; + } -- struct shash_node *node, *next; -+ struct shash_node *node; ++ /* MTU larger then current mbufsize not supported */ ++ if (mtu > hw->flbufsz) { ++ PMD_DRV_LOG(ERR, "MTU (%u) larger then current mbufsize (%u) not supported", ++ mtu, hw->flbufsz); ++ return -ERANGE; ++ } ++ + /* writing to configuration space */ + nn_cfg_writel(hw, NFP_NET_CFG_MTU, mtu); -- SHASH_FOR_EACH_SAFE (node, next, replication_dbs) { -+ SHASH_FOR_EACH_SAFE (node, replication_dbs) { - hmap_remove(&replication_dbs->map, &node->node); - struct replication_db *rdb = node->data; - if (rdb->active_db_schema) { -diff --git a/ovsdb/table.c b/ovsdb/table.c -index 455a3663fe..2184701ec1 100644 ---- a/ovsdb/table.c -+++ b/ovsdb/table.c -@@ -309,10 +309,10 @@ void - ovsdb_table_destroy(struct ovsdb_table *table) +@@ -969,22 +989,25 @@ nfp_net_vlan_offload_set(struct rte_eth_dev *dev, int mask) { - if (table) { -- struct ovsdb_row *row, *next; -+ struct ovsdb_row *row; - size_t i; + uint32_t new_ctrl, update; + struct nfp_net_hw *hw; ++ struct rte_eth_conf *dev_conf; + int ret; -- HMAP_FOR_EACH_SAFE (row, next, hmap_node, &table->rows) { -+ HMAP_FOR_EACH_SAFE (row, hmap_node, &table->rows) { - ovsdb_row_destroy(row); - } - hmap_destroy(&table->rows); -diff --git a/ovsdb/transaction-forward.c b/ovsdb/transaction-forward.c -index d15f2f1d6d..963e937957 100644 ---- a/ovsdb/transaction-forward.c -+++ b/ovsdb/transaction-forward.c -@@ -126,10 +126,10 @@ ovsdb_txn_forward_steal_reply(struct ovsdb_txn_forward *txn_fwd) - void - ovsdb_txn_forward_run(struct ovsdb *db, struct ovsdb_cs *cs) - { -- struct ovsdb_txn_forward *t, *next; -+ struct ovsdb_txn_forward *t; + hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private); +- new_ctrl = 0; +- +- /* Enable vlan strip if it is not configured yet */ +- if ((mask & RTE_ETH_VLAN_STRIP_OFFLOAD) && +- !(hw->ctrl & NFP_NET_CFG_CTRL_RXVLAN)) +- new_ctrl = hw->ctrl | NFP_NET_CFG_CTRL_RXVLAN; ++ dev_conf = &dev->data->dev_conf; ++ new_ctrl = hw->ctrl; + +- /* Disable vlan strip just if it is configured */ +- if (!(mask & RTE_ETH_VLAN_STRIP_OFFLOAD) && +- (hw->ctrl & NFP_NET_CFG_CTRL_RXVLAN)) +- new_ctrl = hw->ctrl & ~NFP_NET_CFG_CTRL_RXVLAN; ++ /* ++ * Vlan stripping setting ++ * Enable or disable VLAN stripping ++ */ ++ if (mask & RTE_ETH_VLAN_STRIP_MASK) { ++ if (dev_conf->rxmode.offloads & RTE_ETH_RX_OFFLOAD_VLAN_STRIP) ++ new_ctrl |= NFP_NET_CFG_CTRL_RXVLAN; ++ else ++ new_ctrl &= ~NFP_NET_CFG_CTRL_RXVLAN; ++ } + +- if (new_ctrl == 0) ++ if (new_ctrl == hw->ctrl) + return 0; - /* Send all transactions that needs to be forwarded. */ -- LIST_FOR_EACH_SAFE (t, next, new_node, &db->txn_forward_new) { -+ LIST_FOR_EACH_SAFE (t, new_node, &db->txn_forward_new) { - if (!ovsdb_cs_may_send_transaction(cs)) { - break; - } -@@ -167,9 +167,9 @@ ovsdb_txn_forward_cancel(struct ovsdb *db, struct ovsdb_txn_forward *txn_fwd) - void - ovsdb_txn_forward_cancel_all(struct ovsdb *db, bool sent_only) - { -- struct ovsdb_txn_forward *t, *next; -+ struct ovsdb_txn_forward *t; + update = NFP_NET_CFG_UPDATE_GEN; +diff --git a/dpdk/drivers/net/nfp/nfp_common.h b/dpdk/drivers/net/nfp/nfp_common.h +index 8b35fa119c..8db5ec23f8 100644 +--- a/dpdk/drivers/net/nfp/nfp_common.h ++++ b/dpdk/drivers/net/nfp/nfp_common.h +@@ -98,6 +98,9 @@ struct nfp_net_adapter; + /* Number of supported physical ports */ + #define NFP_MAX_PHYPORTS 12 + ++/* Maximum supported NFP frame size (MTU + layer 2 headers) */ ++#define NFP_FRAME_SIZE_MAX 10048 ++ + #include + #include + +diff --git a/dpdk/drivers/net/nfp/nfp_ethdev.c b/dpdk/drivers/net/nfp/nfp_ethdev.c +index 8e81cc498f..1a9f7581a7 100644 +--- a/dpdk/drivers/net/nfp/nfp_ethdev.c ++++ b/dpdk/drivers/net/nfp/nfp_ethdev.c +@@ -302,11 +302,13 @@ nfp_net_close(struct rte_eth_dev *dev) + for (i = 0; i < dev->data->nb_tx_queues; i++) { + this_tx_q = (struct nfp_net_txq *)dev->data->tx_queues[i]; + nfp_net_reset_tx_queue(this_tx_q); ++ nfp_net_tx_queue_release(dev, i); + } -- HMAP_FOR_EACH_SAFE (t, next, sent_node, &db->txn_forward_sent) { -+ HMAP_FOR_EACH_SAFE (t, sent_node, &db->txn_forward_sent) { - ovsdb_txn_forward_cancel(db, t); - } + for (i = 0; i < dev->data->nb_rx_queues; i++) { + this_rx_q = (struct nfp_net_rxq *)dev->data->rx_queues[i]; + nfp_net_reset_rx_queue(this_rx_q); ++ nfp_net_rx_queue_release(dev, i); + } -@@ -177,7 +177,7 @@ ovsdb_txn_forward_cancel_all(struct ovsdb *db, bool sent_only) - return; - } + /* Cancel possible impending LSC work here before releasing the port*/ +@@ -500,6 +502,7 @@ nfp_net_init(struct rte_eth_dev *eth_dev) + hw->cap = nn_cfg_readl(hw, NFP_NET_CFG_CAP); + hw->max_mtu = nn_cfg_readl(hw, NFP_NET_CFG_MAX_MTU); + hw->mtu = RTE_ETHER_MTU; ++ hw->flbufsz = RTE_ETHER_MTU; + + /* VLAN insertion is incompatible with LSOv2 */ + if (hw->cap & NFP_NET_CFG_CTRL_LSO2) +diff --git a/dpdk/drivers/net/nfp/nfp_ethdev_vf.c b/dpdk/drivers/net/nfp/nfp_ethdev_vf.c +index 303ef72b1b..0781f34764 100644 +--- a/dpdk/drivers/net/nfp/nfp_ethdev_vf.c ++++ b/dpdk/drivers/net/nfp/nfp_ethdev_vf.c +@@ -219,11 +219,13 @@ nfp_netvf_close(struct rte_eth_dev *dev) + for (i = 0; i < dev->data->nb_tx_queues; i++) { + this_tx_q = (struct nfp_net_txq *)dev->data->tx_queues[i]; + nfp_net_reset_tx_queue(this_tx_q); ++ nfp_net_tx_queue_release(dev, i); + } -- LIST_FOR_EACH_SAFE (t, next, new_node, &db->txn_forward_new) { -+ LIST_FOR_EACH_SAFE (t, new_node, &db->txn_forward_new) { - ovsdb_txn_forward_cancel(db, t); - } - } -diff --git a/ovsdb/transaction.c b/ovsdb/transaction.c -index db86d847c3..3a6ddfa1df 100644 ---- a/ovsdb/transaction.c -+++ b/ovsdb/transaction.c -@@ -159,15 +159,15 @@ ovsdb_txn_row_abort(struct ovsdb_txn *txn OVS_UNUSED, - hmap_replace(&new->table->rows, &new->hmap_node, &old->hmap_node); - } + for (i = 0; i < dev->data->nb_rx_queues; i++) { + this_rx_q = (struct nfp_net_rxq *)dev->data->rx_queues[i]; + nfp_net_reset_rx_queue(this_rx_q); ++ nfp_net_rx_queue_release(dev, i); + } -- struct ovsdb_weak_ref *weak, *next; -- LIST_FOR_EACH_SAFE (weak, next, src_node, &txn_row->deleted_refs) { -+ struct ovsdb_weak_ref *weak; -+ LIST_FOR_EACH_SAFE (weak, src_node, &txn_row->deleted_refs) { - ovs_list_remove(&weak->src_node); - ovs_list_init(&weak->src_node); - if (hmap_node_is_null(&weak->dst_node)) { - ovsdb_weak_ref_destroy(weak); - } - } -- LIST_FOR_EACH_SAFE (weak, next, src_node, &txn_row->added_refs) { -+ LIST_FOR_EACH_SAFE (weak, src_node, &txn_row->added_refs) { - ovs_list_remove(&weak->src_node); - ovs_list_init(&weak->src_node); - if (hmap_node_is_null(&weak->dst_node)) { -@@ -508,11 +508,11 @@ static struct ovsdb_error * - ovsdb_txn_update_weak_refs(struct ovsdb_txn *txn OVS_UNUSED, - struct ovsdb_txn_row *txn_row) + rte_intr_disable(pci_dev->intr_handle); +@@ -367,6 +369,7 @@ nfp_netvf_init(struct rte_eth_dev *eth_dev) + hw->cap = nn_cfg_readl(hw, NFP_NET_CFG_CAP); + hw->max_mtu = nn_cfg_readl(hw, NFP_NET_CFG_MAX_MTU); + hw->mtu = RTE_ETHER_MTU; ++ hw->flbufsz = RTE_ETHER_MTU; + + /* VLAN insertion is incompatible with LSOv2 */ + if (hw->cap & NFP_NET_CFG_CTRL_LSO2) +diff --git a/dpdk/drivers/net/nfp/nfp_rxtx.c b/dpdk/drivers/net/nfp/nfp_rxtx.c +index 0fe1415596..335a90b2c9 100644 +--- a/dpdk/drivers/net/nfp/nfp_rxtx.c ++++ b/dpdk/drivers/net/nfp/nfp_rxtx.c +@@ -470,6 +470,7 @@ nfp_net_rx_queue_release(struct rte_eth_dev *dev, uint16_t queue_idx) + + if (rxq) { + nfp_net_rx_queue_release_mbufs(rxq); ++ rte_eth_dma_zone_free(dev, "rx_ring", queue_idx); + rte_free(rxq->rxbufs); + rte_free(rxq); + } +@@ -660,6 +661,7 @@ nfp_net_tx_queue_release(struct rte_eth_dev *dev, uint16_t queue_idx) + + if (txq) { + nfp_net_tx_queue_release_mbufs(txq); ++ rte_eth_dma_zone_free(dev, "tx_ring", queue_idx); + rte_free(txq->txbufs); + rte_free(txq); + } +diff --git a/dpdk/drivers/net/nfp/nfpcore/nfp-common/nfp_resid.h b/dpdk/drivers/net/nfp/nfpcore/nfp-common/nfp_resid.h +index 0e03948ec7..394a7628e0 100644 +--- a/dpdk/drivers/net/nfp/nfpcore/nfp-common/nfp_resid.h ++++ b/dpdk/drivers/net/nfp/nfpcore/nfp-common/nfp_resid.h +@@ -63,7 +63,7 @@ + * Wildcard indicating a CPP read or write action + * + * The action used will be either read or write depending on whether a read or +- * write instruction/call is performed on the NFP_CPP_ID. It is recomended that ++ * write instruction/call is performed on the NFP_CPP_ID. It is recommended that + * the RW action is used even if all actions to be performed on a NFP_CPP_ID are + * known to be only reads or writes. Doing so will in many cases save NFP CPP + * internal software resources. +@@ -405,7 +405,7 @@ int nfp_idstr2meid(int chip_family, const char *s, const char **endptr); + * @param chip_family Chip family ID + * @param s A string of format "iX.anything" or "iX" + * @param endptr If non-NULL, *endptr will point to the trailing +- * striong after the ME ID part of the string, which ++ * string after the ME ID part of the string, which + * is either an empty string or the first character + * after the separating period. + * @return The island ID on succes, -1 on error. +@@ -425,7 +425,7 @@ int nfp_idstr2island(int chip_family, const char *s, const char **endptr); + * @param chip_family Chip family ID + * @param s A string of format "meX.anything" or "meX" + * @param endptr If non-NULL, *endptr will point to the trailing +- * striong after the ME ID part of the string, which ++ * string after the ME ID part of the string, which + * is either an empty string or the first character + * after the separating period. + * @return The ME number on succes, -1 on error. +diff --git a/dpdk/drivers/net/nfp/nfpcore/nfp_cpp_pcie_ops.c b/dpdk/drivers/net/nfp/nfpcore/nfp_cpp_pcie_ops.c +index bad80a5a1c..08bc4e8ef2 100644 +--- a/dpdk/drivers/net/nfp/nfpcore/nfp_cpp_pcie_ops.c ++++ b/dpdk/drivers/net/nfp/nfpcore/nfp_cpp_pcie_ops.c +@@ -16,9 +16,6 @@ + + #include + #include +-#if defined(RTE_BACKTRACE) +-#include +-#endif + #include + #include + #include +diff --git a/dpdk/drivers/net/nfp/nfpcore/nfp_cppcore.c b/dpdk/drivers/net/nfp/nfpcore/nfp_cppcore.c +index f91049383e..37799af558 100644 +--- a/dpdk/drivers/net/nfp/nfpcore/nfp_cppcore.c ++++ b/dpdk/drivers/net/nfp/nfpcore/nfp_cppcore.c +@@ -202,7 +202,7 @@ nfp_cpp_area_alloc(struct nfp_cpp *cpp, uint32_t dest, + * @address: start address on CPP target + * @size: size of area + * +- * Allocate and initilizae a CPP area structure, and lock it down so ++ * Allocate and initialize a CPP area structure, and lock it down so + * that it can be accessed directly. + * + * NOTE: @address and @size must be 32-bit aligned values. +diff --git a/dpdk/drivers/net/nfp/nfpcore/nfp_nsp.h b/dpdk/drivers/net/nfp/nfpcore/nfp_nsp.h +index c9c7b0d0fb..e74cdeb191 100644 +--- a/dpdk/drivers/net/nfp/nfpcore/nfp_nsp.h ++++ b/dpdk/drivers/net/nfp/nfpcore/nfp_nsp.h +@@ -272,7 +272,7 @@ int __nfp_eth_set_split(struct nfp_nsp *nsp, unsigned int lanes); + * @br_primary: branch id of primary bootloader + * @br_secondary: branch id of secondary bootloader + * @br_nsp: branch id of NSP +- * @primary: version of primarary bootloader ++ * @primary: version of primary bootloader + * @secondary: version id of secondary bootloader + * @nsp: version id of NSP + * @sensor_mask: mask of present sensors available on NIC +diff --git a/dpdk/drivers/net/nfp/nfpcore/nfp_resource.c b/dpdk/drivers/net/nfp/nfpcore/nfp_resource.c +index dd41fa4de4..7b5630fd86 100644 +--- a/dpdk/drivers/net/nfp/nfpcore/nfp_resource.c ++++ b/dpdk/drivers/net/nfp/nfpcore/nfp_resource.c +@@ -207,7 +207,7 @@ nfp_resource_acquire(struct nfp_cpp *cpp, const char *name) + * nfp_resource_release() - Release a NFP Resource handle + * @res: NFP Resource handle + * +- * NOTE: This function implictly unlocks the resource handle ++ * NOTE: This function implicitly unlocks the resource handle + */ + void + nfp_resource_release(struct nfp_resource *res) +diff --git a/dpdk/drivers/net/nfp/nfpcore/nfp_rtsym.c b/dpdk/drivers/net/nfp/nfpcore/nfp_rtsym.c +index cb7d83db51..2feca2ed81 100644 +--- a/dpdk/drivers/net/nfp/nfpcore/nfp_rtsym.c ++++ b/dpdk/drivers/net/nfp/nfpcore/nfp_rtsym.c +@@ -236,7 +236,7 @@ nfp_rtsym_lookup(struct nfp_rtsym_table *rtbl, const char *name) + * nfp_rtsym_read_le() - Read a simple unsigned scalar value from symbol + * @rtbl: NFP RTsym table + * @name: Symbol name +- * @error: Poniter to error code (optional) ++ * @error: Pointer to error code (optional) + * + * Lookup a symbol, map, read it and return it's value. Value of the symbol + * will be interpreted as a simple little-endian unsigned value. Symbol can +diff --git a/dpdk/drivers/net/nfp/nfpcore/nfp_target.h b/dpdk/drivers/net/nfp/nfpcore/nfp_target.h +index 2884a0034f..e8dcc9ad1e 100644 +--- a/dpdk/drivers/net/nfp/nfpcore/nfp_target.h ++++ b/dpdk/drivers/net/nfp/nfpcore/nfp_target.h +@@ -37,7 +37,7 @@ pushpull_width(int pp) + static inline int + target_rw(uint32_t cpp_id, int pp, int start, int len) { -- struct ovsdb_weak_ref *weak, *next, *dst_weak; -+ struct ovsdb_weak_ref *weak, *dst_weak; - struct ovsdb_row *dst_row; +- int island = NFP_CPP_ID_ISLAND_of(cpp_id); ++ uint8_t island = NFP_CPP_ID_ISLAND_of(cpp_id); + + if (island && (island < start || island > (start + len))) + return NFP_ERRNO(EINVAL); +@@ -117,7 +117,7 @@ nfp6000_nbi_ppc(uint32_t cpp_id) + static inline int + nfp6000_nbi(uint32_t cpp_id, uint64_t address) + { +- int island = NFP_CPP_ID_ISLAND_of(cpp_id); ++ uint8_t island = NFP_CPP_ID_ISLAND_of(cpp_id); + uint64_t rel_addr = address & 0x3fFFFF; - /* Find and clean up deleted references from destination rows. */ -- LIST_FOR_EACH_SAFE (weak, next, src_node, &txn_row->deleted_refs) { -+ LIST_FOR_EACH_SAFE (weak, src_node, &txn_row->deleted_refs) { - dst_row = CONST_CAST(struct ovsdb_row *, - ovsdb_table_get_row(weak->dst_table, &weak->dst)); - if (dst_row) { -@@ -529,7 +529,7 @@ ovsdb_txn_update_weak_refs(struct ovsdb_txn *txn OVS_UNUSED, - } + if (island && (island < 8 || island > 9)) +@@ -281,7 +281,7 @@ static inline int + nfp6000_mu(uint32_t cpp_id, uint64_t address) + { + int pp; +- int island = NFP_CPP_ID_ISLAND_of(cpp_id); ++ uint8_t island = NFP_CPP_ID_ISLAND_of(cpp_id); + + if (island == 0) { + if (address < 0x2000000000ULL) +@@ -316,7 +316,7 @@ nfp6000_mu(uint32_t cpp_id, uint64_t address) + static inline int + nfp6000_ila(uint32_t cpp_id) + { +- int island = NFP_CPP_ID_ISLAND_of(cpp_id); ++ uint8_t island = NFP_CPP_ID_ISLAND_of(cpp_id); + + if (island && (island < 48 || island > 51)) + return NFP_ERRNO(EINVAL); +@@ -336,7 +336,7 @@ nfp6000_ila(uint32_t cpp_id) + static inline int + nfp6000_pci(uint32_t cpp_id) + { +- int island = NFP_CPP_ID_ISLAND_of(cpp_id); ++ uint8_t island = NFP_CPP_ID_ISLAND_of(cpp_id); + + if (island && (island < 4 || island > 7)) + return NFP_ERRNO(EINVAL); +@@ -354,7 +354,7 @@ nfp6000_pci(uint32_t cpp_id) + static inline int + nfp6000_crypto(uint32_t cpp_id) + { +- int island = NFP_CPP_ID_ISLAND_of(cpp_id); ++ uint8_t island = NFP_CPP_ID_ISLAND_of(cpp_id); + + if (island && (island < 12 || island > 15)) + return NFP_ERRNO(EINVAL); +@@ -370,9 +370,9 @@ nfp6000_crypto(uint32_t cpp_id) + static inline int + nfp6000_cap_xpb(uint32_t cpp_id) + { +- int island = NFP_CPP_ID_ISLAND_of(cpp_id); ++ uint8_t island = NFP_CPP_ID_ISLAND_of(cpp_id); - /* Insert the weak references added in the new version of the row. */ -- LIST_FOR_EACH_SAFE (weak, next, src_node, &txn_row->added_refs) { -+ LIST_FOR_EACH_SAFE (weak, src_node, &txn_row->added_refs) { - dst_row = CONST_CAST(struct ovsdb_row *, - ovsdb_table_get_row(weak->dst_table, &weak->dst)); +- if (island && (island < 1 || island > 63)) ++ if (island > 63) + return NFP_ERRNO(EINVAL); -@@ -597,7 +597,7 @@ find_and_add_weak_ref(struct ovsdb_txn_row *txn_row, - static struct ovsdb_error * OVS_WARN_UNUSED_RESULT - assess_weak_refs(struct ovsdb_txn *txn, struct ovsdb_txn_row *txn_row) + switch (cpp_id & NFP_CPP_ID(0, ~0, ~0)) { +@@ -410,9 +410,9 @@ nfp6000_cap_xpb(uint32_t cpp_id) + static inline int + nfp6000_cls(uint32_t cpp_id) { -- struct ovsdb_weak_ref *weak, *next; -+ struct ovsdb_weak_ref *weak; - struct ovsdb_table *table; - struct shash_node *node; +- int island = NFP_CPP_ID_ISLAND_of(cpp_id); ++ uint8_t island = NFP_CPP_ID_ISLAND_of(cpp_id); -@@ -642,7 +642,7 @@ assess_weak_refs(struct ovsdb_txn *txn, struct ovsdb_txn_row *txn_row) +- if (island && (island < 1 || island > 63)) ++ if (island > 63) + return NFP_ERRNO(EINVAL); - /* Collecting all key-value pairs that references deleted rows. */ - ovsdb_datum_init_empty(&deleted_refs); -- LIST_FOR_EACH_SAFE (weak, next, src_node, &txn_row->deleted_refs) { -+ LIST_FOR_EACH_SAFE (weak, src_node, &txn_row->deleted_refs) { - if (column->index == weak->column_idx) { - ovsdb_datum_add_unsafe(&deleted_refs, &weak->key, &weak->value, - &column->type, NULL); -@@ -1094,10 +1094,10 @@ static void - ovsdb_txn_destroy_cloned(struct ovsdb_txn *txn) + switch (cpp_id & NFP_CPP_ID(0, ~0, ~0)) { +@@ -540,11 +540,11 @@ nfp_target_cpp(uint32_t cpp_island_id, uint64_t cpp_island_address, + const uint32_t *imb_table) { - ovs_assert(!txn->db); -- struct ovsdb_txn_table *t, *next_txn_table; -- LIST_FOR_EACH_SAFE (t, next_txn_table, node, &txn->txn_tables) { -- struct ovsdb_txn_row *r, *next_txn_row; -- HMAP_FOR_EACH_SAFE (r, next_txn_row, hmap_node, &t->txn_rows) { -+ struct ovsdb_txn_table *t; -+ LIST_FOR_EACH_SAFE (t, node, &txn->txn_tables) { -+ struct ovsdb_txn_row *r; -+ HMAP_FOR_EACH_SAFE (r, hmap_node, &t->txn_rows) { - if (r->old) { - ovsdb_row_destroy(r->old); - } -@@ -1549,19 +1549,19 @@ for_each_txn_row(struct ovsdb_txn *txn, - serial++; + int err; +- int island = NFP_CPP_ID_ISLAND_of(cpp_island_id); +- int target = NFP_CPP_ID_TARGET_of(cpp_island_id); ++ uint8_t island = NFP_CPP_ID_ISLAND_of(cpp_island_id); ++ uint8_t target = NFP_CPP_ID_TARGET_of(cpp_island_id); + uint32_t imb; + +- if (target < 0 || target >= 16) ++ if (target >= 16) + return NFP_ERRNO(EINVAL); + + if (island == 0) { +diff --git a/dpdk/drivers/net/ngbe/base/ngbe_dummy.h b/dpdk/drivers/net/ngbe/base/ngbe_dummy.h +index 61b0d82bfb..d74c9f7b54 100644 +--- a/dpdk/drivers/net/ngbe/base/ngbe_dummy.h ++++ b/dpdk/drivers/net/ngbe/base/ngbe_dummy.h +@@ -114,6 +114,9 @@ static inline s32 ngbe_mac_get_link_capabilities_dummy(struct ngbe_hw *TUP0, + { + return NGBE_ERR_OPS_DUMMY; + } ++static inline void ngbe_setup_pba_dummy(struct ngbe_hw *TUP0) ++{ ++} + static inline s32 ngbe_mac_led_on_dummy(struct ngbe_hw *TUP0, u32 TUP1) + { + return NGBE_ERR_OPS_DUMMY; +@@ -298,6 +301,7 @@ static inline void ngbe_init_ops_dummy(struct ngbe_hw *hw) + hw->mac.setup_link = ngbe_mac_setup_link_dummy; + hw->mac.check_link = ngbe_mac_check_link_dummy; + hw->mac.get_link_capabilities = ngbe_mac_get_link_capabilities_dummy; ++ hw->mac.setup_pba = ngbe_setup_pba_dummy; + hw->mac.led_on = ngbe_mac_led_on_dummy; + hw->mac.led_off = ngbe_mac_led_off_dummy; + hw->mac.set_rar = ngbe_mac_set_rar_dummy; +diff --git a/dpdk/drivers/net/ngbe/base/ngbe_eeprom.c b/dpdk/drivers/net/ngbe/base/ngbe_eeprom.c +index f9a876e9bd..6375ee9b29 100644 +--- a/dpdk/drivers/net/ngbe/base/ngbe_eeprom.c ++++ b/dpdk/drivers/net/ngbe/base/ngbe_eeprom.c +@@ -20,8 +20,6 @@ s32 ngbe_init_eeprom_params(struct ngbe_hw *hw) + u32 eec; + u16 eeprom_size; + +- DEBUGFUNC("ngbe_init_eeprom_params"); +- + if (eeprom->type != ngbe_eeprom_unknown) + return 0; - do { -- struct ovsdb_txn_table *t, *next_txn_table; -+ struct ovsdb_txn_table *t; +@@ -52,8 +50,8 @@ s32 ngbe_init_eeprom_params(struct ngbe_hw *hw) + eeprom->address_bits = 16; + eeprom->sw_addr = 0x80; - any_work = false; -- LIST_FOR_EACH_SAFE (t, next_txn_table, node, &txn->txn_tables) { -+ LIST_FOR_EACH_SAFE (t, node, &txn->txn_tables) { - if (t->serial != serial) { - t->serial = serial; - t->n_processed = 0; - } +- DEBUGOUT("eeprom params: type = %d, size = %d, address bits: " +- "%d %d\n", eeprom->type, eeprom->word_size, ++ DEBUGOUT("eeprom params: type = %d, size = %d, address bits: %d %d", ++ eeprom->type, eeprom->word_size, + eeprom->address_bits, eeprom->sw_addr); - while (t->n_processed < hmap_count(&t->txn_rows)) { -- struct ovsdb_txn_row *r, *next_txn_row; -+ struct ovsdb_txn_row *r; + return 0; +@@ -72,9 +70,6 @@ s32 ngbe_get_eeprom_semaphore(struct ngbe_hw *hw) + u32 i; + u32 swsm; -- HMAP_FOR_EACH_SAFE (r, next_txn_row, hmap_node, &t->txn_rows) { -+ HMAP_FOR_EACH_SAFE (r, hmap_node, &t->txn_rows) { - if (r->serial != serial) { - struct ovsdb_error *error; +- DEBUGFUNC("ngbe_get_eeprom_semaphore"); +- +- + /* Get SMBI software semaphore between device drivers first */ + for (i = 0; i < timeout; i++) { + /* +@@ -90,8 +85,7 @@ s32 ngbe_get_eeprom_semaphore(struct ngbe_hw *hw) + } -@@ -1629,8 +1629,8 @@ ovsdb_txn_history_destroy(struct ovsdb *db) - return; - } + if (i == timeout) { +- DEBUGOUT("Driver can't access the eeprom - SMBI Semaphore " +- "not granted.\n"); ++ DEBUGOUT("Driver can't access the eeprom - SMBI Semaphore not granted."); + /* + * this release is particularly important because our attempts + * above to get the semaphore may have succeeded, and if there +@@ -134,13 +128,12 @@ s32 ngbe_get_eeprom_semaphore(struct ngbe_hw *hw) + * was not granted because we don't have access to the EEPROM + */ + if (i >= timeout) { +- DEBUGOUT("SWESMBI Software EEPROM semaphore not granted.\n"); ++ DEBUGOUT("SWESMBI Software EEPROM semaphore not granted."); + ngbe_release_eeprom_semaphore(hw); + status = NGBE_ERR_EEPROM; + } + } else { +- DEBUGOUT("Software semaphore SMBI between device drivers " +- "not granted.\n"); ++ DEBUGOUT("Software semaphore SMBI between device drivers not granted."); + } -- struct ovsdb_txn_history_node *txn_h_node, *next; -- LIST_FOR_EACH_SAFE (txn_h_node, next, node, &db->txn_history) { -+ struct ovsdb_txn_history_node *txn_h_node; -+ LIST_FOR_EACH_SAFE (txn_h_node, node, &db->txn_history) { - ovs_list_remove(&txn_h_node->node); - ovsdb_txn_destroy_cloned(txn_h_node->txn); - free(txn_h_node); -diff --git a/ovsdb/trigger.c b/ovsdb/trigger.c -index 726c138bf0..7d3003bca3 100644 ---- a/ovsdb/trigger.c -+++ b/ovsdb/trigger.c -@@ -146,14 +146,14 @@ ovsdb_trigger_prereplace_db(struct ovsdb_trigger *trigger) - bool - ovsdb_trigger_run(struct ovsdb *db, long long int now) + return status; +@@ -154,8 +147,6 @@ s32 ngbe_get_eeprom_semaphore(struct ngbe_hw *hw) + **/ + void ngbe_release_eeprom_semaphore(struct ngbe_hw *hw) { -- struct ovsdb_trigger *t, *next; -+ struct ovsdb_trigger *t; +- DEBUGFUNC("ngbe_release_eeprom_semaphore"); +- + wr32m(hw, NGBE_MNGSWSYNC, NGBE_MNGSWSYNC_REQ, 0); + wr32m(hw, NGBE_SWSEM, NGBE_SWSEM_PF, 0); + ngbe_flush(hw); +@@ -276,7 +267,6 @@ s32 ngbe_validate_eeprom_checksum_em(struct ngbe_hw *hw, + u32 eeprom_cksum_devcap = 0; + int err = 0; + +- DEBUGFUNC("ngbe_validate_eeprom_checksum_em"); + UNREFERENCED_PARAMETER(checksum_val); + + /* Check EEPROM only once */ +@@ -315,8 +305,6 @@ s32 ngbe_save_eeprom_version(struct ngbe_hw *hw) + u32 etrack_id = 0; + u32 offset = (hw->rom.sw_addr + NGBE_EEPROM_VERSION_L) << 1; + +- DEBUGFUNC("ngbe_save_eeprom_version"); +- + if (hw->bus.lan_id == 0) { + hw->rom.read32(hw, offset, &eeprom_verl); + etrack_id = eeprom_verl; +diff --git a/dpdk/drivers/net/ngbe/base/ngbe_hw.c b/dpdk/drivers/net/ngbe/base/ngbe_hw.c +index 0716357725..08a7e02943 100644 +--- a/dpdk/drivers/net/ngbe/base/ngbe_hw.c ++++ b/dpdk/drivers/net/ngbe/base/ngbe_hw.c +@@ -20,8 +20,6 @@ s32 ngbe_start_hw(struct ngbe_hw *hw) + { + s32 err; - bool run_triggers = db->run_triggers; - db->run_triggers_now = db->run_triggers = false; +- DEBUGFUNC("ngbe_start_hw"); +- + /* Clear the VLAN filter table */ + hw->mac.clear_vfta(hw); + +@@ -31,7 +29,7 @@ s32 ngbe_start_hw(struct ngbe_hw *hw) + /* Setup flow control */ + err = hw->mac.setup_fc(hw); + if (err != 0 && err != NGBE_NOT_IMPLEMENTED) { +- DEBUGOUT("Flow control setup failed, returning %d\n", err); ++ DEBUGOUT("Flow control setup failed, returning %d", err); + return err; + } - bool disconnect_all = false; +@@ -55,8 +53,6 @@ s32 ngbe_init_hw(struct ngbe_hw *hw) + { + s32 status; -- LIST_FOR_EACH_SAFE (t, next, node, &db->triggers) { -+ LIST_FOR_EACH_SAFE (t, node, &db->triggers) { - if (run_triggers - || now - t->created >= t->timeout_msec - || t->progress || t->txn_forward) { -diff --git a/python/ovs/db/idl.py b/python/ovs/db/idl.py -index 4ecdcaa197..b87099ff52 100644 ---- a/python/ovs/db/idl.py -+++ b/python/ovs/db/idl.py -@@ -140,6 +140,47 @@ class ConditionState(object): - return False +- DEBUGFUNC("ngbe_init_hw"); +- + ngbe_save_eeprom_version(hw); + /* Reset the hardware */ +@@ -67,7 +63,7 @@ s32 ngbe_init_hw(struct ngbe_hw *hw) + } -+class IdlTable(object): -+ def __init__(self, idl, table): -+ assert(isinstance(table, ovs.db.schema.TableSchema)) -+ self._table = table -+ self.need_table = False -+ self.rows = custom_index.IndexedRows(self) -+ self.idl = idl -+ self._condition_state = ConditionState() -+ self.columns = {k: IdlColumn(v) for k, v in table.columns.items()} -+ -+ def __getattr__(self, attr): -+ return getattr(self._table, attr) -+ -+ @property -+ def condition_state(self): -+ # read-only, no setter -+ return self._condition_state -+ -+ @property -+ def condition(self): -+ return self.condition_state.latest -+ -+ @condition.setter -+ def condition(self, condition): -+ assert(isinstance(condition, list)) -+ self.idl.cond_change(self.name, condition) -+ -+ @classmethod -+ def schema_tables(cls, idl, schema): -+ return {k: cls(idl, v) for k, v in schema.tables.items()} -+ -+ -+class IdlColumn(object): -+ def __init__(self, column): -+ self._column = column -+ self.alert = True -+ -+ def __getattr__(self, attr): -+ return getattr(self._column, attr) -+ -+ - class Idl(object): - """Open vSwitch Database Interface Definition Language (OVSDB IDL). + if (status != 0) +- DEBUGOUT("Failed to initialize HW, STATUS = %d\n", status); ++ DEBUGOUT("Failed to initialize HW, STATUS = %d", status); -@@ -241,7 +282,7 @@ class Idl(object): - assert isinstance(schema_helper, SchemaHelper) - schema = schema_helper.get_idl_schema() + return status; + } +@@ -156,8 +152,6 @@ s32 ngbe_reset_hw_em(struct ngbe_hw *hw) + { + s32 status; -- self.tables = schema.tables -+ self.tables = IdlTable.schema_tables(self, schema) - self.readonly = schema.readonly - self._db = schema - remotes = self._parse_remotes(remote) -@@ -282,15 +323,6 @@ class Idl(object): - self.cond_changed = False - self.cond_seqno = 0 +- DEBUGFUNC("ngbe_reset_hw_em"); +- + /* Call adapter stop to disable tx/rx and clear interrupts */ + status = hw->mac.stop_hw(hw); + if (status != 0) +@@ -205,8 +199,6 @@ s32 ngbe_clear_hw_cntrs(struct ngbe_hw *hw) + { + u16 i = 0; -- for table in schema.tables.values(): -- for column in table.columns.values(): -- if not hasattr(column, 'alert'): -- column.alert = True -- table.need_table = False -- table.rows = custom_index.IndexedRows(table) -- table.idl = self -- table.condition = ConditionState() +- DEBUGFUNC("ngbe_clear_hw_cntrs"); - - def _parse_remotes(self, remote): - # If remote is - - # "tcp:10.0.0.1:6641,unix:/tmp/db.sock,t,s,tcp:10.0.0.2:6642" -@@ -330,7 +362,7 @@ class Idl(object): - def ack_conditions(self): - """Mark all requested table conditions as acked""" - for table in self.tables.values(): -- table.condition.ack() -+ table.condition_state.ack() + /* QP Stats */ + /* don't write clear queue stats */ + for (i = 0; i < NGBE_MAX_QP; i++) { +@@ -305,8 +297,6 @@ s32 ngbe_get_mac_addr(struct ngbe_hw *hw, u8 *mac_addr) + u32 rar_low; + u16 i; + +- DEBUGFUNC("ngbe_get_mac_addr"); +- + wr32(hw, NGBE_ETHADDRIDX, 0); + rar_high = rd32(hw, NGBE_ETHADDRH); + rar_low = rd32(hw, NGBE_ETHADDRL); +@@ -332,8 +322,6 @@ void ngbe_set_lan_id_multi_port(struct ngbe_hw *hw) + struct ngbe_bus_info *bus = &hw->bus; + u32 reg = 0; + +- DEBUGFUNC("ngbe_set_lan_id_multi_port"); +- + reg = rd32(hw, NGBE_PORTSTAT); + bus->lan_id = NGBE_PORTSTAT_ID(reg); + bus->func = bus->lan_id; +@@ -350,10 +338,8 @@ void ngbe_set_lan_id_multi_port(struct ngbe_hw *hw) + **/ + s32 ngbe_stop_hw(struct ngbe_hw *hw) + { +- u32 reg_val; + u16 i; +- +- DEBUGFUNC("ngbe_stop_hw"); ++ s32 status = 0; + + /* + * Set the adapter_stopped flag so other driver functions stop touching +@@ -372,16 +358,27 @@ s32 ngbe_stop_hw(struct ngbe_hw *hw) + wr32(hw, NGBE_ICRMISC, NGBE_ICRMISC_MASK); + wr32(hw, NGBE_ICR(0), NGBE_ICR_MASK); + +- /* Disable the transmit unit. Each queue must be disabled. */ +- for (i = 0; i < hw->mac.max_tx_queues; i++) +- wr32(hw, NGBE_TXCFG(i), NGBE_TXCFG_FLUSH); ++ wr32(hw, NGBE_BMECTL, 0x3); + + /* Disable the receive unit by stopping each queue */ +- for (i = 0; i < hw->mac.max_rx_queues; i++) { +- reg_val = rd32(hw, NGBE_RXCFG(i)); +- reg_val &= ~NGBE_RXCFG_ENA; +- wr32(hw, NGBE_RXCFG(i), reg_val); +- } ++ for (i = 0; i < hw->mac.max_rx_queues; i++) ++ wr32(hw, NGBE_RXCFG(i), 0); ++ ++ /* flush all queues disables */ ++ ngbe_flush(hw); ++ msec_delay(2); ++ ++ /* ++ * Prevent the PCI-E bus from hanging by disabling PCI-E master ++ * access and verify no pending requests ++ */ ++ status = ngbe_set_pcie_master(hw, false); ++ if (status) ++ return status; ++ ++ /* Disable the transmit unit. Each queue must be disabled. */ ++ for (i = 0; i < hw->mac.max_tx_queues; i++) ++ wr32(hw, NGBE_TXCFG(i), 0); + + /* flush all queues disables */ + ngbe_flush(hw); +@@ -399,8 +396,6 @@ s32 ngbe_led_on(struct ngbe_hw *hw, u32 index) + { + u32 led_reg = rd32(hw, NGBE_LEDCTL); - def sync_conditions(self): - """Synchronize condition state when the FSM is restarted -@@ -356,14 +388,17 @@ class Idl(object): - flushing the local cached DB contents. - """ - ack_all = self.last_id == str(uuid.UUID(int=0)) -+ if ack_all: -+ self.cond_changed = False -+ - for table in self.tables.values(): - if ack_all: -- table.condition.request() -- table.condition.ack() -+ table.condition_state.request() -+ table.condition_state.ack() - else: -- if table.condition.reset(): -+ if table.condition_state.reset(): - self.last_id = str(uuid.UUID(int=0)) -- self.cond_changed = True -+ self.cond_changed = True +- DEBUGFUNC("ngbe_led_on"); +- + if (index > 3) + return NGBE_ERR_PARAM; - def restart_fsm(self): - # Resync data DB table conditions to avoid missing updated due to -@@ -482,7 +517,7 @@ class Idl(object): - sh.register_table(self._server_db_table) - schema = sh.get_idl_schema() - self._server_db = schema -- self.server_tables = schema.tables -+ self.server_tables = IdlTable.schema_tables(self, schema) - self.__send_server_monitor_request() - except error.Error as e: - vlog.err("%s: error receiving server schema: %s" -@@ -588,10 +623,10 @@ class Idl(object): - for table in self.tables.values(): - # Always use the most recent conditions set by the IDL client when - # requesting monitor_cond_change -- if table.condition.new is not None: -+ if table.condition_state.new is not None: - change_requests[table.name] = [ -- {"where": table.condition.new}] -- table.condition.request() -+ {"where": table.condition_state.new}] -+ table.condition_state.request() +@@ -421,8 +416,6 @@ s32 ngbe_led_off(struct ngbe_hw *hw, u32 index) + { + u32 led_reg = rd32(hw, NGBE_LEDCTL); - if not change_requests: - return -@@ -627,19 +662,20 @@ class Idl(object): - cond = [False] +- DEBUGFUNC("ngbe_led_off"); +- + if (index > 3) + return NGBE_ERR_PARAM; - # Compare the new condition to the last known condition -- if table.condition.latest != cond: -- table.condition.init(cond) -+ if table.condition_state.latest != cond: -+ table.condition_state.init(cond) - self.cond_changed = True +@@ -444,8 +437,6 @@ s32 ngbe_validate_mac_addr(u8 *mac_addr) + { + s32 status = 0; - # New condition will be sent out after all already requested ones - # are acked. -- if table.condition.new: -- any_reqs = any(t.condition.request for t in self.tables.values()) -+ if table.condition_state.new: -+ any_reqs = any(t.condition_state.request -+ for t in self.tables.values()) - return self.cond_seqno + int(any_reqs) + 1 +- DEBUGFUNC("ngbe_validate_mac_addr"); +- + /* Make sure it is not a multicast address */ + if (NGBE_IS_MULTICAST((struct rte_ether_addr *)mac_addr)) { + status = NGBE_ERR_INVALID_MAC_ADDR; +@@ -476,11 +467,9 @@ s32 ngbe_set_rar(struct ngbe_hw *hw, u32 index, u8 *addr, u32 vmdq, + u32 rar_low, rar_high; + u32 rar_entries = hw->mac.num_rar_entries; + +- DEBUGFUNC("ngbe_set_rar"); +- + /* Make sure we are using a valid rar index range */ + if (index >= rar_entries) { +- DEBUGOUT("RAR index %d is out of range.\n", index); ++ DEBUGOUT("RAR index %d is out of range.", index); + return NGBE_ERR_INVALID_ARGUMENT; + } - # Already requested conditions should be up to date at - # self.cond_seqno + 1 while acked conditions are already up to date -- return self.cond_seqno + int(bool(table.condition.requested)) -+ return self.cond_seqno + int(bool(table.condition_state.requested)) +@@ -528,11 +517,9 @@ s32 ngbe_clear_rar(struct ngbe_hw *hw, u32 index) + u32 rar_high; + u32 rar_entries = hw->mac.num_rar_entries; - def wait(self, poller): - """Arranges for poller.block() to wake up when self.run() has something -@@ -811,8 +847,8 @@ class Idl(object): - columns.append(column) - monitor_request = {"columns": columns} - if method in ("monitor_cond", "monitor_cond_since") and ( -- not ConditionState.is_true(table.condition.acked)): -- monitor_request["where"] = table.condition.acked -+ not ConditionState.is_true(table.condition_state.acked)): -+ monitor_request["where"] = table.condition_state.acked - monitor_requests[table.name] = [monitor_request] +- DEBUGFUNC("ngbe_clear_rar"); +- + /* Make sure we are using a valid rar index range */ + if (index >= rar_entries) { +- DEBUGOUT("RAR index %d is out of range.\n", index); ++ DEBUGOUT("RAR index %d is out of range.", index); + return NGBE_ERR_INVALID_ARGUMENT; + } - args = [self._db.name, str(self.uuid), monitor_requests] -@@ -1148,13 +1184,6 @@ class Idl(object): - return True +@@ -568,8 +555,6 @@ s32 ngbe_init_rx_addrs(struct ngbe_hw *hw) + u32 psrctl; + u32 rar_entries = hw->mac.num_rar_entries; +- DEBUGFUNC("ngbe_init_rx_addrs"); +- + /* + * If the current mac address is valid, assume it is a software override + * to the permanent address. +@@ -580,18 +565,18 @@ s32 ngbe_init_rx_addrs(struct ngbe_hw *hw) + /* Get the MAC address from the RAR0 for later reference */ + hw->mac.get_mac_addr(hw, hw->mac.addr); + +- DEBUGOUT(" Keeping Current RAR0 Addr =%.2X %.2X %.2X ", ++ DEBUGOUT(" Keeping Current RAR0 Addr = " ++ RTE_ETHER_ADDR_PRT_FMT, + hw->mac.addr[0], hw->mac.addr[1], +- hw->mac.addr[2]); +- DEBUGOUT("%.2X %.2X %.2X\n", hw->mac.addr[3], ++ hw->mac.addr[2], hw->mac.addr[3], + hw->mac.addr[4], hw->mac.addr[5]); + } else { + /* Setup the receive address. */ +- DEBUGOUT("Overriding MAC Address in RAR[0]\n"); +- DEBUGOUT(" New MAC Addr =%.2X %.2X %.2X ", ++ DEBUGOUT("Overriding MAC Address in RAR[0]"); ++ DEBUGOUT(" New MAC Addr = " ++ RTE_ETHER_ADDR_PRT_FMT, + hw->mac.addr[0], hw->mac.addr[1], +- hw->mac.addr[2]); +- DEBUGOUT("%.2X %.2X %.2X\n", hw->mac.addr[3], ++ hw->mac.addr[2], hw->mac.addr[3], + hw->mac.addr[4], hw->mac.addr[5]); + + hw->mac.set_rar(hw, 0, hw->mac.addr, 0, true); +@@ -601,7 +586,7 @@ s32 ngbe_init_rx_addrs(struct ngbe_hw *hw) + hw->mac.clear_vmdq(hw, 0, BIT_MASK32); + + /* Zero out the other receive addresses. */ +- DEBUGOUT("Clearing RAR[1-%d]\n", rar_entries - 1); ++ DEBUGOUT("Clearing RAR[1-%d]", rar_entries - 1); + for (i = 1; i < rar_entries; i++) { + wr32(hw, NGBE_ETHADDRIDX, i); + wr32(hw, NGBE_ETHADDRL, 0); +@@ -615,7 +600,7 @@ s32 ngbe_init_rx_addrs(struct ngbe_hw *hw) + psrctl |= NGBE_PSRCTL_ADHF12(hw->mac.mc_filter_type); + wr32(hw, NGBE_PSRCTL, psrctl); + +- DEBUGOUT(" Clearing MTA\n"); ++ DEBUGOUT(" Clearing MTA"); + for (i = 0; i < hw->mac.mcft_size; i++) + wr32(hw, NGBE_MCADDRTBL(i), 0); + +@@ -640,8 +625,6 @@ static s32 ngbe_mta_vector(struct ngbe_hw *hw, u8 *mc_addr) + { + u32 vector = 0; --def _uuid_to_row(atom, base): -- if base.ref_table: -- return base.ref_table.rows.get(atom) -- else: -- return atom +- DEBUGFUNC("ngbe_mta_vector"); - + switch (hw->mac.mc_filter_type) { + case 0: /* use bits [47:36] of the address */ + vector = ((mc_addr[4] >> 4) | (((u16)mc_addr[5]) << 4)); +@@ -656,7 +639,7 @@ static s32 ngbe_mta_vector(struct ngbe_hw *hw, u8 *mc_addr) + vector = ((mc_addr[4]) | (((u16)mc_addr[5]) << 8)); + break; + default: /* Invalid mc_filter_type */ +- DEBUGOUT("MC filter type param set incorrectly\n"); ++ DEBUGOUT("MC filter type param set incorrectly"); + ASSERT(0); + break; + } +@@ -679,12 +662,10 @@ void ngbe_set_mta(struct ngbe_hw *hw, u8 *mc_addr) + u32 vector_bit; + u32 vector_reg; + +- DEBUGFUNC("ngbe_set_mta"); - - def _row_to_uuid(value): - if isinstance(value, Row): - return value.uuid -@@ -1268,6 +1297,17 @@ class Row(object): - data=", ".join("{col}={val}".format(col=c, val=getattr(self, c)) - for c in sorted(self._table.columns))) + hw->addr_ctrl.mta_in_use++; -+ def _uuid_to_row(self, atom, base): -+ if base.ref_table: -+ try: -+ table = self._idl.tables[base.ref_table.name] -+ except KeyError as e: -+ msg = "Table {} is not registered".format(base.ref_table.name) -+ raise AttributeError(msg) from e -+ return table.rows.get(atom) -+ else: -+ return atom -+ - def __getattr__(self, column_name): - assert self._changes is not None - assert self._mutations is not None -@@ -1309,7 +1349,7 @@ class Row(object): - datum = data.Datum.from_python(column.type, dlist, - _row_to_uuid) - elif column.type.is_map(): -- dmap = datum.to_python(_uuid_to_row) -+ dmap = datum.to_python(self._uuid_to_row) - if inserts is not None: - dmap.update(inserts) - if removes is not None: -@@ -1326,7 +1366,7 @@ class Row(object): - else: - datum = inserts + vector = ngbe_mta_vector(hw, mc_addr); +- DEBUGOUT(" bit-vector = 0x%03X\n", vector); ++ DEBUGOUT(" bit-vector = 0x%03X", vector); + + /* + * The MTA is a register array of 128 32-bit registers. It is treated +@@ -718,8 +699,6 @@ s32 ngbe_update_mc_addr_list(struct ngbe_hw *hw, u8 *mc_addr_list, + u32 i; + u32 vmdq; + +- DEBUGFUNC("ngbe_update_mc_addr_list"); +- + /* + * Set the new number of MC addresses that we are being requested to + * use. +@@ -729,13 +708,13 @@ s32 ngbe_update_mc_addr_list(struct ngbe_hw *hw, u8 *mc_addr_list, + + /* Clear mta_shadow */ + if (clear) { +- DEBUGOUT(" Clearing MTA\n"); ++ DEBUGOUT(" Clearing MTA"); + memset(&hw->mac.mta_shadow, 0, sizeof(hw->mac.mta_shadow)); + } + + /* Update mta_shadow */ + for (i = 0; i < mc_addr_count; i++) { +- DEBUGOUT(" Adding the multicast addresses:\n"); ++ DEBUGOUT(" Adding the multicast addresses:"); + ngbe_set_mta(hw, next(hw, &mc_addr_list, &vmdq)); + } + +@@ -752,7 +731,7 @@ s32 ngbe_update_mc_addr_list(struct ngbe_hw *hw, u8 *mc_addr_list, + wr32(hw, NGBE_PSRCTL, psrctl); + } + +- DEBUGOUT("ngbe update mc addr list complete\n"); ++ DEBUGOUT("ngbe update mc addr list complete"); + return 0; + } + +@@ -767,11 +746,9 @@ s32 ngbe_setup_fc_em(struct ngbe_hw *hw) + s32 err = 0; + u16 reg_cu = 0; + +- DEBUGFUNC("ngbe_setup_fc"); +- + /* Validate the requested mode */ + if (hw->fc.strict_ieee && hw->fc.requested_mode == ngbe_fc_rx_pause) { +- DEBUGOUT("ngbe_fc_rx_pause not valid in strict IEEE mode\n"); ++ DEBUGOUT("ngbe_fc_rx_pause not valid in strict IEEE mode"); + err = NGBE_ERR_INVALID_LINK_SETTINGS; + goto out; + } +@@ -827,7 +804,7 @@ s32 ngbe_setup_fc_em(struct ngbe_hw *hw) + reg_cu |= 0xC00; /*need to merge rtl and mvl on page 0*/ + break; + default: +- DEBUGOUT("Flow control param set incorrectly\n"); ++ DEBUGOUT("Flow control param set incorrectly"); + err = NGBE_ERR_CONFIG; + goto out; + } +@@ -851,8 +828,6 @@ s32 ngbe_fc_enable(struct ngbe_hw *hw) + u32 pause_time; + u32 fcrtl, fcrth; + +- DEBUGFUNC("ngbe_fc_enable"); +- + /* Validate the water mark configuration */ + if (!hw->fc.pause_time) { + err = NGBE_ERR_INVALID_LINK_SETTINGS; +@@ -863,7 +838,7 @@ s32 ngbe_fc_enable(struct ngbe_hw *hw) + if ((hw->fc.current_mode & ngbe_fc_tx_pause) && hw->fc.high_water) { + if (!hw->fc.low_water || + hw->fc.low_water >= hw->fc.high_water) { +- DEBUGOUT("Invalid water mark configuration\n"); ++ DEBUGOUT("Invalid water mark configuration"); + err = NGBE_ERR_INVALID_LINK_SETTINGS; + goto out; + } +@@ -919,7 +894,7 @@ s32 ngbe_fc_enable(struct ngbe_hw *hw) + fccfg_reg |= NGBE_TXFCCFG_FC; + break; + default: +- DEBUGOUT("Flow control param set incorrectly\n"); ++ DEBUGOUT("Flow control param set incorrectly"); + err = NGBE_ERR_CONFIG; + goto out; + } +@@ -977,8 +952,7 @@ s32 ngbe_negotiate_fc(struct ngbe_hw *hw, u32 adv_reg, u32 lp_reg, + u32 adv_sym, u32 adv_asm, u32 lp_sym, u32 lp_asm) + { + if ((!(adv_reg)) || (!(lp_reg))) { +- DEBUGOUT("Local or link partner's advertised flow control " +- "settings are NULL. Local: %x, link partner: %x\n", ++ DEBUGOUT("Local or link partner's advertised flow control settings are NULL. Local: %x, link partner: %x", + adv_reg, lp_reg); + return NGBE_ERR_FC_NOT_NEGOTIATED; + } +@@ -993,22 +967,22 @@ s32 ngbe_negotiate_fc(struct ngbe_hw *hw, u32 adv_reg, u32 lp_reg, + */ + if (hw->fc.requested_mode == ngbe_fc_full) { + hw->fc.current_mode = ngbe_fc_full; +- DEBUGOUT("Flow Control = FULL.\n"); ++ DEBUGOUT("Flow Control = FULL."); + } else { + hw->fc.current_mode = ngbe_fc_rx_pause; +- DEBUGOUT("Flow Control=RX PAUSE frames only\n"); ++ DEBUGOUT("Flow Control=RX PAUSE frames only"); + } + } else if (!(adv_reg & adv_sym) && (adv_reg & adv_asm) && + (lp_reg & lp_sym) && (lp_reg & lp_asm)) { + hw->fc.current_mode = ngbe_fc_tx_pause; +- DEBUGOUT("Flow Control = TX PAUSE frames only.\n"); ++ DEBUGOUT("Flow Control = TX PAUSE frames only."); + } else if ((adv_reg & adv_sym) && (adv_reg & adv_asm) && + !(lp_reg & lp_sym) && (lp_reg & lp_asm)) { + hw->fc.current_mode = ngbe_fc_rx_pause; +- DEBUGOUT("Flow Control = RX PAUSE frames only.\n"); ++ DEBUGOUT("Flow Control = RX PAUSE frames only."); + } else { + hw->fc.current_mode = ngbe_fc_none; +- DEBUGOUT("Flow Control = NONE.\n"); ++ DEBUGOUT("Flow Control = NONE."); + } + return 0; + } +@@ -1046,8 +1020,6 @@ void ngbe_fc_autoneg(struct ngbe_hw *hw) + u32 speed; + bool link_up; + +- DEBUGFUNC("ngbe_fc_autoneg"); +- + /* + * AN should have completed when the cable was plugged in. + * Look for reasons to bail out. Bail out if: +@@ -1076,6 +1048,64 @@ void ngbe_fc_autoneg(struct ngbe_hw *hw) + } + } + ++/** ++ * ngbe_set_pcie_master - Disable or Enable PCI-express master access ++ * @hw: pointer to hardware structure ++ * ++ * Disables PCI-Express master access and verifies there are no pending ++ * requests. NGBE_ERR_MASTER_REQUESTS_PENDING is returned if master disable ++ * bit hasn't caused the master requests to be disabled, else 0 ++ * is returned signifying master requests disabled. ++ **/ ++s32 ngbe_set_pcie_master(struct ngbe_hw *hw, bool enable) ++{ ++ struct rte_pci_device *pci_dev = (struct rte_pci_device *)hw->back; ++ s32 status = 0; ++ s32 ret = 0; ++ u32 i; ++ u16 reg; ++ ++ ret = rte_pci_read_config(pci_dev, ®, ++ sizeof(reg), PCI_COMMAND); ++ if (ret != sizeof(reg)) { ++ DEBUGOUT("Cannot read command from PCI config space!\n"); ++ return -1; ++ } ++ ++ if (enable) ++ reg |= PCI_COMMAND_MASTER; ++ else ++ reg &= ~PCI_COMMAND_MASTER; ++ ++ ret = rte_pci_write_config(pci_dev, ®, ++ sizeof(reg), PCI_COMMAND); ++ if (ret != sizeof(reg)) { ++ DEBUGOUT("Cannot write command to PCI config space!\n"); ++ return -1; ++ } ++ ++ if (enable) ++ goto out; ++ ++ /* Exit if master requests are blocked */ ++ if (!(rd32(hw, NGBE_BMEPEND)) || ++ NGBE_REMOVED(hw->hw_addr)) ++ goto out; ++ ++ /* Poll for master request bit to clear */ ++ for (i = 0; i < NGBE_PCI_MASTER_DISABLE_TIMEOUT; i++) { ++ usec_delay(100); ++ if (!(rd32(hw, NGBE_BMEPEND))) ++ goto out; ++ } ++ ++ DEBUGOUT("PCIe transaction pending bit also did not clear."); ++ status = NGBE_ERR_MASTER_REQUESTS_PENDING; ++ ++out: ++ return status; ++} ++ + /** + * ngbe_acquire_swfw_sync - Acquire SWFW semaphore + * @hw: pointer to hardware structure +@@ -1092,8 +1122,6 @@ s32 ngbe_acquire_swfw_sync(struct ngbe_hw *hw, u32 mask) + u32 timeout = 200; + u32 i; + +- DEBUGFUNC("ngbe_acquire_swfw_sync"); +- + for (i = 0; i < timeout; i++) { + /* + * SW NVM semaphore bit is used for access to all +@@ -1136,8 +1164,6 @@ void ngbe_release_swfw_sync(struct ngbe_hw *hw, u32 mask) + u32 mngsem; + u32 swmask = mask; + +- DEBUGFUNC("ngbe_release_swfw_sync"); +- + ngbe_get_eeprom_semaphore(hw); + + mngsem = rd32(hw, NGBE_MNGSEM); +@@ -1161,9 +1187,6 @@ s32 ngbe_disable_sec_rx_path(struct ngbe_hw *hw) + int i; + u32 secrxreg; + +- DEBUGFUNC("ngbe_disable_sec_rx_path"); +- +- + secrxreg = rd32(hw, NGBE_SECRXCTL); + secrxreg |= NGBE_SECRXCTL_XDSA; + wr32(hw, NGBE_SECRXCTL, secrxreg); +@@ -1178,8 +1201,7 @@ s32 ngbe_disable_sec_rx_path(struct ngbe_hw *hw) + + /* For informational purposes only */ + if (i >= NGBE_MAX_SECRX_POLL) +- DEBUGOUT("Rx unit being enabled before security " +- "path fully disabled. Continuing with init.\n"); ++ DEBUGOUT("Rx unit being enabled before security path fully disabled. Continuing with init."); + + return 0; + } +@@ -1194,8 +1216,6 @@ s32 ngbe_enable_sec_rx_path(struct ngbe_hw *hw) + { + u32 secrxreg; + +- DEBUGFUNC("ngbe_enable_sec_rx_path"); +- + secrxreg = rd32(hw, NGBE_SECRXCTL); + secrxreg &= ~NGBE_SECRXCTL_XDSA; + wr32(hw, NGBE_SECRXCTL, secrxreg); +@@ -1215,11 +1235,9 @@ s32 ngbe_clear_vmdq(struct ngbe_hw *hw, u32 rar, u32 vmdq) + u32 mpsar; + u32 rar_entries = hw->mac.num_rar_entries; + +- DEBUGFUNC("ngbe_clear_vmdq"); +- + /* Make sure we are using a valid rar index range */ + if (rar >= rar_entries) { +- DEBUGOUT("RAR index %d is out of range.\n", rar); ++ DEBUGOUT("RAR index %d is out of range.", rar); + return NGBE_ERR_INVALID_ARGUMENT; + } + +@@ -1253,11 +1271,9 @@ s32 ngbe_set_vmdq(struct ngbe_hw *hw, u32 rar, u32 vmdq) + u32 mpsar; + u32 rar_entries = hw->mac.num_rar_entries; + +- DEBUGFUNC("ngbe_set_vmdq"); +- + /* Make sure we are using a valid rar index range */ + if (rar >= rar_entries) { +- DEBUGOUT("RAR index %d is out of range.\n", rar); ++ DEBUGOUT("RAR index %d is out of range.", rar); + return NGBE_ERR_INVALID_ARGUMENT; + } + +@@ -1278,8 +1294,7 @@ s32 ngbe_init_uta_tables(struct ngbe_hw *hw) + { + int i; + +- DEBUGFUNC("ngbe_init_uta_tables"); +- DEBUGOUT(" Clearing UTA\n"); ++ DEBUGOUT(" Clearing UTA"); + + for (i = 0; i < 128; i++) + wr32(hw, NGBE_UCADDRTBL(i), 0); +@@ -1334,7 +1349,7 @@ s32 ngbe_find_vlvf_slot(struct ngbe_hw *hw, u32 vlan, bool vlvf_bypass) + * slot we found during our search, else error. + */ + if (!first_empty_slot) +- DEBUGOUT("No space in VLVF.\n"); ++ DEBUGOUT("No space in VLVF."); + + return first_empty_slot ? first_empty_slot : NGBE_ERR_NO_SPACE; + } +@@ -1355,8 +1370,6 @@ s32 ngbe_set_vfta(struct ngbe_hw *hw, u32 vlan, u32 vind, + u32 regidx, vfta_delta, vfta; + s32 err; + +- DEBUGFUNC("ngbe_set_vfta"); +- + if (vlan > 4095 || vind > 63) + return NGBE_ERR_PARAM; + +@@ -1424,8 +1437,6 @@ s32 ngbe_set_vlvf(struct ngbe_hw *hw, u32 vlan, u32 vind, + u32 portctl; + s32 vlvf_index; + +- DEBUGFUNC("ngbe_set_vlvf"); +- + if (vlan > 4095 || vind > 63) + return NGBE_ERR_PARAM; + +@@ -1505,8 +1516,6 @@ s32 ngbe_clear_vfta(struct ngbe_hw *hw) + { + u32 offset; + +- DEBUGFUNC("ngbe_clear_vfta"); +- + for (offset = 0; offset < hw->mac.vft_size; offset++) + wr32(hw, NGBE_VLANTBL(offset), 0); + +@@ -1534,8 +1543,6 @@ s32 ngbe_check_mac_link_em(struct ngbe_hw *hw, u32 *speed, + u32 i, reg; + s32 status = 0; + +- DEBUGFUNC("ngbe_check_mac_link_em"); +- + reg = rd32(hw, NGBE_GPIOINTSTAT); + wr32(hw, NGBE_GPIOEOI, reg); + +@@ -1559,7 +1566,6 @@ s32 ngbe_get_link_capabilities_em(struct ngbe_hw *hw, + { + s32 status = 0; + +- DEBUGFUNC("\n"); + + hw->mac.autoneg = *autoneg; + +@@ -1582,8 +1588,6 @@ s32 ngbe_setup_mac_link_em(struct ngbe_hw *hw, + { + s32 status; + +- DEBUGFUNC("\n"); +- + /* Setup the PHY according to input speed */ + status = hw->phy.setup_link(hw, speed, autoneg_wait_to_complete); + +@@ -1609,6 +1613,30 @@ void ngbe_set_mac_anti_spoofing(struct ngbe_hw *hw, bool enable, int vf) + wr32(hw, NGBE_POOLTXASMAC, pfvfspoof); + } + ++/** ++ * ngbe_set_pba - Initialize Rx packet buffer ++ * @hw: pointer to hardware structure ++ * @headroom: reserve n KB of headroom ++ **/ ++void ngbe_set_pba(struct ngbe_hw *hw) ++{ ++ u32 rxpktsize = hw->mac.rx_pb_size; ++ u32 txpktsize, txpbthresh; ++ ++ /* Reserve 256 KB of headroom */ ++ rxpktsize -= 256; ++ ++ rxpktsize <<= 10; ++ wr32(hw, NGBE_PBRXSIZE, rxpktsize); ++ ++ /* Only support an equally distributed Tx packet buffer strategy. */ ++ txpktsize = NGBE_PBTXSIZE_MAX; ++ txpbthresh = (txpktsize / 1024) - NGBE_TXPKT_SIZE_MAX; ++ ++ wr32(hw, NGBE_PBTXSIZE, txpktsize); ++ wr32(hw, NGBE_PBTXDMATH, txpbthresh); ++} ++ + /** + * ngbe_set_vlan_anti_spoofing - Enable/Disable VLAN anti-spoofing + * @hw: pointer to hardware structure +@@ -1639,8 +1667,6 @@ s32 ngbe_init_thermal_sensor_thresh(struct ngbe_hw *hw) + { + struct ngbe_thermal_sensor_data *data = &hw->mac.thermal_sensor_data; + +- DEBUGFUNC("ngbe_init_thermal_sensor_thresh"); +- + memset(data, 0, sizeof(struct ngbe_thermal_sensor_data)); + + if (hw->bus.lan_id != 0) +@@ -1664,8 +1690,6 @@ s32 ngbe_mac_check_overtemp(struct ngbe_hw *hw) + s32 status = 0; + u32 ts_state; + +- DEBUGFUNC("ngbe_mac_check_overtemp"); +- + /* Check that the LASI temp alarm status was triggered */ + ts_state = rd32(hw, NGBE_TSALM); + +@@ -1720,8 +1744,6 @@ s32 ngbe_set_mac_type(struct ngbe_hw *hw) + { + s32 err = 0; + +- DEBUGFUNC("ngbe_set_mac_type"); +- + if (hw->vendor_id != PCI_VENDOR_ID_WANGXUN) { + DEBUGOUT("Unsupported vendor id: %x", hw->vendor_id); + return NGBE_ERR_DEVICE_NOT_SUPPORTED; +@@ -1750,7 +1772,7 @@ s32 ngbe_set_mac_type(struct ngbe_hw *hw) + break; + } + +- DEBUGOUT("found mac: %d media: %d, returns: %d\n", ++ DEBUGOUT("found mac: %d media: %d, returns: %d", + hw->mac.type, hw->phy.media_type, err); + return err; + } +@@ -1764,15 +1786,12 @@ s32 ngbe_set_mac_type(struct ngbe_hw *hw) + **/ + s32 ngbe_enable_rx_dma(struct ngbe_hw *hw, u32 regval) + { +- DEBUGFUNC("ngbe_enable_rx_dma"); +- + /* + * Workaround silicon errata when enabling the Rx datapath. + * If traffic is incoming before we enable the Rx unit, it could hang + * the Rx DMA unit. Therefore, make sure the security engine is + * completely disabled prior to enabling the Rx unit. + */ +- + hw->mac.disable_sec_rx_path(hw); + + if (regval & NGBE_PBRXCTL_ENA) +@@ -1853,8 +1872,6 @@ s32 ngbe_init_ops_pf(struct ngbe_hw *hw) + struct ngbe_rom_info *rom = &hw->rom; + struct ngbe_mbx_info *mbx = &hw->mbx; + +- DEBUGFUNC("ngbe_init_ops_pf"); +- + /* BUS */ + bus->set_lan_id = ngbe_set_lan_id_multi_port; + +@@ -1907,6 +1924,8 @@ s32 ngbe_init_ops_pf(struct ngbe_hw *hw) + mac->check_link = ngbe_check_mac_link_em; + mac->setup_link = ngbe_setup_mac_link_em; + ++ mac->setup_pba = ngbe_set_pba; ++ + /* Manageability interface */ + mac->init_thermal_sensor_thresh = ngbe_init_thermal_sensor_thresh; + mac->check_overtemp = ngbe_mac_check_overtemp; +@@ -1928,6 +1947,7 @@ s32 ngbe_init_ops_pf(struct ngbe_hw *hw) + mac->mcft_size = NGBE_EM_MC_TBL_SIZE; + mac->vft_size = NGBE_EM_VFT_TBL_SIZE; + mac->num_rar_entries = NGBE_EM_RAR_ENTRIES; ++ mac->rx_pb_size = NGBE_EM_RX_PB_SIZE; + mac->max_rx_queues = NGBE_EM_MAX_RX_QUEUES; + mac->max_tx_queues = NGBE_EM_MAX_TX_QUEUES; + +@@ -1953,8 +1973,6 @@ s32 ngbe_init_shared_code(struct ngbe_hw *hw) + { + s32 status = 0; + +- DEBUGFUNC("ngbe_init_shared_code"); +- + /* + * Set the mac type + */ +diff --git a/dpdk/drivers/net/ngbe/base/ngbe_hw.h b/dpdk/drivers/net/ngbe/base/ngbe_hw.h +index ad7e8fc2d9..7e0e23b195 100644 +--- a/dpdk/drivers/net/ngbe/base/ngbe_hw.h ++++ b/dpdk/drivers/net/ngbe/base/ngbe_hw.h +@@ -13,6 +13,7 @@ + #define NGBE_EM_RAR_ENTRIES 32 + #define NGBE_EM_MC_TBL_SIZE 32 + #define NGBE_EM_VFT_TBL_SIZE 128 ++#define NGBE_EM_RX_PB_SIZE 42 /*KB*/ + + s32 ngbe_init_hw(struct ngbe_hw *hw); + s32 ngbe_start_hw(struct ngbe_hw *hw); +@@ -44,6 +45,7 @@ s32 ngbe_update_mc_addr_list(struct ngbe_hw *hw, u8 *mc_addr_list, + ngbe_mc_addr_itr func, bool clear); + s32 ngbe_disable_sec_rx_path(struct ngbe_hw *hw); + s32 ngbe_enable_sec_rx_path(struct ngbe_hw *hw); ++void ngbe_set_pba(struct ngbe_hw *hw); + + s32 ngbe_setup_fc_em(struct ngbe_hw *hw); + s32 ngbe_fc_enable(struct ngbe_hw *hw); +@@ -52,6 +54,7 @@ void ngbe_fc_autoneg(struct ngbe_hw *hw); + s32 ngbe_validate_mac_addr(u8 *mac_addr); + s32 ngbe_acquire_swfw_sync(struct ngbe_hw *hw, u32 mask); + void ngbe_release_swfw_sync(struct ngbe_hw *hw, u32 mask); ++s32 ngbe_set_pcie_master(struct ngbe_hw *hw, bool enable); + + s32 ngbe_set_vmdq(struct ngbe_hw *hw, u32 rar, u32 vmdq); + s32 ngbe_clear_vmdq(struct ngbe_hw *hw, u32 rar, u32 vmdq); +diff --git a/dpdk/drivers/net/ngbe/base/ngbe_mbx.c b/dpdk/drivers/net/ngbe/base/ngbe_mbx.c +index 764ae81319..bc0adbb3ec 100644 +--- a/dpdk/drivers/net/ngbe/base/ngbe_mbx.c ++++ b/dpdk/drivers/net/ngbe/base/ngbe_mbx.c +@@ -21,8 +21,6 @@ s32 ngbe_read_mbx(struct ngbe_hw *hw, u32 *msg, u16 size, u16 mbx_id) + struct ngbe_mbx_info *mbx = &hw->mbx; + s32 ret_val = NGBE_ERR_MBX; + +- DEBUGFUNC("ngbe_read_mbx"); +- + /* limit read to size of mailbox */ + if (size > mbx->size) + size = mbx->size; +@@ -47,8 +45,6 @@ s32 ngbe_write_mbx(struct ngbe_hw *hw, u32 *msg, u16 size, u16 mbx_id) + struct ngbe_mbx_info *mbx = &hw->mbx; + s32 ret_val = 0; + +- DEBUGFUNC("ngbe_write_mbx"); +- + if (size > mbx->size) { + ret_val = NGBE_ERR_MBX; + DEBUGOUT("Invalid mailbox message size %d", size); +@@ -71,8 +67,6 @@ s32 ngbe_check_for_msg(struct ngbe_hw *hw, u16 mbx_id) + struct ngbe_mbx_info *mbx = &hw->mbx; + s32 ret_val = NGBE_ERR_MBX; + +- DEBUGFUNC("ngbe_check_for_msg"); +- + if (mbx->check_for_msg) + ret_val = mbx->check_for_msg(hw, mbx_id); + +@@ -91,8 +85,6 @@ s32 ngbe_check_for_ack(struct ngbe_hw *hw, u16 mbx_id) + struct ngbe_mbx_info *mbx = &hw->mbx; + s32 ret_val = NGBE_ERR_MBX; + +- DEBUGFUNC("ngbe_check_for_ack"); +- + if (mbx->check_for_ack) + ret_val = mbx->check_for_ack(hw, mbx_id); + +@@ -111,8 +103,6 @@ s32 ngbe_check_for_rst(struct ngbe_hw *hw, u16 mbx_id) + struct ngbe_mbx_info *mbx = &hw->mbx; + s32 ret_val = NGBE_ERR_MBX; + +- DEBUGFUNC("ngbe_check_for_rst"); +- + if (mbx->check_for_rst) + ret_val = mbx->check_for_rst(hw, mbx_id); + +@@ -144,8 +134,6 @@ s32 ngbe_check_for_msg_pf(struct ngbe_hw *hw, u16 vf_number) + s32 ret_val = NGBE_ERR_MBX; + u32 vf_bit = vf_number; + +- DEBUGFUNC("ngbe_check_for_msg_pf"); +- + if (!ngbe_check_for_bit_pf(hw, NGBE_MBVFICR_VFREQ_VF1 << vf_bit)) { + ret_val = 0; + hw->mbx.stats.reqs++; +@@ -166,8 +154,6 @@ s32 ngbe_check_for_ack_pf(struct ngbe_hw *hw, u16 vf_number) + s32 ret_val = NGBE_ERR_MBX; + u32 vf_bit = vf_number; + +- DEBUGFUNC("ngbe_check_for_ack_pf"); +- + if (!ngbe_check_for_bit_pf(hw, NGBE_MBVFICR_VFACK_VF1 << vf_bit)) { + ret_val = 0; + hw->mbx.stats.acks++; +@@ -188,8 +174,6 @@ s32 ngbe_check_for_rst_pf(struct ngbe_hw *hw, u16 vf_number) + u32 vflre = 0; + s32 ret_val = NGBE_ERR_MBX; + +- DEBUGFUNC("ngbe_check_for_rst_pf"); +- + vflre = rd32(hw, NGBE_FLRVFE); + if (vflre & (1 << vf_number)) { + ret_val = 0; +@@ -212,8 +196,6 @@ STATIC s32 ngbe_obtain_mbx_lock_pf(struct ngbe_hw *hw, u16 vf_number) + s32 ret_val = NGBE_ERR_MBX; + u32 p2v_mailbox; + +- DEBUGFUNC("ngbe_obtain_mbx_lock_pf"); +- + /* Take ownership of the buffer */ + wr32(hw, NGBE_MBCTL(vf_number), NGBE_MBCTL_PFU); + +@@ -242,8 +224,6 @@ s32 ngbe_write_mbx_pf(struct ngbe_hw *hw, u32 *msg, u16 size, u16 vf_number) + s32 ret_val; + u16 i; + +- DEBUGFUNC("ngbe_write_mbx_pf"); +- + /* lock the mailbox to prevent pf/vf race condition */ + ret_val = ngbe_obtain_mbx_lock_pf(hw, vf_number); + if (ret_val) +@@ -283,8 +263,6 @@ s32 ngbe_read_mbx_pf(struct ngbe_hw *hw, u32 *msg, u16 size, u16 vf_number) + s32 ret_val; + u16 i; + +- DEBUGFUNC("ngbe_read_mbx_pf"); +- + /* lock the mailbox to prevent pf/vf race condition */ + ret_val = ngbe_obtain_mbx_lock_pf(hw, vf_number); + if (ret_val) +diff --git a/dpdk/drivers/net/ngbe/base/ngbe_mng.c b/dpdk/drivers/net/ngbe/base/ngbe_mng.c +index a3dd8093ce..dd32644bfe 100644 +--- a/dpdk/drivers/net/ngbe/base/ngbe_mng.c ++++ b/dpdk/drivers/net/ngbe/base/ngbe_mng.c +@@ -26,10 +26,8 @@ ngbe_hic_unlocked(struct ngbe_hw *hw, u32 *buffer, u32 length, u32 timeout) + u32 value, loop; + u16 i, dword_len; + +- DEBUGFUNC("ngbe_hic_unlocked"); +- + if (!length || length > NGBE_PMMBX_BSIZE) { +- DEBUGOUT("Buffer length failure buffersize=%d.\n", length); ++ DEBUGOUT("Buffer length failure buffersize=%d.", length); + return NGBE_ERR_HOST_INTERFACE_COMMAND; + } + +@@ -59,7 +57,7 @@ ngbe_hic_unlocked(struct ngbe_hw *hw, u32 *buffer, u32 length, u32 timeout) + NGBE_MNGMBXCTL_FWRDY, NGBE_MNGMBXCTL_FWRDY, + &value, timeout, 1000); + if (!loop || !(value & NGBE_MNGMBXCTL_FWACK)) { +- DEBUGOUT("Command has failed with no status valid.\n"); ++ DEBUGOUT("Command has failed with no status valid."); + return NGBE_ERR_HOST_INTERFACE_COMMAND; + } + +@@ -95,10 +93,8 @@ ngbe_host_interface_command(struct ngbe_hw *hw, u32 *buffer, + u32 bi; + u32 dword_len; + +- DEBUGFUNC("ngbe_host_interface_command"); +- + if (length == 0 || length > NGBE_PMMBX_BSIZE) { +- DEBUGOUT("Buffer length failure buffersize=%d.\n", length); ++ DEBUGOUT("Buffer length failure buffersize=%d.", length); + return NGBE_ERR_HOST_INTERFACE_COMMAND; + } + +@@ -140,7 +136,7 @@ ngbe_host_interface_command(struct ngbe_hw *hw, u32 *buffer, + goto rel_out; + + if (length < buf_len + hdr_size) { +- DEBUGOUT("Buffer not large enough for reply message.\n"); ++ DEBUGOUT("Buffer not large enough for reply message."); + err = NGBE_ERR_HOST_INTERFACE_COMMAND; + goto rel_out; + } +@@ -243,14 +239,69 @@ s32 ngbe_hic_sr_write(struct ngbe_hw *hw, u32 addr, u8 *buf, int len) + return err; + } + ++s32 ngbe_hic_pcie_read(struct ngbe_hw *hw, u16 addr, u32 *buf, int len) ++{ ++ struct ngbe_hic_read_pcie command; ++ u32 value = 0; ++ int err, i = 0; ++ ++ if (len > NGBE_PMMBX_DATA_SIZE) ++ return NGBE_ERR_HOST_INTERFACE_COMMAND; ++ ++ memset(&command, 0, sizeof(command)); ++ command.hdr.cmd = FW_PCIE_READ_CMD; ++ command.hdr.buf_len = sizeof(command) - sizeof(command.hdr); ++ command.hdr.checksum = FW_DEFAULT_CHECKSUM; ++ command.lan_id = hw->bus.lan_id; ++ command.addr = addr; ++ ++ err = ngbe_host_interface_command(hw, (u32 *)&command, ++ sizeof(command), NGBE_HI_COMMAND_TIMEOUT, false); ++ if (err) ++ return err; ++ ++ while (i < (len >> 2)) { ++ value = rd32a(hw, NGBE_MNGMBX, FW_PCIE_BUSMASTER_OFFSET + i); ++ ((u32 *)buf)[i] = value; ++ i++; ++ } ++ ++ return 0; ++} ++ ++s32 ngbe_hic_pcie_write(struct ngbe_hw *hw, u16 addr, u32 *buf, int len) ++{ ++ struct ngbe_hic_write_pcie command; ++ u32 value = 0; ++ int err, i = 0; ++ ++ while (i < (len >> 2)) { ++ value = ((u32 *)buf)[i]; ++ i++; ++ } ++ ++ memset(&command, 0, sizeof(command)); ++ command.hdr.cmd = FW_PCIE_WRITE_CMD; ++ command.hdr.buf_len = sizeof(command) - sizeof(command.hdr); ++ command.hdr.checksum = FW_DEFAULT_CHECKSUM; ++ command.lan_id = hw->bus.lan_id; ++ command.addr = addr; ++ command.data = value; ++ ++ err = ngbe_host_interface_command(hw, (u32 *)&command, ++ sizeof(command), NGBE_HI_COMMAND_TIMEOUT, false); ++ if (err) ++ return err; ++ ++ return 0; ++} ++ + s32 ngbe_hic_check_cap(struct ngbe_hw *hw) + { + struct ngbe_hic_read_shadow_ram command; + s32 err; + int i; + +- DEBUGFUNC("\n"); +- + command.hdr.req.cmd = FW_EEPROM_CHECK_STATUS; + command.hdr.req.buf_lenh = 0; + command.hdr.req.buf_lenl = 0; +diff --git a/dpdk/drivers/net/ngbe/base/ngbe_mng.h b/dpdk/drivers/net/ngbe/base/ngbe_mng.h +index e3d0309cbc..321338a051 100644 +--- a/dpdk/drivers/net/ngbe/base/ngbe_mng.h ++++ b/dpdk/drivers/net/ngbe/base/ngbe_mng.h +@@ -20,6 +20,9 @@ + #define FW_READ_SHADOW_RAM_LEN 0x6 + #define FW_WRITE_SHADOW_RAM_CMD 0x33 + #define FW_WRITE_SHADOW_RAM_LEN 0xA /* 8 plus 1 WORD to write */ ++#define FW_PCIE_READ_CMD 0xEC ++#define FW_PCIE_WRITE_CMD 0xED ++#define FW_PCIE_BUSMASTER_OFFSET 2 + #define FW_DEFAULT_CHECKSUM 0xFF /* checksum always 0xFF */ + #define FW_NVM_DATA_OFFSET 3 + #define FW_EEPROM_CHECK_STATUS 0xE9 +@@ -76,8 +79,26 @@ struct ngbe_hic_write_shadow_ram { + u16 pad3; + }; + ++struct ngbe_hic_read_pcie { ++ struct ngbe_hic_hdr hdr; ++ u8 lan_id; ++ u8 rsvd; ++ u16 addr; ++ u32 data; ++}; ++ ++struct ngbe_hic_write_pcie { ++ struct ngbe_hic_hdr hdr; ++ u8 lan_id; ++ u8 rsvd; ++ u16 addr; ++ u32 data; ++}; ++ + s32 ngbe_hic_sr_read(struct ngbe_hw *hw, u32 addr, u8 *buf, int len); + s32 ngbe_hic_sr_write(struct ngbe_hw *hw, u32 addr, u8 *buf, int len); ++s32 ngbe_hic_pcie_read(struct ngbe_hw *hw, u16 addr, u32 *buf, int len); ++s32 ngbe_hic_pcie_write(struct ngbe_hw *hw, u16 addr, u32 *buf, int len); + + s32 ngbe_hic_check_cap(struct ngbe_hw *hw); + #endif /* _NGBE_MNG_H_ */ +diff --git a/dpdk/drivers/net/ngbe/base/ngbe_osdep.h b/dpdk/drivers/net/ngbe/base/ngbe_osdep.h +index b62d793191..bf1fa30312 100644 +--- a/dpdk/drivers/net/ngbe/base/ngbe_osdep.h ++++ b/dpdk/drivers/net/ngbe/base/ngbe_osdep.h +@@ -19,6 +19,7 @@ + #include + #include + #include ++#include + + #include "../ngbe_logs.h" + +@@ -180,4 +181,7 @@ static inline u64 REVERT_BIT_MASK64(u64 mask) + #define ETH_P_8021Q 0x8100 + #define ETH_P_8021AD 0x88A8 + ++#define PCI_COMMAND 0x04 ++#define PCI_COMMAND_MASTER 0x4 ++ + #endif /* _NGBE_OS_H_ */ +diff --git a/dpdk/drivers/net/ngbe/base/ngbe_phy.c b/dpdk/drivers/net/ngbe/base/ngbe_phy.c +index 51b0a2ec60..e2b6b05ddf 100644 +--- a/dpdk/drivers/net/ngbe/base/ngbe_phy.c ++++ b/dpdk/drivers/net/ngbe/base/ngbe_phy.c +@@ -46,7 +46,7 @@ s32 ngbe_mdi_map_register(mdi_reg_t *reg, mdi_reg_22_t *reg22) + static bool ngbe_probe_phy(struct ngbe_hw *hw, u16 phy_addr) + { + if (!ngbe_validate_phy_addr(hw, phy_addr)) { +- DEBUGOUT("Unable to validate PHY address 0x%04X\n", ++ DEBUGOUT("Unable to validate PHY address 0x%04X", + phy_addr); + return false; + } +@@ -72,8 +72,6 @@ s32 ngbe_identify_phy(struct ngbe_hw *hw) + s32 err = NGBE_ERR_PHY_ADDR_INVALID; + u16 phy_addr; + +- DEBUGFUNC("ngbe_identify_phy"); +- + if (hw->phy.type != ngbe_phy_unknown) + return 0; + +@@ -103,11 +101,9 @@ s32 ngbe_check_reset_blocked(struct ngbe_hw *hw) + { + u32 mmngc; + +- DEBUGFUNC("ngbe_check_reset_blocked"); +- + mmngc = rd32(hw, NGBE_STAT); + if (mmngc & NGBE_STAT_MNGVETO) { +- DEBUGOUT("MNG_VETO bit detected.\n"); ++ DEBUGOUT("MNG_VETO bit detected."); + return true; + } + +@@ -125,19 +121,14 @@ bool ngbe_validate_phy_addr(struct ngbe_hw *hw, u32 phy_addr) + u16 phy_id = 0; + bool valid = false; + +- DEBUGFUNC("ngbe_validate_phy_addr"); +- +- if (hw->sub_device_id == NGBE_SUB_DEV_ID_EM_YT8521S_SFP) +- return true; +- + hw->phy.addr = phy_addr; +- hw->phy.read_reg(hw, NGBE_MD_PHY_ID_HIGH, ++ hw->phy.read_reg(hw, NGBE_MD_PHY_ID_LOW, + NGBE_MD_DEV_PMA_PMD, &phy_id); + + if (phy_id != 0xFFFF && phy_id != 0x0) + valid = true; + +- DEBUGOUT("PHY ID HIGH is 0x%04X\n", phy_id); ++ DEBUGOUT("PHY ID LOW is 0x%04X", phy_id); + + return valid; + } +@@ -153,8 +144,6 @@ s32 ngbe_get_phy_id(struct ngbe_hw *hw) + u16 phy_id_high = 0; + u16 phy_id_low = 0; + +- DEBUGFUNC("ngbe_get_phy_id"); +- + err = hw->phy.read_reg(hw, NGBE_MD_PHY_ID_HIGH, + NGBE_MD_DEV_PMA_PMD, + &phy_id_high); +@@ -166,7 +155,7 @@ s32 ngbe_get_phy_id(struct ngbe_hw *hw) + hw->phy.id |= (u32)(phy_id_low & NGBE_PHY_REVISION_MASK); + hw->phy.revision = (u32)(phy_id_low & ~NGBE_PHY_REVISION_MASK); + +- DEBUGOUT("PHY_ID_HIGH 0x%04X, PHY_ID_LOW 0x%04X\n", ++ DEBUGOUT("PHY_ID_HIGH 0x%04X, PHY_ID_LOW 0x%04X", + phy_id_high, phy_id_low); + + return err; +@@ -181,8 +170,6 @@ enum ngbe_phy_type ngbe_get_phy_type_from_id(struct ngbe_hw *hw) + { + enum ngbe_phy_type phy_type; + +- DEBUGFUNC("ngbe_get_phy_type_from_id"); +- + switch (hw->phy.id) { + case NGBE_PHYID_RTL: + phy_type = ngbe_phy_rtl; +@@ -215,8 +202,6 @@ s32 ngbe_reset_phy(struct ngbe_hw *hw) + { + s32 err = 0; + +- DEBUGFUNC("ngbe_reset_phy"); +- + if (hw->phy.type == ngbe_phy_unknown) + err = ngbe_identify_phy(hw); + +@@ -281,7 +266,7 @@ s32 ngbe_read_phy_reg_mdi(struct ngbe_hw *hw, u32 reg_addr, u32 device_type, + */ + if (!po32m(hw, NGBE_MDIOSCD, NGBE_MDIOSCD_BUSY, + 0, NULL, 100, 100)) { +- DEBUGOUT("PHY address command did not complete\n"); ++ DEBUGOUT("PHY address command did not complete"); + return NGBE_ERR_PHY; + } + +@@ -305,8 +290,6 @@ s32 ngbe_read_phy_reg(struct ngbe_hw *hw, u32 reg_addr, + s32 err; + u32 gssr = hw->phy.phy_semaphore_mask; + +- DEBUGFUNC("ngbe_read_phy_reg"); +- + if (hw->mac.acquire_swfw_sync(hw, gssr)) + return NGBE_ERR_SWFW_SYNC; + +@@ -346,7 +329,7 @@ s32 ngbe_write_phy_reg_mdi(struct ngbe_hw *hw, u32 reg_addr, + /* wait for completion */ + if (!po32m(hw, NGBE_MDIOSCD, NGBE_MDIOSCD_BUSY, + 0, NULL, 100, 100)) { +- TLOG_DEBUG("PHY write cmd didn't complete\n"); ++ DEBUGOUT("PHY write cmd didn't complete"); + return NGBE_ERR_PHY; + } + +@@ -367,8 +350,6 @@ s32 ngbe_write_phy_reg(struct ngbe_hw *hw, u32 reg_addr, + s32 err; + u32 gssr = hw->phy.phy_semaphore_mask; + +- DEBUGFUNC("ngbe_write_phy_reg"); +- + if (hw->mac.acquire_swfw_sync(hw, gssr)) + err = NGBE_ERR_SWFW_SYNC; + +@@ -394,8 +375,6 @@ s32 ngbe_init_phy(struct ngbe_hw *hw) + struct ngbe_phy_info *phy = &hw->phy; + s32 err = 0; + +- DEBUGFUNC("ngbe_init_phy"); +- + hw->phy.addr = 0; + + switch (hw->sub_device_id) { +diff --git a/dpdk/drivers/net/ngbe/base/ngbe_phy_mvl.c b/dpdk/drivers/net/ngbe/base/ngbe_phy_mvl.c +index 2eb351d258..b1da0e3167 100644 +--- a/dpdk/drivers/net/ngbe/base/ngbe_phy_mvl.c ++++ b/dpdk/drivers/net/ngbe/base/ngbe_phy_mvl.c +@@ -54,8 +54,6 @@ s32 ngbe_init_phy_mvl(struct ngbe_hw *hw) + u16 value = 0; + int i; + +- DEBUGFUNC("ngbe_init_phy_mvl"); +- + /* enable interrupts, only link status change and an done is allowed */ + ngbe_write_phy_reg_mdi(hw, MVL_PAGE_SEL, 0, 2); + ngbe_read_phy_reg_mdi(hw, MVL_RGM_CTL2, 0, &value); +@@ -73,7 +71,7 @@ s32 ngbe_init_phy_mvl(struct ngbe_hw *hw) + } + + if (i == 15) { +- DEBUGOUT("phy reset exceeds maximum waiting period.\n"); ++ DEBUGOUT("phy reset exceeds maximum waiting period."); + return NGBE_ERR_TIMEOUT; + } + +@@ -119,7 +117,6 @@ s32 ngbe_setup_phy_link_mvl(struct ngbe_hw *hw, u32 speed, + u16 value_r9 = 0; + u16 value; + +- DEBUGFUNC("ngbe_setup_phy_link_mvl"); + UNREFERENCED_PARAMETER(autoneg_wait_to_complete); + + hw->phy.autoneg_advertised = 0; +@@ -176,8 +173,6 @@ s32 ngbe_reset_phy_mvl(struct ngbe_hw *hw) + u16 ctrl = 0; + s32 status = 0; + +- DEBUGFUNC("ngbe_reset_phy_mvl"); +- + if (hw->phy.type != ngbe_phy_mvl && hw->phy.type != ngbe_phy_mvl_sfi) + return NGBE_ERR_PHY_TYPE; + +@@ -202,7 +197,7 @@ s32 ngbe_reset_phy_mvl(struct ngbe_hw *hw) + } + + if (i == MVL_PHY_RST_WAIT_PERIOD) { +- DEBUGOUT("PHY reset polling failed to complete.\n"); ++ DEBUGOUT("PHY reset polling failed to complete."); + return NGBE_ERR_RESET_FAILED; + } + +@@ -250,8 +245,6 @@ s32 ngbe_set_phy_pause_adv_mvl(struct ngbe_hw *hw, u16 pause_bit) + u16 value; + s32 status = 0; + +- DEBUGFUNC("ngbe_set_phy_pause_adv_mvl"); +- + if (hw->phy.type == ngbe_phy_mvl) { + status = hw->phy.read_reg(hw, MVL_ANA, 0, &value); + value &= ~(MVL_CANA_ASM_PAUSE | MVL_CANA_PAUSE); +@@ -275,8 +268,6 @@ s32 ngbe_check_phy_link_mvl(struct ngbe_hw *hw, + u16 phy_data = 0; + u16 insr = 0; + +- DEBUGFUNC("ngbe_check_phy_link_mvl"); +- + /* Initialize speed and link to default case */ + *link_up = false; + *speed = NGBE_LINK_SPEED_UNKNOWN; +diff --git a/dpdk/drivers/net/ngbe/base/ngbe_phy_rtl.c b/dpdk/drivers/net/ngbe/base/ngbe_phy_rtl.c +index 7b08b7a46c..a95efcbab6 100644 +--- a/dpdk/drivers/net/ngbe/base/ngbe_phy_rtl.c ++++ b/dpdk/drivers/net/ngbe/base/ngbe_phy_rtl.c +@@ -57,7 +57,7 @@ s32 ngbe_init_phy_rtl(struct ngbe_hw *hw) + msec_delay(10); + } + if (i == 15) { +- DEBUGOUT("GPhy reset exceeds maximum times.\n"); ++ DEBUGOUT("GPhy reset exceeds maximum times."); + return NGBE_ERR_PHY_TIMEOUT; + } + +@@ -108,8 +108,6 @@ s32 ngbe_setup_phy_link_rtl(struct ngbe_hw *hw, + u16 autoneg_reg = NGBE_MII_AUTONEG_REG; + u16 value = 0; + +- DEBUGFUNC("ngbe_setup_phy_link_rtl"); +- + UNREFERENCED_PARAMETER(autoneg_wait_to_complete); + + hw->phy.read_reg(hw, RTL_INSR, 0xa43, &autoneg_reg); +@@ -129,7 +127,7 @@ s32 ngbe_setup_phy_link_rtl(struct ngbe_hw *hw, + break; + default: + value = RTL_BMCR_SPEED_SELECT1 | RTL_BMCR_SPEED_SELECT0; +- DEBUGOUT("unknown speed = 0x%x.\n", speed); ++ DEBUGOUT("unknown speed = 0x%x.", speed); + break; + } + /* duplex full */ +@@ -229,8 +227,6 @@ s32 ngbe_reset_phy_rtl(struct ngbe_hw *hw) + u16 value = 0, i; + s32 status = 0; + +- DEBUGFUNC("ngbe_reset_phy_rtl"); +- + value |= RTL_BMCR_RESET; + status = hw->phy.write_reg(hw, RTL_BMCR, RTL_DEV_ZERO, value); + +@@ -299,8 +295,6 @@ s32 ngbe_check_phy_link_rtl(struct ngbe_hw *hw, u32 *speed, bool *link_up) + u16 phy_data = 0; + u16 insr = 0; + +- DEBUGFUNC("ngbe_check_phy_link_rtl"); +- + hw->phy.read_reg(hw, RTL_INSR, 0xa43, &insr); + + /* Initialize speed and link to default case */ +diff --git a/dpdk/drivers/net/ngbe/base/ngbe_phy_yt.c b/dpdk/drivers/net/ngbe/base/ngbe_phy_yt.c +index 8db0f9ce48..40f1725f61 100644 +--- a/dpdk/drivers/net/ngbe/base/ngbe_phy_yt.c ++++ b/dpdk/drivers/net/ngbe/base/ngbe_phy_yt.c +@@ -102,8 +102,6 @@ s32 ngbe_init_phy_yt(struct ngbe_hw *hw) + { + u16 value = 0; + +- DEBUGFUNC("ngbe_init_phy_yt"); +- + if (hw->phy.type != ngbe_phy_yt8521s_sfi) + return 0; + +@@ -131,7 +129,6 @@ s32 ngbe_setup_phy_link_yt(struct ngbe_hw *hw, u32 speed, + u16 value_r9 = 0; + u16 value; + +- DEBUGFUNC("ngbe_setup_phy_link_yt"); + UNREFERENCED_PARAMETER(autoneg_wait_to_complete); + + hw->phy.autoneg_advertised = 0; +@@ -208,8 +205,6 @@ s32 ngbe_reset_phy_yt(struct ngbe_hw *hw) + u16 ctrl = 0; + s32 status = 0; + +- DEBUGFUNC("ngbe_reset_phy_yt"); +- + if (hw->phy.type != ngbe_phy_yt8521s && + hw->phy.type != ngbe_phy_yt8521s_sfi) + return NGBE_ERR_PHY_TYPE; +@@ -227,7 +222,7 @@ s32 ngbe_reset_phy_yt(struct ngbe_hw *hw) + } + + if (i == YT_PHY_RST_WAIT_PERIOD) { +- DEBUGOUT("PHY reset polling failed to complete.\n"); ++ DEBUGOUT("PHY reset polling failed to complete."); + return NGBE_ERR_RESET_FAILED; + } + +@@ -239,8 +234,6 @@ s32 ngbe_get_phy_advertised_pause_yt(struct ngbe_hw *hw, u8 *pause_bit) + u16 value; + s32 status = 0; + +- DEBUGFUNC("ngbe_get_phy_advertised_pause_yt"); +- + status = hw->phy.read_reg(hw, YT_ANA, 0, &value); + value &= YT_FANA_PAUSE_MASK; + *pause_bit = (u8)(value >> 7); +@@ -253,8 +246,6 @@ s32 ngbe_get_phy_lp_advertised_pause_yt(struct ngbe_hw *hw, u8 *pause_bit) + u16 value; + s32 status = 0; + +- DEBUGFUNC("ngbe_get_phy_lp_advertised_pause_yt"); +- + status = hw->phy.read_reg(hw, YT_LPAR, 0, &value); + value &= YT_FLPAR_PAUSE_MASK; + *pause_bit = (u8)(value >> 7); +@@ -267,9 +258,6 @@ s32 ngbe_set_phy_pause_adv_yt(struct ngbe_hw *hw, u16 pause_bit) + u16 value; + s32 status = 0; + +- DEBUGFUNC("ngbe_set_phy_pause_adv_yt"); +- +- + status = hw->phy.read_reg(hw, YT_ANA, 0, &value); + value &= ~YT_FANA_PAUSE_MASK; + value |= pause_bit; +@@ -287,8 +275,6 @@ s32 ngbe_check_phy_link_yt(struct ngbe_hw *hw, + u16 phy_data = 0; + u16 insr = 0; + +- DEBUGFUNC("ngbe_check_phy_link_yt"); +- + /* Initialize speed and link to default case */ + *link_up = false; + *speed = NGBE_LINK_SPEED_UNKNOWN; +diff --git a/dpdk/drivers/net/ngbe/base/ngbe_regs.h b/dpdk/drivers/net/ngbe/base/ngbe_regs.h +index 872b008c46..640e385990 100644 +--- a/dpdk/drivers/net/ngbe/base/ngbe_regs.h ++++ b/dpdk/drivers/net/ngbe/base/ngbe_regs.h +@@ -785,30 +785,30 @@ enum ngbe_5tuple_protocol { + #define NGBE_MACRXERRCRCH 0x01192C + #define NGBE_MACRXERRLENL 0x011978 + #define NGBE_MACRXERRLENH 0x01197C +-#define NGBE_MACRX1TO64L 0x001940 +-#define NGBE_MACRX1TO64H 0x001944 +-#define NGBE_MACRX65TO127L 0x001948 +-#define NGBE_MACRX65TO127H 0x00194C +-#define NGBE_MACRX128TO255L 0x001950 +-#define NGBE_MACRX128TO255H 0x001954 +-#define NGBE_MACRX256TO511L 0x001958 +-#define NGBE_MACRX256TO511H 0x00195C +-#define NGBE_MACRX512TO1023L 0x001960 +-#define NGBE_MACRX512TO1023H 0x001964 +-#define NGBE_MACRX1024TOMAXL 0x001968 +-#define NGBE_MACRX1024TOMAXH 0x00196C +-#define NGBE_MACTX1TO64L 0x001834 +-#define NGBE_MACTX1TO64H 0x001838 +-#define NGBE_MACTX65TO127L 0x00183C +-#define NGBE_MACTX65TO127H 0x001840 +-#define NGBE_MACTX128TO255L 0x001844 +-#define NGBE_MACTX128TO255H 0x001848 +-#define NGBE_MACTX256TO511L 0x00184C +-#define NGBE_MACTX256TO511H 0x001850 +-#define NGBE_MACTX512TO1023L 0x001854 +-#define NGBE_MACTX512TO1023H 0x001858 +-#define NGBE_MACTX1024TOMAXL 0x00185C +-#define NGBE_MACTX1024TOMAXH 0x001860 ++#define NGBE_MACRX1TO64L 0x011940 ++#define NGBE_MACRX1TO64H 0x011944 ++#define NGBE_MACRX65TO127L 0x011948 ++#define NGBE_MACRX65TO127H 0x01194C ++#define NGBE_MACRX128TO255L 0x011950 ++#define NGBE_MACRX128TO255H 0x011954 ++#define NGBE_MACRX256TO511L 0x011958 ++#define NGBE_MACRX256TO511H 0x01195C ++#define NGBE_MACRX512TO1023L 0x011960 ++#define NGBE_MACRX512TO1023H 0x011964 ++#define NGBE_MACRX1024TOMAXL 0x011968 ++#define NGBE_MACRX1024TOMAXH 0x01196C ++#define NGBE_MACTX1TO64L 0x011834 ++#define NGBE_MACTX1TO64H 0x011838 ++#define NGBE_MACTX65TO127L 0x01183C ++#define NGBE_MACTX65TO127H 0x011840 ++#define NGBE_MACTX128TO255L 0x011844 ++#define NGBE_MACTX128TO255H 0x011848 ++#define NGBE_MACTX256TO511L 0x01184C ++#define NGBE_MACTX256TO511H 0x011850 ++#define NGBE_MACTX512TO1023L 0x011854 ++#define NGBE_MACTX512TO1023H 0x011858 ++#define NGBE_MACTX1024TOMAXL 0x01185C ++#define NGBE_MACTX1024TOMAXH 0x011860 + + #define NGBE_MACRXUNDERSIZE 0x011938 + #define NGBE_MACRXOVERSIZE 0x01193C +@@ -866,6 +866,9 @@ enum ngbe_5tuple_protocol { + * PF(Physical Function) Registers + ******************************************************************************/ + /* Interrupt */ ++#define NGBE_BMECTL 0x012020 ++#define NGBE_BMECTL_VFDRP MS(1, 0x1) ++#define NGBE_BMECTL_PFDRP MS(0, 0x1) + #define NGBE_ICRMISC 0x000100 + #define NGBE_ICRMISC_MASK MS(8, 0xFFFFFF) + #define NGBE_ICRMISC_RST MS(10, 0x1) /* device reset event */ +@@ -1419,8 +1422,13 @@ po32m(struct ngbe_hw *hw, u32 reg, u32 mask, u32 expect, u32 *actual, + } + + do { +- all |= rd32(hw, reg); +- value |= mask & all; ++ if (expect != 0) { ++ all |= rd32(hw, reg); ++ value |= mask & all; ++ } else { ++ all = rd32(hw, reg); ++ value = mask & all; ++ } + if (value == expect) + break; + +diff --git a/dpdk/drivers/net/ngbe/base/ngbe_type.h b/dpdk/drivers/net/ngbe/base/ngbe_type.h +index 12847b7272..4c995e7397 100644 +--- a/dpdk/drivers/net/ngbe/base/ngbe_type.h ++++ b/dpdk/drivers/net/ngbe/base/ngbe_type.h +@@ -11,9 +11,15 @@ + #define NGBE_FRAME_SIZE_MAX (9728) /* Maximum frame size, +FCS */ + #define NGBE_FRAME_SIZE_DFT (1522) /* Default frame size, +FCS */ + #define NGBE_NUM_POOL (32) ++#define NGBE_PBRXSIZE_MAX 0x00080000 /* 512KB Packet Buffer */ ++#define NGBE_PBTXSIZE_MAX 0x00005000 /* 20KB Packet Buffer */ ++#define NGBE_TXPKT_SIZE_MAX 0xA /* Max Tx Packet size */ + #define NGBE_MAX_QP (8) + #define NGBE_MAX_UTA 128 + ++#define NGBE_PCI_MASTER_DISABLE_TIMEOUT 800 ++ ++ + #define NGBE_ALIGN 128 /* as intel did */ + #define NGBE_ISB_SIZE 16 + +@@ -269,6 +275,9 @@ struct ngbe_mac_info { + s32 (*get_link_capabilities)(struct ngbe_hw *hw, + u32 *speed, bool *autoneg); + ++ /* Packet Buffer manipulation */ ++ void (*setup_pba)(struct ngbe_hw *hw); ++ + /* LED */ + s32 (*led_on)(struct ngbe_hw *hw, u32 index); + s32 (*led_off)(struct ngbe_hw *hw, u32 index); +@@ -311,6 +320,7 @@ struct ngbe_mac_info { + u32 mcft_size; + u32 vft_size; + u32 num_rar_entries; ++ u32 rx_pb_size; + u32 max_tx_queues; + u32 max_rx_queues; + bool get_link_status; +diff --git a/dpdk/drivers/net/ngbe/ngbe_ethdev.c b/dpdk/drivers/net/ngbe/ngbe_ethdev.c +index 981592f7f4..b930326379 100644 +--- a/dpdk/drivers/net/ngbe/ngbe_ethdev.c ++++ b/dpdk/drivers/net/ngbe/ngbe_ethdev.c +@@ -89,7 +89,6 @@ static int ngbe_dev_macsec_interrupt_setup(struct rte_eth_dev *dev); + static int ngbe_dev_misc_interrupt_setup(struct rte_eth_dev *dev); + static int ngbe_dev_rxq_interrupt_setup(struct rte_eth_dev *dev); + static void ngbe_dev_interrupt_handler(void *param); +-static void ngbe_dev_interrupt_delayed_handler(void *param); + static void ngbe_configure_msix(struct rte_eth_dev *dev); + + #define NGBE_SET_HWSTRIP(h, q) do {\ +@@ -165,6 +164,8 @@ static const struct rte_ngbe_xstats_name_off rte_ngbe_stats_strings[] = { + HW_XSTAT(rx_management_packets), + HW_XSTAT(tx_management_packets), + HW_XSTAT(rx_management_dropped), ++ HW_XSTAT(rx_dma_drop), ++ HW_XSTAT(tx_secdrp_packets), + + /* Basic Error */ + HW_XSTAT(rx_crc_errors), +@@ -180,6 +181,12 @@ static const struct rte_ngbe_xstats_name_off rte_ngbe_stats_strings[] = { + HW_XSTAT(mac_local_errors), + HW_XSTAT(mac_remote_errors), + ++ /* PB Stats */ ++ HW_XSTAT(rx_up_dropped), ++ HW_XSTAT(rdb_pkt_cnt), ++ HW_XSTAT(rdb_repli_cnt), ++ HW_XSTAT(rdb_drp_cnt), ++ + /* MACSEC */ + HW_XSTAT(tx_macsec_pkts_untagged), + HW_XSTAT(tx_macsec_pkts_encrypted), +@@ -356,6 +363,7 @@ eth_ngbe_dev_init(struct rte_eth_dev *eth_dev, void *init_params __rte_unused) + eth_dev->data->dev_flags |= RTE_ETH_DEV_AUTOFILL_QUEUE_XSTATS; + + /* Vendor and Device ID need to be set before init of shared code */ ++ hw->back = pci_dev; + hw->device_id = pci_dev->id.device_id; + hw->vendor_id = pci_dev->id.vendor_id; + hw->sub_system_id = pci_dev->id.subsystem_device_id; +@@ -943,12 +951,14 @@ ngbe_dev_start(struct rte_eth_dev *dev) + + PMD_INIT_FUNC_TRACE(); + ++ /* Stop the link setup handler before resetting the HW. */ ++ rte_eal_alarm_cancel(ngbe_dev_setup_link_alarm_handler, dev); ++ + /* disable uio/vfio intr/eventfd mapping */ + rte_intr_disable(intr_handle); + + /* stop adapter */ + hw->adapter_stopped = 0; +- ngbe_stop_hw(hw); + + /* reinitialize adapter, this calls reset and start */ + hw->nb_rx_queues = dev->data->nb_rx_queues; +@@ -959,6 +969,8 @@ ngbe_dev_start(struct rte_eth_dev *dev) + hw->mac.start_hw(hw); + hw->mac.get_link_status = true; + ++ ngbe_set_pcie_master(hw, true); ++ + /* configure PF module if SRIOV enabled */ + ngbe_pf_host_configure(dev); + +@@ -983,7 +995,7 @@ ngbe_dev_start(struct rte_eth_dev *dev) + } + } + +- /* confiugre MSI-X for sleep until Rx interrupt */ ++ /* configure MSI-X for sleep until Rx interrupt */ + ngbe_configure_msix(dev); + + /* initialize transmission unit */ +@@ -1004,6 +1016,7 @@ ngbe_dev_start(struct rte_eth_dev *dev) + goto error; + } + ++ hw->mac.setup_pba(hw); + ngbe_configure_port(dev); + + err = ngbe_dev_rxtx_start(dev); +@@ -1037,7 +1050,7 @@ ngbe_dev_start(struct rte_eth_dev *dev) + if (hw->mac.default_speeds & NGBE_LINK_SPEED_10M_FULL) + allowed_speeds |= RTE_ETH_LINK_SPEED_10M; + +- if (*link_speeds & ~allowed_speeds) { ++ if (((*link_speeds) >> 1) & ~(allowed_speeds >> 1)) { + PMD_INIT_LOG(ERR, "Invalid link setting"); + goto error; + } +@@ -1131,6 +1144,8 @@ ngbe_dev_stop(struct rte_eth_dev *dev) + + PMD_INIT_FUNC_TRACE(); + ++ rte_eal_alarm_cancel(ngbe_dev_setup_link_alarm_handler, dev); ++ + if ((hw->sub_system_id & NGBE_OEM_MASK) == NGBE_LY_M88E1512_SFP || + (hw->sub_system_id & NGBE_OEM_MASK) == NGBE_LY_YT8521S_SFP) { + /* gpio0 is used to power on/off control*/ +@@ -1169,6 +1184,8 @@ ngbe_dev_stop(struct rte_eth_dev *dev) + rte_intr_efd_disable(intr_handle); + rte_intr_vec_list_free(intr_handle); + ++ ngbe_set_pcie_master(hw, true); ++ + adapter->rss_reta_updated = 0; + + hw->adapter_stopped = true; +@@ -1197,6 +1214,8 @@ ngbe_dev_close(struct rte_eth_dev *dev) + + ngbe_dev_free_queues(dev); + ++ ngbe_set_pcie_master(hw, false); ++ + /* reprogram the RAR[0] in case user changed it. */ + ngbe_set_rar(hw, 0, hw->mac.addr, 0, true); + +@@ -1800,6 +1819,24 @@ ngbe_dev_supported_ptypes_get(struct rte_eth_dev *dev) + return NULL; + } + ++void ++ngbe_dev_setup_link_alarm_handler(void *param) ++{ ++ struct rte_eth_dev *dev = (struct rte_eth_dev *)param; ++ struct ngbe_hw *hw = ngbe_dev_hw(dev); ++ struct ngbe_interrupt *intr = ngbe_dev_intr(dev); ++ u32 speed; ++ bool autoneg = false; ++ ++ speed = hw->phy.autoneg_advertised; ++ if (!speed) ++ hw->mac.get_link_capabilities(hw, &speed, &autoneg); ++ ++ hw->mac.setup_link(hw, speed, true); ++ ++ intr->flags &= ~NGBE_FLAG_NEED_LINK_CONFIG; ++} ++ + /* return 0 means link status changed, -1 means not changed */ + int + ngbe_dev_link_update_share(struct rte_eth_dev *dev, +@@ -1837,8 +1874,16 @@ ngbe_dev_link_update_share(struct rte_eth_dev *dev, + return rte_eth_linkstatus_set(dev, &link); + } + +- if (!link_up) ++ if (!link_up) { ++ if (hw->phy.media_type == ngbe_media_type_fiber && ++ hw->phy.type != ngbe_phy_mvl_sfi) { ++ intr->flags |= NGBE_FLAG_NEED_LINK_CONFIG; ++ rte_eal_alarm_set(10, ++ ngbe_dev_setup_link_alarm_handler, dev); ++ } ++ + return rte_eth_linkstatus_set(dev, &link); ++ } + + intr->flags &= ~NGBE_FLAG_NEED_LINK_CONFIG; + link.link_status = RTE_ETH_LINK_UP; +@@ -2061,9 +2106,6 @@ ngbe_dev_interrupt_get_status(struct rte_eth_dev *dev) + struct ngbe_hw *hw = ngbe_dev_hw(dev); + struct ngbe_interrupt *intr = ngbe_dev_intr(dev); + +- /* clear all cause mask */ +- ngbe_disable_intr(hw); +- + /* read-on-clear nic registers here */ + eicr = ((u32 *)hw->isb_mem)[NGBE_ISB_MISC]; + PMD_DRV_LOG(DEBUG, "eicr %x", eicr); +@@ -2083,6 +2125,8 @@ ngbe_dev_interrupt_get_status(struct rte_eth_dev *dev) + if (eicr & NGBE_ICRMISC_GPIO) + intr->flags |= NGBE_FLAG_NEED_LINK_UPDATE; + ++ ((u32 *)hw->isb_mem)[NGBE_ISB_MISC] = 0; ++ + return 0; + } + +@@ -2135,7 +2179,6 @@ static int + ngbe_dev_interrupt_action(struct rte_eth_dev *dev) + { + struct ngbe_interrupt *intr = ngbe_dev_intr(dev); +- int64_t timeout; + + PMD_DRV_LOG(DEBUG, "intr action type %d", intr->flags); + +@@ -2151,31 +2194,11 @@ ngbe_dev_interrupt_action(struct rte_eth_dev *dev) + rte_eth_linkstatus_get(dev, &link); + + ngbe_dev_link_update(dev, 0); +- +- /* likely to up */ +- if (link.link_status != RTE_ETH_LINK_UP) +- /* handle it 1 sec later, wait it being stable */ +- timeout = NGBE_LINK_UP_CHECK_TIMEOUT; +- /* likely to down */ +- else +- /* handle it 4 sec later, wait it being stable */ +- timeout = NGBE_LINK_DOWN_CHECK_TIMEOUT; +- ++ intr->flags &= ~NGBE_FLAG_NEED_LINK_UPDATE; + ngbe_dev_link_status_print(dev); +- if (rte_eal_alarm_set(timeout * 1000, +- ngbe_dev_interrupt_delayed_handler, +- (void *)dev) < 0) { +- PMD_DRV_LOG(ERR, "Error setting alarm"); +- } else { +- /* remember original mask */ +- intr->mask_misc_orig = intr->mask_misc; +- /* only disable lsc interrupt */ +- intr->mask_misc &= ~NGBE_ICRMISC_PHY; +- +- intr->mask_orig = intr->mask; +- /* only disable all misc interrupts */ +- intr->mask &= ~(1ULL << NGBE_MISC_VEC_ID); +- } ++ if (dev->data->dev_link.link_speed != link.link_speed) ++ rte_eth_dev_callback_process(dev, ++ RTE_ETH_EVENT_INTR_LSC, NULL); + } + + PMD_DRV_LOG(DEBUG, "enable intr immediately"); +@@ -2184,53 +2207,6 @@ ngbe_dev_interrupt_action(struct rte_eth_dev *dev) + return 0; + } + +-/** +- * Interrupt handler which shall be registered for alarm callback for delayed +- * handling specific interrupt to wait for the stable nic state. As the +- * NIC interrupt state is not stable for ngbe after link is just down, +- * it needs to wait 4 seconds to get the stable status. +- * +- * @param param +- * The address of parameter (struct rte_eth_dev *) registered before. +- */ +-static void +-ngbe_dev_interrupt_delayed_handler(void *param) +-{ +- struct rte_eth_dev *dev = (struct rte_eth_dev *)param; +- struct ngbe_interrupt *intr = ngbe_dev_intr(dev); +- struct ngbe_hw *hw = ngbe_dev_hw(dev); +- uint32_t eicr; +- +- ngbe_disable_intr(hw); +- +- eicr = ((u32 *)hw->isb_mem)[NGBE_ISB_MISC]; +- if (eicr & NGBE_ICRMISC_VFMBX) +- ngbe_pf_mbx_process(dev); +- +- if (intr->flags & NGBE_FLAG_NEED_LINK_UPDATE) { +- ngbe_dev_link_update(dev, 0); +- intr->flags &= ~NGBE_FLAG_NEED_LINK_UPDATE; +- ngbe_dev_link_status_print(dev); +- rte_eth_dev_callback_process(dev, RTE_ETH_EVENT_INTR_LSC, +- NULL); +- } +- +- if (intr->flags & NGBE_FLAG_MACSEC) { +- rte_eth_dev_callback_process(dev, RTE_ETH_EVENT_MACSEC, +- NULL); +- intr->flags &= ~NGBE_FLAG_MACSEC; +- } +- +- /* restore original mask */ +- intr->mask_misc = intr->mask_misc_orig; +- intr->mask_misc_orig = 0; +- intr->mask = intr->mask_orig; +- intr->mask_orig = 0; +- +- PMD_DRV_LOG(DEBUG, "enable intr in delayed handler S[%08x]", eicr); +- ngbe_enable_intr(dev); +-} +- + /** + * Interrupt handler triggered by NIC for handling + * specific interrupt. +@@ -2641,7 +2617,7 @@ ngbe_set_ivar_map(struct ngbe_hw *hw, int8_t direction, + wr32(hw, NGBE_IVARMISC, tmp); + } else { + /* rx or tx causes */ +- /* Workround for ICR lost */ ++ /* Workaround for ICR lost */ + idx = ((16 * (queue & 1)) + (8 * direction)); + tmp = rd32(hw, NGBE_IVAR(queue >> 1)); + tmp &= ~(0xFF << idx); +@@ -2893,7 +2869,7 @@ ngbe_timesync_disable(struct rte_eth_dev *dev) + /* Disable L2 filtering of IEEE1588/802.1AS Ethernet frame types. */ + wr32(hw, NGBE_ETFLT(NGBE_ETF_ID_1588), 0); + +- /* Stop incrementating the System Time registers. */ ++ /* Stop incrementing the System Time registers. */ + wr32(hw, NGBE_TSTIMEINC, 0); + + return 0; +diff --git a/dpdk/drivers/net/ngbe/ngbe_ethdev.h b/dpdk/drivers/net/ngbe/ngbe_ethdev.h +index bb96f6a5e7..8d500fd38c 100644 +--- a/dpdk/drivers/net/ngbe/ngbe_ethdev.h ++++ b/dpdk/drivers/net/ngbe/ngbe_ethdev.h +@@ -341,6 +341,7 @@ void ngbe_vlan_hw_strip_bitmap_set(struct rte_eth_dev *dev, + uint16_t queue, bool on); + void ngbe_config_vlan_strip_on_all_queues(struct rte_eth_dev *dev, + int mask); ++void ngbe_dev_setup_link_alarm_handler(void *param); + void ngbe_read_stats_registers(struct ngbe_hw *hw, + struct ngbe_hw_stats *hw_stats); + +diff --git a/dpdk/drivers/net/ngbe/ngbe_logs.h b/dpdk/drivers/net/ngbe/ngbe_logs.h +index fd306419e6..e5165ffd60 100644 +--- a/dpdk/drivers/net/ngbe/ngbe_logs.h ++++ b/dpdk/drivers/net/ngbe/ngbe_logs.h +@@ -37,10 +37,7 @@ extern int ngbe_logtype_tx; + #define PMD_TX_LOG(level, fmt, args...) do { } while (0) + #endif + +-#define TLOG_DEBUG(fmt, args...) PMD_DRV_LOG(DEBUG, fmt, ##args) +- +-#define DEBUGOUT(fmt, args...) TLOG_DEBUG(fmt, ##args) +-#define PMD_INIT_FUNC_TRACE() TLOG_DEBUG(" >>") +-#define DEBUGFUNC(fmt) TLOG_DEBUG(fmt) ++#define DEBUGOUT(fmt, args...) PMD_DRV_LOG(DEBUG, fmt, ##args) ++#define PMD_INIT_FUNC_TRACE() PMD_DRV_LOG(DEBUG, ">>") + + #endif /* _NGBE_LOGS_H_ */ +diff --git a/dpdk/drivers/net/ngbe/ngbe_pf.c b/dpdk/drivers/net/ngbe/ngbe_pf.c +index 7f9c04fb0e..12a18de31d 100644 +--- a/dpdk/drivers/net/ngbe/ngbe_pf.c ++++ b/dpdk/drivers/net/ngbe/ngbe_pf.c +@@ -163,7 +163,7 @@ int ngbe_pf_host_configure(struct rte_eth_dev *eth_dev) + + wr32(hw, NGBE_PSRCTL, NGBE_PSRCTL_LBENA); + +- /* clear VMDq map to perment rar 0 */ ++ /* clear VMDq map to permanent rar 0 */ + hw->mac.clear_vmdq(hw, 0, BIT_MASK32); + + /* clear VMDq map to scan rar 31 */ +diff --git a/dpdk/drivers/net/octeontx/octeontx_ethdev.c b/dpdk/drivers/net/octeontx/octeontx_ethdev.c +index 4f1e368c61..f879a0c9fc 100644 +--- a/dpdk/drivers/net/octeontx/octeontx_ethdev.c ++++ b/dpdk/drivers/net/octeontx/octeontx_ethdev.c +@@ -26,6 +26,11 @@ + #include "octeontx_rxtx.h" + #include "octeontx_logs.h" + ++/* Useful in stopping/closing event device if no of ++ * eth ports are using it. ++ */ ++uint16_t evdev_refcnt; ++ + struct evdev_priv_data { + OFFLOAD_FLAGS; /*Sequence should not be changed */ + } __rte_cache_aligned; +@@ -491,7 +496,11 @@ octeontx_dev_close(struct rte_eth_dev *dev) + if (rte_eal_process_type() != RTE_PROC_PRIMARY) + return 0; + +- rte_event_dev_close(nic->evdev); ++ /* Stopping/closing event device once all eth ports are closed. */ ++ if (__atomic_sub_fetch(&evdev_refcnt, 1, __ATOMIC_ACQUIRE) == 0) { ++ rte_event_dev_stop(nic->evdev); ++ rte_event_dev_close(nic->evdev); ++ } + + octeontx_dev_flow_ctrl_fini(dev); + +@@ -671,8 +680,6 @@ octeontx_dev_stop(struct rte_eth_dev *dev) + + PMD_INIT_FUNC_TRACE(); + +- rte_event_dev_stop(nic->evdev); +- + ret = octeontx_port_stop(nic); + if (ret < 0) { + octeontx_log_err("failed to req stop port %d res=%d", +@@ -1090,7 +1097,7 @@ octeontx_dev_rx_queue_setup(struct rte_eth_dev *dev, uint16_t qidx, + + /* Verify queue index */ + if (qidx >= dev->data->nb_rx_queues) { +- octeontx_log_err("QID %d not supporteded (0 - %d available)\n", ++ octeontx_log_err("QID %d not supported (0 - %d available)\n", + qidx, (dev->data->nb_rx_queues - 1)); + return -ENOTSUP; + } +@@ -1334,6 +1341,7 @@ octeontx_create(struct rte_vdev_device *dev, int port, uint8_t evdev, + nic->pko_vfid = pko_vfid; + nic->port_id = port; + nic->evdev = evdev; ++ __atomic_add_fetch(&evdev_refcnt, 1, __ATOMIC_ACQUIRE); + + res = octeontx_port_open(nic); + if (res < 0) +@@ -1583,6 +1591,7 @@ octeontx_probe(struct rte_vdev_device *dev) + } + } + ++ __atomic_store_n(&evdev_refcnt, 0, __ATOMIC_RELEASE); + /* + * Do 1:1 links for ports & queues. All queues would be mapped to + * one port. If there are more ports than queues, then some ports +diff --git a/dpdk/drivers/net/octeontx2/otx2_ethdev_irq.c b/dpdk/drivers/net/octeontx2/otx2_ethdev_irq.c +index cc573bb2e8..f56d5b2a38 100644 +--- a/dpdk/drivers/net/octeontx2/otx2_ethdev_irq.c ++++ b/dpdk/drivers/net/octeontx2/otx2_ethdev_irq.c +@@ -369,7 +369,7 @@ oxt2_nix_register_cq_irqs(struct rte_eth_dev *eth_dev) + "rc=%d", rc); + return rc; + } +- /* VFIO vector zero is resereved for misc interrupt so ++ /* VFIO vector zero is reserved for misc interrupt so + * doing required adjustment. (b13bfab4cd) + */ + if (rte_intr_vec_list_index_set(handle, q, +diff --git a/dpdk/drivers/net/octeontx2/otx2_ptp.c b/dpdk/drivers/net/octeontx2/otx2_ptp.c +index abb2130587..974018f97e 100644 +--- a/dpdk/drivers/net/octeontx2/otx2_ptp.c ++++ b/dpdk/drivers/net/octeontx2/otx2_ptp.c +@@ -440,7 +440,7 @@ otx2_nix_read_clock(struct rte_eth_dev *eth_dev, uint64_t *clock) + /* This API returns the raw PTP HI clock value. Since LFs doesn't + * have direct access to PTP registers and it requires mbox msg + * to AF for this value. In fastpath reading this value for every +- * packet (which involes mbox call) becomes very expensive, hence ++ * packet (which involves mbox call) becomes very expensive, hence + * we should be able to derive PTP HI clock value from tsc by + * using freq_mult and clk_delta calculated during configure stage. + */ +diff --git a/dpdk/drivers/net/octeontx2/otx2_tx.h b/dpdk/drivers/net/octeontx2/otx2_tx.h +index 4bbd5a390f..a2fb7ce3cb 100644 +--- a/dpdk/drivers/net/octeontx2/otx2_tx.h ++++ b/dpdk/drivers/net/octeontx2/otx2_tx.h +@@ -61,7 +61,7 @@ otx2_nix_xmit_prepare_tstamp(uint64_t *cmd, const uint64_t *send_mem_desc, + /* Retrieving the default desc values */ + cmd[off] = send_mem_desc[6]; + +- /* Using compiler barier to avoid voilation of C ++ /* Using compiler barrier to avoid violation of C + * aliasing rules. + */ + rte_compiler_barrier(); +@@ -70,7 +70,7 @@ otx2_nix_xmit_prepare_tstamp(uint64_t *cmd, const uint64_t *send_mem_desc, + /* Packets for which RTE_MBUF_F_TX_IEEE1588_TMST is not set, tx tstamp + * should not be recorded, hence changing the alg type to + * NIX_SENDMEMALG_SET and also changing send mem addr field to +- * next 8 bytes as it corrpt the actual tx tstamp registered ++ * next 8 bytes as it corrupts the actual tx tstamp registered + * address. + */ + send_mem->alg = NIX_SENDMEMALG_SETTSTMP - (is_ol_tstamp); +diff --git a/dpdk/drivers/net/octeontx2/otx2_vlan.c b/dpdk/drivers/net/octeontx2/otx2_vlan.c +index cce643b7b5..359680de5c 100644 +--- a/dpdk/drivers/net/octeontx2/otx2_vlan.c ++++ b/dpdk/drivers/net/octeontx2/otx2_vlan.c +@@ -953,7 +953,7 @@ static void nix_vlan_reinstall_vlan_filters(struct rte_eth_dev *eth_dev) + struct vlan_entry *entry; + int rc; + +- /* VLAN filters can't be set without setting filtern on */ ++ /* VLAN filters can't be set without setting filters on */ + rc = nix_vlan_handle_default_rx_entry(eth_dev, false, true, true); + if (rc) { + otx2_err("Failed to reinstall vlan filters"); +diff --git a/dpdk/drivers/net/octeontx_ep/otx2_ep_vf.c b/dpdk/drivers/net/octeontx_ep/otx2_ep_vf.c +index 0716beb9b1..85e14a998f 100644 +--- a/dpdk/drivers/net/octeontx_ep/otx2_ep_vf.c ++++ b/dpdk/drivers/net/octeontx_ep/otx2_ep_vf.c +@@ -104,7 +104,7 @@ otx2_vf_setup_iq_regs(struct otx_ep_device *otx_ep, uint32_t iq_no) + iq->inst_cnt_reg = (uint8_t *)otx_ep->hw_addr + + SDP_VF_R_IN_CNTS(iq_no); + +- otx_ep_dbg("InstQ[%d]:dbell reg @ 0x%p instcnt_reg @ 0x%p", ++ otx_ep_dbg("InstQ[%d]:dbell reg @ 0x%p inst_cnt_reg @ 0x%p", + iq_no, iq->doorbell_reg, iq->inst_cnt_reg); + + do { +diff --git a/dpdk/drivers/net/octeontx_ep/otx_ep_vf.c b/dpdk/drivers/net/octeontx_ep/otx_ep_vf.c +index c9b91fef9e..96366b2a7f 100644 +--- a/dpdk/drivers/net/octeontx_ep/otx_ep_vf.c ++++ b/dpdk/drivers/net/octeontx_ep/otx_ep_vf.c +@@ -117,7 +117,7 @@ otx_ep_setup_iq_regs(struct otx_ep_device *otx_ep, uint32_t iq_no) + iq->inst_cnt_reg = (uint8_t *)otx_ep->hw_addr + + OTX_EP_R_IN_CNTS(iq_no); + +- otx_ep_dbg("InstQ[%d]:dbell reg @ 0x%p instcnt_reg @ 0x%p\n", ++ otx_ep_dbg("InstQ[%d]:dbell reg @ 0x%p inst_cnt_reg @ 0x%p\n", + iq_no, iq->doorbell_reg, iq->inst_cnt_reg); + + do { +diff --git a/dpdk/drivers/net/pfe/pfe_ethdev.c b/dpdk/drivers/net/pfe/pfe_ethdev.c +index 047010e15e..ebb5d1ae0e 100644 +--- a/dpdk/drivers/net/pfe/pfe_ethdev.c ++++ b/dpdk/drivers/net/pfe/pfe_ethdev.c +@@ -769,7 +769,7 @@ pfe_eth_init(struct rte_vdev_device *vdev, struct pfe *pfe, int id) + if (eth_dev == NULL) + return -ENOMEM; + +- /* Extract pltform data */ ++ /* Extract platform data */ + pfe_info = (struct ls1012a_pfe_platform_data *)&pfe->platform_data; + if (!pfe_info) { + PFE_PMD_ERR("pfe missing additional platform data"); +diff --git a/dpdk/drivers/net/pfe/pfe_hal.c b/dpdk/drivers/net/pfe/pfe_hal.c +index 41d783dbff..6431dec47e 100644 +--- a/dpdk/drivers/net/pfe/pfe_hal.c ++++ b/dpdk/drivers/net/pfe/pfe_hal.c +@@ -187,7 +187,7 @@ gemac_set_mode(void *base, __rte_unused int mode) + { + u32 val = readl(base + EMAC_RCNTRL_REG); + +- /*Remove loopbank*/ ++ /* Remove loopback */ + val &= ~EMAC_RCNTRL_LOOP; + + /*Enable flow control and MII mode*/ +diff --git a/dpdk/drivers/net/pfe/pfe_hif.c b/dpdk/drivers/net/pfe/pfe_hif.c +index c4a7154ba7..69b1d0edde 100644 +--- a/dpdk/drivers/net/pfe/pfe_hif.c ++++ b/dpdk/drivers/net/pfe/pfe_hif.c +@@ -114,9 +114,9 @@ pfe_hif_init_buffers(struct pfe_hif *hif) + * results, eth id, queue id from PFE block along with data. + * so we have to provide additional memory for each packet to + * HIF rx rings so that PFE block can write its headers. +- * so, we are giving the data pointor to HIF rings whose ++ * so, we are giving the data pointer to HIF rings whose + * calculation is as below: +- * mbuf->data_pointor - Required_header_size ++ * mbuf->data_pointer - Required_header_size + * + * We are utilizing the HEADROOM area to receive the PFE + * block headers. On packet reception, HIF driver will use +diff --git a/dpdk/drivers/net/pfe/pfe_hif.h b/dpdk/drivers/net/pfe/pfe_hif.h +index 6aaf904bb1..e8d5ba10e1 100644 +--- a/dpdk/drivers/net/pfe/pfe_hif.h ++++ b/dpdk/drivers/net/pfe/pfe_hif.h +@@ -8,7 +8,7 @@ + #define HIF_CLIENT_QUEUES_MAX 16 + #define HIF_RX_PKT_MIN_SIZE RTE_CACHE_LINE_SIZE + /* +- * HIF_TX_DESC_NT value should be always greter than 4, ++ * HIF_TX_DESC_NT value should be always greater than 4, + * Otherwise HIF_TX_POLL_MARK will become zero. + */ + #define HIF_RX_DESC_NT 64 +diff --git a/dpdk/drivers/net/pfe/pfe_hif_lib.c b/dpdk/drivers/net/pfe/pfe_hif_lib.c +index 799050dce3..6fe6d33d23 100644 +--- a/dpdk/drivers/net/pfe/pfe_hif_lib.c ++++ b/dpdk/drivers/net/pfe/pfe_hif_lib.c +@@ -38,7 +38,7 @@ pfe_hif_shm_clean(struct hif_shm *hif_shm) + * This function should be called before initializing HIF driver. + * + * @param[in] hif_shm Shared memory address location in DDR +- * @rerurn 0 - on succes, <0 on fail to initialize ++ * @return 0 - on succes, <0 on fail to initialize + */ + int + pfe_hif_shm_init(struct hif_shm *hif_shm, struct rte_mempool *mb_pool) +@@ -109,9 +109,9 @@ hif_lib_client_release_rx_buffers(struct hif_client_s *client) + for (ii = 0; ii < client->rx_q[qno].size; ii++) { + buf = (void *)desc->data; + if (buf) { +- /* Data pointor to mbuf pointor calculation: ++ /* Data pointer to mbuf pointer calculation: + * "Data - User private data - headroom - mbufsize" +- * Actual data pointor given to HIF BDs was ++ * Actual data pointer given to HIF BDs was + * "mbuf->data_offset - PFE_PKT_HEADER_SZ" + */ + buf = buf + PFE_PKT_HEADER_SZ +@@ -477,7 +477,7 @@ hif_hdr_write(struct hif_hdr *pkt_hdr, unsigned int + client_id, unsigned int qno, + u32 client_ctrl) + { +- /* Optimize the write since the destinaton may be non-cacheable */ ++ /* Optimize the write since the destination may be non-cacheable */ + if (!((unsigned long)pkt_hdr & 0x3)) { + ((u32 *)pkt_hdr)[0] = (client_ctrl << 16) | (qno << 8) | + client_id; +diff --git a/dpdk/drivers/net/qede/base/bcm_osal.h b/dpdk/drivers/net/qede/base/bcm_osal.h +index c5b5399282..9ea579bfc8 100644 +--- a/dpdk/drivers/net/qede/base/bcm_osal.h ++++ b/dpdk/drivers/net/qede/base/bcm_osal.h +@@ -14,7 +14,6 @@ + #include + #include + #include +-#include + #include + #include + #include +@@ -99,7 +98,7 @@ typedef intptr_t osal_int_ptr_t; + } while (0) + #define OSAL_VFREE(dev, memory) OSAL_FREE(dev, memory) + #define OSAL_MEM_ZERO(mem, size) bzero(mem, size) +-#define OSAL_MEMCPY(dst, src, size) rte_memcpy(dst, src, size) ++#define OSAL_MEMCPY(dst, src, size) memcpy(dst, src, size) + #define OSAL_MEMCMP(s1, s2, size) memcmp(s1, s2, size) + #define OSAL_MEMSET(dst, val, length) \ + memset(dst, val, length) +diff --git a/dpdk/drivers/net/qede/qede_debug.c b/dpdk/drivers/net/qede/qede_debug.c +index 2297d245c4..18f2d988fb 100644 +--- a/dpdk/drivers/net/qede/qede_debug.c ++++ b/dpdk/drivers/net/qede/qede_debug.c +@@ -1809,7 +1809,8 @@ static u32 qed_grc_dump_addr_range(struct ecore_hwfn *p_hwfn, + u8 split_id) + { + struct dbg_tools_data *dev_data = &p_hwfn->dbg_info; +- u8 port_id = 0, pf_id = 0, vf_id = 0, fid = 0; ++ u8 port_id = 0, pf_id = 0; ++ u16 vf_id = 0, fid = 0; + bool read_using_dmae = false; + u32 thresh; + +@@ -3522,7 +3523,7 @@ static enum dbg_status qed_grc_dump(struct ecore_hwfn *p_hwfn, + + /* Dump MCP HW Dump */ + if (qed_grc_is_included(p_hwfn, DBG_GRC_PARAM_DUMP_MCP_HW_DUMP) && +- !qed_grc_get_param(p_hwfn, DBG_GRC_PARAM_NO_MCP) && 1) ++ !qed_grc_get_param(p_hwfn, DBG_GRC_PARAM_NO_MCP)) + offset += qed_grc_dump_mcp_hw_dump(p_hwfn, + p_ptt, + dump_buf + offset, dump); +@@ -5983,7 +5984,7 @@ static char *qed_get_buf_ptr(void *buf, u32 offset) + /* Reads a param from the specified buffer. Returns the number of dwords read. + * If the returned str_param is NULL, the param is numeric and its value is + * returned in num_param. +- * Otheriwise, the param is a string and its pointer is returned in str_param. ++ * Otherwise, the param is a string and its pointer is returned in str_param. + */ + static u32 qed_read_param(u32 *dump_buf, + const char **param_name, +@@ -7558,7 +7559,7 @@ static enum dbg_status format_feature(struct ecore_hwfn *p_hwfn, + text_buf[i] = '\n'; + + +- /* Free the old dump_buf and point the dump_buf to the newly allocagted ++ /* Free the old dump_buf and point the dump_buf to the newly allocated + * and formatted text buffer. + */ + OSAL_VFREE(p_hwfn, feature->dump_buf); +diff --git a/dpdk/drivers/net/qede/qede_ethdev.c b/dpdk/drivers/net/qede/qede_ethdev.c +index 3e9aaeecd3..2a3123f0c8 100644 +--- a/dpdk/drivers/net/qede/qede_ethdev.c ++++ b/dpdk/drivers/net/qede/qede_ethdev.c +@@ -358,7 +358,7 @@ qede_assign_rxtx_handlers(struct rte_eth_dev *dev, bool is_dummy) + static void + qede_alloc_etherdev(struct qede_dev *qdev, struct qed_dev_eth_info *info) + { +- rte_memcpy(&qdev->dev_info, info, sizeof(*info)); ++ qdev->dev_info = *info; + qdev->ops = qed_ops; + } + +@@ -2338,7 +2338,7 @@ static int qede_set_mtu(struct rte_eth_dev *dev, uint16_t mtu) + if (fp->rxq != NULL) { + bufsz = (uint16_t)rte_pktmbuf_data_room_size( + fp->rxq->mb_pool) - RTE_PKTMBUF_HEADROOM; +- /* cache align the mbuf size to simplfy rx_buf_size ++ /* cache align the mbuf size to simplify rx_buf_size + * calculation + */ + bufsz = QEDE_FLOOR_TO_CACHE_LINE_SIZE(bufsz); +diff --git a/dpdk/drivers/net/qede/qede_filter.c b/dpdk/drivers/net/qede/qede_filter.c +index 440440423a..ca3165d972 100644 +--- a/dpdk/drivers/net/qede/qede_filter.c ++++ b/dpdk/drivers/net/qede/qede_filter.c +@@ -388,10 +388,8 @@ qede_arfs_construct_pkt(struct rte_eth_dev *eth_dev, + ip6->vtc_flow = + rte_cpu_to_be_32(QEDE_FDIR_IPV6_DEFAULT_VTC_FLOW); + +- rte_memcpy(&ip6->src_addr, arfs->tuple.src_ipv6, +- IPV6_ADDR_LEN); +- rte_memcpy(&ip6->dst_addr, arfs->tuple.dst_ipv6, +- IPV6_ADDR_LEN); ++ memcpy(&ip6->src_addr, arfs->tuple.src_ipv6, IPV6_ADDR_LEN); ++ memcpy(&ip6->dst_addr, arfs->tuple.dst_ipv6, IPV6_ADDR_LEN); + len += sizeof(struct rte_ipv6_hdr); + params->ipv6 = true; + +@@ -821,12 +819,10 @@ qede_flow_parse_pattern(__rte_unused struct rte_eth_dev *dev, + const struct rte_flow_item_ipv6 *spec; + + spec = pattern->spec; +- rte_memcpy(flow->entry.tuple.src_ipv6, +- spec->hdr.src_addr, +- IPV6_ADDR_LEN); +- rte_memcpy(flow->entry.tuple.dst_ipv6, +- spec->hdr.dst_addr, +- IPV6_ADDR_LEN); ++ memcpy(flow->entry.tuple.src_ipv6, ++ spec->hdr.src_addr, IPV6_ADDR_LEN); ++ memcpy(flow->entry.tuple.dst_ipv6, ++ spec->hdr.dst_addr, IPV6_ADDR_LEN); + flow->entry.tuple.eth_proto = + RTE_ETHER_TYPE_IPV6; + } +diff --git a/dpdk/drivers/net/qede/qede_main.c b/dpdk/drivers/net/qede/qede_main.c +index 2d1f70693a..c5afdb00d5 100644 +--- a/dpdk/drivers/net/qede/qede_main.c ++++ b/dpdk/drivers/net/qede/qede_main.c +@@ -373,7 +373,7 @@ qed_fill_dev_info(struct ecore_dev *edev, struct qed_dev_info *dev_info) + dev_info->mtu = ECORE_LEADING_HWFN(edev)->hw_info.mtu; + dev_info->dev_type = edev->type; + +- rte_memcpy(&dev_info->hw_mac, &edev->hwfns[0].hw_info.hw_mac_addr, ++ memcpy(&dev_info->hw_mac, &edev->hwfns[0].hw_info.hw_mac_addr, + RTE_ETHER_ADDR_LEN); + + dev_info->fw_major = FW_MAJOR_VERSION; +@@ -441,7 +441,7 @@ qed_fill_eth_dev_info(struct ecore_dev *edev, struct qed_dev_eth_info *info) + info->num_vlan_filters = RESC_NUM(&edev->hwfns[0], ECORE_VLAN) - + max_vf_vlan_filters; + +- rte_memcpy(&info->port_mac, &edev->hwfns[0].hw_info.hw_mac_addr, ++ memcpy(&info->port_mac, &edev->hwfns[0].hw_info.hw_mac_addr, + RTE_ETHER_ADDR_LEN); + } else { + ecore_vf_get_num_rxqs(ECORE_LEADING_HWFN(edev), +@@ -472,7 +472,7 @@ static void qed_set_name(struct ecore_dev *edev, char name[NAME_SIZE]) + { + int i; + +- rte_memcpy(edev->name, name, NAME_SIZE); ++ memcpy(edev->name, name, NAME_SIZE); + for_each_hwfn(edev, i) { + snprintf(edev->hwfns[i].name, NAME_SIZE, "%s-%d", name, i); + } +@@ -514,10 +514,9 @@ static void qed_fill_link(struct ecore_hwfn *hwfn, + + /* Prepare source inputs */ + if (IS_PF(hwfn->p_dev)) { +- rte_memcpy(¶ms, ecore_mcp_get_link_params(hwfn), +- sizeof(params)); +- rte_memcpy(&link, ecore_mcp_get_link_state(hwfn), sizeof(link)); +- rte_memcpy(&link_caps, ecore_mcp_get_link_capabilities(hwfn), ++ memcpy(¶ms, ecore_mcp_get_link_params(hwfn), sizeof(params)); ++ memcpy(&link, ecore_mcp_get_link_state(hwfn), sizeof(link)); ++ memcpy(&link_caps, ecore_mcp_get_link_capabilities(hwfn), + sizeof(link_caps)); + } else { + ecore_vf_read_bulletin(hwfn, &change); +diff --git a/dpdk/drivers/net/qede/qede_rxtx.c b/dpdk/drivers/net/qede/qede_rxtx.c +index c0eeea896e..936d5c2dc1 100644 +--- a/dpdk/drivers/net/qede/qede_rxtx.c ++++ b/dpdk/drivers/net/qede/qede_rxtx.c +@@ -38,48 +38,40 @@ static inline int qede_alloc_rx_buffer(struct qede_rx_queue *rxq) + + static inline int qede_alloc_rx_bulk_mbufs(struct qede_rx_queue *rxq, int count) + { ++ void *obj_p[QEDE_MAX_BULK_ALLOC_COUNT] __rte_cache_aligned; + struct rte_mbuf *mbuf = NULL; + struct eth_rx_bd *rx_bd; + dma_addr_t mapping; + int i, ret = 0; + uint16_t idx; +- uint16_t mask = NUM_RX_BDS(rxq); +- +- if (count > QEDE_MAX_BULK_ALLOC_COUNT) +- count = QEDE_MAX_BULK_ALLOC_COUNT; + + idx = rxq->sw_rx_prod & NUM_RX_BDS(rxq); + +- if (count > mask - idx + 1) +- count = mask - idx + 1; +- +- ret = rte_mempool_get_bulk(rxq->mb_pool, (void **)&rxq->sw_rx_ring[idx], +- count); +- ++ ret = rte_mempool_get_bulk(rxq->mb_pool, obj_p, count); + if (unlikely(ret)) { + PMD_RX_LOG(ERR, rxq, + "Failed to allocate %d rx buffers " + "sw_rx_prod %u sw_rx_cons %u mp entries %u free %u", +- count, +- rxq->sw_rx_prod & NUM_RX_BDS(rxq), +- rxq->sw_rx_cons & NUM_RX_BDS(rxq), ++ count, idx, rxq->sw_rx_cons & NUM_RX_BDS(rxq), + rte_mempool_avail_count(rxq->mb_pool), + rte_mempool_in_use_count(rxq->mb_pool)); + return -ENOMEM; + } + + for (i = 0; i < count; i++) { +- rte_prefetch0(rxq->sw_rx_ring[(idx + 1) & NUM_RX_BDS(rxq)]); +- mbuf = rxq->sw_rx_ring[idx & NUM_RX_BDS(rxq)]; ++ mbuf = obj_p[i]; ++ if (likely(i < count - 1)) ++ rte_prefetch0(obj_p[i + 1]); + ++ idx = rxq->sw_rx_prod & NUM_RX_BDS(rxq); ++ rxq->sw_rx_ring[idx] = mbuf; + mapping = rte_mbuf_data_iova_default(mbuf); + rx_bd = (struct eth_rx_bd *) + ecore_chain_produce(&rxq->rx_bd_ring); + rx_bd->addr.hi = rte_cpu_to_le_32(U64_HI(mapping)); + rx_bd->addr.lo = rte_cpu_to_le_32(U64_LO(mapping)); +- idx++; ++ rxq->sw_rx_prod++; + } +- rxq->sw_rx_prod = idx; + + return 0; + } +@@ -90,7 +82,7 @@ static inline int qede_alloc_rx_bulk_mbufs(struct qede_rx_queue *rxq, int count) + * (MTU + Maximum L2 Header Size + 2) / ETH_RX_MAX_BUFF_PER_PKT + * 3) In regular mode - minimum rx_buf_size should be + * (MTU + Maximum L2 Header Size + 2) +- * In above cases +2 corrosponds to 2 bytes padding in front of L2 ++ * In above cases +2 corresponds to 2 bytes padding in front of L2 + * header. + * 4) rx_buf_size should be cacheline-size aligned. So considering + * criteria 1, we need to adjust the size to floor instead of ceil, +@@ -106,7 +98,7 @@ qede_calc_rx_buf_size(struct rte_eth_dev *dev, uint16_t mbufsz, + + if (dev->data->scattered_rx) { + /* per HW limitation, only ETH_RX_MAX_BUFF_PER_PKT number of +- * bufferes can be used for single packet. So need to make sure ++ * buffers can be used for single packet. So need to make sure + * mbuf size is sufficient enough for this. + */ + if ((mbufsz * ETH_RX_MAX_BUFF_PER_PKT) < +@@ -243,11 +235,11 @@ qede_rx_queue_setup(struct rte_eth_dev *dev, uint16_t qid, + dev->data->rx_queues[qid] = NULL; + } + +- max_rx_pktlen = dev->data->mtu + RTE_ETHER_HDR_LEN; ++ max_rx_pktlen = dev->data->mtu + RTE_ETHER_HDR_LEN + RTE_ETHER_CRC_LEN; + + /* Fix up RX buffer size */ + bufsz = (uint16_t)rte_pktmbuf_data_room_size(mp) - RTE_PKTMBUF_HEADROOM; +- /* cache align the mbuf size to simplfy rx_buf_size calculation */ ++ /* cache align the mbuf size to simplify rx_buf_size calculation */ + bufsz = QEDE_FLOOR_TO_CACHE_LINE_SIZE(bufsz); + if ((rxmode->offloads & RTE_ETH_RX_OFFLOAD_SCATTER) || + (max_rx_pktlen + QEDE_ETH_OVERHEAD) > bufsz) { +@@ -887,68 +879,55 @@ qede_tx_queue_start(struct rte_eth_dev *eth_dev, uint16_t tx_queue_id) + } + + static inline void +-qede_process_tx_compl(__rte_unused struct ecore_dev *edev, +- struct qede_tx_queue *txq) ++qede_free_tx_pkt(struct qede_tx_queue *txq) + { +- uint16_t hw_bd_cons; +- uint16_t sw_tx_cons; +- uint16_t remaining; +- uint16_t mask; + struct rte_mbuf *mbuf; + uint16_t nb_segs; + uint16_t idx; +- uint16_t first_idx; +- +- rte_compiler_barrier(); +- rte_prefetch0(txq->hw_cons_ptr); +- sw_tx_cons = ecore_chain_get_cons_idx(&txq->tx_pbl); +- hw_bd_cons = rte_le_to_cpu_16(*txq->hw_cons_ptr); +-#ifdef RTE_LIBRTE_QEDE_DEBUG_TX +- PMD_TX_LOG(DEBUG, txq, "Tx Completions = %u\n", +- abs(hw_bd_cons - sw_tx_cons)); +-#endif + +- mask = NUM_TX_BDS(txq); +- idx = txq->sw_tx_cons & mask; +- +- remaining = hw_bd_cons - sw_tx_cons; +- txq->nb_tx_avail += remaining; +- first_idx = idx; +- +- while (remaining) { +- mbuf = txq->sw_tx_ring[idx]; +- RTE_ASSERT(mbuf); ++ idx = TX_CONS(txq); ++ mbuf = txq->sw_tx_ring[idx]; ++ if (mbuf) { + nb_segs = mbuf->nb_segs; +- remaining -= nb_segs; +- +- /* Prefetch the next mbuf. Note that at least the last 4 mbufs +- * that are prefetched will not be used in the current call. +- */ +- rte_mbuf_prefetch_part1(txq->sw_tx_ring[(idx + 4) & mask]); +- rte_mbuf_prefetch_part2(txq->sw_tx_ring[(idx + 4) & mask]); +- + PMD_TX_LOG(DEBUG, txq, "nb_segs to free %u\n", nb_segs); +- + while (nb_segs) { ++ /* It's like consuming rxbuf in recv() */ + ecore_chain_consume(&txq->tx_pbl); ++ txq->nb_tx_avail++; + nb_segs--; + } +- +- idx = (idx + 1) & mask; ++ rte_pktmbuf_free(mbuf); ++ txq->sw_tx_ring[idx] = NULL; ++ txq->sw_tx_cons++; + PMD_TX_LOG(DEBUG, txq, "Freed tx packet\n"); +- } +- txq->sw_tx_cons = idx; +- +- if (first_idx > idx) { +- rte_pktmbuf_free_bulk(&txq->sw_tx_ring[first_idx], +- mask - first_idx + 1); +- rte_pktmbuf_free_bulk(&txq->sw_tx_ring[0], idx); + } else { +- rte_pktmbuf_free_bulk(&txq->sw_tx_ring[first_idx], +- idx - first_idx); ++ ecore_chain_consume(&txq->tx_pbl); ++ txq->nb_tx_avail++; + } + } + ++static inline void ++qede_process_tx_compl(__rte_unused struct ecore_dev *edev, ++ struct qede_tx_queue *txq) ++{ ++ uint16_t hw_bd_cons; ++#ifdef RTE_LIBRTE_QEDE_DEBUG_TX ++ uint16_t sw_tx_cons; ++#endif ++ ++ hw_bd_cons = rte_le_to_cpu_16(*txq->hw_cons_ptr); ++ /* read barrier prevents speculative execution on stale data */ ++ rte_rmb(); ++ ++#ifdef RTE_LIBRTE_QEDE_DEBUG_TX ++ sw_tx_cons = ecore_chain_get_cons_idx(&txq->tx_pbl); ++ PMD_TX_LOG(DEBUG, txq, "Tx Completions = %u\n", ++ abs(hw_bd_cons - sw_tx_cons)); ++#endif ++ while (hw_bd_cons != ecore_chain_get_cons_idx(&txq->tx_pbl)) ++ qede_free_tx_pkt(txq); ++} ++ + static int qede_drain_txq(struct qede_dev *qdev, + struct qede_tx_queue *txq, bool allow_drain) + { +@@ -1559,25 +1538,26 @@ qede_recv_pkts_regular(void *p_rxq, struct rte_mbuf **rx_pkts, uint16_t nb_pkts) + uint8_t bitfield_val; + #endif + uint8_t offset, flags, bd_num; +- ++ uint16_t count = 0; + + /* Allocate buffers that we used in previous loop */ + if (rxq->rx_alloc_count) { +- if (unlikely(qede_alloc_rx_bulk_mbufs(rxq, +- rxq->rx_alloc_count))) { ++ count = rxq->rx_alloc_count > QEDE_MAX_BULK_ALLOC_COUNT ? ++ QEDE_MAX_BULK_ALLOC_COUNT : rxq->rx_alloc_count; ++ ++ if (unlikely(qede_alloc_rx_bulk_mbufs(rxq, count))) { + struct rte_eth_dev *dev; + + PMD_RX_LOG(ERR, rxq, +- "New buffer allocation failed," +- "dropping incoming packetn"); ++ "New buffers allocation failed," ++ "dropping incoming packets\n"); + dev = &rte_eth_devices[rxq->port_id]; +- dev->data->rx_mbuf_alloc_failed += +- rxq->rx_alloc_count; +- rxq->rx_alloc_errors += rxq->rx_alloc_count; ++ dev->data->rx_mbuf_alloc_failed += count; ++ rxq->rx_alloc_errors += count; + return 0; + } + qede_update_rx_prod(qdev, rxq); +- rxq->rx_alloc_count = 0; ++ rxq->rx_alloc_count -= count; + } + + hw_comp_cons = rte_le_to_cpu_16(*rxq->hw_cons_ptr); +@@ -1745,8 +1725,8 @@ qede_recv_pkts_regular(void *p_rxq, struct rte_mbuf **rx_pkts, uint16_t nb_pkts) + } + } + +- /* Request number of bufferes to be allocated in next loop */ +- rxq->rx_alloc_count = rx_alloc_count; ++ /* Request number of buffers to be allocated in next loop */ ++ rxq->rx_alloc_count += rx_alloc_count; + + rxq->rcv_pkts += rx_pkt; + rxq->rx_segs += rx_pkt; +@@ -1786,25 +1766,26 @@ qede_recv_pkts(void *p_rxq, struct rte_mbuf **rx_pkts, uint16_t nb_pkts) + struct qede_agg_info *tpa_info = NULL; + uint32_t rss_hash; + int rx_alloc_count = 0; +- ++ uint16_t count = 0; + + /* Allocate buffers that we used in previous loop */ + if (rxq->rx_alloc_count) { +- if (unlikely(qede_alloc_rx_bulk_mbufs(rxq, +- rxq->rx_alloc_count))) { ++ count = rxq->rx_alloc_count > QEDE_MAX_BULK_ALLOC_COUNT ? ++ QEDE_MAX_BULK_ALLOC_COUNT : rxq->rx_alloc_count; ++ ++ if (unlikely(qede_alloc_rx_bulk_mbufs(rxq, count))) { + struct rte_eth_dev *dev; + + PMD_RX_LOG(ERR, rxq, +- "New buffer allocation failed," +- "dropping incoming packetn"); ++ "New buffers allocation failed," ++ "dropping incoming packets\n"); + dev = &rte_eth_devices[rxq->port_id]; +- dev->data->rx_mbuf_alloc_failed += +- rxq->rx_alloc_count; +- rxq->rx_alloc_errors += rxq->rx_alloc_count; ++ dev->data->rx_mbuf_alloc_failed += count; ++ rxq->rx_alloc_errors += count; + return 0; + } + qede_update_rx_prod(qdev, rxq); +- rxq->rx_alloc_count = 0; ++ rxq->rx_alloc_count -= count; + } + + hw_comp_cons = rte_le_to_cpu_16(*rxq->hw_cons_ptr); +@@ -2042,8 +2023,8 @@ qede_recv_pkts(void *p_rxq, struct rte_mbuf **rx_pkts, uint16_t nb_pkts) + } + } + +- /* Request number of bufferes to be allocated in next loop */ +- rxq->rx_alloc_count = rx_alloc_count; ++ /* Request number of buffers to be allocated in next loop */ ++ rxq->rx_alloc_count += rx_alloc_count; + + rxq->rcv_pkts += rx_pkt; + +@@ -2506,7 +2487,7 @@ qede_xmit_pkts(void *p_txq, struct rte_mbuf **tx_pkts, uint16_t nb_pkts) + /* Inner L2 header size in two byte words */ + inner_l2_hdr_size = (mbuf->l2_len - + MPLSINUDP_HDR_SIZE) / 2; +- /* Inner L4 header offset from the beggining ++ /* Inner L4 header offset from the beginning + * of inner packet in two byte words + */ + inner_l4_hdr_offset = (mbuf->l2_len - +diff --git a/dpdk/drivers/net/qede/qede_rxtx.h b/dpdk/drivers/net/qede/qede_rxtx.h +index 754efe793f..11ed1d9b9c 100644 +--- a/dpdk/drivers/net/qede/qede_rxtx.h ++++ b/dpdk/drivers/net/qede/qede_rxtx.h +@@ -225,7 +225,7 @@ struct qede_fastpath { + struct qede_tx_queue *txq; + }; + +-/* This structure holds the inforation of fast path queues ++/* This structure holds the information of fast path queues + * belonging to individual engines in CMT mode. + */ + struct qede_fastpath_cmt { +diff --git a/dpdk/drivers/net/qede/qede_sriov.c b/dpdk/drivers/net/qede/qede_sriov.c +index 0b99a8d6fe..937d339fb8 100644 +--- a/dpdk/drivers/net/qede/qede_sriov.c ++++ b/dpdk/drivers/net/qede/qede_sriov.c +@@ -203,10 +203,10 @@ void qed_inform_vf_link_state(struct ecore_hwfn *hwfn) + if (!hwfn->pf_iov_info) + return; + +- rte_memcpy(¶ms, ecore_mcp_get_link_params(lead_hwfn), ++ memcpy(¶ms, ecore_mcp_get_link_params(lead_hwfn), + sizeof(params)); +- rte_memcpy(&link, ecore_mcp_get_link_state(lead_hwfn), sizeof(link)); +- rte_memcpy(&caps, ecore_mcp_get_link_capabilities(lead_hwfn), ++ memcpy(&link, ecore_mcp_get_link_state(lead_hwfn), sizeof(link)); ++ memcpy(&caps, ecore_mcp_get_link_capabilities(lead_hwfn), + sizeof(caps)); + + /* Update bulletin of all future possible VFs with link configuration */ +diff --git a/dpdk/drivers/net/sfc/sfc.c b/dpdk/drivers/net/sfc/sfc.c +index ed714fe02f..2cead4e045 100644 +--- a/dpdk/drivers/net/sfc/sfc.c ++++ b/dpdk/drivers/net/sfc/sfc.c +@@ -371,7 +371,7 @@ sfc_set_drv_limits(struct sfc_adapter *sa) + + /* + * Limits are strict since take into account initial estimation. +- * Resource allocation stategy is described in ++ * Resource allocation strategy is described in + * sfc_estimate_resource_limits(). + */ + lim.edl_min_evq_count = lim.edl_max_evq_count = +diff --git a/dpdk/drivers/net/sfc/sfc_dp.c b/dpdk/drivers/net/sfc/sfc_dp.c +index d4cd162541..da2d1603cf 100644 +--- a/dpdk/drivers/net/sfc/sfc_dp.c ++++ b/dpdk/drivers/net/sfc/sfc_dp.c +@@ -68,7 +68,7 @@ sfc_dp_register(struct sfc_dp_list *head, struct sfc_dp *entry) + { + if (sfc_dp_find_by_name(head, entry->type, entry->name) != NULL) { + SFC_GENERIC_LOG(ERR, +- "sfc %s dapapath '%s' already registered", ++ "sfc %s datapath '%s' already registered", + entry->type == SFC_DP_RX ? "Rx" : + entry->type == SFC_DP_TX ? "Tx" : + "unknown", +diff --git a/dpdk/drivers/net/sfc/sfc_dp_rx.h b/dpdk/drivers/net/sfc/sfc_dp_rx.h +index 760540ba22..246adbd87c 100644 +--- a/dpdk/drivers/net/sfc/sfc_dp_rx.h ++++ b/dpdk/drivers/net/sfc/sfc_dp_rx.h +@@ -158,7 +158,7 @@ typedef int (sfc_dp_rx_qcreate_t)(uint16_t port_id, uint16_t queue_id, + struct sfc_dp_rxq **dp_rxqp); + + /** +- * Free resources allocated for datapath recevie queue. ++ * Free resources allocated for datapath receive queue. + */ + typedef void (sfc_dp_rx_qdestroy_t)(struct sfc_dp_rxq *dp_rxq); + +@@ -191,7 +191,7 @@ typedef bool (sfc_dp_rx_qrx_ps_ev_t)(struct sfc_dp_rxq *dp_rxq, + /** + * Receive queue purge function called after queue flush. + * +- * Should be used to free unused recevie buffers. ++ * Should be used to free unused receive buffers. + */ + typedef void (sfc_dp_rx_qpurge_t)(struct sfc_dp_rxq *dp_rxq); + +diff --git a/dpdk/drivers/net/sfc/sfc_ef100.h b/dpdk/drivers/net/sfc/sfc_ef100.h +index 5e2052d142..e81847e75a 100644 +--- a/dpdk/drivers/net/sfc/sfc_ef100.h ++++ b/dpdk/drivers/net/sfc/sfc_ef100.h +@@ -19,7 +19,7 @@ extern "C" { + * + * @param evq_prime Global address of the prime register + * @param evq_hw_index Event queue index +- * @param evq_read_ptr Masked event qeueu read pointer ++ * @param evq_read_ptr Masked event queue read pointer + */ + static inline void + sfc_ef100_evq_prime(volatile void *evq_prime, unsigned int evq_hw_index, +diff --git a/dpdk/drivers/net/sfc/sfc_ef100_rx.c b/dpdk/drivers/net/sfc/sfc_ef100_rx.c +index 5d16bf281d..45253ed7dc 100644 +--- a/dpdk/drivers/net/sfc/sfc_ef100_rx.c ++++ b/dpdk/drivers/net/sfc/sfc_ef100_rx.c +@@ -851,7 +851,7 @@ sfc_ef100_rx_qstart(struct sfc_dp_rxq *dp_rxq, unsigned int evq_read_ptr, + unsup_rx_prefix_fields = + efx_rx_prefix_layout_check(pinfo, &sfc_ef100_rx_prefix_layout); + +- /* LENGTH and CLASS filds must always be present */ ++ /* LENGTH and CLASS fields must always be present */ + if ((unsup_rx_prefix_fields & + ((1U << EFX_RX_PREFIX_FIELD_LENGTH) | + (1U << EFX_RX_PREFIX_FIELD_CLASS))) != 0) +diff --git a/dpdk/drivers/net/sfc/sfc_ef10_essb_rx.c b/dpdk/drivers/net/sfc/sfc_ef10_essb_rx.c +index 712c207617..78bd430363 100644 +--- a/dpdk/drivers/net/sfc/sfc_ef10_essb_rx.c ++++ b/dpdk/drivers/net/sfc/sfc_ef10_essb_rx.c +@@ -630,7 +630,7 @@ sfc_ef10_essb_rx_qcreate(uint16_t port_id, uint16_t queue_id, + rxq->block_size, rxq->buf_stride); + sfc_ef10_essb_rx_info(&rxq->dp.dpq, + "max fill level is %u descs (%u bufs), " +- "refill threashold %u descs (%u bufs)", ++ "refill threshold %u descs (%u bufs)", + rxq->max_fill_level, + rxq->max_fill_level * rxq->block_size, + rxq->refill_threshold, +diff --git a/dpdk/drivers/net/sfc/sfc_ef10_rx_ev.h b/dpdk/drivers/net/sfc/sfc_ef10_rx_ev.h +index 821e2227bb..412254e3d7 100644 +--- a/dpdk/drivers/net/sfc/sfc_ef10_rx_ev.h ++++ b/dpdk/drivers/net/sfc/sfc_ef10_rx_ev.h +@@ -40,7 +40,7 @@ sfc_ef10_rx_ev_to_offloads(const efx_qword_t rx_ev, struct rte_mbuf *m, + rte_cpu_to_le_64((1ull << ESF_DZ_RX_ECC_ERR_LBN) | + (1ull << ESF_DZ_RX_ECRC_ERR_LBN) | + (1ull << ESF_DZ_RX_PARSE_INCOMPLETE_LBN)))) { +- /* Zero packet type is used as a marker to dicard bad packets */ ++ /* Zero packet type is used as a marker to discard bad packets */ + goto done; + } + +diff --git a/dpdk/drivers/net/sfc/sfc_ethdev.c b/dpdk/drivers/net/sfc/sfc_ethdev.c +index d4210b63dd..184f6e7c67 100644 +--- a/dpdk/drivers/net/sfc/sfc_ethdev.c ++++ b/dpdk/drivers/net/sfc/sfc_ethdev.c +@@ -94,7 +94,6 @@ sfc_dev_infos_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info) + struct sfc_adapter *sa = sfc_adapter_by_eth_dev(dev); + struct sfc_rss *rss = &sas->rss; + struct sfc_mae *mae = &sa->mae; +- uint64_t txq_offloads_def = 0; + + sfc_log_init(sa, "entry"); + +@@ -146,11 +145,6 @@ sfc_dev_infos_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info) + dev_info->tx_offload_capa = sfc_tx_get_dev_offload_caps(sa) | + dev_info->tx_queue_offload_capa; + +- if (dev_info->tx_offload_capa & RTE_ETH_TX_OFFLOAD_MBUF_FAST_FREE) +- txq_offloads_def |= RTE_ETH_TX_OFFLOAD_MBUF_FAST_FREE; +- +- dev_info->default_txconf.offloads |= txq_offloads_def; +- + if (rss->context_type != EFX_RX_SCALE_UNAVAILABLE) { + uint64_t rte_hf = 0; + unsigned int i; +diff --git a/dpdk/drivers/net/sfc/sfc_flow.c b/dpdk/drivers/net/sfc/sfc_flow.c +index fc74c8035e..509fde4a86 100644 +--- a/dpdk/drivers/net/sfc/sfc_flow.c ++++ b/dpdk/drivers/net/sfc/sfc_flow.c +@@ -1477,6 +1477,9 @@ sfc_flow_parse_rss(struct sfc_adapter *sa, + rxq_hw_index_max = rxq->hw_index; + } + ++ if (rxq_hw_index_max - rxq_hw_index_min + 1 > EFX_MAXRSS) ++ return -EINVAL; ++ + switch (action_rss->func) { + case RTE_ETH_HASH_FUNCTION_DEFAULT: + case RTE_ETH_HASH_FUNCTION_TOEPLITZ: +@@ -1612,9 +1615,8 @@ sfc_flow_filter_insert(struct sfc_adapter *sa, + uint8_t *rss_key; + + if (spec_filter->rss) { +- rss_spread = MIN(flow_rss->rxq_hw_index_max - +- flow_rss->rxq_hw_index_min + 1, +- EFX_MAXRSS); ++ rss_spread = flow_rss->rxq_hw_index_max - ++ flow_rss->rxq_hw_index_min + 1; + rss_hash_types = flow_rss->rss_hash_types; + rss_key = flow_rss->rss_key; + } else { +diff --git a/dpdk/drivers/net/sfc/sfc_flow_tunnel.c b/dpdk/drivers/net/sfc/sfc_flow_tunnel.c +index 463b01c596..af5941c1ba 100644 +--- a/dpdk/drivers/net/sfc/sfc_flow_tunnel.c ++++ b/dpdk/drivers/net/sfc/sfc_flow_tunnel.c +@@ -21,7 +21,7 @@ sfc_flow_tunnel_is_supported(struct sfc_adapter *sa) + SFC_ASSERT(sfc_adapter_is_locked(sa)); + + return ((sa->priv.dp_rx->features & SFC_DP_RX_FEAT_FLOW_MARK) != 0 && +- sa->mae.status == SFC_MAE_STATUS_SUPPORTED); ++ sa->mae.status == SFC_MAE_STATUS_ADMIN); + } + + bool +@@ -433,7 +433,7 @@ sfc_flow_tunnel_get_restore_info(struct rte_eth_dev *dev, + ft = &sa->flow_tunnels[ft_id]; + + if (ft->refcnt == 0) { +- sfc_err(sa, "tunnel offload: get_restore_info: tunnel=%u does not exist", ++ sfc_dbg(sa, "tunnel offload: get_restore_info: tunnel=%u does not exist", + ft_id); + rc = ENOENT; + goto fail; +diff --git a/dpdk/drivers/net/sfc/sfc_intr.c b/dpdk/drivers/net/sfc/sfc_intr.c +index ab67aa9237..ddddefad7b 100644 +--- a/dpdk/drivers/net/sfc/sfc_intr.c ++++ b/dpdk/drivers/net/sfc/sfc_intr.c +@@ -8,7 +8,7 @@ + */ + + /* +- * At the momemt of writing DPDK v16.07 has notion of two types of ++ * At the moment of writing DPDK v16.07 has notion of two types of + * interrupts: LSC (link status change) and RXQ (receive indication). + * It allows to register interrupt callback for entire device which is + * not intended to be used for receive indication (i.e. link status +diff --git a/dpdk/drivers/net/sfc/sfc_repr_proxy.c b/dpdk/drivers/net/sfc/sfc_repr_proxy.c +index 535b07ea52..8660d419a3 100644 +--- a/dpdk/drivers/net/sfc/sfc_repr_proxy.c ++++ b/dpdk/drivers/net/sfc/sfc_repr_proxy.c +@@ -1413,6 +1413,7 @@ sfc_repr_proxy_add_rxq(uint16_t pf_port_id, uint16_t repr_id, + port = sfc_repr_proxy_find_port(rp, repr_id); + if (port == NULL) { + sfc_err(sa, "%s() failed: no such port", __func__); ++ sfc_put_adapter(sa); + return ENOENT; + } + +@@ -1451,6 +1452,7 @@ sfc_repr_proxy_del_rxq(uint16_t pf_port_id, uint16_t repr_id, + port = sfc_repr_proxy_find_port(rp, repr_id); + if (port == NULL) { + sfc_err(sa, "%s() failed: no such port", __func__); ++ sfc_put_adapter(sa); + return; + } + +@@ -1484,6 +1486,7 @@ sfc_repr_proxy_add_txq(uint16_t pf_port_id, uint16_t repr_id, + port = sfc_repr_proxy_find_port(rp, repr_id); + if (port == NULL) { + sfc_err(sa, "%s() failed: no such port", __func__); ++ sfc_put_adapter(sa); + return ENOENT; + } + +@@ -1516,6 +1519,7 @@ sfc_repr_proxy_del_txq(uint16_t pf_port_id, uint16_t repr_id, + port = sfc_repr_proxy_find_port(rp, repr_id); + if (port == NULL) { + sfc_err(sa, "%s() failed: no such port", __func__); ++ sfc_put_adapter(sa); + return; + } + +@@ -1614,6 +1618,7 @@ sfc_repr_proxy_stop_repr(uint16_t pf_port_id, uint16_t repr_id) + port = sfc_repr_proxy_find_port(rp, repr_id); + if (port == NULL) { + sfc_err(sa, "%s() failed: no such port", __func__); ++ sfc_put_adapter(sa); + return ENOENT; + } + +diff --git a/dpdk/drivers/net/sfc/sfc_rx.c b/dpdk/drivers/net/sfc/sfc_rx.c +index 7104284106..cd58d60a36 100644 +--- a/dpdk/drivers/net/sfc/sfc_rx.c ++++ b/dpdk/drivers/net/sfc/sfc_rx.c +@@ -1057,7 +1057,7 @@ sfc_rx_mb_pool_buf_size(struct sfc_adapter *sa, struct rte_mempool *mb_pool) + /* Make sure that end padding does not write beyond the buffer */ + if (buf_aligned < nic_align_end) { + /* +- * Estimate space which can be lost. If guarnteed buffer ++ * Estimate space which can be lost. If guaranteed buffer + * size is odd, lost space is (nic_align_end - 1). More + * accurate formula is below. + */ +@@ -1702,7 +1702,7 @@ sfc_rx_fini_queues(struct sfc_adapter *sa, unsigned int nb_rx_queues) + + /* + * Finalize only ethdev queues since other ones are finalized only +- * on device close and they may require additional deinitializaton. ++ * on device close and they may require additional deinitialization. + */ + ethdev_qid = sas->ethdev_rxq_count; + while (--ethdev_qid >= (int)nb_rx_queues) { +@@ -1775,7 +1775,7 @@ sfc_rx_configure(struct sfc_adapter *sa) + + reconfigure = true; + +- /* Do not ununitialize reserved queues */ ++ /* Do not uninitialize reserved queues */ + if (nb_rx_queues < sas->ethdev_rxq_count) + sfc_rx_fini_queues(sa, nb_rx_queues); + +diff --git a/dpdk/drivers/net/sfc/sfc_sw_stats.c b/dpdk/drivers/net/sfc/sfc_sw_stats.c +index 70259660c0..81f5aa3cc4 100644 +--- a/dpdk/drivers/net/sfc/sfc_sw_stats.c ++++ b/dpdk/drivers/net/sfc/sfc_sw_stats.c +@@ -777,7 +777,7 @@ sfc_sw_xstats_configure(struct sfc_adapter *sa) + + memset(*reset_vals, 0, nb_supported * sizeof(**reset_vals)); + +- *cache = rte_realloc(*cache, cache_count * sizeof(*cache), 0); ++ *cache = rte_realloc(*cache, cache_count * sizeof(**cache), 0); + if (*cache == NULL) { + rc = ENOMEM; + goto fail_cache; +diff --git a/dpdk/drivers/net/sfc/sfc_tx.c b/dpdk/drivers/net/sfc/sfc_tx.c +index 0dccf21f7c..f376f24f7b 100644 +--- a/dpdk/drivers/net/sfc/sfc_tx.c ++++ b/dpdk/drivers/net/sfc/sfc_tx.c +@@ -308,6 +308,7 @@ sfc_tx_qinit_info(struct sfc_adapter *sa, sfc_sw_index_t sw_index) + static int + sfc_tx_check_mode(struct sfc_adapter *sa, const struct rte_eth_txmode *txmode) + { ++ uint64_t dev_tx_offload_cap = sfc_tx_get_dev_offload_caps(sa); + int rc = 0; + + switch (txmode->mq_mode) { +@@ -319,6 +320,13 @@ sfc_tx_check_mode(struct sfc_adapter *sa, const struct rte_eth_txmode *txmode) + rc = EINVAL; + } + ++ if ((dev_tx_offload_cap & RTE_ETH_TX_OFFLOAD_MBUF_FAST_FREE) != 0 && ++ (txmode->offloads & RTE_ETH_TX_OFFLOAD_MBUF_FAST_FREE) == 0) { ++ sfc_err(sa, "There is no FAST_FREE flag in the attempted Tx mode configuration"); ++ sfc_err(sa, "FAST_FREE is always active as per the current Tx datapath variant"); ++ rc = EINVAL; ++ } ++ + /* + * These features are claimed to be i40e-specific, + * but it does make sense to double-check their absence +@@ -356,7 +364,7 @@ sfc_tx_fini_queues(struct sfc_adapter *sa, unsigned int nb_tx_queues) + + /* + * Finalize only ethdev queues since other ones are finalized only +- * on device close and they may require additional deinitializaton. ++ * on device close and they may require additional deinitialization. + */ + ethdev_qid = sas->ethdev_txq_count; + while (--ethdev_qid >= (int)nb_tx_queues) { +diff --git a/dpdk/drivers/net/softnic/rte_eth_softnic_flow.c b/dpdk/drivers/net/softnic/rte_eth_softnic_flow.c +index ca70eab678..ad96288e7e 100644 +--- a/dpdk/drivers/net/softnic/rte_eth_softnic_flow.c ++++ b/dpdk/drivers/net/softnic/rte_eth_softnic_flow.c +@@ -930,7 +930,7 @@ flow_rule_match_acl_get(struct pmd_internals *softnic __rte_unused, + * Both *tmask* and *fmask* are byte arrays of size *tsize* and *fsize* + * respectively. + * They are located within a larger buffer at offsets *toffset* and *foffset* +- * respectivelly. Both *tmask* and *fmask* represent bitmasks for the larger ++ * respectively. Both *tmask* and *fmask* represent bitmasks for the larger + * buffer. + * Question: are the two masks equivalent? + * +diff --git a/dpdk/drivers/net/tap/rte_eth_tap.c b/dpdk/drivers/net/tap/rte_eth_tap.c +index f1b48cae82..e020a2417b 100644 +--- a/dpdk/drivers/net/tap/rte_eth_tap.c ++++ b/dpdk/drivers/net/tap/rte_eth_tap.c +@@ -67,6 +67,7 @@ + + /* IPC key for queue fds sync */ + #define TAP_MP_KEY "tap_mp_sync_queues" ++#define TAP_MP_REQ_START_RXTX "tap_mp_req_start_rxtx" + + #define TAP_IOV_DEFAULT_MAX 1024 + +@@ -525,7 +526,7 @@ tap_tx_l4_cksum(uint16_t *l4_cksum, uint16_t l4_phdr_cksum, + } + } + +-/* Accumaulate L4 raw checksums */ ++/* Accumulate L4 raw checksums */ + static void + tap_tx_l4_add_rcksum(char *l4_data, unsigned int l4_len, uint16_t *l4_cksum, + uint32_t *l4_raw_cksum) +@@ -880,11 +881,49 @@ tap_link_set_up(struct rte_eth_dev *dev) + return tap_ioctl(pmd, SIOCSIFFLAGS, &ifr, 1, LOCAL_AND_REMOTE); + } + ++static int ++tap_mp_req_on_rxtx(struct rte_eth_dev *dev) ++{ ++ struct rte_mp_msg msg; ++ struct ipc_queues *request_param = (struct ipc_queues *)msg.param; ++ int err; ++ int fd_iterator = 0; ++ struct pmd_process_private *process_private = dev->process_private; ++ int i; ++ ++ memset(&msg, 0, sizeof(msg)); ++ strlcpy(msg.name, TAP_MP_REQ_START_RXTX, sizeof(msg.name)); ++ strlcpy(request_param->port_name, dev->data->name, sizeof(request_param->port_name)); ++ msg.len_param = sizeof(*request_param); ++ for (i = 0; i < dev->data->nb_tx_queues; i++) { ++ msg.fds[fd_iterator++] = process_private->txq_fds[i]; ++ msg.num_fds++; ++ request_param->txq_count++; ++ } ++ for (i = 0; i < dev->data->nb_rx_queues; i++) { ++ msg.fds[fd_iterator++] = process_private->rxq_fds[i]; ++ msg.num_fds++; ++ request_param->rxq_count++; ++ } ++ ++ err = rte_mp_sendmsg(&msg); ++ if (err < 0) { ++ TAP_LOG(ERR, "Failed to send start req to secondary %d", ++ rte_errno); ++ return -1; ++ } ++ ++ return 0; ++} ++ + static int + tap_dev_start(struct rte_eth_dev *dev) + { + int err, i; + ++ if (rte_eal_process_type() == RTE_PROC_PRIMARY) ++ tap_mp_req_on_rxtx(dev); ++ + err = tap_intr_handle_set(dev, 1); + if (err) + return err; +@@ -901,6 +940,34 @@ tap_dev_start(struct rte_eth_dev *dev) + return err; + } + ++static int ++tap_mp_req_start_rxtx(const struct rte_mp_msg *request, __rte_unused const void *peer) ++{ ++ struct rte_eth_dev *dev; ++ const struct ipc_queues *request_param = ++ (const struct ipc_queues *)request->param; ++ int fd_iterator; ++ int queue; ++ struct pmd_process_private *process_private; ++ ++ dev = rte_eth_dev_get_by_name(request_param->port_name); ++ if (!dev) { ++ TAP_LOG(ERR, "Failed to get dev for %s", ++ request_param->port_name); ++ return -1; ++ } ++ process_private = dev->process_private; ++ fd_iterator = 0; ++ TAP_LOG(DEBUG, "tap_attach rx_q:%d tx_q:%d\n", request_param->rxq_count, ++ request_param->txq_count); ++ for (queue = 0; queue < request_param->txq_count; queue++) ++ process_private->txq_fds[queue] = request->fds[fd_iterator++]; ++ for (queue = 0; queue < request_param->rxq_count; queue++) ++ process_private->rxq_fds[queue] = request->fds[fd_iterator++]; ++ ++ return 0; ++} ++ + /* This function gets called when the current port gets stopped. + */ + static int +@@ -1084,6 +1151,9 @@ tap_dev_close(struct rte_eth_dev *dev) + + if (rte_eal_process_type() != RTE_PROC_PRIMARY) { + rte_free(dev->process_private); ++ if (tap_devices_count == 1) ++ rte_mp_action_unregister(TAP_MP_REQ_START_RXTX); ++ tap_devices_count--; + return 0; + } + +@@ -1135,6 +1205,8 @@ tap_dev_close(struct rte_eth_dev *dev) + TAP_LOG(DEBUG, "Closing %s Ethernet device on numa %u", + tuntap_types[internals->type], rte_socket_id()); + ++ rte_intr_instance_free(internals->intr_handle); ++ + if (internals->ioctl_sock != -1) { + close(internals->ioctl_sock); + internals->ioctl_sock = -1; +@@ -2099,8 +2171,8 @@ eth_dev_tap_create(struct rte_vdev_device *vdev, const char *tap_name, + close(pmd->ioctl_sock); + /* mac_addrs must not be freed alone because part of dev_private */ + dev->data->mac_addrs = NULL; +- rte_eth_dev_release_port(dev); + rte_intr_instance_free(pmd->intr_handle); ++ rte_eth_dev_release_port(dev); + + error_exit_nodev: + TAP_LOG(ERR, "%s Unable to initialize %s", +@@ -2445,6 +2517,16 @@ rte_pmd_tap_probe(struct rte_vdev_device *dev) + ret = tap_mp_attach_queues(name, eth_dev); + if (ret != 0) + return -1; ++ ++ if (!tap_devices_count) { ++ ret = rte_mp_action_register(TAP_MP_REQ_START_RXTX, tap_mp_req_start_rxtx); ++ if (ret < 0 && rte_errno != ENOTSUP) { ++ TAP_LOG(ERR, "tap: Failed to register IPC callback: %s", ++ strerror(rte_errno)); ++ return -1; ++ } ++ } ++ tap_devices_count++; + rte_eth_dev_probing_finish(eth_dev); + return 0; + } +diff --git a/dpdk/drivers/net/tap/tap_bpf_api.c b/dpdk/drivers/net/tap/tap_bpf_api.c +index 98f6a76011..15283f8917 100644 +--- a/dpdk/drivers/net/tap/tap_bpf_api.c ++++ b/dpdk/drivers/net/tap/tap_bpf_api.c +@@ -96,7 +96,7 @@ static inline int sys_bpf(enum bpf_cmd cmd, union bpf_attr *attr, + * Load BPF instructions to kernel + * + * @param[in] type +- * BPF program type: classifieir or action ++ * BPF program type: classifier or action + * + * @param[in] insns + * Array of BPF instructions (equivalent to BPF instructions) +@@ -104,7 +104,7 @@ static inline int sys_bpf(enum bpf_cmd cmd, union bpf_attr *attr, + * @param[in] insns_cnt + * Number of BPF instructions (size of array) + * +- * @param[in] lincense ++ * @param[in] license + * License string that must be acknowledged by the kernel + * + * @return +diff --git a/dpdk/drivers/net/tap/tap_flow.c b/dpdk/drivers/net/tap/tap_flow.c +index c4f60ce98e..7673823945 100644 +--- a/dpdk/drivers/net/tap/tap_flow.c ++++ b/dpdk/drivers/net/tap/tap_flow.c +@@ -961,7 +961,7 @@ add_action(struct rte_flow *flow, size_t *act_index, struct action_data *adata) + } + + /** +- * Helper function to send a serie of TC actions to the kernel ++ * Helper function to send a series of TC actions to the kernel + * + * @param[in] flow + * Pointer to rte flow containing the netlink message +@@ -2017,7 +2017,7 @@ static int bpf_rss_key(enum bpf_rss_key_e cmd, __u32 *key_idx) + break; + + /* +- * Subtract offest to restore real key index ++ * Subtract offset to restore real key index + * If a non RSS flow is falsely trying to release map + * entry 0 - the offset subtraction will calculate the real + * map index as an out-of-range value and the release operation +diff --git a/dpdk/drivers/net/tap/tap_intr.c b/dpdk/drivers/net/tap/tap_intr.c +index 56c343acea..a9097def1a 100644 +--- a/dpdk/drivers/net/tap/tap_intr.c ++++ b/dpdk/drivers/net/tap/tap_intr.c +@@ -34,8 +34,6 @@ tap_rx_intr_vec_uninstall(struct rte_eth_dev *dev) + rte_intr_free_epoll_fd(intr_handle); + rte_intr_vec_list_free(intr_handle); + rte_intr_nb_efd_set(intr_handle, 0); +- +- rte_intr_instance_free(intr_handle); + } + + /** +diff --git a/dpdk/drivers/net/thunderx/nicvf_svf.c b/dpdk/drivers/net/thunderx/nicvf_svf.c +index bccf290599..1bcf73d9fc 100644 +--- a/dpdk/drivers/net/thunderx/nicvf_svf.c ++++ b/dpdk/drivers/net/thunderx/nicvf_svf.c +@@ -21,7 +21,7 @@ nicvf_svf_push(struct nicvf *vf) + + entry = rte_zmalloc("nicvf", sizeof(*entry), RTE_CACHE_LINE_SIZE); + if (entry == NULL) +- rte_panic("Cannoc allocate memory for svf_entry\n"); ++ rte_panic("Cannot allocate memory for svf_entry\n"); + + entry->vf = vf; + +diff --git a/dpdk/drivers/net/txgbe/base/meson.build b/dpdk/drivers/net/txgbe/base/meson.build +index 7a30191472..a81d6890fe 100644 +--- a/dpdk/drivers/net/txgbe/base/meson.build ++++ b/dpdk/drivers/net/txgbe/base/meson.build +@@ -22,6 +22,6 @@ foreach flag: error_cflags + endforeach + + base_lib = static_library('txgbe_base', sources, +- dependencies: static_rte_eal, ++ dependencies: [static_rte_eal, static_rte_net], + c_args: c_args) + base_objs = base_lib.extract_all_objects(recursive: true) +diff --git a/dpdk/drivers/net/txgbe/base/txgbe_eeprom.c b/dpdk/drivers/net/txgbe/base/txgbe_eeprom.c +index 72901cd0b0..4ed6bd6728 100644 +--- a/dpdk/drivers/net/txgbe/base/txgbe_eeprom.c ++++ b/dpdk/drivers/net/txgbe/base/txgbe_eeprom.c +@@ -21,8 +21,6 @@ s32 txgbe_init_eeprom_params(struct txgbe_hw *hw) + u16 eeprom_size; + int err = 0; + +- DEBUGFUNC("txgbe_init_eeprom_params"); +- + if (eeprom->type != txgbe_eeprom_unknown) + return 0; + +@@ -54,12 +52,12 @@ s32 txgbe_init_eeprom_params(struct txgbe_hw *hw) + + err = eeprom->read32(hw, TXGBE_SW_REGION_PTR << 1, &eeprom->sw_addr); + if (err) { +- DEBUGOUT("EEPROM read failed.\n"); ++ DEBUGOUT("EEPROM read failed."); + return err; + } + +- DEBUGOUT("eeprom params: type = %d, size = %d, address bits: " +- "%d %d\n", eeprom->type, eeprom->word_size, ++ DEBUGOUT("eeprom params: type = %d, size = %d, address bits: %d %d", ++ eeprom->type, eeprom->word_size, + eeprom->address_bits, eeprom->sw_addr); + + return 0; +@@ -78,9 +76,6 @@ s32 txgbe_get_eeprom_semaphore(struct txgbe_hw *hw) + u32 i; + u32 swsm; + +- DEBUGFUNC("txgbe_get_eeprom_semaphore"); +- +- + /* Get SMBI software semaphore between device drivers first */ + for (i = 0; i < timeout; i++) { + /* +@@ -96,8 +91,7 @@ s32 txgbe_get_eeprom_semaphore(struct txgbe_hw *hw) + } + + if (i == timeout) { +- DEBUGOUT("Driver can't access the eeprom - SMBI Semaphore " +- "not granted.\n"); ++ DEBUGOUT("Driver can't access the eeprom - SMBI Semaphore not granted."); + /* + * this release is particularly important because our attempts + * above to get the semaphore may have succeeded, and if there +@@ -140,13 +134,12 @@ s32 txgbe_get_eeprom_semaphore(struct txgbe_hw *hw) + * was not granted because we don't have access to the EEPROM + */ + if (i >= timeout) { +- DEBUGOUT("SWESMBI Software EEPROM semaphore not granted.\n"); ++ DEBUGOUT("SWESMBI Software EEPROM semaphore not granted."); + txgbe_release_eeprom_semaphore(hw); + status = TXGBE_ERR_EEPROM; + } + } else { +- DEBUGOUT("Software semaphore SMBI between device drivers " +- "not granted.\n"); ++ DEBUGOUT("Software semaphore SMBI between device drivers not granted."); + } + + return status; +@@ -160,8 +153,6 @@ s32 txgbe_get_eeprom_semaphore(struct txgbe_hw *hw) + **/ + void txgbe_release_eeprom_semaphore(struct txgbe_hw *hw) + { +- DEBUGFUNC("txgbe_release_eeprom_semaphore"); +- + wr32m(hw, TXGBE_MNGSWSYNC, TXGBE_MNGSWSYNC_REQ, 0); + wr32m(hw, TXGBE_SWSEM, TXGBE_SWSEM_PF, 0); + txgbe_flush(hw); +@@ -290,8 +281,6 @@ s32 txgbe_ee_write16(struct txgbe_hw *hw, u32 offset, + u32 addr = (offset << 1); + int err; + +- DEBUGFUNC("\n"); +- + err = hw->mac.acquire_swfw_sync(hw, mask); + if (err) + return err; +@@ -348,8 +337,6 @@ s32 txgbe_ee_writew_sw(struct txgbe_hw *hw, u32 offset, + u32 addr = hw->rom.sw_addr + (offset << 1); + int err; + +- DEBUGFUNC("\n"); +- + err = hw->mac.acquire_swfw_sync(hw, mask); + if (err) + return err; +@@ -399,11 +386,9 @@ s32 txgbe_calc_eeprom_checksum(struct txgbe_hw *hw) + int err; + u16 buffer[BUFF_SIZE]; + +- DEBUGFUNC("txgbe_calc_eeprom_checksum"); +- + err = hw->rom.readw_sw(hw, TXGBE_EEPROM_CHECKSUM, &read_checksum); + if (err) { +- DEBUGOUT("EEPROM read failed\n"); ++ DEBUGOUT("EEPROM read failed"); + return err; + } + +@@ -437,15 +422,13 @@ s32 txgbe_validate_eeprom_checksum(struct txgbe_hw *hw, + u16 read_checksum = 0; + int err; + +- DEBUGFUNC("txgbe_validate_eeprom_checksum"); +- + /* Read the first word from the EEPROM. If this times out or fails, do + * not continue or we could be in for a very long wait while every + * EEPROM read fails + */ + err = hw->rom.read16(hw, 0, &checksum); + if (err) { +- DEBUGOUT("EEPROM read failed\n"); ++ DEBUGOUT("EEPROM read failed"); + return err; + } + +@@ -457,7 +440,7 @@ s32 txgbe_validate_eeprom_checksum(struct txgbe_hw *hw, + + err = hw->rom.readw_sw(hw, TXGBE_EEPROM_CHECKSUM, &read_checksum); + if (err) { +- DEBUGOUT("EEPROM read failed\n"); ++ DEBUGOUT("EEPROM read failed"); + return err; + } + +@@ -466,7 +449,7 @@ s32 txgbe_validate_eeprom_checksum(struct txgbe_hw *hw, + */ + if (read_checksum != checksum) { + err = TXGBE_ERR_EEPROM_CHECKSUM; +- DEBUGOUT("EEPROM checksum error\n"); ++ DEBUGOUT("EEPROM checksum error"); + } + + /* If the user cares, return the calculated checksum */ +@@ -485,15 +468,13 @@ s32 txgbe_update_eeprom_checksum(struct txgbe_hw *hw) + s32 status; + u16 checksum; + +- DEBUGFUNC("txgbe_update_eeprom_checksum"); +- + /* Read the first word from the EEPROM. If this times out or fails, do + * not continue or we could be in for a very long wait while every + * EEPROM read fails + */ + status = hw->rom.read16(hw, 0, &checksum); + if (status) { +- DEBUGOUT("EEPROM read failed\n"); ++ DEBUGOUT("EEPROM read failed"); + return status; + } + +diff --git a/dpdk/drivers/net/txgbe/base/txgbe_hw.c b/dpdk/drivers/net/txgbe/base/txgbe_hw.c +index 00a8db78bf..776891ee7e 100644 +--- a/dpdk/drivers/net/txgbe/base/txgbe_hw.c ++++ b/dpdk/drivers/net/txgbe/base/txgbe_hw.c +@@ -42,8 +42,6 @@ bool txgbe_device_supports_autoneg_fc(struct txgbe_hw *hw) + u32 speed; + bool link_up; + +- DEBUGFUNC("txgbe_device_supports_autoneg_fc"); +- + switch (hw->phy.media_type) { + case txgbe_media_type_fiber_qsfp: + case txgbe_media_type_fiber: +@@ -93,11 +91,9 @@ s32 txgbe_setup_fc(struct txgbe_hw *hw) + u32 value = 0; + u64 reg_bp = 0; + +- DEBUGFUNC("txgbe_setup_fc"); +- + /* Validate the requested mode */ + if (hw->fc.strict_ieee && hw->fc.requested_mode == txgbe_fc_rx_pause) { +- DEBUGOUT("txgbe_fc_rx_pause not valid in strict IEEE mode\n"); ++ DEBUGOUT("txgbe_fc_rx_pause not valid in strict IEEE mode"); + err = TXGBE_ERR_INVALID_LINK_SETTINGS; + goto out; + } +@@ -149,7 +145,7 @@ s32 txgbe_setup_fc(struct txgbe_hw *hw) + SR_AN_MMD_ADV_REG1_PAUSE_ASM; + break; + default: +- DEBUGOUT("Flow control param set incorrectly\n"); ++ DEBUGOUT("Flow control param set incorrectly"); + err = TXGBE_ERR_CONFIG; + goto out; + } +@@ -180,7 +176,7 @@ s32 txgbe_setup_fc(struct txgbe_hw *hw) + TXGBE_MD_DEV_AUTO_NEG, reg_cu); + } + +- DEBUGOUT("Set up FC; reg = 0x%08X\n", reg); ++ DEBUGOUT("Set up FC; reg = 0x%08X", reg); + out: + return err; + } +@@ -199,8 +195,6 @@ s32 txgbe_start_hw(struct txgbe_hw *hw) + s32 err; + u16 device_caps; + +- DEBUGFUNC("txgbe_start_hw"); +- + /* Set the media type */ + hw->phy.media_type = hw->phy.get_media_type(hw); + +@@ -213,7 +207,7 @@ s32 txgbe_start_hw(struct txgbe_hw *hw) + /* Setup flow control */ + err = txgbe_setup_fc(hw); + if (err != 0 && err != TXGBE_NOT_IMPLEMENTED) { +- DEBUGOUT("Flow control setup failed, returning %d\n", err); ++ DEBUGOUT("Flow control setup failed, returning %d", err); + return err; + } + +@@ -275,8 +269,6 @@ s32 txgbe_init_hw(struct txgbe_hw *hw) + { + s32 status; + +- DEBUGFUNC("txgbe_init_hw"); +- + /* Get firmware version */ + hw->phy.get_fw_version(hw, &hw->fw_version); + +@@ -288,7 +280,7 @@ s32 txgbe_init_hw(struct txgbe_hw *hw) + } + + if (status != 0) +- DEBUGOUT("Failed to initialize HW, STATUS = %d\n", status); ++ DEBUGOUT("Failed to initialize HW, STATUS = %d", status); + + return status; + } +@@ -304,8 +296,6 @@ s32 txgbe_clear_hw_cntrs(struct txgbe_hw *hw) + { + u16 i = 0; + +- DEBUGFUNC("txgbe_clear_hw_cntrs"); +- + /* QP Stats */ + /* don't write clear queue stats */ + for (i = 0; i < TXGBE_MAX_QP; i++) { +@@ -425,8 +415,6 @@ s32 txgbe_get_mac_addr(struct txgbe_hw *hw, u8 *mac_addr) + u32 rar_low; + u16 i; + +- DEBUGFUNC("txgbe_get_mac_addr"); +- + wr32(hw, TXGBE_ETHADDRIDX, 0); + rar_high = rd32(hw, TXGBE_ETHADDRH); + rar_low = rd32(hw, TXGBE_ETHADDRL); +@@ -452,8 +440,6 @@ void txgbe_set_lan_id_multi_port(struct txgbe_hw *hw) + struct txgbe_bus_info *bus = &hw->bus; + u32 reg; + +- DEBUGFUNC("txgbe_set_lan_id_multi_port_pcie"); +- + reg = rd32(hw, TXGBE_PORTSTAT); + bus->lan_id = TXGBE_PORTSTAT_ID(reg); + +@@ -479,8 +465,6 @@ s32 txgbe_stop_hw(struct txgbe_hw *hw) + u32 reg_val; + u16 i; + +- DEBUGFUNC("txgbe_stop_hw"); +- + /* + * Set the adapter_stopped flag so other driver functions stop touching + * the hardware +@@ -527,8 +511,6 @@ s32 txgbe_led_on(struct txgbe_hw *hw, u32 index) + { + u32 led_reg = rd32(hw, TXGBE_LEDCTL); + +- DEBUGFUNC("txgbe_led_on"); +- + if (index > 4) + return TXGBE_ERR_PARAM; + +@@ -550,8 +532,6 @@ s32 txgbe_led_off(struct txgbe_hw *hw, u32 index) + { + u32 led_reg = rd32(hw, TXGBE_LEDCTL); + +- DEBUGFUNC("txgbe_led_off"); +- + if (index > 4) + return TXGBE_ERR_PARAM; + +@@ -574,8 +554,6 @@ s32 txgbe_validate_mac_addr(u8 *mac_addr) + { + s32 status = 0; + +- DEBUGFUNC("txgbe_validate_mac_addr"); +- + /* Make sure it is not a multicast address */ + if (TXGBE_IS_MULTICAST(mac_addr)) { + status = TXGBE_ERR_INVALID_MAC_ADDR; +@@ -606,11 +584,9 @@ s32 txgbe_set_rar(struct txgbe_hw *hw, u32 index, u8 *addr, u32 vmdq, + u32 rar_low, rar_high; + u32 rar_entries = hw->mac.num_rar_entries; + +- DEBUGFUNC("txgbe_set_rar"); +- + /* Make sure we are using a valid rar index range */ + if (index >= rar_entries) { +- DEBUGOUT("RAR index %d is out of range.\n", index); ++ DEBUGOUT("RAR index %d is out of range.", index); + return TXGBE_ERR_INVALID_ARGUMENT; + } + +@@ -658,11 +634,9 @@ s32 txgbe_clear_rar(struct txgbe_hw *hw, u32 index) + u32 rar_high; + u32 rar_entries = hw->mac.num_rar_entries; + +- DEBUGFUNC("txgbe_clear_rar"); +- + /* Make sure we are using a valid rar index range */ + if (index >= rar_entries) { +- DEBUGOUT("RAR index %d is out of range.\n", index); ++ DEBUGOUT("RAR index %d is out of range.", index); + return TXGBE_ERR_INVALID_ARGUMENT; + } + +@@ -698,8 +672,6 @@ s32 txgbe_init_rx_addrs(struct txgbe_hw *hw) + u32 psrctl; + u32 rar_entries = hw->mac.num_rar_entries; + +- DEBUGFUNC("txgbe_init_rx_addrs"); +- + /* + * If the current mac address is valid, assume it is a software override + * to the permanent address. +@@ -710,18 +682,18 @@ s32 txgbe_init_rx_addrs(struct txgbe_hw *hw) + /* Get the MAC address from the RAR0 for later reference */ + hw->mac.get_mac_addr(hw, hw->mac.addr); + +- DEBUGOUT(" Keeping Current RAR0 Addr =%.2X %.2X %.2X ", ++ DEBUGOUT(" Keeping Current RAR0 Addr = " ++ RTE_ETHER_ADDR_PRT_FMT, + hw->mac.addr[0], hw->mac.addr[1], +- hw->mac.addr[2]); +- DEBUGOUT("%.2X %.2X %.2X\n", hw->mac.addr[3], ++ hw->mac.addr[2], hw->mac.addr[3], + hw->mac.addr[4], hw->mac.addr[5]); + } else { + /* Setup the receive address. */ +- DEBUGOUT("Overriding MAC Address in RAR[0]\n"); +- DEBUGOUT(" New MAC Addr =%.2X %.2X %.2X ", ++ DEBUGOUT("Overriding MAC Address in RAR[0]"); ++ DEBUGOUT(" New MAC Addr = " ++ RTE_ETHER_ADDR_PRT_FMT, + hw->mac.addr[0], hw->mac.addr[1], +- hw->mac.addr[2]); +- DEBUGOUT("%.2X %.2X %.2X\n", hw->mac.addr[3], ++ hw->mac.addr[2], hw->mac.addr[3], + hw->mac.addr[4], hw->mac.addr[5]); + + hw->mac.set_rar(hw, 0, hw->mac.addr, 0, true); +@@ -735,7 +707,7 @@ s32 txgbe_init_rx_addrs(struct txgbe_hw *hw) + hw->addr_ctrl.rar_used_count = 1; + + /* Zero out the other receive addresses. */ +- DEBUGOUT("Clearing RAR[1-%d]\n", rar_entries - 1); ++ DEBUGOUT("Clearing RAR[1-%d]", rar_entries - 1); + for (i = 1; i < rar_entries; i++) { + wr32(hw, TXGBE_ETHADDRIDX, i); + wr32(hw, TXGBE_ETHADDRL, 0); +@@ -749,7 +721,7 @@ s32 txgbe_init_rx_addrs(struct txgbe_hw *hw) + psrctl |= TXGBE_PSRCTL_ADHF12(hw->mac.mc_filter_type); + wr32(hw, TXGBE_PSRCTL, psrctl); + +- DEBUGOUT(" Clearing MTA\n"); ++ DEBUGOUT(" Clearing MTA"); + for (i = 0; i < hw->mac.mcft_size; i++) + wr32(hw, TXGBE_MCADDRTBL(i), 0); + +@@ -774,8 +746,6 @@ static s32 txgbe_mta_vector(struct txgbe_hw *hw, u8 *mc_addr) + { + u32 vector = 0; + +- DEBUGFUNC("txgbe_mta_vector"); +- + switch (hw->mac.mc_filter_type) { + case 0: /* use bits [47:36] of the address */ + vector = ((mc_addr[4] >> 4) | (((u16)mc_addr[5]) << 4)); +@@ -790,7 +760,7 @@ static s32 txgbe_mta_vector(struct txgbe_hw *hw, u8 *mc_addr) + vector = ((mc_addr[4]) | (((u16)mc_addr[5]) << 8)); + break; + default: /* Invalid mc_filter_type */ +- DEBUGOUT("MC filter type param set incorrectly\n"); ++ DEBUGOUT("MC filter type param set incorrectly"); + ASSERT(0); + break; + } +@@ -813,12 +783,10 @@ void txgbe_set_mta(struct txgbe_hw *hw, u8 *mc_addr) + u32 vector_bit; + u32 vector_reg; + +- DEBUGFUNC("txgbe_set_mta"); +- + hw->addr_ctrl.mta_in_use++; + + vector = txgbe_mta_vector(hw, mc_addr); +- DEBUGOUT(" bit-vector = 0x%03X\n", vector); ++ DEBUGOUT(" bit-vector = 0x%03X", vector); + + /* + * The MTA is a register array of 128 32-bit registers. It is treated +@@ -852,8 +820,6 @@ s32 txgbe_update_mc_addr_list(struct txgbe_hw *hw, u8 *mc_addr_list, + u32 i; + u32 vmdq; + +- DEBUGFUNC("txgbe_update_mc_addr_list"); +- + /* + * Set the new number of MC addresses that we are being requested to + * use. +@@ -863,13 +829,13 @@ s32 txgbe_update_mc_addr_list(struct txgbe_hw *hw, u8 *mc_addr_list, + + /* Clear mta_shadow */ + if (clear) { +- DEBUGOUT(" Clearing MTA\n"); ++ DEBUGOUT(" Clearing MTA"); + memset(&hw->mac.mta_shadow, 0, sizeof(hw->mac.mta_shadow)); + } + + /* Update mta_shadow */ + for (i = 0; i < mc_addr_count; i++) { +- DEBUGOUT(" Adding the multicast addresses:\n"); ++ DEBUGOUT(" Adding the multicast addresses:"); + txgbe_set_mta(hw, next(hw, &mc_addr_list, &vmdq)); + } + +@@ -886,7 +852,7 @@ s32 txgbe_update_mc_addr_list(struct txgbe_hw *hw, u8 *mc_addr_list, + wr32(hw, TXGBE_PSRCTL, psrctl); + } + +- DEBUGOUT("txgbe update mc addr list complete\n"); ++ DEBUGOUT("txgbe update mc addr list complete"); + return 0; + } + +@@ -904,8 +870,6 @@ s32 txgbe_fc_enable(struct txgbe_hw *hw) + u32 fcrtl, fcrth; + int i; + +- DEBUGFUNC("txgbe_fc_enable"); +- + /* Validate the water mark configuration */ + if (!hw->fc.pause_time) { + err = TXGBE_ERR_INVALID_LINK_SETTINGS; +@@ -918,7 +882,7 @@ s32 txgbe_fc_enable(struct txgbe_hw *hw) + hw->fc.high_water[i]) { + if (!hw->fc.low_water[i] || + hw->fc.low_water[i] >= hw->fc.high_water[i]) { +- DEBUGOUT("Invalid water mark configuration\n"); ++ DEBUGOUT("Invalid water mark configuration"); + err = TXGBE_ERR_INVALID_LINK_SETTINGS; + goto out; + } +@@ -976,7 +940,7 @@ s32 txgbe_fc_enable(struct txgbe_hw *hw) + fccfg_reg |= TXGBE_TXFCCFG_FC; + break; + default: +- DEBUGOUT("Flow control param set incorrectly\n"); ++ DEBUGOUT("Flow control param set incorrectly"); + err = TXGBE_ERR_CONFIG; + goto out; + } +@@ -1037,8 +1001,7 @@ s32 txgbe_negotiate_fc(struct txgbe_hw *hw, u32 adv_reg, u32 lp_reg, + u32 adv_sym, u32 adv_asm, u32 lp_sym, u32 lp_asm) + { + if ((!(adv_reg)) || (!(lp_reg))) { +- DEBUGOUT("Local or link partner's advertised flow control " +- "settings are NULL. Local: %x, link partner: %x\n", ++ DEBUGOUT("Local or link partner's advertised flow control settings are NULL. Local: %x, link partner: %x", + adv_reg, lp_reg); + return TXGBE_ERR_FC_NOT_NEGOTIATED; + } +@@ -1053,22 +1016,22 @@ s32 txgbe_negotiate_fc(struct txgbe_hw *hw, u32 adv_reg, u32 lp_reg, + */ + if (hw->fc.requested_mode == txgbe_fc_full) { + hw->fc.current_mode = txgbe_fc_full; +- DEBUGOUT("Flow Control = FULL.\n"); ++ DEBUGOUT("Flow Control = FULL."); + } else { + hw->fc.current_mode = txgbe_fc_rx_pause; +- DEBUGOUT("Flow Control=RX PAUSE frames only\n"); ++ DEBUGOUT("Flow Control=RX PAUSE frames only"); + } + } else if (!(adv_reg & adv_sym) && (adv_reg & adv_asm) && + (lp_reg & lp_sym) && (lp_reg & lp_asm)) { + hw->fc.current_mode = txgbe_fc_tx_pause; +- DEBUGOUT("Flow Control = TX PAUSE frames only.\n"); ++ DEBUGOUT("Flow Control = TX PAUSE frames only."); + } else if ((adv_reg & adv_sym) && (adv_reg & adv_asm) && + !(lp_reg & lp_sym) && (lp_reg & lp_asm)) { + hw->fc.current_mode = txgbe_fc_rx_pause; +- DEBUGOUT("Flow Control = RX PAUSE frames only.\n"); ++ DEBUGOUT("Flow Control = RX PAUSE frames only."); + } else { + hw->fc.current_mode = txgbe_fc_none; +- DEBUGOUT("Flow Control = NONE.\n"); ++ DEBUGOUT("Flow Control = NONE."); + } + return 0; + } +@@ -1168,8 +1131,6 @@ void txgbe_fc_autoneg(struct txgbe_hw *hw) + u32 speed; + bool link_up; + +- DEBUGFUNC("txgbe_fc_autoneg"); +- + /* + * AN should have completed when the cable was plugged in. + * Look for reasons to bail out. Bail out if: +@@ -1235,8 +1196,6 @@ s32 txgbe_acquire_swfw_sync(struct txgbe_hw *hw, u32 mask) + u32 timeout = 200; + u32 i; + +- DEBUGFUNC("txgbe_acquire_swfw_sync"); +- + for (i = 0; i < timeout; i++) { + /* + * SW NVM semaphore bit is used for access to all +@@ -1279,8 +1238,6 @@ void txgbe_release_swfw_sync(struct txgbe_hw *hw, u32 mask) + u32 mngsem; + u32 swmask = mask; + +- DEBUGFUNC("txgbe_release_swfw_sync"); +- + txgbe_get_eeprom_semaphore(hw); + + mngsem = rd32(hw, TXGBE_MNGSEM); +@@ -1304,8 +1261,6 @@ s32 txgbe_disable_sec_rx_path(struct txgbe_hw *hw) + int i; + u32 secrxreg; + +- DEBUGFUNC("txgbe_disable_sec_rx_path"); +- + secrxreg = rd32(hw, TXGBE_SECRXCTL); + secrxreg |= TXGBE_SECRXCTL_XDSA; + wr32(hw, TXGBE_SECRXCTL, secrxreg); +@@ -1320,8 +1275,7 @@ s32 txgbe_disable_sec_rx_path(struct txgbe_hw *hw) + + /* For informational purposes only */ + if (i >= TXGBE_MAX_SECRX_POLL) +- DEBUGOUT("Rx unit being enabled before security " +- "path fully disabled. Continuing with init.\n"); ++ DEBUGOUT("Rx unit being enabled before security path fully disabled. Continuing with init."); + + return 0; + } +@@ -1336,8 +1290,6 @@ s32 txgbe_enable_sec_rx_path(struct txgbe_hw *hw) + { + u32 secrxreg; + +- DEBUGFUNC("txgbe_enable_sec_rx_path"); +- + secrxreg = rd32(hw, TXGBE_SECRXCTL); + secrxreg &= ~TXGBE_SECRXCTL_XDSA; + wr32(hw, TXGBE_SECRXCTL, secrxreg); +@@ -1373,8 +1325,7 @@ int txgbe_disable_sec_tx_path(struct txgbe_hw *hw) + + /* For informational purposes only */ + if (i >= TXGBE_MAX_SECTX_POLL) +- PMD_DRV_LOG(DEBUG, "Tx unit being enabled before security " +- "path fully disabled. Continuing with init."); ++ DEBUGOUT("Tx unit being enabled before security path fully disabled. Continuing with init."); + + return 0; + } +@@ -1411,8 +1362,6 @@ static s32 txgbe_get_san_mac_addr_offset(struct txgbe_hw *hw, + { + s32 err; + +- DEBUGFUNC("txgbe_get_san_mac_addr_offset"); +- + /* + * First read the EEPROM pointer to see if the MAC addresses are + * available. +@@ -1443,8 +1392,6 @@ s32 txgbe_get_san_mac_addr(struct txgbe_hw *hw, u8 *san_mac_addr) + u8 i; + s32 err; + +- DEBUGFUNC("txgbe_get_san_mac_addr"); +- + /* + * First read the EEPROM pointer to see if the MAC addresses are + * available. If they're not, no point in calling set_lan_id() here. +@@ -1493,8 +1440,6 @@ s32 txgbe_set_san_mac_addr(struct txgbe_hw *hw, u8 *san_mac_addr) + u16 san_mac_data, san_mac_offset; + u8 i; + +- DEBUGFUNC("txgbe_set_san_mac_addr"); +- + /* Look for SAN mac address pointer. If not defined, return */ + err = txgbe_get_san_mac_addr_offset(hw, &san_mac_offset); + if (err || san_mac_offset == 0 || san_mac_offset == 0xFFFF) +@@ -1525,11 +1470,9 @@ s32 txgbe_clear_vmdq(struct txgbe_hw *hw, u32 rar, u32 vmdq) + u32 mpsar_lo, mpsar_hi; + u32 rar_entries = hw->mac.num_rar_entries; + +- DEBUGFUNC("txgbe_clear_vmdq"); +- + /* Make sure we are using a valid rar index range */ + if (rar >= rar_entries) { +- DEBUGOUT("RAR index %d is out of range.\n", rar); ++ DEBUGOUT("RAR index %d is out of range.", rar); + return TXGBE_ERR_INVALID_ARGUMENT; + } + +@@ -1579,11 +1522,9 @@ s32 txgbe_set_vmdq(struct txgbe_hw *hw, u32 rar, u32 vmdq) + u32 mpsar; + u32 rar_entries = hw->mac.num_rar_entries; + +- DEBUGFUNC("txgbe_set_vmdq"); +- + /* Make sure we are using a valid rar index range */ + if (rar >= rar_entries) { +- DEBUGOUT("RAR index %d is out of range.\n", rar); ++ DEBUGOUT("RAR index %d is out of range.", rar); + return TXGBE_ERR_INVALID_ARGUMENT; + } + +@@ -1608,8 +1549,7 @@ s32 txgbe_init_uta_tables(struct txgbe_hw *hw) + { + int i; + +- DEBUGFUNC("txgbe_init_uta_tables"); +- DEBUGOUT(" Clearing UTA\n"); ++ DEBUGOUT(" Clearing UTA"); + + for (i = 0; i < 128; i++) + wr32(hw, TXGBE_UCADDRTBL(i), 0); +@@ -1664,7 +1604,7 @@ s32 txgbe_find_vlvf_slot(struct txgbe_hw *hw, u32 vlan, bool vlvf_bypass) + * slot we found during our search, else error. + */ + if (!first_empty_slot) +- DEBUGOUT("No space in VLVF.\n"); ++ DEBUGOUT("No space in VLVF."); + + return first_empty_slot ? first_empty_slot : TXGBE_ERR_NO_SPACE; + } +@@ -1685,8 +1625,6 @@ s32 txgbe_set_vfta(struct txgbe_hw *hw, u32 vlan, u32 vind, + u32 regidx, vfta_delta, vfta; + s32 err; + +- DEBUGFUNC("txgbe_set_vfta"); +- + if (vlan > 4095 || vind > 63) + return TXGBE_ERR_PARAM; + +@@ -1754,8 +1692,6 @@ s32 txgbe_set_vlvf(struct txgbe_hw *hw, u32 vlan, u32 vind, + u32 portctl; + s32 vlvf_index; + +- DEBUGFUNC("txgbe_set_vlvf"); +- + if (vlan > 4095 || vind > 63) + return TXGBE_ERR_PARAM; + +@@ -1835,8 +1771,6 @@ s32 txgbe_clear_vfta(struct txgbe_hw *hw) + { + u32 offset; + +- DEBUGFUNC("txgbe_clear_vfta"); +- + for (offset = 0; offset < hw->mac.vft_size; offset++) + wr32(hw, TXGBE_VLANTBL(offset), 0); + +@@ -1890,8 +1824,6 @@ s32 txgbe_check_mac_link(struct txgbe_hw *hw, u32 *speed, + u32 links_reg, links_orig; + u32 i; + +- DEBUGFUNC("txgbe_check_mac_link"); +- + /* If Crosstalk fix enabled do the sanity check of making sure + * the SFP+ cage is full. + */ +@@ -1922,7 +1854,7 @@ s32 txgbe_check_mac_link(struct txgbe_hw *hw, u32 *speed, + links_reg = rd32(hw, TXGBE_PORTSTAT); + + if (links_orig != links_reg) { +- DEBUGOUT("LINKS changed from %08X to %08X\n", ++ DEBUGOUT("LINKS changed from %08X to %08X", + links_orig, links_reg); + } + +@@ -1977,8 +1909,6 @@ s32 txgbe_get_wwn_prefix(struct txgbe_hw *hw, u16 *wwnn_prefix, + u16 offset, caps; + u16 alt_san_mac_blk_offset; + +- DEBUGFUNC("txgbe_get_wwn_prefix"); +- + /* clear output first */ + *wwnn_prefix = 0xFFFF; + *wwpn_prefix = 0xFFFF; +@@ -2068,8 +1998,6 @@ void txgbe_set_ethertype_anti_spoofing(struct txgbe_hw *hw, + **/ + s32 txgbe_get_device_caps(struct txgbe_hw *hw, u16 *device_caps) + { +- DEBUGFUNC("txgbe_get_device_caps"); +- + hw->rom.readw_sw(hw, TXGBE_DEVICE_CAPS, device_caps); + + return 0; +@@ -2191,8 +2119,6 @@ s32 txgbe_get_thermal_sensor_data(struct txgbe_hw *hw) + s64 tsv; + u32 ts_stat; + +- DEBUGFUNC("txgbe_get_thermal_sensor_data"); +- + /* Only support thermal sensors attached to physical port 0 */ + if (hw->bus.lan_id != 0) + return TXGBE_NOT_IMPLEMENTED; +@@ -2223,8 +2149,6 @@ s32 txgbe_init_thermal_sensor_thresh(struct txgbe_hw *hw) + { + struct txgbe_thermal_sensor_data *data = &hw->mac.thermal_sensor_data; + +- DEBUGFUNC("txgbe_init_thermal_sensor_thresh"); +- + memset(data, 0, sizeof(struct txgbe_thermal_sensor_data)); + + if (hw->bus.lan_id != 0) +@@ -2295,8 +2219,6 @@ s32 txgbe_setup_mac_link_multispeed_fiber(struct txgbe_hw *hw, + u32 i = 0; + bool autoneg, link_up = false; + +- DEBUGFUNC("txgbe_setup_mac_link_multispeed_fiber"); +- + /* Mask off requested but non-supported speeds */ + status = hw->mac.get_link_capabilities(hw, &link_speed, &autoneg); + if (status != 0) +@@ -2321,7 +2243,7 @@ s32 txgbe_setup_mac_link_multispeed_fiber(struct txgbe_hw *hw, + /* QSFP module automatically detects MAC link speed */ + break; + default: +- DEBUGOUT("Unexpected media type.\n"); ++ DEBUGOUT("Unexpected media type."); + break; + } + +@@ -2371,7 +2293,7 @@ s32 txgbe_setup_mac_link_multispeed_fiber(struct txgbe_hw *hw, + /* QSFP module automatically detects link speed */ + break; + default: +- DEBUGOUT("Unexpected media type.\n"); ++ DEBUGOUT("Unexpected media type."); + break; + } + +@@ -2437,8 +2359,6 @@ s32 txgbe_init_shared_code(struct txgbe_hw *hw) + { + s32 status; + +- DEBUGFUNC("txgbe_init_shared_code"); +- + /* + * Set the mac type + */ +@@ -2474,8 +2394,6 @@ s32 txgbe_set_mac_type(struct txgbe_hw *hw) + { + s32 err = 0; + +- DEBUGFUNC("txgbe_set_mac_type"); +- + if (hw->vendor_id != PCI_VENDOR_ID_WANGXUN) { + DEBUGOUT("Unsupported vendor id: %x", hw->vendor_id); + return TXGBE_ERR_DEVICE_NOT_SUPPORTED; +@@ -2497,7 +2415,7 @@ s32 txgbe_set_mac_type(struct txgbe_hw *hw) + break; + } + +- DEBUGOUT("found mac: %d, returns: %d\n", ++ DEBUGOUT("found mac: %d, returns: %d", + hw->mac.type, err); + return err; + } +@@ -2506,8 +2424,6 @@ void txgbe_init_mac_link_ops(struct txgbe_hw *hw) + { + struct txgbe_mac_info *mac = &hw->mac; + +- DEBUGFUNC("txgbe_init_mac_link_ops"); +- + /* + * enable the laser control functions for SFP+ fiber + * and MNG not enabled +@@ -2550,8 +2466,6 @@ s32 txgbe_init_phy_raptor(struct txgbe_hw *hw) + struct txgbe_phy_info *phy = &hw->phy; + s32 err = 0; + +- DEBUGFUNC("txgbe_init_phy_raptor"); +- + if ((hw->device_id & 0xFF) == TXGBE_DEV_ID_QSFP) { + /* Store flag indicating I2C bus access control unit. */ + hw->phy.qsfp_shared_i2c_bus = TRUE; +@@ -2598,8 +2512,6 @@ s32 txgbe_setup_sfp_modules(struct txgbe_hw *hw) + { + s32 err = 0; + +- DEBUGFUNC("txgbe_setup_sfp_modules"); +- + if (hw->phy.sfp_type == txgbe_sfp_type_unknown) + return 0; + +@@ -2619,7 +2531,7 @@ s32 txgbe_setup_sfp_modules(struct txgbe_hw *hw) + msec_delay(hw->rom.semaphore_delay); + + if (err) { +- DEBUGOUT("sfp module setup not complete\n"); ++ DEBUGOUT("sfp module setup not complete"); + return TXGBE_ERR_SFP_SETUP_NOT_COMPLETE; + } + +@@ -2717,8 +2629,6 @@ s32 txgbe_init_ops_pf(struct txgbe_hw *hw) + struct txgbe_rom_info *rom = &hw->rom; + struct txgbe_mbx_info *mbx = &hw->mbx; + +- DEBUGFUNC("txgbe_init_ops_pf"); +- + /* BUS */ + bus->set_lan_id = txgbe_set_lan_id_multi_port; + +@@ -2845,8 +2755,6 @@ s32 txgbe_get_link_capabilities_raptor(struct txgbe_hw *hw, + s32 status = 0; + u32 autoc = 0; + +- DEBUGFUNC("txgbe_get_link_capabilities_raptor"); +- + /* Check if 1G SFP module. */ + if (hw->phy.sfp_type == txgbe_sfp_type_1g_cu_core0 || + hw->phy.sfp_type == txgbe_sfp_type_1g_cu_core1 || +@@ -2950,8 +2858,6 @@ u32 txgbe_get_media_type_raptor(struct txgbe_hw *hw) + { + u32 media_type; + +- DEBUGFUNC("txgbe_get_media_type_raptor"); +- + if (hw->phy.ffe_set) + txgbe_bp_mode_set(hw); + +@@ -3010,8 +2916,6 @@ s32 txgbe_start_mac_link_raptor(struct txgbe_hw *hw, + s32 status = 0; + bool got_lock = false; + +- DEBUGFUNC("txgbe_start_mac_link_raptor"); +- + UNREFERENCED_PARAMETER(autoneg_wait_to_complete); + + /* reset_pipeline requires us to hold this lock as it writes to +@@ -3094,8 +2998,6 @@ void txgbe_enable_tx_laser_multispeed_fiber(struct txgbe_hw *hw) + **/ + void txgbe_flap_tx_laser_multispeed_fiber(struct txgbe_hw *hw) + { +- DEBUGFUNC("txgbe_flap_tx_laser_multispeed_fiber"); +- + /* Blocked by MNG FW so bail */ + if (txgbe_check_reset_blocked(hw)) + return; +@@ -3127,7 +3029,7 @@ void txgbe_set_hard_rate_select_speed(struct txgbe_hw *hw, + esdp_reg &= ~(TXGBE_GPIOBIT_4 | TXGBE_GPIOBIT_5); + break; + default: +- DEBUGOUT("Invalid fixed module speed\n"); ++ DEBUGOUT("Invalid fixed module speed"); + return; + } + +@@ -3153,8 +3055,6 @@ s32 txgbe_setup_mac_link_smartspeed(struct txgbe_hw *hw, + bool link_up = false; + u32 autoc_reg = rd32_epcs(hw, SR_AN_MMD_ADV_REG1); + +- DEBUGFUNC("txgbe_setup_mac_link_smartspeed"); +- + /* Set autoneg_advertised value based on input link speed */ + hw->phy.autoneg_advertised = 0; + +@@ -3243,8 +3143,7 @@ s32 txgbe_setup_mac_link_smartspeed(struct txgbe_hw *hw, + + out: + if (link_up && link_speed == TXGBE_LINK_SPEED_1GB_FULL) +- DEBUGOUT("Smartspeed has downgraded the link speed " +- "from the maximum advertised\n"); ++ DEBUGOUT("Smartspeed has downgraded the link speed from the maximum advertised"); + return status; + } + +@@ -3270,7 +3169,6 @@ s32 txgbe_setup_mac_link(struct txgbe_hw *hw, + u64 orig_autoc = 0; + u32 link_capabilities = TXGBE_LINK_SPEED_UNKNOWN; + +- DEBUGFUNC("txgbe_setup_mac_link"); + UNREFERENCED_PARAMETER(autoneg_wait_to_complete); + + /* Check to see if speed passed in is supported. */ +@@ -3357,8 +3255,6 @@ static s32 txgbe_setup_copper_link_raptor(struct txgbe_hw *hw, + { + s32 status; + +- DEBUGFUNC("txgbe_setup_copper_link_raptor"); +- + /* Setup the PHY according to input speed */ + status = hw->phy.setup_link_speed(hw, speed, + autoneg_wait_to_complete); +@@ -3467,8 +3363,6 @@ s32 txgbe_reset_hw(struct txgbe_hw *hw) + s32 status; + u32 autoc; + +- DEBUGFUNC("txgbe_reset_hw"); +- + /* Call adapter stop to disable tx/rx and clear interrupts */ + status = hw->mac.stop_hw(hw); + if (status != 0) +@@ -3624,15 +3518,13 @@ s32 txgbe_reinit_fdir_tables(struct txgbe_hw *hw) + u32 fdircmd; + fdirctrl &= ~TXGBE_FDIRCTL_INITDONE; + +- DEBUGFUNC("txgbe_reinit_fdir_tables"); +- + /* + * Before starting reinitialization process, + * FDIRPICMD.OP must be zero. + */ + err = txgbe_fdir_check_cmd_complete(hw, &fdircmd); + if (err) { +- DEBUGOUT("Flow Director previous command did not complete, aborting table re-initialization.\n"); ++ DEBUGOUT("Flow Director previous command did not complete, aborting table re-initialization."); + return err; + } + +@@ -3666,7 +3558,7 @@ s32 txgbe_reinit_fdir_tables(struct txgbe_hw *hw) + msec_delay(1); + } + if (i >= TXGBE_FDIR_INIT_DONE_POLL) { +- DEBUGOUT("Flow Director Signature poll time exceeded!\n"); ++ DEBUGOUT("Flow Director Signature poll time exceeded!"); + return TXGBE_ERR_FDIR_REINIT_FAILED; + } + +@@ -3692,8 +3584,6 @@ s32 txgbe_start_hw_raptor(struct txgbe_hw *hw) + { + s32 err = 0; + +- DEBUGFUNC("txgbe_start_hw_raptor"); +- + err = txgbe_start_hw(hw); + if (err != 0) + goto out; +@@ -3718,8 +3608,6 @@ s32 txgbe_start_hw_raptor(struct txgbe_hw *hw) + **/ + s32 txgbe_enable_rx_dma_raptor(struct txgbe_hw *hw, u32 regval) + { +- DEBUGFUNC("txgbe_enable_rx_dma_raptor"); +- + /* + * Workaround silicon errata when enabling the Rx datapath. + * If traffic is incoming before we enable the Rx unit, it could hang +@@ -3752,8 +3640,6 @@ bool txgbe_verify_lesm_fw_enabled_raptor(struct txgbe_hw *hw) + u16 fw_offset, fw_lesm_param_offset, fw_lesm_state; + s32 status; + +- DEBUGFUNC("txgbe_verify_lesm_fw_enabled_raptor"); +- + /* get the offset to the Firmware Module block */ + status = hw->rom.read16(hw, TXGBE_FW_PTR, &fw_offset); + +diff --git a/dpdk/drivers/net/txgbe/base/txgbe_mbx.c b/dpdk/drivers/net/txgbe/base/txgbe_mbx.c +index 4d64c6c3e9..7f2489a13f 100644 +--- a/dpdk/drivers/net/txgbe/base/txgbe_mbx.c ++++ b/dpdk/drivers/net/txgbe/base/txgbe_mbx.c +@@ -21,8 +21,6 @@ s32 txgbe_read_mbx(struct txgbe_hw *hw, u32 *msg, u16 size, u16 mbx_id) + struct txgbe_mbx_info *mbx = &hw->mbx; + s32 ret_val = TXGBE_ERR_MBX; + +- DEBUGFUNC("txgbe_read_mbx"); +- + /* limit read to size of mailbox */ + if (size > mbx->size) + size = mbx->size; +@@ -47,8 +45,6 @@ s32 txgbe_write_mbx(struct txgbe_hw *hw, u32 *msg, u16 size, u16 mbx_id) + struct txgbe_mbx_info *mbx = &hw->mbx; + s32 ret_val = 0; + +- DEBUGFUNC("txgbe_write_mbx"); +- + if (size > mbx->size) { + ret_val = TXGBE_ERR_MBX; + DEBUGOUT("Invalid mailbox message size %d", size); +@@ -71,8 +67,6 @@ s32 txgbe_check_for_msg(struct txgbe_hw *hw, u16 mbx_id) + struct txgbe_mbx_info *mbx = &hw->mbx; + s32 ret_val = TXGBE_ERR_MBX; + +- DEBUGFUNC("txgbe_check_for_msg"); +- + if (mbx->check_for_msg) + ret_val = mbx->check_for_msg(hw, mbx_id); + +@@ -91,8 +85,6 @@ s32 txgbe_check_for_ack(struct txgbe_hw *hw, u16 mbx_id) + struct txgbe_mbx_info *mbx = &hw->mbx; + s32 ret_val = TXGBE_ERR_MBX; + +- DEBUGFUNC("txgbe_check_for_ack"); +- + if (mbx->check_for_ack) + ret_val = mbx->check_for_ack(hw, mbx_id); + +@@ -111,8 +103,6 @@ s32 txgbe_check_for_rst(struct txgbe_hw *hw, u16 mbx_id) + struct txgbe_mbx_info *mbx = &hw->mbx; + s32 ret_val = TXGBE_ERR_MBX; + +- DEBUGFUNC("txgbe_check_for_rst"); +- + if (mbx->check_for_rst) + ret_val = mbx->check_for_rst(hw, mbx_id); + +@@ -131,8 +121,6 @@ STATIC s32 txgbe_poll_for_msg(struct txgbe_hw *hw, u16 mbx_id) + struct txgbe_mbx_info *mbx = &hw->mbx; + int countdown = mbx->timeout; + +- DEBUGFUNC("txgbe_poll_for_msg"); +- + if (!countdown || !mbx->check_for_msg) + goto out; + +@@ -162,8 +150,6 @@ STATIC s32 txgbe_poll_for_ack(struct txgbe_hw *hw, u16 mbx_id) + struct txgbe_mbx_info *mbx = &hw->mbx; + int countdown = mbx->timeout; + +- DEBUGFUNC("txgbe_poll_for_ack"); +- + if (!countdown || !mbx->check_for_ack) + goto out; + +@@ -196,8 +182,6 @@ s32 txgbe_read_posted_mbx(struct txgbe_hw *hw, u32 *msg, u16 size, u16 mbx_id) + struct txgbe_mbx_info *mbx = &hw->mbx; + s32 ret_val = TXGBE_ERR_MBX; + +- DEBUGFUNC("txgbe_read_posted_mbx"); +- + if (!mbx->read) + goto out; + +@@ -226,8 +210,6 @@ s32 txgbe_write_posted_mbx(struct txgbe_hw *hw, u32 *msg, u16 size, + struct txgbe_mbx_info *mbx = &hw->mbx; + s32 ret_val = TXGBE_ERR_MBX; + +- DEBUGFUNC("txgbe_write_posted_mbx"); +- + /* exit if either we can't write or there isn't a defined timeout */ + if (!mbx->write || !mbx->timeout) + goto out; +@@ -292,7 +274,6 @@ s32 txgbe_check_for_msg_vf(struct txgbe_hw *hw, u16 mbx_id) + s32 ret_val = TXGBE_ERR_MBX; + + UNREFERENCED_PARAMETER(mbx_id); +- DEBUGFUNC("txgbe_check_for_msg_vf"); + + if (!txgbe_check_for_bit_vf(hw, TXGBE_VFMBCTL_PFSTS)) { + ret_val = 0; +@@ -314,7 +295,6 @@ s32 txgbe_check_for_ack_vf(struct txgbe_hw *hw, u16 mbx_id) + s32 ret_val = TXGBE_ERR_MBX; + + UNREFERENCED_PARAMETER(mbx_id); +- DEBUGFUNC("txgbe_check_for_ack_vf"); + + if (!txgbe_check_for_bit_vf(hw, TXGBE_VFMBCTL_PFACK)) { + ret_val = 0; +@@ -336,7 +316,6 @@ s32 txgbe_check_for_rst_vf(struct txgbe_hw *hw, u16 mbx_id) + s32 ret_val = TXGBE_ERR_MBX; + + UNREFERENCED_PARAMETER(mbx_id); +- DEBUGFUNC("txgbe_check_for_rst_vf"); + + if (!txgbe_check_for_bit_vf(hw, (TXGBE_VFMBCTL_RSTD | + TXGBE_VFMBCTL_RSTI))) { +@@ -357,8 +336,6 @@ STATIC s32 txgbe_obtain_mbx_lock_vf(struct txgbe_hw *hw) + { + s32 ret_val = TXGBE_ERR_MBX; + +- DEBUGFUNC("txgbe_obtain_mbx_lock_vf"); +- + /* Take ownership of the buffer */ + wr32(hw, TXGBE_VFMBCTL, TXGBE_VFMBCTL_VFU); + +@@ -386,8 +363,6 @@ s32 txgbe_write_mbx_vf(struct txgbe_hw *hw, u32 *msg, u16 size, + + UNREFERENCED_PARAMETER(mbx_id); + +- DEBUGFUNC("txgbe_write_mbx_vf"); +- + /* lock the mailbox to prevent pf/vf race condition */ + ret_val = txgbe_obtain_mbx_lock_vf(hw); + if (ret_val) +@@ -426,7 +401,6 @@ s32 txgbe_read_mbx_vf(struct txgbe_hw *hw, u32 *msg, u16 size, + s32 ret_val = 0; + u16 i; + +- DEBUGFUNC("txgbe_read_mbx_vf"); + UNREFERENCED_PARAMETER(mbx_id); + + /* lock the mailbox to prevent pf/vf race condition */ +@@ -499,8 +473,6 @@ s32 txgbe_check_for_msg_pf(struct txgbe_hw *hw, u16 vf_number) + s32 index = TXGBE_MBVFICR_INDEX(vf_number); + u32 vf_bit = vf_number % 16; + +- DEBUGFUNC("txgbe_check_for_msg_pf"); +- + if (!txgbe_check_for_bit_pf(hw, TXGBE_MBVFICR_VFREQ_VF1 << vf_bit, + index)) { + ret_val = 0; +@@ -523,8 +495,6 @@ s32 txgbe_check_for_ack_pf(struct txgbe_hw *hw, u16 vf_number) + s32 index = TXGBE_MBVFICR_INDEX(vf_number); + u32 vf_bit = vf_number % 16; + +- DEBUGFUNC("txgbe_check_for_ack_pf"); +- + if (!txgbe_check_for_bit_pf(hw, TXGBE_MBVFICR_VFACK_VF1 << vf_bit, + index)) { + ret_val = 0; +@@ -548,8 +518,6 @@ s32 txgbe_check_for_rst_pf(struct txgbe_hw *hw, u16 vf_number) + u32 vflre = 0; + s32 ret_val = TXGBE_ERR_MBX; + +- DEBUGFUNC("txgbe_check_for_rst_pf"); +- + vflre = rd32(hw, TXGBE_FLRVFE(reg_offset)); + if (vflre & (1 << vf_shift)) { + ret_val = 0; +@@ -572,8 +540,6 @@ STATIC s32 txgbe_obtain_mbx_lock_pf(struct txgbe_hw *hw, u16 vf_number) + s32 ret_val = TXGBE_ERR_MBX; + u32 p2v_mailbox; + +- DEBUGFUNC("txgbe_obtain_mbx_lock_pf"); +- + /* Take ownership of the buffer */ + wr32(hw, TXGBE_MBCTL(vf_number), TXGBE_MBCTL_PFU); + +@@ -602,8 +568,6 @@ s32 txgbe_write_mbx_pf(struct txgbe_hw *hw, u32 *msg, u16 size, u16 vf_number) + s32 ret_val; + u16 i; + +- DEBUGFUNC("txgbe_write_mbx_pf"); +- + /* lock the mailbox to prevent pf/vf race condition */ + ret_val = txgbe_obtain_mbx_lock_pf(hw, vf_number); + if (ret_val) +@@ -643,8 +607,6 @@ s32 txgbe_read_mbx_pf(struct txgbe_hw *hw, u32 *msg, u16 size, u16 vf_number) + s32 ret_val; + u16 i; + +- DEBUGFUNC("txgbe_read_mbx_pf"); +- + /* lock the mailbox to prevent pf/vf race condition */ + ret_val = txgbe_obtain_mbx_lock_pf(hw, vf_number); + if (ret_val) +diff --git a/dpdk/drivers/net/txgbe/base/txgbe_mng.c b/dpdk/drivers/net/txgbe/base/txgbe_mng.c +index dbe512122c..045a2f5de0 100644 +--- a/dpdk/drivers/net/txgbe/base/txgbe_mng.c ++++ b/dpdk/drivers/net/txgbe/base/txgbe_mng.c +@@ -45,10 +45,8 @@ txgbe_hic_unlocked(struct txgbe_hw *hw, u32 *buffer, u32 length, u32 timeout) + u32 value, loop; + u16 i, dword_len; + +- DEBUGFUNC("txgbe_hic_unlocked"); +- + if (!length || length > TXGBE_PMMBX_BSIZE) { +- DEBUGOUT("Buffer length failure buffersize=%d.\n", length); ++ DEBUGOUT("Buffer length failure buffersize=%d.", length); + return TXGBE_ERR_HOST_INTERFACE_COMMAND; + } + +@@ -78,7 +76,7 @@ txgbe_hic_unlocked(struct txgbe_hw *hw, u32 *buffer, u32 length, u32 timeout) + TXGBE_MNGMBXCTL_FWRDY, TXGBE_MNGMBXCTL_FWRDY, + &value, timeout, 1000); + if (!loop || !(value & TXGBE_MNGMBXCTL_FWACK)) { +- DEBUGOUT("Command has failed with no status valid.\n"); ++ DEBUGOUT("Command has failed with no status valid."); + return TXGBE_ERR_HOST_INTERFACE_COMMAND; + } + +@@ -114,10 +112,8 @@ txgbe_host_interface_command(struct txgbe_hw *hw, u32 *buffer, + u32 bi; + u32 dword_len; + +- DEBUGFUNC("txgbe_host_interface_command"); +- + if (length == 0 || length > TXGBE_PMMBX_BSIZE) { +- DEBUGOUT("Buffer length failure buffersize=%d.\n", length); ++ DEBUGOUT("Buffer length failure buffersize=%d.", length); + return TXGBE_ERR_HOST_INTERFACE_COMMAND; + } + +@@ -159,7 +155,7 @@ txgbe_host_interface_command(struct txgbe_hw *hw, u32 *buffer, + goto rel_out; + + if (length < buf_len + hdr_size) { +- DEBUGOUT("Buffer not large enough for reply message.\n"); ++ DEBUGOUT("Buffer not large enough for reply message."); + err = TXGBE_ERR_HOST_INTERFACE_COMMAND; + goto rel_out; + } +@@ -285,7 +281,6 @@ s32 txgbe_hic_set_drv_ver(struct txgbe_hw *hw, u8 maj, u8 min, + int i; + s32 ret_val = 0; + +- DEBUGFUNC("txgbe_hic_set_drv_ver"); + UNREFERENCED_PARAMETER(len, driver_ver); + + fw_cmd.hdr.cmd = FW_CEM_CMD_DRIVER_INFO; +@@ -338,8 +333,6 @@ txgbe_hic_reset(struct txgbe_hw *hw) + int i; + s32 err = 0; + +- DEBUGFUNC("\n"); +- + reset_cmd.hdr.cmd = FW_RESET_CMD; + reset_cmd.hdr.buf_len = FW_RESET_LEN; + reset_cmd.hdr.cmd_or_resp.cmd_resv = FW_CEM_CMD_RESERVED; +diff --git a/dpdk/drivers/net/txgbe/base/txgbe_osdep.h b/dpdk/drivers/net/txgbe/base/txgbe_osdep.h +index 11fcf7e8fe..b62c0b0824 100644 +--- a/dpdk/drivers/net/txgbe/base/txgbe_osdep.h ++++ b/dpdk/drivers/net/txgbe/base/txgbe_osdep.h +@@ -18,6 +18,7 @@ + #include + #include + #include ++#include + + #include "../txgbe_logs.h" + +diff --git a/dpdk/drivers/net/txgbe/base/txgbe_phy.c b/dpdk/drivers/net/txgbe/base/txgbe_phy.c +index 3f5229ecc2..9f46d5bdb0 100644 +--- a/dpdk/drivers/net/txgbe/base/txgbe_phy.c ++++ b/dpdk/drivers/net/txgbe/base/txgbe_phy.c +@@ -35,7 +35,7 @@ static bool txgbe_identify_extphy(struct txgbe_hw *hw) + u16 phy_addr = 0; + + if (!txgbe_validate_phy_addr(hw, phy_addr)) { +- DEBUGOUT("Unable to validate PHY address 0x%04X\n", ++ DEBUGOUT("Unable to validate PHY address 0x%04X", + phy_addr); + return false; + } +@@ -100,8 +100,6 @@ s32 txgbe_identify_phy(struct txgbe_hw *hw) + { + s32 err = TXGBE_ERR_PHY_ADDR_INVALID; + +- DEBUGFUNC("txgbe_identify_phy"); +- + txgbe_read_phy_if(hw); + + if (hw->phy.type != txgbe_phy_unknown) +@@ -137,11 +135,9 @@ s32 txgbe_check_reset_blocked(struct txgbe_hw *hw) + { + u32 mmngc; + +- DEBUGFUNC("txgbe_check_reset_blocked"); +- + mmngc = rd32(hw, TXGBE_STAT); + if (mmngc & TXGBE_STAT_MNGVETO) { +- DEBUGOUT("MNG_VETO bit detected.\n"); ++ DEBUGOUT("MNG_VETO bit detected."); + return true; + } + +@@ -159,8 +155,6 @@ bool txgbe_validate_phy_addr(struct txgbe_hw *hw, u32 phy_addr) + u16 phy_id = 0; + bool valid = false; + +- DEBUGFUNC("txgbe_validate_phy_addr"); +- + hw->phy.addr = phy_addr; + hw->phy.read_reg(hw, TXGBE_MD_PHY_ID_HIGH, + TXGBE_MD_DEV_PMA_PMD, &phy_id); +@@ -168,7 +162,7 @@ bool txgbe_validate_phy_addr(struct txgbe_hw *hw, u32 phy_addr) + if (phy_id != 0xFFFF && phy_id != 0x0) + valid = true; + +- DEBUGOUT("PHY ID HIGH is 0x%04X\n", phy_id); ++ DEBUGOUT("PHY ID HIGH is 0x%04X", phy_id); + + return valid; + } +@@ -184,8 +178,6 @@ s32 txgbe_get_phy_id(struct txgbe_hw *hw) + u16 phy_id_high = 0; + u16 phy_id_low = 0; + +- DEBUGFUNC("txgbe_get_phy_id"); +- + err = hw->phy.read_reg(hw, TXGBE_MD_PHY_ID_HIGH, + TXGBE_MD_DEV_PMA_PMD, + &phy_id_high); +@@ -198,7 +190,7 @@ s32 txgbe_get_phy_id(struct txgbe_hw *hw) + hw->phy.id |= (u32)(phy_id_low & TXGBE_PHY_REVISION_MASK); + hw->phy.revision = (u32)(phy_id_low & ~TXGBE_PHY_REVISION_MASK); + } +- DEBUGOUT("PHY_ID_HIGH 0x%04X, PHY_ID_LOW 0x%04X\n", ++ DEBUGOUT("PHY_ID_HIGH 0x%04X, PHY_ID_LOW 0x%04X", + phy_id_high, phy_id_low); + + return err; +@@ -213,8 +205,6 @@ enum txgbe_phy_type txgbe_get_phy_type_from_id(u32 phy_id) + { + enum txgbe_phy_type phy_type; + +- DEBUGFUNC("txgbe_get_phy_type_from_id"); +- + switch (phy_id) { + case TXGBE_PHYID_TN1010: + phy_type = txgbe_phy_tn; +@@ -272,7 +262,7 @@ txgbe_reset_extphy(struct txgbe_hw *hw) + + if (ctrl & TXGBE_MD_PORT_CTRL_RESET) { + err = TXGBE_ERR_RESET_FAILED; +- DEBUGOUT("PHY reset polling failed to complete.\n"); ++ DEBUGOUT("PHY reset polling failed to complete."); + } + + return err; +@@ -286,8 +276,6 @@ s32 txgbe_reset_phy(struct txgbe_hw *hw) + { + s32 err = 0; + +- DEBUGFUNC("txgbe_reset_phy"); +- + if (hw->phy.type == txgbe_phy_unknown) + err = txgbe_identify_phy(hw); + +@@ -343,7 +331,7 @@ s32 txgbe_read_phy_reg_mdi(struct txgbe_hw *hw, u32 reg_addr, u32 device_type, + */ + if (!po32m(hw, TXGBE_MDIOSCD, TXGBE_MDIOSCD_BUSY, + 0, NULL, 100, 100)) { +- DEBUGOUT("PHY address command did not complete\n"); ++ DEBUGOUT("PHY address command did not complete"); + return TXGBE_ERR_PHY; + } + +@@ -367,8 +355,6 @@ s32 txgbe_read_phy_reg(struct txgbe_hw *hw, u32 reg_addr, + s32 err; + u32 gssr = hw->phy.phy_semaphore_mask; + +- DEBUGFUNC("txgbe_read_phy_reg"); +- + if (hw->mac.acquire_swfw_sync(hw, gssr)) + return TXGBE_ERR_SWFW_SYNC; + +@@ -406,7 +392,7 @@ s32 txgbe_write_phy_reg_mdi(struct txgbe_hw *hw, u32 reg_addr, + /* wait for completion */ + if (!po32m(hw, TXGBE_MDIOSCD, TXGBE_MDIOSCD_BUSY, + 0, NULL, 100, 100)) { +- TLOG_DEBUG("PHY write cmd didn't complete\n"); ++ DEBUGOUT("PHY write cmd didn't complete"); + return -TERR_PHY; + } + +@@ -427,8 +413,6 @@ s32 txgbe_write_phy_reg(struct txgbe_hw *hw, u32 reg_addr, + s32 err; + u32 gssr = hw->phy.phy_semaphore_mask; + +- DEBUGFUNC("txgbe_write_phy_reg"); +- + if (hw->mac.acquire_swfw_sync(hw, gssr)) + err = TXGBE_ERR_SWFW_SYNC; + +@@ -452,8 +436,6 @@ s32 txgbe_setup_phy_link(struct txgbe_hw *hw) + bool autoneg = false; + u32 speed; + +- DEBUGFUNC("txgbe_setup_phy_link"); +- + txgbe_get_copper_link_capabilities(hw, &speed, &autoneg); + + /* Set or unset auto-negotiation 10G advertisement */ +@@ -539,8 +521,6 @@ s32 txgbe_setup_phy_link_speed(struct txgbe_hw *hw, + { + UNREFERENCED_PARAMETER(autoneg_wait_to_complete); + +- DEBUGFUNC("txgbe_setup_phy_link_speed"); +- + /* + * Clear autoneg_advertised and set new values based on input link + * speed. +@@ -623,8 +603,6 @@ s32 txgbe_get_copper_link_capabilities(struct txgbe_hw *hw, + { + s32 err = 0; + +- DEBUGFUNC("txgbe_get_copper_link_capabilities"); +- + *autoneg = true; + if (!hw->phy.speeds_supported) + err = txgbe_get_copper_speeds_supported(hw); +@@ -652,8 +630,6 @@ s32 txgbe_check_phy_link_tnx(struct txgbe_hw *hw, u32 *speed, + u16 phy_speed = 0; + u16 phy_data = 0; + +- DEBUGFUNC("txgbe_check_phy_link_tnx"); +- + /* Initialize speed and link to default case */ + *link_up = false; + *speed = TXGBE_LINK_SPEED_10GB_FULL; +@@ -697,8 +673,6 @@ s32 txgbe_setup_phy_link_tnx(struct txgbe_hw *hw) + bool autoneg = false; + u32 speed; + +- DEBUGFUNC("txgbe_setup_phy_link_tnx"); +- + txgbe_get_copper_link_capabilities(hw, &speed, &autoneg); + + if (speed & TXGBE_LINK_SPEED_10GB_FULL) { +@@ -772,8 +746,6 @@ s32 txgbe_identify_module(struct txgbe_hw *hw) + { + s32 err = TXGBE_ERR_SFP_NOT_PRESENT; + +- DEBUGFUNC("txgbe_identify_module"); +- + switch (hw->phy.media_type) { + case txgbe_media_type_fiber: + err = txgbe_identify_sfp_module(hw); +@@ -811,8 +783,6 @@ s32 txgbe_identify_sfp_module(struct txgbe_hw *hw) + u8 cable_spec = 0; + u16 enforce_sfp = 0; + +- DEBUGFUNC("txgbe_identify_sfp_module"); +- + if (hw->phy.media_type != txgbe_media_type_fiber) { + hw->phy.sfp_type = txgbe_sfp_type_not_present; + return TXGBE_ERR_SFP_NOT_PRESENT; +@@ -992,7 +962,7 @@ s32 txgbe_identify_sfp_module(struct txgbe_hw *hw) + hw->phy.sfp_type == txgbe_sfp_type_1g_lx_core1 || + hw->phy.sfp_type == txgbe_sfp_type_1g_sx_core0 || + hw->phy.sfp_type == txgbe_sfp_type_1g_sx_core1)) { +- DEBUGOUT("SFP+ module not supported\n"); ++ DEBUGOUT("SFP+ module not supported"); + hw->phy.type = txgbe_phy_sfp_unsupported; + return TXGBE_ERR_SFP_NOT_SUPPORTED; + } +@@ -1021,8 +991,6 @@ s32 txgbe_identify_qsfp_module(struct txgbe_hw *hw) + u8 device_tech = 0; + bool active_cable = false; + +- DEBUGFUNC("txgbe_identify_qsfp_module"); +- + if (hw->phy.media_type != txgbe_media_type_fiber_qsfp) { + hw->phy.sfp_type = txgbe_sfp_type_not_present; + err = TXGBE_ERR_SFP_NOT_PRESENT; +@@ -1165,10 +1133,10 @@ s32 txgbe_identify_qsfp_module(struct txgbe_hw *hw) + if (hw->allow_unsupported_sfp) { + DEBUGOUT("WARNING: Wangxun (R) Network Connections are quality tested using Wangxun (R) Ethernet Optics. " + "Using untested modules is not supported and may cause unstable operation or damage to the module or the adapter. " +- "Wangxun Corporation is not responsible for any harm caused by using untested modules.\n"); ++ "Wangxun Corporation is not responsible for any harm caused by using untested modules."); + err = 0; + } else { +- DEBUGOUT("QSFP module not supported\n"); ++ DEBUGOUT("QSFP module not supported"); + hw->phy.type = + txgbe_phy_sfp_unsupported; + err = TXGBE_ERR_SFP_NOT_SUPPORTED; +@@ -1194,8 +1162,6 @@ s32 txgbe_identify_qsfp_module(struct txgbe_hw *hw) + s32 txgbe_read_i2c_eeprom(struct txgbe_hw *hw, u8 byte_offset, + u8 *eeprom_data) + { +- DEBUGFUNC("txgbe_read_i2c_eeprom"); +- + return hw->phy.read_i2c_byte(hw, byte_offset, + TXGBE_I2C_EEPROM_DEV_ADDR, + eeprom_data); +@@ -1228,8 +1194,6 @@ s32 txgbe_read_i2c_sff8472(struct txgbe_hw *hw, u8 byte_offset, + s32 txgbe_write_i2c_eeprom(struct txgbe_hw *hw, u8 byte_offset, + u8 eeprom_data) + { +- DEBUGFUNC("txgbe_write_i2c_eeprom"); +- + return hw->phy.write_i2c_byte(hw, byte_offset, + TXGBE_I2C_EEPROM_DEV_ADDR, + eeprom_data); +@@ -1248,8 +1212,6 @@ s32 txgbe_write_i2c_eeprom(struct txgbe_hw *hw, u8 byte_offset, + s32 txgbe_read_i2c_byte_unlocked(struct txgbe_hw *hw, u8 byte_offset, + u8 dev_addr, u8 *data) + { +- DEBUGFUNC("txgbe_read_i2c_byte"); +- + txgbe_i2c_start(hw, dev_addr); + + /* wait tx empty */ +@@ -1312,8 +1274,6 @@ s32 txgbe_read_i2c_byte(struct txgbe_hw *hw, u8 byte_offset, + s32 txgbe_write_i2c_byte_unlocked(struct txgbe_hw *hw, u8 byte_offset, + u8 dev_addr, u8 data) + { +- DEBUGFUNC("txgbe_write_i2c_byte"); +- + txgbe_i2c_start(hw, dev_addr); + + /* wait tx empty */ +@@ -1367,8 +1327,6 @@ s32 txgbe_write_i2c_byte(struct txgbe_hw *hw, u8 byte_offset, + **/ + static void txgbe_i2c_start(struct txgbe_hw *hw, u8 dev_addr) + { +- DEBUGFUNC("txgbe_i2c_start"); +- + wr32(hw, TXGBE_I2CENA, 0); + + wr32(hw, TXGBE_I2CCON, +@@ -1396,12 +1354,10 @@ static void txgbe_i2c_start(struct txgbe_hw *hw, u8 dev_addr) + **/ + static void txgbe_i2c_stop(struct txgbe_hw *hw) + { +- DEBUGFUNC("txgbe_i2c_stop"); +- + /* wait for completion */ + if (!po32m(hw, TXGBE_I2CSTAT, TXGBE_I2CSTAT_MST, + 0, NULL, 100, 100)) { +- DEBUGFUNC("i2c stop timeout."); ++ DEBUGOUT("i2c stop timeout."); + } + + wr32(hw, TXGBE_I2CENA, 0); +@@ -1411,9 +1367,17 @@ static void + txgbe_set_sgmii_an37_ability(struct txgbe_hw *hw) + { + u32 value; ++ u8 device_type = hw->subsystem_device_id & 0xF0; + + wr32_epcs(hw, VR_XS_OR_PCS_MMD_DIGI_CTL1, 0x3002); +- wr32_epcs(hw, SR_MII_MMD_AN_CTL, 0x0105); ++ /* for sgmii + external phy, set to 0x0105 (phy sgmii mode) */ ++ /* for sgmii direct link, set to 0x010c (mac sgmii mode) */ ++ if (device_type == TXGBE_DEV_ID_MAC_SGMII || ++ hw->phy.media_type == txgbe_media_type_fiber) ++ wr32_epcs(hw, SR_MII_MMD_AN_CTL, 0x010C); ++ else if (device_type == TXGBE_DEV_ID_SGMII || ++ device_type == TXGBE_DEV_ID_XAUI) ++ wr32_epcs(hw, SR_MII_MMD_AN_CTL, 0x0105); + wr32_epcs(hw, SR_MII_MMD_DIGI_CTL, 0x0200); + value = rd32_epcs(hw, SR_MII_MMD_CTL); + value = (value & ~0x1200) | (0x1 << 12) | (0x1 << 9); +@@ -1455,6 +1419,10 @@ txgbe_set_link_to_kr(struct txgbe_hw *hw, bool autoneg) + if (!(hw->devarg.auto_neg == 1)) { + wr32_epcs(hw, SR_AN_CTRL, 0); + wr32_epcs(hw, VR_AN_KR_MODE_CL, 0); ++ } else { ++ value = rd32_epcs(hw, TXGBE_PHY_TX_EQ_CTL1); ++ value &= ~(1 << 6); ++ wr32_epcs(hw, TXGBE_PHY_TX_EQ_CTL1, value); + } + if (hw->devarg.present == 1) { + value = rd32_epcs(hw, TXGBE_PHY_TX_EQ_CTL1); +@@ -2320,6 +2288,8 @@ void txgbe_autoc_write(struct txgbe_hw *hw, u64 autoc) + } + } else if (hw->phy.media_type == txgbe_media_type_fiber) { + txgbe_set_link_to_sfi(hw, speed); ++ if (speed == TXGBE_LINK_SPEED_1GB_FULL) ++ txgbe_set_sgmii_an37_ability(hw); + } + + if (speed == TXGBE_LINK_SPEED_10GB_FULL) +@@ -2416,8 +2386,6 @@ s32 txgbe_kr_handle(struct txgbe_hw *hw) + u32 value; + s32 status = 0; + +- DEBUGFUNC("txgbe_kr_handle"); +- + value = rd32_epcs(hw, VR_AN_INTR); + BP_LOG("AN INTERRUPT!! value: 0x%x\n", value); + if (!(value & VR_AN_INTR_PG_RCV)) { +@@ -2441,8 +2409,6 @@ static s32 txgbe_handle_bp_flow(u32 link_mode, struct txgbe_hw *hw) + s32 status = 0; + struct txgbe_backplane_ability local_ability, lp_ability; + +- DEBUGFUNC("txgbe_handle_bp_flow"); +- + local_ability.current_link_mode = link_mode; + + /* 1. Get the local AN73 Base Page Ability */ +@@ -2544,8 +2510,6 @@ static void txgbe_get_bp_ability(struct txgbe_backplane_ability *ability, + { + u32 value = 0; + +- DEBUGFUNC("txgbe_get_bp_ability"); +- + /* Link Partner Base Page */ + if (link_partner == 1) { + /* Read the link partner AN73 Base Page Ability Registers */ +@@ -2617,8 +2581,6 @@ static s32 txgbe_check_bp_ability(struct txgbe_backplane_ability *local_ability, + u32 com_link_abi; + s32 ret = 0; + +- DEBUGFUNC("txgbe_check_bp_ability"); +- + com_link_abi = local_ability->link_ability & lp_ability->link_ability; + BP_LOG("com_link_abi = 0x%x, local_ability = 0x%x, lp_ability = 0x%x\n", + com_link_abi, local_ability->link_ability, +@@ -2674,8 +2636,6 @@ static void txgbe_clear_bp_intr(u32 bit, u32 bit_high, struct txgbe_hw *hw) + { + u32 rdata = 0, wdata, i; + +- DEBUGFUNC("txgbe_clear_bp_intr"); +- + rdata = rd32_epcs(hw, VR_AN_INTR); + BP_LOG("[Before clear]Read VR AN MMD Interrupt Register: 0x%x\n", + rdata); +@@ -2700,8 +2660,6 @@ static s32 txgbe_enable_kr_training(struct txgbe_hw *hw) + s32 status = 0; + u32 value = 0; + +- DEBUGFUNC("txgbe_enable_kr_training"); +- + BP_LOG("Enable Clause 72 KR Training ...\n"); + + if (CL72_KRTR_PRBS_MODE_EN != 0xFFFF) { +@@ -2745,8 +2703,6 @@ static s32 txgbe_disable_kr_training(struct txgbe_hw *hw, s32 post, s32 mode) + { + s32 status = 0; + +- DEBUGFUNC("txgbe_disable_kr_training"); +- + BP_LOG("Disable Clause 72 KR Training ...\n"); + /* Read PHY Lane0 TX EQ before Clause 72 KR Training. */ + txgbe_read_phy_lane_tx_eq(0, hw, post, mode); +@@ -2763,8 +2719,6 @@ static s32 txgbe_check_kr_training(struct txgbe_hw *hw) + int i; + int times = hw->devarg.poll ? 35 : 20; + +- DEBUGFUNC("txgbe_check_kr_training"); +- + for (i = 0; i < times; i++) { + value = rd32_epcs(hw, SR_PMA_KR_LP_CEU); + BP_LOG("SR PMA MMD 10GBASE-KR LP Coefficient Update Register: 0x%x\n", +@@ -2822,8 +2776,6 @@ static void txgbe_read_phy_lane_tx_eq(u16 lane, struct txgbe_hw *hw, + u32 addr; + u32 tx_main_cursor, tx_pre_cursor, tx_post_cursor, lmain; + +- DEBUGFUNC("txgbe_read_phy_lane_tx_eq"); +- + addr = TXGBE_PHY_LANE0_TX_EQ_CTL1 | (lane << 8); + value = rd32_ephy(hw, addr); + BP_LOG("PHY LANE TX EQ Read Value: %x\n", lane); +diff --git a/dpdk/drivers/net/txgbe/base/txgbe_regs.h b/dpdk/drivers/net/txgbe/base/txgbe_regs.h +index 144047ba62..dc22ef53e3 100644 +--- a/dpdk/drivers/net/txgbe/base/txgbe_regs.h ++++ b/dpdk/drivers/net/txgbe/base/txgbe_regs.h +@@ -1862,8 +1862,13 @@ po32m(struct txgbe_hw *hw, u32 reg, u32 mask, u32 expect, u32 *actual, + } + + do { +- all |= rd32(hw, reg); +- value |= mask & all; ++ if (expect != 0) { ++ all |= rd32(hw, reg); ++ value |= mask & all; ++ } else { ++ all = rd32(hw, reg); ++ value = mask & all; ++ } + if (value == expect) + break; + +@@ -1896,7 +1901,7 @@ po32m(struct txgbe_hw *hw, u32 reg, u32 mask, u32 expect, u32 *actual, + + #define wr32w(hw, reg, val, mask, slice) do { \ + wr32((hw), reg, val); \ +- po32m((hw), reg, mask, mask, NULL, 5, slice); \ ++ po32m((hw), reg, mask, 0, NULL, 5, slice); \ + } while (0) + + #define TXGBE_XPCS_IDAADDR 0x13000 +diff --git a/dpdk/drivers/net/txgbe/base/txgbe_vf.c b/dpdk/drivers/net/txgbe/base/txgbe_vf.c +index fb6d6d90ea..a73502351e 100644 +--- a/dpdk/drivers/net/txgbe/base/txgbe_vf.c ++++ b/dpdk/drivers/net/txgbe/base/txgbe_vf.c +@@ -107,8 +107,6 @@ s32 txgbe_reset_hw_vf(struct txgbe_hw *hw) + u32 msgbuf[TXGBE_VF_PERMADDR_MSG_LEN]; + u8 *addr = (u8 *)(&msgbuf[1]); + +- DEBUGFUNC("txgbevf_reset_hw_vf"); +- + /* Call adapter stop to disable tx/rx and clear interrupts */ + hw->mac.stop_hw(hw); + +@@ -121,7 +119,7 @@ s32 txgbe_reset_hw_vf(struct txgbe_hw *hw) + mbx->write_posted(hw, msgbuf, 1, 0); + msec_delay(10); + +- DEBUGOUT("Issuing a function level reset to MAC\n"); ++ DEBUGOUT("Issuing a function level reset to MAC"); + wr32(hw, TXGBE_VFRST, TXGBE_VFRST_SET); + txgbe_flush(hw); + msec_delay(50); +@@ -238,7 +236,7 @@ STATIC s32 txgbe_mta_vector(struct txgbe_hw *hw, u8 *mc_addr) + vector = ((mc_addr[4]) | (((u16)mc_addr[5]) << 8)); + break; + default: /* Invalid mc_filter_type */ +- DEBUGOUT("MC filter type param set incorrectly\n"); ++ DEBUGOUT("MC filter type param set incorrectly"); + ASSERT(0); + break; + } +@@ -316,8 +314,6 @@ s32 txgbe_update_mc_addr_list_vf(struct txgbe_hw *hw, u8 *mc_addr_list, + + UNREFERENCED_PARAMETER(clear); + +- DEBUGFUNC("txgbe_update_mc_addr_list_vf"); +- + /* Each entry in the list uses 1 16 bit word. We have 30 + * 16 bit words available in our HW msg buffer (minus 1 for the + * msg type). That's 30 hash values if we pack 'em right. If +@@ -327,7 +323,7 @@ s32 txgbe_update_mc_addr_list_vf(struct txgbe_hw *hw, u8 *mc_addr_list, + * addresses except for in large enterprise network environments. + */ + +- DEBUGOUT("MC Addr Count = %d\n", mc_addr_count); ++ DEBUGOUT("MC Addr Count = %d", mc_addr_count); + + cnt = (mc_addr_count > 30) ? 30 : mc_addr_count; + msgbuf[0] = TXGBE_VF_SET_MULTICAST; +@@ -335,7 +331,7 @@ s32 txgbe_update_mc_addr_list_vf(struct txgbe_hw *hw, u8 *mc_addr_list, + + for (i = 0; i < cnt; i++) { + vector = txgbe_mta_vector(hw, next(hw, &mc_addr_list, &vmdq)); +- DEBUGOUT("Hash value = 0x%03X\n", vector); ++ DEBUGOUT("Hash value = 0x%03X", vector); + vector_list[i] = (u16)vector; + } + +diff --git a/dpdk/drivers/net/txgbe/txgbe_ethdev.c b/dpdk/drivers/net/txgbe/txgbe_ethdev.c +index 47d0e6ea40..49948e62bc 100644 +--- a/dpdk/drivers/net/txgbe/txgbe_ethdev.c ++++ b/dpdk/drivers/net/txgbe/txgbe_ethdev.c +@@ -376,7 +376,7 @@ txgbe_dev_queue_stats_mapping_set(struct rte_eth_dev *eth_dev, + if (hw->mac.type != txgbe_mac_raptor) + return -ENOSYS; + +- if (stat_idx & !QMAP_FIELD_RESERVED_BITS_MASK) ++ if (stat_idx & ~QMAP_FIELD_RESERVED_BITS_MASK) + return -EIO; + + PMD_INIT_LOG(DEBUG, "Setting port %d, %s queue_id %d to stat index %d", +@@ -1678,7 +1678,7 @@ txgbe_dev_start(struct rte_eth_dev *dev) + return -ENOMEM; + } + } +- /* confiugre msix for sleep until rx interrupt */ ++ /* configure msix for sleep until rx interrupt */ + txgbe_configure_msix(dev); + + /* initialize transmission unit */ +@@ -1937,6 +1937,7 @@ txgbe_dev_set_link_up(struct rte_eth_dev *dev) + } else { + /* Turn on the laser */ + hw->mac.enable_tx_laser(hw); ++ hw->dev_start = true; + txgbe_dev_link_update(dev, 0); + } + +@@ -1957,6 +1958,7 @@ txgbe_dev_set_link_down(struct rte_eth_dev *dev) + } else { + /* Turn off the laser */ + hw->mac.disable_tx_laser(hw); ++ hw->dev_start = false; + txgbe_dev_link_update(dev, 0); + } + +@@ -2034,6 +2036,7 @@ txgbe_dev_close(struct rte_eth_dev *dev) + + #ifdef RTE_LIB_SECURITY + rte_free(dev->security_ctx); ++ dev->security_ctx = NULL; + #endif + + return ret; +@@ -3682,7 +3685,7 @@ txgbe_set_ivar_map(struct txgbe_hw *hw, int8_t direction, + wr32(hw, TXGBE_IVARMISC, tmp); + } else { + /* rx or tx causes */ +- /* Workround for ICR lost */ ++ /* Workaround for ICR lost */ + idx = ((16 * (queue & 1)) + (8 * direction)); + tmp = rd32(hw, TXGBE_IVAR(queue >> 1)); + tmp &= ~(0xFF << idx); +@@ -4387,7 +4390,7 @@ txgbe_timesync_disable(struct rte_eth_dev *dev) + /* Disable L2 filtering of IEEE1588/802.1AS Ethernet frame types. */ + wr32(hw, TXGBE_ETFLT(TXGBE_ETF_ID_1588), 0); + +- /* Stop incrementating the System Time registers. */ ++ /* Stop incrementing the System Time registers. */ + wr32(hw, TXGBE_TSTIMEINC, 0); + + return 0; +diff --git a/dpdk/drivers/net/txgbe/txgbe_ethdev_vf.c b/dpdk/drivers/net/txgbe/txgbe_ethdev_vf.c +index 84b960b8f9..f52cd8bc19 100644 +--- a/dpdk/drivers/net/txgbe/txgbe_ethdev_vf.c ++++ b/dpdk/drivers/net/txgbe/txgbe_ethdev_vf.c +@@ -961,7 +961,7 @@ txgbevf_set_ivar_map(struct txgbe_hw *hw, int8_t direction, + wr32(hw, TXGBE_VFIVARMISC, tmp); + } else { + /* rx or tx cause */ +- /* Workround for ICR lost */ ++ /* Workaround for ICR lost */ + idx = ((16 * (queue & 1)) + (8 * direction)); + tmp = rd32(hw, TXGBE_VFIVAR(queue >> 1)); + tmp &= ~(0xFF << idx); +@@ -997,7 +997,7 @@ txgbevf_configure_msix(struct rte_eth_dev *dev) + /* Configure all RX queues of VF */ + for (q_idx = 0; q_idx < dev->data->nb_rx_queues; q_idx++) { + /* Force all queue use vector 0, +- * as TXGBE_VF_MAXMSIVECOTR = 1 ++ * as TXGBE_VF_MAXMSIVECTOR = 1 + */ + txgbevf_set_ivar_map(hw, 0, q_idx, vector_idx); + rte_intr_vec_list_index_set(intr_handle, q_idx, +@@ -1288,7 +1288,7 @@ txgbevf_dev_interrupt_get_status(struct rte_eth_dev *dev) + + /* only one misc vector supported - mailbox */ + eicr &= TXGBE_VFICR_MASK; +- /* Workround for ICR lost */ ++ /* Workaround for ICR lost */ + intr->flags |= TXGBE_FLAG_MAILBOX; + + /* To avoid compiler warnings set eicr to used. */ +diff --git a/dpdk/drivers/net/txgbe/txgbe_ipsec.c b/dpdk/drivers/net/txgbe/txgbe_ipsec.c +index 445733f3ba..3ca3d85ed5 100644 +--- a/dpdk/drivers/net/txgbe/txgbe_ipsec.c ++++ b/dpdk/drivers/net/txgbe/txgbe_ipsec.c +@@ -288,7 +288,7 @@ txgbe_crypto_remove_sa(struct rte_eth_dev *dev, + return -1; + } + +- /* Disable and clear Rx SPI and key table entryes*/ ++ /* Disable and clear Rx SPI and key table entries */ + reg_val = TXGBE_IPSRXIDX_WRITE | + TXGBE_IPSRXIDX_TB_SPI | (sa_index << 3); + wr32(hw, TXGBE_IPSRXSPI, 0); +diff --git a/dpdk/drivers/net/txgbe/txgbe_logs.h b/dpdk/drivers/net/txgbe/txgbe_logs.h +index 67e9bfb3af..74f49ab9ef 100644 +--- a/dpdk/drivers/net/txgbe/txgbe_logs.h ++++ b/dpdk/drivers/net/txgbe/txgbe_logs.h +@@ -48,11 +48,8 @@ extern int txgbe_logtype_tx_free; + #define PMD_TX_FREE_LOG(level, fmt, args...) do { } while (0) + #endif + +-#define TLOG_DEBUG(fmt, args...) PMD_DRV_LOG(DEBUG, fmt, ##args) +- +-#define DEBUGOUT(fmt, args...) TLOG_DEBUG(fmt, ##args) +-#define PMD_INIT_FUNC_TRACE() TLOG_DEBUG(" >>") +-#define DEBUGFUNC(fmt) TLOG_DEBUG(fmt) ++#define DEBUGOUT(fmt, args...) PMD_DRV_LOG(DEBUG, fmt, ##args) ++#define PMD_INIT_FUNC_TRACE() PMD_DRV_LOG(DEBUG, ">>") + + extern int txgbe_logtype_bp; + #define BP_LOG(fmt, args...) \ +diff --git a/dpdk/drivers/net/txgbe/txgbe_pf.c b/dpdk/drivers/net/txgbe/txgbe_pf.c +index 30be287330..0b82fb1a88 100644 +--- a/dpdk/drivers/net/txgbe/txgbe_pf.c ++++ b/dpdk/drivers/net/txgbe/txgbe_pf.c +@@ -108,7 +108,7 @@ int txgbe_pf_host_init(struct rte_eth_dev *eth_dev) + nb_queue = 4; + RTE_ETH_DEV_SRIOV(eth_dev).active = RTE_ETH_32_POOLS; + } else { +- nb_queue = 8; ++ nb_queue = 4; + RTE_ETH_DEV_SRIOV(eth_dev).active = RTE_ETH_16_POOLS; + } + +@@ -236,7 +236,7 @@ int txgbe_pf_host_configure(struct rte_eth_dev *eth_dev) + + wr32(hw, TXGBE_PSRCTL, TXGBE_PSRCTL_LBENA); + +- /* clear VMDq map to perment rar 0 */ ++ /* clear VMDq map to permanent rar 0 */ + hw->mac.clear_vmdq(hw, 0, BIT_MASK32); + + /* clear VMDq map to scan rar 127 */ +diff --git a/dpdk/drivers/net/vhost/rte_eth_vhost.c b/dpdk/drivers/net/vhost/rte_eth_vhost.c +index 070f0e6dfd..a280e788fb 100644 +--- a/dpdk/drivers/net/vhost/rte_eth_vhost.c ++++ b/dpdk/drivers/net/vhost/rte_eth_vhost.c +@@ -716,10 +716,11 @@ eth_vhost_install_intr(struct rte_eth_dev *dev) + } + + static void +-update_queuing_status(struct rte_eth_dev *dev) ++update_queuing_status(struct rte_eth_dev *dev, bool wait_queuing) + { + struct pmd_internal *internal = dev->data->dev_private; + struct vhost_queue *vq; ++ struct rte_vhost_vring_state *state; + unsigned int i; + int allow_queuing = 1; + +@@ -730,13 +731,18 @@ update_queuing_status(struct rte_eth_dev *dev) + rte_atomic32_read(&internal->dev_attached) == 0) + allow_queuing = 0; + ++ state = vring_states[dev->data->port_id]; ++ + /* Wait until rx/tx_pkt_burst stops accessing vhost device */ + for (i = 0; i < dev->data->nb_rx_queues; i++) { + vq = dev->data->rx_queues[i]; + if (vq == NULL) + continue; +- rte_atomic32_set(&vq->allow_queuing, allow_queuing); +- while (rte_atomic32_read(&vq->while_queuing)) ++ if (allow_queuing && state->cur[vq->virtqueue_id]) ++ rte_atomic32_set(&vq->allow_queuing, 1); ++ else ++ rte_atomic32_set(&vq->allow_queuing, 0); ++ while (wait_queuing && rte_atomic32_read(&vq->while_queuing)) + rte_pause(); + } + +@@ -744,8 +750,11 @@ update_queuing_status(struct rte_eth_dev *dev) + vq = dev->data->tx_queues[i]; + if (vq == NULL) + continue; +- rte_atomic32_set(&vq->allow_queuing, allow_queuing); +- while (rte_atomic32_read(&vq->while_queuing)) ++ if (allow_queuing && state->cur[vq->virtqueue_id]) ++ rte_atomic32_set(&vq->allow_queuing, 1); ++ else ++ rte_atomic32_set(&vq->allow_queuing, 0); ++ while (wait_queuing && rte_atomic32_read(&vq->while_queuing)) + rte_pause(); + } + } +@@ -827,7 +836,7 @@ new_device(int vid) + eth_dev->data->dev_link.link_status = RTE_ETH_LINK_UP; + + rte_atomic32_set(&internal->dev_attached, 1); +- update_queuing_status(eth_dev); ++ update_queuing_status(eth_dev, false); + + VHOST_LOG(INFO, "Vhost device %d created\n", vid); + +@@ -857,7 +866,7 @@ destroy_device(int vid) + internal = eth_dev->data->dev_private; + + rte_atomic32_set(&internal->dev_attached, 0); +- update_queuing_status(eth_dev); ++ update_queuing_status(eth_dev, true); + + eth_dev->data->dev_link.link_status = RTE_ETH_LINK_DOWN; + +@@ -967,6 +976,8 @@ vring_state_changed(int vid, uint16_t vring, int enable) + state->max_vring = RTE_MAX(vring, state->max_vring); + rte_spinlock_unlock(&state->lock); + ++ update_queuing_status(eth_dev, false); ++ + VHOST_LOG(INFO, "vring%u is %s\n", + vring, enable ? "enabled" : "disabled"); + +@@ -1152,7 +1163,7 @@ eth_dev_start(struct rte_eth_dev *eth_dev) + } + + rte_atomic32_set(&internal->started, 1); +- update_queuing_status(eth_dev); ++ update_queuing_status(eth_dev, false); + + return 0; + } +@@ -1164,7 +1175,7 @@ eth_dev_stop(struct rte_eth_dev *dev) + + dev->data->dev_started = 0; + rte_atomic32_set(&internal->started, 0); +- update_queuing_status(dev); ++ update_queuing_status(dev, true); + + return 0; + } +@@ -1643,11 +1654,11 @@ rte_pmd_vhost_probe(struct rte_vdev_device *dev) + &open_int, &tso); + if (ret < 0) + goto out_free; ++ } + +- if (tso == 0) { +- disable_flags |= (1ULL << VIRTIO_NET_F_HOST_TSO4); +- disable_flags |= (1ULL << VIRTIO_NET_F_HOST_TSO6); +- } ++ if (tso == 0) { ++ disable_flags |= (1ULL << VIRTIO_NET_F_HOST_TSO4); ++ disable_flags |= (1ULL << VIRTIO_NET_F_HOST_TSO6); + } + + if (rte_kvargs_count(kvlist, ETH_VHOST_LINEAR_BUF) == 1) { +diff --git a/dpdk/drivers/net/virtio/meson.build b/dpdk/drivers/net/virtio/meson.build +index 01a333ada2..d78b8278c6 100644 +--- a/dpdk/drivers/net/virtio/meson.build ++++ b/dpdk/drivers/net/virtio/meson.build +@@ -30,11 +30,11 @@ if arch_subdir == 'x86' + c_args: [cflags, '-mavx512f', '-mavx512bw', '-mavx512vl']) + objs += virtio_avx512_lib.extract_objects('virtio_rxtx_packed.c') + if (toolchain == 'gcc' and cc.version().version_compare('>=8.3.0')) +- cflags += '-DVHOST_GCC_UNROLL_PRAGMA' ++ cflags += '-DVIRTIO_GCC_UNROLL_PRAGMA' + elif (toolchain == 'clang' and cc.version().version_compare('>=3.7.0')) +- cflags += '-DVHOST_CLANG_UNROLL_PRAGMA' ++ cflags += '-DVIRTIO_CLANG_UNROLL_PRAGMA' + elif (toolchain == 'icc' and cc.version().version_compare('>=16.0.0')) +- cflags += '-DVHOST_ICC_UNROLL_PRAGMA' ++ cflags += '-DVIRTIO_ICC_UNROLL_PRAGMA' + endif + endif + endif +diff --git a/dpdk/drivers/net/virtio/virtio_ethdev.c b/dpdk/drivers/net/virtio/virtio_ethdev.c +index c2588369b2..d180162abd 100644 +--- a/dpdk/drivers/net/virtio/virtio_ethdev.c ++++ b/dpdk/drivers/net/virtio/virtio_ethdev.c +@@ -2028,7 +2028,8 @@ virtio_dev_rss_hash_update(struct rte_eth_dev *dev, + + return 0; + restore_key: +- memcpy(hw->rss_key, old_rss_key, VIRTIO_NET_RSS_KEY_SIZE); ++ if (rss_conf->rss_key && rss_conf->rss_key_len) ++ memcpy(hw->rss_key, old_rss_key, VIRTIO_NET_RSS_KEY_SIZE); + restore_types: + hw->rss_hash_types = old_hash_types; + +@@ -2657,7 +2658,7 @@ virtio_dev_configure(struct rte_eth_dev *dev) + hw->has_rx_offload = rx_offload_enabled(hw); + + if (dev->data->dev_flags & RTE_ETH_DEV_INTR_LSC) +- /* Enable vector (0) for Link State Intrerrupt */ ++ /* Enable vector (0) for Link State Interrupt */ + if (VIRTIO_OPS(hw)->set_config_irq(hw, 0) == + VIRTIO_MSI_NO_VECTOR) { + PMD_DRV_LOG(ERR, "failed to set config vector"); +@@ -2775,7 +2776,7 @@ virtio_dev_start(struct rte_eth_dev *dev) + } + } + +- /* Enable uio/vfio intr/eventfd mapping: althrough we already did that ++ /* Enable uio/vfio intr/eventfd mapping: although we already did that + * in device configure, but it could be unmapped when device is + * stopped. + */ +diff --git a/dpdk/drivers/net/virtio/virtio_pci.c b/dpdk/drivers/net/virtio/virtio_pci.c +index 182cfc9eae..632451dcbe 100644 +--- a/dpdk/drivers/net/virtio/virtio_pci.c ++++ b/dpdk/drivers/net/virtio/virtio_pci.c +@@ -235,7 +235,7 @@ legacy_get_isr(struct virtio_hw *hw) + return dst; + } + +-/* Enable one vector (0) for Link State Intrerrupt */ ++/* Enable one vector (0) for Link State Interrupt */ + static uint16_t + legacy_set_config_irq(struct virtio_hw *hw, uint16_t vec) + { +diff --git a/dpdk/drivers/net/virtio/virtio_pci_ethdev.c b/dpdk/drivers/net/virtio/virtio_pci_ethdev.c +index 54645dc62e..1f6bdeddda 100644 +--- a/dpdk/drivers/net/virtio/virtio_pci_ethdev.c ++++ b/dpdk/drivers/net/virtio/virtio_pci_ethdev.c +@@ -122,10 +122,20 @@ static int + eth_virtio_pci_uninit(struct rte_eth_dev *eth_dev) + { + int ret; ++ struct virtio_pci_dev *dev; ++ struct virtio_hw *hw; + PMD_INIT_FUNC_TRACE(); + +- if (rte_eal_process_type() == RTE_PROC_SECONDARY) ++ if (rte_eal_process_type() == RTE_PROC_SECONDARY) { ++ dev = eth_dev->data->dev_private; ++ hw = &dev->hw; ++ ++ if (dev->modern) ++ rte_pci_unmap_device(RTE_ETH_DEV_TO_PCI(eth_dev)); ++ else ++ vtpci_legacy_ioport_unmap(hw); + return 0; ++ } + + ret = virtio_dev_stop(eth_dev); + virtio_dev_close(eth_dev); +diff --git a/dpdk/drivers/net/virtio/virtio_rxtx.c b/dpdk/drivers/net/virtio/virtio_rxtx.c +index 2e115ded02..4795893ec7 100644 +--- a/dpdk/drivers/net/virtio/virtio_rxtx.c ++++ b/dpdk/drivers/net/virtio/virtio_rxtx.c +@@ -814,7 +814,7 @@ virtio_dev_tx_queue_setup(struct rte_eth_dev *dev, + unsigned int socket_id __rte_unused, + const struct rte_eth_txconf *tx_conf) + { +- uint8_t vq_idx = 2 * queue_idx + VTNET_SQ_TQ_QUEUE_IDX; ++ uint16_t vq_idx = 2 * queue_idx + VTNET_SQ_TQ_QUEUE_IDX; + struct virtio_hw *hw = dev->data->dev_private; + struct virtqueue *vq = hw->vqs[vq_idx]; + struct virtnet_tx *txvq; +@@ -858,7 +858,7 @@ int + virtio_dev_tx_queue_setup_finish(struct rte_eth_dev *dev, + uint16_t queue_idx) + { +- uint8_t vq_idx = 2 * queue_idx + VTNET_SQ_TQ_QUEUE_IDX; ++ uint16_t vq_idx = 2 * queue_idx + VTNET_SQ_TQ_QUEUE_IDX; + struct virtio_hw *hw = dev->data->dev_private; + struct virtqueue *vq = hw->vqs[vq_idx]; + +@@ -962,7 +962,7 @@ virtio_rx_offload(struct rte_mbuf *m, struct virtio_net_hdr *hdr) + return -EINVAL; + } + +- /* Update mss lengthes in mbuf */ ++ /* Update mss lengths in mbuf */ + m->tso_segsz = hdr->gso_size; + switch (hdr->gso_type & ~VIRTIO_NET_HDR_GSO_ECN) { + case VIRTIO_NET_HDR_GSO_TCPV4: +diff --git a/dpdk/drivers/net/virtio/virtio_rxtx_packed.h b/dpdk/drivers/net/virtio/virtio_rxtx_packed.h +index d5c259a1f6..536112983c 100644 +--- a/dpdk/drivers/net/virtio/virtio_rxtx_packed.h ++++ b/dpdk/drivers/net/virtio/virtio_rxtx_packed.h +@@ -125,13 +125,12 @@ virtqueue_enqueue_single_packed_vec(struct virtnet_tx *txvq, + * any_layout => number of segments + * default => number of segments + 1 + */ +- slots = use_indirect ? 1 : (txm->nb_segs + !can_push); + can_push = rte_mbuf_refcnt_read(txm) == 1 && + RTE_MBUF_DIRECT(txm) && + txm->nb_segs == 1 && + rte_pktmbuf_headroom(txm) >= hdr_size; + +- slots = txm->nb_segs + !can_push; ++ slots = use_indirect ? 1 : (txm->nb_segs + !can_push); + need = slots - vq->vq_free_cnt; + + /* Positive value indicates it need free vring descriptors */ +diff --git a/dpdk/drivers/net/virtio/virtio_rxtx_packed_avx.h b/dpdk/drivers/net/virtio/virtio_rxtx_packed_avx.h +index 8cb71f3fe6..584ac72f95 100644 +--- a/dpdk/drivers/net/virtio/virtio_rxtx_packed_avx.h ++++ b/dpdk/drivers/net/virtio/virtio_rxtx_packed_avx.h +@@ -192,7 +192,7 @@ virtqueue_dequeue_batch_packed_vec(struct virtnet_rx *rxvq, + + /* + * load len from desc, store into mbuf pkt_len and data_len +- * len limiated by l6bit buf_len, pkt_len[16:31] can be ignored ++ * len limited by l6bit buf_len, pkt_len[16:31] can be ignored + */ + const __mmask16 mask = 0x6 | 0x6 << 4 | 0x6 << 8 | 0x6 << 12; + __m512i values = _mm512_maskz_shuffle_epi32(mask, v_desc, 0xAA); +diff --git a/dpdk/drivers/net/virtio/virtio_rxtx_simple_altivec.c b/dpdk/drivers/net/virtio/virtio_rxtx_simple_altivec.c +index 7534974ef4..e7f0ed6068 100644 +--- a/dpdk/drivers/net/virtio/virtio_rxtx_simple_altivec.c ++++ b/dpdk/drivers/net/virtio/virtio_rxtx_simple_altivec.c +@@ -50,9 +50,9 @@ virtio_recv_pkts_vec(void *rx_queue, struct rte_mbuf **rx_pkts, + struct rte_mbuf **sw_ring_end; + struct rte_mbuf **ref_rx_pkts; + uint16_t nb_pkts_received = 0; +- const vector unsigned char zero = {0}; ++ const __vector unsigned char zero = {0}; + +- const vector unsigned char shuf_msk1 = { ++ const __vector unsigned char shuf_msk1 = { + 0xFF, 0xFF, 0xFF, 0xFF, /* packet type */ + 4, 5, 0xFF, 0xFF, /* vlan tci */ + 4, 5, /* dat len */ +@@ -60,7 +60,7 @@ virtio_recv_pkts_vec(void *rx_queue, struct rte_mbuf **rx_pkts, + 0xFF, 0xFF, 0xFF, 0xFF + }; + +- const vector unsigned char shuf_msk2 = { ++ const __vector unsigned char shuf_msk2 = { + 0xFF, 0xFF, 0xFF, 0xFF, /* packet type */ + 12, 13, 0xFF, 0xFF, /* pkt len */ + 12, 13, /* dat len */ +@@ -72,7 +72,7 @@ virtio_recv_pkts_vec(void *rx_queue, struct rte_mbuf **rx_pkts, + * Subtract the header length. + * In which case do we need the header length in used->len ? + */ +- const vector unsigned short len_adjust = { ++ const __vector unsigned short len_adjust = { + 0, 0, + (uint16_t)-vq->hw->vtnet_hdr_size, 0, + (uint16_t)-vq->hw->vtnet_hdr_size, 0, +@@ -112,68 +112,68 @@ virtio_recv_pkts_vec(void *rx_queue, struct rte_mbuf **rx_pkts, + ref_rx_pkts = rx_pkts; + for (nb_pkts_received = 0; + nb_pkts_received < nb_total;) { +- vector unsigned char desc[RTE_VIRTIO_DESC_PER_LOOP / 2]; +- vector unsigned char mbp[RTE_VIRTIO_DESC_PER_LOOP / 2]; +- vector unsigned char pkt_mb[RTE_VIRTIO_DESC_PER_LOOP]; ++ __vector unsigned char desc[RTE_VIRTIO_DESC_PER_LOOP / 2]; ++ __vector unsigned char mbp[RTE_VIRTIO_DESC_PER_LOOP / 2]; ++ __vector unsigned char pkt_mb[RTE_VIRTIO_DESC_PER_LOOP]; + + mbp[0] = vec_vsx_ld(0, (unsigned char const *)(sw_ring + 0)); + desc[0] = vec_vsx_ld(0, (unsigned char const *)(rused + 0)); +- *(vector unsigned char *)&rx_pkts[0] = mbp[0]; ++ *(__vector unsigned char *)&rx_pkts[0] = mbp[0]; + + mbp[1] = vec_vsx_ld(0, (unsigned char const *)(sw_ring + 2)); + desc[1] = vec_vsx_ld(0, (unsigned char const *)(rused + 2)); +- *(vector unsigned char *)&rx_pkts[2] = mbp[1]; ++ *(__vector unsigned char *)&rx_pkts[2] = mbp[1]; + + mbp[2] = vec_vsx_ld(0, (unsigned char const *)(sw_ring + 4)); + desc[2] = vec_vsx_ld(0, (unsigned char const *)(rused + 4)); +- *(vector unsigned char *)&rx_pkts[4] = mbp[2]; ++ *(__vector unsigned char *)&rx_pkts[4] = mbp[2]; + + mbp[3] = vec_vsx_ld(0, (unsigned char const *)(sw_ring + 6)); + desc[3] = vec_vsx_ld(0, (unsigned char const *)(rused + 6)); +- *(vector unsigned char *)&rx_pkts[6] = mbp[3]; ++ *(__vector unsigned char *)&rx_pkts[6] = mbp[3]; + + pkt_mb[0] = vec_perm(desc[0], zero, shuf_msk1); + pkt_mb[1] = vec_perm(desc[0], zero, shuf_msk2); +- pkt_mb[0] = (vector unsigned char) +- ((vector unsigned short)pkt_mb[0] + len_adjust); +- pkt_mb[1] = (vector unsigned char) +- ((vector unsigned short)pkt_mb[1] + len_adjust); +- *(vector unsigned char *)&rx_pkts[0]->rx_descriptor_fields1 = ++ pkt_mb[0] = (__vector unsigned char) ++ ((__vector unsigned short)pkt_mb[0] + len_adjust); ++ pkt_mb[1] = (__vector unsigned char) ++ ((__vector unsigned short)pkt_mb[1] + len_adjust); ++ *(__vector unsigned char *)&rx_pkts[0]->rx_descriptor_fields1 = + pkt_mb[0]; +- *(vector unsigned char *)&rx_pkts[1]->rx_descriptor_fields1 = ++ *(__vector unsigned char *)&rx_pkts[1]->rx_descriptor_fields1 = + pkt_mb[1]; + + pkt_mb[2] = vec_perm(desc[1], zero, shuf_msk1); + pkt_mb[3] = vec_perm(desc[1], zero, shuf_msk2); +- pkt_mb[2] = (vector unsigned char) +- ((vector unsigned short)pkt_mb[2] + len_adjust); +- pkt_mb[3] = (vector unsigned char) +- ((vector unsigned short)pkt_mb[3] + len_adjust); +- *(vector unsigned char *)&rx_pkts[2]->rx_descriptor_fields1 = ++ pkt_mb[2] = (__vector unsigned char) ++ ((__vector unsigned short)pkt_mb[2] + len_adjust); ++ pkt_mb[3] = (__vector unsigned char) ++ ((__vector unsigned short)pkt_mb[3] + len_adjust); ++ *(__vector unsigned char *)&rx_pkts[2]->rx_descriptor_fields1 = + pkt_mb[2]; +- *(vector unsigned char *)&rx_pkts[3]->rx_descriptor_fields1 = ++ *(__vector unsigned char *)&rx_pkts[3]->rx_descriptor_fields1 = + pkt_mb[3]; + + pkt_mb[4] = vec_perm(desc[2], zero, shuf_msk1); + pkt_mb[5] = vec_perm(desc[2], zero, shuf_msk2); +- pkt_mb[4] = (vector unsigned char) +- ((vector unsigned short)pkt_mb[4] + len_adjust); +- pkt_mb[5] = (vector unsigned char) +- ((vector unsigned short)pkt_mb[5] + len_adjust); +- *(vector unsigned char *)&rx_pkts[4]->rx_descriptor_fields1 = ++ pkt_mb[4] = (__vector unsigned char) ++ ((__vector unsigned short)pkt_mb[4] + len_adjust); ++ pkt_mb[5] = (__vector unsigned char) ++ ((__vector unsigned short)pkt_mb[5] + len_adjust); ++ *(__vector unsigned char *)&rx_pkts[4]->rx_descriptor_fields1 = + pkt_mb[4]; +- *(vector unsigned char *)&rx_pkts[5]->rx_descriptor_fields1 = ++ *(__vector unsigned char *)&rx_pkts[5]->rx_descriptor_fields1 = + pkt_mb[5]; + + pkt_mb[6] = vec_perm(desc[3], zero, shuf_msk1); + pkt_mb[7] = vec_perm(desc[3], zero, shuf_msk2); +- pkt_mb[6] = (vector unsigned char) +- ((vector unsigned short)pkt_mb[6] + len_adjust); +- pkt_mb[7] = (vector unsigned char) +- ((vector unsigned short)pkt_mb[7] + len_adjust); +- *(vector unsigned char *)&rx_pkts[6]->rx_descriptor_fields1 = ++ pkt_mb[6] = (__vector unsigned char) ++ ((__vector unsigned short)pkt_mb[6] + len_adjust); ++ pkt_mb[7] = (__vector unsigned char) ++ ((__vector unsigned short)pkt_mb[7] + len_adjust); ++ *(__vector unsigned char *)&rx_pkts[6]->rx_descriptor_fields1 = + pkt_mb[6]; +- *(vector unsigned char *)&rx_pkts[7]->rx_descriptor_fields1 = ++ *(__vector unsigned char *)&rx_pkts[7]->rx_descriptor_fields1 = + pkt_mb[7]; + + if (unlikely(nb_used <= RTE_VIRTIO_DESC_PER_LOOP)) { +diff --git a/dpdk/drivers/net/virtio/virtio_user/vhost_user.c b/dpdk/drivers/net/virtio/virtio_user/vhost_user.c +index cc830a660f..77820bf967 100644 +--- a/dpdk/drivers/net/virtio/virtio_user/vhost_user.c ++++ b/dpdk/drivers/net/virtio/virtio_user/vhost_user.c +@@ -840,8 +840,10 @@ vhost_user_setup(struct virtio_user_dev *dev) + } + + flag = fcntl(fd, F_GETFD); +- if (fcntl(fd, F_SETFD, flag | FD_CLOEXEC) < 0) +- PMD_DRV_LOG(WARNING, "fcntl failed, %s", strerror(errno)); ++ if (flag == -1) ++ PMD_DRV_LOG(WARNING, "fcntl get fd failed, %s", strerror(errno)); ++ else if (fcntl(fd, F_SETFD, flag | FD_CLOEXEC) < 0) ++ PMD_DRV_LOG(WARNING, "fcntl set fd failed, %s", strerror(errno)); + + memset(&un, 0, sizeof(un)); + un.sun_family = AF_UNIX; +@@ -940,15 +942,8 @@ vhost_user_update_link_state(struct virtio_user_dev *dev) + + if (data->vhostfd >= 0) { + int r; +- int flags; + +- flags = fcntl(data->vhostfd, F_GETFL); +- if (fcntl(data->vhostfd, F_SETFL, flags | O_NONBLOCK) == -1) { +- PMD_DRV_LOG(ERR, "error setting O_NONBLOCK flag"); +- return -1; +- } +- +- r = recv(data->vhostfd, buf, 128, MSG_PEEK); ++ r = recv(data->vhostfd, buf, 128, MSG_PEEK | MSG_DONTWAIT); + if (r == 0 || (r < 0 && errno != EAGAIN)) { + dev->net_status &= (~VIRTIO_NET_S_LINK_UP); + PMD_DRV_LOG(ERR, "virtio-user port %u is down", dev->hw.port_id); +@@ -963,12 +958,6 @@ vhost_user_update_link_state(struct virtio_user_dev *dev) + } else { + dev->net_status |= VIRTIO_NET_S_LINK_UP; + } +- +- if (fcntl(data->vhostfd, F_SETFL, +- flags & ~O_NONBLOCK) == -1) { +- PMD_DRV_LOG(ERR, "error clearing O_NONBLOCK flag"); +- return -1; +- } + } else if (dev->is_server) { + dev->net_status &= (~VIRTIO_NET_S_LINK_UP); + if (virtio_user_dev_server_reconnect(dev) >= 0) +diff --git a/dpdk/drivers/net/virtio/virtio_user/virtio_user_dev.c b/dpdk/drivers/net/virtio/virtio_user/virtio_user_dev.c +index 35aa76b1ff..f9cada05e4 100644 +--- a/dpdk/drivers/net/virtio/virtio_user/virtio_user_dev.c ++++ b/dpdk/drivers/net/virtio/virtio_user/virtio_user_dev.c +@@ -417,7 +417,7 @@ virtio_user_fill_intr_handle(struct virtio_user_dev *dev) + + for (i = 0; i < dev->max_queue_pairs; ++i) { + if (rte_intr_efds_index_set(eth_dev->intr_handle, i, +- dev->callfds[i])) ++ dev->callfds[2 * i + VTNET_SQ_RQ_QUEUE_IDX])) + return -rte_errno; + } + +diff --git a/dpdk/drivers/net/virtio/virtio_user_ethdev.c b/dpdk/drivers/net/virtio/virtio_user_ethdev.c +index 0271098f0d..16eca2f940 100644 +--- a/dpdk/drivers/net/virtio/virtio_user_ethdev.c ++++ b/dpdk/drivers/net/virtio/virtio_user_ethdev.c +@@ -666,6 +666,7 @@ virtio_user_pmd_probe(struct rte_vdev_device *vdev) + /* previously called by pci probing for physical dev */ + if (eth_virtio_dev_init(eth_dev) < 0) { + PMD_INIT_LOG(ERR, "eth_virtio_dev_init fails"); ++ virtio_user_dev_uninit(dev); + virtio_user_eth_dev_free(eth_dev); + goto end; + } +diff --git a/dpdk/drivers/net/virtio/virtqueue.c b/dpdk/drivers/net/virtio/virtqueue.c +index 65bf792eb0..c98d696e62 100644 +--- a/dpdk/drivers/net/virtio/virtqueue.c ++++ b/dpdk/drivers/net/virtio/virtqueue.c +@@ -13,7 +13,7 @@ + /* + * Two types of mbuf to be cleaned: + * 1) mbuf that has been consumed by backend but not used by virtio. +- * 2) mbuf that hasn't been consued by backend. ++ * 2) mbuf that hasn't been consumed by backend. + */ + struct rte_mbuf * + virtqueue_detach_unused(struct virtqueue *vq) +diff --git a/dpdk/drivers/net/virtio/virtqueue.h b/dpdk/drivers/net/virtio/virtqueue.h +index 855f57a956..99c68cf622 100644 +--- a/dpdk/drivers/net/virtio/virtqueue.h ++++ b/dpdk/drivers/net/virtio/virtqueue.h +@@ -227,7 +227,7 @@ struct virtio_net_ctrl_rss { + * Control link announce acknowledgement + * + * The command VIRTIO_NET_CTRL_ANNOUNCE_ACK is used to indicate that +- * driver has recevied the notification; device would clear the ++ * driver has received the notification; device would clear the + * VIRTIO_NET_S_ANNOUNCE bit in the status field after it receives + * this command. + */ +@@ -312,7 +312,7 @@ struct virtqueue { + struct vq_desc_extra vq_descx[0]; + }; + +-/* If multiqueue is provided by host, then we suppport it. */ ++/* If multiqueue is provided by host, then we support it. */ + #define VIRTIO_NET_CTRL_MQ 4 + + #define VIRTIO_NET_CTRL_MQ_VQ_PAIRS_SET 0 +diff --git a/dpdk/drivers/net/vmxnet3/vmxnet3_ethdev.c b/dpdk/drivers/net/vmxnet3/vmxnet3_ethdev.c +index d1ef1cad08..d4aea74026 100644 +--- a/dpdk/drivers/net/vmxnet3/vmxnet3_ethdev.c ++++ b/dpdk/drivers/net/vmxnet3/vmxnet3_ethdev.c +@@ -822,6 +822,11 @@ vmxnet3_setup_driver_shared(struct rte_eth_dev *dev) + rqd->conf.rxRingSize[1] = rxq->cmd_ring[1].size; + rqd->conf.compRingSize = rxq->comp_ring.size; + ++ if (VMXNET3_VERSION_GE_3(hw)) { ++ rqd->conf.rxDataRingBasePA = rxq->data_ring.basePA; ++ rqd->conf.rxDataRingDescSize = rxq->data_desc_size; ++ } ++ + if (hw->intr.lsc_only) + rqd->conf.intrIdx = 1; + else +diff --git a/dpdk/drivers/raw/dpaa2_qdma/dpaa2_qdma.c b/dpdk/drivers/raw/dpaa2_qdma/dpaa2_qdma.c +index de26d2aef3..ebc2cd5d0d 100644 +--- a/dpdk/drivers/raw/dpaa2_qdma/dpaa2_qdma.c ++++ b/dpdk/drivers/raw/dpaa2_qdma/dpaa2_qdma.c +@@ -653,7 +653,7 @@ dpdmai_dev_dequeue_multijob_prefetch( + rte_prefetch0((void *)(size_t)(dq_storage + 1)); + + /* Prepare next pull descriptor. This will give space for the +- * prefething done on DQRR entries ++ * prefetching done on DQRR entries + */ + q_storage->toggle ^= 1; + dq_storage1 = q_storage->dq_storage[q_storage->toggle]; +diff --git a/dpdk/drivers/raw/dpaa2_qdma/dpaa2_qdma.h b/dpdk/drivers/raw/dpaa2_qdma/dpaa2_qdma.h +index d6f6bb5522..1973d5d2b2 100644 +--- a/dpdk/drivers/raw/dpaa2_qdma/dpaa2_qdma.h ++++ b/dpdk/drivers/raw/dpaa2_qdma/dpaa2_qdma.h +@@ -82,7 +82,7 @@ struct qdma_device { + /** total number of hw queues. */ + uint16_t num_hw_queues; + /** +- * Maximum number of hw queues to be alocated per core. ++ * Maximum number of hw queues to be allocated per core. + * This is limited by MAX_HW_QUEUE_PER_CORE + */ + uint16_t max_hw_queues_per_core; +@@ -268,7 +268,7 @@ struct dpaa2_dpdmai_dev { + struct fsl_mc_io dpdmai; + /** HW ID for DPDMAI object */ + uint32_t dpdmai_id; +- /** Tocken of this device */ ++ /** Token of this device */ + uint16_t token; + /** Number of queue in this DPDMAI device */ + uint8_t num_queues; +diff --git a/dpdk/drivers/raw/ifpga/base/ifpga_defines.h b/dpdk/drivers/raw/ifpga/base/ifpga_defines.h +index dca1518a83..8f6203392b 100644 +--- a/dpdk/drivers/raw/ifpga/base/ifpga_defines.h ++++ b/dpdk/drivers/raw/ifpga/base/ifpga_defines.h +@@ -93,9 +93,9 @@ enum fpga_id_type { + + #define PORT_FEATURE_ID_HEADER FEATURE_ID_FIU_HEADER + #define PORT_FEATURE_ID_ERROR 0x10 +-#define PORT_FEATURE_ID_UMSG 0x12 +-#define PORT_FEATURE_ID_UINT 0x13 +-#define PORT_FEATURE_ID_STP 0x14 ++#define PORT_FEATURE_ID_UMSG 0x11 ++#define PORT_FEATURE_ID_UINT 0x12 ++#define PORT_FEATURE_ID_STP 0x13 + #define PORT_FEATURE_ID_UAFU FEATURE_ID_AFU + + /* +diff --git a/dpdk/drivers/raw/ifpga/base/opae_spi.c b/dpdk/drivers/raw/ifpga/base/opae_spi.c +index 9efeecb791..ca3d41fb92 100644 +--- a/dpdk/drivers/raw/ifpga/base/opae_spi.c ++++ b/dpdk/drivers/raw/ifpga/base/opae_spi.c +@@ -239,6 +239,18 @@ int spi_command(struct altera_spi_device *dev, unsigned int chip_select, + return 0; + } + ++int spi_write(struct altera_spi_device *dev, unsigned int chip_select, ++ unsigned int wlen, void *wdata) ++{ ++ return spi_command(dev, chip_select, wlen, wdata, 0, NULL); ++} ++ ++int spi_read(struct altera_spi_device *dev, unsigned int chip_select, ++ unsigned int rlen, void *rdata) ++{ ++ return spi_command(dev, chip_select, 0, NULL, rlen, rdata); ++} ++ + struct altera_spi_device *altera_spi_alloc(void *base, int type) + { + struct altera_spi_device *spi_dev = +diff --git a/dpdk/drivers/raw/ifpga/base/opae_spi.h b/dpdk/drivers/raw/ifpga/base/opae_spi.h +index af11656e4d..bcff67dd66 100644 +--- a/dpdk/drivers/raw/ifpga/base/opae_spi.h ++++ b/dpdk/drivers/raw/ifpga/base/opae_spi.h +@@ -117,6 +117,10 @@ struct spi_tran_header { + u32 addr; + }; + ++int spi_read(struct altera_spi_device *dev, unsigned int chip_select, ++ unsigned int rlen, void *rdata); ++int spi_write(struct altera_spi_device *dev, unsigned int chip_select, ++ unsigned int wlen, void *wdata); + int spi_command(struct altera_spi_device *dev, unsigned int chip_select, + unsigned int wlen, void *wdata, unsigned int rlen, void *rdata); + void spi_cs_deactivate(struct altera_spi_device *dev); +diff --git a/dpdk/drivers/raw/ifpga/base/opae_spi_transaction.c b/dpdk/drivers/raw/ifpga/base/opae_spi_transaction.c +index 006cdb4c1a..cd50d40629 100644 +--- a/dpdk/drivers/raw/ifpga/base/opae_spi_transaction.c ++++ b/dpdk/drivers/raw/ifpga/base/opae_spi_transaction.c +@@ -40,7 +40,7 @@ static void print_buffer(const char *string, void *buffer, int len) + printf("%s print buffer, len=%d\n", string, len); + + for (i = 0; i < len; i++) +- printf("%x ", *(p+i)); ++ printf("%02x ", *(p+i)); + printf("\n"); + } + #else +@@ -72,43 +72,6 @@ static void reorder_phy_data(u8 bits_per_word, + } + } + +-enum { +- SPI_FOUND_SOP, +- SPI_FOUND_EOP, +- SPI_NOT_FOUND, +-}; +- +-static int resp_find_sop_eop(unsigned char *resp, unsigned int len, +- int flags) +-{ +- int ret = SPI_NOT_FOUND; +- +- unsigned char *b = resp; +- +- /* find SOP */ +- if (flags != SPI_FOUND_SOP) { +- while (b < resp + len && *b != SPI_PACKET_SOP) +- b++; +- +- if (*b != SPI_PACKET_SOP) +- goto done; +- +- ret = SPI_FOUND_SOP; +- } +- +- /* find EOP */ +- while (b < resp + len && *b != SPI_PACKET_EOP) +- b++; +- +- if (*b != SPI_PACKET_EOP) +- goto done; +- +- ret = SPI_FOUND_EOP; +- +-done: +- return ret; +-} +- + static void phy_tx_pad(unsigned char *phy_buf, unsigned int phy_buf_len, + unsigned int *aligned_len) + { +@@ -137,6 +100,104 @@ static void phy_tx_pad(unsigned char *phy_buf, unsigned int phy_buf_len, + *p++ = SPI_BYTE_IDLE; + } + ++#define RX_ALL_IDLE_DATA (SPI_BYTE_IDLE << 24 | SPI_BYTE_IDLE << 16 | \ ++ SPI_BYTE_IDLE << 8 | SPI_BYTE_IDLE) ++ ++static bool all_idle_data(u8 *rxbuf) ++{ ++ return *(u32 *)rxbuf == RX_ALL_IDLE_DATA; ++} ++ ++static unsigned char *find_eop(u8 *rxbuf, u32 BPW) ++{ ++ return memchr(rxbuf, SPI_PACKET_EOP, BPW); ++} ++ ++static int do_spi_txrx(struct spi_transaction_dev *dev, ++ unsigned char *tx_buffer, ++ unsigned int tx_len, unsigned char *rx_buffer, ++ unsigned int rx_len, ++ unsigned int *actual_rx) ++{ ++ unsigned int rx_cnt = 0; ++ int ret = 0; ++ unsigned int BPW = 4; ++ bool eop_found = false; ++ unsigned char *eop; ++ unsigned char *ptr; ++ unsigned char *rxbuf = rx_buffer; ++ int add_byte = 0; ++ unsigned long ticks; ++ unsigned long timeout; ++ ++ /* send command */ ++ ret = spi_write(dev->dev, dev->chipselect, tx_len, tx_buffer); ++ if (ret) ++ return -EBUSY; ++ ++ timeout = rte_get_timer_cycles() + ++ msecs_to_timer_cycles(2000); ++ ++ /* read out data */ ++ while (rx_cnt < rx_len) { ++ ret = spi_read(dev->dev, dev->chipselect, BPW, rxbuf); ++ if (ret) ++ return -EBUSY; ++ ++ /* skip all of invalid data */ ++ if (!eop_found && all_idle_data(rxbuf)) { ++ ticks = rte_get_timer_cycles(); ++ if (!time_after(ticks, timeout)) { ++ continue; ++ } else { ++ dev_err(dev, "read spi data timeout\n"); ++ return -ETIMEDOUT; ++ } ++ } ++ ++ rx_cnt += BPW; ++ if (!eop_found) { ++ /* EOP is found, we read 2 more bytes and exit. */ ++ eop = find_eop(rxbuf, BPW); ++ if (eop) { ++ if ((BPW + rxbuf - eop) > 2) { ++ /* ++ * check if the last 2 bytes are already ++ * received in current word. ++ */ ++ break; ++ } else if ((BPW + rxbuf - eop) == 2) { ++ /* ++ * skip if last byte is not SPI_BYTE_ESC ++ * or SPI_PACKET_ESC. this is the valid ++ * end of a response too. ++ */ ++ ptr = eop + 1; ++ ++ if (*ptr != SPI_BYTE_ESC && ++ *ptr != SPI_PACKET_ESC) ++ break; ++ ++ add_byte = 1; ++ } else { ++ add_byte = 2; ++ } ++ ++ rx_len = min(rx_len, ++ IFPGA_ALIGN(rx_cnt + ++ add_byte, BPW)); ++ eop_found = true; ++ } ++ } ++ rxbuf += BPW; ++ } ++ ++ *actual_rx = rx_cnt; ++ print_buffer("found valid data:", rx_buffer, rx_cnt); ++ ++ return ret; ++} ++ + static int byte_to_core_convert(struct spi_transaction_dev *dev, + unsigned int send_len, unsigned char *send_data, + unsigned int resp_len, unsigned char *resp_data, +@@ -148,15 +209,9 @@ static int byte_to_core_convert(struct spi_transaction_dev *dev, + unsigned char *resp_packet = dev->buffer->bytes_resp; + unsigned char *p; + unsigned char current_byte; +- unsigned char *tx_buffer; + unsigned int tx_len = 0; +- unsigned char *rx_buffer; +- unsigned int rx_len = 0; +- int retry = 0; +- int spi_flags; +- unsigned long timeout = msecs_to_timer_cycles(1000); +- unsigned long ticks; + unsigned int resp_max_len = 2 * resp_len; ++ unsigned int actual_rx; + + print_buffer("before bytes:", send_data, send_len); + +@@ -190,48 +245,15 @@ static int byte_to_core_convert(struct spi_transaction_dev *dev, + + print_buffer("after order to spi:", send_packet, tx_len); + +- /* call spi */ +- tx_buffer = send_packet; +- rx_buffer = resp_packet; +- rx_len = resp_max_len; +- spi_flags = SPI_NOT_FOUND; +- +-read_again: +- ret = spi_command(dev->dev, dev->chipselect, tx_len, tx_buffer, +- rx_len, rx_buffer); ++ ret = do_spi_txrx(dev, send_packet, tx_len, resp_packet, ++ resp_max_len, &actual_rx); + if (ret) +- return -EBUSY; +- +- print_buffer("read from spi:", rx_buffer, rx_len); +- +- /* look for SOP firstly*/ +- ret = resp_find_sop_eop(rx_buffer, rx_len - 1, spi_flags); +- if (ret != SPI_FOUND_EOP) { +- tx_buffer = NULL; +- tx_len = 0; +- ticks = rte_get_timer_cycles(); +- if (time_after(ticks, timeout) && +- retry++ > SPI_MAX_RETRY) { +- dev_err(NULL, "Have retry %d, found invalid packet data\n", +- retry); +- return -EBUSY; +- } +- +- if (ret == SPI_FOUND_SOP) { +- rx_buffer += rx_len; +- resp_max_len += rx_len; +- } +- +- spi_flags = ret; +- goto read_again; +- } +- +- print_buffer("found valid data:", resp_packet, resp_max_len); ++ return ret; + + /* analyze response packet */ + i = 0; + p = resp_data; +- while (i < resp_max_len) { ++ while (i < actual_rx) { + current_byte = resp_packet[i]; + switch (current_byte) { + case SPI_BYTE_IDLE: +@@ -337,9 +359,13 @@ static int packet_to_byte_conver(struct spi_transaction_dev *dev, + current_byte = resp_packet[i]; + + switch (current_byte) { +- case SPI_PACKET_ESC: +- case SPI_PACKET_CHANNEL: + case SPI_PACKET_SOP: ++ dev_err(dev, "error on get SOP after SOP\n"); ++ return -EINVAL; ++ case SPI_PACKET_CHANNEL: ++ i += 2; ++ break; ++ case SPI_PACKET_ESC: + i++; + current_byte = resp_packet[i]; + *p++ = xor_20(current_byte); +@@ -348,23 +374,30 @@ static int packet_to_byte_conver(struct spi_transaction_dev *dev, + case SPI_PACKET_EOP: + i++; + current_byte = resp_packet[i]; +- if (current_byte == SPI_PACKET_ESC || +- current_byte == SPI_PACKET_CHANNEL || +- current_byte == SPI_PACKET_SOP) { ++ switch (current_byte) { ++ case SPI_PACKET_ESC: + i++; + current_byte = resp_packet[i]; + *p++ = xor_20(current_byte); +- } else ++ break; ++ case SPI_PACKET_CHANNEL: ++ case SPI_PACKET_SOP: ++ case SPI_PACKET_EOP: ++ dev_err(dev, "error get SOP/EOP after EOP\n"); ++ return -EINVAL; ++ default: + *p++ = current_byte; +- i = valid_resp_len; +- break; ++ break; ++ } ++ goto done; ++ + default: + *p++ = current_byte; + i++; + } +- + } + ++done: + *valid = p - resp_buf; + + print_buffer("after packet:", resp_buf, *valid); +diff --git a/dpdk/drivers/raw/ifpga/ifpga_rawdev.c b/dpdk/drivers/raw/ifpga/ifpga_rawdev.c +index 8d9db585a4..cb0427157a 100644 +--- a/dpdk/drivers/raw/ifpga/ifpga_rawdev.c ++++ b/dpdk/drivers/raw/ifpga/ifpga_rawdev.c +@@ -68,13 +68,9 @@ static const struct rte_pci_id pci_ifpga_map[] = { + + static struct ifpga_rawdev ifpga_rawdevices[IFPGA_RAWDEV_NUM]; + +-static int ifpga_monitor_start; ++static int ifpga_monitor_refcnt; + static pthread_t ifpga_monitor_start_thread; + +-#define IFPGA_MAX_IRQ 12 +-/* 0 for FME interrupt, others are reserved for AFU irq */ +-static struct rte_intr_handle *ifpga_irq_handle[IFPGA_MAX_IRQ]; +- + static struct ifpga_rawdev * + ifpga_rawdev_allocate(struct rte_rawdev *rawdev); + static int set_surprise_link_check_aer( +@@ -82,6 +78,7 @@ static int set_surprise_link_check_aer( + static int ifpga_pci_find_next_ext_capability(unsigned int fd, + int start, uint32_t cap); + static int ifpga_pci_find_ext_capability(unsigned int fd, uint32_t cap); ++static void fme_interrupt_handler(void *param); + + struct ifpga_rawdev * + ifpga_rawdev_get(const struct rte_rawdev *rawdev) +@@ -118,6 +115,7 @@ ifpga_rawdev_allocate(struct rte_rawdev *rawdev) + { + struct ifpga_rawdev *dev; + uint16_t dev_id; ++ int i = 0; + + dev = ifpga_rawdev_get(rawdev); + if (dev != NULL) { +@@ -134,6 +132,11 @@ ifpga_rawdev_allocate(struct rte_rawdev *rawdev) + dev = &ifpga_rawdevices[dev_id]; + dev->rawdev = rawdev; + dev->dev_id = dev_id; ++ for (i = 0; i < IFPGA_MAX_IRQ; i++) ++ dev->intr_handle[i] = NULL; ++ dev->poll_enabled = 0; ++ for (i = 0; i < IFPGA_MAX_VDEV; i++) ++ dev->vdev_name[i] = NULL; + + return dev; + } +@@ -208,15 +211,16 @@ static int ifpga_get_dev_vendor_id(const char *bdf, + + return 0; + } +-static int ifpga_rawdev_fill_info(struct ifpga_rawdev *ifpga_dev, +- const char *bdf) ++ ++static int ifpga_rawdev_fill_info(struct ifpga_rawdev *ifpga_dev) + { +- char path[1024] = "/sys/bus/pci/devices/0000:"; ++ struct opae_adapter *adapter = NULL; ++ char path[1024] = "/sys/bus/pci/devices/"; + char link[1024], link1[1024]; + char dir[1024] = "/sys/devices/"; + char *c; + int ret; +- char sub_brg_bdf[4][16]; ++ char sub_brg_bdf[4][16] = {{0}}; + int point; + DIR *dp = NULL; + struct dirent *entry; +@@ -224,9 +228,14 @@ static int ifpga_rawdev_fill_info(struct ifpga_rawdev *ifpga_dev, + + unsigned int dom, bus, dev; + int func; +- uint32_t dev_id, vendor_id; ++ uint32_t dev_id = 0; ++ uint32_t vendor_id = 0; + +- strlcat(path, bdf, sizeof(path)); ++ adapter = ifpga_dev ? ifpga_rawdev_get_priv(ifpga_dev->rawdev) : NULL; ++ if (!adapter) ++ return -ENODEV; ++ ++ strlcat(path, adapter->name, sizeof(path)); + memset(link, 0, sizeof(link)); + memset(link1, 0, sizeof(link1)); + ret = readlink(path, link, (sizeof(link)-1)); +@@ -376,13 +385,13 @@ ifpga_monitor_sensor(struct rte_rawdev *raw_dev, + /* monitor temperature sensors */ + if (!strcmp(sensor->name, "Board Temperature") || + !strcmp(sensor->name, "FPGA Die Temperature")) { +- IFPGA_RAWDEV_PMD_INFO("read sensor %s %d %d %d\n", ++ IFPGA_RAWDEV_PMD_DEBUG("read sensor %s %d %d %d\n", + sensor->name, value, sensor->high_warn, + sensor->high_fatal); + + if (HIGH_WARN(sensor, value) || + LOW_WARN(sensor, value)) { +- IFPGA_RAWDEV_PMD_INFO("%s reach theshold %d\n", ++ IFPGA_RAWDEV_PMD_INFO("%s reach threshold %d\n", + sensor->name, value); + *gsd_start = true; + break; +@@ -393,7 +402,7 @@ ifpga_monitor_sensor(struct rte_rawdev *raw_dev, + if (!strcmp(sensor->name, "12V AUX Voltage")) { + if (value < AUX_VOLTAGE_WARN) { + IFPGA_RAWDEV_PMD_INFO( +- "%s reach theshold %d mV\n", ++ "%s reach threshold %d mV\n", + sensor->name, value); + *gsd_start = true; + break; +@@ -418,7 +427,7 @@ static int set_surprise_link_check_aer( + bool enable = 0; + uint32_t aer_new0, aer_new1; + +- if (!ifpga_rdev) { ++ if (!ifpga_rdev || !ifpga_rdev->rawdev) { + printf("\n device does not exist\n"); + return -EFAULT; + } +@@ -441,12 +450,12 @@ static int set_surprise_link_check_aer( + pos = ifpga_pci_find_ext_capability(fd, RTE_PCI_EXT_CAP_ID_ERR); + if (!pos) + goto end; +- /* save previout ECAP_AER+0x08 */ ++ /* save previous ECAP_AER+0x08 */ + ret = pread(fd, &data, sizeof(data), pos+0x08); + if (ret == -1) + goto end; + ifpga_rdev->aer_old[0] = data; +- /* save previout ECAP_AER+0x14 */ ++ /* save previous ECAP_AER+0x14 */ + ret = pread(fd, &data, sizeof(data), pos+0x14); + if (ret == -1) + goto end; +@@ -497,11 +506,11 @@ ifpga_rawdev_gsd_handle(__rte_unused void *param) + int gsd_enable, ret; + #define MS 1000 + +- while (1) { ++ while (__atomic_load_n(&ifpga_monitor_refcnt, __ATOMIC_RELAXED)) { + gsd_enable = 0; + for (i = 0; i < IFPGA_RAWDEV_NUM; i++) { + ifpga_rdev = &ifpga_rawdevices[i]; +- if (ifpga_rdev->rawdev) { ++ if (ifpga_rdev->poll_enabled) { + ret = set_surprise_link_check_aer(ifpga_rdev, + gsd_enable); + if (ret == 1 && !gsd_enable) { +@@ -521,30 +530,46 @@ ifpga_rawdev_gsd_handle(__rte_unused void *param) + } + + static int +-ifpga_monitor_start_func(void) ++ifpga_monitor_start_func(struct ifpga_rawdev *dev) + { + int ret; + +- if (ifpga_monitor_start == 0) { ++ if (!dev) ++ return -ENODEV; ++ ++ ret = ifpga_rawdev_fill_info(dev); ++ if (ret) ++ return ret; ++ ++ dev->poll_enabled = 1; ++ ++ if (!__atomic_fetch_add(&ifpga_monitor_refcnt, 1, __ATOMIC_RELAXED)) { + ret = rte_ctrl_thread_create(&ifpga_monitor_start_thread, + "ifpga-monitor", NULL, + ifpga_rawdev_gsd_handle, NULL); + if (ret != 0) { ++ ifpga_monitor_start_thread = 0; + IFPGA_RAWDEV_PMD_ERR( +- "Fail to create ifpga nonitor thread"); ++ "Fail to create ifpga monitor thread"); + return -1; + } +- ifpga_monitor_start = 1; + } + + return 0; + } ++ + static int +-ifpga_monitor_stop_func(void) ++ifpga_monitor_stop_func(struct ifpga_rawdev *dev) + { + int ret; + +- if (ifpga_monitor_start == 1) { ++ if (!dev || !dev->poll_enabled) ++ return 0; ++ ++ dev->poll_enabled = 0; ++ ++ if (!__atomic_sub_fetch(&ifpga_monitor_refcnt, 1, __ATOMIC_RELAXED) && ++ ifpga_monitor_start_thread) { + ret = pthread_cancel(ifpga_monitor_start_thread); + if (ret) + IFPGA_RAWDEV_PMD_ERR("Can't cancel the thread"); +@@ -553,8 +578,6 @@ ifpga_monitor_stop_func(void) + if (ret) + IFPGA_RAWDEV_PMD_ERR("Can't join the thread"); + +- ifpga_monitor_start = 0; +- + return ret; + } + +@@ -716,17 +739,38 @@ ifpga_rawdev_stop(struct rte_rawdev *dev) + static int + ifpga_rawdev_close(struct rte_rawdev *dev) + { ++ struct ifpga_rawdev *ifpga_rdev = NULL; + struct opae_adapter *adapter; ++ struct opae_manager *mgr; ++ char *vdev_name = NULL; ++ int i, ret = 0; + + if (dev) { ++ ifpga_rdev = ifpga_rawdev_get(dev); ++ if (ifpga_rdev) { ++ for (i = 0; i < IFPGA_MAX_VDEV; i++) { ++ vdev_name = ifpga_rdev->vdev_name[i]; ++ if (vdev_name) ++ rte_vdev_uninit(vdev_name); ++ } ++ ifpga_monitor_stop_func(ifpga_rdev); ++ ifpga_rdev->rawdev = NULL; ++ } + adapter = ifpga_rawdev_get_priv(dev); + if (adapter) { ++ mgr = opae_adapter_get_mgr(adapter); ++ if (ifpga_rdev && mgr) { ++ if (ifpga_unregister_msix_irq(ifpga_rdev, ++ IFPGA_FME_IRQ, 0, ++ fme_interrupt_handler, mgr) < 0) ++ ret = -EINVAL; ++ } + opae_adapter_destroy(adapter); + opae_adapter_data_free(adapter->data); + } + } + +- return dev ? 0:1; ++ return ret; + } + + static int +@@ -1341,49 +1385,62 @@ fme_interrupt_handler(void *param) + } + + int +-ifpga_unregister_msix_irq(enum ifpga_irq_type type, ++ifpga_unregister_msix_irq(struct ifpga_rawdev *dev, enum ifpga_irq_type type, + int vec_start, rte_intr_callback_fn handler, void *arg) + { +- struct rte_intr_handle *intr_handle; +- int rc, i; ++ struct rte_intr_handle **intr_handle; ++ int rc = 0; ++ int i = vec_start + 1; ++ ++ if (!dev) ++ return -ENODEV; + + if (type == IFPGA_FME_IRQ) +- intr_handle = ifpga_irq_handle[0]; ++ intr_handle = (struct rte_intr_handle **)&dev->intr_handle[0]; + else if (type == IFPGA_AFU_IRQ) +- intr_handle = ifpga_irq_handle[vec_start + 1]; ++ intr_handle = (struct rte_intr_handle **)&dev->intr_handle[i]; + else +- return 0; ++ return -EINVAL; + +- rte_intr_efd_disable(intr_handle); ++ if ((*intr_handle) == NULL) { ++ IFPGA_RAWDEV_PMD_ERR("%s interrupt %d not registered\n", ++ type == IFPGA_FME_IRQ ? "FME" : "AFU", ++ type == IFPGA_FME_IRQ ? 0 : vec_start); ++ return -ENOENT; ++ } + +- rc = rte_intr_callback_unregister(intr_handle, handler, arg); ++ rte_intr_efd_disable(*intr_handle); ++ ++ rc = rte_intr_callback_unregister(*intr_handle, handler, arg); ++ if (rc < 0) { ++ IFPGA_RAWDEV_PMD_ERR("Failed to unregister %s interrupt %d\n", ++ type == IFPGA_FME_IRQ ? "FME" : "AFU", ++ type == IFPGA_FME_IRQ ? 0 : vec_start); ++ } else { ++ rte_intr_instance_free(*intr_handle); ++ *intr_handle = NULL; ++ } + +- for (i = 0; i < IFPGA_MAX_IRQ; i++) +- rte_intr_instance_free(ifpga_irq_handle[i]); + return rc; + } + + int +-ifpga_register_msix_irq(struct rte_rawdev *dev, int port_id, ++ifpga_register_msix_irq(struct ifpga_rawdev *dev, int port_id, + enum ifpga_irq_type type, int vec_start, int count, + rte_intr_callback_fn handler, const char *name, + void *arg) + { + int ret; +- struct rte_intr_handle *intr_handle; ++ struct rte_intr_handle **intr_handle; + struct opae_adapter *adapter; + struct opae_manager *mgr; + struct opae_accelerator *acc; + int *intr_efds = NULL, nb_intr, i; + +- for (i = 0; i < IFPGA_MAX_IRQ; i++) { +- ifpga_irq_handle[i] = +- rte_intr_instance_alloc(RTE_INTR_INSTANCE_F_PRIVATE); +- if (ifpga_irq_handle[i] == NULL) +- return -ENOMEM; +- } ++ if (!dev || !dev->rawdev) ++ return -ENODEV; + +- adapter = ifpga_rawdev_get_priv(dev); ++ adapter = ifpga_rawdev_get_priv(dev->rawdev); + if (!adapter) + return -ENODEV; + +@@ -1392,32 +1449,40 @@ ifpga_register_msix_irq(struct rte_rawdev *dev, int port_id, + return -ENODEV; + + if (type == IFPGA_FME_IRQ) { +- intr_handle = ifpga_irq_handle[0]; ++ intr_handle = (struct rte_intr_handle **)&dev->intr_handle[0]; + count = 1; + } else if (type == IFPGA_AFU_IRQ) { +- intr_handle = ifpga_irq_handle[vec_start + 1]; ++ i = vec_start + 1; ++ intr_handle = (struct rte_intr_handle **)&dev->intr_handle[i]; + } else { + return -EINVAL; + } + +- if (rte_intr_type_set(intr_handle, RTE_INTR_HANDLE_VFIO_MSIX)) ++ if (*intr_handle) ++ return -EBUSY; ++ ++ *intr_handle = rte_intr_instance_alloc(RTE_INTR_INSTANCE_F_PRIVATE); ++ if (!(*intr_handle)) ++ return -ENOMEM; ++ ++ if (rte_intr_type_set(*intr_handle, RTE_INTR_HANDLE_VFIO_MSIX)) + return -rte_errno; + +- ret = rte_intr_efd_enable(intr_handle, count); ++ ret = rte_intr_efd_enable(*intr_handle, count); + if (ret) + return -ENODEV; + +- if (rte_intr_fd_set(intr_handle, +- rte_intr_efds_index_get(intr_handle, 0))) ++ if (rte_intr_fd_set(*intr_handle, ++ rte_intr_efds_index_get(*intr_handle, 0))) + return -rte_errno; + + IFPGA_RAWDEV_PMD_DEBUG("register %s irq, vfio_fd=%d, fd=%d\n", +- name, rte_intr_dev_fd_get(intr_handle), +- rte_intr_fd_get(intr_handle)); ++ name, rte_intr_dev_fd_get(*intr_handle), ++ rte_intr_fd_get(*intr_handle)); + + if (type == IFPGA_FME_IRQ) { + struct fpga_fme_err_irq_set err_irq_set; +- err_irq_set.evtfd = rte_intr_efds_index_get(intr_handle, ++ err_irq_set.evtfd = rte_intr_efds_index_get(*intr_handle, + 0); + + ret = opae_manager_ifpga_set_err_irq(mgr, &err_irq_set); +@@ -1428,14 +1493,14 @@ ifpga_register_msix_irq(struct rte_rawdev *dev, int port_id, + if (!acc) + return -EINVAL; + +- nb_intr = rte_intr_nb_intr_get(intr_handle); ++ nb_intr = rte_intr_nb_intr_get(*intr_handle); + + intr_efds = calloc(nb_intr, sizeof(int)); + if (!intr_efds) + return -ENOMEM; + + for (i = 0; i < nb_intr; i++) +- intr_efds[i] = rte_intr_efds_index_get(intr_handle, i); ++ intr_efds[i] = rte_intr_efds_index_get(*intr_handle, i); + + ret = opae_acc_set_irq(acc, vec_start, count, intr_efds); + if (ret) { +@@ -1445,7 +1510,7 @@ ifpga_register_msix_irq(struct rte_rawdev *dev, int port_id, + } + + /* register interrupt handler using DPDK API */ +- ret = rte_intr_callback_register(intr_handle, ++ ret = rte_intr_callback_register(*intr_handle, + handler, (void *)arg); + if (ret) { + free(intr_efds); +@@ -1547,11 +1612,15 @@ ifpga_rawdev_create(struct rte_pci_device *pci_dev, + IFPGA_RAWDEV_PMD_INFO("this is a PF function"); + } + +- ret = ifpga_register_msix_irq(rawdev, 0, IFPGA_FME_IRQ, 0, 0, ++ ret = ifpga_register_msix_irq(dev, 0, IFPGA_FME_IRQ, 0, 0, + fme_interrupt_handler, "fme_irq", mgr); + if (ret) + goto free_adapter_data; + ++ ret = ifpga_monitor_start_func(dev); ++ if (ret) ++ goto free_adapter_data; ++ + return ret; + + free_adapter_data: +@@ -1570,9 +1639,6 @@ ifpga_rawdev_destroy(struct rte_pci_device *pci_dev) + int ret; + struct rte_rawdev *rawdev; + char name[RTE_RAWDEV_NAME_MAX_LEN]; +- struct opae_adapter *adapter; +- struct opae_manager *mgr; +- struct ifpga_rawdev *dev; + + if (!pci_dev) { + IFPGA_RAWDEV_PMD_ERR("Invalid pci_dev of the device!"); +@@ -1592,21 +1658,6 @@ ifpga_rawdev_destroy(struct rte_pci_device *pci_dev) + IFPGA_RAWDEV_PMD_ERR("Invalid device name (%s)", name); + return -EINVAL; + } +- dev = ifpga_rawdev_get(rawdev); +- if (dev) +- dev->rawdev = NULL; +- +- adapter = ifpga_rawdev_get_priv(rawdev); +- if (!adapter) +- return -ENODEV; +- +- mgr = opae_adapter_get_mgr(adapter); +- if (!mgr) +- return -ENODEV; +- +- if (ifpga_unregister_msix_irq(IFPGA_FME_IRQ, 0, +- fme_interrupt_handler, mgr) < 0) +- return -EINVAL; + + /* rte_rawdev_close is called by pmd_release */ + ret = rte_rawdev_pmd_release(rawdev); +@@ -1627,7 +1678,7 @@ ifpga_rawdev_pci_probe(struct rte_pci_driver *pci_drv __rte_unused, + static int + ifpga_rawdev_pci_remove(struct rte_pci_device *pci_dev) + { +- ifpga_monitor_stop_func(); ++ IFPGA_RAWDEV_PMD_INFO("remove pci_dev %s", pci_dev->device.name); + return ifpga_rawdev_destroy(pci_dev); + } + +@@ -1669,80 +1720,118 @@ static int ifpga_rawdev_get_string_arg(const char *key __rte_unused, + + return 0; + } ++ + static int +-ifpga_cfg_probe(struct rte_vdev_device *dev) ++ifpga_vdev_parse_devargs(struct rte_devargs *devargs, ++ struct ifpga_vdev_args *args) + { +- struct rte_devargs *devargs; +- struct rte_kvargs *kvlist = NULL; +- struct rte_rawdev *rawdev = NULL; +- struct ifpga_rawdev *ifpga_dev; +- int port; ++ struct rte_kvargs *kvlist; + char *name = NULL; +- const char *bdf; +- char dev_name[RTE_RAWDEV_NAME_MAX_LEN]; +- int ret = -1; ++ int port = 0; ++ int ret = -EINVAL; + +- devargs = dev->device.devargs; ++ if (!devargs || !args) ++ return ret; + + kvlist = rte_kvargs_parse(devargs->args, valid_args); + if (!kvlist) { +- IFPGA_RAWDEV_PMD_LOG(ERR, "error when parsing param"); +- goto end; ++ IFPGA_RAWDEV_PMD_ERR("error when parsing devargs"); ++ return ret; + } + + if (rte_kvargs_count(kvlist, IFPGA_ARG_NAME) == 1) { + if (rte_kvargs_process(kvlist, IFPGA_ARG_NAME, +- &ifpga_rawdev_get_string_arg, +- &name) < 0) { ++ &ifpga_rawdev_get_string_arg, &name) < 0) { + IFPGA_RAWDEV_PMD_ERR("error to parse %s", +- IFPGA_ARG_NAME); ++ IFPGA_ARG_NAME); + goto end; ++ } else { ++ strlcpy(args->bdf, name, sizeof(args->bdf)); ++ rte_free(name); + } + } else { + IFPGA_RAWDEV_PMD_ERR("arg %s is mandatory for ifpga bus", +- IFPGA_ARG_NAME); ++ IFPGA_ARG_NAME); + goto end; + } + + if (rte_kvargs_count(kvlist, IFPGA_ARG_PORT) == 1) { +- if (rte_kvargs_process(kvlist, +- IFPGA_ARG_PORT, +- &rte_ifpga_get_integer32_arg, +- &port) < 0) { ++ if (rte_kvargs_process(kvlist, IFPGA_ARG_PORT, ++ &rte_ifpga_get_integer32_arg, &port) < 0) { + IFPGA_RAWDEV_PMD_ERR("error to parse %s", + IFPGA_ARG_PORT); + goto end; ++ } else { ++ args->port = port; + } + } else { + IFPGA_RAWDEV_PMD_ERR("arg %s is mandatory for ifpga bus", +- IFPGA_ARG_PORT); ++ IFPGA_ARG_PORT); + goto end; + } + ++ ret = 0; ++ ++end: ++ if (kvlist) ++ rte_kvargs_free(kvlist); ++ ++ return ret; ++} ++ ++static int ++ifpga_cfg_probe(struct rte_vdev_device *vdev) ++{ ++ struct rte_rawdev *rawdev = NULL; ++ struct ifpga_rawdev *ifpga_dev; ++ struct ifpga_vdev_args args; ++ char dev_name[RTE_RAWDEV_NAME_MAX_LEN]; ++ const char *vdev_name = NULL; ++ int i, n, ret = 0; ++ ++ vdev_name = rte_vdev_device_name(vdev); ++ if (!vdev_name) ++ return -EINVAL; ++ ++ IFPGA_RAWDEV_PMD_INFO("probe ifpga virtual device %s", vdev_name); ++ ++ ret = ifpga_vdev_parse_devargs(vdev->device.devargs, &args); ++ if (ret) ++ return ret; ++ + memset(dev_name, 0, sizeof(dev_name)); +- snprintf(dev_name, RTE_RAWDEV_NAME_MAX_LEN, "IFPGA:%s", name); ++ snprintf(dev_name, RTE_RAWDEV_NAME_MAX_LEN, "IFPGA:%s", args.bdf); + rawdev = rte_rawdev_pmd_get_named_dev(dev_name); + if (!rawdev) +- goto end; ++ return -ENODEV; + ifpga_dev = ifpga_rawdev_get(rawdev); + if (!ifpga_dev) +- goto end; +- bdf = name; +- ifpga_rawdev_fill_info(ifpga_dev, bdf); ++ return -ENODEV; + +- ifpga_monitor_start_func(); ++ for (i = 0; i < IFPGA_MAX_VDEV; i++) { ++ if (ifpga_dev->vdev_name[i] == NULL) { ++ n = strlen(vdev_name) + 1; ++ ifpga_dev->vdev_name[i] = rte_malloc(NULL, n, 0); ++ if (ifpga_dev->vdev_name[i] == NULL) ++ return -ENOMEM; ++ strlcpy(ifpga_dev->vdev_name[i], vdev_name, n); ++ break; ++ } ++ } + +- memset(dev_name, 0, sizeof(dev_name)); +- snprintf(dev_name, RTE_RAWDEV_NAME_MAX_LEN, "%d|%s", +- port, name); ++ if (i >= IFPGA_MAX_VDEV) { ++ IFPGA_RAWDEV_PMD_ERR("Can't create more virtual device!"); ++ return -ENOENT; ++ } + ++ snprintf(dev_name, RTE_RAWDEV_NAME_MAX_LEN, "%d|%s", ++ args.port, args.bdf); + ret = rte_eal_hotplug_add(RTE_STR(IFPGA_BUS_NAME), +- dev_name, devargs->args); +-end: +- if (kvlist) +- rte_kvargs_free(kvlist); +- if (name) +- free(name); ++ dev_name, vdev->device.devargs->args); ++ if (ret) { ++ rte_free(ifpga_dev->vdev_name[i]); ++ ifpga_dev->vdev_name[i] = NULL; ++ } + + return ret; + } +@@ -1750,10 +1839,47 @@ ifpga_cfg_probe(struct rte_vdev_device *dev) + static int + ifpga_cfg_remove(struct rte_vdev_device *vdev) + { +- IFPGA_RAWDEV_PMD_INFO("Remove ifpga_cfg %p", +- vdev); ++ struct rte_rawdev *rawdev = NULL; ++ struct ifpga_rawdev *ifpga_dev; ++ struct ifpga_vdev_args args; ++ char dev_name[RTE_RAWDEV_NAME_MAX_LEN]; ++ const char *vdev_name = NULL; ++ char *tmp_vdev = NULL; ++ int i, ret = 0; + +- return 0; ++ vdev_name = rte_vdev_device_name(vdev); ++ if (!vdev_name) ++ return -EINVAL; ++ ++ IFPGA_RAWDEV_PMD_INFO("remove ifpga virtual device %s", vdev_name); ++ ++ ret = ifpga_vdev_parse_devargs(vdev->device.devargs, &args); ++ if (ret) ++ return ret; ++ ++ memset(dev_name, 0, sizeof(dev_name)); ++ snprintf(dev_name, RTE_RAWDEV_NAME_MAX_LEN, "IFPGA:%s", args.bdf); ++ rawdev = rte_rawdev_pmd_get_named_dev(dev_name); ++ if (!rawdev) ++ return -ENODEV; ++ ifpga_dev = ifpga_rawdev_get(rawdev); ++ if (!ifpga_dev) ++ return -ENODEV; ++ ++ snprintf(dev_name, RTE_RAWDEV_NAME_MAX_LEN, "%d|%s", ++ args.port, args.bdf); ++ ret = rte_eal_hotplug_remove(RTE_STR(IFPGA_BUS_NAME), dev_name); ++ ++ for (i = 0; i < IFPGA_MAX_VDEV; i++) { ++ tmp_vdev = ifpga_dev->vdev_name[i]; ++ if (tmp_vdev && !strcmp(tmp_vdev, vdev_name)) { ++ free(tmp_vdev); ++ ifpga_dev->vdev_name[i] = NULL; ++ break; ++ } ++ } ++ ++ return ret; + } + + static struct rte_vdev_driver ifpga_cfg_driver = { +diff --git a/dpdk/drivers/raw/ifpga/ifpga_rawdev.h b/dpdk/drivers/raw/ifpga/ifpga_rawdev.h +index 61c8366707..4c191190ca 100644 +--- a/dpdk/drivers/raw/ifpga/ifpga_rawdev.h ++++ b/dpdk/drivers/raw/ifpga/ifpga_rawdev.h +@@ -50,6 +50,8 @@ ifpga_rawdev_get_priv(const struct rte_rawdev *rawdev) + + #define IFPGA_RAWDEV_MSIX_IRQ_NUM 7 + #define IFPGA_RAWDEV_NUM 32 ++#define IFPGA_MAX_VDEV 4 ++#define IFPGA_MAX_IRQ 12 + + struct ifpga_rawdev { + int dev_id; +@@ -59,6 +61,17 @@ struct ifpga_rawdev { + uint32_t aer_old[2]; + char fvl_bdf[8][16]; + char parent_bdf[16]; ++ /* 0 for FME interrupt, others are reserved for AFU irq */ ++ void *intr_handle[IFPGA_MAX_IRQ]; ++ /* enable monitor thread poll device's sensors or not */ ++ int poll_enabled; ++ /* name of virtual devices created on raw device */ ++ char *vdev_name[IFPGA_MAX_VDEV]; ++}; ++ ++struct ifpga_vdev_args { ++ char bdf[PCI_PRI_STR_SIZE]; ++ int port; + }; + + struct ifpga_rawdev * +@@ -70,12 +83,12 @@ enum ifpga_irq_type { + }; + + int +-ifpga_register_msix_irq(struct rte_rawdev *dev, int port_id, ++ifpga_register_msix_irq(struct ifpga_rawdev *dev, int port_id, + enum ifpga_irq_type type, int vec_start, int count, + rte_intr_callback_fn handler, const char *name, + void *arg); + int +-ifpga_unregister_msix_irq(enum ifpga_irq_type type, ++ifpga_unregister_msix_irq(struct ifpga_rawdev *dev, enum ifpga_irq_type type, + int vec_start, rte_intr_callback_fn handler, void *arg); + + struct rte_pci_bus *ifpga_get_pci_bus(void); +diff --git a/dpdk/drivers/raw/ioat/ioat_common.c b/dpdk/drivers/raw/ioat/ioat_common.c +index 60de41152a..e44e181e58 100644 +--- a/dpdk/drivers/raw/ioat/ioat_common.c ++++ b/dpdk/drivers/raw/ioat/ioat_common.c +@@ -9,6 +9,8 @@ + + #include "ioat_private.h" + ++RTE_LOG_REGISTER_DEFAULT(ioat_rawdev_logtype, INFO); ++ + static const char * const xstat_names[] = { + "failed_enqueues", "successful_enqueues", + "copies_started", "copies_completed" +diff --git a/dpdk/drivers/raw/ioat/ioat_private.h b/dpdk/drivers/raw/ioat/ioat_private.h +index f032d5fe3d..7fb685a3d7 100644 +--- a/dpdk/drivers/raw/ioat/ioat_private.h ++++ b/dpdk/drivers/raw/ioat/ioat_private.h +@@ -18,10 +18,10 @@ + #include + #include "rte_ioat_rawdev.h" + +-extern int ioat_pmd_logtype; ++extern int ioat_rawdev_logtype; + + #define IOAT_PMD_LOG(level, fmt, args...) rte_log(RTE_LOG_ ## level, \ +- ioat_pmd_logtype, "IOAT: %s(): " fmt "\n", __func__, ##args) ++ ioat_rawdev_logtype, "IOAT: %s(): " fmt "\n", __func__, ##args) + + #define IOAT_PMD_DEBUG(fmt, args...) IOAT_PMD_LOG(DEBUG, fmt, ## args) + #define IOAT_PMD_INFO(fmt, args...) IOAT_PMD_LOG(INFO, fmt, ## args) +diff --git a/dpdk/drivers/raw/ioat/ioat_rawdev.c b/dpdk/drivers/raw/ioat/ioat_rawdev.c +index 5396671d4f..11341fcf5d 100644 +--- a/dpdk/drivers/raw/ioat/ioat_rawdev.c ++++ b/dpdk/drivers/raw/ioat/ioat_rawdev.c +@@ -28,8 +28,6 @@ static struct rte_pci_driver ioat_pmd_drv; + #define IOAT_DEVICE_ID_BDXF 0x6f2F + #define IOAT_DEVICE_ID_ICX 0x0b00 + +-RTE_LOG_REGISTER_DEFAULT(ioat_pmd_logtype, INFO); +- + #define DESC_SZ sizeof(struct rte_ioat_generic_hw_desc) + #define COMPLETION_SZ sizeof(__m128i) + +diff --git a/dpdk/drivers/raw/ioat/rte_idxd_rawdev_fns.h b/dpdk/drivers/raw/ioat/rte_idxd_rawdev_fns.h +index 3ea25f6ca9..a07892faa0 100644 +--- a/dpdk/drivers/raw/ioat/rte_idxd_rawdev_fns.h ++++ b/dpdk/drivers/raw/ioat/rte_idxd_rawdev_fns.h +@@ -16,6 +16,7 @@ + */ + + #include ++#include + + /* + * Defines used in the data path for interacting with IDXD hardware. +diff --git a/dpdk/drivers/raw/ntb/ntb.c b/dpdk/drivers/raw/ntb/ntb.c +index 46ac02e5ab..f5e773c53b 100644 +--- a/dpdk/drivers/raw/ntb/ntb.c ++++ b/dpdk/drivers/raw/ntb/ntb.c +@@ -1398,6 +1398,10 @@ ntb_init_hw(struct rte_rawdev *dev, struct rte_pci_device *pci_dev) + + /* Init doorbell. */ + hw->db_valid_mask = RTE_LEN2MASK(hw->db_cnt, uint64_t); ++ /* Clear all valid doorbell bits before registering intr handler */ ++ if (hw->ntb_ops->db_clear == NULL) ++ return -ENOTSUP; ++ (*hw->ntb_ops->db_clear)(dev, hw->db_valid_mask); + + intr_handle = pci_dev->intr_handle; + /* Register callback func to eal lib */ +diff --git a/dpdk/drivers/raw/ntb/ntb.h b/dpdk/drivers/raw/ntb/ntb.h +index cdf7667d5d..c9ff33aa59 100644 +--- a/dpdk/drivers/raw/ntb/ntb.h ++++ b/dpdk/drivers/raw/ntb/ntb.h +@@ -95,7 +95,7 @@ enum ntb_spad_idx { + * @spad_write: Write val to local/peer spad register. + * @db_read: Read doorbells status. + * @db_clear: Clear local doorbells. +- * @db_set_mask: Set bits in db mask, preventing db interrpts generated ++ * @db_set_mask: Set bits in db mask, preventing db interrupts generated + * for those db bits. + * @peer_db_set: Set doorbell bit to generate peer interrupt for that bit. + * @vector_bind: Bind vector source [intr] to msix vector [msix]. +diff --git a/dpdk/drivers/regex/mlx5/mlx5_rxp.c b/dpdk/drivers/regex/mlx5/mlx5_rxp.c +index 5ead3ca341..51b6e71376 100644 +--- a/dpdk/drivers/regex/mlx5/mlx5_rxp.c ++++ b/dpdk/drivers/regex/mlx5/mlx5_rxp.c +@@ -148,7 +148,7 @@ mlx5_regex_configure(struct rte_regexdev *dev, + dev->data->dev_conf.nb_queue_pairs = priv->nb_queues; + priv->qps = rte_zmalloc(NULL, sizeof(struct mlx5_regex_qp) * + priv->nb_queues, 0); +- if (!priv->nb_queues) { ++ if (!priv->qps) { + DRV_LOG(ERR, "can't allocate qps memory"); + rte_errno = ENOMEM; + return -rte_errno; +diff --git a/dpdk/drivers/vdpa/ifc/base/ifcvf.c b/dpdk/drivers/vdpa/ifc/base/ifcvf.c +index 721cb1da8a..f3c29f94b3 100644 +--- a/dpdk/drivers/vdpa/ifc/base/ifcvf.c ++++ b/dpdk/drivers/vdpa/ifc/base/ifcvf.c +@@ -87,6 +87,8 @@ ifcvf_init_hw(struct ifcvf_hw *hw, PCI_DEV *dev) + } + + hw->lm_cfg = hw->mem_resource[4].addr; ++ if (!hw->lm_cfg) ++ WARNINGOUT("HW support live migration not support!\n"); + + if (hw->common_cfg == NULL || hw->notify_base == NULL || + hw->isr == NULL || hw->dev_cfg == NULL) { +@@ -94,12 +96,14 @@ ifcvf_init_hw(struct ifcvf_hw *hw, PCI_DEV *dev) + return -1; + } + +- DEBUGOUT("capability mapping:\ncommon cfg: %p\n" +- "notify base: %p\nisr cfg: %p\ndevice cfg: %p\n" +- "multiplier: %u\n", +- hw->common_cfg, hw->dev_cfg, +- hw->isr, hw->notify_base, +- hw->notify_off_multiplier); ++ DEBUGOUT("capability mapping:\n" ++ "common cfg: %p\n" ++ "notify base: %p\n" ++ "isr cfg: %p\n" ++ "device cfg: %p\n" ++ "multiplier: %u\n", ++ hw->common_cfg, hw->notify_base, hw->isr, hw->dev_cfg, ++ hw->notify_off_multiplier); + + return 0; + } +@@ -216,10 +220,12 @@ ifcvf_hw_enable(struct ifcvf_hw *hw) + &cfg->queue_used_hi); + IFCVF_WRITE_REG16(hw->vring[i].size, &cfg->queue_size); + +- *(u32 *)(lm_cfg + IFCVF_LM_RING_STATE_OFFSET + +- (i / 2) * IFCVF_LM_CFG_SIZE + (i % 2) * 4) = +- (u32)hw->vring[i].last_avail_idx | +- ((u32)hw->vring[i].last_used_idx << 16); ++ if (lm_cfg) { ++ *(u32 *)(lm_cfg + IFCVF_LM_RING_STATE_OFFSET + ++ (i / 2) * IFCVF_LM_CFG_SIZE + (i % 2) * 4) = ++ (u32)hw->vring[i].last_avail_idx | ++ ((u32)hw->vring[i].last_used_idx << 16); ++ } + + IFCVF_WRITE_REG16(i + 1, &cfg->queue_msix_vector); + if (IFCVF_READ_REG16(&cfg->queue_msix_vector) == +@@ -289,6 +295,8 @@ ifcvf_enable_logging(struct ifcvf_hw *hw, u64 log_base, u64 log_size) + u8 *lm_cfg; + + lm_cfg = hw->lm_cfg; ++ if (!lm_cfg) ++ return; + + *(u32 *)(lm_cfg + IFCVF_LM_BASE_ADDR_LOW) = + log_base & IFCVF_32_BIT_MASK; +@@ -311,6 +319,9 @@ ifcvf_disable_logging(struct ifcvf_hw *hw) + u8 *lm_cfg; + + lm_cfg = hw->lm_cfg; ++ if (!lm_cfg) ++ return; ++ + *(u32 *)(lm_cfg + IFCVF_LM_LOGGING_CTRL) = IFCVF_LM_DISABLE; + } + +diff --git a/dpdk/drivers/vdpa/ifc/base/ifcvf_osdep.h b/dpdk/drivers/vdpa/ifc/base/ifcvf_osdep.h +index 6aef25ea45..3d567695cc 100644 +--- a/dpdk/drivers/vdpa/ifc/base/ifcvf_osdep.h ++++ b/dpdk/drivers/vdpa/ifc/base/ifcvf_osdep.h +@@ -14,6 +14,7 @@ + #include + #include + ++#define WARNINGOUT(S, args...) RTE_LOG(WARNING, PMD, S, ##args) + #define DEBUGOUT(S, args...) RTE_LOG(DEBUG, PMD, S, ##args) + #define STATIC static + +diff --git a/dpdk/drivers/vdpa/ifc/ifcvf_vdpa.c b/dpdk/drivers/vdpa/ifc/ifcvf_vdpa.c +index 3853c4cf7e..6a915b0d5e 100644 +--- a/dpdk/drivers/vdpa/ifc/ifcvf_vdpa.c ++++ b/dpdk/drivers/vdpa/ifc/ifcvf_vdpa.c +@@ -356,6 +356,8 @@ vdpa_enable_vfio_intr(struct ifcvf_internal *internal, bool m_rx) + vring.callfd = -1; + + nr_vring = rte_vhost_get_vring_num(internal->vid); ++ if (nr_vring > IFCVF_MAX_QUEUES * 2) ++ return -1; + + irq_set = (struct vfio_irq_set *)irq_set_buf; + irq_set->argsz = sizeof(irq_set_buf); +diff --git a/dpdk/drivers/vdpa/mlx5/mlx5_vdpa.c b/dpdk/drivers/vdpa/mlx5/mlx5_vdpa.c +index b9e84dd9bf..9c1c70037c 100644 +--- a/dpdk/drivers/vdpa/mlx5/mlx5_vdpa.c ++++ b/dpdk/drivers/vdpa/mlx5/mlx5_vdpa.c +@@ -81,7 +81,7 @@ mlx5_vdpa_get_queue_num(struct rte_vdpa_device *vdev, uint32_t *queue_num) + DRV_LOG(ERR, "Invalid vDPA device: %s.", vdev->device->name); + return -1; + } +- *queue_num = priv->caps.max_num_virtio_queues; ++ *queue_num = priv->caps.max_num_virtio_queues / 2; + return 0; + } + +@@ -138,7 +138,7 @@ mlx5_vdpa_set_vring_state(int vid, int vring, int state) + DRV_LOG(ERR, "Invalid vDPA device: %s.", vdev->device->name); + return -EINVAL; + } +- if (vring >= (int)priv->caps.max_num_virtio_queues * 2) { ++ if (vring >= (int)priv->caps.max_num_virtio_queues) { + DRV_LOG(ERR, "Too big vring id: %d.", vring); + return -E2BIG; + } +@@ -250,10 +250,10 @@ mlx5_vdpa_dev_close(int vid) + DRV_LOG(ERR, "Invalid vDPA device: %s.", vdev->device->name); + return -1; + } +- if (priv->configured) +- ret |= mlx5_vdpa_lm_log(priv); + mlx5_vdpa_err_event_unset(priv); + mlx5_vdpa_cqe_event_unset(priv); ++ if (priv->configured) ++ ret |= mlx5_vdpa_lm_log(priv); + mlx5_vdpa_steer_unset(priv); + mlx5_vdpa_virtqs_release(priv); + mlx5_vdpa_event_qp_global_release(priv); +@@ -518,7 +518,7 @@ mlx5_vdpa_dev_probe(struct mlx5_common_device *cdev) + DRV_LOG(DEBUG, "No capability to support virtq statistics."); + priv = rte_zmalloc("mlx5 vDPA device private", sizeof(*priv) + + sizeof(struct mlx5_vdpa_virtq) * +- attr->vdpa.max_num_virtio_queues * 2, ++ attr->vdpa.max_num_virtio_queues, + RTE_CACHE_LINE_SIZE); + if (!priv) { + DRV_LOG(ERR, "Failed to allocate private memory."); +diff --git a/dpdk/drivers/vdpa/mlx5/mlx5_vdpa_event.c b/dpdk/drivers/vdpa/mlx5/mlx5_vdpa_event.c +index f8d910b33f..ecad139768 100644 +--- a/dpdk/drivers/vdpa/mlx5/mlx5_vdpa_event.c ++++ b/dpdk/drivers/vdpa/mlx5/mlx5_vdpa_event.c +@@ -460,6 +460,7 @@ mlx5_vdpa_cqe_event_setup(struct mlx5_vdpa_priv *priv) + { + int ret; + rte_cpuset_t cpuset; ++ pthread_attr_t *attrp = NULL; + pthread_attr_t attr; + char name[16]; + const struct sched_param sp = { +@@ -469,22 +470,27 @@ mlx5_vdpa_cqe_event_setup(struct mlx5_vdpa_priv *priv) + if (!priv->eventc) + /* All virtqs are in poll mode. */ + return 0; +- pthread_attr_init(&attr); +- ret = pthread_attr_setschedpolicy(&attr, SCHED_RR); ++ ret = pthread_attr_init(&attr); ++ if (ret != 0) { ++ DRV_LOG(ERR, "Failed to initialize thread attributes"); ++ goto out; ++ } ++ attrp = &attr; ++ ret = pthread_attr_setschedpolicy(attrp, SCHED_RR); + if (ret) { + DRV_LOG(ERR, "Failed to set thread sched policy = RR."); +- return -1; ++ goto out; + } +- ret = pthread_attr_setschedparam(&attr, &sp); ++ ret = pthread_attr_setschedparam(attrp, &sp); + if (ret) { + DRV_LOG(ERR, "Failed to set thread priority."); +- return -1; ++ goto out; + } +- ret = pthread_create(&priv->timer_tid, &attr, mlx5_vdpa_event_handle, ++ ret = pthread_create(&priv->timer_tid, attrp, mlx5_vdpa_event_handle, + (void *)priv); + if (ret) { + DRV_LOG(ERR, "Failed to create timer thread."); +- return -1; ++ goto out; + } + CPU_ZERO(&cpuset); + if (priv->event_core != -1) +@@ -494,12 +500,16 @@ mlx5_vdpa_cqe_event_setup(struct mlx5_vdpa_priv *priv) + ret = pthread_setaffinity_np(priv->timer_tid, sizeof(cpuset), &cpuset); + if (ret) { + DRV_LOG(ERR, "Failed to set thread affinity."); +- return -1; ++ goto out; + } + snprintf(name, sizeof(name), "vDPA-mlx5-%d", priv->vid); +- ret = rte_thread_setname(priv->timer_tid, name); +- if (ret) ++ if (rte_thread_setname(priv->timer_tid, name) != 0) + DRV_LOG(DEBUG, "Cannot set timer thread name."); ++out: ++ if (attrp != NULL) ++ pthread_attr_destroy(attrp); ++ if (ret != 0) ++ return -1; + return 0; + } + +diff --git a/dpdk/drivers/vdpa/mlx5/mlx5_vdpa_mem.c b/dpdk/drivers/vdpa/mlx5/mlx5_vdpa_mem.c +index b1b9053bff..130d201a85 100644 +--- a/dpdk/drivers/vdpa/mlx5/mlx5_vdpa_mem.c ++++ b/dpdk/drivers/vdpa/mlx5/mlx5_vdpa_mem.c +@@ -160,7 +160,7 @@ mlx5_vdpa_vhost_mem_regions_prepare(int vid, uint8_t *mode, uint64_t *mem_size, + * The target here is to group all the physical memory regions of the + * virtio device in one indirect mkey. + * For KLM Fixed Buffer Size mode (HW find the translation entry in one +- * read according to the guest phisical address): ++ * read according to the guest physical address): + * All the sub-direct mkeys of it must be in the same size, hence, each + * one of them should be in the GCD size of all the virtio memory + * regions and the holes between them. +diff --git a/dpdk/drivers/vdpa/mlx5/mlx5_vdpa_virtq.c b/dpdk/drivers/vdpa/mlx5/mlx5_vdpa_virtq.c +index db971bad48..98d788d734 100644 +--- a/dpdk/drivers/vdpa/mlx5/mlx5_vdpa_virtq.c ++++ b/dpdk/drivers/vdpa/mlx5/mlx5_vdpa_virtq.c +@@ -9,6 +9,7 @@ + #include + #include + #include ++#include + + #include + +@@ -17,17 +18,17 @@ + + + static void +-mlx5_vdpa_virtq_handler(void *cb_arg) ++mlx5_vdpa_virtq_kick_handler(void *cb_arg) + { + struct mlx5_vdpa_virtq *virtq = cb_arg; + struct mlx5_vdpa_priv *priv = virtq->priv; + uint64_t buf; + int nbytes; ++ int retry; + + if (rte_intr_fd_get(virtq->intr_handle) < 0) + return; +- +- do { ++ for (retry = 0; retry < 3; ++retry) { + nbytes = read(rte_intr_fd_get(virtq->intr_handle), &buf, + 8); + if (nbytes < 0) { +@@ -39,7 +40,9 @@ mlx5_vdpa_virtq_handler(void *cb_arg) + virtq->index, strerror(errno)); + } + break; +- } while (1); ++ } ++ if (nbytes < 0) ++ return; + rte_write32(virtq->index, priv->virtq_db_addr); + if (virtq->notifier_state == MLX5_VDPA_NOTIFIER_STATE_DISABLED) { + if (rte_vhost_host_notifier_ctrl(priv->vid, virtq->index, true)) +@@ -59,20 +62,16 @@ static int + mlx5_vdpa_virtq_unset(struct mlx5_vdpa_virtq *virtq) + { + unsigned int i; +- int retries = MLX5_VDPA_INTR_RETRIES; + int ret = -EAGAIN; + +- if (rte_intr_fd_get(virtq->intr_handle) != -1) { +- while (retries-- && ret == -EAGAIN) { ++ if (rte_intr_fd_get(virtq->intr_handle) >= 0) { ++ while (ret == -EAGAIN) { + ret = rte_intr_callback_unregister(virtq->intr_handle, +- mlx5_vdpa_virtq_handler, +- virtq); ++ mlx5_vdpa_virtq_kick_handler, virtq); + if (ret == -EAGAIN) { +- DRV_LOG(DEBUG, "Try again to unregister fd %d " +- "of virtq %d interrupt, retries = %d.", +- rte_intr_fd_get(virtq->intr_handle), +- (int)virtq->index, retries); +- ++ DRV_LOG(DEBUG, "Try again to unregister fd %d of virtq %hu interrupt", ++ rte_intr_fd_get(virtq->intr_handle), ++ virtq->index); + usleep(MLX5_VDPA_INTR_RETRIES_USEC); + } + } +@@ -124,7 +123,9 @@ mlx5_vdpa_virtqs_release(struct mlx5_vdpa_priv *priv) + priv->td = NULL; + } + if (priv->virtq_db_addr) { +- claim_zero(munmap(priv->virtq_db_addr, priv->var->length)); ++ /* Mask out the within page offset for munmap. */ ++ claim_zero(munmap((void *)((uintptr_t)priv->virtq_db_addr & ++ ~(rte_mem_page_size() - 1)), priv->var->length)); + priv->virtq_db_addr = NULL; + } + priv->features = 0; +@@ -360,7 +361,7 @@ mlx5_vdpa_virtq_setup(struct mlx5_vdpa_priv *priv, int index) + goto error; + + if (rte_intr_callback_register(virtq->intr_handle, +- mlx5_vdpa_virtq_handler, ++ mlx5_vdpa_virtq_kick_handler, + virtq)) { + rte_intr_fd_set(virtq->intr_handle, -1); + DRV_LOG(ERR, "Failed to register virtq %d interrupt.", +@@ -403,7 +404,7 @@ mlx5_vdpa_features_validate(struct mlx5_vdpa_priv *priv) + if (priv->features & (1ULL << VIRTIO_F_RING_PACKED)) { + if (!(priv->caps.virtio_queue_type & (1 << + MLX5_VIRTQ_TYPE_PACKED))) { +- DRV_LOG(ERR, "Failed to configur PACKED mode for vdev " ++ DRV_LOG(ERR, "Failed to configure PACKED mode for vdev " + "%d - it was not reported by HW/driver" + " capability.", priv->vid); + return -ENOTSUP; +@@ -472,9 +473,9 @@ mlx5_vdpa_virtqs_prepare(struct mlx5_vdpa_priv *priv) + DRV_LOG(INFO, "TSO is enabled without CSUM, force CSUM."); + priv->features |= (1ULL << VIRTIO_NET_F_CSUM); + } +- if (nr_vring > priv->caps.max_num_virtio_queues * 2) { ++ if (nr_vring > priv->caps.max_num_virtio_queues) { + DRV_LOG(ERR, "Do not support more than %d virtqs(%d).", +- (int)priv->caps.max_num_virtio_queues * 2, ++ (int)priv->caps.max_num_virtio_queues, + (int)nr_vring); + return -1; + } +@@ -487,6 +488,10 @@ mlx5_vdpa_virtqs_prepare(struct mlx5_vdpa_priv *priv) + priv->virtq_db_addr = NULL; + goto error; + } else { ++ /* Add within page offset for 64K page system. */ ++ priv->virtq_db_addr = (char *)priv->virtq_db_addr + ++ ((rte_mem_page_size() - 1) & ++ priv->caps.doorbell_bar_offset); + DRV_LOG(DEBUG, "VAR address of doorbell mapping is %p.", + priv->virtq_db_addr); + } +diff --git a/dpdk/drivers/vdpa/sfc/sfc_vdpa.c b/dpdk/drivers/vdpa/sfc/sfc_vdpa.c +index fccdd8c687..53f598facc 100644 +--- a/dpdk/drivers/vdpa/sfc/sfc_vdpa.c ++++ b/dpdk/drivers/vdpa/sfc/sfc_vdpa.c +@@ -328,7 +328,8 @@ sfc_vdpa_pci_remove(struct rte_pci_device *pci_dev) + + sva = sfc_vdpa_get_adapter_by_dev(pci_dev); + if (sva == NULL) { +- sfc_vdpa_info(sva, "invalid device: %s", pci_dev->name); ++ SFC_VDPA_GENERIC_LOG(INFO, ++ "Invalid device: %s.", pci_dev->name); + return -1; + } + +diff --git a/dpdk/drivers/vdpa/sfc/sfc_vdpa_ops.c b/dpdk/drivers/vdpa/sfc/sfc_vdpa_ops.c +index c4ce4474ef..b84699d234 100644 +--- a/dpdk/drivers/vdpa/sfc/sfc_vdpa_ops.c ++++ b/dpdk/drivers/vdpa/sfc/sfc_vdpa_ops.c +@@ -222,6 +222,7 @@ static int + sfc_vdpa_virtq_start(struct sfc_vdpa_ops_data *ops_data, int vq_num) + { + int rc; ++ uint32_t doorbell; + efx_virtio_vq_t *vq; + struct sfc_vdpa_vring_info vring; + efx_virtio_vq_cfg_t vq_cfg; +@@ -270,22 +271,35 @@ sfc_vdpa_virtq_start(struct sfc_vdpa_ops_data *ops_data, int vq_num) + /* Start virtqueue */ + rc = efx_virtio_qstart(vq, &vq_cfg, &vq_dyncfg); + if (rc != 0) { +- /* destroy virtqueue */ + sfc_vdpa_err(ops_data->dev_handle, + "virtqueue start failed: %s", + rte_strerror(rc)); +- efx_virtio_qdestroy(vq); + goto fail_virtio_qstart; + } + + sfc_vdpa_info(ops_data->dev_handle, + "virtqueue started successfully for vq_num %d", vq_num); + ++ rc = efx_virtio_get_doorbell_offset(vq, &doorbell); ++ if (rc != 0) { ++ sfc_vdpa_err(ops_data->dev_handle, ++ "failed to get doorbell offset: %s", ++ rte_strerror(rc)); ++ goto fail_doorbell; ++ } ++ ++ /* ++ * Cache the bar_offset here for each VQ here, it will come ++ * in handy when sfc_vdpa_get_notify_area() is invoked. ++ */ ++ ops_data->vq_cxt[vq_num].doorbell = (void *)(uintptr_t)doorbell; + ops_data->vq_cxt[vq_num].enable = B_TRUE; + + return rc; + ++fail_doorbell: + fail_virtio_qstart: ++ efx_virtio_qdestroy(vq); + fail_vring_info: + return rc; + } +@@ -611,7 +625,7 @@ sfc_vdpa_dev_config(int vid) + + ops_data = sfc_vdpa_get_data_by_dev(vdpa_dev); + if (ops_data == NULL) { +- sfc_vdpa_err(ops_data->dev_handle, ++ SFC_VDPA_GENERIC_LOG(ERR, + "invalid vDPA device : %p, vid : %d", + vdpa_dev, vid); + return -1; +@@ -666,7 +680,7 @@ sfc_vdpa_dev_close(int vid) + + ops_data = sfc_vdpa_get_data_by_dev(vdpa_dev); + if (ops_data == NULL) { +- sfc_vdpa_err(ops_data->dev_handle, ++ SFC_VDPA_GENERIC_LOG(ERR, + "invalid vDPA device : %p, vid : %d", + vdpa_dev, vid); + return -1; +@@ -792,8 +806,8 @@ sfc_vdpa_get_notify_area(int vid, int qid, uint64_t *offset, uint64_t *size) + int ret; + efx_nic_t *nic; + int vfio_dev_fd; +- efx_rc_t rc; +- unsigned int bar_offset; ++ volatile void *doorbell; ++ struct rte_pci_device *pci_dev; + struct rte_vdpa_device *vdpa_dev; + struct sfc_vdpa_ops_data *ops_data; + struct vfio_region_info reg = { .argsz = sizeof(reg) }; +@@ -822,19 +836,6 @@ sfc_vdpa_get_notify_area(int vid, int qid, uint64_t *offset, uint64_t *size) + return -1; + } + +- if (ops_data->vq_cxt[qid].enable != B_TRUE) { +- sfc_vdpa_err(dev, "vq is not enabled"); +- return -1; +- } +- +- rc = efx_virtio_get_doorbell_offset(ops_data->vq_cxt[qid].vq, +- &bar_offset); +- if (rc != 0) { +- sfc_vdpa_err(dev, "failed to get doorbell offset: %s", +- rte_strerror(rc)); +- return rc; +- } +- + reg.index = sfc_vdpa_adapter_by_dev_handle(dev)->mem_bar.esb_rid; + ret = ioctl(vfio_dev_fd, VFIO_DEVICE_GET_REGION_INFO, ®); + if (ret != 0) { +@@ -843,7 +844,8 @@ sfc_vdpa_get_notify_area(int vid, int qid, uint64_t *offset, uint64_t *size) + return ret; + } + +- *offset = reg.offset + bar_offset; ++ /* Use bar_offset that was cached during sfc_vdpa_virtq_start() */ ++ *offset = reg.offset + (uint64_t)ops_data->vq_cxt[qid].doorbell; + + len = (1U << encp->enc_vi_window_shift) / 2; + if (len >= sysconf(_SC_PAGESIZE)) { +@@ -856,6 +858,18 @@ sfc_vdpa_get_notify_area(int vid, int qid, uint64_t *offset, uint64_t *size) + sfc_vdpa_info(dev, "vDPA ops get_notify_area :: offset : 0x%" PRIx64, + *offset); + ++ pci_dev = sfc_vdpa_adapter_by_dev_handle(dev)->pdev; ++ doorbell = (uint8_t *)pci_dev->mem_resource[reg.index].addr + *offset; ++ ++ /* ++ * virtio-net driver in VM sends queue notifications before ++ * vDPA has a chance to setup the queues and notification area, ++ * and hence the HW misses these doorbell notifications. ++ * Since, it is safe to send duplicate doorbell, send another ++ * doorbell from vDPA driver as workaround for this timing issue. ++ */ ++ rte_write16(qid, doorbell); ++ + return 0; + } + +diff --git a/dpdk/drivers/vdpa/sfc/sfc_vdpa_ops.h b/dpdk/drivers/vdpa/sfc/sfc_vdpa_ops.h +index 6d790fd8be..9dbd5b84dd 100644 +--- a/dpdk/drivers/vdpa/sfc/sfc_vdpa_ops.h ++++ b/dpdk/drivers/vdpa/sfc/sfc_vdpa_ops.h +@@ -35,6 +35,7 @@ struct sfc_vdpa_vring_info { + }; + + typedef struct sfc_vdpa_vq_context_s { ++ volatile void *doorbell; + uint8_t enable; + uint32_t pidx; + uint32_t cidx; +diff --git a/dpdk/examples/bbdev_app/main.c b/dpdk/examples/bbdev_app/main.c +index ecafc5e4f1..fc7e8b8174 100644 +--- a/dpdk/examples/bbdev_app/main.c ++++ b/dpdk/examples/bbdev_app/main.c +@@ -372,7 +372,7 @@ add_awgn(struct rte_mbuf **mbufs, uint16_t num_pkts) + /* Encoder output to Decoder input adapter. The Decoder accepts only soft input + * so each bit of the encoder output must be translated into one byte of LLR. If + * Sub-block Deinterleaver is bypassed, which is the case, the padding bytes +- * must additionally be insterted at the end of each sub-block. ++ * must additionally be inserted at the end of each sub-block. + */ + static inline void + transform_enc_out_dec_in(struct rte_mbuf **mbufs, uint8_t *temp_buf, +diff --git a/dpdk/examples/bond/main.c b/dpdk/examples/bond/main.c +index 1087b0dad1..4efebb3902 100644 +--- a/dpdk/examples/bond/main.c ++++ b/dpdk/examples/bond/main.c +@@ -230,7 +230,7 @@ bond_port_init(struct rte_mempool *mbuf_pool) + 0 /*SOCKET_ID_ANY*/); + if (retval < 0) + rte_exit(EXIT_FAILURE, +- "Faled to create bond port\n"); ++ "Failed to create bond port\n"); + + BOND_PORT = retval; + +@@ -373,7 +373,7 @@ static int lcore_main(__rte_unused void *arg1) + bond_ip = BOND_IP_1 | (BOND_IP_2 << 8) | + (BOND_IP_3 << 16) | (BOND_IP_4 << 24); + +- rte_spinlock_trylock(&global_flag_stru_p->lock); ++ rte_spinlock_lock(&global_flag_stru_p->lock); + + while (global_flag_stru_p->LcoreMainIsRunning) { + rte_spinlock_unlock(&global_flag_stru_p->lock); +@@ -405,7 +405,7 @@ static int lcore_main(__rte_unused void *arg1) + struct rte_ether_hdr *); + ether_type = eth_hdr->ether_type; + if (ether_type == rte_cpu_to_be_16(RTE_ETHER_TYPE_VLAN)) +- printf("VLAN taged frame, offset:"); ++ printf("VLAN tagged frame, offset:"); + offset = get_vlan_offset(eth_hdr, ðer_type); + if (offset > 0) + printf("%d\n", offset); +@@ -456,7 +456,7 @@ static int lcore_main(__rte_unused void *arg1) + if (is_free == 0) + rte_pktmbuf_free(pkts[i]); + } +- rte_spinlock_trylock(&global_flag_stru_p->lock); ++ rte_spinlock_lock(&global_flag_stru_p->lock); + } + rte_spinlock_unlock(&global_flag_stru_p->lock); + printf("BYE lcore_main\n"); +@@ -571,7 +571,7 @@ static void cmd_start_parsed(__rte_unused void *parsed_result, + { + int worker_core_id = rte_lcore_id(); + +- rte_spinlock_trylock(&global_flag_stru_p->lock); ++ rte_spinlock_lock(&global_flag_stru_p->lock); + if (global_flag_stru_p->LcoreMainIsRunning == 0) { + if (rte_eal_get_lcore_state(global_flag_stru_p->LcoreMainCore) + != WAIT) { +@@ -591,7 +591,7 @@ static void cmd_start_parsed(__rte_unused void *parsed_result, + if ((worker_core_id >= RTE_MAX_LCORE) || (worker_core_id == 0)) + return; + +- rte_spinlock_trylock(&global_flag_stru_p->lock); ++ rte_spinlock_lock(&global_flag_stru_p->lock); + global_flag_stru_p->LcoreMainIsRunning = 1; + rte_spinlock_unlock(&global_flag_stru_p->lock); + cmdline_printf(cl, +@@ -659,7 +659,7 @@ static void cmd_stop_parsed(__rte_unused void *parsed_result, + struct cmdline *cl, + __rte_unused void *data) + { +- rte_spinlock_trylock(&global_flag_stru_p->lock); ++ rte_spinlock_lock(&global_flag_stru_p->lock); + if (global_flag_stru_p->LcoreMainIsRunning == 0) { + cmdline_printf(cl, + "lcore_main not running on core:%d\n", +@@ -700,7 +700,7 @@ static void cmd_quit_parsed(__rte_unused void *parsed_result, + struct cmdline *cl, + __rte_unused void *data) + { +- rte_spinlock_trylock(&global_flag_stru_p->lock); ++ rte_spinlock_lock(&global_flag_stru_p->lock); + if (global_flag_stru_p->LcoreMainIsRunning == 0) { + cmdline_printf(cl, + "lcore_main not running on core:%d\n", +@@ -762,7 +762,7 @@ static void cmd_show_parsed(__rte_unused void *parsed_result, + printf("\n"); + } + +- rte_spinlock_trylock(&global_flag_stru_p->lock); ++ rte_spinlock_lock(&global_flag_stru_p->lock); + cmdline_printf(cl, + "Active_slaves:%d " + "packets received:Tot:%d Arp:%d IPv4:%d\n", +diff --git a/dpdk/examples/distributor/main.c b/dpdk/examples/distributor/main.c +index c681e237ea..8995806b4e 100644 +--- a/dpdk/examples/distributor/main.c ++++ b/dpdk/examples/distributor/main.c +@@ -108,7 +108,7 @@ static inline int + port_init(uint16_t port, struct rte_mempool *mbuf_pool) + { + struct rte_eth_conf port_conf = port_conf_default; +- const uint16_t rxRings = 1, txRings = rte_lcore_count() - 1; ++ const uint16_t rxRings = 1, txRings = 1; + int retval; + uint16_t q; + uint16_t nb_rxd = RX_RING_SIZE; +@@ -261,8 +261,8 @@ lcore_rx(struct lcore_params *p) + * packets are then send straight to the tx core. + */ + #if 0 +- rte_distributor_process(d, bufs, nb_rx); +- const uint16_t nb_ret = rte_distributor_returned_pktsd, ++ rte_distributor_process(p->d, bufs, nb_rx); ++ const uint16_t nb_ret = rte_distributor_returned_pkts(p->d, + bufs, BURST_SIZE*2); + + app_stats.rx.returned_pkts += nb_ret; +diff --git a/dpdk/examples/dma/dmafwd.c b/dpdk/examples/dma/dmafwd.c +index d074acc905..dd576bcf77 100644 +--- a/dpdk/examples/dma/dmafwd.c ++++ b/dpdk/examples/dma/dmafwd.c +@@ -87,7 +87,7 @@ static uint16_t nb_queues = 1; + /* MAC updating enabled by default. */ + static int mac_updating = 1; + +-/* hardare copy mode enabled by default. */ ++/* hardware copy mode enabled by default. */ + static copy_mode_t copy_mode = COPY_MODE_DMA_NUM; + + /* size of descriptor ring for hardware copy mode or +@@ -117,12 +117,11 @@ static uint16_t nb_txd = TX_DEFAULT_RINGSIZE; + static volatile bool force_quit; + + static uint32_t dma_batch_sz = MAX_PKT_BURST; +-static uint32_t max_frame_size = RTE_ETHER_MAX_LEN; ++static uint32_t max_frame_size; + + /* ethernet addresses of ports */ + static struct rte_ether_addr dma_ports_eth_addr[RTE_MAX_ETHPORTS]; + +-static struct rte_eth_dev_tx_buffer *tx_buffer[RTE_MAX_ETHPORTS]; + struct rte_mempool *dma_pktmbuf_pool; + + /* Print out statistics for one port. */ +@@ -484,10 +483,13 @@ dma_tx_port(struct rxtx_port_config *tx_config) + + port_statistics.tx[tx_config->rxtx_port] += nb_tx; + +- /* Free any unsent packets. */ +- if (unlikely(nb_tx < nb_dq)) ++ if (unlikely(nb_tx < nb_dq)) { ++ port_statistics.tx_dropped[tx_config->rxtx_port] += ++ (nb_dq - nb_tx); ++ /* Free any unsent packets. */ + rte_mempool_put_bulk(dma_pktmbuf_pool, + (void *)&mbufs[nb_tx], nb_dq - nb_tx); ++ } + } + } + /* >8 End of transmitting packets from dmadev. */ +@@ -851,6 +853,38 @@ assign_rings(void) + } + /* >8 End of assigning ring structures for packet exchanging. */ + ++static uint32_t ++eth_dev_get_overhead_len(uint32_t max_rx_pktlen, uint16_t max_mtu) ++{ ++ uint32_t overhead_len; ++ ++ if (max_mtu != UINT16_MAX && max_rx_pktlen > max_mtu) ++ overhead_len = max_rx_pktlen - max_mtu; ++ else ++ overhead_len = RTE_ETHER_HDR_LEN + RTE_ETHER_CRC_LEN; ++ ++ return overhead_len; ++} ++ ++static int ++config_port_max_pkt_len(struct rte_eth_conf *conf, ++ struct rte_eth_dev_info *dev_info) ++{ ++ uint32_t overhead_len; ++ ++ if (max_frame_size == 0) ++ return 0; ++ ++ if (max_frame_size < RTE_ETHER_MIN_LEN) ++ return -1; ++ ++ overhead_len = eth_dev_get_overhead_len(dev_info->max_rx_pktlen, ++ dev_info->max_mtu); ++ conf->rxmode.mtu = max_frame_size - overhead_len; ++ ++ return 0; ++} ++ + /* + * Initializes a given port using global settings and with the RX buffers + * coming from the mbuf_pool passed as a parameter. +@@ -878,9 +912,6 @@ port_init(uint16_t portid, struct rte_mempool *mbuf_pool, uint16_t nb_queues) + struct rte_eth_dev_info dev_info; + int ret, i; + +- if (max_frame_size > local_port_conf.rxmode.mtu) +- local_port_conf.rxmode.mtu = max_frame_size; +- + /* Skip ports that are not enabled */ + if ((dma_enabled_port_mask & (1 << portid)) == 0) { + printf("Skipping disabled port %u\n", portid); +@@ -895,6 +926,12 @@ port_init(uint16_t portid, struct rte_mempool *mbuf_pool, uint16_t nb_queues) + rte_exit(EXIT_FAILURE, "Cannot get device info: %s, port=%u\n", + rte_strerror(-ret), portid); + ++ ret = config_port_max_pkt_len(&local_port_conf, &dev_info); ++ if (ret != 0) ++ rte_exit(EXIT_FAILURE, ++ "Invalid max frame size: %u (port %u)\n", ++ max_frame_size, portid); ++ + local_port_conf.rx_adv_conf.rss_conf.rss_hf &= + dev_info.flow_type_rss_offloads; + ret = rte_eth_dev_configure(portid, nb_queues, 1, &local_port_conf); +@@ -935,25 +972,6 @@ port_init(uint16_t portid, struct rte_mempool *mbuf_pool, uint16_t nb_queues) + "rte_eth_tx_queue_setup:err=%d,port=%u\n", + ret, portid); + +- /* Initialize TX buffers */ +- tx_buffer[portid] = rte_zmalloc_socket("tx_buffer", +- RTE_ETH_TX_BUFFER_SIZE(MAX_PKT_BURST), 0, +- rte_eth_dev_socket_id(portid)); +- if (tx_buffer[portid] == NULL) +- rte_exit(EXIT_FAILURE, +- "Cannot allocate buffer for tx on port %u\n", +- portid); +- +- rte_eth_tx_buffer_init(tx_buffer[portid], MAX_PKT_BURST); +- +- ret = rte_eth_tx_buffer_set_err_callback(tx_buffer[portid], +- rte_eth_tx_buffer_count_callback, +- &port_statistics.tx_dropped[portid]); +- if (ret < 0) +- rte_exit(EXIT_FAILURE, +- "Cannot set error callback for tx buffer on port %u\n", +- portid); +- + /* Start device. 8< */ + ret = rte_eth_dev_start(portid); + if (ret < 0) +diff --git a/dpdk/examples/ethtool/lib/rte_ethtool.c b/dpdk/examples/ethtool/lib/rte_ethtool.c +index 86286d38a6..ffaad96498 100644 +--- a/dpdk/examples/ethtool/lib/rte_ethtool.c ++++ b/dpdk/examples/ethtool/lib/rte_ethtool.c +@@ -402,7 +402,7 @@ rte_ethtool_net_set_rx_mode(uint16_t port_id) + #endif + } + +- /* Enable Rx vlan filter, VF unspport status is discard */ ++ /* Enable Rx vlan filter, VF unsupported status is discard */ + ret = rte_eth_dev_set_vlan_offload(port_id, RTE_ETH_VLAN_FILTER_MASK); + if (ret != 0) + return ret; +diff --git a/dpdk/examples/ethtool/lib/rte_ethtool.h b/dpdk/examples/ethtool/lib/rte_ethtool.h +index f177096636..d27e0102b1 100644 +--- a/dpdk/examples/ethtool/lib/rte_ethtool.h ++++ b/dpdk/examples/ethtool/lib/rte_ethtool.h +@@ -189,7 +189,7 @@ int rte_ethtool_get_module_eeprom(uint16_t port_id, + + /** + * Retrieve the Ethernet device pause frame configuration according to +- * parameter attributes desribed by ethtool data structure, ++ * parameter attributes described by ethtool data structure, + * ethtool_pauseparam. + * + * @param port_id +@@ -209,7 +209,7 @@ int rte_ethtool_get_pauseparam(uint16_t port_id, + + /** + * Setting the Ethernet device pause frame configuration according to +- * parameter attributes desribed by ethtool data structure, ethtool_pauseparam. ++ * parameter attributes described by ethtool data structure, ethtool_pauseparam. + * + * @param port_id + * The port identifier of the Ethernet device. +diff --git a/dpdk/examples/fips_validation/fips_validation.c b/dpdk/examples/fips_validation/fips_validation.c +index 52a7bf952d..511c176a2a 100644 +--- a/dpdk/examples/fips_validation/fips_validation.c ++++ b/dpdk/examples/fips_validation/fips_validation.c +@@ -522,7 +522,7 @@ parse_uint8_hex_str(const char *key, char *src, struct fips_val *val) + val->val = NULL; + } + +- val->val = rte_zmalloc(NULL, len, 0); ++ val->val = rte_zmalloc(NULL, len + 1, 0); + if (!val->val) + return -ENOMEM; + +diff --git a/dpdk/examples/flow_classify/flow_classify.c b/dpdk/examples/flow_classify/flow_classify.c +index 6185b34060..97708b7084 100644 +--- a/dpdk/examples/flow_classify/flow_classify.c ++++ b/dpdk/examples/flow_classify/flow_classify.c +@@ -430,7 +430,7 @@ parse_ipv4_5tuple_rule(char *str, struct rte_eth_ntuple_filter *ntuple_filter) + &ntuple_filter->dst_ip, + &ntuple_filter->dst_ip_mask); + if (ret != 0) { +- flow_classify_log("failed to read source address/mask: %s\n", ++ flow_classify_log("failed to read destination address/mask: %s\n", + in[CB_FLD_DST_ADDR]); + return ret; + } +diff --git a/dpdk/examples/ip_reassembly/main.c b/dpdk/examples/ip_reassembly/main.c +index fb3cac3bd0..6e4c11c3c7 100644 +--- a/dpdk/examples/ip_reassembly/main.c ++++ b/dpdk/examples/ip_reassembly/main.c +@@ -244,7 +244,7 @@ static struct rte_lpm6 *socket_lpm6[RTE_MAX_NUMA_NODES]; + #endif /* RTE_LIBRTE_IP_FRAG_TBL_STAT */ + + /* +- * If number of queued packets reached given threahold, then ++ * If number of queued packets reached given threshold, then + * send burst of packets on an output interface. + */ + static inline uint32_t +@@ -873,11 +873,11 @@ setup_queue_tbl(struct rx_queue *rxq, uint32_t lcore, uint32_t queue) + + /* + * At any given moment up to +- * mbufs could be stored int the fragment table. ++ * mbufs could be stored in the fragment table. + * Plus, each TX queue can hold up to packets. + */ + +- /* mbufs stored int the gragment table. 8< */ ++ /* mbufs stored in the fragment table. 8< */ + nb_mbuf = RTE_MAX(max_flow_num, 2UL * MAX_PKT_BURST) * MAX_FRAG_NUM; + nb_mbuf *= (port_conf.rxmode.mtu + RTE_ETHER_HDR_LEN + RTE_ETHER_CRC_LEN + + BUF_SIZE - 1) / BUF_SIZE; +@@ -895,7 +895,7 @@ setup_queue_tbl(struct rx_queue *rxq, uint32_t lcore, uint32_t queue) + "rte_pktmbuf_pool_create(%s) failed", buf); + return -1; + } +- /* >8 End of mbufs stored int the fragmentation table. */ ++ /* >8 End of mbufs stored in the fragmentation table. */ + + return 0; + } +diff --git a/dpdk/examples/ipsec-secgw/event_helper.c b/dpdk/examples/ipsec-secgw/event_helper.c +index e8600f5e90..8947e41803 100644 +--- a/dpdk/examples/ipsec-secgw/event_helper.c ++++ b/dpdk/examples/ipsec-secgw/event_helper.c +@@ -716,6 +716,16 @@ eh_initialize_eventdev(struct eventmode_conf *em_conf) + } + } + ++ return 0; ++} ++ ++static int ++eh_start_eventdev(struct eventmode_conf *em_conf) ++{ ++ struct eventdev_params *eventdev_config; ++ int nb_eventdev = em_conf->nb_eventdev; ++ int i, ret; ++ + /* Start event devices */ + for (i = 0; i < nb_eventdev; i++) { + +@@ -1353,7 +1363,7 @@ eh_display_rx_adapter_conf(struct eventmode_conf *em_conf) + for (i = 0; i < nb_rx_adapter; i++) { + adapter = &(em_conf->rx_adapter[i]); + sprintf(print_buf, +- "\tRx adaper ID: %-2d\tConnections: %-2d\tEvent dev ID: %-2d", ++ "\tRx adapter ID: %-2d\tConnections: %-2d\tEvent dev ID: %-2d", + adapter->adapter_id, + adapter->nb_connections, + adapter->eventdev_id); +@@ -1688,6 +1698,13 @@ eh_devs_init(struct eh_conf *conf) + return ret; + } + ++ /* Start eventdev */ ++ ret = eh_start_eventdev(em_conf); ++ if (ret < 0) { ++ EH_LOG_ERR("Failed to start event dev %d", ret); ++ return ret; ++ } ++ + /* Start eth devices after setting up adapter */ + RTE_ETH_FOREACH_DEV(port_id) { + +diff --git a/dpdk/examples/ipsec-secgw/flow.c b/dpdk/examples/ipsec-secgw/flow.c +index 1a1ec7861c..c217b9e475 100644 +--- a/dpdk/examples/ipsec-secgw/flow.c ++++ b/dpdk/examples/ipsec-secgw/flow.c +@@ -214,7 +214,7 @@ flow_init_single(struct flow_rule_entry *rule) + struct rte_flow_item pattern[MAX_RTE_FLOW_PATTERN] = {}; + struct rte_flow_action action[MAX_RTE_FLOW_ACTIONS] = {}; + struct rte_flow_attr attr = {}; +- struct rte_flow_error err; ++ struct rte_flow_error err = {}; + int ret; + + attr.egress = 0; +diff --git a/dpdk/examples/ipsec-secgw/ipsec-secgw.c b/dpdk/examples/ipsec-secgw/ipsec-secgw.c +index bf3dbf6b5c..a1faff6a59 100644 +--- a/dpdk/examples/ipsec-secgw/ipsec-secgw.c ++++ b/dpdk/examples/ipsec-secgw/ipsec-secgw.c +@@ -169,7 +169,7 @@ uint32_t single_sa_idx; + /* mask of enabled ports */ + static uint32_t enabled_port_mask; + static uint64_t enabled_cryptodev_mask = UINT64_MAX; +-static int32_t promiscuous_on = 1; ++static int32_t promiscuous_on; + static int32_t numa_on = 1; /**< NUMA is enabled by default. */ + static uint32_t nb_lcores; + static uint32_t single_sa; +@@ -265,7 +265,7 @@ struct socket_ctx socket_ctx[NB_SOCKETS]; + /* + * Determine is multi-segment support required: + * - either frame buffer size is smaller then mtu +- * - or reassmeble support is requested ++ * - or reassemble support is requested + */ + static int + multi_seg_required(void) +@@ -2050,7 +2050,7 @@ add_mapping(struct rte_hash *map, const char *str, uint16_t cdev_id, + + ret = rte_hash_add_key_data(map, &key, (void *)i); + if (ret < 0) { +- printf("Faled to insert cdev mapping for (lcore %u, " ++ printf("Failed to insert cdev mapping for (lcore %u, " + "cdev %u, qp %u), errno %d\n", + key.lcore_id, ipsec_ctx->tbl[i].id, + ipsec_ctx->tbl[i].qp, ret); +@@ -2083,7 +2083,7 @@ add_cdev_mapping(struct rte_cryptodev_info *dev_info, uint16_t cdev_id, + str = "Inbound"; + } + +- /* Required cryptodevs with operation chainning */ ++ /* Required cryptodevs with operation chaining */ + if (!(dev_info->feature_flags & + RTE_CRYPTODEV_FF_SYM_OPERATION_CHAINING)) + return ret; +@@ -2251,7 +2251,7 @@ port_init(uint16_t portid, uint64_t req_rx_offloads, uint64_t req_tx_offloads) + "Error during getting device (port %u) info: %s\n", + portid, strerror(-ret)); + +- /* limit allowed HW offloafs, as user requested */ ++ /* limit allowed HW offloads, as user requested */ + dev_info.rx_offload_capa &= dev_rx_offload; + dev_info.tx_offload_capa &= dev_tx_offload; + +@@ -2298,7 +2298,7 @@ port_init(uint16_t portid, uint64_t req_rx_offloads, uint64_t req_tx_offloads) + local_port_conf.rxmode.offloads) + rte_exit(EXIT_FAILURE, + "Error: port %u required RX offloads: 0x%" PRIx64 +- ", avaialbe RX offloads: 0x%" PRIx64 "\n", ++ ", available RX offloads: 0x%" PRIx64 "\n", + portid, local_port_conf.rxmode.offloads, + dev_info.rx_offload_capa); + +@@ -2306,7 +2306,7 @@ port_init(uint16_t portid, uint64_t req_rx_offloads, uint64_t req_tx_offloads) + local_port_conf.txmode.offloads) + rte_exit(EXIT_FAILURE, + "Error: port %u required TX offloads: 0x%" PRIx64 +- ", avaialbe TX offloads: 0x%" PRIx64 "\n", ++ ", available TX offloads: 0x%" PRIx64 "\n", + portid, local_port_conf.txmode.offloads, + dev_info.tx_offload_capa); + +@@ -2317,7 +2317,7 @@ port_init(uint16_t portid, uint64_t req_rx_offloads, uint64_t req_tx_offloads) + if (dev_info.tx_offload_capa & RTE_ETH_TX_OFFLOAD_IPV4_CKSUM) + local_port_conf.txmode.offloads |= RTE_ETH_TX_OFFLOAD_IPV4_CKSUM; + +- printf("port %u configurng rx_offloads=0x%" PRIx64 ++ printf("port %u configuring rx_offloads=0x%" PRIx64 + ", tx_offloads=0x%" PRIx64 "\n", + portid, local_port_conf.rxmode.offloads, + local_port_conf.txmode.offloads); +@@ -3379,13 +3379,14 @@ main(int32_t argc, char **argv) + if ((enabled_port_mask & (1 << portid)) == 0) + continue; + +- /* Create flow before starting the device */ +- create_default_ipsec_flow(portid, req_rx_offloads[portid]); +- + ret = rte_eth_dev_start(portid); + if (ret < 0) + rte_exit(EXIT_FAILURE, "rte_eth_dev_start: " + "err=%d, port=%d\n", ret, portid); ++ ++ /* Create flow after starting the device */ ++ create_default_ipsec_flow(portid, req_rx_offloads[portid]); ++ + /* + * If enabled, put device in promiscuous mode. + * This allows IO forwarding mode to forward packets +diff --git a/dpdk/examples/ipsec-secgw/ipsec.c b/dpdk/examples/ipsec-secgw/ipsec.c +index 2d4a26c962..b66ff2b650 100644 +--- a/dpdk/examples/ipsec-secgw/ipsec.c ++++ b/dpdk/examples/ipsec-secgw/ipsec.c +@@ -496,7 +496,7 @@ int + create_ipsec_esp_flow(struct ipsec_sa *sa) + { + int ret = 0; +- struct rte_flow_error err; ++ struct rte_flow_error err = {}; + if (sa->direction == RTE_SECURITY_IPSEC_SA_DIR_EGRESS) { + RTE_LOG(ERR, IPSEC, + "No Flow director rule for Egress traffic\n"); +diff --git a/dpdk/examples/ipsec-secgw/ipsec_process.c b/dpdk/examples/ipsec-secgw/ipsec_process.c +index 3fc4b3a84f..285e9c7f4c 100644 +--- a/dpdk/examples/ipsec-secgw/ipsec_process.c ++++ b/dpdk/examples/ipsec-secgw/ipsec_process.c +@@ -232,8 +232,10 @@ prep_process_group(void *sa, struct rte_mbuf *mb[], uint32_t cnt) + mb[j]->outer_l3_len = mb[j]->l3_len; + mb[j]->outer_l2_len = mb[j]->l2_len; + mb[j]->ol_flags |= +- (RTE_MBUF_F_TX_OUTER_IP_CKSUM | +- RTE_MBUF_F_TX_TUNNEL_ESP); ++ RTE_MBUF_F_TX_TUNNEL_ESP; ++ if (RTE_ETH_IS_IPV4_HDR(ptype)) ++ mb[j]->ol_flags |= ++ RTE_MBUF_F_TX_OUTER_IP_CKSUM; + } + mb[j]->l4_len = sizeof(struct rte_tcp_hdr); + mb[j]->ol_flags |= (RTE_MBUF_F_TX_TCP_SEG | +diff --git a/dpdk/examples/ipsec-secgw/ipsec_worker.c b/dpdk/examples/ipsec-secgw/ipsec_worker.c +index 7419e85db2..5fe91b62e4 100644 +--- a/dpdk/examples/ipsec-secgw/ipsec_worker.c ++++ b/dpdk/examples/ipsec-secgw/ipsec_worker.c +@@ -205,12 +205,16 @@ check_sp_sa_bulk(struct sp_ctx *sp, struct sa_ctx *sa_ctx, + ip->pkts[j++] = m; + else { + sa = *(struct ipsec_sa **)rte_security_dynfield(m); +- if (sa == NULL) ++ if (sa == NULL) { + free_pkts(&m, 1); ++ continue; ++ } + + /* SPI on the packet should match with the one in SA */ +- if (unlikely(sa->spi != sa_ctx->sa[res - 1].spi)) ++ if (unlikely(sa->spi != sa_ctx->sa[res - 1].spi)) { + free_pkts(&m, 1); ++ continue; ++ } + + ip->pkts[j++] = m; + } +@@ -535,6 +539,7 @@ ipsec_ev_route_pkts(struct rte_event_vector *vec, struct route_table *rt, + RTE_SECURITY_ACTION_TYPE_INLINE_PROTOCOL)) { + RTE_LOG(ERR, IPSEC, "SA type not supported\n"); + free_pkts(&pkt, 1); ++ continue; + } + rte_security_set_pkt_metadata(sess->security.ctx, + sess->security.ses, pkt, NULL); +@@ -693,11 +698,13 @@ ipsec_ev_vector_process(struct lcore_conf_ev_tx_int_port_wrkr *lconf, + ret = process_ipsec_ev_outbound_vector(&lconf->outbound, + &lconf->rt, vec); + +- if (ret > 0) { ++ if (likely(ret > 0)) { + vec->nb_elem = ret; + rte_event_eth_tx_adapter_enqueue(links[0].eventdev_id, + links[0].event_port_id, + ev, 1, 0); ++ } else { ++ rte_mempool_put(rte_mempool_from_obj(vec), vec); + } + } + +@@ -718,6 +725,8 @@ ipsec_ev_vector_drv_mode_process(struct eh_event_link_info *links, + rte_event_eth_tx_adapter_enqueue(links[0].eventdev_id, + links[0].event_port_id, + ev, 1, 0); ++ else ++ rte_mempool_put(rte_mempool_from_obj(vec), vec); + } + + /* +diff --git a/dpdk/examples/ipsec-secgw/sa.c b/dpdk/examples/ipsec-secgw/sa.c +index 30bc693e06..89131f71e5 100644 +--- a/dpdk/examples/ipsec-secgw/sa.c ++++ b/dpdk/examples/ipsec-secgw/sa.c +@@ -897,7 +897,7 @@ parse_sa_tokens(char **tokens, uint32_t n_tokens, + continue; + } + +- /* unrecognizeable input */ ++ /* unrecognizable input */ + APP_CHECK(0, status, "unrecognized input \"%s\"", + tokens[ti]); + return; +@@ -1145,7 +1145,7 @@ get_spi_proto(uint32_t spi, enum rte_security_ipsec_sa_direction dir, + if (rc4 >= 0) { + if (rc6 >= 0) { + RTE_LOG(ERR, IPSEC, +- "%s: SPI %u used simultaeously by " ++ "%s: SPI %u used simultaneously by " + "IPv4(%d) and IPv6 (%d) SP rules\n", + __func__, spi, rc4, rc6); + return -EINVAL; +@@ -1458,9 +1458,16 @@ fill_ipsec_sa_prm(struct rte_ipsec_sa_prm *prm, const struct ipsec_sa *ss, + RTE_SECURITY_IPSEC_SA_MODE_TRANSPORT : + RTE_SECURITY_IPSEC_SA_MODE_TUNNEL; + prm->ipsec_xform.options.udp_encap = ss->udp_encap; ++ prm->ipsec_xform.udp.dport = ss->udp.dport; ++ prm->ipsec_xform.udp.sport = ss->udp.sport; + prm->ipsec_xform.options.ecn = 1; + prm->ipsec_xform.options.copy_dscp = 1; + ++ if (ss->esn > 0) { ++ prm->ipsec_xform.options.esn = 1; ++ prm->ipsec_xform.esn.value = ss->esn; ++ } ++ + if (IS_IP4_TUNNEL(ss->flags)) { + prm->ipsec_xform.tunnel.type = RTE_SECURITY_IPSEC_TUNNEL_IPV4; + prm->tun.hdr_len = sizeof(*v4); +@@ -1513,13 +1520,13 @@ ipsec_sa_init(struct ipsec_sa *lsa, struct rte_ipsec_sa *sa, uint32_t sa_size) + .version_ihl = IPVERSION << 4 | + sizeof(v4) / RTE_IPV4_IHL_MULTIPLIER, + .time_to_live = IPDEFTTL, +- .next_proto_id = IPPROTO_ESP, ++ .next_proto_id = lsa->udp_encap ? IPPROTO_UDP : IPPROTO_ESP, + .src_addr = lsa->src.ip.ip4, + .dst_addr = lsa->dst.ip.ip4, + }; + struct rte_ipv6_hdr v6 = { + .vtc_flow = htonl(IP6_VERSION << 28), +- .proto = IPPROTO_ESP, ++ .proto = lsa->udp_encap ? IPPROTO_UDP : IPPROTO_ESP, + }; + + if (IS_IP6_TUNNEL(lsa->flags)) { +@@ -1550,7 +1557,7 @@ ipsec_sa_init(struct ipsec_sa *lsa, struct rte_ipsec_sa *sa, uint32_t sa_size) + } + + /* +- * Allocate space and init rte_ipsec_sa strcutures, ++ * Allocate space and init rte_ipsec_sa structures, + * one per session. + */ + static int +diff --git a/dpdk/examples/ipsec-secgw/sp4.c b/dpdk/examples/ipsec-secgw/sp4.c +index beddd7bc1d..fc4101a4a2 100644 +--- a/dpdk/examples/ipsec-secgw/sp4.c ++++ b/dpdk/examples/ipsec-secgw/sp4.c +@@ -410,7 +410,7 @@ parse_sp4_tokens(char **tokens, uint32_t n_tokens, + continue; + } + +- /* unrecognizeable input */ ++ /* unrecognizable input */ + APP_CHECK(0, status, "unrecognized input \"%s\"", + tokens[ti]); + return; +diff --git a/dpdk/examples/ipsec-secgw/sp6.c b/dpdk/examples/ipsec-secgw/sp6.c +index 328e085288..cce4da7862 100644 +--- a/dpdk/examples/ipsec-secgw/sp6.c ++++ b/dpdk/examples/ipsec-secgw/sp6.c +@@ -515,7 +515,7 @@ parse_sp6_tokens(char **tokens, uint32_t n_tokens, + continue; + } + +- /* unrecognizeable input */ ++ /* unrecognizable input */ + APP_CHECK(0, status, "unrecognized input \"%s\"", + tokens[ti]); + return; +diff --git a/dpdk/examples/ipsec-secgw/test/common_defs.sh b/dpdk/examples/ipsec-secgw/test/common_defs.sh +index f22eb3ab12..3ef06bc761 100644 +--- a/dpdk/examples/ipsec-secgw/test/common_defs.sh ++++ b/dpdk/examples/ipsec-secgw/test/common_defs.sh +@@ -20,7 +20,7 @@ REMOTE_MAC=`ssh ${REMOTE_HOST} ip addr show dev ${REMOTE_IFACE}` + st=$? + REMOTE_MAC=`echo ${REMOTE_MAC} | sed -e 's/^.*ether //' -e 's/ brd.*$//'` + if [[ $st -ne 0 || -z "${REMOTE_MAC}" ]]; then +- echo "coouldn't retrieve ether addr from ${REMOTE_IFACE}" ++ echo "couldn't retrieve ether addr from ${REMOTE_IFACE}" + exit 127 + fi + +@@ -40,7 +40,7 @@ DPDK_VARS="" + + # by default ipsec-secgw can't deal with multi-segment packets + # make sure our local/remote host wouldn't generate fragmented packets +-# if reassmebly option is not enabled ++# if reassembly option is not enabled + DEF_MTU_LEN=1400 + DEF_PING_LEN=1200 + +diff --git a/dpdk/examples/kni/main.c b/dpdk/examples/kni/main.c +index d324ee2241..e99ef5c38a 100644 +--- a/dpdk/examples/kni/main.c ++++ b/dpdk/examples/kni/main.c +@@ -719,7 +719,7 @@ log_link_state(struct rte_kni *kni, int prev, struct rte_eth_link *link) + + rte_eth_link_to_str(link_status_text, sizeof(link_status_text), link); + if (prev != link->link_status) +- RTE_LOG(INFO, APP, "%s NIC %s", ++ RTE_LOG(INFO, APP, "%s NIC %s\n", + rte_kni_get_name(kni), + link_status_text); + } +@@ -1039,7 +1039,7 @@ main(int argc, char** argv) + pthread_t kni_link_tid; + int pid; + +- /* Associate signal_hanlder function with USR signals */ ++ /* Associate signal_handler function with USR signals */ + signal(SIGUSR1, signal_handler); + signal(SIGUSR2, signal_handler); + signal(SIGRTMIN, signal_handler); +diff --git a/dpdk/examples/l2fwd-cat/l2fwd-cat.c b/dpdk/examples/l2fwd-cat/l2fwd-cat.c +index d9cf00c9df..6e16705e99 100644 +--- a/dpdk/examples/l2fwd-cat/l2fwd-cat.c ++++ b/dpdk/examples/l2fwd-cat/l2fwd-cat.c +@@ -157,7 +157,7 @@ main(int argc, char *argv[]) + int ret = rte_eal_init(argc, argv); + if (ret < 0) + rte_exit(EXIT_FAILURE, "Error with EAL initialization\n"); +- /* >8 End of initializion the Environment Abstraction Layer (EAL). */ ++ /* >8 End of initialization the Environment Abstraction Layer (EAL). */ + + argc -= ret; + argv += ret; +diff --git a/dpdk/examples/l2fwd-crypto/main.c b/dpdk/examples/l2fwd-crypto/main.c +index 4d9f8861af..b1e2613ccf 100644 +--- a/dpdk/examples/l2fwd-crypto/main.c ++++ b/dpdk/examples/l2fwd-crypto/main.c +@@ -252,11 +252,9 @@ struct l2fwd_port_statistics port_statistics[RTE_MAX_ETHPORTS]; + struct l2fwd_crypto_statistics crypto_statistics[RTE_CRYPTO_MAX_DEVS]; + + /* A tsc-based timer responsible for triggering statistics printout */ +-#define TIMER_MILLISECOND 2000000ULL /* around 1ms at 2 Ghz */ ++#define TIMER_MILLISECOND (rte_get_tsc_hz() / 1000) + #define MAX_TIMER_PERIOD 86400UL /* 1 day max */ +- +-/* default period is 10 seconds */ +-static int64_t timer_period = 10 * TIMER_MILLISECOND * 1000; ++#define DEFAULT_TIMER_PERIOD 10UL + + /* Print out statistics on packets dropped */ + static void +@@ -894,18 +892,17 @@ l2fwd_main_loop(struct l2fwd_crypto_options *options) + } + + /* if timer is enabled */ +- if (timer_period > 0) { ++ if (options->refresh_period > 0) { + + /* advance the timer */ + timer_tsc += diff_tsc; + + /* if timer has reached its timeout */ + if (unlikely(timer_tsc >= +- (uint64_t)timer_period)) { ++ options->refresh_period)) { + + /* do this only on main core */ +- if (lcore_id == rte_get_main_lcore() +- && options->refresh_period) { ++ if (lcore_id == rte_get_main_lcore()) { + print_stats(); + timer_tsc = 0; + } +@@ -1481,7 +1478,8 @@ l2fwd_crypto_default_options(struct l2fwd_crypto_options *options) + { + options->portmask = 0xffffffff; + options->nb_ports_per_lcore = 1; +- options->refresh_period = 10000; ++ options->refresh_period = DEFAULT_TIMER_PERIOD * ++ TIMER_MILLISECOND * 1000; + options->single_lcore = 0; + options->sessionless = 0; + +@@ -2719,7 +2717,7 @@ initialize_ports(struct l2fwd_crypto_options *options) + last_portid = portid; + } + +- l2fwd_enabled_port_mask |= (1 << portid); ++ l2fwd_enabled_port_mask |= (1ULL << portid); + enabled_portcount++; + } + +diff --git a/dpdk/examples/l2fwd-event/l2fwd_event_generic.c b/dpdk/examples/l2fwd-event/l2fwd_event_generic.c +index f31569a744..1977e23261 100644 +--- a/dpdk/examples/l2fwd-event/l2fwd_event_generic.c ++++ b/dpdk/examples/l2fwd-event/l2fwd_event_generic.c +@@ -42,7 +42,7 @@ l2fwd_event_device_setup_generic(struct l2fwd_resources *rsrc) + ethdev_count++; + } + +- /* Event device configurtion */ ++ /* Event device configuration */ + rte_event_dev_info_get(event_d_id, &dev_info); + + /* Enable implicit release */ +diff --git a/dpdk/examples/l2fwd-event/l2fwd_event_internal_port.c b/dpdk/examples/l2fwd-event/l2fwd_event_internal_port.c +index 86d772d817..717a7bceb8 100644 +--- a/dpdk/examples/l2fwd-event/l2fwd_event_internal_port.c ++++ b/dpdk/examples/l2fwd-event/l2fwd_event_internal_port.c +@@ -40,7 +40,7 @@ l2fwd_event_device_setup_internal_port(struct l2fwd_resources *rsrc) + ethdev_count++; + } + +- /* Event device configurtion */ ++ /* Event device configuration */ + rte_event_dev_info_get(event_d_id, &dev_info); + + /* Enable implicit release */ +diff --git a/dpdk/examples/l2fwd-jobstats/main.c b/dpdk/examples/l2fwd-jobstats/main.c +index d8eabe4c86..9e71ba2d4e 100644 +--- a/dpdk/examples/l2fwd-jobstats/main.c ++++ b/dpdk/examples/l2fwd-jobstats/main.c +@@ -468,7 +468,7 @@ l2fwd_flush_job(__rte_unused struct rte_timer *timer, __rte_unused void *arg) + qconf->next_flush_time[portid] = rte_get_timer_cycles() + drain_tsc; + } + +- /* Pass target to indicate that this job is happy of time interwal ++ /* Pass target to indicate that this job is happy of time interval + * in which it was called. */ + rte_jobstats_finish(&qconf->flush_job, qconf->flush_job.target); + } +diff --git a/dpdk/examples/l3fwd-acl/main.c b/dpdk/examples/l3fwd-acl/main.c +index 1fb1807235..2d2ecc7635 100644 +--- a/dpdk/examples/l3fwd-acl/main.c ++++ b/dpdk/examples/l3fwd-acl/main.c +@@ -801,8 +801,8 @@ send_packets(struct rte_mbuf **m, uint32_t *res, int num) + } + + /* +- * Parses IPV6 address, exepcts the following format: +- * XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX (where X - is a hexedecimal digit). ++ * Parse IPv6 address, expects the following format: ++ * XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX (where X is a hexadecimal digit). + */ + static int + parse_ipv6_addr(const char *in, const char **end, uint32_t v[IPV6_ADDR_U32], +@@ -1959,7 +1959,7 @@ check_all_ports_link_status(uint32_t port_mask) + } + + /* +- * build-up default vaues for dest MACs. ++ * build-up default values for dest MACs. + */ + static void + set_default_dest_mac(void) +diff --git a/dpdk/examples/l3fwd-power/main.c b/dpdk/examples/l3fwd-power/main.c +index b8b3be2b8a..20e5b59af9 100644 +--- a/dpdk/examples/l3fwd-power/main.c ++++ b/dpdk/examples/l3fwd-power/main.c +@@ -433,7 +433,7 @@ signal_exit_now(int sigtype) + + } + +-/* Freqency scale down timer callback */ ++/* Frequency scale down timer callback */ + static void + power_timer_cb(__rte_unused struct rte_timer *tim, + __rte_unused void *arg) +@@ -2358,7 +2358,7 @@ update_telemetry(__rte_unused struct rte_timer *tim, + ret = rte_metrics_update_values(RTE_METRICS_GLOBAL, telstats_index, + values, RTE_DIM(values)); + if (ret < 0) +- RTE_LOG(WARNING, POWER, "failed to update metrcis\n"); ++ RTE_LOG(WARNING, POWER, "failed to update metrics\n"); + } + + static int +diff --git a/dpdk/examples/l3fwd/l3fwd.h b/dpdk/examples/l3fwd/l3fwd.h +index 38ca19133c..4ccdb28b4a 100644 +--- a/dpdk/examples/l3fwd/l3fwd.h ++++ b/dpdk/examples/l3fwd/l3fwd.h +@@ -81,6 +81,10 @@ struct lcore_conf { + + extern volatile bool force_quit; + ++/* RX and TX queue depths */ ++extern uint16_t nb_rxd; ++extern uint16_t nb_txd; ++ + /* ethernet addresses of ports */ + extern uint64_t dest_eth_addr[RTE_MAX_ETHPORTS]; + extern struct rte_ether_addr ports_eth_addr[RTE_MAX_ETHPORTS]; +diff --git a/dpdk/examples/l3fwd/l3fwd_altivec.h b/dpdk/examples/l3fwd/l3fwd_altivec.h +index ed9b0b8f25..88fb41843b 100644 +--- a/dpdk/examples/l3fwd/l3fwd_altivec.h ++++ b/dpdk/examples/l3fwd/l3fwd_altivec.h +@@ -17,50 +17,50 @@ + static inline void + processx4_step3(struct rte_mbuf *pkt[FWDSTEP], uint16_t dst_port[FWDSTEP]) + { +- vector unsigned int te[FWDSTEP]; +- vector unsigned int ve[FWDSTEP]; +- vector unsigned int *p[FWDSTEP]; ++ __vector unsigned int te[FWDSTEP]; ++ __vector unsigned int ve[FWDSTEP]; ++ __vector unsigned int *p[FWDSTEP]; + +- p[0] = rte_pktmbuf_mtod(pkt[0], vector unsigned int *); +- p[1] = rte_pktmbuf_mtod(pkt[1], vector unsigned int *); +- p[2] = rte_pktmbuf_mtod(pkt[2], vector unsigned int *); +- p[3] = rte_pktmbuf_mtod(pkt[3], vector unsigned int *); ++ p[0] = rte_pktmbuf_mtod(pkt[0], __vector unsigned int *); ++ p[1] = rte_pktmbuf_mtod(pkt[1], __vector unsigned int *); ++ p[2] = rte_pktmbuf_mtod(pkt[2], __vector unsigned int *); ++ p[3] = rte_pktmbuf_mtod(pkt[3], __vector unsigned int *); + +- ve[0] = (vector unsigned int)val_eth[dst_port[0]]; ++ ve[0] = (__vector unsigned int)val_eth[dst_port[0]]; + te[0] = *p[0]; + +- ve[1] = (vector unsigned int)val_eth[dst_port[1]]; ++ ve[1] = (__vector unsigned int)val_eth[dst_port[1]]; + te[1] = *p[1]; + +- ve[2] = (vector unsigned int)val_eth[dst_port[2]]; ++ ve[2] = (__vector unsigned int)val_eth[dst_port[2]]; + te[2] = *p[2]; + +- ve[3] = (vector unsigned int)val_eth[dst_port[3]]; ++ ve[3] = (__vector unsigned int)val_eth[dst_port[3]]; + te[3] = *p[3]; + + /* Update first 12 bytes, keep rest bytes intact. */ +- te[0] = (vector unsigned int)vec_sel( +- (vector unsigned short)ve[0], +- (vector unsigned short)te[0], +- (vector unsigned short) {0, 0, 0, 0, ++ te[0] = (__vector unsigned int)vec_sel( ++ (__vector unsigned short)ve[0], ++ (__vector unsigned short)te[0], ++ (__vector unsigned short) {0, 0, 0, 0, + 0, 0, 0xffff, 0xffff}); + +- te[1] = (vector unsigned int)vec_sel( +- (vector unsigned short)ve[1], +- (vector unsigned short)te[1], +- (vector unsigned short) {0, 0, 0, 0, ++ te[1] = (__vector unsigned int)vec_sel( ++ (__vector unsigned short)ve[1], ++ (__vector unsigned short)te[1], ++ (__vector unsigned short) {0, 0, 0, 0, + 0, 0, 0xffff, 0xffff}); + +- te[2] = (vector unsigned int)vec_sel( +- (vector unsigned short)ve[2], +- (vector unsigned short)te[2], +- (vector unsigned short) {0, 0, 0, 0, 0, ++ te[2] = (__vector unsigned int)vec_sel( ++ (__vector unsigned short)ve[2], ++ (__vector unsigned short)te[2], ++ (__vector unsigned short) {0, 0, 0, 0, 0, + 0, 0xffff, 0xffff}); + +- te[3] = (vector unsigned int)vec_sel( +- (vector unsigned short)ve[3], +- (vector unsigned short)te[3], +- (vector unsigned short) {0, 0, 0, 0, ++ te[3] = (__vector unsigned int)vec_sel( ++ (__vector unsigned short)ve[3], ++ (__vector unsigned short)te[3], ++ (__vector unsigned short) {0, 0, 0, 0, + 0, 0, 0xffff, 0xffff}); + + *p[0] = te[0]; +@@ -91,8 +91,9 @@ processx4_step3(struct rte_mbuf *pkt[FWDSTEP], uint16_t dst_port[FWDSTEP]) + * This mask is used as an index into prebuild array of pnum values. + */ + static inline uint16_t * +-port_groupx4(uint16_t pn[FWDSTEP + 1], uint16_t *lp, vector unsigned short dp1, +- vector unsigned short dp2) ++port_groupx4(uint16_t pn[FWDSTEP + 1], uint16_t *lp, ++ __vector unsigned short dp1, ++ __vector unsigned short dp2) + { + union { + uint16_t u16[FWDSTEP + 1]; +@@ -126,24 +127,24 @@ static inline void + process_packet(struct rte_mbuf *pkt, uint16_t *dst_port) + { + struct rte_ether_hdr *eth_hdr; +- vector unsigned int te, ve; ++ __vector unsigned int te, ve; + + eth_hdr = rte_pktmbuf_mtod(pkt, struct rte_ether_hdr *); + +- te = *(vector unsigned int *)eth_hdr; +- ve = (vector unsigned int)val_eth[dst_port[0]]; ++ te = *(__vector unsigned int *)eth_hdr; ++ ve = (__vector unsigned int)val_eth[dst_port[0]]; + + rfc1812_process((struct rte_ipv4_hdr *)(eth_hdr + 1), dst_port, + pkt->packet_type); + + /* dynamically vec_sel te and ve for MASK_ETH (0x3f) */ +- te = (vector unsigned int)vec_sel( +- (vector unsigned short)ve, +- (vector unsigned short)te, +- (vector unsigned short){0, 0, 0, 0, ++ te = (__vector unsigned int)vec_sel( ++ (__vector unsigned short)ve, ++ (__vector unsigned short)te, ++ (__vector unsigned short){0, 0, 0, 0, + 0, 0, 0xffff, 0xffff}); + +- *(vector unsigned int *)eth_hdr = te; ++ *(__vector unsigned int *)eth_hdr = te; + } + + /** +@@ -165,7 +166,7 @@ send_packets_multi(struct lcore_conf *qconf, struct rte_mbuf **pkts_burst, + */ + k = RTE_ALIGN_FLOOR(nb_rx, FWDSTEP); + if (k != 0) { +- vector unsigned short dp1, dp2; ++ __vector unsigned short dp1, dp2; + + lp = pnum; + lp[0] = 1; +@@ -173,7 +174,7 @@ send_packets_multi(struct lcore_conf *qconf, struct rte_mbuf **pkts_burst, + processx4_step3(pkts_burst, dst_port); + + /* dp1: */ +- dp1 = *(vector unsigned short *)dst_port; ++ dp1 = *(__vector unsigned short *)dst_port; + + for (j = FWDSTEP; j != k; j += FWDSTEP) { + processx4_step3(&pkts_burst[j], &dst_port[j]); +@@ -182,7 +183,7 @@ send_packets_multi(struct lcore_conf *qconf, struct rte_mbuf **pkts_burst, + * dp2: + * + */ +- dp2 = *((vector unsigned short *) ++ dp2 = *((__vector unsigned short *) + &dst_port[j - FWDSTEP + 1]); + lp = port_groupx4(&pnum[j - FWDSTEP], lp, dp1, dp2); + +@@ -190,7 +191,7 @@ send_packets_multi(struct lcore_conf *qconf, struct rte_mbuf **pkts_burst, + * dp1: + * + */ +- dp1 = vec_sro(dp2, (vector unsigned char) { ++ dp1 = vec_sro(dp2, (__vector unsigned char) { + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, (FWDSTEP - 1) * sizeof(dst_port[0])}); + } +@@ -198,8 +199,8 @@ send_packets_multi(struct lcore_conf *qconf, struct rte_mbuf **pkts_burst, + /* + * dp2: + */ +- dp2 = vec_perm(dp1, (vector unsigned short){}, +- (vector unsigned char){0xf9}); ++ dp2 = vec_perm(dp1, (__vector unsigned short){}, ++ (__vector unsigned char){0xf9}); + lp = port_groupx4(&pnum[j - FWDSTEP], lp, dp1, dp2); + + /* +diff --git a/dpdk/examples/l3fwd/l3fwd_common.h b/dpdk/examples/l3fwd/l3fwd_common.h +index 7d83ff641a..8e4c27218f 100644 +--- a/dpdk/examples/l3fwd/l3fwd_common.h ++++ b/dpdk/examples/l3fwd/l3fwd_common.h +@@ -51,7 +51,7 @@ rfc1812_process(struct rte_ipv4_hdr *ipv4_hdr, uint16_t *dp, uint32_t ptype) + #endif /* DO_RFC_1812_CHECKS */ + + /* +- * We group consecutive packets with the same destionation port into one burst. ++ * We group consecutive packets with the same destination port into one burst. + * To avoid extra latency this is done together with some other packet + * processing, but after we made a final decision about packet's destination. + * To do this we maintain: +@@ -76,7 +76,7 @@ rfc1812_process(struct rte_ipv4_hdr *ipv4_hdr, uint16_t *dp, uint32_t ptype) + + static const struct { + uint64_t pnum; /* prebuild 4 values for pnum[]. */ +- int32_t idx; /* index for new last updated elemnet. */ ++ int32_t idx; /* index for new last updated element. */ + uint16_t lpv; /* add value to the last updated element. */ + } gptbl[GRPSZ] = { + { +@@ -236,6 +236,9 @@ send_packetsx4(struct lcore_conf *qconf, uint16_t port, struct rte_mbuf *m[], + + /* copy rest of the packets into the TX buffer. */ + len = num - n; ++ if (len == 0) ++ goto exit; ++ + j = 0; + switch (len % FWDSTEP) { + while (j < len) { +@@ -258,6 +261,7 @@ send_packetsx4(struct lcore_conf *qconf, uint16_t port, struct rte_mbuf *m[], + } + } + ++exit: + qconf->tx_mbufs[port].len = len; + } + +diff --git a/dpdk/examples/l3fwd/l3fwd_event.c b/dpdk/examples/l3fwd/l3fwd_event.c +index dd9eafa9b9..7a401290f8 100644 +--- a/dpdk/examples/l3fwd/l3fwd_event.c ++++ b/dpdk/examples/l3fwd/l3fwd_event.c +@@ -43,8 +43,6 @@ l3fwd_eth_dev_port_setup(struct rte_eth_conf *port_conf) + { + struct l3fwd_event_resources *evt_rsrc = l3fwd_get_eventdev_rsrc(); + uint16_t nb_ports = rte_eth_dev_count_avail(); +- uint16_t nb_rxd = RTE_TEST_RX_DESC_DEFAULT; +- uint16_t nb_txd = RTE_TEST_TX_DESC_DEFAULT; + unsigned int nb_lcores = rte_lcore_count(); + struct rte_eth_conf local_port_conf; + struct rte_eth_dev_info dev_info; +diff --git a/dpdk/examples/l3fwd/l3fwd_event_internal_port.c b/dpdk/examples/l3fwd/l3fwd_event_internal_port.c +index 1e8f46bc11..32cf657148 100644 +--- a/dpdk/examples/l3fwd/l3fwd_event_internal_port.c ++++ b/dpdk/examples/l3fwd/l3fwd_event_internal_port.c +@@ -118,6 +118,8 @@ l3fwd_event_port_setup_internal_port(void) + event_p_conf.event_port_cfg |= + RTE_EVENT_PORT_CFG_DISABLE_IMPL_REL; + ++ evt_rsrc->deq_depth = def_p_conf.dequeue_depth; ++ + for (event_p_id = 0; event_p_id < evt_rsrc->evp.nb_ports; + event_p_id++) { + ret = rte_event_port_setup(event_d_id, event_p_id, +diff --git a/dpdk/examples/l3fwd/l3fwd_lpm_altivec.h b/dpdk/examples/l3fwd/l3fwd_lpm_altivec.h +index 7c6814252a..0c6852a7bb 100644 +--- a/dpdk/examples/l3fwd/l3fwd_lpm_altivec.h ++++ b/dpdk/examples/l3fwd/l3fwd_lpm_altivec.h +@@ -14,7 +14,7 @@ + */ + static inline void + processx4_step1(struct rte_mbuf *pkt[FWDSTEP], +- vector unsigned int *dip, ++ __vector unsigned int *dip, + uint32_t *ipv4_flag) + { + struct rte_ipv4_hdr *ipv4_hdr; +@@ -45,7 +45,7 @@ processx4_step1(struct rte_mbuf *pkt[FWDSTEP], + ipv4_flag[0] &= pkt[3]->packet_type; + + rte_compiler_barrier(); +- dip[0] = (vector unsigned int){x0, x1, x2, x3}; ++ dip[0] = (__vector unsigned int){x0, x1, x2, x3}; + } + + /* +@@ -54,22 +54,22 @@ processx4_step1(struct rte_mbuf *pkt[FWDSTEP], + */ + static inline void + processx4_step2(const struct lcore_conf *qconf, +- vector unsigned int dip, ++ __vector unsigned int dip, + uint32_t ipv4_flag, + uint8_t portid, + struct rte_mbuf *pkt[FWDSTEP], + uint16_t dprt[FWDSTEP]) + { + rte_xmm_t dst; +- const vector unsigned char bswap_mask = (vector unsigned char){ ++ const __vector unsigned char bswap_mask = (__vector unsigned char){ + 3, 2, 1, 0, + 7, 6, 5, 4, + 11, 10, 9, 8, + 15, 14, 13, 12}; + + /* Byte swap 4 IPV4 addresses. */ +- dip = (vector unsigned int)vec_perm(*(vector unsigned char *)&dip, +- (vector unsigned char){}, bswap_mask); ++ dip = (__vector unsigned int)vec_perm(*(__vector unsigned char *)&dip, ++ (__vector unsigned char){}, bswap_mask); + + /* if all 4 packets are IPV4. */ + if (likely(ipv4_flag)) { +@@ -101,7 +101,7 @@ l3fwd_lpm_send_packets(int nb_rx, struct rte_mbuf **pkts_burst, + { + int32_t j; + uint16_t dst_port[MAX_PKT_BURST]; +- vector unsigned int dip[MAX_PKT_BURST / FWDSTEP]; ++ __vector unsigned int dip[MAX_PKT_BURST / FWDSTEP]; + uint32_t ipv4_flag[MAX_PKT_BURST / FWDSTEP]; + const int32_t k = RTE_ALIGN_FLOOR(nb_rx, FWDSTEP); + +diff --git a/dpdk/examples/l3fwd/l3fwd_neon.h b/dpdk/examples/l3fwd/l3fwd_neon.h +index 86ac5971d7..e3d33a5229 100644 +--- a/dpdk/examples/l3fwd/l3fwd_neon.h ++++ b/dpdk/examples/l3fwd/l3fwd_neon.h +@@ -64,7 +64,7 @@ processx4_step3(struct rte_mbuf *pkt[FWDSTEP], uint16_t dst_port[FWDSTEP]) + + /* + * Group consecutive packets with the same destination port in bursts of 4. +- * Suppose we have array of destionation ports: ++ * Suppose we have array of destination ports: + * dst_port[] = {a, b, c, d,, e, ... } + * dp1 should contain: , dp2: . + * We doing 4 comparisons at once and the result is 4 bit mask. +diff --git a/dpdk/examples/l3fwd/l3fwd_sse.h b/dpdk/examples/l3fwd/l3fwd_sse.h +index bb565ed546..d5a717e18c 100644 +--- a/dpdk/examples/l3fwd/l3fwd_sse.h ++++ b/dpdk/examples/l3fwd/l3fwd_sse.h +@@ -64,7 +64,7 @@ processx4_step3(struct rte_mbuf *pkt[FWDSTEP], uint16_t dst_port[FWDSTEP]) + + /* + * Group consecutive packets with the same destination port in bursts of 4. +- * Suppose we have array of destionation ports: ++ * Suppose we have array of destination ports: + * dst_port[] = {a, b, c, d,, e, ... } + * dp1 should contain: , dp2: . + * We doing 4 comparisons at once and the result is 4 bit mask. +diff --git a/dpdk/examples/l3fwd/main.c b/dpdk/examples/l3fwd/main.c +index eb68ffc5aa..3a0e15109b 100644 +--- a/dpdk/examples/l3fwd/main.c ++++ b/dpdk/examples/l3fwd/main.c +@@ -53,9 +53,8 @@ + + #define MAX_LCORE_PARAMS 1024 + +-/* Static global variables used within this file. */ +-static uint16_t nb_rxd = RTE_TEST_RX_DESC_DEFAULT; +-static uint16_t nb_txd = RTE_TEST_TX_DESC_DEFAULT; ++uint16_t nb_rxd = RTE_TEST_RX_DESC_DEFAULT; ++uint16_t nb_txd = RTE_TEST_TX_DESC_DEFAULT; + + /**< Ports set in promiscuous mode off by default. */ + static int promiscuous_on; +@@ -342,6 +341,8 @@ print_usage(const char *prgname) + " [-P]" + " [--lookup]" + " --config (port,queue,lcore)[,(port,queue,lcore)]" ++ " [--rx-queue-size NPKTS]" ++ " [--tx-queue-size NPKTS]" + " [--eth-dest=X,MM:MM:MM:MM:MM:MM]" + " [--max-pkt-len PKTLEN]" + " [--no-numa]" +@@ -361,6 +362,10 @@ print_usage(const char *prgname) + " Default: lpm\n" + " Accepted: em (Exact Match), lpm (Longest Prefix Match), fib (Forwarding Information Base)\n" + " --config (port,queue,lcore): Rx queue configuration\n" ++ " --rx-queue-size NPKTS: Rx queue size in decimal\n" ++ " Default: %d\n" ++ " --tx-queue-size NPKTS: Tx queue size in decimal\n" ++ " Default: %d\n" + " --eth-dest=X,MM:MM:MM:MM:MM:MM: Ethernet destination for port X\n" + " --max-pkt-len PKTLEN: maximum packet length in decimal (64-9600)\n" + " --no-numa: Disable numa awareness\n" +@@ -382,7 +387,7 @@ print_usage(const char *prgname) + " --event-vector-tmo: Max timeout to form vector in nanoseconds if event vectorization is enabled\n" + " -E : Enable exact match, legacy flag please use --lookup=em instead\n" + " -L : Enable longest prefix match, legacy flag please use --lookup=lpm instead\n\n", +- prgname); ++ prgname, RTE_TEST_RX_DESC_DEFAULT, RTE_TEST_TX_DESC_DEFAULT); + } + + static int +@@ -525,6 +530,38 @@ parse_mode(const char *optarg) + evt_rsrc->enabled = true; + } + ++static void ++parse_queue_size(const char *queue_size_arg, uint16_t *queue_size, int rx) ++{ ++ char *end = NULL; ++ unsigned long value; ++ ++ /* parse decimal string */ ++ value = strtoul(queue_size_arg, &end, 10); ++ if ((queue_size_arg[0] == '\0') || (end == NULL) || ++ (*end != '\0') || (value == 0)) { ++ if (rx == 1) ++ rte_exit(EXIT_FAILURE, "Invalid rx-queue-size\n"); ++ else ++ rte_exit(EXIT_FAILURE, "Invalid tx-queue-size\n"); ++ ++ return; ++ } ++ ++ if (value > UINT16_MAX) { ++ if (rx == 1) ++ rte_exit(EXIT_FAILURE, "rx-queue-size %lu > %d\n", ++ value, UINT16_MAX); ++ else ++ rte_exit(EXIT_FAILURE, "tx-queue-size %lu > %d\n", ++ value, UINT16_MAX); ++ ++ return; ++ } ++ ++ *queue_size = value; ++} ++ + static void + parse_eventq_sched(const char *optarg) + { +@@ -582,6 +619,8 @@ static const char short_options[] = + ; + + #define CMD_LINE_OPT_CONFIG "config" ++#define CMD_LINE_OPT_RX_QUEUE_SIZE "rx-queue-size" ++#define CMD_LINE_OPT_TX_QUEUE_SIZE "tx-queue-size" + #define CMD_LINE_OPT_ETH_DEST "eth-dest" + #define CMD_LINE_OPT_NO_NUMA "no-numa" + #define CMD_LINE_OPT_IPV6 "ipv6" +@@ -604,6 +643,8 @@ enum { + * conflict with short options */ + CMD_LINE_OPT_MIN_NUM = 256, + CMD_LINE_OPT_CONFIG_NUM, ++ CMD_LINE_OPT_RX_QUEUE_SIZE_NUM, ++ CMD_LINE_OPT_TX_QUEUE_SIZE_NUM, + CMD_LINE_OPT_ETH_DEST_NUM, + CMD_LINE_OPT_NO_NUMA_NUM, + CMD_LINE_OPT_IPV6_NUM, +@@ -622,6 +663,8 @@ enum { + + static const struct option lgopts[] = { + {CMD_LINE_OPT_CONFIG, 1, 0, CMD_LINE_OPT_CONFIG_NUM}, ++ {CMD_LINE_OPT_RX_QUEUE_SIZE, 1, 0, CMD_LINE_OPT_RX_QUEUE_SIZE_NUM}, ++ {CMD_LINE_OPT_TX_QUEUE_SIZE, 1, 0, CMD_LINE_OPT_TX_QUEUE_SIZE_NUM}, + {CMD_LINE_OPT_ETH_DEST, 1, 0, CMD_LINE_OPT_ETH_DEST_NUM}, + {CMD_LINE_OPT_NO_NUMA, 0, 0, CMD_LINE_OPT_NO_NUMA_NUM}, + {CMD_LINE_OPT_IPV6, 0, 0, CMD_LINE_OPT_IPV6_NUM}, +@@ -715,6 +758,14 @@ parse_args(int argc, char **argv) + lcore_params = 1; + break; + ++ case CMD_LINE_OPT_RX_QUEUE_SIZE_NUM: ++ parse_queue_size(optarg, &nb_rxd, 1); ++ break; ++ ++ case CMD_LINE_OPT_TX_QUEUE_SIZE_NUM: ++ parse_queue_size(optarg, &nb_txd, 0); ++ break; ++ + case CMD_LINE_OPT_ETH_DEST_NUM: + parse_eth_dest(optarg); + break; +diff --git a/dpdk/examples/link_status_interrupt/main.c b/dpdk/examples/link_status_interrupt/main.c +index 551f0524da..9699e14ce6 100644 +--- a/dpdk/examples/link_status_interrupt/main.c ++++ b/dpdk/examples/link_status_interrupt/main.c +@@ -101,9 +101,10 @@ struct lsi_port_statistics { + struct lsi_port_statistics port_statistics[RTE_MAX_ETHPORTS]; + + /* A tsc-based timer responsible for triggering statistics printout */ +-#define TIMER_MILLISECOND 2000000ULL /* around 1ms at 2 Ghz */ ++#define TIMER_MILLISECOND (rte_get_timer_hz() / 1000) + #define MAX_TIMER_PERIOD 86400 /* 1 day max */ +-static int64_t timer_period = 10 * TIMER_MILLISECOND * 1000; /* default period is 10 seconds */ ++#define DEFAULT_TIMER_PERIOD 10UL /* default period is 10 seconds */ ++static int64_t timer_period; + + /* Print out statistics on packets dropped */ + static void +@@ -370,6 +371,8 @@ lsi_parse_args(int argc, char **argv) + {NULL, 0, 0, 0} + }; + ++ timer_period = DEFAULT_TIMER_PERIOD * TIMER_MILLISECOND * 1000; ++ + argvopt = argv; + + while ((opt = getopt_long(argc, argvopt, "p:q:T:", +diff --git a/dpdk/examples/multi_process/hotplug_mp/commands.c b/dpdk/examples/multi_process/hotplug_mp/commands.c +index 48fd329583..41ea265e45 100644 +--- a/dpdk/examples/multi_process/hotplug_mp/commands.c ++++ b/dpdk/examples/multi_process/hotplug_mp/commands.c +@@ -175,7 +175,7 @@ static void cmd_dev_detach_parsed(void *parsed_result, + cmdline_printf(cl, "detached device %s\n", + da.name); + else +- cmdline_printf(cl, "failed to dettach device %s\n", ++ cmdline_printf(cl, "failed to detach device %s\n", + da.name); + rte_devargs_reset(&da); + } +diff --git a/dpdk/examples/multi_process/simple_mp/main.c b/dpdk/examples/multi_process/simple_mp/main.c +index 5df2a39000..9d5f1088b0 100644 +--- a/dpdk/examples/multi_process/simple_mp/main.c ++++ b/dpdk/examples/multi_process/simple_mp/main.c +@@ -4,7 +4,7 @@ + + /* + * This sample application is a simple multi-process application which +- * demostrates sharing of queues and memory pools between processes, and ++ * demonstrates sharing of queues and memory pools between processes, and + * using those queues/pools for communication between the processes. + * + * Application is designed to run with two processes, a primary and a +diff --git a/dpdk/examples/multi_process/symmetric_mp/main.c b/dpdk/examples/multi_process/symmetric_mp/main.c +index b35886a77b..050337765f 100644 +--- a/dpdk/examples/multi_process/symmetric_mp/main.c ++++ b/dpdk/examples/multi_process/symmetric_mp/main.c +@@ -3,7 +3,7 @@ + */ + + /* +- * Sample application demostrating how to do packet I/O in a multi-process ++ * Sample application demonstrating how to do packet I/O in a multi-process + * environment. The same code can be run as a primary process and as a + * secondary process, just with a different proc-id parameter in each case + * (apart from the EAL flag to indicate a secondary process). +diff --git a/dpdk/examples/ntb/ntb_fwd.c b/dpdk/examples/ntb/ntb_fwd.c +index f110fc129f..81964d0308 100644 +--- a/dpdk/examples/ntb/ntb_fwd.c ++++ b/dpdk/examples/ntb/ntb_fwd.c +@@ -696,7 +696,7 @@ assign_stream_to_lcores(void) + break; + } + +- /* Print packet forwading config. */ ++ /* Print packet forwarding config. */ + RTE_LCORE_FOREACH_WORKER(lcore_id) { + conf = &fwd_lcore_conf[lcore_id]; + +diff --git a/dpdk/examples/packet_ordering/main.c b/dpdk/examples/packet_ordering/main.c +index b01ac60fd1..99e67ef67b 100644 +--- a/dpdk/examples/packet_ordering/main.c ++++ b/dpdk/examples/packet_ordering/main.c +@@ -686,7 +686,7 @@ main(int argc, char **argv) + if (ret < 0) + rte_exit(EXIT_FAILURE, "Invalid packet_ordering arguments\n"); + +- /* Check if we have enought cores */ ++ /* Check if we have enough cores */ + if (rte_lcore_count() < 3) + rte_exit(EXIT_FAILURE, "Error, This application needs at " + "least 3 logical cores to run:\n" +diff --git a/dpdk/examples/performance-thread/common/lthread.c b/dpdk/examples/performance-thread/common/lthread.c +index 009374a8c3..b02e0fc13a 100644 +--- a/dpdk/examples/performance-thread/common/lthread.c ++++ b/dpdk/examples/performance-thread/common/lthread.c +@@ -178,7 +178,7 @@ lthread_create(struct lthread **new_lt, int lcore_id, + bzero(lt, sizeof(struct lthread)); + lt->root_sched = THIS_SCHED; + +- /* set the function args and exit handlder */ ++ /* set the function args and exit handler */ + _lthread_init(lt, fun, arg, _lthread_exit_handler); + + /* put it in the ready queue */ +@@ -384,7 +384,7 @@ void lthread_exit(void *ptr) + } + + +- /* wait until the joinging thread has collected the exit value */ ++ /* wait until the joining thread has collected the exit value */ + while (lt->join != LT_JOIN_EXIT_VAL_READ) + _reschedule(); + +@@ -410,7 +410,7 @@ int lthread_join(struct lthread *lt, void **ptr) + /* invalid to join a detached thread, or a thread that is joined */ + if ((lt_state & BIT(ST_LT_DETACH)) || (lt->join == LT_JOIN_THREAD_SET)) + return POSIX_ERRNO(EINVAL); +- /* pointer to the joining thread and a poingter to return a value */ ++ /* pointer to the joining thread and a pointer to return a value */ + lt->lt_join = current; + current->lt_exit_ptr = ptr; + /* There is a race between lthread_join() and lthread_exit() +diff --git a/dpdk/examples/performance-thread/common/lthread_diag.c b/dpdk/examples/performance-thread/common/lthread_diag.c +index 57760a1e23..b1bdf7a30c 100644 +--- a/dpdk/examples/performance-thread/common/lthread_diag.c ++++ b/dpdk/examples/performance-thread/common/lthread_diag.c +@@ -232,7 +232,7 @@ lthread_sched_stats_display(void) + } + + /* +- * Defafult diagnostic callback ++ * Default diagnostic callback + */ + static uint64_t + _lthread_diag_default_cb(uint64_t time, struct lthread *lt, int diag_event, +diff --git a/dpdk/examples/performance-thread/common/lthread_int.h b/dpdk/examples/performance-thread/common/lthread_int.h +index d010126f16..ec018e34a1 100644 +--- a/dpdk/examples/performance-thread/common/lthread_int.h ++++ b/dpdk/examples/performance-thread/common/lthread_int.h +@@ -107,7 +107,7 @@ enum join_st { + LT_JOIN_EXIT_VAL_READ, /* joining thread has collected ret val */ + }; + +-/* defnition of an lthread stack object */ ++/* definition of an lthread stack object */ + struct lthread_stack { + uint8_t stack[LTHREAD_MAX_STACK_SIZE]; + size_t stack_size; +diff --git a/dpdk/examples/performance-thread/common/lthread_tls.c b/dpdk/examples/performance-thread/common/lthread_tls.c +index 4ab2e3558b..bae45f2aa9 100644 +--- a/dpdk/examples/performance-thread/common/lthread_tls.c ++++ b/dpdk/examples/performance-thread/common/lthread_tls.c +@@ -215,7 +215,7 @@ void _lthread_tls_alloc(struct lthread *lt) + tls->root_sched = (THIS_SCHED); + lt->tls = tls; + +- /* allocate data for TLS varaiables using RTE_PER_LTHREAD macros */ ++ /* allocate data for TLS variables using RTE_PER_LTHREAD macros */ + if (sizeof(void *) < (uint64_t)RTE_PER_LTHREAD_SECTION_SIZE) { + lt->per_lthread_data = + _lthread_objcache_alloc((THIS_SCHED)->per_lthread_cache); +diff --git a/dpdk/examples/performance-thread/l3fwd-thread/main.c b/dpdk/examples/performance-thread/l3fwd-thread/main.c +index 8a35040597..1ddb2a9138 100644 +--- a/dpdk/examples/performance-thread/l3fwd-thread/main.c ++++ b/dpdk/examples/performance-thread/l3fwd-thread/main.c +@@ -125,7 +125,7 @@ cb_parse_ptype(__rte_unused uint16_t port, __rte_unused uint16_t queue, + } + + /* +- * When set to zero, simple forwaring path is eanbled. ++ * When set to zero, simple forwarding path is enabled. + * When set to one, optimized forwarding path is enabled. + * Note that LPM optimisation path uses SSE4.1 instructions. + */ +@@ -1529,7 +1529,7 @@ processx4_step3(struct rte_mbuf *pkt[FWDSTEP], uint16_t dst_port[FWDSTEP]) + } + + /* +- * We group consecutive packets with the same destionation port into one burst. ++ * We group consecutive packets with the same destination port into one burst. + * To avoid extra latency this is done together with some other packet + * processing, but after we made a final decision about packet's destination. + * To do this we maintain: +@@ -1554,7 +1554,7 @@ processx4_step3(struct rte_mbuf *pkt[FWDSTEP], uint16_t dst_port[FWDSTEP]) + + /* + * Group consecutive packets with the same destination port in bursts of 4. +- * Suppose we have array of destionation ports: ++ * Suppose we have array of destination ports: + * dst_port[] = {a, b, c, d,, e, ... } + * dp1 should contain: , dp2: . + * We doing 4 comparisons at once and the result is 4 bit mask. +@@ -1565,7 +1565,7 @@ port_groupx4(uint16_t pn[FWDSTEP + 1], uint16_t *lp, __m128i dp1, __m128i dp2) + { + static const struct { + uint64_t pnum; /* prebuild 4 values for pnum[]. */ +- int32_t idx; /* index for new last updated elemnet. */ ++ int32_t idx; /* index for new last updated element. */ + uint16_t lpv; /* add value to the last updated element. */ + } gptbl[GRPSZ] = { + { +@@ -1834,7 +1834,7 @@ process_burst(struct rte_mbuf *pkts_burst[MAX_PKT_BURST], int nb_rx, + + /* + * Send packets out, through destination port. +- * Consecuteve pacekts with the same destination port ++ * Consecutive packets with the same destination port + * are already grouped together. + * If destination port for the packet equals BAD_PORT, + * then free the packet without sending it out. +@@ -3514,7 +3514,7 @@ main(int argc, char **argv) + + ret = rte_timer_subsystem_init(); + if (ret < 0) +- rte_exit(EXIT_FAILURE, "Failed to initialize timer subystem\n"); ++ rte_exit(EXIT_FAILURE, "Failed to initialize timer subsystem\n"); + + /* pre-init dst MACs for all ports to 02:00:00:00:00:xx */ + for (portid = 0; portid < RTE_MAX_ETHPORTS; portid++) { +diff --git a/dpdk/examples/performance-thread/pthread_shim/pthread_shim.c b/dpdk/examples/performance-thread/pthread_shim/pthread_shim.c +index bbc076584b..a44cb8244d 100644 +--- a/dpdk/examples/performance-thread/pthread_shim/pthread_shim.c ++++ b/dpdk/examples/performance-thread/pthread_shim/pthread_shim.c +@@ -586,6 +586,11 @@ pthread_t pthread_self(void) + return _sys_pthread_funcs.f_pthread_self(); + } + ++#if defined(RTE_TOOLCHAIN_GCC) && (GCC_VERSION >= 120000) ++#pragma GCC diagnostic push ++#pragma GCC diagnostic ignored "-Wmaybe-uninitialized" ++#endif ++ + int pthread_setspecific(pthread_key_t key, const void *data) + { + if (override) { +@@ -595,6 +600,10 @@ int pthread_setspecific(pthread_key_t key, const void *data) + return _sys_pthread_funcs.f_pthread_setspecific(key, data); + } + ++#if defined(RTE_TOOLCHAIN_GCC) && (GCC_VERSION >= 120000) ++#pragma GCC diagnostic pop ++#endif ++ + int pthread_spin_init(pthread_spinlock_t *a, int b) + { + NOT_IMPLEMENTED; +diff --git a/dpdk/examples/performance-thread/pthread_shim/pthread_shim.h b/dpdk/examples/performance-thread/pthread_shim/pthread_shim.h +index e90fb15fc1..ce51627a5b 100644 +--- a/dpdk/examples/performance-thread/pthread_shim/pthread_shim.h ++++ b/dpdk/examples/performance-thread/pthread_shim/pthread_shim.h +@@ -41,7 +41,7 @@ + * + * The decision whether to invoke the real library function or the lthread + * function is controlled by a per pthread flag that can be switched +- * on of off by the pthread_override_set() API described below. Typcially ++ * on of off by the pthread_override_set() API described below. Typically + * this should be done as the first action of the initial lthread. + * + * N.B In general it would be poor practice to revert to invoke a real +diff --git a/dpdk/examples/pipeline/examples/registers.spec b/dpdk/examples/pipeline/examples/registers.spec +index 74a014ad06..59998fef03 100644 +--- a/dpdk/examples/pipeline/examples/registers.spec ++++ b/dpdk/examples/pipeline/examples/registers.spec +@@ -4,7 +4,7 @@ + ; This program is setting up two register arrays called "pkt_counters" and "byte_counters". + ; On every input packet (Ethernet/IPv4), the "pkt_counters" register at location indexed by + ; the IPv4 header "Source Address" field is incremented, while the same location in the +-; "byte_counters" array accummulates the value of the IPv4 header "Total Length" field. ++; "byte_counters" array accumulates the value of the IPv4 header "Total Length" field. + ; + ; The "regrd" and "regwr" CLI commands can be used to read and write the current value of + ; any register array location. +diff --git a/dpdk/examples/qos_sched/args.c b/dpdk/examples/qos_sched/args.c +index 10ca7bea61..ff51d0215a 100644 +--- a/dpdk/examples/qos_sched/args.c ++++ b/dpdk/examples/qos_sched/args.c +@@ -11,6 +11,7 @@ + #include + #include + ++#include + #include + #include + #include +@@ -427,13 +428,13 @@ app_parse_args(int argc, char **argv) + + /* check main core index validity */ + for (i = 0; i <= app_main_core; i++) { +- if (app_used_core_mask & (1u << app_main_core)) { ++ if (app_used_core_mask & RTE_BIT64(app_main_core)) { + RTE_LOG(ERR, APP, "Main core index is not configured properly\n"); + app_usage(prgname); + return -1; + } + } +- app_used_core_mask |= 1u << app_main_core; ++ app_used_core_mask |= RTE_BIT64(app_main_core); + + if ((app_used_core_mask != app_eal_core_mask()) || + (app_main_core != rte_get_main_lcore())) { +diff --git a/dpdk/examples/qos_sched/cmdline.c b/dpdk/examples/qos_sched/cmdline.c +index 257b87a7cf..6691b02d89 100644 +--- a/dpdk/examples/qos_sched/cmdline.c ++++ b/dpdk/examples/qos_sched/cmdline.c +@@ -41,7 +41,7 @@ static void cmd_help_parsed(__rte_unused void *parsed_result, + " qavg port X subport Y pipe Z : Show average queue size per pipe.\n" + " qavg port X subport Y pipe Z tc A : Show average queue size per pipe and TC.\n" + " qavg port X subport Y pipe Z tc A q B : Show average queue size of a specific queue.\n" +- " qavg [n|period] X : Set number of times and peiod (us).\n\n" ++ " qavg [n|period] X : Set number of times and period (us).\n\n" + ); + + } +diff --git a/dpdk/examples/server_node_efd/node/node.c b/dpdk/examples/server_node_efd/node/node.c +index ba1c7e5153..fc2aa5ffef 100644 +--- a/dpdk/examples/server_node_efd/node/node.c ++++ b/dpdk/examples/server_node_efd/node/node.c +@@ -296,7 +296,7 @@ handle_packets(struct rte_hash *h, struct rte_mbuf **bufs, uint16_t num_packets) + } + } + } +-/* >8 End of packets dequeueing. */ ++/* >8 End of packets dequeuing. */ + + /* + * Application main function - loops through +diff --git a/dpdk/examples/skeleton/basicfwd.c b/dpdk/examples/skeleton/basicfwd.c +index 16435ee3cc..518cd72179 100644 +--- a/dpdk/examples/skeleton/basicfwd.c ++++ b/dpdk/examples/skeleton/basicfwd.c +@@ -179,7 +179,7 @@ main(int argc, char *argv[]) + int ret = rte_eal_init(argc, argv); + if (ret < 0) + rte_exit(EXIT_FAILURE, "Error with EAL initialization\n"); +- /* >8 End of initializion the Environment Abstraction Layer (EAL). */ ++ /* >8 End of initialization the Environment Abstraction Layer (EAL). */ + + argc -= ret; + argv += ret; +diff --git a/dpdk/examples/vhost/main.c b/dpdk/examples/vhost/main.c +index 33d023aa39..f9e932061f 100644 +--- a/dpdk/examples/vhost/main.c ++++ b/dpdk/examples/vhost/main.c +@@ -32,6 +32,8 @@ + #define MAX_QUEUES 128 + #endif + ++#define NUM_MBUFS_DEFAULT 0x24000 ++ + /* the maximum number of external ports supported */ + #define MAX_SUP_PORTS 1 + +@@ -57,6 +59,9 @@ + + #define INVALID_PORT_ID 0xFF + ++/* number of mbufs in all pools - if specified on command-line. */ ++static int total_num_mbufs = NUM_MBUFS_DEFAULT; ++ + /* mask of enabled ports */ + static uint32_t enabled_port_mask = 0; + +@@ -107,7 +112,7 @@ static uint32_t burst_rx_retry_num = BURST_RX_RETRIES; + static char *socket_files; + static int nb_sockets; + +-/* empty vmdq configuration structure. Filled in programatically */ ++/* empty VMDq configuration structure. Filled in programmatically */ + static struct rte_eth_conf vmdq_conf_default = { + .rxmode = { + .mq_mode = RTE_ETH_MQ_RX_VMDQ_ONLY, +@@ -115,7 +120,7 @@ static struct rte_eth_conf vmdq_conf_default = { + /* + * VLAN strip is necessary for 1G NIC such as I350, + * this fixes bug of ipv4 forwarding in guest can't +- * forward pakets from one virtio dev to another virtio dev. ++ * forward packets from one virtio dev to another virtio dev. + */ + .offloads = RTE_ETH_RX_OFFLOAD_VLAN_STRIP, + }, +@@ -259,6 +264,10 @@ port_init(uint16_t port) + + return retval; + } ++ if (dev_info.max_vmdq_pools == 0) { ++ RTE_LOG(ERR, VHOST_PORT, "Failed to get VMDq info.\n"); ++ return -1; ++ } + + rxconf = &dev_info.default_rxconf; + txconf = &dev_info.default_txconf; +@@ -463,7 +472,7 @@ us_vhost_usage(const char *prgname) + " --nb-devices ND\n" + " -p PORTMASK: Set mask for ports to be used by application\n" + " --vm2vm [0|1|2]: disable/software(default)/hardware vm2vm comms\n" +- " --rx-retry [0|1]: disable/enable(default) retries on rx. Enable retry if destintation queue is full\n" ++ " --rx-retry [0|1]: disable/enable(default) retries on Rx. Enable retry if destination queue is full\n" + " --rx-retry-delay [0-N]: timeout(in usecond) between retries on RX. This makes effect only if retries on rx enabled\n" + " --rx-retry-num [0-N]: the number of retries on rx. This makes effect only if retries on rx enabled\n" + " --mergeable [0|1]: disable(default)/enable RX mergeable buffers\n" +@@ -473,7 +482,8 @@ us_vhost_usage(const char *prgname) + " --tso [0|1] disable/enable TCP segment offload.\n" + " --client register a vhost-user socket as client mode.\n" + " --dma-type register dma type for your vhost async driver. For example \"ioat\" for now.\n" +- " --dmas register dma channel for specific vhost device.\n", ++ " --dmas register dma channel for specific vhost device.\n" ++ " --total-num-mbufs [0-N] set the number of mbufs to be allocated in mbuf pools, the default value is 147456.\n", + prgname); + } + +@@ -504,6 +514,8 @@ enum { + OPT_DMA_TYPE_NUM, + #define OPT_DMAS "dmas" + OPT_DMAS_NUM, ++#define OPT_NUM_MBUFS "total-num-mbufs" ++ OPT_NUM_MBUFS_NUM, + }; + + /* +@@ -543,6 +555,8 @@ us_vhost_parse_args(int argc, char **argv) + NULL, OPT_DMA_TYPE_NUM}, + {OPT_DMAS, required_argument, + NULL, OPT_DMAS_NUM}, ++ {OPT_NUM_MBUFS, required_argument, ++ NULL, OPT_NUM_MBUFS_NUM}, + {NULL, 0, 0, 0}, + }; + +@@ -675,6 +689,19 @@ us_vhost_parse_args(int argc, char **argv) + async_vhost_driver = 1; + break; + ++ case OPT_NUM_MBUFS_NUM: ++ ret = parse_num_opt(optarg, INT32_MAX); ++ if (ret == -1) { ++ RTE_LOG(INFO, VHOST_CONFIG, ++ "Invalid argument for total-num-mbufs [0..N]\n"); ++ us_vhost_usage(prgname); ++ return -1; ++ } ++ ++ if (total_num_mbufs < ret) ++ total_num_mbufs = ret; ++ break; ++ + case OPT_CLIENT_NUM: + client_mode = 1; + break; +@@ -873,31 +900,43 @@ sync_virtio_xmit(struct vhost_dev *dst_vdev, struct vhost_dev *src_vdev, + } + } + +-static __rte_always_inline void +-drain_vhost(struct vhost_dev *vdev) ++static __rte_always_inline uint16_t ++enqueue_pkts(struct vhost_dev *vdev, struct rte_mbuf **pkts, uint16_t rx_count) + { +- uint16_t ret; +- uint32_t buff_idx = rte_lcore_id() * MAX_VHOST_DEVICE + vdev->vid; +- uint16_t nr_xmit = vhost_txbuff[buff_idx]->len; +- struct rte_mbuf **m = vhost_txbuff[buff_idx]->m_table; ++ uint16_t enqueue_count; + + if (builtin_net_driver) { +- ret = vs_enqueue_pkts(vdev, VIRTIO_RXQ, m, nr_xmit); ++ enqueue_count = vs_enqueue_pkts(vdev, VIRTIO_RXQ, pkts, rx_count); + } else if (async_vhost_driver) { + uint16_t enqueue_fail = 0; + + complete_async_pkts(vdev); +- ret = rte_vhost_submit_enqueue_burst(vdev->vid, VIRTIO_RXQ, m, nr_xmit); +- __atomic_add_fetch(&vdev->pkts_inflight, ret, __ATOMIC_SEQ_CST); ++ enqueue_count = rte_vhost_submit_enqueue_burst(vdev->vid, ++ VIRTIO_RXQ, pkts, rx_count); ++ __atomic_add_fetch(&vdev->pkts_inflight, enqueue_count, __ATOMIC_SEQ_CST); + +- enqueue_fail = nr_xmit - ret; ++ enqueue_fail = rx_count - enqueue_count; + if (enqueue_fail) +- free_pkts(&m[ret], nr_xmit - ret); ++ free_pkts(&pkts[enqueue_count], enqueue_fail); ++ + } else { +- ret = rte_vhost_enqueue_burst(vdev->vid, VIRTIO_RXQ, +- m, nr_xmit); ++ enqueue_count = rte_vhost_enqueue_burst(vdev->vid, VIRTIO_RXQ, ++ pkts, rx_count); + } + ++ return enqueue_count; ++} ++ ++static __rte_always_inline void ++drain_vhost(struct vhost_dev *vdev) ++{ ++ uint16_t ret; ++ uint32_t buff_idx = rte_lcore_id() * MAX_VHOST_DEVICE + vdev->vid; ++ uint16_t nr_xmit = vhost_txbuff[buff_idx]->len; ++ struct rte_mbuf **m = vhost_txbuff[buff_idx]->m_table; ++ ++ ret = enqueue_pkts(vdev, m, nr_xmit); ++ + if (enable_stats) { + __atomic_add_fetch(&vdev->stats.rx_total_atomic, nr_xmit, + __ATOMIC_SEQ_CST); +@@ -1190,44 +1229,19 @@ drain_eth_rx(struct vhost_dev *vdev) + if (!rx_count) + return; + +- /* +- * When "enable_retry" is set, here we wait and retry when there +- * is no enough free slots in the queue to hold @rx_count packets, +- * to diminish packet loss. +- */ +- if (enable_retry && +- unlikely(rx_count > rte_vhost_avail_entries(vdev->vid, +- VIRTIO_RXQ))) { +- uint32_t retry; ++ enqueue_count = enqueue_pkts(vdev, pkts, rx_count); ++ ++ /* Retry if necessary */ ++ if (enable_retry && unlikely(enqueue_count < rx_count)) { ++ uint32_t retry = 0; + +- for (retry = 0; retry < burst_rx_retry_num; retry++) { ++ while (enqueue_count < rx_count && retry++ < burst_rx_retry_num) { + rte_delay_us(burst_rx_delay_time); +- if (rx_count <= rte_vhost_avail_entries(vdev->vid, +- VIRTIO_RXQ)) +- break; ++ enqueue_count += enqueue_pkts(vdev, &pkts[enqueue_count], ++ rx_count - enqueue_count); + } + } + +- if (builtin_net_driver) { +- enqueue_count = vs_enqueue_pkts(vdev, VIRTIO_RXQ, +- pkts, rx_count); +- } else if (async_vhost_driver) { +- uint16_t enqueue_fail = 0; +- +- complete_async_pkts(vdev); +- enqueue_count = rte_vhost_submit_enqueue_burst(vdev->vid, +- VIRTIO_RXQ, pkts, rx_count); +- __atomic_add_fetch(&vdev->pkts_inflight, enqueue_count, __ATOMIC_SEQ_CST); +- +- enqueue_fail = rx_count - enqueue_count; +- if (enqueue_fail) +- free_pkts(&pkts[enqueue_count], enqueue_fail); +- +- } else { +- enqueue_count = rte_vhost_enqueue_burst(vdev->vid, VIRTIO_RXQ, +- pkts, rx_count); +- } +- + if (enable_stats) { + __atomic_add_fetch(&vdev->stats.rx_total_atomic, rx_count, + __ATOMIC_SEQ_CST); +@@ -1289,7 +1303,7 @@ switch_worker(void *arg __rte_unused) + struct vhost_dev *vdev; + struct mbuf_table *tx_q; + +- RTE_LOG(INFO, VHOST_DATA, "Procesing on Core %u started\n", lcore_id); ++ RTE_LOG(INFO, VHOST_DATA, "Processing on Core %u started\n", lcore_id); + + tx_q = &lcore_tx_queue[lcore_id]; + for (i = 0; i < rte_lcore_count(); i++) { +@@ -1333,7 +1347,7 @@ switch_worker(void *arg __rte_unused) + + /* + * Remove a device from the specific data core linked list and from the +- * main linked list. Synchonization occurs through the use of the ++ * main linked list. Synchronization occurs through the use of the + * lcore dev_removal_flag. Device is made volatile here to avoid re-ordering + * of dev->remove=1 which can cause an infinite loop in the rte_pause loop. + */ +@@ -1606,57 +1620,6 @@ sigint_handler(__rte_unused int signum) + exit(0); + } + +-/* +- * While creating an mbuf pool, one key thing is to figure out how +- * many mbuf entries is enough for our use. FYI, here are some +- * guidelines: +- * +- * - Each rx queue would reserve @nr_rx_desc mbufs at queue setup stage +- * +- * - For each switch core (A CPU core does the packet switch), we need +- * also make some reservation for receiving the packets from virtio +- * Tx queue. How many is enough depends on the usage. It's normally +- * a simple calculation like following: +- * +- * MAX_PKT_BURST * max packet size / mbuf size +- * +- * So, we definitely need allocate more mbufs when TSO is enabled. +- * +- * - Similarly, for each switching core, we should serve @nr_rx_desc +- * mbufs for receiving the packets from physical NIC device. +- * +- * - We also need make sure, for each switch core, we have allocated +- * enough mbufs to fill up the mbuf cache. +- */ +-static void +-create_mbuf_pool(uint16_t nr_port, uint32_t nr_switch_core, uint32_t mbuf_size, +- uint32_t nr_queues, uint32_t nr_rx_desc, uint32_t nr_mbuf_cache) +-{ +- uint32_t nr_mbufs; +- uint32_t nr_mbufs_per_core; +- uint32_t mtu = 1500; +- +- if (mergeable) +- mtu = 9000; +- if (enable_tso) +- mtu = 64 * 1024; +- +- nr_mbufs_per_core = (mtu + mbuf_size) * MAX_PKT_BURST / +- (mbuf_size - RTE_PKTMBUF_HEADROOM); +- nr_mbufs_per_core += nr_rx_desc; +- nr_mbufs_per_core = RTE_MAX(nr_mbufs_per_core, nr_mbuf_cache); +- +- nr_mbufs = nr_queues * nr_rx_desc; +- nr_mbufs += nr_mbufs_per_core * nr_switch_core; +- nr_mbufs *= nr_port; +- +- mbuf_pool = rte_pktmbuf_pool_create("MBUF_POOL", nr_mbufs, +- nr_mbuf_cache, 0, mbuf_size, +- rte_socket_id()); +- if (mbuf_pool == NULL) +- rte_exit(EXIT_FAILURE, "Cannot create mbuf pool\n"); +-} +- + /* + * Main function, does initialisation and calls the per-lcore functions. + */ +@@ -1715,8 +1678,11 @@ main(int argc, char *argv[]) + * many queues here. We probably should only do allocation for + * those queues we are going to use. + */ +- create_mbuf_pool(valid_num_ports, rte_lcore_count() - 1, MBUF_DATA_SIZE, +- MAX_QUEUES, RTE_TEST_RX_DESC_DEFAULT, MBUF_CACHE_SIZE); ++ mbuf_pool = rte_pktmbuf_pool_create("MBUF_POOL", total_num_mbufs, ++ MBUF_CACHE_SIZE, 0, MBUF_DATA_SIZE, ++ rte_socket_id()); ++ if (mbuf_pool == NULL) ++ rte_exit(EXIT_FAILURE, "Cannot create mbuf pool\n"); + + if (vm2vm_mode == VM2VM_HARDWARE) { + /* Enable VT loop back to let L2 switch to do it. */ +diff --git a/dpdk/examples/vm_power_manager/channel_monitor.c b/dpdk/examples/vm_power_manager/channel_monitor.c +index d767423a40..97b8def7ca 100644 +--- a/dpdk/examples/vm_power_manager/channel_monitor.c ++++ b/dpdk/examples/vm_power_manager/channel_monitor.c +@@ -404,7 +404,7 @@ get_pcpu_to_control(struct policy *pol) + + /* + * So now that we're handling virtual and physical cores, we need to +- * differenciate between them when adding them to the branch monitor. ++ * differentiate between them when adding them to the branch monitor. + * Virtual cores need to be converted to physical cores. + */ + if (pol->pkt.core_type == RTE_POWER_CORE_TYPE_VIRTUAL) { +diff --git a/dpdk/examples/vm_power_manager/power_manager.h b/dpdk/examples/vm_power_manager/power_manager.h +index d35f8cbe01..d51039e2c6 100644 +--- a/dpdk/examples/vm_power_manager/power_manager.h ++++ b/dpdk/examples/vm_power_manager/power_manager.h +@@ -224,7 +224,7 @@ int power_manager_enable_turbo_core(unsigned int core_num); + int power_manager_disable_turbo_core(unsigned int core_num); + + /** +- * Get the current freuency of the core specified by core_num ++ * Get the current frequency of the core specified by core_num + * + * @param core_num + * The core number to get the current frequency +diff --git a/dpdk/examples/vmdq/main.c b/dpdk/examples/vmdq/main.c +index 2c00a942f1..10410b8783 100644 +--- a/dpdk/examples/vmdq/main.c ++++ b/dpdk/examples/vmdq/main.c +@@ -62,7 +62,7 @@ static uint8_t rss_enable; + + /* Default structure for VMDq. 8< */ + +-/* empty vmdq configuration structure. Filled in programatically */ ++/* empty VMDq configuration structure. Filled in programmatically */ + static const struct rte_eth_conf vmdq_conf_default = { + .rxmode = { + .mq_mode = RTE_ETH_MQ_RX_VMDQ_ONLY, +diff --git a/dpdk/kernel/freebsd/meson.build b/dpdk/kernel/freebsd/meson.build +index bf5aa20a55..1f612711be 100644 +--- a/dpdk/kernel/freebsd/meson.build ++++ b/dpdk/kernel/freebsd/meson.build +@@ -9,8 +9,8 @@ kmods = ['contigmem', 'nic_uio'] + # right now, which allows us to simplify things. We pull in the sourcer + # files from the individual meson.build files, and then use a custom + # target to call make, passing in the values as env parameters. +-kmod_cflags = ['-I' + meson.build_root(), +- '-I' + join_paths(meson.source_root(), 'config'), ++kmod_cflags = ['-I' + dpdk_build_root, ++ '-I' + join_paths(dpdk_source_root, 'config'), + '-include rte_config.h'] + + # to avoid warnings due to race conditions with creating the dev_if.h, etc. +diff --git a/dpdk/kernel/linux/kni/compat.h b/dpdk/kernel/linux/kni/compat.h +index 664785674f..3a86d12bbc 100644 +--- a/dpdk/kernel/linux/kni/compat.h ++++ b/dpdk/kernel/linux/kni/compat.h +@@ -141,3 +141,11 @@ + #if KERNEL_VERSION(5, 9, 0) > LINUX_VERSION_CODE + #define HAVE_TSK_IN_GUP + #endif ++ ++#if KERNEL_VERSION(5, 15, 0) <= LINUX_VERSION_CODE ++#define HAVE_ETH_HW_ADDR_SET ++#endif ++ ++#if KERNEL_VERSION(5, 18, 0) > LINUX_VERSION_CODE ++#define HAVE_NETIF_RX_NI ++#endif +diff --git a/dpdk/kernel/linux/kni/kni_fifo.h b/dpdk/kernel/linux/kni/kni_fifo.h +index 5c91b55379..1ba5172002 100644 +--- a/dpdk/kernel/linux/kni/kni_fifo.h ++++ b/dpdk/kernel/linux/kni/kni_fifo.h +@@ -41,7 +41,7 @@ kni_fifo_put(struct rte_kni_fifo *fifo, void **data, uint32_t num) + } + + /** +- * Get up to num elements from the fifo. Return the number actully read ++ * Get up to num elements from the FIFO. Return the number actually read + */ + static inline uint32_t + kni_fifo_get(struct rte_kni_fifo *fifo, void **data, uint32_t num) +diff --git a/dpdk/kernel/linux/kni/kni_misc.c b/dpdk/kernel/linux/kni/kni_misc.c +index f10dcd069d..feed12b568 100644 +--- a/dpdk/kernel/linux/kni/kni_misc.c ++++ b/dpdk/kernel/linux/kni/kni_misc.c +@@ -184,13 +184,17 @@ kni_dev_remove(struct kni_dev *dev) + if (!dev) + return -ENODEV; + ++ /* ++ * The memory of kni device is allocated and released together ++ * with net device. Release mbuf before freeing net device. ++ */ ++ kni_net_release_fifo_phy(dev); ++ + if (dev->net_dev) { + unregister_netdev(dev->net_dev); + free_netdev(dev->net_dev); + } + +- kni_net_release_fifo_phy(dev); +- + return 0; + } + +@@ -220,8 +224,8 @@ kni_release(struct inode *inode, struct file *file) + dev->pthread = NULL; + } + +- kni_dev_remove(dev); + list_del(&dev->list); ++ kni_dev_remove(dev); + } + up_write(&knet->kni_list_lock); + +@@ -400,11 +404,16 @@ kni_ioctl_create(struct net *net, uint32_t ioctl_num, + pr_debug("mbuf_size: %u\n", kni->mbuf_size); + + /* if user has provided a valid mac address */ +- if (is_valid_ether_addr(dev_info.mac_addr)) ++ if (is_valid_ether_addr(dev_info.mac_addr)) { ++#ifdef HAVE_ETH_HW_ADDR_SET ++ eth_hw_addr_set(net_dev, dev_info.mac_addr); ++#else + memcpy(net_dev->dev_addr, dev_info.mac_addr, ETH_ALEN); +- else +- /* Generate random MAC address. */ +- eth_random_addr(net_dev->dev_addr); ++#endif ++ } else { ++ /* Assign random MAC address. */ ++ eth_hw_addr_random(net_dev); ++ } + + if (dev_info.mtu) + net_dev->mtu = dev_info.mtu; +@@ -470,8 +479,8 @@ kni_ioctl_release(struct net *net, uint32_t ioctl_num, + dev->pthread = NULL; + } + +- kni_dev_remove(dev); + list_del(&dev->list); ++ kni_dev_remove(dev); + ret = 0; + break; + } +diff --git a/dpdk/kernel/linux/kni/kni_net.c b/dpdk/kernel/linux/kni/kni_net.c +index 29e5b9e21f..779ee3451a 100644 +--- a/dpdk/kernel/linux/kni/kni_net.c ++++ b/dpdk/kernel/linux/kni/kni_net.c +@@ -441,7 +441,11 @@ kni_net_rx_normal(struct kni_dev *kni) + skb->ip_summed = CHECKSUM_UNNECESSARY; + + /* Call netif interface */ ++#ifdef HAVE_NETIF_RX_NI + netif_rx_ni(skb); ++#else ++ netif_rx(skb); ++#endif + + /* Update statistics */ + dev->stats.rx_bytes += len; +@@ -779,7 +783,11 @@ kni_net_set_mac(struct net_device *netdev, void *p) + return -EADDRNOTAVAIL; + + memcpy(req.mac_addr, addr->sa_data, netdev->addr_len); ++#ifdef HAVE_ETH_HW_ADDR_SET ++ eth_hw_addr_set(netdev, addr->sa_data); ++#else + memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len); ++#endif + + ret = kni_net_process_request(netdev, &req); + +diff --git a/dpdk/kernel/linux/kni/meson.build b/dpdk/kernel/linux/kni/meson.build +index c683fc7b36..4c90069e99 100644 +--- a/dpdk/kernel/linux/kni/meson.build ++++ b/dpdk/kernel/linux/kni/meson.build +@@ -5,7 +5,7 @@ + # Ref: https://jira.devtools.intel.com/browse/DPDK-29263 + kmod_cflags = '' + file_path = kernel_source_dir + '/include/linux/netdevice.h' +-run_cmd = run_command('grep', 'ndo_tx_timeout', file_path) ++run_cmd = run_command('grep', 'ndo_tx_timeout', file_path, check: false) + + if run_cmd.stdout().contains('txqueue') == true + kmod_cflags = '-DHAVE_ARG_TX_QUEUE' +@@ -29,10 +29,10 @@ custom_target('rte_kni', + 'M=' + meson.current_build_dir(), + 'src=' + meson.current_source_dir(), + ' '.join(['MODULE_CFLAGS=', kmod_cflags,'-include ']) +- + meson.source_root() + '/config/rte_config.h' + +- ' -I' + meson.source_root() + '/lib/eal/include' + +- ' -I' + meson.source_root() + '/lib/kni' + +- ' -I' + meson.build_root() + ++ + dpdk_source_root + '/config/rte_config.h' + ++ ' -I' + dpdk_source_root + '/lib/eal/include' + ++ ' -I' + dpdk_source_root + '/lib/kni' + ++ ' -I' + dpdk_build_root + + ' -I' + meson.current_source_dir(), + 'modules'] + cross_args, + depends: kni_mkfile, +diff --git a/dpdk/kernel/linux/meson.build b/dpdk/kernel/linux/meson.build +index 0637452e95..d8fb20c1c3 100644 +--- a/dpdk/kernel/linux/meson.build ++++ b/dpdk/kernel/linux/meson.build +@@ -11,7 +11,7 @@ cross_args = [] + + if not meson.is_cross_build() + # native build +- kernel_version = run_command('uname', '-r').stdout().strip() ++ kernel_version = run_command('uname', '-r', check: true).stdout().strip() + kernel_install_dir = '/lib/modules/' + kernel_version + '/extra/dpdk' + if kernel_build_dir == '' + # use default path for native builds +@@ -24,14 +24,14 @@ if not meson.is_cross_build() + + # test running make in kernel directory, using "make kernelversion" + make_returncode = run_command('make', '-sC', kernel_build_dir, +- 'kernelversion').returncode() ++ 'kernelversion', check: true).returncode() + if make_returncode != 0 + # backward compatibility: + # the headers could still be in the 'build' subdir + if not kernel_build_dir.endswith('build') and not kernel_build_dir.endswith('build/') + kernel_build_dir = join_paths(kernel_build_dir, 'build') + make_returncode = run_command('make', '-sC', kernel_build_dir, +- 'kernelversion').returncode() ++ 'kernelversion', check: true).returncode() + endif + endif + +@@ -54,7 +54,8 @@ if kernel_build_dir == '' + endif + cross_compiler = find_program('c').path() + if cross_compiler.endswith('gcc') +- cross_prefix = run_command([py3, '-c', 'print("' + cross_compiler + '"[:-3])']).stdout().strip() ++ cross_prefix = run_command([py3, '-c', 'print("' + cross_compiler + '"[:-3])'], ++ check: true).stdout().strip() + elif cross_compiler.endswith('clang') + cross_prefix = '' + found_target = false +diff --git a/dpdk/lib/acl/acl_bld.c b/dpdk/lib/acl/acl_bld.c +index f316d3e875..2816632803 100644 +--- a/dpdk/lib/acl/acl_bld.c ++++ b/dpdk/lib/acl/acl_bld.c +@@ -12,6 +12,9 @@ + /* number of pointers per alloc */ + #define ACL_PTR_ALLOC 32 + ++/* account for situation when all fields are 8B long */ ++#define ACL_MAX_INDEXES (2 * RTE_ACL_MAX_FIELDS) ++ + /* macros for dividing rule sets heuristics */ + #define NODE_MAX 0x4000 + #define NODE_MIN 0x800 +@@ -80,7 +83,7 @@ struct acl_build_context { + struct tb_mem_pool pool; + struct rte_acl_trie tries[RTE_ACL_MAX_TRIES]; + struct rte_acl_bld_trie bld_tries[RTE_ACL_MAX_TRIES]; +- uint32_t data_indexes[RTE_ACL_MAX_TRIES][RTE_ACL_MAX_FIELDS]; ++ uint32_t data_indexes[RTE_ACL_MAX_TRIES][ACL_MAX_INDEXES]; + + /* memory free lists for nodes and blocks used for node ptrs */ + struct acl_mem_block blocks[MEM_BLOCK_NUM]; +@@ -885,7 +888,7 @@ acl_gen_range_trie(struct acl_build_context *context, + return root; + } + +- /* gather information about divirgent paths */ ++ /* gather information about divergent paths */ + lo_00 = 0; + hi_ff = UINT8_MAX; + for (k = n - 1; k >= 0; k--) { +@@ -988,7 +991,7 @@ build_trie(struct acl_build_context *context, struct rte_acl_build_rule *head, + */ + uint64_t mask; + mask = RTE_ACL_MASKLEN_TO_BITMASK( +- fld->mask_range.u32, ++ fld->mask_range.u64, + rule->config->defs[n].size); + + /* gen a mini-trie for this field */ +@@ -1301,6 +1304,9 @@ acl_build_index(const struct rte_acl_config *config, uint32_t *data_index) + if (last_header != config->defs[n].input_index) { + last_header = config->defs[n].input_index; + data_index[m++] = config->defs[n].offset; ++ if (config->defs[n].size > sizeof(uint32_t)) ++ data_index[m++] = config->defs[n].offset + ++ sizeof(uint32_t); + } + } + +@@ -1487,7 +1493,7 @@ acl_set_data_indexes(struct rte_acl_ctx *ctx) + memcpy(ctx->data_indexes + ofs, ctx->trie[i].data_index, + n * sizeof(ctx->data_indexes[0])); + ctx->trie[i].data_index = ctx->data_indexes + ofs; +- ofs += RTE_ACL_MAX_FIELDS; ++ ofs += ACL_MAX_INDEXES; + } + } + +@@ -1643,7 +1649,7 @@ rte_acl_build(struct rte_acl_ctx *ctx, const struct rte_acl_config *cfg) + /* allocate and fill run-time structures. */ + rc = rte_acl_gen(ctx, bcx.tries, bcx.bld_tries, + bcx.num_tries, bcx.cfg.num_categories, +- RTE_ACL_MAX_FIELDS * RTE_DIM(bcx.tries) * ++ ACL_MAX_INDEXES * RTE_DIM(bcx.tries) * + sizeof(ctx->data_indexes[0]), max_size); + if (rc == 0) { + /* set data indexes. */ +diff --git a/dpdk/lib/acl/acl_run_altivec.h b/dpdk/lib/acl/acl_run_altivec.h +index 2de6f27b1f..4dfe7a14b4 100644 +--- a/dpdk/lib/acl/acl_run_altivec.h ++++ b/dpdk/lib/acl/acl_run_altivec.h +@@ -41,7 +41,7 @@ resolve_priority_altivec(uint64_t transition, int n, + { + uint32_t x; + xmm_t results, priority, results1, priority1; +- vector bool int selector; ++ __vector bool int selector; + xmm_t *saved_results, *saved_priority; + + for (x = 0; x < categories; x += RTE_ACL_RESULTS_MULTIPLIER) { +@@ -110,8 +110,8 @@ transition4(xmm_t next_input, const uint64_t *trans, + xmm_t in, node_type, r, t; + xmm_t dfa_ofs, quad_ofs; + xmm_t *index_mask, *tp; +- vector bool int dfa_msk; +- vector signed char zeroes = {}; ++ __vector bool int dfa_msk; ++ __vector signed char zeroes = {}; + union { + uint64_t d64[2]; + uint32_t d32[4]; +@@ -127,7 +127,7 @@ transition4(xmm_t next_input, const uint64_t *trans, + index_mask = (xmm_t *)&altivec_acl_const.xmm_index_mask.u32; + t = vec_xor(*index_mask, *index_mask); + in = vec_perm(next_input, (xmm_t){}, +- *(vector unsigned char *)&altivec_acl_const.xmm_shuffle_input); ++ *(__vector unsigned char *)&altivec_acl_const.xmm_shuffle_input); + + /* Calc node type and node addr */ + node_type = vec_and(vec_nor(*index_mask, *index_mask), tr_lo); +@@ -137,30 +137,30 @@ transition4(xmm_t next_input, const uint64_t *trans, + dfa_msk = vec_cmpeq(node_type, t); + + /* DFA calculations. */ +- r = vec_sr(in, (vector unsigned int){30, 30, 30, 30}); ++ r = vec_sr(in, (__vector unsigned int){30, 30, 30, 30}); + tp = (xmm_t *)&altivec_acl_const.range_base.u32; + r = vec_add(r, *tp); +- t = vec_sr(in, (vector unsigned int){24, 24, 24, 24}); ++ t = vec_sr(in, (__vector unsigned int){24, 24, 24, 24}); + r = vec_perm(tr_hi, (xmm_t){(uint16_t)0 << 16}, +- (vector unsigned char)r); ++ (__vector unsigned char)r); + + dfa_ofs = vec_sub(t, r); + +- /* QUAD/SINGLE caluclations. */ +- t = (xmm_t)vec_cmpgt((vector signed char)in, (vector signed char)tr_hi); ++ /* QUAD/SINGLE calculations. */ ++ t = (xmm_t)vec_cmpgt((__vector signed char)in, (__vector signed char)tr_hi); + t = (xmm_t)vec_sel( + vec_sel( +- (vector signed char)vec_sub( +- zeroes, (vector signed char)t), +- (vector signed char)t, +- vec_cmpgt((vector signed char)t, zeroes)), ++ (__vector signed char)vec_sub( ++ zeroes, (__vector signed char)t), ++ (__vector signed char)t, ++ vec_cmpgt((__vector signed char)t, zeroes)), + zeroes, +- vec_cmpeq((vector signed char)t, zeroes)); ++ vec_cmpeq((__vector signed char)t, zeroes)); + +- t = (xmm_t)vec_msum((vector signed char)t, +- (vector unsigned char)t, (xmm_t){}); +- quad_ofs = (xmm_t)vec_msum((vector signed short)t, +- *(vector signed short *)&altivec_acl_const.xmm_ones_16.u16, ++ t = (xmm_t)vec_msum((__vector signed char)t, ++ (__vector unsigned char)t, (xmm_t){}); ++ quad_ofs = (xmm_t)vec_msum((__vector signed short)t, ++ *(__vector signed short *)&altivec_acl_const.xmm_ones_16.u16, + (xmm_t){}); + + /* blend DFA and QUAD/SINGLE. */ +@@ -177,7 +177,7 @@ transition4(xmm_t next_input, const uint64_t *trans, + *indices2 = (xmm_t){v.d32[0], v.d32[1], v.d32[2], v.d32[3]}; + + return vec_sr(next_input, +- (vector unsigned int){CHAR_BIT, CHAR_BIT, CHAR_BIT, CHAR_BIT}); ++ (__vector unsigned int){CHAR_BIT, CHAR_BIT, CHAR_BIT, CHAR_BIT}); + } + + /* +diff --git a/dpdk/lib/acl/acl_run_avx512.c b/dpdk/lib/acl/acl_run_avx512.c +index 78fbe34f7c..3b8795561b 100644 +--- a/dpdk/lib/acl/acl_run_avx512.c ++++ b/dpdk/lib/acl/acl_run_avx512.c +@@ -64,7 +64,7 @@ update_flow_mask(const struct acl_flow_avx512 *flow, uint32_t *fmsk, + } + + /* +- * Resolve matches for multiple categories (LE 8, use 128b instuctions/regs) ++ * Resolve matches for multiple categories (LE 8, use 128b instructions/regs) + */ + static inline void + resolve_mcle8_avx512x1(uint32_t result[], +diff --git a/dpdk/lib/acl/acl_run_avx512x16.h b/dpdk/lib/acl/acl_run_avx512x16.h +index 48bb6fed85..f87293eeb7 100644 +--- a/dpdk/lib/acl/acl_run_avx512x16.h ++++ b/dpdk/lib/acl/acl_run_avx512x16.h +@@ -10,7 +10,7 @@ + */ + + /* +- * This implementation uses 512-bit registers(zmm) and instrincts. ++ * This implementation uses 512-bit registers(zmm) and intrinsics. + * So our main SIMD type is 512-bit width and each such variable can + * process sizeof(__m512i) / sizeof(uint32_t) == 16 entries in parallel. + */ +@@ -25,20 +25,20 @@ + #define _F_(x) x##_avx512x16 + + /* +- * Same instrincts have different syntaxis (depending on the bit-width), ++ * Same intrinsics have different syntaxes (depending on the bit-width), + * so to overcome that few macros need to be defined. + */ + +-/* Naming convention for generic epi(packed integers) type instrincts. */ ++/* Naming convention for generic epi(packed integers) type intrinsics. */ + #define _M_I_(x) _mm512_##x + +-/* Naming convention for si(whole simd integer) type instrincts. */ ++/* Naming convention for si(whole simd integer) type intrinsics. */ + #define _M_SI_(x) _mm512_##x##_si512 + +-/* Naming convention for masked gather type instrincts. */ ++/* Naming convention for masked gather type intrinsics. */ + #define _M_MGI_(x) _mm512_##x + +-/* Naming convention for gather type instrincts. */ ++/* Naming convention for gather type intrinsics. */ + #define _M_GI_(name, idx, base, scale) _mm512_##name(idx, base, scale) + + /* num/mask of transitions per SIMD regs */ +@@ -239,7 +239,7 @@ _F_(gather_bytes)(__m512i zero, const __m512i p[2], const uint32_t m[2], + } + + /* +- * Resolve matches for multiple categories (GT 8, use 512b instuctions/regs) ++ * Resolve matches for multiple categories (GT 8, use 512b instructions/regs) + */ + static inline void + resolve_mcgt8_avx512x1(uint32_t result[], +diff --git a/dpdk/lib/acl/acl_run_avx512x8.h b/dpdk/lib/acl/acl_run_avx512x8.h +index 61ac9d1b47..5da2bbfdeb 100644 +--- a/dpdk/lib/acl/acl_run_avx512x8.h ++++ b/dpdk/lib/acl/acl_run_avx512x8.h +@@ -10,7 +10,7 @@ + */ + + /* +- * This implementation uses 256-bit registers(ymm) and instrincts. ++ * This implementation uses 256-bit registers(ymm) and intrinsics. + * So our main SIMD type is 256-bit width and each such variable can + * process sizeof(__m256i) / sizeof(uint32_t) == 8 entries in parallel. + */ +@@ -25,20 +25,20 @@ + #define _F_(x) x##_avx512x8 + + /* +- * Same instrincts have different syntaxis (depending on the bit-width), ++ * Same intrinsics have different syntaxes (depending on the bit-width), + * so to overcome that few macros need to be defined. + */ + +-/* Naming convention for generic epi(packed integers) type instrincts. */ ++/* Naming convention for generic epi(packed integers) type intrinsics. */ + #define _M_I_(x) _mm256_##x + +-/* Naming convention for si(whole simd integer) type instrincts. */ ++/* Naming convention for si(whole simd integer) type intrinsics. */ + #define _M_SI_(x) _mm256_##x##_si256 + +-/* Naming convention for masked gather type instrincts. */ ++/* Naming convention for masked gather type intrinsics. */ + #define _M_MGI_(x) _mm256_m##x + +-/* Naming convention for gather type instrincts. */ ++/* Naming convention for gather type intrinsics. */ + #define _M_GI_(name, idx, base, scale) _mm256_##name(base, idx, scale) + + /* num/mask of transitions per SIMD regs */ +diff --git a/dpdk/lib/acl/rte_acl_osdep.h b/dpdk/lib/acl/rte_acl_osdep.h +index b2c262dee7..3c1dc402ca 100644 +--- a/dpdk/lib/acl/rte_acl_osdep.h ++++ b/dpdk/lib/acl/rte_acl_osdep.h +@@ -5,6 +5,10 @@ + #ifndef _RTE_ACL_OSDEP_H_ + #define _RTE_ACL_OSDEP_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + /** + * @file + * +@@ -45,4 +49,8 @@ + #include + #include + ++#ifdef __cplusplus ++} ++#endif ++ + #endif /* _RTE_ACL_OSDEP_H_ */ +diff --git a/dpdk/lib/bpf/bpf_convert.c b/dpdk/lib/bpf/bpf_convert.c +index db84add7dc..9563274c9c 100644 +--- a/dpdk/lib/bpf/bpf_convert.c ++++ b/dpdk/lib/bpf/bpf_convert.c +@@ -412,7 +412,7 @@ static int bpf_convert_filter(const struct bpf_insn *prog, size_t len, + BPF_EMIT_JMP; + break; + +- /* ldxb 4 * ([14] & 0xf) is remaped into 6 insns. */ ++ /* ldxb 4 * ([14] & 0xf) is remapped into 6 insns. */ + case BPF_LDX | BPF_MSH | BPF_B: + /* tmp = A */ + *insn++ = BPF_MOV64_REG(BPF_REG_TMP, BPF_REG_A); +@@ -428,7 +428,7 @@ static int bpf_convert_filter(const struct bpf_insn *prog, size_t len, + *insn = BPF_MOV64_REG(BPF_REG_A, BPF_REG_TMP); + break; + +- /* RET_K is remaped into 2 insns. RET_A case doesn't need an ++ /* RET_K is remapped into 2 insns. RET_A case doesn't need an + * extra mov as EBPF_REG_0 is already mapped into BPF_REG_A. + */ + case BPF_RET | BPF_A: +diff --git a/dpdk/lib/bpf/bpf_def.h b/dpdk/lib/bpf/bpf_def.h +index fa9125307e..f08cd9106b 100644 +--- a/dpdk/lib/bpf/bpf_def.h ++++ b/dpdk/lib/bpf/bpf_def.h +@@ -7,6 +7,10 @@ + #ifndef _RTE_BPF_DEF_H_ + #define _RTE_BPF_DEF_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + /** + * @file + * +@@ -140,4 +144,8 @@ struct ebpf_insn { + */ + #define EBPF_FUNC_MAX_ARGS (EBPF_REG_6 - EBPF_REG_1) + ++#ifdef __cplusplus ++} ++#endif ++ + #endif /* RTE_BPF_DEF_H_ */ +diff --git a/dpdk/lib/bpf/bpf_impl.h b/dpdk/lib/bpf/bpf_impl.h +index 26d165ad5c..b4d8e87c6d 100644 +--- a/dpdk/lib/bpf/bpf_impl.h ++++ b/dpdk/lib/bpf/bpf_impl.h +@@ -2,8 +2,8 @@ + * Copyright(c) 2018 Intel Corporation + */ + +-#ifndef _BPF_H_ +-#define _BPF_H_ ++#ifndef BPF_IMPL_H ++#define BPF_IMPL_H + + #include + #include +@@ -43,4 +43,4 @@ bpf_size(uint32_t bpf_op_sz) + return 0; + } + +-#endif /* _BPF_H_ */ ++#endif /* BPF_IMPL_H */ +diff --git a/dpdk/lib/compressdev/rte_compressdev_internal.h b/dpdk/lib/compressdev/rte_compressdev_internal.h +index 22ceac66e2..b3b193e3ee 100644 +--- a/dpdk/lib/compressdev/rte_compressdev_internal.h ++++ b/dpdk/lib/compressdev/rte_compressdev_internal.h +@@ -5,6 +5,10 @@ + #ifndef _RTE_COMPRESSDEV_INTERNAL_H_ + #define _RTE_COMPRESSDEV_INTERNAL_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + /* rte_compressdev_internal.h + * This file holds Compressdev private data structures. + */ +@@ -18,7 +22,7 @@ + /* Logging Macros */ + extern int compressdev_logtype; + #define COMPRESSDEV_LOG(level, fmt, args...) \ +- rte_log(RTE_LOG_ ## level, compressdev_logtype, "%s(): "fmt "\n", \ ++ rte_log(RTE_LOG_ ## level, compressdev_logtype, "%s(): " fmt "\n", \ + __func__, ##args) + + /** +@@ -94,7 +98,7 @@ struct rte_compressdev { + struct rte_compressdev_data { + uint8_t dev_id; + /**< Compress device identifier */ +- uint8_t socket_id; ++ int socket_id; + /**< Socket identifier where memory is allocated */ + char name[RTE_COMPRESSDEV_NAME_MAX_LEN]; + /**< Unique identifier name */ +@@ -111,4 +115,9 @@ struct rte_compressdev_data { + void *dev_private; + /**< PMD-specific private data */ + } __rte_cache_aligned; ++ ++#ifdef __cplusplus ++} ++#endif ++ + #endif +diff --git a/dpdk/lib/cryptodev/cryptodev_pmd.h b/dpdk/lib/cryptodev/cryptodev_pmd.h +index b9146f652c..56e659b474 100644 +--- a/dpdk/lib/cryptodev/cryptodev_pmd.h ++++ b/dpdk/lib/cryptodev/cryptodev_pmd.h +@@ -5,6 +5,10 @@ + #ifndef _CRYPTODEV_PMD_H_ + #define _CRYPTODEV_PMD_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + /** @file + * RTE Crypto PMD APIs + * +@@ -640,4 +644,8 @@ set_asym_session_private_data(struct rte_cryptodev_asym_session *sess, + sess->sess_private_data[driver_id] = private_data; + } + ++#ifdef __cplusplus ++} ++#endif ++ + #endif /* _CRYPTODEV_PMD_H_ */ +diff --git a/dpdk/lib/cryptodev/rte_crypto.h b/dpdk/lib/cryptodev/rte_crypto.h +index a864f5036f..aeb3bf6e38 100644 +--- a/dpdk/lib/cryptodev/rte_crypto.h ++++ b/dpdk/lib/cryptodev/rte_crypto.h +@@ -123,15 +123,24 @@ struct rte_crypto_op { + rte_iova_t phys_addr; + /**< physical address of crypto operation */ + ++/* empty structures do not have zero size in C++ leading to compilation errors ++ * with clang about structure/union having different sizes in C and C++. ++ * While things are clearer with an explicit union, since each field is ++ * zero-sized it's not actually needed, so omit it for C++ ++ */ ++#ifndef __cplusplus + __extension__ + union { ++#endif + struct rte_crypto_sym_op sym[0]; + /**< Symmetric operation parameters */ + + struct rte_crypto_asym_op asym[0]; + /**< Asymmetric operation parameters */ + ++#ifndef __cplusplus + }; /**< operation specific parameters */ ++#endif + }; + + /** +diff --git a/dpdk/lib/cryptodev/rte_crypto_asym.h b/dpdk/lib/cryptodev/rte_crypto_asym.h +index 9c866f553f..9c5bb9233a 100644 +--- a/dpdk/lib/cryptodev/rte_crypto_asym.h ++++ b/dpdk/lib/cryptodev/rte_crypto_asym.h +@@ -146,7 +146,7 @@ enum rte_crypto_rsa_padding_type { + enum rte_crypto_rsa_priv_key_type { + RTE_RSA_KEY_TYPE_EXP, + /**< RSA private key is an exponent */ +- RTE_RSA_KET_TYPE_QT, ++ RTE_RSA_KEY_TYPE_QT, + /**< RSA private key is in quintuple format + * See rte_crypto_rsa_priv_key_qt + */ +diff --git a/dpdk/lib/distributor/rte_distributor_single.c b/dpdk/lib/distributor/rte_distributor_single.c +index b653620688..60ca86152f 100644 +--- a/dpdk/lib/distributor/rte_distributor_single.c ++++ b/dpdk/lib/distributor/rte_distributor_single.c +@@ -247,8 +247,7 @@ rte_distributor_process_single(struct rte_distributor_single *d, + * worker given by the bit-position + */ + for (i = 0; i < d->num_workers; i++) +- match |= (!(d->in_flight_tags[i] ^ new_tag) +- << i); ++ match |= ((uint64_t)!(d->in_flight_tags[i] ^ new_tag) << i); + + /* Only turned-on bits are considered as match */ + match &= d->in_flight_bitmask; +diff --git a/dpdk/lib/dmadev/rte_dmadev.h b/dpdk/lib/dmadev/rte_dmadev.h +index 9942c6ec21..4abe79c536 100644 +--- a/dpdk/lib/dmadev/rte_dmadev.h ++++ b/dpdk/lib/dmadev/rte_dmadev.h +@@ -533,7 +533,7 @@ struct rte_dma_port_param { + * @note If some fields can not be supported by the + * hardware/driver, then the driver ignores those fields. + * Please check driver-specific documentation for limitations +- * and capablites. ++ * and capabilities. + */ + __extension__ + struct { +@@ -731,7 +731,7 @@ enum rte_dma_status_code { + /** The operation completed successfully. */ + RTE_DMA_STATUS_SUCCESSFUL, + /** The operation failed to complete due abort by user. +- * This is mainly used when processing dev_stop, user could modidy the ++ * This is mainly used when processing dev_stop, user could modify the + * descriptors (e.g. change one bit to tell hardware abort this job), + * it allows outstanding requests to be complete as much as possible, + * so reduce the time to stop the device. +diff --git a/dpdk/lib/dmadev/rte_dmadev_pmd.h b/dpdk/lib/dmadev/rte_dmadev_pmd.h +index 5316ad5b5f..82ab7a8cc7 100644 +--- a/dpdk/lib/dmadev/rte_dmadev_pmd.h ++++ b/dpdk/lib/dmadev/rte_dmadev_pmd.h +@@ -14,6 +14,8 @@ + * by any application. + */ + ++#include ++ + #include "rte_dmadev.h" + + #ifdef __cplusplus +diff --git a/dpdk/lib/eal/arm/include/rte_cycles_32.h b/dpdk/lib/eal/arm/include/rte_cycles_32.h +index f79718ce8c..cec4d69e7a 100644 +--- a/dpdk/lib/eal/arm/include/rte_cycles_32.h ++++ b/dpdk/lib/eal/arm/include/rte_cycles_32.h +@@ -30,7 +30,7 @@ extern "C" { + + /** + * This call is easily portable to any architecture, however, +- * it may require a system call and inprecise for some tasks. ++ * it may require a system call and imprecise for some tasks. + */ + static inline uint64_t + __rte_rdtsc_syscall(void) +diff --git a/dpdk/lib/eal/common/eal_common_dev.c b/dpdk/lib/eal/common/eal_common_dev.c +index e1e9976d8d..07f285f862 100644 +--- a/dpdk/lib/eal/common/eal_common_dev.c ++++ b/dpdk/lib/eal/common/eal_common_dev.c +@@ -185,8 +185,10 @@ local_dev_probe(const char *devargs, struct rte_device **new_dev) + return ret; + + err_devarg: +- if (rte_devargs_remove(da) != 0) ++ if (rte_devargs_remove(da) != 0) { + rte_devargs_reset(da); ++ free(da); ++ } + return ret; + } + +diff --git a/dpdk/lib/eal/common/eal_common_devargs.c b/dpdk/lib/eal/common/eal_common_devargs.c +index 8c7650cf6c..184fe676aa 100644 +--- a/dpdk/lib/eal/common/eal_common_devargs.c ++++ b/dpdk/lib/eal/common/eal_common_devargs.c +@@ -191,6 +191,7 @@ rte_devargs_parse(struct rte_devargs *da, const char *dev) + + if (da == NULL) + return -EINVAL; ++ memset(da, 0, sizeof(*da)); + + /* First parse according global device syntax. */ + if (rte_devargs_layers_parse(da, dev) == 0) { +diff --git a/dpdk/lib/eal/common/eal_common_dynmem.c b/dpdk/lib/eal/common/eal_common_dynmem.c +index 7c5437ddfa..c1e1889f5c 100644 +--- a/dpdk/lib/eal/common/eal_common_dynmem.c ++++ b/dpdk/lib/eal/common/eal_common_dynmem.c +@@ -304,6 +304,10 @@ eal_dynmem_hugepage_init(void) + needed = num_pages - num_pages_alloc; + + pages = malloc(sizeof(*pages) * needed); ++ if (pages == NULL) { ++ RTE_LOG(ERR, EAL, "Failed to malloc pages\n"); ++ return -1; ++ } + + /* do not request exact number of pages */ + cur_pages = eal_memalloc_alloc_seg_bulk(pages, +diff --git a/dpdk/lib/eal/common/eal_common_proc.c b/dpdk/lib/eal/common/eal_common_proc.c +index ebd0f6673b..b33d58ea0a 100644 +--- a/dpdk/lib/eal/common/eal_common_proc.c ++++ b/dpdk/lib/eal/common/eal_common_proc.c +@@ -282,8 +282,17 @@ read_msg(struct mp_msg_internal *m, struct sockaddr_un *s) + msgh.msg_control = control; + msgh.msg_controllen = sizeof(control); + ++retry: + msglen = recvmsg(mp_fd, &msgh, 0); ++ ++ /* zero length message means socket was closed */ ++ if (msglen == 0) ++ return 0; ++ + if (msglen < 0) { ++ if (errno == EINTR) ++ goto retry; ++ + RTE_LOG(ERR, EAL, "recvmsg failed, %s\n", strerror(errno)); + return -1; + } +@@ -311,7 +320,7 @@ read_msg(struct mp_msg_internal *m, struct sockaddr_un *s) + RTE_LOG(ERR, EAL, "invalid received data length\n"); + return -1; + } +- return 0; ++ return msglen; + } + + static void +@@ -385,8 +394,13 @@ mp_handle(void *arg __rte_unused) + struct sockaddr_un sa; + + while (mp_fd >= 0) { +- if (read_msg(&msg, &sa) == 0) +- process_msg(&msg, &sa); ++ int ret; ++ ++ ret = read_msg(&msg, &sa); ++ if (ret <= 0) ++ break; ++ ++ process_msg(&msg, &sa); + } + + return NULL; +diff --git a/dpdk/lib/eal/common/eal_common_trace_utils.c b/dpdk/lib/eal/common/eal_common_trace_utils.c +index 64f58fb66a..2b55dbec65 100644 +--- a/dpdk/lib/eal/common/eal_common_trace_utils.c ++++ b/dpdk/lib/eal/common/eal_common_trace_utils.c +@@ -104,13 +104,15 @@ trace_session_name_generate(char *trace_dir) + rc = rte_strscpy(trace_dir, eal_get_hugefile_prefix(), + TRACE_PREFIX_LEN); + if (rc == -E2BIG) +- rc = TRACE_PREFIX_LEN; ++ rc = TRACE_PREFIX_LEN - 1; + trace_dir[rc++] = '-'; + + rc = strftime(trace_dir + rc, TRACE_DIR_STR_LEN - rc, + "%Y-%m-%d-%p-%I-%M-%S", tm_result); +- if (rc == 0) ++ if (rc == 0) { ++ errno = ENOSPC; + goto fail; ++ } + + return rc; + fail: +diff --git a/dpdk/lib/eal/common/malloc_elem.h b/dpdk/lib/eal/common/malloc_elem.h +index 15d8ba7af2..503fe5c470 100644 +--- a/dpdk/lib/eal/common/malloc_elem.h ++++ b/dpdk/lib/eal/common/malloc_elem.h +@@ -7,6 +7,8 @@ + + #include + ++#include ++ + #define MIN_DATA_SIZE (RTE_CACHE_LINE_SIZE) + + /* dummy definition of struct so we can use pointers to it in malloc_elem struct */ +@@ -125,12 +127,6 @@ malloc_elem_cookies_ok(const struct malloc_elem *elem) + #define ASAN_MEM_TO_SHADOW(mem) \ + RTE_PTR_ADD(ASAN_MEM_SHIFT(mem), ASAN_SHADOW_OFFSET) + +-#if defined(__clang__) +-#define __rte_no_asan __attribute__((no_sanitize("address", "hwaddress"))) +-#else +-#define __rte_no_asan __attribute__((no_sanitize_address)) +-#endif +- + __rte_no_asan + static inline void + asan_set_shadow(void *addr, char val) +@@ -270,7 +266,9 @@ old_malloc_size(struct malloc_elem *elem) + + #else /* !RTE_MALLOC_ASAN */ + +-#define __rte_no_asan ++static inline void ++asan_set_zone(void *ptr __rte_unused, size_t len __rte_unused, ++ uint32_t val __rte_unused) { } + + static inline void + asan_set_freezone(void *ptr __rte_unused, size_t size __rte_unused) { } +diff --git a/dpdk/lib/eal/common/malloc_heap.c b/dpdk/lib/eal/common/malloc_heap.c +index 55aad2711b..55063ccf81 100644 +--- a/dpdk/lib/eal/common/malloc_heap.c ++++ b/dpdk/lib/eal/common/malloc_heap.c +@@ -402,7 +402,7 @@ try_expand_heap_primary(struct malloc_heap *heap, uint64_t pg_sz, + bool callback_triggered = false; + + alloc_sz = RTE_ALIGN_CEIL(align + elt_size + +- MALLOC_ELEM_TRAILER_LEN, pg_sz); ++ MALLOC_ELEM_OVERHEAD, pg_sz); + n_segs = alloc_sz / pg_sz; + + /* we can't know in advance how many pages we'll need, so we malloc */ +@@ -860,6 +860,7 @@ malloc_heap_free(struct malloc_elem *elem) + struct rte_memseg_list *msl; + unsigned int i, n_segs, before_space, after_space; + int ret; ++ bool unmapped = false; + const struct internal_config *internal_conf = + eal_get_internal_configuration(); + +@@ -1026,6 +1027,9 @@ malloc_heap_free(struct malloc_elem *elem) + request_to_primary(&req); + } + ++ /* we didn't exit early, meaning we have unmapped some pages */ ++ unmapped = true; ++ + RTE_LOG(DEBUG, EAL, "Heap on socket %d was shrunk by %zdMB\n", + msl->socket_id, aligned_len >> 20ULL); + +@@ -1033,6 +1037,37 @@ malloc_heap_free(struct malloc_elem *elem) + free_unlock: + asan_set_freezone(asan_ptr, asan_data_len); + ++ /* if we unmapped some memory, we need to do additional work for ASan */ ++ if (unmapped) { ++ void *asan_end = RTE_PTR_ADD(asan_ptr, asan_data_len); ++ void *aligned_end = RTE_PTR_ADD(aligned_start, aligned_len); ++ void *aligned_trailer = RTE_PTR_SUB(aligned_start, ++ MALLOC_ELEM_TRAILER_LEN); ++ ++ /* ++ * There was a memory area that was unmapped. This memory area ++ * will have to be marked as available for ASan, because we will ++ * want to use it next time it gets mapped again. The OS memory ++ * protection should trigger a fault on access to these areas ++ * anyway, so we are not giving up any protection. ++ */ ++ asan_set_zone(aligned_start, aligned_len, 0x00); ++ ++ /* ++ * ...however, when we unmap pages, we create new free elements ++ * which might have been marked as "freed" with an earlier ++ * `asan_set_freezone` call. So, if there is an area past the ++ * unmapped space that was marked as freezone for ASan, we need ++ * to mark the malloc header as available. ++ */ ++ if (asan_end > aligned_end) ++ asan_set_zone(aligned_end, MALLOC_ELEM_HEADER_LEN, 0x00); ++ ++ /* if there's space before unmapped memory, mark as available */ ++ if (asan_ptr < aligned_start) ++ asan_set_zone(aligned_trailer, MALLOC_ELEM_TRAILER_LEN, 0x00); ++ } ++ + rte_spinlock_unlock(&(heap->lock)); + return ret; + } +diff --git a/dpdk/lib/eal/common/malloc_mp.c b/dpdk/lib/eal/common/malloc_mp.c +index 2e597a17a2..774bd1132f 100644 +--- a/dpdk/lib/eal/common/malloc_mp.c ++++ b/dpdk/lib/eal/common/malloc_mp.c +@@ -251,7 +251,7 @@ handle_alloc_request(const struct malloc_mp_req *m, + } + + alloc_sz = RTE_ALIGN_CEIL(ar->align + ar->elt_size + +- MALLOC_ELEM_TRAILER_LEN, ar->page_sz); ++ MALLOC_ELEM_OVERHEAD, ar->page_sz); + n_segs = alloc_sz / ar->page_sz; + + /* we can't know in advance how many pages we'll need, so we malloc */ +diff --git a/dpdk/lib/eal/common/rte_service.c b/dpdk/lib/eal/common/rte_service.c +index bd8fb72e78..e76c2baffc 100644 +--- a/dpdk/lib/eal/common/rte_service.c ++++ b/dpdk/lib/eal/common/rte_service.c +@@ -764,7 +764,9 @@ rte_service_lcore_stop(uint32_t lcore) + return -EALREADY; + + uint32_t i; +- uint64_t service_mask = lcore_states[lcore].service_mask; ++ struct core_state *cs = &lcore_states[lcore]; ++ uint64_t service_mask = cs->service_mask; ++ + for (i = 0; i < RTE_SERVICE_NUM_MAX; i++) { + int32_t enabled = service_mask & (UINT64_C(1) << i); + int32_t service_running = rte_service_runstate_get(i); +@@ -772,6 +774,11 @@ rte_service_lcore_stop(uint32_t lcore) + __atomic_load_n(&rte_services[i].num_mapped_cores, + __ATOMIC_RELAXED)); + ++ /* Switch off this core for all services, to ensure that future ++ * calls to may_be_active() know this core is switched off. ++ */ ++ cs->service_active_on_lcore[i] = 0; ++ + /* if the core is mapped, and the service is running, and this + * is the only core that is mapped, the service would cease to + * run if this core stopped, so fail instead. +diff --git a/dpdk/lib/eal/freebsd/eal.c b/dpdk/lib/eal/freebsd/eal.c +index a1cd2462db..414aad3dd3 100644 +--- a/dpdk/lib/eal/freebsd/eal.c ++++ b/dpdk/lib/eal/freebsd/eal.c +@@ -986,11 +986,11 @@ rte_eal_cleanup(void) + eal_get_internal_configuration(); + rte_service_finalize(); + rte_mp_channel_cleanup(); ++ rte_trace_save(); ++ eal_trace_fini(); + /* after this point, any DPDK pointers will become dangling */ + rte_eal_memory_detach(); + rte_eal_alarm_cleanup(); +- rte_trace_save(); +- eal_trace_fini(); + eal_cleanup_config(internal_conf); + return 0; + } +diff --git a/dpdk/lib/eal/freebsd/eal_interrupts.c b/dpdk/lib/eal/freebsd/eal_interrupts.c +index 10aa91cc09..9f720bdc8f 100644 +--- a/dpdk/lib/eal/freebsd/eal_interrupts.c ++++ b/dpdk/lib/eal/freebsd/eal_interrupts.c +@@ -234,7 +234,7 @@ rte_intr_callback_unregister_pending(const struct rte_intr_handle *intr_handle, + + rte_spinlock_lock(&intr_lock); + +- /* check if the insterrupt source for the fd is existent */ ++ /* check if the interrupt source for the fd is existent */ + TAILQ_FOREACH(src, &intr_sources, next) + if (rte_intr_fd_get(src->intr_handle) == rte_intr_fd_get(intr_handle)) + break; +@@ -288,7 +288,7 @@ rte_intr_callback_unregister(const struct rte_intr_handle *intr_handle, + + rte_spinlock_lock(&intr_lock); + +- /* check if the insterrupt source for the fd is existent */ ++ /* check if the interrupt source for the fd is existent */ + TAILQ_FOREACH(src, &intr_sources, next) + if (rte_intr_fd_get(src->intr_handle) == rte_intr_fd_get(intr_handle)) + break; +diff --git a/dpdk/lib/eal/freebsd/eal_memory.c b/dpdk/lib/eal/freebsd/eal_memory.c +index 78ac142b82..17ab10e0ca 100644 +--- a/dpdk/lib/eal/freebsd/eal_memory.c ++++ b/dpdk/lib/eal/freebsd/eal_memory.c +@@ -446,8 +446,8 @@ memseg_secondary_init(void) + + msl = &mcfg->memsegs[msl_idx]; + +- /* skip empty memseg lists */ +- if (msl->memseg_arr.len == 0) ++ /* skip empty and external memseg lists */ ++ if (msl->memseg_arr.len == 0 || msl->external) + continue; + + if (rte_fbarray_attach(&msl->memseg_arr)) { +diff --git a/dpdk/lib/eal/freebsd/include/rte_os.h b/dpdk/lib/eal/freebsd/include/rte_os.h +index 9d8a69008c..003468caff 100644 +--- a/dpdk/lib/eal/freebsd/include/rte_os.h ++++ b/dpdk/lib/eal/freebsd/include/rte_os.h +@@ -5,6 +5,10 @@ + #ifndef _RTE_OS_H_ + #define _RTE_OS_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + /** + * This header should contain any definition + * which is not supported natively or named differently in FreeBSD. +@@ -24,6 +28,8 @@ + + typedef cpuset_t rte_cpuset_t; + #define RTE_HAS_CPUSET ++ ++#ifdef RTE_EAL_FREEBSD_CPUSET_LEGACY + #define RTE_CPU_AND(dst, src1, src2) do \ + { \ + cpuset_t tmp; \ +@@ -57,6 +63,23 @@ typedef cpuset_t rte_cpuset_t; + CPU_ANDNOT(&tmp, src); \ + CPU_COPY(&tmp, dst); \ + } while (0) ++#endif /* CPU_NAND */ ++ ++#else /* RTE_EAL_FREEBSD_CPUSET_LEGACY */ ++ ++#define RTE_CPU_AND CPU_AND ++#define RTE_CPU_OR CPU_OR ++#define RTE_CPU_FILL CPU_FILL ++#define RTE_CPU_NOT(dst, src) do { \ ++ cpu_set_t tmp; \ ++ CPU_FILL(&tmp); \ ++ CPU_XOR(dst, src, &tmp); \ ++} while (0) ++ ++#endif /* RTE_EAL_FREEBSD_CPUSET_LEGACY */ ++ ++#ifdef __cplusplus ++} + #endif + + #endif /* _RTE_OS_H_ */ +diff --git a/dpdk/lib/eal/freebsd/meson.build b/dpdk/lib/eal/freebsd/meson.build +index 398ceab71d..fe9097303a 100644 +--- a/dpdk/lib/eal/freebsd/meson.build ++++ b/dpdk/lib/eal/freebsd/meson.build +@@ -19,3 +19,14 @@ sources += files( + ) + + deps += ['kvargs', 'telemetry'] ++ ++# test for version of cpuset macros ++cpuset_test_code = ''' ++ #include ++ #include ++ void cpu_test_or(cpuset_t *s) { CPU_OR(s, s, s); } ++''' ++ ++if not cc.compiles(cpuset_test_code, name: 'Detect argument count for CPU_OR') ++ dpdk_conf.set('RTE_EAL_FREEBSD_CPUSET_LEGACY', 1) ++endif +diff --git a/dpdk/lib/eal/include/generic/rte_pflock.h b/dpdk/lib/eal/include/generic/rte_pflock.h +index b9de063c89..e7bb29b3c5 100644 +--- a/dpdk/lib/eal/include/generic/rte_pflock.h ++++ b/dpdk/lib/eal/include/generic/rte_pflock.h +@@ -157,7 +157,7 @@ rte_pflock_write_lock(rte_pflock_t *pf) + uint16_t ticket, w; + + /* Acquire ownership of write-phase. +- * This is same as rte_tickelock_lock(). ++ * This is same as rte_ticketlock_lock(). + */ + ticket = __atomic_fetch_add(&pf->wr.in, 1, __ATOMIC_RELAXED); + rte_wait_until_equal_16(&pf->wr.out, ticket, __ATOMIC_ACQUIRE); +diff --git a/dpdk/lib/eal/include/generic/rte_ticketlock.h b/dpdk/lib/eal/include/generic/rte_ticketlock.h +index c1b8808f51..693c67b517 100644 +--- a/dpdk/lib/eal/include/generic/rte_ticketlock.h ++++ b/dpdk/lib/eal/include/generic/rte_ticketlock.h +@@ -91,13 +91,13 @@ rte_ticketlock_unlock(rte_ticketlock_t *tl) + static inline int + rte_ticketlock_trylock(rte_ticketlock_t *tl) + { +- rte_ticketlock_t old, new; +- old.tickets = __atomic_load_n(&tl->tickets, __ATOMIC_RELAXED); +- new.tickets = old.tickets; +- new.s.next++; +- if (old.s.next == old.s.current) { +- if (__atomic_compare_exchange_n(&tl->tickets, &old.tickets, +- new.tickets, 0, __ATOMIC_ACQUIRE, __ATOMIC_RELAXED)) ++ rte_ticketlock_t oldl, newl; ++ oldl.tickets = __atomic_load_n(&tl->tickets, __ATOMIC_RELAXED); ++ newl.tickets = oldl.tickets; ++ newl.s.next++; ++ if (oldl.s.next == oldl.s.current) { ++ if (__atomic_compare_exchange_n(&tl->tickets, &oldl.tickets, ++ newl.tickets, 0, __ATOMIC_ACQUIRE, __ATOMIC_RELAXED)) + return 1; + } + +diff --git a/dpdk/lib/eal/include/meson.build b/dpdk/lib/eal/include/meson.build +index 86468d1a2b..9700494816 100644 +--- a/dpdk/lib/eal/include/meson.build ++++ b/dpdk/lib/eal/include/meson.build +@@ -60,6 +60,7 @@ generic_headers = files( + 'generic/rte_mcslock.h', + 'generic/rte_memcpy.h', + 'generic/rte_pause.h', ++ 'generic/rte_pflock.h', + 'generic/rte_power_intrinsics.h', + 'generic/rte_prefetch.h', + 'generic/rte_rwlock.h', +diff --git a/dpdk/lib/eal/include/rte_bitops.h b/dpdk/lib/eal/include/rte_bitops.h +index 141e8ea730..f50dbe4388 100644 +--- a/dpdk/lib/eal/include/rte_bitops.h ++++ b/dpdk/lib/eal/include/rte_bitops.h +@@ -17,6 +17,10 @@ + #include + #include + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + /** + * Get the uint64_t value for a specified bit set. + * +@@ -271,4 +275,8 @@ rte_bit_relaxed_test_and_clear64(unsigned int nr, volatile uint64_t *addr) + return val & mask; + } + ++#ifdef __cplusplus ++} ++#endif ++ + #endif /* _RTE_BITOPS_H_ */ +diff --git a/dpdk/lib/eal/include/rte_branch_prediction.h b/dpdk/lib/eal/include/rte_branch_prediction.h +index 854ef9e5dd..0256a9de60 100644 +--- a/dpdk/lib/eal/include/rte_branch_prediction.h ++++ b/dpdk/lib/eal/include/rte_branch_prediction.h +@@ -10,6 +10,10 @@ + #ifndef _RTE_BRANCH_PREDICTION_H_ + #define _RTE_BRANCH_PREDICTION_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + /** + * Check if a branch is likely to be taken. + * +@@ -38,4 +42,8 @@ + #define unlikely(x) __builtin_expect(!!(x), 0) + #endif /* unlikely */ + ++#ifdef __cplusplus ++} ++#endif ++ + #endif /* _RTE_BRANCH_PREDICTION_H_ */ +diff --git a/dpdk/lib/eal/include/rte_common.h b/dpdk/lib/eal/include/rte_common.h +index 4a399cc7c8..6f004f6cb3 100644 +--- a/dpdk/lib/eal/include/rte_common.h ++++ b/dpdk/lib/eal/include/rte_common.h +@@ -85,6 +85,11 @@ typedef uint16_t unaligned_uint16_t; + */ + #define __rte_packed __attribute__((__packed__)) + ++/** ++ * Macro to mark a type that is not subject to type-based aliasing rules ++ */ ++#define __rte_may_alias __attribute__((__may_alias__)) ++ + /******* Macro to mark functions and fields scheduled for removal *****/ + #define __rte_deprecated __attribute__((__deprecated__)) + #define __rte_deprecated_msg(msg) __attribute__((__deprecated__(msg))) +@@ -242,6 +247,19 @@ static void __attribute__((destructor(RTE_PRIO(prio)), used)) func(void) + */ + #define __rte_cold __attribute__((cold)) + ++/** ++ * Disable AddressSanitizer on some code ++ */ ++#ifdef RTE_MALLOC_ASAN ++#ifdef RTE_CC_CLANG ++#define __rte_no_asan __attribute__((no_sanitize("address", "hwaddress"))) ++#else ++#define __rte_no_asan __attribute__((no_sanitize_address)) ++#endif ++#else /* ! RTE_MALLOC_ASAN */ ++#define __rte_no_asan ++#endif ++ + /*********** Macros for pointer arithmetic ********/ + + /** +diff --git a/dpdk/lib/eal/include/rte_compat.h b/dpdk/lib/eal/include/rte_compat.h +index 2718612cce..a7dbe23449 100644 +--- a/dpdk/lib/eal/include/rte_compat.h ++++ b/dpdk/lib/eal/include/rte_compat.h +@@ -6,6 +6,10 @@ + #ifndef _RTE_COMPAT_H_ + #define _RTE_COMPAT_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + #ifndef ALLOW_EXPERIMENTAL_API + + #define __rte_experimental \ +@@ -43,4 +47,8 @@ __attribute__((section(".text.internal"))) + + #endif + ++#ifdef __cplusplus ++} ++#endif ++ + #endif /* _RTE_COMPAT_H_ */ +diff --git a/dpdk/lib/eal/include/rte_dev.h b/dpdk/lib/eal/include/rte_dev.h +index 448a41cb0e..e6ff1218f9 100644 +--- a/dpdk/lib/eal/include/rte_dev.h ++++ b/dpdk/lib/eal/include/rte_dev.h +@@ -320,10 +320,6 @@ rte_dev_iterator_next(struct rte_dev_iterator *it); + dev != NULL; \ + dev = rte_dev_iterator_next(it)) + +-#ifdef __cplusplus +-} +-#endif +- + /** + * @warning + * @b EXPERIMENTAL: this API may change without prior notice +@@ -496,4 +492,8 @@ int + rte_dev_dma_unmap(struct rte_device *dev, void *addr, uint64_t iova, + size_t len); + ++#ifdef __cplusplus ++} ++#endif ++ + #endif /* _RTE_DEV_H_ */ +diff --git a/dpdk/lib/eal/include/rte_hypervisor.h b/dpdk/lib/eal/include/rte_hypervisor.h +index 5fe719c1d4..1666431ce3 100644 +--- a/dpdk/lib/eal/include/rte_hypervisor.h ++++ b/dpdk/lib/eal/include/rte_hypervisor.h +@@ -5,6 +5,10 @@ + #ifndef RTE_HYPERVISOR_H + #define RTE_HYPERVISOR_H + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + /** + * @file + * Hypervisor awareness. +@@ -30,4 +34,8 @@ rte_hypervisor_get(void); + const char * + rte_hypervisor_get_name(enum rte_hypervisor id); + ++#ifdef __cplusplus ++} ++#endif ++ + #endif /* RTE_HYPERVISOR_H */ +diff --git a/dpdk/lib/eal/include/rte_keepalive.h b/dpdk/lib/eal/include/rte_keepalive.h +index bd25508da8..538fb09095 100644 +--- a/dpdk/lib/eal/include/rte_keepalive.h ++++ b/dpdk/lib/eal/include/rte_keepalive.h +@@ -11,6 +11,10 @@ + #ifndef _KEEPALIVE_H_ + #define _KEEPALIVE_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + #include + #include + +@@ -139,4 +143,8 @@ rte_keepalive_register_relay_callback(struct rte_keepalive *keepcfg, + rte_keepalive_relay_callback_t callback, + void *data); + ++#ifdef __cplusplus ++} ++#endif ++ + #endif /* _KEEPALIVE_H_ */ +diff --git a/dpdk/lib/eal/include/rte_malloc.h b/dpdk/lib/eal/include/rte_malloc.h +index ed02e15119..3892519fab 100644 +--- a/dpdk/lib/eal/include/rte_malloc.h ++++ b/dpdk/lib/eal/include/rte_malloc.h +@@ -58,7 +58,7 @@ rte_malloc(const char *type, size_t size, unsigned align) + __rte_alloc_size(2); + + /** +- * Allocate zero'ed memory from the heap. ++ * Allocate zeroed memory from the heap. + * + * Equivalent to rte_malloc() except that the memory zone is + * initialised with zeros. In NUMA systems, the memory allocated resides on the +@@ -189,7 +189,7 @@ rte_malloc_socket(const char *type, size_t size, unsigned align, int socket) + __rte_alloc_size(2); + + /** +- * Allocate zero'ed memory from the heap. ++ * Allocate zeroed memory from the heap. + * + * Equivalent to rte_malloc() except that the memory zone is + * initialised with zeros. +diff --git a/dpdk/lib/eal/include/rte_pci_dev_feature_defs.h b/dpdk/lib/eal/include/rte_pci_dev_feature_defs.h +index e12c22081f..c5bb631286 100644 +--- a/dpdk/lib/eal/include/rte_pci_dev_feature_defs.h ++++ b/dpdk/lib/eal/include/rte_pci_dev_feature_defs.h +@@ -5,6 +5,10 @@ + #ifndef _RTE_PCI_DEV_DEFS_H_ + #define _RTE_PCI_DEV_DEFS_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + /* interrupt mode */ + enum rte_intr_mode { + RTE_INTR_MODE_NONE = 0, +@@ -13,4 +17,8 @@ enum rte_intr_mode { + RTE_INTR_MODE_MSIX + }; + ++#ifdef __cplusplus ++} ++#endif ++ + #endif /* _RTE_PCI_DEV_DEFS_H_ */ +diff --git a/dpdk/lib/eal/include/rte_pci_dev_features.h b/dpdk/lib/eal/include/rte_pci_dev_features.h +index 6104123d27..ee6e10590c 100644 +--- a/dpdk/lib/eal/include/rte_pci_dev_features.h ++++ b/dpdk/lib/eal/include/rte_pci_dev_features.h +@@ -5,6 +5,10 @@ + #ifndef _RTE_PCI_DEV_FEATURES_H + #define _RTE_PCI_DEV_FEATURES_H + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + #include + + #define RTE_INTR_MODE_NONE_NAME "none" +@@ -12,4 +16,8 @@ + #define RTE_INTR_MODE_MSI_NAME "msi" + #define RTE_INTR_MODE_MSIX_NAME "msix" + ++#ifdef __cplusplus ++} ++#endif ++ + #endif +diff --git a/dpdk/lib/eal/include/rte_time.h b/dpdk/lib/eal/include/rte_time.h +index 5ad7c8841a..ec25f7b93d 100644 +--- a/dpdk/lib/eal/include/rte_time.h ++++ b/dpdk/lib/eal/include/rte_time.h +@@ -5,6 +5,10 @@ + #ifndef _RTE_TIME_H_ + #define _RTE_TIME_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + #include + #include + +@@ -98,4 +102,8 @@ rte_ns_to_timespec(uint64_t nsec) + return ts; + } + ++#ifdef __cplusplus ++} ++#endif ++ + #endif /* _RTE_TIME_H_ */ +diff --git a/dpdk/lib/eal/include/rte_trace_point.h b/dpdk/lib/eal/include/rte_trace_point.h +index e226f073f7..0f8700974f 100644 +--- a/dpdk/lib/eal/include/rte_trace_point.h ++++ b/dpdk/lib/eal/include/rte_trace_point.h +@@ -370,7 +370,7 @@ do { \ + do { \ + if (unlikely(in == NULL)) \ + return; \ +- rte_strscpy(mem, in, __RTE_TRACE_EMIT_STRING_LEN_MAX); \ ++ rte_strscpy((char *)mem, in, __RTE_TRACE_EMIT_STRING_LEN_MAX); \ + mem = RTE_PTR_ADD(mem, __RTE_TRACE_EMIT_STRING_LEN_MAX); \ + } while (0) + +diff --git a/dpdk/lib/eal/include/rte_trace_point_register.h b/dpdk/lib/eal/include/rte_trace_point_register.h +index 4f5c86552d..2e61439940 100644 +--- a/dpdk/lib/eal/include/rte_trace_point_register.h ++++ b/dpdk/lib/eal/include/rte_trace_point_register.h +@@ -5,6 +5,10 @@ + #ifndef _RTE_TRACE_POINT_REGISTER_H_ + #define _RTE_TRACE_POINT_REGISTER_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + #ifdef _RTE_TRACE_POINT_H_ + #error for registration, include this file first before + #endif +@@ -42,4 +46,8 @@ do { \ + RTE_STR(in)"[32]", "string_bounded_t"); \ + } while (0) + ++#ifdef __cplusplus ++} ++#endif ++ + #endif /* _RTE_TRACE_POINT_REGISTER_H_ */ +diff --git a/dpdk/lib/eal/linux/eal.c b/dpdk/lib/eal/linux/eal.c +index 60b4924838..e3d34f7b7c 100644 +--- a/dpdk/lib/eal/linux/eal.c ++++ b/dpdk/lib/eal/linux/eal.c +@@ -1362,13 +1362,17 @@ rte_eal_cleanup(void) + + if (rte_eal_process_type() == RTE_PROC_PRIMARY) + rte_memseg_walk(mark_freeable, NULL); ++ + rte_service_finalize(); ++#ifdef VFIO_PRESENT ++ vfio_mp_sync_cleanup(); ++#endif + rte_mp_channel_cleanup(); ++ rte_trace_save(); ++ eal_trace_fini(); + /* after this point, any DPDK pointers will become dangling */ + rte_eal_memory_detach(); + rte_eal_alarm_cleanup(); +- rte_trace_save(); +- eal_trace_fini(); + eal_cleanup_config(internal_conf); + return 0; + } +diff --git a/dpdk/lib/eal/linux/eal_dev.c b/dpdk/lib/eal/linux/eal_dev.c +index bde55a3d92..52fe336572 100644 +--- a/dpdk/lib/eal/linux/eal_dev.c ++++ b/dpdk/lib/eal/linux/eal_dev.c +@@ -231,13 +231,13 @@ dev_uev_handler(__rte_unused void *param) + { + struct rte_dev_event uevent; + int ret; +- char buf[EAL_UEV_MSG_LEN]; ++ char buf[EAL_UEV_MSG_LEN + 1]; + struct rte_bus *bus; + struct rte_device *dev; + const char *busname = ""; + + memset(&uevent, 0, sizeof(struct rte_dev_event)); +- memset(buf, 0, EAL_UEV_MSG_LEN); ++ memset(buf, 0, EAL_UEV_MSG_LEN + 1); + + if (rte_intr_fd_get(intr_handle) < 0) + return; +@@ -384,6 +384,7 @@ rte_dev_event_monitor_stop(void) + close(rte_intr_fd_get(intr_handle)); + rte_intr_instance_free(intr_handle); + intr_handle = NULL; ++ ret = 0; + + monitor_refcount--; + +diff --git a/dpdk/lib/eal/linux/eal_interrupts.c b/dpdk/lib/eal/linux/eal_interrupts.c +index 6e3925efd4..70060bf3ef 100644 +--- a/dpdk/lib/eal/linux/eal_interrupts.c ++++ b/dpdk/lib/eal/linux/eal_interrupts.c +@@ -589,7 +589,7 @@ rte_intr_callback_unregister_pending(const struct rte_intr_handle *intr_handle, + + rte_spinlock_lock(&intr_lock); + +- /* check if the insterrupt source for the fd is existent */ ++ /* check if the interrupt source for the fd is existent */ + TAILQ_FOREACH(src, &intr_sources, next) { + if (rte_intr_fd_get(src->intr_handle) == rte_intr_fd_get(intr_handle)) + break; +@@ -639,7 +639,7 @@ rte_intr_callback_unregister(const struct rte_intr_handle *intr_handle, + + rte_spinlock_lock(&intr_lock); + +- /* check if the insterrupt source for the fd is existent */ ++ /* check if the interrupt source for the fd is existent */ + TAILQ_FOREACH(src, &intr_sources, next) + if (rte_intr_fd_get(src->intr_handle) == rte_intr_fd_get(intr_handle)) + break; +diff --git a/dpdk/lib/eal/linux/eal_memalloc.c b/dpdk/lib/eal/linux/eal_memalloc.c +index 337f2bc739..16b58d861b 100644 +--- a/dpdk/lib/eal/linux/eal_memalloc.c ++++ b/dpdk/lib/eal/linux/eal_memalloc.c +@@ -308,8 +308,8 @@ get_seg_fd(char *path, int buflen, struct hugepage_info *hi, + if (fd < 0) { + fd = open(path, O_CREAT | O_RDWR, 0600); + if (fd < 0) { +- RTE_LOG(ERR, EAL, "%s(): open failed: %s\n", +- __func__, strerror(errno)); ++ RTE_LOG(ERR, EAL, "%s(): open '%s' failed: %s\n", ++ __func__, path, strerror(errno)); + return -1; + } + /* take out a read lock and keep it indefinitely */ +@@ -346,8 +346,8 @@ get_seg_fd(char *path, int buflen, struct hugepage_info *hi, + + fd = open(path, O_CREAT | O_RDWR, 0600); + if (fd < 0) { +- RTE_LOG(DEBUG, EAL, "%s(): open failed: %s\n", +- __func__, strerror(errno)); ++ RTE_LOG(ERR, EAL, "%s(): open '%s' failed: %s\n", ++ __func__, path, strerror(errno)); + return -1; + } + /* take out a read lock */ +diff --git a/dpdk/lib/eal/linux/eal_memory.c b/dpdk/lib/eal/linux/eal_memory.c +index 03a4f2dd2d..fda6a159d5 100644 +--- a/dpdk/lib/eal/linux/eal_memory.c ++++ b/dpdk/lib/eal/linux/eal_memory.c +@@ -1883,8 +1883,8 @@ memseg_secondary_init(void) + + msl = &mcfg->memsegs[msl_idx]; + +- /* skip empty memseg lists */ +- if (msl->memseg_arr.len == 0) ++ /* skip empty and external memseg lists */ ++ if (msl->memseg_arr.len == 0 || msl->external) + continue; + + if (rte_fbarray_attach(&msl->memseg_arr)) { +diff --git a/dpdk/lib/eal/linux/eal_vfio.h b/dpdk/lib/eal/linux/eal_vfio.h +index 6ebaca6a0c..bba5c7afa5 100644 +--- a/dpdk/lib/eal/linux/eal_vfio.h ++++ b/dpdk/lib/eal/linux/eal_vfio.h +@@ -103,7 +103,7 @@ struct vfio_group { + typedef int (*vfio_dma_func_t)(int); + + /* Custom memory region DMA mapping function prototype. +- * Takes VFIO container fd, virtual address, phisical address, length and ++ * Takes VFIO container fd, virtual address, physical address, length and + * operation type (0 to unmap 1 for map) as a parameters. + * Returns 0 on success, -1 on error. + **/ +@@ -133,6 +133,7 @@ int + vfio_has_supported_extensions(int vfio_container_fd); + + int vfio_mp_sync_setup(void); ++void vfio_mp_sync_cleanup(void); + + #define EAL_VFIO_MP "eal_vfio_mp_sync" + +diff --git a/dpdk/lib/eal/linux/eal_vfio_mp_sync.c b/dpdk/lib/eal/linux/eal_vfio_mp_sync.c +index a2accfab3a..d12bbaee64 100644 +--- a/dpdk/lib/eal/linux/eal_vfio_mp_sync.c ++++ b/dpdk/lib/eal/linux/eal_vfio_mp_sync.c +@@ -120,4 +120,12 @@ vfio_mp_sync_setup(void) + return 0; + } + ++void ++vfio_mp_sync_cleanup(void) ++{ ++ if (rte_eal_process_type() != RTE_PROC_PRIMARY) ++ return; ++ ++ rte_mp_action_unregister(EAL_VFIO_MP); ++} + #endif +diff --git a/dpdk/lib/eal/linux/include/rte_os.h b/dpdk/lib/eal/linux/include/rte_os.h +index 35c07c70cb..c72bf5b7e6 100644 +--- a/dpdk/lib/eal/linux/include/rte_os.h ++++ b/dpdk/lib/eal/linux/include/rte_os.h +@@ -5,6 +5,10 @@ + #ifndef _RTE_OS_H_ + #define _RTE_OS_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + /** + * This header should contain any definition + * which is not supported natively or named differently in Linux. +@@ -42,4 +46,8 @@ typedef cpu_set_t rte_cpuset_t; + } while (0) + #endif + ++#ifdef __cplusplus ++} ++#endif ++ + #endif /* _RTE_OS_H_ */ +diff --git a/dpdk/lib/eal/ppc/include/rte_cycles.h b/dpdk/lib/eal/ppc/include/rte_cycles.h +index 5585f9273c..666fc9b0bf 100644 +--- a/dpdk/lib/eal/ppc/include/rte_cycles.h ++++ b/dpdk/lib/eal/ppc/include/rte_cycles.h +@@ -10,7 +10,10 @@ + extern "C" { + #endif + ++#include ++#ifdef __GLIBC__ + #include ++#endif + + #include "generic/rte_cycles.h" + +@@ -26,7 +29,11 @@ extern "C" { + static inline uint64_t + rte_rdtsc(void) + { ++#ifdef __GLIBC__ + return __ppc_get_timebase(); ++#else ++ return __builtin_ppc_get_timebase(); ++#endif + } + + static inline uint64_t +diff --git a/dpdk/lib/eal/ppc/include/rte_vect.h b/dpdk/lib/eal/ppc/include/rte_vect.h +index c1f0b0672c..a5f009b7df 100644 +--- a/dpdk/lib/eal/ppc/include/rte_vect.h ++++ b/dpdk/lib/eal/ppc/include/rte_vect.h +@@ -17,7 +17,7 @@ extern "C" { + + #define RTE_VECT_DEFAULT_SIMD_BITWIDTH RTE_VECT_SIMD_256 + +-typedef vector signed int xmm_t; ++typedef __vector signed int xmm_t; + + #define XMM_SIZE (sizeof(xmm_t)) + #define XMM_MASK (XMM_SIZE - 1) +diff --git a/dpdk/lib/eal/ppc/rte_cycles.c b/dpdk/lib/eal/ppc/rte_cycles.c +index 3180adb0ff..cd4bdff8b8 100644 +--- a/dpdk/lib/eal/ppc/rte_cycles.c ++++ b/dpdk/lib/eal/ppc/rte_cycles.c +@@ -2,12 +2,51 @@ + * Copyright (C) IBM Corporation 2019. + */ + ++#include ++#ifdef __GLIBC__ + #include ++#elif RTE_EXEC_ENV_LINUX ++#include ++#include ++#endif + + #include "eal_private.h" + + uint64_t + get_tsc_freq_arch(void) + { ++#ifdef __GLIBC__ + return __ppc_get_timebase_freq(); ++#elif RTE_EXEC_ENV_LINUX ++ static unsigned long base; ++ char buf[512]; ++ ssize_t nr; ++ FILE *f; ++ ++ if (base != 0) ++ goto out; ++ ++ f = fopen("/proc/cpuinfo", "rb"); ++ if (f == NULL) ++ goto out; ++ ++ while (fgets(buf, sizeof(buf), f) != NULL) { ++ char *ret = strstr(buf, "timebase"); ++ ++ if (ret == NULL) ++ continue; ++ ret += sizeof("timebase") - 1; ++ ret = strchr(ret, ':'); ++ if (ret == NULL) ++ continue; ++ base = strtoul(ret + 1, NULL, 10); ++ break; ++ } ++ fclose(f); ++out: ++ return (uint64_t) base; ++#else ++ return 0; ++#endif ++ + } +diff --git a/dpdk/lib/eal/windows/eal_memalloc.c b/dpdk/lib/eal/windows/eal_memalloc.c +index 55d6dcc71c..aa7589b81d 100644 +--- a/dpdk/lib/eal/windows/eal_memalloc.c ++++ b/dpdk/lib/eal/windows/eal_memalloc.c +@@ -17,7 +17,7 @@ eal_memalloc_get_seg_fd(int list_idx, int seg_idx) + RTE_SET_USED(list_idx); + RTE_SET_USED(seg_idx); + EAL_LOG_NOT_IMPLEMENTED(); +- return -1; ++ return -ENOTSUP; + } + + int +@@ -28,7 +28,7 @@ eal_memalloc_get_seg_fd_offset(int list_idx, int seg_idx, size_t *offset) + RTE_SET_USED(seg_idx); + RTE_SET_USED(offset); + EAL_LOG_NOT_IMPLEMENTED(); +- return -1; ++ return -ENOTSUP; + } + + static int +@@ -428,7 +428,7 @@ eal_memalloc_sync_with_primary(void) + { + /* No multi-process support. */ + EAL_LOG_NOT_IMPLEMENTED(); +- return -1; ++ return -ENOTSUP; + } + + int +diff --git a/dpdk/lib/eal/windows/eal_thread.c b/dpdk/lib/eal/windows/eal_thread.c +index 54fa93fa62..ff84cb42af 100644 +--- a/dpdk/lib/eal/windows/eal_thread.c ++++ b/dpdk/lib/eal/windows/eal_thread.c +@@ -150,13 +150,18 @@ eal_thread_create(pthread_t *thread) + + th = CreateThread(NULL, 0, + (LPTHREAD_START_ROUTINE)(ULONG_PTR)eal_thread_loop, +- NULL, 0, (LPDWORD)thread); ++ NULL, CREATE_SUSPENDED, (LPDWORD)thread); + if (!th) + return -1; + + SetPriorityClass(GetCurrentProcess(), NORMAL_PRIORITY_CLASS); + SetThreadPriority(th, THREAD_PRIORITY_NORMAL); + ++ if (ResumeThread(th) == (DWORD)-1) { ++ (void)CloseHandle(th); ++ return -1; ++ } ++ + return 0; + } + +diff --git a/dpdk/lib/eal/windows/eal_windows.h b/dpdk/lib/eal/windows/eal_windows.h +index 23ead6d30c..245aa60344 100644 +--- a/dpdk/lib/eal/windows/eal_windows.h ++++ b/dpdk/lib/eal/windows/eal_windows.h +@@ -63,7 +63,7 @@ unsigned int eal_socket_numa_node(unsigned int socket_id); + * @param arg + * Argument to the called function. + * @return +- * 0 on success, netagive error code on failure. ++ * 0 on success, negative error code on failure. + */ + int eal_intr_thread_schedule(void (*func)(void *arg), void *arg); + +diff --git a/dpdk/lib/eal/windows/include/dirent.h b/dpdk/lib/eal/windows/include/dirent.h +index 869a598378..34eb077f8c 100644 +--- a/dpdk/lib/eal/windows/include/dirent.h ++++ b/dpdk/lib/eal/windows/include/dirent.h +@@ -440,7 +440,7 @@ opendir(const char *dirname) + * display correctly on console. The problem can be fixed in two ways: + * (1) change the character set of console to 1252 using chcp utility + * and use Lucida Console font, or (2) use _cprintf function when +- * writing to console. The _cprinf() will re-encode ANSI strings to the ++ * writing to console. The _cprintf() will re-encode ANSI strings to the + * console code page so many non-ASCII characters will display correctly. + */ + static struct dirent* +@@ -579,7 +579,7 @@ dirent_mbstowcs_s( + wcstr[n] = 0; + } + +- /* Length of resuting multi-byte string WITH zero ++ /* Length of resulting multi-byte string WITH zero + *terminator + */ + if (pReturnValue) +diff --git a/dpdk/lib/eal/windows/include/fnmatch.h b/dpdk/lib/eal/windows/include/fnmatch.h +index c272f65ccd..c6b226bd5d 100644 +--- a/dpdk/lib/eal/windows/include/fnmatch.h ++++ b/dpdk/lib/eal/windows/include/fnmatch.h +@@ -26,14 +26,14 @@ extern "C" { + #define FNM_PREFIX_DIRS 0x20 + + /** +- * This function is used for searhing a given string source ++ * This function is used for searching a given string source + * with the given regular expression pattern. + * + * @param pattern + * regular expression notation describing the pattern to match + * + * @param string +- * source string to searcg for the pattern ++ * source string to search for the pattern + * + * @param flag + * containing information about the pattern +diff --git a/dpdk/lib/eal/windows/include/rte_windows.h b/dpdk/lib/eal/windows/include/rte_windows.h +index 0063b5d78c..83730c3d2e 100644 +--- a/dpdk/lib/eal/windows/include/rte_windows.h ++++ b/dpdk/lib/eal/windows/include/rte_windows.h +@@ -5,6 +5,10 @@ + #ifndef _RTE_WINDOWS_H_ + #define _RTE_WINDOWS_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + /** + * @file Windows-specific facilities + * +@@ -48,4 +52,8 @@ + RTE_FMT_HEAD(__VA_ARGS__,) "\n", GetLastError(), \ + RTE_FMT_TAIL(__VA_ARGS__,))) + ++#ifdef __cplusplus ++} ++#endif ++ + #endif /* _RTE_WINDOWS_H_ */ +diff --git a/dpdk/lib/eal/x86/include/rte_atomic.h b/dpdk/lib/eal/x86/include/rte_atomic.h +index 915afd9d27..f2ee1a9ce9 100644 +--- a/dpdk/lib/eal/x86/include/rte_atomic.h ++++ b/dpdk/lib/eal/x86/include/rte_atomic.h +@@ -60,7 +60,7 @@ extern "C" { + * Basic idea is to use lock prefixed add with some dummy memory location + * as the destination. From their experiments 128B(2 cache lines) below + * current stack pointer looks like a good candidate. +- * So below we use that techinque for rte_smp_mb() implementation. ++ * So below we use that technique for rte_smp_mb() implementation. + */ + + static __rte_always_inline void +diff --git a/dpdk/lib/eal/x86/include/rte_memcpy.h b/dpdk/lib/eal/x86/include/rte_memcpy.h +index 1b6c6e585f..b678b5c942 100644 +--- a/dpdk/lib/eal/x86/include/rte_memcpy.h ++++ b/dpdk/lib/eal/x86/include/rte_memcpy.h +@@ -45,6 +45,52 @@ extern "C" { + static __rte_always_inline void * + rte_memcpy(void *dst, const void *src, size_t n); + ++/** ++ * Copy bytes from one location to another, ++ * locations should not overlap. ++ * Use with n <= 15. ++ */ ++static __rte_always_inline void * ++rte_mov15_or_less(void *dst, const void *src, size_t n) ++{ ++ /** ++ * Use the following structs to avoid violating C standard ++ * alignment requirements and to avoid strict aliasing bugs ++ */ ++ struct rte_uint64_alias { ++ uint64_t val; ++ } __rte_packed __rte_may_alias; ++ struct rte_uint32_alias { ++ uint32_t val; ++ } __rte_packed __rte_may_alias; ++ struct rte_uint16_alias { ++ uint16_t val; ++ } __rte_packed __rte_may_alias; ++ ++ void *ret = dst; ++ if (n & 8) { ++ ((struct rte_uint64_alias *)dst)->val = ++ ((const struct rte_uint64_alias *)src)->val; ++ src = (const uint64_t *)src + 1; ++ dst = (uint64_t *)dst + 1; ++ } ++ if (n & 4) { ++ ((struct rte_uint32_alias *)dst)->val = ++ ((const struct rte_uint32_alias *)src)->val; ++ src = (const uint32_t *)src + 1; ++ dst = (uint32_t *)dst + 1; ++ } ++ if (n & 2) { ++ ((struct rte_uint16_alias *)dst)->val = ++ ((const struct rte_uint16_alias *)src)->val; ++ src = (const uint16_t *)src + 1; ++ dst = (uint16_t *)dst + 1; ++ } ++ if (n & 1) ++ *(uint8_t *)dst = *(const uint8_t *)src; ++ return ret; ++} ++ + #if defined __AVX512F__ && defined RTE_MEMCPY_AVX512 + + #define ALIGNMENT_MASK 0x3F +@@ -171,8 +217,6 @@ rte_mov512blocks(uint8_t *dst, const uint8_t *src, size_t n) + static __rte_always_inline void * + rte_memcpy_generic(void *dst, const void *src, size_t n) + { +- uintptr_t dstu = (uintptr_t)dst; +- uintptr_t srcu = (uintptr_t)src; + void *ret = dst; + size_t dstofss; + size_t bits; +@@ -181,24 +225,7 @@ rte_memcpy_generic(void *dst, const void *src, size_t n) + * Copy less than 16 bytes + */ + if (n < 16) { +- if (n & 0x01) { +- *(uint8_t *)dstu = *(const uint8_t *)srcu; +- srcu = (uintptr_t)((const uint8_t *)srcu + 1); +- dstu = (uintptr_t)((uint8_t *)dstu + 1); +- } +- if (n & 0x02) { +- *(uint16_t *)dstu = *(const uint16_t *)srcu; +- srcu = (uintptr_t)((const uint16_t *)srcu + 1); +- dstu = (uintptr_t)((uint16_t *)dstu + 1); +- } +- if (n & 0x04) { +- *(uint32_t *)dstu = *(const uint32_t *)srcu; +- srcu = (uintptr_t)((const uint32_t *)srcu + 1); +- dstu = (uintptr_t)((uint32_t *)dstu + 1); +- } +- if (n & 0x08) +- *(uint64_t *)dstu = *(const uint64_t *)srcu; +- return ret; ++ return rte_mov15_or_less(dst, src, n); + } + + /** +@@ -379,8 +406,6 @@ rte_mov128blocks(uint8_t *dst, const uint8_t *src, size_t n) + static __rte_always_inline void * + rte_memcpy_generic(void *dst, const void *src, size_t n) + { +- uintptr_t dstu = (uintptr_t)dst; +- uintptr_t srcu = (uintptr_t)src; + void *ret = dst; + size_t dstofss; + size_t bits; +@@ -389,25 +414,7 @@ rte_memcpy_generic(void *dst, const void *src, size_t n) + * Copy less than 16 bytes + */ + if (n < 16) { +- if (n & 0x01) { +- *(uint8_t *)dstu = *(const uint8_t *)srcu; +- srcu = (uintptr_t)((const uint8_t *)srcu + 1); +- dstu = (uintptr_t)((uint8_t *)dstu + 1); +- } +- if (n & 0x02) { +- *(uint16_t *)dstu = *(const uint16_t *)srcu; +- srcu = (uintptr_t)((const uint16_t *)srcu + 1); +- dstu = (uintptr_t)((uint16_t *)dstu + 1); +- } +- if (n & 0x04) { +- *(uint32_t *)dstu = *(const uint32_t *)srcu; +- srcu = (uintptr_t)((const uint32_t *)srcu + 1); +- dstu = (uintptr_t)((uint32_t *)dstu + 1); +- } +- if (n & 0x08) { +- *(uint64_t *)dstu = *(const uint64_t *)srcu; +- } +- return ret; ++ return rte_mov15_or_less(dst, src, n); + } + + /** +@@ -672,8 +679,6 @@ static __rte_always_inline void * + rte_memcpy_generic(void *dst, const void *src, size_t n) + { + __m128i xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8; +- uintptr_t dstu = (uintptr_t)dst; +- uintptr_t srcu = (uintptr_t)src; + void *ret = dst; + size_t dstofss; + size_t srcofs; +@@ -682,25 +687,7 @@ rte_memcpy_generic(void *dst, const void *src, size_t n) + * Copy less than 16 bytes + */ + if (n < 16) { +- if (n & 0x01) { +- *(uint8_t *)dstu = *(const uint8_t *)srcu; +- srcu = (uintptr_t)((const uint8_t *)srcu + 1); +- dstu = (uintptr_t)((uint8_t *)dstu + 1); +- } +- if (n & 0x02) { +- *(uint16_t *)dstu = *(const uint16_t *)srcu; +- srcu = (uintptr_t)((const uint16_t *)srcu + 1); +- dstu = (uintptr_t)((uint16_t *)dstu + 1); +- } +- if (n & 0x04) { +- *(uint32_t *)dstu = *(const uint32_t *)srcu; +- srcu = (uintptr_t)((const uint32_t *)srcu + 1); +- dstu = (uintptr_t)((uint32_t *)dstu + 1); +- } +- if (n & 0x08) { +- *(uint64_t *)dstu = *(const uint64_t *)srcu; +- } +- return ret; ++ return rte_mov15_or_less(dst, src, n); + } + + /** +@@ -818,27 +805,9 @@ rte_memcpy_aligned(void *dst, const void *src, size_t n) + { + void *ret = dst; + +- /* Copy size <= 16 bytes */ ++ /* Copy size < 16 bytes */ + if (n < 16) { +- if (n & 0x01) { +- *(uint8_t *)dst = *(const uint8_t *)src; +- src = (const uint8_t *)src + 1; +- dst = (uint8_t *)dst + 1; +- } +- if (n & 0x02) { +- *(uint16_t *)dst = *(const uint16_t *)src; +- src = (const uint16_t *)src + 1; +- dst = (uint16_t *)dst + 1; +- } +- if (n & 0x04) { +- *(uint32_t *)dst = *(const uint32_t *)src; +- src = (const uint32_t *)src + 1; +- dst = (uint32_t *)dst + 1; +- } +- if (n & 0x08) +- *(uint64_t *)dst = *(const uint64_t *)src; +- +- return ret; ++ return rte_mov15_or_less(dst, src, n); + } + + /* Copy 16 <= size <= 32 bytes */ +@@ -882,6 +851,8 @@ rte_memcpy(void *dst, const void *src, size_t n) + return rte_memcpy_generic(dst, src, n); + } + ++#undef ALIGNMENT_MASK ++ + #if defined(RTE_TOOLCHAIN_GCC) && (GCC_VERSION >= 100000) + #pragma GCC diagnostic pop + #endif +diff --git a/dpdk/lib/efd/rte_efd.c b/dpdk/lib/efd/rte_efd.c +index 86ef46863c..6c794d7750 100644 +--- a/dpdk/lib/efd/rte_efd.c ++++ b/dpdk/lib/efd/rte_efd.c +@@ -1165,7 +1165,7 @@ rte_efd_update(struct rte_efd_table * const table, const unsigned int socket_id, + { + uint32_t chunk_id = 0, group_id = 0, bin_id = 0; + uint8_t new_bin_choice = 0; +- struct efd_online_group_entry entry; ++ struct efd_online_group_entry entry = {{0}}; + + int status = efd_compute_update(table, socket_id, key, value, + &chunk_id, &group_id, &bin_id, +diff --git a/dpdk/lib/ethdev/ethdev_driver.h b/dpdk/lib/ethdev/ethdev_driver.h +index d95605a355..2822fd8c72 100644 +--- a/dpdk/lib/ethdev/ethdev_driver.h ++++ b/dpdk/lib/ethdev/ethdev_driver.h +@@ -5,6 +5,10 @@ + #ifndef _RTE_ETHDEV_DRIVER_H_ + #define _RTE_ETHDEV_DRIVER_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + /** + * @file + * +@@ -1629,6 +1633,24 @@ rte_eth_hairpin_queue_peer_bind(uint16_t cur_port, uint16_t cur_queue, + struct rte_hairpin_peer_info *peer_info, + uint32_t direction); + ++/** ++ * @internal ++ * Get rte_eth_dev from device name. The device name should be specified ++ * as below: ++ * - PCIe address (Domain:Bus:Device.Function), for example 0000:2:00.0 ++ * - SoC device name, for example fsl-gmac0 ++ * - vdev dpdk name, for example net_[pcap0|null0|tap0] ++ * ++ * @param name ++ * PCI address or name of the device ++ * @return ++ * - rte_eth_dev if successful ++ * - NULL on failure ++ */ ++__rte_internal ++struct rte_eth_dev* ++rte_eth_dev_get_by_name(const char *name); ++ + /** + * @internal + * Reset the current queue state and configuration to disconnect (unbind) it +@@ -1750,4 +1772,8 @@ struct rte_eth_tunnel_filter_conf { + uint16_t queue_id; /**< Queue assigned to if match */ + }; + ++#ifdef __cplusplus ++} ++#endif ++ + #endif /* _RTE_ETHDEV_DRIVER_H_ */ +diff --git a/dpdk/lib/ethdev/ethdev_pci.h b/dpdk/lib/ethdev/ethdev_pci.h +index 71aa4b2e98..0549842709 100644 +--- a/dpdk/lib/ethdev/ethdev_pci.h ++++ b/dpdk/lib/ethdev/ethdev_pci.h +@@ -6,6 +6,10 @@ + #ifndef _RTE_ETHDEV_PCI_H_ + #define _RTE_ETHDEV_PCI_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + #include + #include + #include +@@ -46,8 +50,9 @@ rte_eth_copy_pci_info(struct rte_eth_dev *eth_dev, + } + + static inline int +-eth_dev_pci_specific_init(struct rte_eth_dev *eth_dev, void *bus_device) { +- struct rte_pci_device *pci_dev = bus_device; ++eth_dev_pci_specific_init(struct rte_eth_dev *eth_dev, void *bus_device) ++{ ++ struct rte_pci_device *pci_dev = (struct rte_pci_device *)bus_device; + + if (!pci_dev) + return -ENODEV; +@@ -171,4 +176,8 @@ rte_eth_dev_pci_generic_remove(struct rte_pci_device *pci_dev, + return 0; + } + ++#ifdef __cplusplus ++} ++#endif ++ + #endif /* _RTE_ETHDEV_PCI_H_ */ +diff --git a/dpdk/lib/ethdev/ethdev_profile.h b/dpdk/lib/ethdev/ethdev_profile.h +index e5ee4df824..881aec1273 100644 +--- a/dpdk/lib/ethdev/ethdev_profile.h ++++ b/dpdk/lib/ethdev/ethdev_profile.h +@@ -6,6 +6,7 @@ + #define _RTE_ETHDEV_PROFILE_H_ + + #include "rte_ethdev.h" ++#include "ethdev_driver.h" + + /** + * Initialization of the Ethernet device profiling. +diff --git a/dpdk/lib/ethdev/ethdev_vdev.h b/dpdk/lib/ethdev/ethdev_vdev.h +index 2b49e9665b..f5f536ce64 100644 +--- a/dpdk/lib/ethdev/ethdev_vdev.h ++++ b/dpdk/lib/ethdev/ethdev_vdev.h +@@ -6,6 +6,10 @@ + #ifndef _RTE_ETHDEV_VDEV_H_ + #define _RTE_ETHDEV_VDEV_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + #include + #include + #include +@@ -52,4 +56,8 @@ rte_eth_vdev_allocate(struct rte_vdev_device *dev, size_t private_data_size) + return eth_dev; + } + ++#ifdef __cplusplus ++} ++#endif ++ + #endif /* _RTE_ETHDEV_VDEV_H_ */ +diff --git a/dpdk/lib/ethdev/rte_dev_info.h b/dpdk/lib/ethdev/rte_dev_info.h +index cb2fe0ae97..67cf0ae526 100644 +--- a/dpdk/lib/ethdev/rte_dev_info.h ++++ b/dpdk/lib/ethdev/rte_dev_info.h +@@ -5,6 +5,10 @@ + #ifndef _RTE_DEV_INFO_H_ + #define _RTE_DEV_INFO_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + #include + + /* +@@ -48,4 +52,8 @@ struct rte_eth_dev_module_info { + #define RTE_ETH_MODULE_SFF_8436_LEN 256 + #define RTE_ETH_MODULE_SFF_8436_MAX_LEN 640 + ++#ifdef __cplusplus ++} ++#endif ++ + #endif /* _RTE_DEV_INFO_H_ */ +diff --git a/dpdk/lib/ethdev/rte_ethdev.c b/dpdk/lib/ethdev/rte_ethdev.c +index a1d475a292..62e67f006d 100644 +--- a/dpdk/lib/ethdev/rte_ethdev.c ++++ b/dpdk/lib/ethdev/rte_ethdev.c +@@ -894,6 +894,17 @@ rte_eth_dev_get_port_by_name(const char *name, uint16_t *port_id) + return -ENODEV; + } + ++struct rte_eth_dev * ++rte_eth_dev_get_by_name(const char *name) ++{ ++ uint16_t pid; ++ ++ if (rte_eth_dev_get_port_by_name(name, &pid)) ++ return NULL; ++ ++ return &rte_eth_devices[pid]; ++} ++ + static int + eth_err(uint16_t port_id, int ret) + { +@@ -1879,8 +1890,9 @@ rte_eth_dev_stop(uint16_t port_id) + /* point fast-path functions to dummy ones */ + eth_dev_fp_ops_reset(rte_eth_fp_ops + port_id); + +- dev->data->dev_started = 0; + ret = (*dev->dev_ops->dev_stop)(dev); ++ if (ret == 0) ++ dev->data->dev_started = 0; + rte_ethdev_trace_stop(port_id, ret); + + return ret; +@@ -1920,7 +1932,13 @@ rte_eth_dev_close(uint16_t port_id) + RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -ENODEV); + dev = &rte_eth_devices[port_id]; + +- if (dev->data->dev_started) { ++ /* ++ * Secondary process needs to close device to release process private ++ * resources. But secondary process should not be obliged to wait ++ * for device stop before closing ethdev. ++ */ ++ if (rte_eal_process_type() == RTE_PROC_PRIMARY && ++ dev->data->dev_started) { + RTE_ETHDEV_LOG(ERR, "Cannot close started device (port %u)\n", + port_id); + return -EINVAL; +@@ -4070,6 +4088,7 @@ rte_eth_dev_rss_reta_update(uint16_t port_id, + struct rte_eth_rss_reta_entry64 *reta_conf, + uint16_t reta_size) + { ++ enum rte_eth_rx_mq_mode mq_mode; + struct rte_eth_dev *dev; + int ret; + +@@ -4101,6 +4120,12 @@ rte_eth_dev_rss_reta_update(uint16_t port_id, + if (ret < 0) + return ret; + ++ mq_mode = dev->data->dev_conf.rxmode.mq_mode; ++ if (!(mq_mode & RTE_ETH_MQ_RX_RSS_FLAG)) { ++ RTE_ETHDEV_LOG(ERR, "Multi-queue RSS mode isn't enabled.\n"); ++ return -ENOTSUP; ++ } ++ + RTE_FUNC_PTR_OR_ERR_RET(*dev->dev_ops->reta_update, -ENOTSUP); + return eth_err(port_id, (*dev->dev_ops->reta_update)(dev, reta_conf, + reta_size)); +@@ -4140,6 +4165,7 @@ rte_eth_dev_rss_hash_update(uint16_t port_id, + { + struct rte_eth_dev *dev; + struct rte_eth_dev_info dev_info = { .flow_type_rss_offloads = 0, }; ++ enum rte_eth_rx_mq_mode mq_mode; + int ret; + + RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -ENODEV); +@@ -4165,6 +4191,13 @@ rte_eth_dev_rss_hash_update(uint16_t port_id, + dev_info.flow_type_rss_offloads); + return -EINVAL; + } ++ ++ mq_mode = dev->data->dev_conf.rxmode.mq_mode; ++ if (!(mq_mode & RTE_ETH_MQ_RX_RSS_FLAG)) { ++ RTE_ETHDEV_LOG(ERR, "Multi-queue RSS mode isn't enabled.\n"); ++ return -ENOTSUP; ++ } ++ + RTE_FUNC_PTR_OR_ERR_RET(*dev->dev_ops->rss_hash_update, -ENOTSUP); + return eth_err(port_id, (*dev->dev_ops->rss_hash_update)(dev, + rss_conf)); +@@ -6156,6 +6189,8 @@ eth_dev_add_port_queue_stats(struct rte_tel_data *d, uint64_t *q_stats, + { + int q; + struct rte_tel_data *q_data = rte_tel_data_alloc(); ++ if (q_data == NULL) ++ return; + rte_tel_data_start_array(q_data, RTE_TEL_U64_VAL); + for (q = 0; q < RTE_ETHDEV_QUEUE_STAT_CNTRS; q++) + rte_tel_data_add_array_u64(q_data, q_stats[q]); +@@ -6249,6 +6284,7 @@ eth_dev_handle_port_xstats(const char *cmd __rte_unused, + for (i = 0; i < num_xstats; i++) + rte_tel_data_add_dict_u64(d, xstat_names[i].name, + eth_xstats[i].value); ++ free(eth_xstats); + return 0; + } + +@@ -6295,7 +6331,7 @@ eth_dev_handle_port_info(const char *cmd __rte_unused, + struct rte_tel_data *d) + { + struct rte_tel_data *rxq_state, *txq_state; +- char mac_addr[RTE_ETHER_ADDR_LEN]; ++ char mac_addr[RTE_ETHER_ADDR_FMT_SIZE]; + struct rte_eth_dev *eth_dev; + char *end_param; + int port_id, i; +@@ -6312,16 +6348,16 @@ eth_dev_handle_port_info(const char *cmd __rte_unused, + return -EINVAL; + + eth_dev = &rte_eth_devices[port_id]; +- if (!eth_dev) +- return -EINVAL; + + rxq_state = rte_tel_data_alloc(); + if (!rxq_state) + return -ENOMEM; + + txq_state = rte_tel_data_alloc(); +- if (!txq_state) ++ if (!txq_state) { ++ rte_tel_data_free(rxq_state); + return -ENOMEM; ++ } + + rte_tel_data_start_dict(d); + rte_tel_data_add_dict_string(d, "name", eth_dev->data->name); +@@ -6336,13 +6372,8 @@ eth_dev_handle_port_info(const char *cmd __rte_unused, + eth_dev->data->min_rx_buf_size); + rte_tel_data_add_dict_int(d, "rx_mbuf_alloc_fail", + eth_dev->data->rx_mbuf_alloc_failed); +- snprintf(mac_addr, RTE_ETHER_ADDR_LEN, "%02x:%02x:%02x:%02x:%02x:%02x", +- eth_dev->data->mac_addrs->addr_bytes[0], +- eth_dev->data->mac_addrs->addr_bytes[1], +- eth_dev->data->mac_addrs->addr_bytes[2], +- eth_dev->data->mac_addrs->addr_bytes[3], +- eth_dev->data->mac_addrs->addr_bytes[4], +- eth_dev->data->mac_addrs->addr_bytes[5]); ++ rte_ether_format_addr(mac_addr, sizeof(mac_addr), ++ eth_dev->data->mac_addrs); + rte_tel_data_add_dict_string(d, "mac_addr", mac_addr); + rte_tel_data_add_dict_int(d, "promiscuous", + eth_dev->data->promiscuous); +diff --git a/dpdk/lib/ethdev/rte_ethdev.h b/dpdk/lib/ethdev/rte_ethdev.h +index fa299c8ad7..0be04c5809 100644 +--- a/dpdk/lib/ethdev/rte_ethdev.h ++++ b/dpdk/lib/ethdev/rte_ethdev.h +@@ -74,7 +74,7 @@ + * rte_eth_rx_queue_setup()), it must call rte_eth_dev_stop() first to stop the + * device and then do the reconfiguration before calling rte_eth_dev_start() + * again. The transmit and receive functions should not be invoked when the +- * device is stopped. ++ * device or the queue is stopped. + * + * Please note that some configuration is not stored between calls to + * rte_eth_dev_stop()/rte_eth_dev_start(). The following configuration will +@@ -1787,7 +1787,7 @@ enum rte_eth_representor_type { + * device, etc... + */ + struct rte_eth_dev_info { +- struct rte_device *device; /** Generic device information */ ++ struct rte_device *device; /**< Generic device information */ + const char *driver_name; /**< Device Driver name. */ + unsigned int if_index; /**< Index to bound host interface, or 0 if none. + Use if_indextoname() to translate into an interface name. */ +@@ -1801,8 +1801,8 @@ struct rte_eth_dev_info { + uint16_t max_rx_queues; /**< Maximum number of Rx queues. */ + uint16_t max_tx_queues; /**< Maximum number of Tx queues. */ + uint32_t max_mac_addrs; /**< Maximum number of MAC addresses. */ +- uint32_t max_hash_mac_addrs; + /** Maximum number of hash MAC addresses for MTA and UTA. */ ++ uint32_t max_hash_mac_addrs; + uint16_t max_vfs; /**< Maximum number of VFs. */ + uint16_t max_vmdq_pools; /**< Maximum number of VMDq pools. */ + struct rte_eth_rxseg_capa rx_seg_capa; /**< Segmentation capability.*/ +diff --git a/dpdk/lib/ethdev/version.map b/dpdk/lib/ethdev/version.map +index c2fb0669a4..1f7359c846 100644 +--- a/dpdk/lib/ethdev/version.map ++++ b/dpdk/lib/ethdev/version.map +@@ -267,6 +267,7 @@ INTERNAL { + rte_eth_dev_callback_process; + rte_eth_dev_create; + rte_eth_dev_destroy; ++ rte_eth_dev_get_by_name; + rte_eth_dev_is_rx_hairpin_queue; + rte_eth_dev_is_tx_hairpin_queue; + rte_eth_dev_probing_finish; +diff --git a/dpdk/lib/eventdev/eventdev_pmd.h b/dpdk/lib/eventdev/eventdev_pmd.h +index d0b0c00a60..ce469d47a6 100644 +--- a/dpdk/lib/eventdev/eventdev_pmd.h ++++ b/dpdk/lib/eventdev/eventdev_pmd.h +@@ -5,6 +5,10 @@ + #ifndef _RTE_EVENTDEV_PMD_H_ + #define _RTE_EVENTDEV_PMD_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + /** @file + * RTE Event PMD APIs + * +diff --git a/dpdk/lib/eventdev/eventdev_pmd_pci.h b/dpdk/lib/eventdev/eventdev_pmd_pci.h +index 499852db16..24b56faaa9 100644 +--- a/dpdk/lib/eventdev/eventdev_pmd_pci.h ++++ b/dpdk/lib/eventdev/eventdev_pmd_pci.h +@@ -5,6 +5,10 @@ + #ifndef _RTE_EVENTDEV_PMD_PCI_H_ + #define _RTE_EVENTDEV_PMD_PCI_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + /** @file + * RTE Eventdev PCI PMD APIs + * +@@ -150,4 +154,8 @@ rte_event_pmd_pci_remove(struct rte_pci_device *pci_dev, + return 0; + } + ++#ifdef __cplusplus ++} ++#endif ++ + #endif /* _RTE_EVENTDEV_PMD_PCI_H_ */ +diff --git a/dpdk/lib/eventdev/eventdev_pmd_vdev.h b/dpdk/lib/eventdev/eventdev_pmd_vdev.h +index d9ee7277dd..77904910a2 100644 +--- a/dpdk/lib/eventdev/eventdev_pmd_vdev.h ++++ b/dpdk/lib/eventdev/eventdev_pmd_vdev.h +@@ -5,6 +5,10 @@ + #ifndef _RTE_EVENTDEV_PMD_VDEV_H_ + #define _RTE_EVENTDEV_PMD_VDEV_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + /** @file + * RTE Eventdev VDEV PMD APIs + * +@@ -99,4 +103,8 @@ rte_event_pmd_vdev_uninit(const char *name) + return 0; + } + ++#ifdef __cplusplus ++} ++#endif ++ + #endif /* _RTE_EVENTDEV_PMD_VDEV_H_ */ +diff --git a/dpdk/lib/eventdev/rte_event_eth_rx_adapter.c b/dpdk/lib/eventdev/rte_event_eth_rx_adapter.c +index 809416d9b7..2356e2a535 100644 +--- a/dpdk/lib/eventdev/rte_event_eth_rx_adapter.c ++++ b/dpdk/lib/eventdev/rte_event_eth_rx_adapter.c +@@ -293,6 +293,30 @@ rxa_event_buf_get(struct event_eth_rx_adapter *rx_adapter, uint16_t eth_dev_id, + } \ + } while (0) + ++#define RTE_EVENT_ETH_RX_ADAPTER_ID_VALID_OR_GOTO_ERR_RET(id, retval) do { \ ++ if (!rxa_validate_id(id)) { \ ++ RTE_EDEV_LOG_ERR("Invalid eth Rx adapter id = %d\n", id); \ ++ ret = retval; \ ++ goto error; \ ++ } \ ++} while (0) ++ ++#define RTE_EVENT_ETH_RX_ADAPTER_TOKEN_VALID_OR_GOTO_ERR_RET(token, retval) do { \ ++ if ((token) == NULL || strlen(token) == 0 || !isdigit(*token)) { \ ++ RTE_EDEV_LOG_ERR("Invalid eth Rx adapter token\n"); \ ++ ret = retval; \ ++ goto error; \ ++ } \ ++} while (0) ++ ++#define RTE_ETH_VALID_PORTID_OR_GOTO_ERR_RET(port_id, retval) do { \ ++ if (!rte_eth_dev_is_valid_port(port_id)) { \ ++ RTE_ETHDEV_LOG(ERR, "Invalid port_id=%u\n", port_id); \ ++ ret = retval; \ ++ goto error; \ ++ } \ ++} while (0) ++ + static inline int + rxa_sw_adapter_queue_count(struct event_eth_rx_adapter *rx_adapter) + { +@@ -2984,15 +3008,17 @@ rte_event_eth_rx_adapter_queue_stats_get(uint8_t id, + return -EINVAL; + } + +- queue_info = &dev_info->rx_queue[rx_queue_id]; +- event_buf = queue_info->event_buf; +- q_stats = queue_info->stats; ++ if (dev_info->internal_event_port == 0) { ++ queue_info = &dev_info->rx_queue[rx_queue_id]; ++ event_buf = queue_info->event_buf; ++ q_stats = queue_info->stats; + +- stats->rx_event_buf_count = event_buf->count; +- stats->rx_event_buf_size = event_buf->events_size; +- stats->rx_packets = q_stats->rx_packets; +- stats->rx_poll_count = q_stats->rx_poll_count; +- stats->rx_dropped = q_stats->rx_dropped; ++ stats->rx_event_buf_count = event_buf->count; ++ stats->rx_event_buf_size = event_buf->events_size; ++ stats->rx_packets = q_stats->rx_packets; ++ stats->rx_poll_count = q_stats->rx_poll_count; ++ stats->rx_dropped = q_stats->rx_dropped; ++ } + + dev = &rte_eventdevs[rx_adapter->eventdev_id]; + if (dev->dev_ops->eth_rx_adapter_queue_stats_get != NULL) { +@@ -3086,8 +3112,10 @@ rte_event_eth_rx_adapter_queue_stats_reset(uint8_t id, + return -EINVAL; + } + +- queue_info = &dev_info->rx_queue[rx_queue_id]; +- rxa_queue_stats_reset(queue_info); ++ if (dev_info->internal_event_port == 0) { ++ queue_info = &dev_info->rx_queue[rx_queue_id]; ++ rxa_queue_stats_reset(queue_info); ++ } + + dev = &rte_eventdevs[rx_adapter->eventdev_id]; + if (dev->dev_ops->eth_rx_adapter_queue_stats_reset != NULL) { +@@ -3170,11 +3198,11 @@ rte_event_eth_rx_adapter_queue_conf_get(uint8_t id, + uint16_t rx_queue_id, + struct rte_event_eth_rx_adapter_queue_conf *queue_conf) + { ++#define TICK2NSEC(_ticks, _freq) (((_ticks) * (1E9)) / (_freq)) + struct rte_eventdev *dev; + struct event_eth_rx_adapter *rx_adapter; + struct eth_device_info *dev_info; + struct eth_rx_queue_info *queue_info; +- struct rte_event *qi_ev; + int ret; + + if (rxa_memzone_lookup()) +@@ -3205,7 +3233,6 @@ rte_event_eth_rx_adapter_queue_conf_get(uint8_t id, + } + + queue_info = &dev_info->rx_queue[rx_queue_id]; +- qi_ev = (struct rte_event *)&queue_info->event; + + memset(queue_conf, 0, sizeof(*queue_conf)); + queue_conf->rx_queue_flags = 0; +@@ -3214,7 +3241,18 @@ rte_event_eth_rx_adapter_queue_conf_get(uint8_t id, + RTE_EVENT_ETH_RX_ADAPTER_QUEUE_FLOW_ID_VALID; + queue_conf->servicing_weight = queue_info->wt; + +- memcpy(&queue_conf->ev, qi_ev, sizeof(*qi_ev)); ++ queue_conf->ev.event = queue_info->event; ++ ++ queue_conf->vector_sz = queue_info->vector_data.max_vector_count; ++ queue_conf->vector_mp = queue_info->vector_data.vector_pool; ++ /* need to be converted from ticks to ns */ ++ queue_conf->vector_timeout_ns = TICK2NSEC( ++ queue_info->vector_data.vector_timeout_ticks, rte_get_timer_hz()); ++ ++ if (queue_info->event_buf != NULL) ++ queue_conf->event_buf_size = queue_info->event_buf->events_size; ++ else ++ queue_conf->event_buf_size = 0; + + dev = &rte_eventdevs[rx_adapter->eventdev_id]; + if (dev->dev_ops->eth_rx_adapter_queue_conf_get != NULL) { +@@ -3276,7 +3314,7 @@ handle_rxa_stats_reset(const char *cmd __rte_unused, + { + uint8_t rx_adapter_id; + +- if (params == NULL || strlen(params) == 0 || ~isdigit(*params)) ++ if (params == NULL || strlen(params) == 0 || !isdigit(*params)) + return -1; + + /* Get Rx adapter ID from parameter string */ +@@ -3299,7 +3337,7 @@ handle_rxa_get_queue_conf(const char *cmd __rte_unused, + { + uint8_t rx_adapter_id; + uint16_t rx_queue_id; +- int eth_dev_id; ++ int eth_dev_id, ret = -1; + char *token, *l_params; + struct rte_event_eth_rx_adapter_queue_conf queue_conf; + +@@ -3308,33 +3346,37 @@ handle_rxa_get_queue_conf(const char *cmd __rte_unused, + + /* Get Rx adapter ID from parameter string */ + l_params = strdup(params); ++ if (l_params == NULL) ++ return -ENOMEM; + token = strtok(l_params, ","); ++ RTE_EVENT_ETH_RX_ADAPTER_TOKEN_VALID_OR_GOTO_ERR_RET(token, -1); + rx_adapter_id = strtoul(token, NULL, 10); +- RTE_EVENT_ETH_RX_ADAPTER_ID_VALID_OR_ERR_RET(rx_adapter_id, -EINVAL); ++ RTE_EVENT_ETH_RX_ADAPTER_ID_VALID_OR_GOTO_ERR_RET(rx_adapter_id, -EINVAL); + + token = strtok(NULL, ","); +- if (token == NULL || strlen(token) == 0 || !isdigit(*token)) +- return -1; ++ RTE_EVENT_ETH_RX_ADAPTER_TOKEN_VALID_OR_GOTO_ERR_RET(token, -1); + + /* Get device ID from parameter string */ + eth_dev_id = strtoul(token, NULL, 10); +- RTE_EVENTDEV_VALID_DEVID_OR_ERR_RET(eth_dev_id, -EINVAL); ++ RTE_ETH_VALID_PORTID_OR_GOTO_ERR_RET(eth_dev_id, -EINVAL); + + token = strtok(NULL, ","); +- if (token == NULL || strlen(token) == 0 || !isdigit(*token)) +- return -1; ++ RTE_EVENT_ETH_RX_ADAPTER_TOKEN_VALID_OR_GOTO_ERR_RET(token, -1); + + /* Get Rx queue ID from parameter string */ + rx_queue_id = strtoul(token, NULL, 10); + if (rx_queue_id >= rte_eth_devices[eth_dev_id].data->nb_rx_queues) { + RTE_EDEV_LOG_ERR("Invalid rx queue_id %u", rx_queue_id); +- return -EINVAL; ++ ret = -EINVAL; ++ goto error; + } + + token = strtok(NULL, "\0"); + if (token != NULL) + RTE_EDEV_LOG_ERR("Extra parameters passed to eventdev" +- " telemetry command, igrnoring"); ++ " telemetry command, ignoring"); ++ /* Parsing parameter finished */ ++ free(l_params); + + if (rte_event_eth_rx_adapter_queue_conf_get(rx_adapter_id, eth_dev_id, + rx_queue_id, &queue_conf)) { +@@ -3354,6 +3396,10 @@ handle_rxa_get_queue_conf(const char *cmd __rte_unused, + RXA_ADD_DICT(queue_conf.ev, flow_id); + + return 0; ++ ++error: ++ free(l_params); ++ return ret; + } + + static int +@@ -3363,7 +3409,7 @@ handle_rxa_get_queue_stats(const char *cmd __rte_unused, + { + uint8_t rx_adapter_id; + uint16_t rx_queue_id; +- int eth_dev_id; ++ int eth_dev_id, ret = -1; + char *token, *l_params; + struct rte_event_eth_rx_adapter_queue_stats q_stats; + +@@ -3372,33 +3418,37 @@ handle_rxa_get_queue_stats(const char *cmd __rte_unused, + + /* Get Rx adapter ID from parameter string */ + l_params = strdup(params); ++ if (l_params == NULL) ++ return -ENOMEM; + token = strtok(l_params, ","); ++ RTE_EVENT_ETH_RX_ADAPTER_TOKEN_VALID_OR_GOTO_ERR_RET(token, -1); + rx_adapter_id = strtoul(token, NULL, 10); +- RTE_EVENT_ETH_RX_ADAPTER_ID_VALID_OR_ERR_RET(rx_adapter_id, -EINVAL); ++ RTE_EVENT_ETH_RX_ADAPTER_ID_VALID_OR_GOTO_ERR_RET(rx_adapter_id, -EINVAL); + + token = strtok(NULL, ","); +- if (token == NULL || strlen(token) == 0 || !isdigit(*token)) +- return -1; ++ RTE_EVENT_ETH_RX_ADAPTER_TOKEN_VALID_OR_GOTO_ERR_RET(token, -1); + + /* Get device ID from parameter string */ + eth_dev_id = strtoul(token, NULL, 10); +- RTE_EVENTDEV_VALID_DEVID_OR_ERR_RET(eth_dev_id, -EINVAL); ++ RTE_ETH_VALID_PORTID_OR_GOTO_ERR_RET(eth_dev_id, -EINVAL); + + token = strtok(NULL, ","); +- if (token == NULL || strlen(token) == 0 || !isdigit(*token)) +- return -1; ++ RTE_EVENT_ETH_RX_ADAPTER_TOKEN_VALID_OR_GOTO_ERR_RET(token, -1); + + /* Get Rx queue ID from parameter string */ + rx_queue_id = strtoul(token, NULL, 10); + if (rx_queue_id >= rte_eth_devices[eth_dev_id].data->nb_rx_queues) { + RTE_EDEV_LOG_ERR("Invalid rx queue_id %u", rx_queue_id); +- return -EINVAL; ++ ret = -EINVAL; ++ goto error; + } + + token = strtok(NULL, "\0"); + if (token != NULL) + RTE_EDEV_LOG_ERR("Extra parameters passed to eventdev" +- " telemetry command, igrnoring"); ++ " telemetry command, ignoring"); ++ /* Parsing parameter finished */ ++ free(l_params); + + if (rte_event_eth_rx_adapter_queue_stats_get(rx_adapter_id, eth_dev_id, + rx_queue_id, &q_stats)) { +@@ -3417,6 +3467,10 @@ handle_rxa_get_queue_stats(const char *cmd __rte_unused, + RXA_ADD_DICT(q_stats, rx_dropped); + + return 0; ++ ++error: ++ free(l_params); ++ return ret; + } + + static int +@@ -3426,7 +3480,7 @@ handle_rxa_queue_stats_reset(const char *cmd __rte_unused, + { + uint8_t rx_adapter_id; + uint16_t rx_queue_id; +- int eth_dev_id; ++ int eth_dev_id, ret = -1; + char *token, *l_params; + + if (params == NULL || strlen(params) == 0 || !isdigit(*params)) +@@ -3434,33 +3488,37 @@ handle_rxa_queue_stats_reset(const char *cmd __rte_unused, + + /* Get Rx adapter ID from parameter string */ + l_params = strdup(params); ++ if (l_params == NULL) ++ return -ENOMEM; + token = strtok(l_params, ","); ++ RTE_EVENT_ETH_RX_ADAPTER_TOKEN_VALID_OR_GOTO_ERR_RET(token, -1); + rx_adapter_id = strtoul(token, NULL, 10); +- RTE_EVENT_ETH_RX_ADAPTER_ID_VALID_OR_ERR_RET(rx_adapter_id, -EINVAL); ++ RTE_EVENT_ETH_RX_ADAPTER_ID_VALID_OR_GOTO_ERR_RET(rx_adapter_id, -EINVAL); + + token = strtok(NULL, ","); +- if (token == NULL || strlen(token) == 0 || !isdigit(*token)) +- return -1; ++ RTE_EVENT_ETH_RX_ADAPTER_TOKEN_VALID_OR_GOTO_ERR_RET(token, -1); + + /* Get device ID from parameter string */ + eth_dev_id = strtoul(token, NULL, 10); +- RTE_EVENTDEV_VALID_DEVID_OR_ERR_RET(eth_dev_id, -EINVAL); ++ RTE_ETH_VALID_PORTID_OR_GOTO_ERR_RET(eth_dev_id, -EINVAL); + + token = strtok(NULL, ","); +- if (token == NULL || strlen(token) == 0 || !isdigit(*token)) +- return -1; ++ RTE_EVENT_ETH_RX_ADAPTER_TOKEN_VALID_OR_GOTO_ERR_RET(token, -1); + + /* Get Rx queue ID from parameter string */ + rx_queue_id = strtoul(token, NULL, 10); + if (rx_queue_id >= rte_eth_devices[eth_dev_id].data->nb_rx_queues) { + RTE_EDEV_LOG_ERR("Invalid rx queue_id %u", rx_queue_id); +- return -EINVAL; ++ ret = -EINVAL; ++ goto error; + } + + token = strtok(NULL, "\0"); + if (token != NULL) + RTE_EDEV_LOG_ERR("Extra parameters passed to eventdev" +- " telemetry command, igrnoring"); ++ " telemetry command, ignoring"); ++ /* Parsing parameter finished */ ++ free(l_params); + + if (rte_event_eth_rx_adapter_queue_stats_reset(rx_adapter_id, + eth_dev_id, +@@ -3470,6 +3528,10 @@ handle_rxa_queue_stats_reset(const char *cmd __rte_unused, + } + + return 0; ++ ++error: ++ free(l_params); ++ return ret; + } + + RTE_INIT(rxa_init_telemetry) +diff --git a/dpdk/lib/eventdev/rte_event_eth_tx_adapter.c b/dpdk/lib/eventdev/rte_event_eth_tx_adapter.c +index c17f33f098..b4b37f1cae 100644 +--- a/dpdk/lib/eventdev/rte_event_eth_tx_adapter.c ++++ b/dpdk/lib/eventdev/rte_event_eth_tx_adapter.c +@@ -224,7 +224,7 @@ txa_service_data_init(void) + if (txa_service_data_array == NULL) { + txa_service_data_array = + txa_memzone_array_get("txa_service_data_array", +- sizeof(int), ++ sizeof(*txa_service_data_array), + RTE_EVENT_ETH_TX_ADAPTER_MAX_INSTANCE); + if (txa_service_data_array == NULL) + return -ENOMEM; +@@ -806,10 +806,8 @@ txa_service_queue_add(uint8_t id, + + rte_spinlock_lock(&txa->tx_lock); + +- if (txa_service_is_queue_added(txa, eth_dev, tx_queue_id)) { +- rte_spinlock_unlock(&txa->tx_lock); +- return 0; +- } ++ if (txa_service_is_queue_added(txa, eth_dev, tx_queue_id)) ++ goto ret_unlock; + + ret = txa_service_queue_array_alloc(txa, eth_dev->data->port_id); + if (ret) +@@ -821,6 +819,8 @@ txa_service_queue_add(uint8_t id, + + tdi = &txa->txa_ethdev[eth_dev->data->port_id]; + tqi = txa_service_queue(txa, eth_dev->data->port_id, tx_queue_id); ++ if (tqi == NULL) ++ goto err_unlock; + + txa_retry = &tqi->txa_retry; + txa_retry->id = txa->id; +@@ -836,6 +836,10 @@ txa_service_queue_add(uint8_t id, + tdi->nb_queues++; + txa->nb_queues++; + ++ret_unlock: ++ rte_spinlock_unlock(&txa->tx_lock); ++ return 0; ++ + err_unlock: + if (txa->nb_queues == 0) { + txa_service_queue_array_free(txa, +@@ -844,7 +848,7 @@ txa_service_queue_add(uint8_t id, + } + + rte_spinlock_unlock(&txa->tx_lock); +- return 0; ++ return -1; + } + + static int +@@ -887,9 +891,10 @@ txa_service_queue_del(uint8_t id, + + txa = txa_service_id_to_data(id); + ++ rte_spinlock_lock(&txa->tx_lock); + tqi = txa_service_queue(txa, port_id, tx_queue_id); + if (tqi == NULL || !tqi->added) +- return 0; ++ goto ret_unlock; + + tb = tqi->tx_buf; + tqi->added = 0; +@@ -899,6 +904,9 @@ txa_service_queue_del(uint8_t id, + txa->txa_ethdev[port_id].nb_queues--; + + txa_service_queue_array_free(txa, port_id); ++ ++ret_unlock: ++ rte_spinlock_unlock(&txa->tx_lock); + return 0; + } + +diff --git a/dpdk/lib/eventdev/rte_event_ring.h b/dpdk/lib/eventdev/rte_event_ring.h +index c0861b0ec2..0b9aefb000 100644 +--- a/dpdk/lib/eventdev/rte_event_ring.h ++++ b/dpdk/lib/eventdev/rte_event_ring.h +@@ -14,6 +14,10 @@ + #ifndef _RTE_EVENT_RING_ + #define _RTE_EVENT_RING_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + #include + + #include +@@ -266,4 +270,9 @@ rte_event_ring_get_capacity(const struct rte_event_ring *r) + { + return rte_ring_get_capacity(&r->r); + } ++ ++#ifdef __cplusplus ++} ++#endif ++ + #endif +diff --git a/dpdk/lib/eventdev/rte_event_timer_adapter.h b/dpdk/lib/eventdev/rte_event_timer_adapter.h +index 1551741820..e68d02da72 100644 +--- a/dpdk/lib/eventdev/rte_event_timer_adapter.h ++++ b/dpdk/lib/eventdev/rte_event_timer_adapter.h +@@ -678,4 +678,8 @@ rte_event_timer_cancel_burst(const struct rte_event_timer_adapter *adapter, + return adapter->cancel_burst(adapter, evtims, nb_evtims); + } + ++#ifdef __cplusplus ++} ++#endif ++ + #endif /* __RTE_EVENT_TIMER_ADAPTER_H__ */ +diff --git a/dpdk/lib/eventdev/rte_eventdev.h b/dpdk/lib/eventdev/rte_eventdev.h +index eef47d8acc..476bcbcc21 100644 +--- a/dpdk/lib/eventdev/rte_eventdev.h ++++ b/dpdk/lib/eventdev/rte_eventdev.h +@@ -986,21 +986,31 @@ struct rte_event_vector { + }; + /**< Union to hold common attributes of the vector array. */ + uint64_t impl_opaque; ++ ++/* empty structures do not have zero size in C++ leading to compilation errors ++ * with clang about structure having different sizes in C and C++. ++ * Since these are all zero-sized arrays, we can omit the "union" wrapper for ++ * C++ builds, removing the warning. ++ */ ++#ifndef __cplusplus + /**< Implementation specific opaque value. + * An implementation may use this field to hold implementation specific + * value to share between dequeue and enqueue operation. + * The application should not modify this field. + */ + union { ++#endif + struct rte_mbuf *mbufs[0]; + void *ptrs[0]; + uint64_t *u64s[0]; ++#ifndef __cplusplus + } __rte_aligned(16); ++#endif + /**< Start of the vector array union. Depending upon the event type the + * vector array can be an array of mbufs or pointers or opaque u64 + * values. + */ +-}; ++} __rte_aligned(16); + + /* Scheduler type definitions */ + #define RTE_SCHED_TYPE_ORDERED 0 +@@ -1805,7 +1815,7 @@ __rte_event_enqueue_burst(uint8_t dev_id, uint8_t port_id, + return 0; + } + #endif +- rte_eventdev_trace_enq_burst(dev_id, port_id, ev, nb_events, fn); ++ rte_eventdev_trace_enq_burst(dev_id, port_id, ev, nb_events, (void *)fn); + /* + * Allow zero cost non burst mode routine invocation if application + * requests nb_events as const one +diff --git a/dpdk/lib/fib/rte_fib.c b/dpdk/lib/fib/rte_fib.c +index 6ca180d7e7..0cced97a77 100644 +--- a/dpdk/lib/fib/rte_fib.c ++++ b/dpdk/lib/fib/rte_fib.c +@@ -40,10 +40,10 @@ EAL_REGISTER_TAILQ(rte_fib_tailq) + struct rte_fib { + char name[RTE_FIB_NAMESIZE]; + enum rte_fib_type type; /**< Type of FIB struct */ +- struct rte_rib *rib; /**< RIB helper datastruct */ ++ struct rte_rib *rib; /**< RIB helper datastructure */ + void *dp; /**< pointer to the dataplane struct*/ +- rte_fib_lookup_fn_t lookup; /**< fib lookup function */ +- rte_fib_modify_fn_t modify; /**< modify fib datastruct */ ++ rte_fib_lookup_fn_t lookup; /**< FIB lookup function */ ++ rte_fib_modify_fn_t modify; /**< modify FIB datastructure */ + uint64_t def_nh; + }; + +diff --git a/dpdk/lib/fib/rte_fib.h b/dpdk/lib/fib/rte_fib.h +index b3c59dfaaa..e592d3251a 100644 +--- a/dpdk/lib/fib/rte_fib.h ++++ b/dpdk/lib/fib/rte_fib.h +@@ -189,7 +189,7 @@ rte_fib_lookup_bulk(struct rte_fib *fib, uint32_t *ips, + * FIB object handle + * @return + * Pointer on the dataplane struct on success +- * NULL othervise ++ * NULL otherwise + */ + void * + rte_fib_get_dp(struct rte_fib *fib); +@@ -201,7 +201,7 @@ rte_fib_get_dp(struct rte_fib *fib); + * FIB object handle + * @return + * Pointer on the RIB on success +- * NULL othervise ++ * NULL otherwise + */ + struct rte_rib * + rte_fib_get_rib(struct rte_fib *fib); +diff --git a/dpdk/lib/fib/rte_fib6.c b/dpdk/lib/fib/rte_fib6.c +index be79efe004..eebee297d6 100644 +--- a/dpdk/lib/fib/rte_fib6.c ++++ b/dpdk/lib/fib/rte_fib6.c +@@ -40,10 +40,10 @@ EAL_REGISTER_TAILQ(rte_fib6_tailq) + struct rte_fib6 { + char name[FIB6_NAMESIZE]; + enum rte_fib6_type type; /**< Type of FIB struct */ +- struct rte_rib6 *rib; /**< RIB helper datastruct */ ++ struct rte_rib6 *rib; /**< RIB helper datastructure */ + void *dp; /**< pointer to the dataplane struct*/ +- rte_fib6_lookup_fn_t lookup; /**< fib lookup function */ +- rte_fib6_modify_fn_t modify; /**< modify fib datastruct */ ++ rte_fib6_lookup_fn_t lookup; /**< FIB lookup function */ ++ rte_fib6_modify_fn_t modify; /**< modify FIB datastructure */ + uint64_t def_nh; + }; + +diff --git a/dpdk/lib/fib/rte_fib6.h b/dpdk/lib/fib/rte_fib6.h +index 95879af96d..cb133719e1 100644 +--- a/dpdk/lib/fib/rte_fib6.h ++++ b/dpdk/lib/fib/rte_fib6.h +@@ -184,7 +184,7 @@ rte_fib6_lookup_bulk(struct rte_fib6 *fib, + * FIB6 object handle + * @return + * Pointer on the dataplane struct on success +- * NULL othervise ++ * NULL otherwise + */ + void * + rte_fib6_get_dp(struct rte_fib6 *fib); +@@ -196,7 +196,7 @@ rte_fib6_get_dp(struct rte_fib6 *fib); + * FIB object handle + * @return + * Pointer on the RIB6 on success +- * NULL othervise ++ * NULL otherwise + */ + struct rte_rib6 * + rte_fib6_get_rib(struct rte_fib6 *fib); +diff --git a/dpdk/lib/gpudev/version.map b/dpdk/lib/gpudev/version.map +index 2e414c65cc..34186ab7f1 100644 +--- a/dpdk/lib/gpudev/version.map ++++ b/dpdk/lib/gpudev/version.map +@@ -35,4 +35,6 @@ INTERNAL { + rte_gpu_get_by_name; + rte_gpu_notify; + rte_gpu_release; ++ ++ local: *; + }; +diff --git a/dpdk/lib/graph/rte_graph_worker.h b/dpdk/lib/graph/rte_graph_worker.h +index eef77f732a..0c0b9c095a 100644 +--- a/dpdk/lib/graph/rte_graph_worker.h ++++ b/dpdk/lib/graph/rte_graph_worker.h +@@ -155,7 +155,7 @@ rte_graph_walk(struct rte_graph *graph) + * +-----+ <= cir_start + mask + */ + while (likely(head != graph->tail)) { +- node = RTE_PTR_ADD(graph, cir_start[(int32_t)head++]); ++ node = (struct rte_node *)RTE_PTR_ADD(graph, cir_start[(int32_t)head++]); + RTE_ASSERT(node->fence == RTE_GRAPH_FENCE); + objs = node->objs; + rte_prefetch0(objs); +diff --git a/dpdk/lib/gro/rte_gro.c b/dpdk/lib/gro/rte_gro.c +index 8ca4da67e9..7a788523ad 100644 +--- a/dpdk/lib/gro/rte_gro.c ++++ b/dpdk/lib/gro/rte_gro.c +@@ -33,6 +33,7 @@ static gro_tbl_pkt_count_fn tbl_pkt_count_fn[RTE_GRO_TYPE_MAX_NUM] = { + + #define IS_IPV4_TCP_PKT(ptype) (RTE_ETH_IS_IPV4_HDR(ptype) && \ + ((ptype & RTE_PTYPE_L4_TCP) == RTE_PTYPE_L4_TCP) && \ ++ ((ptype & RTE_PTYPE_L4_FRAG) != RTE_PTYPE_L4_FRAG) && \ + (RTE_ETH_IS_TUNNEL_PKT(ptype) == 0)) + + #define IS_IPV4_UDP_PKT(ptype) (RTE_ETH_IS_IPV4_HDR(ptype) && \ +@@ -41,6 +42,7 @@ static gro_tbl_pkt_count_fn tbl_pkt_count_fn[RTE_GRO_TYPE_MAX_NUM] = { + + #define IS_IPV4_VXLAN_TCP4_PKT(ptype) (RTE_ETH_IS_IPV4_HDR(ptype) && \ + ((ptype & RTE_PTYPE_L4_UDP) == RTE_PTYPE_L4_UDP) && \ ++ ((ptype & RTE_PTYPE_L4_FRAG) != RTE_PTYPE_L4_FRAG) && \ + ((ptype & RTE_PTYPE_TUNNEL_VXLAN) == \ + RTE_PTYPE_TUNNEL_VXLAN) && \ + ((ptype & RTE_PTYPE_INNER_L4_TCP) == \ +diff --git a/dpdk/lib/ipsec/esp_outb.c b/dpdk/lib/ipsec/esp_outb.c +index 672e56aba0..28bd58e3c7 100644 +--- a/dpdk/lib/ipsec/esp_outb.c ++++ b/dpdk/lib/ipsec/esp_outb.c +@@ -197,7 +197,7 @@ outb_tun_pkt_prepare(struct rte_ipsec_sa *sa, rte_be64_t sqc, + /* if UDP encap is enabled update the dgram_len */ + if (sa->type & RTE_IPSEC_SATP_NATT_ENABLE) { + struct rte_udp_hdr *udph = (struct rte_udp_hdr *) +- (ph - sizeof(struct rte_udp_hdr)); ++ (ph + sa->hdr_len - sizeof(struct rte_udp_hdr)); + udph->dgram_len = rte_cpu_to_be_16(mb->pkt_len - sqh_len - + sa->hdr_l3_off - sa->hdr_len); + } +diff --git a/dpdk/lib/ipsec/ipsec_telemetry.c b/dpdk/lib/ipsec/ipsec_telemetry.c +index b8b08404b6..9a91e47122 100644 +--- a/dpdk/lib/ipsec/ipsec_telemetry.c ++++ b/dpdk/lib/ipsec/ipsec_telemetry.c +@@ -236,7 +236,7 @@ RTE_INIT(rte_ipsec_telemetry_init) + "Return list of IPsec SAs with telemetry enabled."); + rte_telemetry_register_cmd("/ipsec/sa/stats", + handle_telemetry_cmd_ipsec_sa_stats, +- "Returns IPsec SA stastistics. Parameters: int sa_spi"); ++ "Returns IPsec SA statistics. Parameters: int sa_spi"); + rte_telemetry_register_cmd("/ipsec/sa/details", + handle_telemetry_cmd_ipsec_sa_details, + "Returns IPsec SA configuration. Parameters: int sa_spi"); +diff --git a/dpdk/lib/ipsec/rte_ipsec_group.h b/dpdk/lib/ipsec/rte_ipsec_group.h +index 60ab297710..62c2bd7217 100644 +--- a/dpdk/lib/ipsec/rte_ipsec_group.h ++++ b/dpdk/lib/ipsec/rte_ipsec_group.h +@@ -49,10 +49,10 @@ rte_ipsec_ses_from_crypto(const struct rte_crypto_op *cop) + + if (cop->sess_type == RTE_CRYPTO_OP_SECURITY_SESSION) { + ss = cop->sym[0].sec_session; +- return (void *)(uintptr_t)ss->opaque_data; ++ return (struct rte_ipsec_session *)(uintptr_t)ss->opaque_data; + } else if (cop->sess_type == RTE_CRYPTO_OP_WITH_SESSION) { + cs = cop->sym[0].session; +- return (void *)(uintptr_t)cs->opaque_data; ++ return (struct rte_ipsec_session *)(uintptr_t)cs->opaque_data; + } + return NULL; + } +diff --git a/dpdk/lib/ipsec/rte_ipsec_sad.h b/dpdk/lib/ipsec/rte_ipsec_sad.h +index b65d295831..a3ae57df7e 100644 +--- a/dpdk/lib/ipsec/rte_ipsec_sad.h ++++ b/dpdk/lib/ipsec/rte_ipsec_sad.h +@@ -153,7 +153,7 @@ rte_ipsec_sad_destroy(struct rte_ipsec_sad *sad); + * @param keys + * Array of keys to be looked up in the SAD + * @param sa +- * Pointer assocoated with the keys. ++ * Pointer associated with the keys. + * If the lookup for the given key failed, then corresponding sa + * will be NULL + * @param n +diff --git a/dpdk/lib/ipsec/sa.c b/dpdk/lib/ipsec/sa.c +index 1e51482c92..c921699390 100644 +--- a/dpdk/lib/ipsec/sa.c ++++ b/dpdk/lib/ipsec/sa.c +@@ -362,13 +362,13 @@ esp_outb_tun_init(struct rte_ipsec_sa *sa, const struct rte_ipsec_sa_prm *prm) + + memcpy(sa->hdr, prm->tun.hdr, prm->tun.hdr_len); + +- /* insert UDP header if UDP encapsulation is inabled */ ++ /* insert UDP header if UDP encapsulation is enabled */ + if (sa->type & RTE_IPSEC_SATP_NATT_ENABLE) { + struct rte_udp_hdr *udph = (struct rte_udp_hdr *) + &sa->hdr[prm->tun.hdr_len]; + sa->hdr_len += sizeof(struct rte_udp_hdr); +- udph->src_port = prm->ipsec_xform.udp.sport; +- udph->dst_port = prm->ipsec_xform.udp.dport; ++ udph->src_port = rte_cpu_to_be_16(prm->ipsec_xform.udp.sport); ++ udph->dst_port = rte_cpu_to_be_16(prm->ipsec_xform.udp.dport); + udph->dgram_cksum = 0; + } + +diff --git a/dpdk/lib/kni/rte_kni_common.h b/dpdk/lib/kni/rte_kni_common.h +index b547ea5501..8d3ee0fa4f 100644 +--- a/dpdk/lib/kni/rte_kni_common.h ++++ b/dpdk/lib/kni/rte_kni_common.h +@@ -6,6 +6,10 @@ + #ifndef _RTE_KNI_COMMON_H_ + #define _RTE_KNI_COMMON_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + #ifdef __KERNEL__ + #include + #include +@@ -136,4 +140,8 @@ struct rte_kni_device_info { + #define RTE_KNI_IOCTL_CREATE _IOWR(0, 2, struct rte_kni_device_info) + #define RTE_KNI_IOCTL_RELEASE _IOWR(0, 3, struct rte_kni_device_info) + ++#ifdef __cplusplus ++} ++#endif ++ + #endif /* _RTE_KNI_COMMON_H_ */ +diff --git a/dpdk/lib/lpm/rte_lpm_altivec.h b/dpdk/lib/lpm/rte_lpm_altivec.h +index 4fbc1b595d..bab8929495 100644 +--- a/dpdk/lib/lpm/rte_lpm_altivec.h ++++ b/dpdk/lib/lpm/rte_lpm_altivec.h +@@ -19,14 +19,14 @@ static inline void + rte_lpm_lookupx4(const struct rte_lpm *lpm, xmm_t ip, uint32_t hop[4], + uint32_t defv) + { +- vector signed int i24; ++ xmm_t i24; + rte_xmm_t i8; + uint32_t tbl[4]; + uint64_t idx, pt, pt2; + const uint32_t *ptbl; + + const uint32_t mask = UINT8_MAX; +- const vector signed int mask8 = (xmm_t){mask, mask, mask, mask}; ++ const xmm_t mask8 = (xmm_t){mask, mask, mask, mask}; + + /* + * RTE_LPM_VALID_EXT_ENTRY_BITMASK for 2 LPM entries +@@ -46,7 +46,7 @@ rte_lpm_lookupx4(const struct rte_lpm *lpm, xmm_t ip, uint32_t hop[4], + + /* get 4 indexes for tbl24[]. */ + i24 = vec_sr((xmm_t) ip, +- (vector unsigned int){CHAR_BIT, CHAR_BIT, CHAR_BIT, CHAR_BIT}); ++ (__vector unsigned int){CHAR_BIT, CHAR_BIT, CHAR_BIT, CHAR_BIT}); + + /* extract values from tbl24[] */ + idx = (uint32_t)i24[0]; +diff --git a/dpdk/lib/mbuf/rte_mbuf.c b/dpdk/lib/mbuf/rte_mbuf.c +index 604d77bbda..dce900f28f 100644 +--- a/dpdk/lib/mbuf/rte_mbuf.c ++++ b/dpdk/lib/mbuf/rte_mbuf.c +@@ -685,6 +685,9 @@ rte_pktmbuf_dump(FILE *f, const struct rte_mbuf *m, unsigned dump_len) + fprintf(f, " pkt_len=%u, ol_flags=%#"PRIx64", nb_segs=%u, port=%u", + m->pkt_len, m->ol_flags, m->nb_segs, m->port); + ++ if (m->ol_flags & (RTE_MBUF_F_RX_QINQ | RTE_MBUF_F_TX_QINQ)) ++ fprintf(f, ", vlan_tci_outer=%u", m->vlan_tci_outer); ++ + if (m->ol_flags & (RTE_MBUF_F_RX_VLAN | RTE_MBUF_F_TX_VLAN)) + fprintf(f, ", vlan_tci=%u", m->vlan_tci); + +diff --git a/dpdk/lib/mbuf/rte_mbuf_core.h b/dpdk/lib/mbuf/rte_mbuf_core.h +index 321a419c71..3d6ddd6773 100644 +--- a/dpdk/lib/mbuf/rte_mbuf_core.h ++++ b/dpdk/lib/mbuf/rte_mbuf_core.h +@@ -8,7 +8,7 @@ + + /** + * @file +- * This file contains definion of RTE mbuf structure itself, ++ * This file contains definition of RTE mbuf structure itself, + * packet offload flags and some related macros. + * For majority of DPDK entities, it is not recommended to include + * this file directly, use include instead. +diff --git a/dpdk/lib/meson.build b/dpdk/lib/meson.build +index 018976df17..fbaa6ef7c2 100644 +--- a/dpdk/lib/meson.build ++++ b/dpdk/lib/meson.build +@@ -3,7 +3,7 @@ + + + # process all libraries equally, as far as possible +-# "core" libs first, then others alphebetically as far as possible ++# "core" libs first, then others alphabetically as far as possible + # NOTE: for speed of meson runs, the dependencies in the subdirectories + # sometimes skip deps that would be implied by others, e.g. if mempool is + # given as a dep, no need to mention ring. This is especially true for the +diff --git a/dpdk/lib/metrics/rte_metrics_telemetry.h b/dpdk/lib/metrics/rte_metrics_telemetry.h +index 2b6eb1ccc8..09b14d9336 100644 +--- a/dpdk/lib/metrics/rte_metrics_telemetry.h ++++ b/dpdk/lib/metrics/rte_metrics_telemetry.h +@@ -13,6 +13,9 @@ + #ifndef _RTE_METRICS_TELEMETRY_H_ + #define _RTE_METRICS_TELEMETRY_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif + + enum rte_telemetry_stats_type { + PORT_STATS = 0, +@@ -60,4 +63,8 @@ __rte_experimental + int32_t + rte_metrics_tel_extract_data(struct telemetry_encode_param *ep, json_t *data); + ++#ifdef __cplusplus ++} ++#endif ++ + #endif +diff --git a/dpdk/lib/net/rte_gtp.h b/dpdk/lib/net/rte_gtp.h +index dca940c2c5..9849872366 100644 +--- a/dpdk/lib/net/rte_gtp.h ++++ b/dpdk/lib/net/rte_gtp.h +@@ -75,11 +75,11 @@ struct rte_gtp_psc_generic_hdr { + uint8_t spare:2; /**< type specific spare bits */ + uint8_t qfi:6; /**< Qos Flow Identifier */ + #else +- uint8_t qfi:6; /**< Qos Flow Identifier */ +- uint8_t spare:2; /**< type specific spare bits */ + uint8_t pad:3; /**< type specific pad bits */ + uint8_t qmp:1; /**< Qos Monitoring Packet */ + uint8_t type:4; /**< PDU type */ ++ uint8_t qfi:6; /**< Qos Flow Identifier */ ++ uint8_t spare:2; /**< type specific spare bits */ + #endif + uint8_t data[0]; /**< variable length data fields */ + } __rte_packed; +@@ -100,12 +100,13 @@ struct rte_gtp_psc_type0_hdr { + uint8_t rqi:1; /**< Reflective Qos Indicator */ + uint8_t qfi:6; /**< Qos Flow Identifier */ + #else +- uint8_t qfi:6; /**< Qos Flow Identifier */ +- uint8_t rqi:1; /**< Reflective Qos Indicator */ +- uint8_t ppp:1; /**< Paging policy presence */ + uint8_t spare_dl1:2; /**< spare down link bits */ + uint8_t snp:1; /**< Sequence number presence */ ++ uint8_t qmp:1; /**< Qos Monitoring Packet */ + uint8_t type:4; /**< PDU type */ ++ uint8_t qfi:6; /**< Qos Flow Identifier */ ++ uint8_t rqi:1; /**< Reflective Qos Indicator */ ++ uint8_t ppp:1; /**< Paging policy presence */ + #endif + uint8_t data[0]; /**< variable length data fields */ + } __rte_packed; +@@ -127,14 +128,14 @@ struct rte_gtp_psc_type1_hdr { + uint8_t spare_ul2:1; /**< spare up link bits */ + uint8_t qfi:6; /**< Qos Flow Identifier */ + #else +- uint8_t qfi:6; /**< Qos Flow Identifier */ +- uint8_t spare_ul2:1; /**< spare up link bits */ +- uint8_t n_delay_ind:1; /**< N3/N9 delay result presence */ + uint8_t snp:1; /**< Sequence number presence ul */ + uint8_t ul_delay_ind:1; /**< ul delay result presence */ + uint8_t dl_delay_ind:1; /**< dl delay result presence */ + uint8_t qmp:1; /**< Qos Monitoring Packet */ + uint8_t type:4; /**< PDU type */ ++ uint8_t qfi:6; /**< Qos Flow Identifier */ ++ uint8_t spare_ul2:1; /**< spare up link bits */ ++ uint8_t n_delay_ind:1; /**< N3/N9 delay result presence */ + #endif + uint8_t data[0]; /**< variable length data fields */ + } __rte_packed; +diff --git a/dpdk/lib/net/rte_l2tpv2.h b/dpdk/lib/net/rte_l2tpv2.h +index b90e36cf12..1f3ad3f03c 100644 +--- a/dpdk/lib/net/rte_l2tpv2.h ++++ b/dpdk/lib/net/rte_l2tpv2.h +@@ -89,16 +89,6 @@ struct rte_l2tpv2_common_hdr { + __extension__ + struct { + #if RTE_BYTE_ORDER == RTE_LITTLE_ENDIAN +- uint16_t t:1; /**< message Type */ +- uint16_t l:1; /**< length option bit */ +- uint16_t res1:2; /**< reserved */ +- uint16_t s:1; /**< ns/nr option bit */ +- uint16_t res2:1; /**< reserved */ +- uint16_t o:1; /**< offset option bit */ +- uint16_t p:1; /**< priority option bit */ +- uint16_t res3:4; /**< reserved */ +- uint16_t ver:4; /**< protocol version */ +-#elif RTE_BYTE_ORDER == RTE_BIG_ENDIAN + uint16_t ver:4; /**< protocol version */ + uint16_t res3:4; /**< reserved */ + uint16_t p:1; /**< priority option bit */ +@@ -108,6 +98,16 @@ struct rte_l2tpv2_common_hdr { + uint16_t res1:2; /**< reserved */ + uint16_t l:1; /**< length option bit */ + uint16_t t:1; /**< message Type */ ++#elif RTE_BYTE_ORDER == RTE_BIG_ENDIAN ++ uint16_t t:1; /**< message Type */ ++ uint16_t l:1; /**< length option bit */ ++ uint16_t res1:2; /**< reserved */ ++ uint16_t s:1; /**< ns/nr option bit */ ++ uint16_t res2:1; /**< reserved */ ++ uint16_t o:1; /**< offset option bit */ ++ uint16_t p:1; /**< priority option bit */ ++ uint16_t res3:4; /**< reserved */ ++ uint16_t ver:4; /**< protocol version */ + #endif + }; + }; +@@ -143,7 +143,7 @@ struct rte_l2tpv2_msg_without_length { + /** + * L2TPv2 message Header contains all options except ns_nr(length, + * offset size, offset padding). +- * Ns and Nr MUST be toghter. ++ * Ns and Nr MUST be together. + */ + struct rte_l2tpv2_msg_without_ns_nr { + rte_be16_t length; /**< length(16) */ +@@ -155,7 +155,7 @@ struct rte_l2tpv2_msg_without_ns_nr { + + /** + * L2TPv2 message Header contains all options except ns_nr(length, ns, nr). +- * offset size and offset padding MUST be toghter. ++ * offset size and offset padding MUST be together. + */ + struct rte_l2tpv2_msg_without_offset { + rte_be16_t length; /**< length(16) */ +diff --git a/dpdk/lib/pcapng/rte_pcapng.c b/dpdk/lib/pcapng/rte_pcapng.c +index 03edabe73e..0caf3d31f8 100644 +--- a/dpdk/lib/pcapng/rte_pcapng.c ++++ b/dpdk/lib/pcapng/rte_pcapng.c +@@ -20,6 +20,7 @@ + #include + #include + #include ++#include + #include + + #include "pcapng_proto.h" +@@ -35,27 +36,63 @@ struct rte_pcapng { + }; + + /* For converting TSC cycles to PCAPNG ns format */ +-struct pcapng_time { ++static struct pcapng_time { + uint64_t ns; + uint64_t cycles; ++ uint64_t tsc_hz; ++ struct rte_reciprocal_u64 tsc_hz_inverse; + } pcapng_time; + +-RTE_INIT(pcapng_init) ++static inline void ++pcapng_init(void) + { + struct timespec ts; + + pcapng_time.cycles = rte_get_tsc_cycles(); + clock_gettime(CLOCK_REALTIME, &ts); ++ pcapng_time.cycles = (pcapng_time.cycles + rte_get_tsc_cycles()) / 2; + pcapng_time.ns = rte_timespec_to_ns(&ts); ++ ++ pcapng_time.tsc_hz = rte_get_tsc_hz(); ++ pcapng_time.tsc_hz_inverse = rte_reciprocal_value_u64(pcapng_time.tsc_hz); + } + + /* PCAPNG timestamps are in nanoseconds */ + static uint64_t pcapng_tsc_to_ns(uint64_t cycles) + { +- uint64_t delta; +- ++ uint64_t delta, secs; ++ ++ if (!pcapng_time.tsc_hz) ++ pcapng_init(); ++ ++ /* In essence the calculation is: ++ * delta = (cycles - pcapng_time.cycles) * NSEC_PRE_SEC / rte_get_tsc_hz() ++ * but this overflows within 4 to 8 seconds depending on TSC frequency. ++ * Instead, if delta >= pcapng_time.tsc_hz: ++ * Increase pcapng_time.ns and pcapng_time.cycles by the number of ++ * whole seconds in delta and reduce delta accordingly. ++ * delta will therefore always lie in the interval [0, pcapng_time.tsc_hz), ++ * which will not overflow when multiplied by NSEC_PER_SEC provided the ++ * TSC frequency < approx 18.4GHz. ++ * ++ * Currently all TSCs operate below 5GHz. ++ */ + delta = cycles - pcapng_time.cycles; +- return pcapng_time.ns + (delta * NSEC_PER_SEC) / rte_get_tsc_hz(); ++ if (unlikely(delta >= pcapng_time.tsc_hz)) { ++ if (likely(delta < pcapng_time.tsc_hz * 2)) { ++ delta -= pcapng_time.tsc_hz; ++ pcapng_time.cycles += pcapng_time.tsc_hz; ++ pcapng_time.ns += NSEC_PER_SEC; ++ } else { ++ secs = rte_reciprocal_divide_u64(delta, &pcapng_time.tsc_hz_inverse); ++ delta -= secs * pcapng_time.tsc_hz; ++ pcapng_time.cycles += secs * pcapng_time.tsc_hz; ++ pcapng_time.ns += secs * NSEC_PER_SEC; ++ } ++ } ++ ++ return pcapng_time.ns + rte_reciprocal_divide_u64(delta * NSEC_PER_SEC, ++ &pcapng_time.tsc_hz_inverse); + } + + /* length of option including padding */ +@@ -177,8 +214,8 @@ pcapng_add_interface(rte_pcapng_t *self, uint16_t port) + "%s-%s", dev->bus->name, dev->name); + + /* DPDK reports in units of Mbps */ +- rte_eth_link_get(port, &link); +- if (link.link_status == RTE_ETH_LINK_UP) ++ if (rte_eth_link_get(port, &link) == 0 && ++ link.link_status == RTE_ETH_LINK_UP) + speed = link.link_speed * PCAPNG_MBPS_SPEED; + + if (rte_eth_macaddr_get(port, &macaddr) < 0) +diff --git a/dpdk/lib/pipeline/rte_swx_ctl.c b/dpdk/lib/pipeline/rte_swx_ctl.c +index 1c908e3e3f..f52ccffd75 100644 +--- a/dpdk/lib/pipeline/rte_swx_ctl.c ++++ b/dpdk/lib/pipeline/rte_swx_ctl.c +@@ -372,18 +372,34 @@ table_entry_check(struct rte_swx_ctl_pipeline *ctl, + + if (data_check) { + struct action *a; ++ struct rte_swx_ctl_table_action_info *tai; + uint32_t i; + + /* action_id. */ +- for (i = 0; i < table->info.n_actions; i++) +- if (entry->action_id == table->actions[i].action_id) ++ for (i = 0; i < table->info.n_actions; i++) { ++ tai = &table->actions[i]; ++ ++ if (entry->action_id == tai->action_id) + break; ++ } + + CHECK(i < table->info.n_actions, EINVAL); + + /* action_data. */ + a = &ctl->actions[entry->action_id]; + CHECK(!(a->data_size && !entry->action_data), EINVAL); ++ ++ /* When both key_check and data_check are true, we are interested in both the entry ++ * key and data, which means the operation is _regular_ table entry add. ++ */ ++ if (key_check && !tai->action_is_for_table_entries) ++ return -EINVAL; ++ ++ /* When key_check is false while data_check is true, we are only interested in the ++ * entry data, which means the operation is _default_ table entry add. ++ */ ++ if (!key_check && !tai->action_is_for_default_entry) ++ return -EINVAL; + } + + return 0; +@@ -1005,15 +1021,16 @@ learner_action_data_size_get(struct rte_swx_ctl_pipeline *ctl, struct learner *l + static void + table_state_free(struct rte_swx_ctl_pipeline *ctl) + { +- uint32_t i; ++ uint32_t table_base_index, selector_base_index, learner_base_index, i; + + if (!ctl->ts_next) + return; + + /* For each table, free its table state. */ ++ table_base_index = 0; + for (i = 0; i < ctl->info.n_tables; i++) { + struct table *table = &ctl->tables[i]; +- struct rte_swx_table_state *ts = &ctl->ts_next[i]; ++ struct rte_swx_table_state *ts = &ctl->ts_next[table_base_index + i]; + + /* Default action data. */ + free(ts->default_action_data); +@@ -1024,8 +1041,9 @@ table_state_free(struct rte_swx_ctl_pipeline *ctl) + } + + /* For each selector table, free its table state. */ ++ selector_base_index = ctl->info.n_tables; + for (i = 0; i < ctl->info.n_selectors; i++) { +- struct rte_swx_table_state *ts = &ctl->ts_next[i]; ++ struct rte_swx_table_state *ts = &ctl->ts_next[selector_base_index + i]; + + /* Table object. */ + if (ts->obj) +@@ -1033,8 +1051,9 @@ table_state_free(struct rte_swx_ctl_pipeline *ctl) + } + + /* For each learner table, free its table state. */ ++ learner_base_index = ctl->info.n_tables + ctl->info.n_selectors; + for (i = 0; i < ctl->info.n_learners; i++) { +- struct rte_swx_table_state *ts = &ctl->ts_next[i]; ++ struct rte_swx_table_state *ts = &ctl->ts_next[learner_base_index + i]; + + /* Default action data. */ + free(ts->default_action_data); +@@ -1047,10 +1066,10 @@ table_state_free(struct rte_swx_ctl_pipeline *ctl) + static int + table_state_create(struct rte_swx_ctl_pipeline *ctl) + { ++ uint32_t table_base_index, selector_base_index, learner_base_index, i; + int status = 0; +- uint32_t i; + +- ctl->ts_next = calloc(ctl->info.n_tables + ctl->info.n_selectors, ++ ctl->ts_next = calloc(ctl->info.n_tables + ctl->info.n_selectors + ctl->info.n_learners, + sizeof(struct rte_swx_table_state)); + if (!ctl->ts_next) { + status = -ENOMEM; +@@ -1058,10 +1077,11 @@ table_state_create(struct rte_swx_ctl_pipeline *ctl) + } + + /* Tables. */ ++ table_base_index = 0; + for (i = 0; i < ctl->info.n_tables; i++) { + struct table *table = &ctl->tables[i]; +- struct rte_swx_table_state *ts = &ctl->ts[i]; +- struct rte_swx_table_state *ts_next = &ctl->ts_next[i]; ++ struct rte_swx_table_state *ts = &ctl->ts[table_base_index + i]; ++ struct rte_swx_table_state *ts_next = &ctl->ts_next[table_base_index + i]; + + /* Table object. */ + if (!table->is_stub && table->ops.add) { +@@ -1094,9 +1114,10 @@ table_state_create(struct rte_swx_ctl_pipeline *ctl) + } + + /* Selector tables. */ ++ selector_base_index = ctl->info.n_tables; + for (i = 0; i < ctl->info.n_selectors; i++) { + struct selector *s = &ctl->selectors[i]; +- struct rte_swx_table_state *ts_next = &ctl->ts_next[ctl->info.n_tables + i]; ++ struct rte_swx_table_state *ts_next = &ctl->ts_next[selector_base_index + i]; + + /* Table object. */ + ts_next->obj = rte_swx_table_selector_create(&s->params, NULL, ctl->numa_node); +@@ -1107,10 +1128,11 @@ table_state_create(struct rte_swx_ctl_pipeline *ctl) + } + + /* Learner tables. */ ++ learner_base_index = ctl->info.n_tables + ctl->info.n_selectors; + for (i = 0; i < ctl->info.n_learners; i++) { + struct learner *l = &ctl->learners[i]; +- struct rte_swx_table_state *ts = &ctl->ts[i]; +- struct rte_swx_table_state *ts_next = &ctl->ts_next[i]; ++ struct rte_swx_table_state *ts = &ctl->ts[learner_base_index + i]; ++ struct rte_swx_table_state *ts_next = &ctl->ts_next[learner_base_index + i]; + + /* Table object: duplicate from the current table state. */ + ts_next->obj = ts->obj; +@@ -1446,8 +1468,6 @@ rte_swx_ctl_pipeline_table_entry_add(struct rte_swx_ctl_pipeline *ctl, + CHECK(entry, EINVAL); + CHECK(!table_entry_check(ctl, table_id, entry, 1, 1), EINVAL); + +- CHECK(table->actions[entry->action_id].action_is_for_table_entries, EINVAL); +- + new_entry = table_entry_duplicate(ctl, table_id, entry, 1, 1); + CHECK(new_entry, ENOMEM); + +@@ -1653,8 +1673,6 @@ rte_swx_ctl_pipeline_table_default_entry_add(struct rte_swx_ctl_pipeline *ctl, + CHECK(entry, EINVAL); + CHECK(!table_entry_check(ctl, table_id, entry, 0, 1), EINVAL); + +- CHECK(table->actions[entry->action_id].action_is_for_default_entry, EINVAL); +- + new_entry = table_entry_duplicate(ctl, table_id, entry, 0, 1); + CHECK(new_entry, ENOMEM); + +diff --git a/dpdk/lib/pipeline/rte_swx_ctl.h b/dpdk/lib/pipeline/rte_swx_ctl.h +index 46d05823e1..82e62e70a7 100644 +--- a/dpdk/lib/pipeline/rte_swx_ctl.h ++++ b/dpdk/lib/pipeline/rte_swx_ctl.h +@@ -369,7 +369,7 @@ struct rte_swx_table_stats { + uint64_t n_pkts_miss; + + /** Number of packets (with either lookup hit or miss) per pipeline +- * action. Array of pipeline *n_actions* elements indedex by the ++ * action. Array of pipeline *n_actions* elements indexed by the + * pipeline-level *action_id*, therefore this array has the same size + * for all the tables within the same pipeline. + */ +@@ -629,7 +629,7 @@ struct rte_swx_learner_stats { + uint64_t n_pkts_forget; + + /** Number of packets (with either lookup hit or miss) per pipeline action. Array of +- * pipeline *n_actions* elements indedex by the pipeline-level *action_id*, therefore this ++ * pipeline *n_actions* elements indexed by the pipeline-level *action_id*, therefore this + * array has the same size for all the tables within the same pipeline. + */ + uint64_t *n_pkts_action; +diff --git a/dpdk/lib/pipeline/rte_swx_pipeline.c b/dpdk/lib/pipeline/rte_swx_pipeline.c +index 2145ca0a42..8d5073cf19 100644 +--- a/dpdk/lib/pipeline/rte_swx_pipeline.c ++++ b/dpdk/lib/pipeline/rte_swx_pipeline.c +@@ -8531,7 +8531,7 @@ table_state_build(struct rte_swx_pipeline *p) + struct selector *s; + struct learner *l; + +- p->table_state = calloc(p->n_tables + p->n_selectors, ++ p->table_state = calloc(p->n_tables + p->n_selectors + p->n_learners, + sizeof(struct rte_swx_table_state)); + CHECK(p->table_state, ENOMEM); + +diff --git a/dpdk/lib/pipeline/rte_swx_pipeline_internal.h b/dpdk/lib/pipeline/rte_swx_pipeline_internal.h +index 1921fdcd78..c8fa978580 100644 +--- a/dpdk/lib/pipeline/rte_swx_pipeline_internal.h ++++ b/dpdk/lib/pipeline/rte_swx_pipeline_internal.h +@@ -309,7 +309,7 @@ enum instruction_type { + */ + INSTR_ALU_CKADD_FIELD, /* src = H */ + INSTR_ALU_CKADD_STRUCT20, /* src = h.header, with sizeof(header) = 20 */ +- INSTR_ALU_CKADD_STRUCT, /* src = h.hdeader, with any sizeof(header) */ ++ INSTR_ALU_CKADD_STRUCT, /* src = h.header, with any sizeof(header) */ + + /* cksub dst src + * dst = dst '- src +@@ -1562,7 +1562,7 @@ emit_handler(struct thread *t) + return; + } + +- /* Header encapsulation (optionally, with prior header decasulation). */ ++ /* Header encapsulation (optionally, with prior header decapsulation). */ + if ((t->n_headers_out == 2) && + (h1->ptr + h1->n_bytes == t->ptr) && + (h0->ptr == h0->ptr0)) { +@@ -1820,9 +1820,9 @@ __instr_hdr_emit_many_exec(struct rte_swx_pipeline *p __rte_unused, + { + uint64_t valid_headers = t->valid_headers; + uint32_t n_headers_out = t->n_headers_out; +- struct header_out_runtime *ho = &t->headers_out[n_headers_out - 1]; ++ struct header_out_runtime *ho = NULL; + uint8_t *ho_ptr = NULL; +- uint32_t ho_nbytes = 0, first = 1, i; ++ uint32_t ho_nbytes = 0, i; + + for (i = 0; i < n_emit; i++) { + uint32_t header_id = ip->io.hdr.header_id[i]; +@@ -1834,18 +1834,21 @@ __instr_hdr_emit_many_exec(struct rte_swx_pipeline *p __rte_unused, + + uint8_t *hi_ptr = t->structs[struct_id]; + +- if (!MASK64_BIT_GET(valid_headers, header_id)) ++ if (!MASK64_BIT_GET(valid_headers, header_id)) { ++ TRACE("[Thread %2u]: emit header %u (invalid)\n", ++ p->thread_id, ++ header_id); ++ + continue; ++ } + +- TRACE("[Thread %2u]: emit header %u\n", ++ TRACE("[Thread %2u]: emit header %u (valid)\n", + p->thread_id, + header_id); + + /* Headers. */ +- if (first) { +- first = 0; +- +- if (!t->n_headers_out) { ++ if (!ho) { ++ if (!n_headers_out) { + ho = &t->headers_out[0]; + + ho->ptr0 = hi_ptr0; +@@ -1858,6 +1861,8 @@ __instr_hdr_emit_many_exec(struct rte_swx_pipeline *p __rte_unused, + + continue; + } else { ++ ho = &t->headers_out[n_headers_out - 1]; ++ + ho_ptr = ho->ptr; + ho_nbytes = ho->n_bytes; + } +@@ -1879,7 +1884,8 @@ __instr_hdr_emit_many_exec(struct rte_swx_pipeline *p __rte_unused, + } + } + +- ho->n_bytes = ho_nbytes; ++ if (ho) ++ ho->n_bytes = ho_nbytes; + t->n_headers_out = n_headers_out; + } + +diff --git a/dpdk/lib/pipeline/rte_swx_pipeline_spec.c b/dpdk/lib/pipeline/rte_swx_pipeline_spec.c +index 8e9aa44e30..07a7580ac8 100644 +--- a/dpdk/lib/pipeline/rte_swx_pipeline_spec.c ++++ b/dpdk/lib/pipeline/rte_swx_pipeline_spec.c +@@ -2011,7 +2011,7 @@ rte_swx_pipeline_build_from_spec(struct rte_swx_pipeline *p, + if (err_line) + *err_line = 0; + if (err_msg) +- *err_msg = "Null pipeline arument."; ++ *err_msg = "Null pipeline argument."; + status = -EINVAL; + goto error; + } +diff --git a/dpdk/lib/power/power_cppc_cpufreq.c b/dpdk/lib/power/power_cppc_cpufreq.c +index 6afd310e4e..25185a791c 100644 +--- a/dpdk/lib/power/power_cppc_cpufreq.c ++++ b/dpdk/lib/power/power_cppc_cpufreq.c +@@ -621,7 +621,7 @@ power_cppc_enable_turbo(unsigned int lcore_id) + return -1; + } + +- /* TODO: must set to max once enbling Turbo? Considering add condition: ++ /* TODO: must set to max once enabling Turbo? Considering add condition: + * if ((pi->turbo_available) && (pi->curr_idx <= 1)) + */ + /* Max may have changed, so call to max function */ +diff --git a/dpdk/lib/regexdev/rte_regexdev.h b/dpdk/lib/regexdev/rte_regexdev.h +index 86f0b231b0..513ce5b67c 100644 +--- a/dpdk/lib/regexdev/rte_regexdev.h ++++ b/dpdk/lib/regexdev/rte_regexdev.h +@@ -228,6 +228,9 @@ extern int rte_regexdev_logtype; + } while (0) + + /** ++ * @warning ++ * @b EXPERIMENTAL: this API may change without prior notice. ++ * + * Check if dev_id is ready. + * + * @param dev_id +@@ -237,6 +240,7 @@ extern int rte_regexdev_logtype; + * - 0 if device state is not in ready state. + * - 1 if device state is ready state. + */ ++__rte_experimental + int rte_regexdev_is_valid_dev(uint16_t dev_id); + + /** +@@ -298,14 +302,14 @@ rte_regexdev_get_dev_id(const char *name); + * backtracking positions remembered by any tokens inside the group. + * Example RegEx is `a(?>bc|b)c` if the given patterns are `abc` and `abcc` then + * `a(bc|b)c` matches both where as `a(?>bc|b)c` matches only abcc because +- * atomic groups don't allow backtracing back to `b`. ++ * atomic groups don't allow backtracking back to `b`. + * + * @see struct rte_regexdev_info::regexdev_capa + */ + + #define RTE_REGEXDEV_SUPP_PCRE_BACKTRACKING_CTRL_F (1ULL << 3) + /**< RegEx device support PCRE backtracking control verbs. +- * Some examples of backtracing verbs are (*COMMIT), (*ACCEPT), (*FAIL), ++ * Some examples of backtracking verbs are (*COMMIT), (*ACCEPT), (*FAIL), + * (*SKIP), (*PRUNE). + * + * @see struct rte_regexdev_info::regexdev_capa +@@ -1015,7 +1019,7 @@ rte_regexdev_rule_db_update(uint8_t dev_id, + * @b EXPERIMENTAL: this API may change without prior notice. + * + * Compile local rule set and burn the complied result to the +- * RegEx deive. ++ * RegEx device. + * + * @param dev_id + * RegEx device identifier. +diff --git a/dpdk/lib/regexdev/rte_regexdev_driver.h b/dpdk/lib/regexdev/rte_regexdev_driver.h +index 64742016c0..6246b144a6 100644 +--- a/dpdk/lib/regexdev/rte_regexdev_driver.h ++++ b/dpdk/lib/regexdev/rte_regexdev_driver.h +@@ -32,6 +32,7 @@ extern "C" { + * A pointer to the RegEx device slot case of success, + * NULL otherwise. + */ ++__rte_internal + struct rte_regexdev *rte_regexdev_register(const char *name); + + /** +@@ -41,6 +42,7 @@ struct rte_regexdev *rte_regexdev_register(const char *name); + * @param dev + * Device to be released. + */ ++__rte_internal + void rte_regexdev_unregister(struct rte_regexdev *dev); + + /** +@@ -50,6 +52,7 @@ void rte_regexdev_unregister(struct rte_regexdev *dev); + * @param name + * The device name. + */ ++__rte_internal + struct rte_regexdev *rte_regexdev_get_device_by_name(const char *name); + + #ifdef __cplusplus +diff --git a/dpdk/lib/regexdev/version.map b/dpdk/lib/regexdev/version.map +index 8db9b17018..3c6e9fffa1 100644 +--- a/dpdk/lib/regexdev/version.map ++++ b/dpdk/lib/regexdev/version.map +@@ -1,6 +1,7 @@ + EXPERIMENTAL { + global: + ++ rte_regex_devices; + rte_regexdev_attr_get; + rte_regexdev_attr_set; + rte_regexdev_close; +@@ -11,6 +12,8 @@ EXPERIMENTAL { + rte_regexdev_enqueue_burst; + rte_regexdev_get_dev_id; + rte_regexdev_info_get; ++ rte_regexdev_is_valid_dev; ++ rte_regexdev_logtype; + rte_regexdev_queue_pair_setup; + rte_regexdev_rule_db_compile_activate; + rte_regexdev_rule_db_export; +@@ -23,4 +26,12 @@ EXPERIMENTAL { + rte_regexdev_xstats_get; + rte_regexdev_xstats_names_get; + rte_regexdev_xstats_reset; ++ ++ local: *; ++}; ++ ++INTERNAL { ++ rte_regexdev_get_device_by_name; ++ rte_regexdev_register; ++ rte_regexdev_unregister; + }; +diff --git a/dpdk/lib/rib/rte_rib.c b/dpdk/lib/rib/rte_rib.c +index 6c29e1c49a..1a4b10d728 100644 +--- a/dpdk/lib/rib/rte_rib.c ++++ b/dpdk/lib/rib/rte_rib.c +@@ -73,6 +73,8 @@ is_covered(uint32_t ip1, uint32_t ip2, uint8_t depth) + static inline struct rte_rib_node * + get_nxt_node(struct rte_rib_node *node, uint32_t ip) + { ++ if (node->depth == RIB_MAXDEPTH) ++ return NULL; + return (ip & (1 << (31 - node->depth))) ? node->right : node->left; + } + +diff --git a/dpdk/lib/rib/rte_rib6.h b/dpdk/lib/rib/rte_rib6.h +index 6f532265c6..d52b0b05cc 100644 +--- a/dpdk/lib/rib/rte_rib6.h ++++ b/dpdk/lib/rib/rte_rib6.h +@@ -40,12 +40,12 @@ struct rte_rib6_node; + /** RIB configuration structure */ + struct rte_rib6_conf { + /** +- * Size of extension block inside rte_rib_node. ++ * Size of extension block inside rte_rib6_node. + * This space could be used to store additional user + * defined data. + */ + size_t ext_sz; +- /* size of rte_rib_node's pool */ ++ /* size of rte_rib6_node's pool */ + int max_nodes; + }; + +@@ -307,7 +307,7 @@ rte_rib6_create(const char *name, int socket_id, + * Find an existing RIB object and return a pointer to it. + * + * @param name +- * Name of the rib object as passed to rte_rib_create() ++ * Name of the rib object as passed to rte_rib6_create() + * @return + * Pointer to RIB object on success + * NULL otherwise with rte_errno indicating reason for failure. +diff --git a/dpdk/lib/ring/rte_ring.c b/dpdk/lib/ring/rte_ring.c +index f17bd966be..6a94a038c4 100644 +--- a/dpdk/lib/ring/rte_ring.c ++++ b/dpdk/lib/ring/rte_ring.c +@@ -75,7 +75,7 @@ rte_ring_get_memsize_elem(unsigned int esize, unsigned int count) + return -EINVAL; + } + +- sz = sizeof(struct rte_ring) + count * esize; ++ sz = sizeof(struct rte_ring) + (ssize_t)count * esize; + sz = RTE_ALIGN(sz, RTE_CACHE_LINE_SIZE); + return sz; + } +@@ -267,7 +267,7 @@ rte_ring_create_elem(const char *name, unsigned int esize, unsigned int count, + + ring_size = rte_ring_get_memsize_elem(esize, count); + if (ring_size < 0) { +- rte_errno = ring_size; ++ rte_errno = -ring_size; + return NULL; + } + +diff --git a/dpdk/lib/ring/rte_ring_core.h b/dpdk/lib/ring/rte_ring_core.h +index 46ad584f9c..1252ca9546 100644 +--- a/dpdk/lib/ring/rte_ring_core.h ++++ b/dpdk/lib/ring/rte_ring_core.h +@@ -12,7 +12,7 @@ + + /** + * @file +- * This file contains definion of RTE ring structure itself, ++ * This file contains definition of RTE ring structure itself, + * init flags and some related macros. + * For majority of DPDK entities, it is not recommended to include + * this file directly, use include or +diff --git a/dpdk/lib/sched/rte_pie.c b/dpdk/lib/sched/rte_pie.c +index 934e9aee50..79db6e96b1 100644 +--- a/dpdk/lib/sched/rte_pie.c ++++ b/dpdk/lib/sched/rte_pie.c +@@ -3,6 +3,7 @@ + */ + + #include ++#include + + #include "rte_pie.h" + #include +@@ -17,26 +18,11 @@ int + rte_pie_rt_data_init(struct rte_pie *pie) + { + if (pie == NULL) { +- /* Allocate memory to use the PIE data structure */ +- pie = rte_malloc(NULL, sizeof(struct rte_pie), 0); +- +- if (pie == NULL) +- RTE_LOG(ERR, SCHED, "%s: Memory allocation fails\n", __func__); +- +- return -1; ++ RTE_LOG(ERR, SCHED, "%s: Invalid addr for pie\n", __func__); ++ return -EINVAL; + } + +- pie->active = 0; +- pie->in_measurement = 0; +- pie->departed_bytes_count = 0; +- pie->start_measurement = 0; +- pie->last_measurement = 0; +- pie->qlen = 0; +- pie->avg_dq_time = 0; +- pie->burst_allowance = 0; +- pie->qdelay_old = 0; +- pie->drop_prob = 0; +- pie->accu_prob = 0; ++ memset(pie, 0, sizeof(*pie)); + + return 0; + } +diff --git a/dpdk/lib/sched/rte_pie.h b/dpdk/lib/sched/rte_pie.h +index dfdf572311..3e2c1ef467 100644 +--- a/dpdk/lib/sched/rte_pie.h ++++ b/dpdk/lib/sched/rte_pie.h +@@ -218,7 +218,7 @@ _rte_pie_drop(const struct rte_pie_config *pie_cfg, + struct rte_pie *pie) + { + uint64_t rand_value; +- double qdelay = pie_cfg->qdelay_ref * 0.5; ++ uint64_t qdelay = pie_cfg->qdelay_ref / 2; + + /* PIE is active but the queue is not congested: return 0 */ + if (((pie->qdelay_old < qdelay) && (pie->drop_prob < 0.2)) || +@@ -252,7 +252,7 @@ _rte_pie_drop(const struct rte_pie_config *pie_cfg, + } + + /** +- * @brief Decides if new packet should be enqeued or dropped for non-empty queue ++ * @brief Decides if new packet should be enqueued or dropped for non-empty queue + * + * @param pie_cfg [in] config pointer to a PIE configuration parameter structure + * @param pie [in,out] data pointer to PIE runtime data +@@ -319,7 +319,7 @@ rte_pie_enqueue_nonempty(const struct rte_pie_config *pie_cfg, + } + + /** +- * @brief Decides if new packet should be enqeued or dropped ++ * @brief Decides if new packet should be enqueued or dropped + * Updates run time data and gives verdict whether to enqueue or drop the packet. + * + * @param pie_cfg [in] config pointer to a PIE configuration parameter structure +@@ -330,7 +330,7 @@ rte_pie_enqueue_nonempty(const struct rte_pie_config *pie_cfg, + * + * @return Operation status + * @retval 0 enqueue the packet +- * @retval 1 drop the packet based on drop probility criteria ++ * @retval 1 drop the packet based on drop probability criteria + */ + static inline int + __rte_experimental +diff --git a/dpdk/lib/sched/rte_red.h b/dpdk/lib/sched/rte_red.h +index 36273cac64..f5843dab1b 100644 +--- a/dpdk/lib/sched/rte_red.h ++++ b/dpdk/lib/sched/rte_red.h +@@ -303,7 +303,7 @@ __rte_red_drop(const struct rte_red_config *red_cfg, struct rte_red *red) + } + + /** +- * @brief Decides if new packet should be enqeued or dropped in queue non-empty case ++ * @brief Decides if new packet should be enqueued or dropped in queue non-empty case + * + * @param red_cfg [in] config pointer to a RED configuration parameter structure + * @param red [in,out] data pointer to RED runtime data +@@ -361,7 +361,7 @@ rte_red_enqueue_nonempty(const struct rte_red_config *red_cfg, + } + + /** +- * @brief Decides if new packet should be enqeued or dropped ++ * @brief Decides if new packet should be enqueued or dropped + * Updates run time data based on new queue size value. + * Based on new queue average and RED configuration parameters + * gives verdict whether to enqueue or drop the packet. +diff --git a/dpdk/lib/sched/rte_sched.c b/dpdk/lib/sched/rte_sched.c +index ed44808f7b..62b3d2e315 100644 +--- a/dpdk/lib/sched/rte_sched.c ++++ b/dpdk/lib/sched/rte_sched.c +@@ -239,7 +239,7 @@ struct rte_sched_port { + int socket; + + /* Timing */ +- uint64_t time_cpu_cycles; /* Current CPU time measured in CPU cyles */ ++ uint64_t time_cpu_cycles; /* Current CPU time measured in CPU cycles */ + uint64_t time_cpu_bytes; /* Current CPU time measured in bytes */ + uint64_t time; /* Current NIC TX time measured in bytes */ + struct rte_reciprocal inv_cycles_per_byte; /* CPU cycles per byte */ +diff --git a/dpdk/lib/sched/rte_sched.h b/dpdk/lib/sched/rte_sched.h +index 484dbdcc3d..3c625ba169 100644 +--- a/dpdk/lib/sched/rte_sched.h ++++ b/dpdk/lib/sched/rte_sched.h +@@ -360,7 +360,7 @@ rte_sched_subport_pipe_profile_add(struct rte_sched_port *port, + * + * Hierarchical scheduler subport bandwidth profile add + * Note that this function is safe to use in runtime for adding new +- * subport bandwidth profile as it doesn't have any impact on hiearchical ++ * subport bandwidth profile as it doesn't have any impact on hierarchical + * structure of the scheduler. + * @param port + * Handle to port scheduler instance +diff --git a/dpdk/lib/security/rte_security.h b/dpdk/lib/security/rte_security.h +index 1228b6c8b1..1a15e95267 100644 +--- a/dpdk/lib/security/rte_security.h ++++ b/dpdk/lib/security/rte_security.h +@@ -301,9 +301,9 @@ struct rte_security_ipsec_lifetime { + uint64_t bytes_soft_limit; + /**< Soft expiry limit in bytes */ + uint64_t packets_hard_limit; +- /**< Soft expiry limit in number of packets */ ++ /**< Hard expiry limit in number of packets */ + uint64_t bytes_hard_limit; +- /**< Soft expiry limit in bytes */ ++ /**< Hard expiry limit in bytes */ + }; + + /** +diff --git a/dpdk/lib/stack/meson.build b/dpdk/lib/stack/meson.build +index 2f53f49677..18177a742f 100644 +--- a/dpdk/lib/stack/meson.build ++++ b/dpdk/lib/stack/meson.build +@@ -9,4 +9,5 @@ indirect_headers += files( + 'rte_stack_lf.h', + 'rte_stack_lf_generic.h', + 'rte_stack_lf_c11.h', ++ 'rte_stack_lf_stubs.h', + ) +diff --git a/dpdk/lib/table/rte_swx_table.h b/dpdk/lib/table/rte_swx_table.h +index f93e5f3f95..c1383c2e57 100644 +--- a/dpdk/lib/table/rte_swx_table.h ++++ b/dpdk/lib/table/rte_swx_table.h +@@ -216,7 +216,7 @@ typedef int + * operations into the same table. + * + * The typical reason an implementation may choose to split the table lookup +- * operation into multiple steps is to hide the latency of the inherrent memory ++ * operation into multiple steps is to hide the latency of the inherent memory + * read operations: before a read operation with the source data likely not in + * the CPU cache, the source data prefetch is issued and the table lookup + * operation is postponed in favor of some other unrelated work, which the CPU +diff --git a/dpdk/lib/table/rte_swx_table_selector.h b/dpdk/lib/table/rte_swx_table_selector.h +index 62988d2856..05863cc90b 100644 +--- a/dpdk/lib/table/rte_swx_table_selector.h ++++ b/dpdk/lib/table/rte_swx_table_selector.h +@@ -155,7 +155,7 @@ rte_swx_table_selector_group_set(void *table, + * mechanism allows for multiple concurrent select operations into the same table. + * + * The typical reason an implementation may choose to split the operation into multiple steps is to +- * hide the latency of the inherrent memory read operations: before a read operation with the ++ * hide the latency of the inherent memory read operations: before a read operation with the + * source data likely not in the CPU cache, the source data prefetch is issued and the operation is + * postponed in favor of some other unrelated work, which the CPU executes in parallel with the + * source data being fetched into the CPU cache; later on, the operation is resumed, this time with +diff --git a/dpdk/lib/table/rte_table_hash_func.h b/dpdk/lib/table/rte_table_hash_func.h +index c4c35cc06a..a962ec2f68 100644 +--- a/dpdk/lib/table/rte_table_hash_func.h ++++ b/dpdk/lib/table/rte_table_hash_func.h +@@ -58,8 +58,8 @@ static inline uint64_t + rte_table_hash_crc_key8(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) + { +- uint64_t *k = key; +- uint64_t *m = mask; ++ uint64_t *k = (uint64_t *)key; ++ uint64_t *m = (uint64_t *)mask; + uint64_t crc0; + + crc0 = rte_crc32_u64(seed, k[0] & m[0]); +@@ -72,8 +72,8 @@ static inline uint64_t + rte_table_hash_crc_key16(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) + { +- uint64_t *k = key; +- uint64_t *m = mask; ++ uint64_t *k = (uint64_t *)key; ++ uint64_t *m = (uint64_t *)mask; + uint64_t k0, crc0, crc1; + + k0 = k[0] & m[0]; +@@ -91,8 +91,8 @@ static inline uint64_t + rte_table_hash_crc_key24(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) + { +- uint64_t *k = key; +- uint64_t *m = mask; ++ uint64_t *k = (uint64_t *)key; ++ uint64_t *m = (uint64_t *)mask; + uint64_t k0, k2, crc0, crc1; + + k0 = k[0] & m[0]; +@@ -113,8 +113,8 @@ static inline uint64_t + rte_table_hash_crc_key32(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) + { +- uint64_t *k = key; +- uint64_t *m = mask; ++ uint64_t *k = (uint64_t *)key; ++ uint64_t *m = (uint64_t *)mask; + uint64_t k0, k2, crc0, crc1, crc2, crc3; + + k0 = k[0] & m[0]; +@@ -139,8 +139,8 @@ static inline uint64_t + rte_table_hash_crc_key40(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) + { +- uint64_t *k = key; +- uint64_t *m = mask; ++ uint64_t *k = (uint64_t *)key; ++ uint64_t *m = (uint64_t *)mask; + uint64_t k0, k2, crc0, crc1, crc2, crc3; + + k0 = k[0] & m[0]; +@@ -165,8 +165,8 @@ static inline uint64_t + rte_table_hash_crc_key48(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) + { +- uint64_t *k = key; +- uint64_t *m = mask; ++ uint64_t *k = (uint64_t *)key; ++ uint64_t *m = (uint64_t *)mask; + uint64_t k0, k2, k5, crc0, crc1, crc2, crc3; + + k0 = k[0] & m[0]; +@@ -192,8 +192,8 @@ static inline uint64_t + rte_table_hash_crc_key56(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) + { +- uint64_t *k = key; +- uint64_t *m = mask; ++ uint64_t *k = (uint64_t *)key; ++ uint64_t *m = (uint64_t *)mask; + uint64_t k0, k2, k5, crc0, crc1, crc2, crc3, crc4, crc5; + + k0 = k[0] & m[0]; +@@ -222,8 +222,8 @@ static inline uint64_t + rte_table_hash_crc_key64(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) + { +- uint64_t *k = key; +- uint64_t *m = mask; ++ uint64_t *k = (uint64_t *)key; ++ uint64_t *m = (uint64_t *)mask; + uint64_t k0, k2, k5, crc0, crc1, crc2, crc3, crc4, crc5; + + k0 = k[0] & m[0]; +diff --git a/dpdk/lib/telemetry/rte_telemetry.h b/dpdk/lib/telemetry/rte_telemetry.h +index 7bca8a9a49..3372b32f38 100644 +--- a/dpdk/lib/telemetry/rte_telemetry.h ++++ b/dpdk/lib/telemetry/rte_telemetry.h +@@ -9,6 +9,10 @@ + #ifndef _RTE_TELEMETRY_H_ + #define _RTE_TELEMETRY_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + /** Maximum length for string used in object. */ + #define RTE_TEL_MAX_STRING_LEN 128 + /** Maximum length of string. */ +@@ -294,4 +298,8 @@ rte_tel_data_alloc(void); + void + rte_tel_data_free(struct rte_tel_data *data); + ++#ifdef __cplusplus ++} ++#endif ++ + #endif +diff --git a/dpdk/lib/telemetry/telemetry.c b/dpdk/lib/telemetry/telemetry.c +index a7483167d4..e5ccfe47f7 100644 +--- a/dpdk/lib/telemetry/telemetry.c ++++ b/dpdk/lib/telemetry/telemetry.c +@@ -534,7 +534,7 @@ telemetry_legacy_init(void) + } + rc = pthread_create(&t_old, NULL, socket_listener, &v1_socket); + if (rc != 0) { +- TMTY_LOG(ERR, "Error with create legcay socket thread: %s\n", ++ TMTY_LOG(ERR, "Error with create legacy socket thread: %s\n", + strerror(rc)); + close(v1_socket.sock); + v1_socket.sock = -1; +diff --git a/dpdk/lib/telemetry/telemetry_json.h b/dpdk/lib/telemetry/telemetry_json.h +index f02a12f5b0..db70690274 100644 +--- a/dpdk/lib/telemetry/telemetry_json.h ++++ b/dpdk/lib/telemetry/telemetry_json.h +@@ -23,7 +23,7 @@ + /** + * @internal + * Copies a value into a buffer if the buffer has enough available space. +- * Nothing written to buffer if an overflow ocurs. ++ * Nothing written to buffer if an overflow occurs. + * This function is not for use for values larger than given buffer length. + */ + __rte_format_printf(3, 4) +diff --git a/dpdk/lib/vhost/rte_vdpa.h b/dpdk/lib/vhost/rte_vdpa.h +index 1437f400bf..6ac85d1bbf 100644 +--- a/dpdk/lib/vhost/rte_vdpa.h ++++ b/dpdk/lib/vhost/rte_vdpa.h +@@ -5,6 +5,10 @@ + #ifndef _RTE_VDPA_H_ + #define _RTE_VDPA_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + /** + * @file + * +@@ -183,4 +187,9 @@ rte_vdpa_get_stats(struct rte_vdpa_device *dev, uint16_t qid, + */ + int + rte_vdpa_reset_stats(struct rte_vdpa_device *dev, uint16_t qid); ++ ++#ifdef __cplusplus ++} ++#endif ++ + #endif /* _RTE_VDPA_H_ */ +diff --git a/dpdk/lib/vhost/rte_vhost.h b/dpdk/lib/vhost/rte_vhost.h +index b454c05868..2acb31df2d 100644 +--- a/dpdk/lib/vhost/rte_vhost.h ++++ b/dpdk/lib/vhost/rte_vhost.h +@@ -21,10 +21,12 @@ + extern "C" { + #endif + ++#ifndef __cplusplus + /* These are not C++-aware. */ + #include + #include + #include ++#endif + + #define RTE_VHOST_USER_CLIENT (1ULL << 0) + #define RTE_VHOST_USER_NO_RECONNECT (1ULL << 1) +diff --git a/dpdk/lib/vhost/rte_vhost_async.h b/dpdk/lib/vhost/rte_vhost_async.h +index a87ea6ba37..d20152ca7a 100644 +--- a/dpdk/lib/vhost/rte_vhost_async.h ++++ b/dpdk/lib/vhost/rte_vhost_async.h +@@ -5,6 +5,10 @@ + #ifndef _RTE_VHOST_ASYNC_H_ + #define _RTE_VHOST_ASYNC_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + #include "rte_vhost.h" + + /** +@@ -242,4 +246,8 @@ __rte_experimental + uint16_t rte_vhost_clear_queue_thread_unsafe(int vid, uint16_t queue_id, + struct rte_mbuf **pkts, uint16_t count); + ++#ifdef __cplusplus ++} ++#endif ++ + #endif /* _RTE_VHOST_ASYNC_H_ */ +diff --git a/dpdk/lib/vhost/rte_vhost_crypto.h b/dpdk/lib/vhost/rte_vhost_crypto.h +index f54d731139..b49e389579 100644 +--- a/dpdk/lib/vhost/rte_vhost_crypto.h ++++ b/dpdk/lib/vhost/rte_vhost_crypto.h +@@ -5,6 +5,10 @@ + #ifndef _VHOST_CRYPTO_H_ + #define _VHOST_CRYPTO_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + #include + + #include +@@ -132,4 +136,8 @@ uint16_t + rte_vhost_crypto_finalize_requests(struct rte_crypto_op **ops, + uint16_t nb_ops, int *callfds, uint16_t *nb_callfds); + ++#ifdef __cplusplus ++} ++#endif ++ + #endif /**< _VHOST_CRYPTO_H_ */ +diff --git a/dpdk/lib/vhost/socket.c b/dpdk/lib/vhost/socket.c +index 82963c1e6d..33f54a779b 100644 +--- a/dpdk/lib/vhost/socket.c ++++ b/dpdk/lib/vhost/socket.c +@@ -501,7 +501,7 @@ vhost_user_reconnect_init(void) + + ret = pthread_mutex_init(&reconn_list.mutex, NULL); + if (ret < 0) { +- VHOST_LOG_CONFIG(ERR, "failed to initialize mutex"); ++ VHOST_LOG_CONFIG(ERR, "failed to initialize mutex\n"); + return ret; + } + TAILQ_INIT(&reconn_list.head); +@@ -509,10 +509,10 @@ vhost_user_reconnect_init(void) + ret = rte_ctrl_thread_create(&reconn_tid, "vhost_reconn", NULL, + vhost_user_client_reconnect, NULL); + if (ret != 0) { +- VHOST_LOG_CONFIG(ERR, "failed to create reconnect thread"); ++ VHOST_LOG_CONFIG(ERR, "failed to create reconnect thread\n"); + if (pthread_mutex_destroy(&reconn_list.mutex)) { + VHOST_LOG_CONFIG(ERR, +- "failed to destroy reconnect mutex"); ++ "failed to destroy reconnect mutex\n"); + } + } + +@@ -1147,7 +1147,7 @@ rte_vhost_driver_start(const char *path) + &vhost_user.fdset); + if (ret != 0) { + VHOST_LOG_CONFIG(ERR, +- "failed to create fdset handling thread"); ++ "failed to create fdset handling thread\n"); + + fdset_pipe_uninit(&vhost_user.fdset); + return -1; +diff --git a/dpdk/lib/vhost/vdpa_driver.h b/dpdk/lib/vhost/vdpa_driver.h +index fc2d6acedd..7ba9e28e57 100644 +--- a/dpdk/lib/vhost/vdpa_driver.h ++++ b/dpdk/lib/vhost/vdpa_driver.h +@@ -5,6 +5,10 @@ + #ifndef _VDPA_DRIVER_H_ + #define _VDPA_DRIVER_H_ + ++#ifdef __cplusplus ++extern "C" { ++#endif ++ + #include + + #include +@@ -141,4 +145,8 @@ __rte_internal + int + rte_vdpa_relay_vring_used(int vid, uint16_t qid, void *vring_m); + ++#ifdef __cplusplus ++} ++#endif ++ + #endif /* _VDPA_DRIVER_H_ */ +diff --git a/dpdk/lib/vhost/version.map b/dpdk/lib/vhost/version.map +index a7ef7f1976..0f315ed2a5 100644 +--- a/dpdk/lib/vhost/version.map ++++ b/dpdk/lib/vhost/version.map +@@ -87,7 +87,7 @@ EXPERIMENTAL { + }; + + INTERNAL { +- global; ++ global: + + rte_vdpa_register_device; + rte_vdpa_relay_vring_used; +diff --git a/dpdk/lib/vhost/vhost.c b/dpdk/lib/vhost/vhost.c +index 13a9bb9dd1..24f94495c6 100644 +--- a/dpdk/lib/vhost/vhost.c ++++ b/dpdk/lib/vhost/vhost.c +@@ -1299,11 +1299,15 @@ rte_vhost_vring_call(int vid, uint16_t vring_idx) + if (!vq) + return -1; + ++ rte_spinlock_lock(&vq->access_lock); ++ + if (vq_is_packed(dev)) + vhost_vring_call_packed(dev, vq); + else + vhost_vring_call_split(dev, vq); + ++ rte_spinlock_unlock(&vq->access_lock); ++ + return 0; + } + +@@ -1779,26 +1783,22 @@ rte_vhost_async_channel_unregister(int vid, uint16_t queue_id) + if (vq == NULL) + return ret; + +- ret = 0; +- +- if (!vq->async) +- return ret; +- + if (!rte_spinlock_trylock(&vq->access_lock)) { + VHOST_LOG_CONFIG(ERR, "Failed to unregister async channel. " + "virt queue busy.\n"); +- return -1; ++ return ret; + } + +- if (vq->async->pkts_inflight_n) { ++ if (!vq->async) { ++ ret = 0; ++ } else if (vq->async->pkts_inflight_n) { + VHOST_LOG_CONFIG(ERR, "Failed to unregister async channel. " + "async inflight packets must be completed before unregistration.\n"); +- ret = -1; +- goto out; ++ } else { ++ vhost_free_async_mem(vq); ++ ret = 0; + } + +- vhost_free_async_mem(vq); +-out: + rte_spinlock_unlock(&vq->access_lock); + + return ret; +@@ -1853,16 +1853,15 @@ rte_vhost_async_get_inflight(int vid, uint16_t queue_id) + if (vq == NULL) + return ret; + +- if (!vq->async) +- return ret; +- + if (!rte_spinlock_trylock(&vq->access_lock)) { + VHOST_LOG_CONFIG(DEBUG, "Failed to check in-flight packets. " + "virt queue busy.\n"); + return ret; + } + +- ret = vq->async->pkts_inflight_n; ++ if (vq->async) ++ ret = vq->async->pkts_inflight_n; ++ + rte_spinlock_unlock(&vq->access_lock); + + return ret; +diff --git a/dpdk/lib/vhost/vhost.h b/dpdk/lib/vhost/vhost.h +index 7085e0885c..d4586f3341 100644 +--- a/dpdk/lib/vhost/vhost.h ++++ b/dpdk/lib/vhost/vhost.h +@@ -354,7 +354,8 @@ struct vring_packed_desc_event { + + struct guest_page { + uint64_t guest_phys_addr; +- uint64_t host_phys_addr; ++ uint64_t host_iova; ++ uint64_t host_user_addr; + uint64_t size; + }; + +@@ -587,6 +588,20 @@ static __rte_always_inline int guest_page_addrcmp(const void *p1, + return 0; + } + ++static __rte_always_inline int guest_page_rangecmp(const void *p1, const void *p2) ++{ ++ const struct guest_page *page1 = (const struct guest_page *)p1; ++ const struct guest_page *page2 = (const struct guest_page *)p2; ++ ++ if (page1->guest_phys_addr >= page2->guest_phys_addr) { ++ if (page1->guest_phys_addr < page2->guest_phys_addr + page2->size) ++ return 0; ++ else ++ return 1; ++ } else ++ return -1; ++} ++ + static __rte_always_inline rte_iova_t + gpa_to_first_hpa(struct virtio_net *dev, uint64_t gpa, + uint64_t gpa_size, uint64_t *hpa_size) +@@ -597,20 +612,20 @@ gpa_to_first_hpa(struct virtio_net *dev, uint64_t gpa, + + *hpa_size = gpa_size; + if (dev->nr_guest_pages >= VHOST_BINARY_SEARCH_THRESH) { +- key.guest_phys_addr = gpa & ~(dev->guest_pages[0].size - 1); ++ key.guest_phys_addr = gpa; + page = bsearch(&key, dev->guest_pages, dev->nr_guest_pages, +- sizeof(struct guest_page), guest_page_addrcmp); ++ sizeof(struct guest_page), guest_page_rangecmp); + if (page) { + if (gpa + gpa_size <= + page->guest_phys_addr + page->size) { + return gpa - page->guest_phys_addr + +- page->host_phys_addr; ++ page->host_iova; + } else if (gpa < page->guest_phys_addr + + page->size) { + *hpa_size = page->guest_phys_addr + + page->size - gpa; + return gpa - page->guest_phys_addr + +- page->host_phys_addr; ++ page->host_iova; + } + } + } else { +@@ -621,13 +636,13 @@ gpa_to_first_hpa(struct virtio_net *dev, uint64_t gpa, + if (gpa + gpa_size <= + page->guest_phys_addr + page->size) { + return gpa - page->guest_phys_addr + +- page->host_phys_addr; ++ page->host_iova; + } else if (gpa < page->guest_phys_addr + + page->size) { + *hpa_size = page->guest_phys_addr + + page->size - gpa; + return gpa - page->guest_phys_addr + +- page->host_phys_addr; ++ page->host_iova; + } + } + } +diff --git a/dpdk/lib/vhost/vhost_crypto.c b/dpdk/lib/vhost/vhost_crypto.c +index 926b5c0bd9..7d1d6a1861 100644 +--- a/dpdk/lib/vhost/vhost_crypto.c ++++ b/dpdk/lib/vhost/vhost_crypto.c +@@ -565,94 +565,57 @@ get_data_ptr(struct vhost_crypto_data_req *vc_req, + return data; + } + +-static __rte_always_inline int +-copy_data(void *dst_data, struct vhost_crypto_data_req *vc_req, +- struct vhost_crypto_desc *head, +- struct vhost_crypto_desc **cur_desc, +- uint32_t size, uint32_t max_n_descs) ++static __rte_always_inline uint32_t ++copy_data_from_desc(void *dst, struct vhost_crypto_data_req *vc_req, ++ struct vhost_crypto_desc *desc, uint32_t size) + { +- struct vhost_crypto_desc *desc = *cur_desc; +- uint64_t remain, addr, dlen, len; +- uint32_t to_copy; +- uint8_t *data = dst_data; +- uint8_t *src; +- int left = size; +- +- to_copy = RTE_MIN(desc->len, (uint32_t)left); +- dlen = to_copy; +- src = IOVA_TO_VVA(uint8_t *, vc_req, desc->addr, &dlen, +- VHOST_ACCESS_RO); +- if (unlikely(!src || !dlen)) +- return -1; ++ uint64_t remain; ++ uint64_t addr; ++ ++ remain = RTE_MIN(desc->len, size); ++ addr = desc->addr; ++ do { ++ uint64_t len; ++ void *src; ++ ++ len = remain; ++ src = IOVA_TO_VVA(void *, vc_req, addr, &len, VHOST_ACCESS_RO); ++ if (unlikely(src == NULL || len == 0)) ++ return 0; + +- rte_memcpy((uint8_t *)data, src, dlen); +- data += dlen; ++ rte_memcpy(dst, src, len); ++ remain -= len; ++ /* cast is needed for 32-bit architecture */ ++ dst = RTE_PTR_ADD(dst, (size_t)len); ++ addr += len; ++ } while (unlikely(remain != 0)); + +- if (unlikely(dlen < to_copy)) { +- remain = to_copy - dlen; +- addr = desc->addr + dlen; ++ return RTE_MIN(desc->len, size); ++} + +- while (remain) { +- len = remain; +- src = IOVA_TO_VVA(uint8_t *, vc_req, addr, &len, +- VHOST_ACCESS_RO); +- if (unlikely(!src || !len)) { +- VC_LOG_ERR("Failed to map descriptor"); +- return -1; +- } + +- rte_memcpy(data, src, len); +- addr += len; +- remain -= len; +- data += len; +- } +- } ++static __rte_always_inline int ++copy_data(void *data, struct vhost_crypto_data_req *vc_req, ++ struct vhost_crypto_desc *head, struct vhost_crypto_desc **cur_desc, ++ uint32_t size, uint32_t max_n_descs) ++{ ++ struct vhost_crypto_desc *desc = *cur_desc; ++ uint32_t left = size; + +- left -= to_copy; ++ do { ++ uint32_t copied; + +- while (desc >= head && desc - head < (int)max_n_descs && left) { +- desc++; +- to_copy = RTE_MIN(desc->len, (uint32_t)left); +- dlen = to_copy; +- src = IOVA_TO_VVA(uint8_t *, vc_req, desc->addr, &dlen, +- VHOST_ACCESS_RO); +- if (unlikely(!src || !dlen)) { +- VC_LOG_ERR("Failed to map descriptor"); ++ copied = copy_data_from_desc(data, vc_req, desc, left); ++ if (copied == 0) + return -1; +- } +- +- rte_memcpy(data, src, dlen); +- data += dlen; +- +- if (unlikely(dlen < to_copy)) { +- remain = to_copy - dlen; +- addr = desc->addr + dlen; +- +- while (remain) { +- len = remain; +- src = IOVA_TO_VVA(uint8_t *, vc_req, addr, &len, +- VHOST_ACCESS_RO); +- if (unlikely(!src || !len)) { +- VC_LOG_ERR("Failed to map descriptor"); +- return -1; +- } +- +- rte_memcpy(data, src, len); +- addr += len; +- remain -= len; +- data += len; +- } +- } +- +- left -= to_copy; +- } ++ left -= copied; ++ data = RTE_PTR_ADD(data, copied); ++ } while (left != 0 && ++desc < head + max_n_descs); + +- if (unlikely(left > 0)) { +- VC_LOG_ERR("Incorrect virtio descriptor"); ++ if (unlikely(left != 0)) + return -1; +- } + +- if (unlikely(desc - head == (int)max_n_descs)) ++ if (unlikely(desc == head + max_n_descs)) + *cur_desc = NULL; + else + *cur_desc = desc + 1; +@@ -852,6 +815,7 @@ prepare_sym_cipher_op(struct vhost_crypto *vcrypto, struct rte_crypto_op *op, + /* iv */ + if (unlikely(copy_data(iv_data, vc_req, head, &desc, + cipher->para.iv_len, max_n_descs))) { ++ VC_LOG_ERR("Incorrect virtio descriptor"); + ret = VIRTIO_CRYPTO_BADMSG; + goto error_exit; + } +@@ -883,6 +847,7 @@ prepare_sym_cipher_op(struct vhost_crypto *vcrypto, struct rte_crypto_op *op, + if (unlikely(copy_data(rte_pktmbuf_mtod(m_src, uint8_t *), + vc_req, head, &desc, cipher->para.src_data_len, + max_n_descs) < 0)) { ++ VC_LOG_ERR("Incorrect virtio descriptor"); + ret = VIRTIO_CRYPTO_BADMSG; + goto error_exit; + } +@@ -1006,6 +971,7 @@ prepare_sym_chain_op(struct vhost_crypto *vcrypto, struct rte_crypto_op *op, + /* iv */ + if (unlikely(copy_data(iv_data, vc_req, head, &desc, + chain->para.iv_len, max_n_descs) < 0)) { ++ VC_LOG_ERR("Incorrect virtio descriptor"); + ret = VIRTIO_CRYPTO_BADMSG; + goto error_exit; + } +@@ -1037,6 +1003,7 @@ prepare_sym_chain_op(struct vhost_crypto *vcrypto, struct rte_crypto_op *op, + if (unlikely(copy_data(rte_pktmbuf_mtod(m_src, uint8_t *), + vc_req, head, &desc, chain->para.src_data_len, + max_n_descs) < 0)) { ++ VC_LOG_ERR("Incorrect virtio descriptor"); + ret = VIRTIO_CRYPTO_BADMSG; + goto error_exit; + } +@@ -1121,6 +1088,7 @@ prepare_sym_chain_op(struct vhost_crypto *vcrypto, struct rte_crypto_op *op, + if (unlikely(copy_data(digest_addr, vc_req, head, &digest_desc, + chain->para.hash_result_len, + max_n_descs) < 0)) { ++ VC_LOG_ERR("Incorrect virtio descriptor"); + ret = VIRTIO_CRYPTO_BADMSG; + goto error_exit; + } +diff --git a/dpdk/lib/vhost/vhost_user.c b/dpdk/lib/vhost/vhost_user.c +index a781346c4d..2b45e35d4a 100644 +--- a/dpdk/lib/vhost/vhost_user.c ++++ b/dpdk/lib/vhost/vhost_user.c +@@ -143,57 +143,59 @@ get_blk_size(int fd) + return ret == -1 ? (uint64_t)-1 : (uint64_t)stat.st_blksize; + } + +-static int +-async_dma_map(struct rte_vhost_mem_region *region, bool do_map) ++static void ++async_dma_map(struct virtio_net *dev, bool do_map) + { +- uint64_t host_iova; + int ret = 0; ++ uint32_t i; ++ struct guest_page *page; + +- host_iova = rte_mem_virt2iova((void *)(uintptr_t)region->host_user_addr); + if (do_map) { +- /* Add mapped region into the default container of DPDK. */ +- ret = rte_vfio_container_dma_map(RTE_VFIO_DEFAULT_CONTAINER_FD, +- region->host_user_addr, +- host_iova, +- region->size); +- if (ret) { +- /* +- * DMA device may bind with kernel driver, in this case, +- * we don't need to program IOMMU manually. However, if no +- * device is bound with vfio/uio in DPDK, and vfio kernel +- * module is loaded, the API will still be called and return +- * with ENODEV/ENOSUP. +- * +- * DPDK vfio only returns ENODEV/ENOSUP in very similar +- * situations(vfio either unsupported, or supported +- * but no devices found). Either way, no mappings could be +- * performed. We treat it as normal case in async path. +- */ +- if (rte_errno == ENODEV || rte_errno == ENOTSUP) +- return 0; +- +- VHOST_LOG_CONFIG(ERR, "DMA engine map failed\n"); +- /* DMA mapping errors won't stop VHST_USER_SET_MEM_TABLE. */ +- return 0; ++ for (i = 0; i < dev->nr_guest_pages; i++) { ++ page = &dev->guest_pages[i]; ++ ret = rte_vfio_container_dma_map(RTE_VFIO_DEFAULT_CONTAINER_FD, ++ page->host_user_addr, ++ page->host_iova, ++ page->size); ++ if (ret) { ++ /* ++ * DMA device may bind with kernel driver, in this case, ++ * we don't need to program IOMMU manually. However, if no ++ * device is bound with vfio/uio in DPDK, and vfio kernel ++ * module is loaded, the API will still be called and return ++ * with ENODEV. ++ * ++ * DPDK vfio only returns ENODEV in very similar situations ++ * (vfio either unsupported, or supported but no devices found). ++ * Either way, no mappings could be performed. We treat it as ++ * normal case in async path. This is a workaround. ++ */ ++ if (rte_errno == ENODEV) ++ return; ++ ++ /* DMA mapping errors won't stop VHOST_USER_SET_MEM_TABLE. */ ++ VHOST_LOG_CONFIG(ERR, "(%s) DMA engine map failed\n", ++ dev->ifname); ++ } + } + + } else { +- /* Remove mapped region from the default container of DPDK. */ +- ret = rte_vfio_container_dma_unmap(RTE_VFIO_DEFAULT_CONTAINER_FD, +- region->host_user_addr, +- host_iova, +- region->size); +- if (ret) { +- /* like DMA map, ignore the kernel driver case when unmap. */ +- if (rte_errno == EINVAL) +- return 0; +- +- VHOST_LOG_CONFIG(ERR, "DMA engine unmap failed\n"); +- return ret; ++ for (i = 0; i < dev->nr_guest_pages; i++) { ++ page = &dev->guest_pages[i]; ++ ret = rte_vfio_container_dma_unmap(RTE_VFIO_DEFAULT_CONTAINER_FD, ++ page->host_user_addr, ++ page->host_iova, ++ page->size); ++ if (ret) { ++ /* like DMA map, ignore the kernel driver case when unmap. */ ++ if (rte_errno == EINVAL) ++ return; ++ ++ VHOST_LOG_CONFIG(ERR, "(%s) DMA engine unmap failed\n", ++ dev->ifname); ++ } + } + } +- +- return ret; + } + + static void +@@ -205,12 +207,12 @@ free_mem_region(struct virtio_net *dev) + if (!dev || !dev->mem) + return; + ++ if (dev->async_copy && rte_vfio_is_enabled("vfio")) ++ async_dma_map(dev, false); ++ + for (i = 0; i < dev->mem->nregions; i++) { + reg = &dev->mem->regions[i]; + if (reg->host_user_addr) { +- if (dev->async_copy && rte_vfio_is_enabled("vfio")) +- async_dma_map(reg, false); +- + munmap(reg->mmap_addr, reg->mmap_size); + close(reg->fd); + } +@@ -978,7 +980,7 @@ vhost_user_set_vring_base(struct virtio_net **pdev, + + static int + add_one_guest_page(struct virtio_net *dev, uint64_t guest_phys_addr, +- uint64_t host_phys_addr, uint64_t size) ++ uint64_t host_iova, uint64_t host_user_addr, uint64_t size) + { + struct guest_page *page, *last_page; + struct guest_page *old_pages; +@@ -990,7 +992,8 @@ add_one_guest_page(struct virtio_net *dev, uint64_t guest_phys_addr, + dev->max_guest_pages * sizeof(*page), + RTE_CACHE_LINE_SIZE); + if (dev->guest_pages == NULL) { +- VHOST_LOG_CONFIG(ERR, "cannot realloc guest_pages\n"); ++ VHOST_LOG_CONFIG(ERR, "(%s) cannot realloc guest_pages\n", ++ dev->ifname); + rte_free(old_pages); + return -1; + } +@@ -999,8 +1002,9 @@ add_one_guest_page(struct virtio_net *dev, uint64_t guest_phys_addr, + if (dev->nr_guest_pages > 0) { + last_page = &dev->guest_pages[dev->nr_guest_pages - 1]; + /* merge if the two pages are continuous */ +- if (host_phys_addr == last_page->host_phys_addr + +- last_page->size) { ++ if (host_iova == last_page->host_iova + last_page->size && ++ guest_phys_addr == last_page->guest_phys_addr + last_page->size && ++ host_user_addr == last_page->host_user_addr + last_page->size) { + last_page->size += size; + return 0; + } +@@ -1008,7 +1012,8 @@ add_one_guest_page(struct virtio_net *dev, uint64_t guest_phys_addr, + + page = &dev->guest_pages[dev->nr_guest_pages++]; + page->guest_phys_addr = guest_phys_addr; +- page->host_phys_addr = host_phys_addr; ++ page->host_iova = host_iova; ++ page->host_user_addr = host_user_addr; + page->size = size; + + return 0; +@@ -1021,14 +1026,15 @@ add_guest_pages(struct virtio_net *dev, struct rte_vhost_mem_region *reg, + uint64_t reg_size = reg->size; + uint64_t host_user_addr = reg->host_user_addr; + uint64_t guest_phys_addr = reg->guest_phys_addr; +- uint64_t host_phys_addr; ++ uint64_t host_iova; + uint64_t size; + +- host_phys_addr = rte_mem_virt2iova((void *)(uintptr_t)host_user_addr); ++ host_iova = rte_mem_virt2iova((void *)(uintptr_t)host_user_addr); + size = page_size - (guest_phys_addr & (page_size - 1)); + size = RTE_MIN(size, reg_size); + +- if (add_one_guest_page(dev, guest_phys_addr, host_phys_addr, size) < 0) ++ if (add_one_guest_page(dev, guest_phys_addr, host_iova, ++ host_user_addr, size) < 0) + return -1; + + host_user_addr += size; +@@ -1037,10 +1043,10 @@ add_guest_pages(struct virtio_net *dev, struct rte_vhost_mem_region *reg, + + while (reg_size > 0) { + size = RTE_MIN(reg_size, page_size); +- host_phys_addr = rte_mem_virt2iova((void *)(uintptr_t) ++ host_iova = rte_mem_virt2iova((void *)(uintptr_t) + host_user_addr); +- if (add_one_guest_page(dev, guest_phys_addr, host_phys_addr, +- size) < 0) ++ if (add_one_guest_page(dev, guest_phys_addr, host_iova, ++ host_user_addr, size) < 0) + return -1; + + host_user_addr += size; +@@ -1071,11 +1077,11 @@ dump_guest_pages(struct virtio_net *dev) + VHOST_LOG_CONFIG(INFO, + "guest physical page region %u\n" + "\t guest_phys_addr: %" PRIx64 "\n" +- "\t host_phys_addr : %" PRIx64 "\n" ++ "\t host_iova : %" PRIx64 "\n" + "\t size : %" PRIx64 "\n", + i, + page->guest_phys_addr, +- page->host_phys_addr, ++ page->host_iova, + page->size); + } + } +@@ -1115,7 +1121,7 @@ vhost_user_postcopy_region_register(struct virtio_net *dev, + struct uffdio_register reg_struct; + + /* +- * Let's register all the mmap'ed area to ensure ++ * Let's register all the mmapped area to ensure + * alignment on page boundary. + */ + reg_struct.range.start = (uint64_t)(uintptr_t)reg->mmap_addr; +@@ -1177,7 +1183,7 @@ vhost_user_postcopy_register(struct virtio_net *dev, int main_fd, + msg->fd_num = 0; + send_vhost_reply(main_fd, msg); + +- /* Wait for qemu to acknolwedge it's got the addresses ++ /* Wait for qemu to acknowledge it got the addresses + * we've got to wait before we're allowed to generate faults. + */ + if (read_vhost_message(main_fd, &ack_msg) <= 0) { +@@ -1215,7 +1221,6 @@ vhost_user_mmap_region(struct virtio_net *dev, + uint64_t mmap_size; + uint64_t alignment; + int populate; +- int ret; + + /* Check for memory_size + mmap_offset overflow */ + if (mmap_offset >= -region->size) { +@@ -1274,14 +1279,6 @@ vhost_user_mmap_region(struct virtio_net *dev, + VHOST_LOG_CONFIG(ERR, "adding guest pages to region failed.\n"); + return -1; + } +- +- if (rte_vfio_is_enabled("vfio")) { +- ret = async_dma_map(region, true); +- if (ret) { +- VHOST_LOG_CONFIG(ERR, "Configure IOMMU for DMA engine failed\n"); +- return -1; +- } +- } + } + + VHOST_LOG_CONFIG(INFO, +@@ -1420,6 +1417,9 @@ vhost_user_set_mem_table(struct virtio_net **pdev, struct VhostUserMsg *msg, + dev->mem->nregions++; + } + ++ if (dev->async_copy && rte_vfio_is_enabled("vfio")) ++ async_dma_map(dev, true); ++ + if (vhost_user_postcopy_register(dev, main_fd, msg) < 0) + goto free_mem_table; + +@@ -1603,6 +1603,9 @@ vhost_user_get_inflight_fd(struct virtio_net **pdev, + int numa_node = SOCKET_ID_ANY; + void *addr; + ++ if (validate_msg_fds(msg, 0) != 0) ++ return RTE_VHOST_MSG_RESULT_ERR; ++ + if (msg->size != sizeof(msg->payload.inflight)) { + VHOST_LOG_CONFIG(ERR, + "invalid get_inflight_fd message size is %d\n", +@@ -1704,6 +1707,9 @@ vhost_user_set_inflight_fd(struct virtio_net **pdev, VhostUserMsg *msg, + int fd, i; + int numa_node = SOCKET_ID_ANY; + ++ if (validate_msg_fds(msg, 1) != 0) ++ return RTE_VHOST_MSG_RESULT_ERR; ++ + fd = msg->fds[0]; + if (msg->size != sizeof(msg->payload.inflight) || fd < 0) { + VHOST_LOG_CONFIG(ERR, +@@ -2566,8 +2572,12 @@ vhost_user_iotlb_msg(struct virtio_net **pdev, struct VhostUserMsg *msg, + vhost_user_iotlb_cache_insert(vq, imsg->iova, vva, + len, imsg->perm); + +- if (is_vring_iotlb(dev, vq, imsg)) ++ if (is_vring_iotlb(dev, vq, imsg)) { ++ rte_spinlock_lock(&vq->access_lock); + *pdev = dev = translate_ring_addresses(dev, i); ++ vq = dev->virtqueue[i]; ++ rte_spinlock_unlock(&vq->access_lock); ++ } + } + break; + case VHOST_IOTLB_INVALIDATE: +@@ -2580,8 +2590,11 @@ vhost_user_iotlb_msg(struct virtio_net **pdev, struct VhostUserMsg *msg, + vhost_user_iotlb_cache_remove(vq, imsg->iova, + imsg->size); + +- if (is_vring_iotlb(dev, vq, imsg)) ++ if (is_vring_iotlb(dev, vq, imsg)) { ++ rte_spinlock_lock(&vq->access_lock); + vring_invalidate(dev, vq); ++ rte_spinlock_unlock(&vq->access_lock); ++ } + } + break; + default: +@@ -2873,6 +2886,9 @@ vhost_user_check_and_alloc_queue_pair(struct virtio_net *dev, + case VHOST_USER_SET_VRING_ADDR: + vring_idx = msg->payload.addr.index; + break; ++ case VHOST_USER_SET_INFLIGHT_FD: ++ vring_idx = msg->payload.inflight.num_queues - 1; ++ break; + default: + return 0; + } +@@ -2961,7 +2977,6 @@ vhost_user_msg_handler(int vid, int fd) + return -1; + } + +- ret = 0; + request = msg.request.master; + if (request > VHOST_USER_NONE && request < VHOST_USER_MAX && + vhost_message_str[request]) { +@@ -3103,9 +3118,11 @@ vhost_user_msg_handler(int vid, int fd) + } else if (ret == RTE_VHOST_MSG_RESULT_ERR) { + VHOST_LOG_CONFIG(ERR, + "vhost message handling failed.\n"); +- return -1; ++ ret = -1; ++ goto unlock; + } + ++ ret = 0; + for (i = 0; i < dev->nr_vring; i++) { + struct vhost_virtqueue *vq = dev->virtqueue[i]; + bool cur_ready = vq_is_ready(dev, vq); +@@ -3116,10 +3133,11 @@ vhost_user_msg_handler(int vid, int fd) + } + } + ++unlock: + if (unlock_required) + vhost_user_unlock_all_queue_pairs(dev); + +- if (!virtio_is_ready(dev)) ++ if (ret != 0 || !virtio_is_ready(dev)) + goto out; + + /* +@@ -3146,7 +3164,7 @@ vhost_user_msg_handler(int vid, int fd) + } + + out: +- return 0; ++ return ret; + } + + static int process_slave_message_reply(struct virtio_net *dev, +diff --git a/dpdk/lib/vhost/virtio_net.c b/dpdk/lib/vhost/virtio_net.c +index b3d954aab4..bf4d75b4bd 100644 +--- a/dpdk/lib/vhost/virtio_net.c ++++ b/dpdk/lib/vhost/virtio_net.c +@@ -415,6 +415,16 @@ virtio_enqueue_offload(struct rte_mbuf *m_buf, struct virtio_net_hdr *net_hdr) + csum_l4 |= RTE_MBUF_F_TX_TCP_CKSUM; + + if (csum_l4) { ++ /* ++ * Pseudo-header checksum must be set as per Virtio spec. ++ * ++ * Note: We don't propagate rte_net_intel_cksum_prepare() ++ * errors, as it would have an impact on performance, and an ++ * error would mean the packet is dropped by the guest instead ++ * of being dropped here. ++ */ ++ rte_net_intel_cksum_prepare(m_buf); ++ + net_hdr->flags = VIRTIO_NET_HDR_F_NEEDS_CSUM; + net_hdr->csum_start = m_buf->l2_len + m_buf->l3_len; + +@@ -870,20 +880,21 @@ async_mbuf_to_desc_seg(struct virtio_net *dev, struct vhost_virtqueue *vq, + struct vhost_async *async = vq->async; + uint64_t mapped_len; + uint32_t buf_offset = 0; +- void *hpa; ++ void *host_iova; + + while (cpy_len) { +- hpa = (void *)(uintptr_t)gpa_to_first_hpa(dev, ++ host_iova = (void *)(uintptr_t)gpa_to_first_hpa(dev, + buf_iova + buf_offset, cpy_len, &mapped_len); +- if (unlikely(!hpa)) { +- VHOST_LOG_DATA(ERR, "(%d) %s: failed to get hpa.\n", dev->vid, __func__); ++ if (unlikely(!host_iova)) { ++ VHOST_LOG_DATA(ERR, "(%d) %s: failed to get host_iova.\n", ++ dev->vid, __func__); + return -1; + } + + if (unlikely(async_iter_add_iovec(async, + (void *)(uintptr_t)rte_pktmbuf_iova_offset(m, + mbuf_offset), +- hpa, (size_t)mapped_len))) ++ host_iova, (size_t)mapped_len))) + return -1; + + cpy_len -= (uint32_t)mapped_len; +@@ -1900,16 +1911,22 @@ rte_vhost_poll_enqueue_completed(int vid, uint16_t queue_id, + + vq = dev->virtqueue[queue_id]; + ++ if (!rte_spinlock_trylock(&vq->access_lock)) { ++ VHOST_LOG_DATA(DEBUG, ++ "%s: virtqueue %u is busy.\n", ++ __func__, queue_id); ++ return 0; ++ } ++ + if (unlikely(!vq->async)) { + VHOST_LOG_DATA(ERR, "(%d) %s: async not registered for queue id %d.\n", + dev->vid, __func__, queue_id); +- return 0; ++ goto out; + } + +- rte_spinlock_lock(&vq->access_lock); +- + n_pkts_cpl = vhost_poll_enqueue_completed(dev, queue_id, pkts, count); + ++out: + rte_spinlock_unlock(&vq->access_lock); + + return n_pkts_cpl; +@@ -2305,25 +2322,22 @@ copy_desc_to_mbuf(struct virtio_net *dev, struct vhost_virtqueue *vq, + uint32_t buf_avail, buf_offset; + uint64_t buf_addr, buf_len; + uint32_t mbuf_avail, mbuf_offset; ++ uint32_t hdr_remain = dev->vhost_hlen; + uint32_t cpy_len; + struct rte_mbuf *cur = m, *prev = m; + struct virtio_net_hdr tmp_hdr; + struct virtio_net_hdr *hdr = NULL; +- /* A counter to avoid desc dead loop chain */ +- uint16_t vec_idx = 0; ++ uint16_t vec_idx; + struct batch_copy_elem *batch_copy = vq->batch_copy_elems; + int error = 0; + +- buf_addr = buf_vec[vec_idx].buf_addr; +- buf_len = buf_vec[vec_idx].buf_len; +- +- if (unlikely(buf_len < dev->vhost_hlen && nr_vec <= 1)) { +- error = -1; +- goto out; +- } ++ /* ++ * The caller has checked the descriptors chain is larger than the ++ * header size. ++ */ + + if (virtio_net_with_host_offload(dev)) { +- if (unlikely(buf_len < sizeof(struct virtio_net_hdr))) { ++ if (unlikely(buf_vec[0].buf_len < sizeof(struct virtio_net_hdr))) { + /* + * No luck, the virtio-net header doesn't fit + * in a contiguous virtual area. +@@ -2331,34 +2345,22 @@ copy_desc_to_mbuf(struct virtio_net *dev, struct vhost_virtqueue *vq, + copy_vnet_hdr_from_desc(&tmp_hdr, buf_vec); + hdr = &tmp_hdr; + } else { +- hdr = (struct virtio_net_hdr *)((uintptr_t)buf_addr); ++ hdr = (struct virtio_net_hdr *)((uintptr_t)buf_vec[0].buf_addr); + } + } + +- /* +- * A virtio driver normally uses at least 2 desc buffers +- * for Tx: the first for storing the header, and others +- * for storing the data. +- */ +- if (unlikely(buf_len < dev->vhost_hlen)) { +- buf_offset = dev->vhost_hlen - buf_len; +- vec_idx++; +- buf_addr = buf_vec[vec_idx].buf_addr; +- buf_len = buf_vec[vec_idx].buf_len; +- buf_avail = buf_len - buf_offset; +- } else if (buf_len == dev->vhost_hlen) { +- if (unlikely(++vec_idx >= nr_vec)) +- goto out; +- buf_addr = buf_vec[vec_idx].buf_addr; +- buf_len = buf_vec[vec_idx].buf_len; ++ for (vec_idx = 0; vec_idx < nr_vec; vec_idx++) { ++ if (buf_vec[vec_idx].buf_len > hdr_remain) ++ break; + +- buf_offset = 0; +- buf_avail = buf_len; +- } else { +- buf_offset = dev->vhost_hlen; +- buf_avail = buf_vec[vec_idx].buf_len - dev->vhost_hlen; ++ hdr_remain -= buf_vec[vec_idx].buf_len; + } + ++ buf_addr = buf_vec[vec_idx].buf_addr; ++ buf_len = buf_vec[vec_idx].buf_len; ++ buf_offset = hdr_remain; ++ buf_avail = buf_vec[vec_idx].buf_len - hdr_remain; ++ + PRINT_PACKET(dev, + (uintptr_t)(buf_addr + buf_offset), + (uint32_t)buf_avail, 0); +@@ -2551,6 +2553,14 @@ virtio_dev_tx_split(struct virtio_net *dev, struct vhost_virtqueue *vq, + + update_shadow_used_ring_split(vq, head_idx, 0); + ++ if (unlikely(buf_len <= dev->vhost_hlen)) { ++ dropped += 1; ++ i++; ++ break; ++ } ++ ++ buf_len -= dev->vhost_hlen; ++ + err = virtio_dev_pktmbuf_prep(dev, pkts[i], buf_len); + if (unlikely(err)) { + /* +@@ -2754,6 +2764,11 @@ vhost_dequeue_single_packed(struct virtio_net *dev, + VHOST_ACCESS_RO) < 0)) + return -1; + ++ if (unlikely(buf_len <= dev->vhost_hlen)) ++ return -1; ++ ++ buf_len -= dev->vhost_hlen; ++ + if (unlikely(virtio_dev_pktmbuf_prep(dev, pkts, buf_len))) { + if (!allocerr_warned) { + VHOST_LOG_DATA(ERR, +diff --git a/dpdk/meson.build b/dpdk/meson.build +index 12cb6e0e83..21dc51f00d 100644 +--- a/dpdk/meson.build ++++ b/dpdk/meson.build +@@ -5,7 +5,7 @@ project('DPDK', 'C', + # Get version number from file. + # Fallback to "more" for Windows compatibility. + version: run_command(find_program('cat', 'more'), +- files('VERSION')).stdout().strip(), ++ files('VERSION'), check: true).stdout().strip(), + license: 'BSD', + default_options: ['buildtype=release', 'default_library=static'], + meson_version: '>= 0.49.2' +@@ -27,6 +27,8 @@ endif + + # set up some global vars for compiler, platform, configuration, etc. + cc = meson.get_compiler('c') ++dpdk_source_root = meson.current_source_dir() ++dpdk_build_root = meson.current_build_dir() + dpdk_conf = configuration_data() + dpdk_libraries = [] + dpdk_static_libraries = [] +diff --git a/include/linux/automake.mk b/include/linux/automake.mk +index 8f063f482e..f857c7e088 100644 +--- a/include/linux/automake.mk ++++ b/include/linux/automake.mk +@@ -2,6 +2,7 @@ noinst_HEADERS += \ + include/linux/netlink.h \ + include/linux/netfilter/nf_conntrack_sctp.h \ + include/linux/pkt_cls.h \ ++ include/linux/gen_stats.h \ + include/linux/tc_act/tc_mpls.h \ + include/linux/tc_act/tc_pedit.h \ + include/linux/tc_act/tc_skbedit.h \ +diff --git a/include/linux/gen_stats.h b/include/linux/gen_stats.h +new file mode 100644 +index 0000000000..6fae6f727c +--- /dev/null ++++ b/include/linux/gen_stats.h +@@ -0,0 +1,81 @@ ++/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ ++#ifndef __LINUX_GEN_STATS_WRAPPER_H ++#define __LINUX_GEN_STATS_WRAPPER_H 1 ++ ++#if defined(__KERNEL__) || defined(HAVE_TCA_STATS_PKT64) ++#include_next ++#else ++#include ++ ++enum { ++ TCA_STATS_UNSPEC, ++ TCA_STATS_BASIC, ++ TCA_STATS_RATE_EST, ++ TCA_STATS_QUEUE, ++ TCA_STATS_APP, ++ TCA_STATS_RATE_EST64, ++ TCA_STATS_PAD, ++ TCA_STATS_BASIC_HW, ++ TCA_STATS_PKT64, ++ __TCA_STATS_MAX, ++}; ++#define TCA_STATS_MAX (__TCA_STATS_MAX - 1) ++ ++/** ++ * struct gnet_stats_basic - byte/packet throughput statistics ++ * @bytes: number of seen bytes ++ * @packets: number of seen packets ++ */ ++struct gnet_stats_basic { ++ __u64 bytes; ++ __u32 packets; ++}; ++ ++/** ++ * struct gnet_stats_rate_est - rate estimator ++ * @bps: current byte rate ++ * @pps: current packet rate ++ */ ++struct gnet_stats_rate_est { ++ __u32 bps; ++ __u32 pps; ++}; ++ ++/** ++ * struct gnet_stats_rate_est64 - rate estimator ++ * @bps: current byte rate ++ * @pps: current packet rate ++ */ ++struct gnet_stats_rate_est64 { ++ __u64 bps; ++ __u64 pps; ++}; ++ ++/** ++ * struct gnet_stats_queue - queuing statistics ++ * @qlen: queue length ++ * @backlog: backlog size of queue ++ * @drops: number of dropped packets ++ * @requeues: number of requeues ++ * @overlimits: number of enqueues over the limit ++ */ ++struct gnet_stats_queue { ++ __u32 qlen; ++ __u32 backlog; ++ __u32 drops; ++ __u32 requeues; ++ __u32 overlimits; ++}; ++ ++/** ++ * struct gnet_estimator - rate estimator configuration ++ * @interval: sampling period ++ * @ewma_log: the log of measurement window weight ++ */ ++struct gnet_estimator { ++ signed char interval; ++ unsigned char ewma_log; ++}; ++ ++#endif /* __KERNEL__ || !HAVE_TCA_STATS_PKT64 */ ++#endif /* __LINUX_GEN_STATS_WRAPPER_H */ +diff --git a/include/openvswitch/dynamic-string.h b/include/openvswitch/dynamic-string.h +index ee18217107..1c262b0494 100644 +--- a/include/openvswitch/dynamic-string.h ++++ b/include/openvswitch/dynamic-string.h +@@ -61,6 +61,8 @@ void ds_put_printable(struct ds *, const char *, size_t); + void ds_put_hex(struct ds *ds, const void *buf, size_t size); + void ds_put_hex_dump(struct ds *ds, const void *buf_, size_t size, + uintptr_t ofs, bool ascii); ++void ds_put_sparse_hex_dump(struct ds *ds, const void *buf_, size_t size, ++ uintptr_t ofs, bool ascii); + int ds_get_line(struct ds *, FILE *); + int ds_get_preprocessed_line(struct ds *, FILE *, int *line_number); + int ds_get_test_line(struct ds *, FILE *); +diff --git a/include/openvswitch/flow.h b/include/openvswitch/flow.h +index 3054015d93..df10cf579e 100644 +--- a/include/openvswitch/flow.h ++++ b/include/openvswitch/flow.h +@@ -141,15 +141,14 @@ struct flow { + uint8_t nw_tos; /* IP ToS (including DSCP and ECN). */ + uint8_t nw_ttl; /* IP TTL/Hop Limit. */ + uint8_t nw_proto; /* IP protocol or low 8 bits of ARP opcode. */ ++ /* L4 (64-bit aligned) */ + struct in6_addr nd_target; /* IPv6 neighbor discovery (ND) target. */ + struct eth_addr arp_sha; /* ARP/ND source hardware address. */ + struct eth_addr arp_tha; /* ARP/ND target hardware address. */ +- ovs_be16 tcp_flags; /* TCP flags/ICMPv6 ND options type. +- * With L3 to avoid matching L4. */ ++ ovs_be16 tcp_flags; /* TCP flags/ICMPv6 ND options type. */ + ovs_be16 pad2; /* Pad to 64 bits. */ + struct ovs_key_nsh nsh; /* Network Service Header keys */ + +- /* L4 (64-bit aligned) */ + ovs_be16 tp_src; /* TCP/UDP/SCTP source port/ICMP type. */ + ovs_be16 tp_dst; /* TCP/UDP/SCTP destination port/ICMP code. */ + ovs_be16 ct_tp_src; /* CT original tuple source port/ICMP type. */ +@@ -179,7 +178,7 @@ BUILD_ASSERT_DECL(offsetof(struct flow, igmp_group_ip4) + sizeof(uint32_t) + enum { + FLOW_SEGMENT_1_ENDS_AT = offsetof(struct flow, dl_dst), + FLOW_SEGMENT_2_ENDS_AT = offsetof(struct flow, nw_src), +- FLOW_SEGMENT_3_ENDS_AT = offsetof(struct flow, tp_src), ++ FLOW_SEGMENT_3_ENDS_AT = offsetof(struct flow, nd_target), + }; + BUILD_ASSERT_DECL(FLOW_SEGMENT_1_ENDS_AT % sizeof(uint64_t) == 0); + BUILD_ASSERT_DECL(FLOW_SEGMENT_2_ENDS_AT % sizeof(uint64_t) == 0); +diff --git a/include/openvswitch/hmap.h b/include/openvswitch/hmap.h +index 4e001cc692..beb48295b9 100644 +--- a/include/openvswitch/hmap.h ++++ b/include/openvswitch/hmap.h +@@ -134,17 +134,17 @@ struct hmap_node *hmap_random_node(const struct hmap *); + * without using 'break', NODE will be NULL. This is true for all of the + * HMAP_FOR_EACH_*() macros. + */ +-#define HMAP_FOR_EACH_WITH_HASH(NODE, MEMBER, HASH, HMAP) \ +- for (INIT_CONTAINER(NODE, hmap_first_with_hash(HMAP, HASH), MEMBER); \ +- (NODE != OBJECT_CONTAINING(NULL, NODE, MEMBER)) \ +- || ((NODE = NULL), false); \ +- ASSIGN_CONTAINER(NODE, hmap_next_with_hash(&(NODE)->MEMBER), \ +- MEMBER)) +-#define HMAP_FOR_EACH_IN_BUCKET(NODE, MEMBER, HASH, HMAP) \ +- for (INIT_CONTAINER(NODE, hmap_first_in_bucket(HMAP, HASH), MEMBER); \ +- (NODE != OBJECT_CONTAINING(NULL, NODE, MEMBER)) \ +- || ((NODE = NULL), false); \ +- ASSIGN_CONTAINER(NODE, hmap_next_in_bucket(&(NODE)->MEMBER), MEMBER)) ++#define HMAP_FOR_EACH_WITH_HASH(NODE, MEMBER, HASH, HMAP) \ ++ for (INIT_MULTIVAR(NODE, MEMBER, hmap_first_with_hash(HMAP, HASH), \ ++ struct hmap_node); \ ++ CONDITION_MULTIVAR(NODE, MEMBER, ITER_VAR(NODE) != NULL); \ ++ UPDATE_MULTIVAR(NODE, hmap_next_with_hash(ITER_VAR(NODE)))) ++ ++#define HMAP_FOR_EACH_IN_BUCKET(NODE, MEMBER, HASH, HMAP) \ ++ for (INIT_MULTIVAR(NODE, MEMBER, hmap_first_in_bucket(HMAP, HASH), \ ++ struct hmap_node); \ ++ CONDITION_MULTIVAR(NODE, MEMBER, ITER_VAR(NODE) != NULL); \ ++ UPDATE_MULTIVAR(NODE, hmap_next_in_bucket(ITER_VAR(NODE)))) + + static inline struct hmap_node *hmap_first_with_hash(const struct hmap *, + size_t hash); +@@ -170,54 +170,80 @@ bool hmap_contains(const struct hmap *, const struct hmap_node *); + /* Iterates through every node in HMAP. */ + #define HMAP_FOR_EACH(NODE, MEMBER, HMAP) \ + HMAP_FOR_EACH_INIT(NODE, MEMBER, HMAP, (void) 0) +-#define HMAP_FOR_EACH_INIT(NODE, MEMBER, HMAP, ...) \ +- for (INIT_CONTAINER(NODE, hmap_first(HMAP), MEMBER), __VA_ARGS__; \ +- (NODE != OBJECT_CONTAINING(NULL, NODE, MEMBER)) \ +- || ((NODE = NULL), false); \ +- ASSIGN_CONTAINER(NODE, hmap_next(HMAP, &(NODE)->MEMBER), MEMBER)) ++#define HMAP_FOR_EACH_INIT(NODE, MEMBER, HMAP, ...) \ ++ for (INIT_MULTIVAR_EXP(NODE, MEMBER, hmap_first(HMAP), struct hmap_node, \ ++ __VA_ARGS__); \ ++ CONDITION_MULTIVAR(NODE, MEMBER, ITER_VAR(NODE) != NULL); \ ++ UPDATE_MULTIVAR(NODE, hmap_next(HMAP, ITER_VAR(NODE)))) + + /* Safe when NODE may be freed (not needed when NODE may be removed from the + * hash map but its members remain accessible and intact). */ +-#define HMAP_FOR_EACH_SAFE(NODE, NEXT, MEMBER, HMAP) \ +- HMAP_FOR_EACH_SAFE_INIT(NODE, NEXT, MEMBER, HMAP, (void) 0) +-#define HMAP_FOR_EACH_SAFE_INIT(NODE, NEXT, MEMBER, HMAP, ...) \ +- for (INIT_CONTAINER(NODE, hmap_first(HMAP), MEMBER), __VA_ARGS__; \ +- ((NODE != OBJECT_CONTAINING(NULL, NODE, MEMBER)) \ +- || ((NODE = NULL), false) \ +- ? INIT_CONTAINER(NEXT, hmap_next(HMAP, &(NODE)->MEMBER), MEMBER), 1 \ +- : 0); \ +- (NODE) = (NEXT)) ++#define HMAP_FOR_EACH_SAFE_LONG(NODE, NEXT, MEMBER, HMAP) \ ++ HMAP_FOR_EACH_SAFE_LONG_INIT (NODE, NEXT, MEMBER, HMAP, (void) NEXT) ++ ++#define HMAP_FOR_EACH_SAFE_LONG_INIT(NODE, NEXT, MEMBER, HMAP, ...) \ ++ for (INIT_MULTIVAR_SAFE_LONG_EXP(NODE, NEXT, MEMBER, hmap_first(HMAP), \ ++ struct hmap_node, __VA_ARGS__); \ ++ CONDITION_MULTIVAR_SAFE_LONG(NODE, NEXT, MEMBER, \ ++ ITER_VAR(NODE) != NULL, \ ++ ITER_VAR(NEXT) = hmap_next(HMAP, ITER_VAR(NODE)), \ ++ ITER_VAR(NEXT) != NULL); \ ++ UPDATE_MULTIVAR_SAFE_LONG(NODE, NEXT)) ++ ++/* Short versions of HMAP_FOR_EACH_SAFE. */ ++#define HMAP_FOR_EACH_SAFE_SHORT(NODE, MEMBER, HMAP) \ ++ HMAP_FOR_EACH_SAFE_SHORT_INIT (NODE, MEMBER, HMAP, (void) 0) ++ ++#define HMAP_FOR_EACH_SAFE_SHORT_INIT(NODE, MEMBER, HMAP, ...) \ ++ for (INIT_MULTIVAR_SAFE_SHORT_EXP(NODE, MEMBER, hmap_first(HMAP), \ ++ struct hmap_node, __VA_ARGS__); \ ++ CONDITION_MULTIVAR_SAFE_SHORT(NODE, MEMBER, \ ++ ITER_VAR(NODE) != NULL, \ ++ ITER_NEXT_VAR(NODE) = hmap_next(HMAP, ITER_VAR(NODE))); \ ++ UPDATE_MULTIVAR_SAFE_SHORT(NODE)) ++ ++#define HMAP_FOR_EACH_SAFE(...) \ ++ OVERLOAD_SAFE_MACRO(HMAP_FOR_EACH_SAFE_LONG, \ ++ HMAP_FOR_EACH_SAFE_SHORT, \ ++ 4, __VA_ARGS__) ++ + + /* Continues an iteration from just after NODE. */ + #define HMAP_FOR_EACH_CONTINUE(NODE, MEMBER, HMAP) \ + HMAP_FOR_EACH_CONTINUE_INIT(NODE, MEMBER, HMAP, (void) 0) +-#define HMAP_FOR_EACH_CONTINUE_INIT(NODE, MEMBER, HMAP, ...) \ +- for (ASSIGN_CONTAINER(NODE, hmap_next(HMAP, &(NODE)->MEMBER), MEMBER), \ +- __VA_ARGS__; \ +- (NODE != OBJECT_CONTAINING(NULL, NODE, MEMBER)) \ +- || ((NODE = NULL), false); \ +- ASSIGN_CONTAINER(NODE, hmap_next(HMAP, &(NODE)->MEMBER), MEMBER)) ++#define HMAP_FOR_EACH_CONTINUE_INIT(NODE, MEMBER, HMAP, ...) \ ++ for (INIT_MULTIVAR_EXP(NODE, MEMBER, hmap_next(HMAP, &(NODE)->MEMBER), \ ++ struct hmap_node, __VA_ARGS__); \ ++ CONDITION_MULTIVAR(NODE, MEMBER, ITER_VAR(NODE) != NULL); \ ++ UPDATE_MULTIVAR(NODE, hmap_next(HMAP, ITER_VAR(NODE)))) ++ ++struct hmap_pop_helper_iter__ { ++ size_t bucket; ++ struct hmap_node *node; ++}; + +-static inline struct hmap_node * +-hmap_pop_helper__(struct hmap *hmap, size_t *bucket) { ++static inline void ++hmap_pop_helper__(struct hmap *hmap, struct hmap_pop_helper_iter__ *iter) { + +- for (; *bucket <= hmap->mask; (*bucket)++) { +- struct hmap_node *node = hmap->buckets[*bucket]; ++ for (; iter->bucket <= hmap->mask; (iter->bucket)++) { ++ struct hmap_node *node = hmap->buckets[iter->bucket]; + + if (node) { + hmap_remove(hmap, node); +- return node; ++ iter->node = node; ++ return; + } + } +- +- return NULL; ++ iter->node = NULL; + } + +-#define HMAP_FOR_EACH_POP(NODE, MEMBER, HMAP) \ +- for (size_t bucket__ = 0; \ +- INIT_CONTAINER(NODE, hmap_pop_helper__(HMAP, &bucket__), MEMBER), \ +- (NODE != OBJECT_CONTAINING(NULL, NODE, MEMBER)) \ +- || ((NODE = NULL), false);) ++#define HMAP_FOR_EACH_POP(NODE, MEMBER, HMAP) \ ++ for (struct hmap_pop_helper_iter__ ITER_VAR(NODE) = { 0, NULL }; \ ++ hmap_pop_helper__(HMAP, &ITER_VAR(NODE)), \ ++ (ITER_VAR(NODE).node != NULL) ? \ ++ (((NODE) = OBJECT_CONTAINING(ITER_VAR(NODE).node, \ ++ NODE, MEMBER)),1): \ ++ (((NODE) = NULL), 0);) + + static inline struct hmap_node *hmap_first(const struct hmap *); + static inline struct hmap_node *hmap_next(const struct hmap *, +diff --git a/include/openvswitch/list.h b/include/openvswitch/list.h +index 8ad5eeb327..6272d340cf 100644 +--- a/include/openvswitch/list.h ++++ b/include/openvswitch/list.h +@@ -72,37 +72,74 @@ static inline bool ovs_list_is_empty(const struct ovs_list *); + static inline bool ovs_list_is_singleton(const struct ovs_list *); + static inline bool ovs_list_is_short(const struct ovs_list *); + +-#define LIST_FOR_EACH(ITER, MEMBER, LIST) \ +- for (INIT_CONTAINER(ITER, (LIST)->next, MEMBER); \ +- &(ITER)->MEMBER != (LIST); \ +- ASSIGN_CONTAINER(ITER, (ITER)->MEMBER.next, MEMBER)) +-#define LIST_FOR_EACH_CONTINUE(ITER, MEMBER, LIST) \ +- for (ASSIGN_CONTAINER(ITER, (ITER)->MEMBER.next, MEMBER); \ +- &(ITER)->MEMBER != (LIST); \ +- ASSIGN_CONTAINER(ITER, (ITER)->MEMBER.next, MEMBER)) +-#define LIST_FOR_EACH_REVERSE(ITER, MEMBER, LIST) \ +- for (INIT_CONTAINER(ITER, (LIST)->prev, MEMBER); \ +- &(ITER)->MEMBER != (LIST); \ +- ASSIGN_CONTAINER(ITER, (ITER)->MEMBER.prev, MEMBER)) +-#define LIST_FOR_EACH_REVERSE_SAFE(ITER, PREV, MEMBER, LIST) \ +- for (INIT_CONTAINER(ITER, (LIST)->prev, MEMBER); \ +- (&(ITER)->MEMBER != (LIST) \ +- ? INIT_CONTAINER(PREV, (ITER)->MEMBER.prev, MEMBER), 1 \ +- : 0); \ +- (ITER) = (PREV)) +-#define LIST_FOR_EACH_REVERSE_CONTINUE(ITER, MEMBER, LIST) \ +- for (ASSIGN_CONTAINER(ITER, (ITER)->MEMBER.prev, MEMBER); \ +- &(ITER)->MEMBER != (LIST); \ +- ASSIGN_CONTAINER(ITER, (ITER)->MEMBER.prev, MEMBER)) +-#define LIST_FOR_EACH_SAFE(ITER, NEXT, MEMBER, LIST) \ +- for (INIT_CONTAINER(ITER, (LIST)->next, MEMBER); \ +- (&(ITER)->MEMBER != (LIST) \ +- ? INIT_CONTAINER(NEXT, (ITER)->MEMBER.next, MEMBER), 1 \ +- : 0); \ +- (ITER) = (NEXT)) +-#define LIST_FOR_EACH_POP(ITER, MEMBER, LIST) \ +- while (!ovs_list_is_empty(LIST) \ +- && (INIT_CONTAINER(ITER, ovs_list_pop_front(LIST), MEMBER), 1)) ++#define LIST_FOR_EACH(VAR, MEMBER, LIST) \ ++ for (INIT_MULTIVAR(VAR, MEMBER, (LIST)->next, struct ovs_list); \ ++ CONDITION_MULTIVAR(VAR, MEMBER, ITER_VAR(VAR) != (LIST)); \ ++ UPDATE_MULTIVAR(VAR, ITER_VAR(VAR)->next)) ++ ++#define LIST_FOR_EACH_CONTINUE(VAR, MEMBER, LIST) \ ++ for (INIT_MULTIVAR(VAR, MEMBER, VAR->MEMBER.next, struct ovs_list); \ ++ CONDITION_MULTIVAR(VAR, MEMBER, ITER_VAR(VAR) != (LIST)); \ ++ UPDATE_MULTIVAR(VAR, ITER_VAR(VAR)->next)) ++ ++#define LIST_FOR_EACH_REVERSE(VAR, MEMBER, LIST) \ ++ for (INIT_MULTIVAR(VAR, MEMBER, (LIST)->prev, struct ovs_list); \ ++ CONDITION_MULTIVAR(VAR, MEMBER, ITER_VAR(VAR) != (LIST)); \ ++ UPDATE_MULTIVAR(VAR, ITER_VAR(VAR)->prev)) ++ ++#define LIST_FOR_EACH_REVERSE_CONTINUE(VAR, MEMBER, LIST) \ ++ for (INIT_MULTIVAR(VAR, MEMBER, VAR->MEMBER.prev, struct ovs_list); \ ++ CONDITION_MULTIVAR(VAR, MEMBER, ITER_VAR(VAR) != (LIST)); \ ++ UPDATE_MULTIVAR(VAR, ITER_VAR(VAR)->prev)) ++ ++/* LONG version of SAFE iterators. */ ++#define LIST_FOR_EACH_REVERSE_SAFE_LONG(VAR, PREV, MEMBER, LIST) \ ++ for (INIT_MULTIVAR_SAFE_LONG(VAR, PREV, MEMBER, (LIST)->prev, \ ++ struct ovs_list); \ ++ CONDITION_MULTIVAR_SAFE_LONG(VAR, PREV, MEMBER, \ ++ ITER_VAR(VAR) != (LIST), \ ++ ITER_VAR(PREV) = ITER_VAR(VAR)->prev, \ ++ ITER_VAR(PREV) != (LIST)); \ ++ UPDATE_MULTIVAR_SAFE_LONG(VAR, PREV)) ++ ++#define LIST_FOR_EACH_SAFE_LONG(VAR, NEXT, MEMBER, LIST) \ ++ for (INIT_MULTIVAR_SAFE_LONG(VAR, NEXT, MEMBER, (LIST)->next, \ ++ struct ovs_list); \ ++ CONDITION_MULTIVAR_SAFE_LONG(VAR, NEXT, MEMBER, \ ++ ITER_VAR(VAR) != (LIST), \ ++ ITER_VAR(NEXT) = ITER_VAR(VAR)->next, \ ++ ITER_VAR(NEXT) != (LIST)); \ ++ UPDATE_MULTIVAR_SAFE_LONG(VAR, NEXT)) ++ ++/* SHORT version of SAFE iterators. */ ++#define LIST_FOR_EACH_REVERSE_SAFE_SHORT(VAR, MEMBER, LIST) \ ++ for (INIT_MULTIVAR_SAFE_SHORT(VAR, MEMBER, (LIST)->prev, struct ovs_list);\ ++ CONDITION_MULTIVAR_SAFE_SHORT(VAR, MEMBER, \ ++ ITER_VAR(VAR) != (LIST), \ ++ ITER_NEXT_VAR(VAR) = ITER_VAR(VAR)->prev); \ ++ UPDATE_MULTIVAR_SAFE_SHORT(VAR)) ++ ++#define LIST_FOR_EACH_SAFE_SHORT(VAR, MEMBER, LIST) \ ++ for (INIT_MULTIVAR_SAFE_SHORT(VAR, MEMBER, (LIST)->next, struct ovs_list);\ ++ CONDITION_MULTIVAR_SAFE_SHORT(VAR, MEMBER, \ ++ ITER_VAR(VAR) != (LIST), \ ++ ITER_NEXT_VAR(VAR) = ITER_VAR(VAR)->next); \ ++ UPDATE_MULTIVAR_SAFE_SHORT(VAR)) ++ ++#define LIST_FOR_EACH_SAFE(...) \ ++ OVERLOAD_SAFE_MACRO(LIST_FOR_EACH_SAFE_LONG, \ ++ LIST_FOR_EACH_SAFE_SHORT, \ ++ 4, __VA_ARGS__) ++ ++#define LIST_FOR_EACH_REVERSE_SAFE(...) \ ++ OVERLOAD_SAFE_MACRO(LIST_FOR_EACH_REVERSE_SAFE_LONG, \ ++ LIST_FOR_EACH_REVERSE_SAFE_SHORT, \ ++ 4, __VA_ARGS__) ++ ++#define LIST_FOR_EACH_POP(ITER, MEMBER, LIST) \ ++ while (!ovs_list_is_empty(LIST) ? \ ++ (INIT_CONTAINER(ITER, ovs_list_pop_front(LIST), MEMBER), 1) : \ ++ (ITER = NULL, 0)) + + /* Inline implementations. */ + +diff --git a/include/openvswitch/ofp-actions.h b/include/openvswitch/ofp-actions.h +index 41bcb55d20..b7231c7bb3 100644 +--- a/include/openvswitch/ofp-actions.h ++++ b/include/openvswitch/ofp-actions.h +@@ -218,7 +218,9 @@ struct ofpact *ofpact_next_flattened(const struct ofpact *); + static inline struct ofpact * + ofpact_end(const struct ofpact *ofpacts, size_t ofpacts_len) + { +- return ALIGNED_CAST(struct ofpact *, (uint8_t *) ofpacts + ofpacts_len); ++ return ofpacts ++ ? ALIGNED_CAST(struct ofpact *, (uint8_t *) ofpacts + ofpacts_len) ++ : NULL; + } + + static inline bool +diff --git a/include/openvswitch/ofpbuf.h b/include/openvswitch/ofpbuf.h +index 1136ba04c8..32f03ea837 100644 +--- a/include/openvswitch/ofpbuf.h ++++ b/include/openvswitch/ofpbuf.h +@@ -179,7 +179,11 @@ static inline void ofpbuf_delete(struct ofpbuf *b) + static inline void *ofpbuf_at(const struct ofpbuf *b, size_t offset, + size_t size) + { +- return offset + size <= b->size ? (char *) b->data + offset : NULL; ++ if (offset + size <= b->size) { ++ ovs_assert(b->data); ++ return (char *) b->data + offset; ++ } ++ return NULL; + } + + /* Returns a pointer to byte 'offset' in 'b', which must contain at least +@@ -188,20 +192,23 @@ static inline void *ofpbuf_at_assert(const struct ofpbuf *b, size_t offset, + size_t size) + { + ovs_assert(offset + size <= b->size); +- return ((char *) b->data) + offset; ++ ovs_assert(b->data); ++ return (char *) b->data + offset; + } + + /* Returns a pointer to byte following the last byte of data in use in 'b'. */ + static inline void *ofpbuf_tail(const struct ofpbuf *b) + { +- return (char *) b->data + b->size; ++ ovs_assert(b->data || !b->size); ++ return b->data ? (char *) b->data + b->size : NULL; + } + + /* Returns a pointer to byte following the last byte allocated for use (but + * not necessarily in use) in 'b'. */ + static inline void *ofpbuf_end(const struct ofpbuf *b) + { +- return (char *) b->base + b->allocated; ++ ovs_assert(b->base || !b->allocated); ++ return b->base ? (char *) b->base + b->allocated : NULL; + } + + /* Returns the number of bytes of headroom in 'b', that is, the number of bytes +@@ -249,6 +256,11 @@ static inline void *ofpbuf_pull(struct ofpbuf *b, size_t size) + { + ovs_assert(b->size >= size); + void *data = b->data; ++ ++ if (!size) { ++ return data; ++ } ++ + b->data = (char*)b->data + size; + b->size = b->size - size; + return data; +@@ -270,7 +282,7 @@ static inline struct ofpbuf *ofpbuf_from_list(const struct ovs_list *list) + static inline bool ofpbuf_equal(const struct ofpbuf *a, const struct ofpbuf *b) + { + return a->size == b->size && +- memcmp(a->data, b->data, a->size) == 0; ++ (a->size == 0 || memcmp(a->data, b->data, a->size) == 0); + } + + static inline bool ofpbuf_oversized(const struct ofpbuf *ofpacts) +diff --git a/include/openvswitch/shash.h b/include/openvswitch/shash.h +index c249e13e1f..4e7badd4dc 100644 +--- a/include/openvswitch/shash.h ++++ b/include/openvswitch/shash.h +@@ -41,13 +41,24 @@ struct shash { + BUILD_ASSERT_TYPE(SHASH_NODE, struct shash_node *), \ + BUILD_ASSERT_TYPE(SHASH, struct shash *)) + +-#define SHASH_FOR_EACH_SAFE(SHASH_NODE, NEXT, SHASH) \ +- HMAP_FOR_EACH_SAFE_INIT ( \ ++#define SHASH_FOR_EACH_SAFE_SHORT(SHASH_NODE, SHASH) \ ++ HMAP_FOR_EACH_SAFE_SHORT_INIT ( \ ++ SHASH_NODE, node, &(SHASH)->map, \ ++ BUILD_ASSERT_TYPE(SHASH_NODE, struct shash_node *), \ ++ BUILD_ASSERT_TYPE(SHASH, struct shash *)) ++ ++#define SHASH_FOR_EACH_SAFE_LONG(SHASH_NODE, NEXT, SHASH) \ ++ HMAP_FOR_EACH_SAFE_LONG_INIT ( \ + SHASH_NODE, NEXT, node, &(SHASH)->map, \ + BUILD_ASSERT_TYPE(SHASH_NODE, struct shash_node *), \ + BUILD_ASSERT_TYPE(NEXT, struct shash_node *), \ + BUILD_ASSERT_TYPE(SHASH, struct shash *)) + ++#define SHASH_FOR_EACH_SAFE(...) \ ++ OVERLOAD_SAFE_MACRO(SHASH_FOR_EACH_SAFE_LONG, \ ++ SHASH_FOR_EACH_SAFE_SHORT, \ ++ 3, __VA_ARGS__) ++ + void shash_init(struct shash *); + void shash_destroy(struct shash *); + void shash_destroy_free_data(struct shash *); +diff --git a/include/openvswitch/util.h b/include/openvswitch/util.h +index 228b185c3a..8e6c46a85f 100644 +--- a/include/openvswitch/util.h ++++ b/include/openvswitch/util.h +@@ -145,6 +145,150 @@ OVS_NO_RETURN void ovs_assert_failure(const char *, const char *, const char *); + #define INIT_CONTAINER(OBJECT, POINTER, MEMBER) \ + ((OBJECT) = NULL, ASSIGN_CONTAINER(OBJECT, POINTER, MEMBER)) + ++/* Multi-variable container iterators. ++ * ++ * The following macros facilitate safe iteration over data structures ++ * contained in objects. It does so by using an internal iterator variable of ++ * the type of the member object pointer (i.e: pointer to the data structure). ++ */ ++ ++/* Multi-variable iterator variable name. ++ * Returns the name of the internal iterator variable. ++ */ ++#define ITER_VAR(NAME) NAME ## __iterator__ ++ ++/* Multi-variable initialization. Creates an internal iterator variable that ++ * points to the provided pointer. The type of the iterator variable is ++ * ITER_TYPE*. It must be the same type as &VAR->MEMBER. ++ * ++ * The _EXP version evaluates the extra expressions once. ++ */ ++#define INIT_MULTIVAR(VAR, MEMBER, POINTER, ITER_TYPE) \ ++ INIT_MULTIVAR_EXP(VAR, MEMBER, POINTER, ITER_TYPE, (void) 0) ++ ++#define INIT_MULTIVAR_EXP(VAR, MEMBER, POINTER, ITER_TYPE, ...) \ ++ ITER_TYPE *ITER_VAR(VAR) = ( __VA_ARGS__ , (ITER_TYPE *) POINTER) ++ ++/* Multi-variable condition. ++ * Evaluates the condition expression (that must be based on the internal ++ * iterator variable). Only if the result of expression is true, the OBJECT is ++ * set to the object containing the current value of the iterator variable. ++ * ++ * It is up to the caller to make sure it is safe to run OBJECT_CONTAINING on ++ * the pointers that verify the condition. ++ */ ++#define CONDITION_MULTIVAR(VAR, MEMBER, EXPR) \ ++ ((EXPR) ? \ ++ (((VAR) = OBJECT_CONTAINING(ITER_VAR(VAR), VAR, MEMBER)), 1) : \ ++ (((VAR) = NULL), 0)) ++ ++/* Multi-variable update. ++ * Sets the iterator value to NEXT_ITER. ++ */ ++#define UPDATE_MULTIVAR(VAR, NEXT_ITER) \ ++ (ITER_VAR(VAR) = NEXT_ITER) ++ ++/* In the safe version of the multi-variable container iteration, the next ++ * value of the iterator is precalculated on the condition expression. ++ * This allows for the iterator to be freed inside the loop. ++ * ++ * Two versions of the macros are provided: ++ * ++ * * In the _SHORT version, the user does not have to provide a variable to ++ * store the next value of the iterator. Instead, a second iterator variable ++ * is declared in the INIT_ macro and its name is determined by ++ * ITER_NEXT_VAR(OBJECT). ++ * ++ * * In the _LONG version, the user provides another variable of the same type ++ * as the iterator object variable to store the next containing object. ++ * We still declare an iterator variable inside the loop but in this case it's ++ * name is derived from the name of the next containing variable. ++ * The value of the next containing object will only be set ++ * (via OBJECT_CONTAINING) if an additional condition is statisfied. This ++ * second condition must ensure it is safe to call OBJECT_CONTAINING on the ++ * next iterator variable. ++ * With respect to the value of the next containing object: ++ * - Inside of the loop: the variable is either NULL or safe to use. ++ * - Outside of the loop: the variable is NULL if the loop ends normally. ++ * If the loop ends with a "break;" statement, rules of Inside the loop ++ * apply. ++ */ ++#define ITER_NEXT_VAR(NAME) NAME ## __iterator__next__ ++ ++/* Safe initialization declares both iterators. */ ++#define INIT_MULTIVAR_SAFE_SHORT(VAR, MEMBER, POINTER, ITER_TYPE) \ ++ INIT_MULTIVAR_SAFE_SHORT_EXP(VAR, MEMBER, POINTER, ITER_TYPE, (void) 0) ++ ++#define INIT_MULTIVAR_SAFE_SHORT_EXP(VAR, MEMBER, POINTER, ITER_TYPE, ...) \ ++ ITER_TYPE *ITER_VAR(VAR) = ( __VA_ARGS__ , (ITER_TYPE *) POINTER), \ ++ *ITER_NEXT_VAR(VAR) = NULL ++ ++/* Evaluate the condition expression and, if satisfied, update the _next_ ++ * iterator with the NEXT_EXPR. ++ * Both EXPR and NEXT_EXPR should only use ITER_VAR(VAR) and ++ * ITER_NEXT_VAR(VAR). ++ */ ++#define CONDITION_MULTIVAR_SAFE_SHORT(VAR, MEMBER, EXPR, NEXT_EXPR) \ ++ ((EXPR) ? \ ++ (((VAR) = OBJECT_CONTAINING(ITER_VAR(VAR), VAR, MEMBER)), \ ++ (NEXT_EXPR), 1) : \ ++ (((VAR) = NULL), 0)) ++ ++#define UPDATE_MULTIVAR_SAFE_SHORT(VAR) \ ++ UPDATE_MULTIVAR(VAR, ITER_NEXT_VAR(VAR)) ++ ++/* _LONG versions of the macros. */ ++ ++#define INIT_MULTIVAR_SAFE_LONG(VAR, NEXT_VAR, MEMBER, POINTER, ITER_TYPE) \ ++ INIT_MULTIVAR_SAFE_LONG_EXP(VAR, NEXT_VAR, MEMBER, POINTER, ITER_TYPE, \ ++ (void) 0) \ ++ ++#define INIT_MULTIVAR_SAFE_LONG_EXP(VAR, NEXT_VAR, MEMBER, POINTER, \ ++ ITER_TYPE, ...) \ ++ ITER_TYPE *ITER_VAR(VAR) = ( __VA_ARGS__ , (ITER_TYPE *) POINTER), \ ++ *ITER_VAR(NEXT_VAR) = NULL ++ ++/* Evaluate the condition expression and, if satisfied, update the _next_ ++ * iterator with the NEXT_EXPR. After, evaluate the NEXT_COND and, if ++ * satisfied, set the value to NEXT_VAR. NEXT_COND must use ITER_VAR(NEXT_VAR). ++ * ++ * Both EXPR and NEXT_EXPR should only use ITER_VAR(VAR) and ++ * ITER_VAR(NEXT_VAR). ++ */ ++#define CONDITION_MULTIVAR_SAFE_LONG(VAR, NEXT_VAR, MEMBER, EXPR, NEXT_EXPR, \ ++ NEXT_COND) \ ++ ((EXPR) ? \ ++ (((VAR) = OBJECT_CONTAINING(ITER_VAR(VAR), VAR, MEMBER)), \ ++ (NEXT_EXPR), ((NEXT_COND) ? \ ++ ((NEXT_VAR) = \ ++ OBJECT_CONTAINING(ITER_VAR(NEXT_VAR), NEXT_VAR, MEMBER)) : \ ++ ((NEXT_VAR) = NULL)), 1) : \ ++ (((VAR) = NULL), ((NEXT_VAR) = NULL), 0)) ++ ++#define UPDATE_MULTIVAR_SAFE_LONG(VAR, NEXT_VAR) \ ++ UPDATE_MULTIVAR(VAR, ITER_VAR(NEXT_VAR)) ++ ++/* Helpers to allow overloading the *_SAFE iterator macros and select either ++ * the LONG or the SHORT version depending on the number of arguments. ++ */ ++#define GET_SAFE_MACRO2(_1, _2, NAME, ...) NAME ++#define GET_SAFE_MACRO3(_1, _2, _3, NAME, ...) NAME ++#define GET_SAFE_MACRO4(_1, _2, _3, _4, NAME, ...) NAME ++#define GET_SAFE_MACRO5(_1, _2, _3, _4, _5, NAME, ...) NAME ++#define GET_SAFE_MACRO6(_1, _2, _3, _4, _5, _6, NAME, ...) NAME ++#define GET_SAFE_MACRO(MAX_ARGS) GET_SAFE_MACRO ## MAX_ARGS ++ ++/* MSVC treats __VA_ARGS__ as a simple token in argument lists. Introduce ++ * a level of indirection to work around that. */ ++#define EXPAND_MACRO(name, args) name args ++ ++/* Overload the LONG and the SHORT version of the macros. MAX_ARGS is the ++ * maximum number of arguments (i.e: the number of arguments of the LONG ++ * version). */ ++#define OVERLOAD_SAFE_MACRO(LONG, SHORT, MAX_ARGS, ...) \ ++ EXPAND_MACRO(GET_SAFE_MACRO(MAX_ARGS), \ ++ (__VA_ARGS__, LONG, SHORT))(__VA_ARGS__) ++ + /* Returns the number of elements in ARRAY. */ + #define ARRAY_SIZE(ARRAY) __ARRAY_SIZE(ARRAY) + +@@ -285,6 +429,9 @@ is_pow2(uintmax_t x) + * segfault, so it is important to be aware of correct alignment. */ + #define ALIGNED_CAST(TYPE, ATTR) ((TYPE) (void *) (ATTR)) + ++#define IS_PTR_ALIGNED(OBJ) \ ++ (!(OBJ) || (uintptr_t) (OBJ) % __alignof__(OVS_TYPEOF(OBJ)) == 0) ++ + #ifdef __cplusplus + } + #endif +diff --git a/ipsec/ovs-monitor-ipsec.in b/ipsec/ovs-monitor-ipsec.in +index a8b0705d9f..631a8fca80 100755 +--- a/ipsec/ovs-monitor-ipsec.in ++++ b/ipsec/ovs-monitor-ipsec.in +@@ -337,7 +337,14 @@ conn prevent_unencrypted_vxlan + Once strongSwan vici bindings will be distributed with major + Linux distributions this function could be simplified.""" + vlog.info("Refreshing StrongSwan configuration") +- subprocess.call([self.IPSEC, "update"]) ++ proc = subprocess.Popen([self.IPSEC, "update"], ++ stdout=subprocess.PIPE, ++ stderr=subprocess.PIPE) ++ outs, errs = proc.communicate() ++ if proc.returncode != 0: ++ vlog.err("StrongSwan failed to update configuration:\n" ++ "%s \n %s" % (str(outs), str(errs))) ++ + subprocess.call([self.IPSEC, "rereadsecrets"]) + # "ipsec update" command does not remove those tunnels that were + # updated or that disappeared from the ipsec.conf file. So, we have +@@ -708,6 +715,11 @@ conn prevent_unencrypted_vxlan + not re.match(r".*need --listen.*", pout): + break + ++ if re.match(r".*[F|f]ailed to initiate connection.*", pout): ++ vlog.err('Failed to initiate connection through' ++ ' Interface %s.\n' % (conn.split('-')[0])) ++ vlog.err(pout) ++ + def _nss_clear_database(self): + """Remove all OVS IPsec related state from the NSS database""" + try: +diff --git a/lib/automake.mk b/lib/automake.mk +index a23cdc4ade..e9a5978e88 100644 +--- a/lib/automake.mk ++++ b/lib/automake.mk +@@ -38,8 +38,6 @@ lib_libopenvswitchavx512_la_CFLAGS = \ + -fPIC \ + $(AM_CFLAGS) + lib_libopenvswitchavx512_la_SOURCES = \ +- lib/cpu.c \ +- lib/cpu.h \ + lib/dpif-netdev-lookup-avx512-gather.c \ + lib/dpif-netdev-extract-avx512.c \ + lib/dpif-netdev-avx512.c +@@ -89,6 +87,8 @@ lib_libopenvswitch_la_SOURCES = \ + lib/conntrack.h \ + lib/coverage.c \ + lib/coverage.h \ ++ lib/cpu.c \ ++ lib/cpu.h \ + lib/crc32c.c \ + lib/crc32c.h \ + lib/csum.c \ +diff --git a/lib/cfm.c b/lib/cfm.c +index cc43e70e31..c3742f3de2 100644 +--- a/lib/cfm.c ++++ b/lib/cfm.c +@@ -416,7 +416,7 @@ cfm_run(struct cfm *cfm) OVS_EXCLUDED(mutex) + ovs_mutex_lock(&mutex); + if (timer_expired(&cfm->fault_timer)) { + long long int interval = cfm_fault_interval(cfm); +- struct remote_mp *rmp, *rmp_next; ++ struct remote_mp *rmp; + enum cfm_fault_reason old_cfm_fault = cfm->fault; + uint64_t old_flap_count = cfm->flap_count; + int old_health = cfm->health; +@@ -475,7 +475,7 @@ cfm_run(struct cfm *cfm) OVS_EXCLUDED(mutex) + cfm->rx_packets = rx_packets; + } + +- HMAP_FOR_EACH_SAFE (rmp, rmp_next, node, &cfm->remote_mps) { ++ HMAP_FOR_EACH_SAFE (rmp, node, &cfm->remote_mps) { + if (!rmp->recv) { + VLOG_INFO("%s: Received no CCM from RMP %"PRIu64" in the last" + " %lldms", cfm->name, rmp->mpid, +diff --git a/lib/classifier.c b/lib/classifier.c +index c4790ee6ba..0a89626cc3 100644 +--- a/lib/classifier.c ++++ b/lib/classifier.c +@@ -916,9 +916,9 @@ free_conjunctive_matches(struct hmap *matches, + struct conjunctive_match *cm_stubs, size_t n_cm_stubs) + { + if (hmap_count(matches) > n_cm_stubs) { +- struct conjunctive_match *cm, *next; ++ struct conjunctive_match *cm; + +- HMAP_FOR_EACH_SAFE (cm, next, hmap_node, matches) { ++ HMAP_FOR_EACH_SAFE (cm, hmap_node, matches) { + if (!(cm >= cm_stubs && cm < &cm_stubs[n_cm_stubs])) { + free(cm); + } +diff --git a/lib/cmap.h b/lib/cmap.h +index c502d23112..72e2ec5f71 100644 +--- a/lib/cmap.h ++++ b/lib/cmap.h +@@ -108,6 +108,8 @@ size_t cmap_replace(struct cmap *, struct cmap_node *old_node, + * + * CMAP and HASH are evaluated only once. NODE is evaluated many times. + * ++ * After a normal exit of the loop (not through a "break;" statement) NODE is ++ * NULL. + * + * Thread-safety + * ============= +@@ -128,15 +130,15 @@ size_t cmap_replace(struct cmap *, struct cmap_node *old_node, + * CMAP_FOR_EACH_WITH_HASH_PROTECTED may only be used if CMAP is guaranteed not + * to change during iteration. It may be very slightly faster. + */ +-#define CMAP_NODE_FOR_EACH(NODE, MEMBER, CMAP_NODE) \ +- for (INIT_CONTAINER(NODE, CMAP_NODE, MEMBER); \ +- (NODE) != OBJECT_CONTAINING(NULL, NODE, MEMBER); \ +- ASSIGN_CONTAINER(NODE, cmap_node_next(&(NODE)->MEMBER), MEMBER)) +-#define CMAP_NODE_FOR_EACH_PROTECTED(NODE, MEMBER, CMAP_NODE) \ +- for (INIT_CONTAINER(NODE, CMAP_NODE, MEMBER); \ +- (NODE) != OBJECT_CONTAINING(NULL, NODE, MEMBER); \ +- ASSIGN_CONTAINER(NODE, cmap_node_next_protected(&(NODE)->MEMBER), \ +- MEMBER)) ++#define CMAP_NODE_FOR_EACH(NODE, MEMBER, CMAP_NODE) \ ++ for (INIT_MULTIVAR(NODE, MEMBER, CMAP_NODE, struct cmap_node); \ ++ CONDITION_MULTIVAR(NODE, MEMBER, ITER_VAR(NODE) != NULL); \ ++ UPDATE_MULTIVAR(NODE, cmap_node_next(ITER_VAR(NODE)))) ++#define CMAP_NODE_FOR_EACH_PROTECTED(NODE, MEMBER, CMAP_NODE) \ ++ for (INIT_MULTIVAR(NODE, MEMBER, CMAP_NODE, struct cmap_node); \ ++ CONDITION_MULTIVAR(NODE, MEMBER, ITER_VAR(NODE) != NULL); \ ++ UPDATE_MULTIVAR(NODE, cmap_node_next_protected(ITER_VAR(NODE)))) ++ + #define CMAP_FOR_EACH_WITH_HASH(NODE, MEMBER, HASH, CMAP) \ + CMAP_NODE_FOR_EACH(NODE, MEMBER, cmap_find(CMAP, HASH)) + #define CMAP_FOR_EACH_WITH_HASH_PROTECTED(NODE, MEMBER, HASH, CMAP) \ +@@ -223,7 +225,7 @@ unsigned long cmap_find_batch(const struct cmap *cmap, unsigned long map, + ? (INIT_CONTAINER(NODE, (CURSOR)->node, MEMBER), \ + cmap_cursor_advance(CURSOR), \ + true) \ +- : false) ++ : (NODE = NULL, false)) + + #define CMAP_CURSOR_FOR_EACH(NODE, MEMBER, CURSOR, CMAP) \ + for (*(CURSOR) = cmap_cursor_start(CMAP); \ +diff --git a/lib/conntrack-other.c b/lib/conntrack-other.c +index d3b4601858..7f3e63c384 100644 +--- a/lib/conntrack-other.c ++++ b/lib/conntrack-other.c +@@ -48,18 +48,19 @@ other_conn_update(struct conntrack *ct, struct conn *conn_, + struct dp_packet *pkt OVS_UNUSED, bool reply, long long now) + { + struct conn_other *conn = conn_other_cast(conn_); +- enum ct_update_res ret = CT_UPDATE_VALID; + + if (reply && conn->state != OTHERS_BIDIR) { + conn->state = OTHERS_BIDIR; + } else if (conn->state == OTHERS_FIRST) { + conn->state = OTHERS_MULTIPLE; +- ret = CT_UPDATE_VALID_NEW; + } + + conn_update_expiration(ct, &conn->up, other_timeouts[conn->state], now); + +- return ret; ++ if (conn->state == OTHERS_BIDIR) { ++ return CT_UPDATE_VALID; ++ } ++ return CT_UPDATE_VALID_NEW; + } + + static bool +diff --git a/lib/conntrack.c b/lib/conntrack.c +index 33a1a92953..0103fb5396 100644 +--- a/lib/conntrack.c ++++ b/lib/conntrack.c +@@ -1526,14 +1526,14 @@ set_label(struct dp_packet *pkt, struct conn *conn, + static long long + ct_sweep(struct conntrack *ct, long long now, size_t limit) + { +- struct conn *conn, *next; ++ struct conn *conn; + long long min_expiration = LLONG_MAX; + size_t count = 0; + + ovs_mutex_lock(&ct->ct_lock); + + for (unsigned i = 0; i < N_CT_TM; i++) { +- LIST_FOR_EACH_SAFE (conn, next, exp_node, &ct->exp_lists[i]) { ++ LIST_FOR_EACH_SAFE (conn, exp_node, &ct->exp_lists[i]) { + ovs_mutex_lock(&conn->lock); + if (now < conn->expiration || count >= limit) { + min_expiration = MIN(min_expiration, conn->expiration); +@@ -2242,7 +2242,7 @@ nat_range_hash(const struct conn *conn, uint32_t basis, + hash = ct_addr_hash_add(hash, &nat_info->min_addr); + hash = ct_addr_hash_add(hash, &nat_info->max_addr); + hash = hash_add(hash, +- (nat_info->max_port << 16) ++ ((uint32_t) nat_info->max_port << 16) + | nat_info->min_port); + hash = ct_endpoint_hash_add(hash, &conn->key.src); + hash = ct_endpoint_hash_add(hash, &conn->key.dst); +@@ -2265,8 +2265,16 @@ set_sport_range(const struct nat_action_info_t *ni, const struct conn_key *k, + if (((ni->nat_action & NAT_ACTION_SNAT_ALL) == NAT_ACTION_SRC) || + ((ni->nat_action & NAT_ACTION_DST))) { + *curr = ntohs(k->src.port); +- *min = MIN_NAT_EPHEMERAL_PORT; +- *max = MAX_NAT_EPHEMERAL_PORT; ++ if (*curr < 512) { ++ *min = 1; ++ *max = 511; ++ } else if (*curr < 1024) { ++ *min = 600; ++ *max = 1023; ++ } else { ++ *min = MIN_NAT_EPHEMERAL_PORT; ++ *max = MAX_NAT_EPHEMERAL_PORT; ++ } + } else { + *min = ni->min_port; + *max = ni->max_port; +@@ -2389,6 +2397,26 @@ next_addr_in_range_guarded(union ct_addr *curr, union ct_addr *min, + return exhausted; + } + ++static bool ++nat_get_unique_l4(struct conntrack *ct, struct conn *nat_conn, ++ ovs_be16 *port, uint16_t curr, uint16_t min, ++ uint16_t max) ++{ ++ uint16_t orig = curr; ++ ++ FOR_EACH_PORT_IN_RANGE (curr, min, max) { ++ *port = htons(curr); ++ if (!conn_lookup(ct, &nat_conn->rev_key, ++ time_msec(), NULL, NULL)) { ++ return true; ++ } ++ } ++ ++ *port = htons(orig); ++ ++ return false; ++} ++ + /* This function tries to get a unique tuple. + * Every iteration checks that the reverse tuple doesn't + * collide with any existing one. +@@ -2403,9 +2431,11 @@ next_addr_in_range_guarded(union ct_addr *curr, union ct_addr *min, + * + * In case of DNAT: + * - For each dst IP address in the range (if any). +- * - For each dport in range (if any). +- * - Try to find a source port in the ephemeral range +- * (after testing the port used by the sender). ++ * - For each dport in range (if any) tries to find ++ * an unique tuple. ++ * - Eventually, if the previous attempt fails, ++ * tries to find a source port in the ephemeral ++ * range (after testing the port used by the sender). + * + * If none can be found, return exhaustion to the caller. */ + static bool +@@ -2436,6 +2466,11 @@ nat_get_unique_tuple(struct conntrack *ct, const struct conn *conn, + set_dport_range(nat_info, &conn->key, hash, &curr_dport, + &min_dport, &max_dport); + ++ if (pat_proto) { ++ nat_conn->rev_key.src.port = htons(curr_dport); ++ nat_conn->rev_key.dst.port = htons(curr_sport); ++ } ++ + another_round: + store_addr_to_key(&curr_addr, &nat_conn->rev_key, + nat_info->nat_action); +@@ -2449,15 +2484,19 @@ another_round: + goto next_addr; + } + +- FOR_EACH_PORT_IN_RANGE(curr_dport, min_dport, max_dport) { +- nat_conn->rev_key.src.port = htons(curr_dport); +- FOR_EACH_PORT_IN_RANGE(curr_sport, min_sport, max_sport) { +- nat_conn->rev_key.dst.port = htons(curr_sport); +- if (!conn_lookup(ct, &nat_conn->rev_key, +- time_msec(), NULL, NULL)) { +- return true; +- } +- } ++ bool found = false; ++ if (nat_info->nat_action & NAT_ACTION_DST_PORT) { ++ found = nat_get_unique_l4(ct, nat_conn, &nat_conn->rev_key.src.port, ++ curr_dport, min_dport, max_dport); ++ } ++ ++ if (!found) { ++ found = nat_get_unique_l4(ct, nat_conn, &nat_conn->rev_key.dst.port, ++ curr_sport, min_sport, max_sport); ++ } ++ ++ if (found) { ++ return true; + } + + /* Check if next IP is in range and respin. Otherwise, notify +@@ -2857,8 +2896,8 @@ expectation_clean(struct conntrack *ct, const struct conn_key *parent_key) + { + ovs_rwlock_wrlock(&ct->resources_lock); + +- struct alg_exp_node *node, *next; +- HINDEX_FOR_EACH_WITH_HASH_SAFE (node, next, node_ref, ++ struct alg_exp_node *node; ++ HINDEX_FOR_EACH_WITH_HASH_SAFE (node, node_ref, + conn_key_hash(parent_key, ct->hash_basis), + &ct->alg_expectation_refs) { + if (!conn_key_cmp(&node->parent_key, parent_key)) { +diff --git a/lib/dns-resolve.c b/lib/dns-resolve.c +index d344514343..1afcc65adb 100644 +--- a/lib/dns-resolve.c ++++ b/lib/dns-resolve.c +@@ -189,8 +189,8 @@ dns_resolve_destroy(void) + ub_ctx_delete(ub_ctx__); + ub_ctx__ = NULL; + +- struct resolve_request *req, *next; +- HMAP_FOR_EACH_SAFE (req, next, hmap_node, &all_reqs__) { ++ struct resolve_request *req; ++ HMAP_FOR_EACH_SAFE (req, hmap_node, &all_reqs__) { + ub_resolve_free(req->ub_result); + free(req->addr); + free(req->name); +@@ -265,7 +265,7 @@ resolve_callback__(void *req_, int err, struct ub_result *result) + if (err != 0 || (result->qtype == ns_t_aaaa && !result->havedata)) { + ub_resolve_free(result); + req->state = RESOLVE_ERROR; +- VLOG_ERR_RL(&rl, "%s: failed to resolve", req->name); ++ VLOG_WARN_RL(&rl, "%s: failed to resolve", req->name); + return; + } + +diff --git a/lib/dpif-netdev-avx512.c b/lib/dpif-netdev-avx512.c +index b7131ba3f1..82a4138184 100644 +--- a/lib/dpif-netdev-avx512.c ++++ b/lib/dpif-netdev-avx512.c +@@ -20,7 +20,6 @@ + + #include + +-#include "cpu.h" + #include "dpif-netdev.h" + #include "dpif-netdev-perf.h" + #include "dpif-netdev-private.h" +@@ -59,19 +58,6 @@ struct dpif_userdata { + struct pkt_flow_meta pkt_meta[NETDEV_MAX_BURST]; + }; + +-int32_t +-dp_netdev_input_outer_avx512_probe(void) +-{ +- bool avx512f_available = cpu_has_isa(OVS_CPU_ISA_X86_AVX512F); +- bool bmi2_available = cpu_has_isa(OVS_CPU_ISA_X86_BMI2); +- +- if (!avx512f_available || !bmi2_available) { +- return -ENOTSUP; +- } +- +- return 0; +-} +- + int32_t + dp_netdev_input_outer_avx512(struct dp_netdev_pmd_thread *pmd, + struct dp_packet_batch *packets, +@@ -159,7 +145,7 @@ dp_netdev_input_outer_avx512(struct dp_netdev_pmd_thread *pmd, + mf_mask = mfex_func(packets, keys, batch_size, in_port, pmd); + } + +- uint32_t lookup_pkts_bitmask = (1ULL << batch_size) - 1; ++ uint32_t lookup_pkts_bitmask = (UINT64_C(1) << batch_size) - 1; + uint32_t iter = lookup_pkts_bitmask; + while (iter) { + uint32_t i = raw_ctz(iter); +@@ -183,7 +169,7 @@ dp_netdev_input_outer_avx512(struct dp_netdev_pmd_thread *pmd, + * classifed by vector mfex else do a scalar miniflow extract + * for that packet. + */ +- bool mfex_hit = !!(mf_mask & (1 << i)); ++ bool mfex_hit = !!(mf_mask & (UINT32_C(1) << i)); + + /* Check for a partial hardware offload match. */ + if (hwol_enabled) { +@@ -204,7 +190,7 @@ dp_netdev_input_outer_avx512(struct dp_netdev_pmd_thread *pmd, + + pkt_meta[i].bytes = dp_packet_size(packet); + phwol_hits++; +- hwol_emc_smc_hitmask |= (1 << i); ++ hwol_emc_smc_hitmask |= (UINT32_C(1) << i); + continue; + } + } +@@ -227,7 +213,7 @@ dp_netdev_input_outer_avx512(struct dp_netdev_pmd_thread *pmd, + if (f) { + rules[i] = &f->cr; + emc_hits++; +- hwol_emc_smc_hitmask |= (1 << i); ++ hwol_emc_smc_hitmask |= (UINT32_C(1) << i); + continue; + } + } +@@ -237,7 +223,7 @@ dp_netdev_input_outer_avx512(struct dp_netdev_pmd_thread *pmd, + if (f) { + rules[i] = &f->cr; + smc_hits++; +- smc_hitmask |= (1 << i); ++ smc_hitmask |= (UINT32_C(1) << i); + continue; + } + } +diff --git a/lib/dpif-netdev-extract-avx512.c b/lib/dpif-netdev-extract-avx512.c +index c1c1fefb6a..92980ca1b9 100644 +--- a/lib/dpif-netdev-extract-avx512.c ++++ b/lib/dpif-netdev-extract-avx512.c +@@ -42,7 +42,6 @@ + #include + #include + +-#include "cpu.h" + #include "flow.h" + + #include "dpif-netdev-private-dpcls.h" +@@ -544,7 +543,11 @@ mfex_avx512_process(struct dp_packet_batch *packets, + */ + __m512i v512_zeros = _mm512_setzero_si512(); + __m512i v_blk0; ++#if __GNUC__ >= 4 + if (__builtin_constant_p(use_vbmi) && use_vbmi) { ++#else ++ if (use_vbmi) { ++#endif + v_blk0 = _mm512_maskz_permutexvar_epi8_wrap(k_shuf, v_shuf, + v_pkt0); + } else { +@@ -619,7 +622,7 @@ mfex_avx512_process(struct dp_packet_batch *packets, + }; + + /* This packet has its miniflow created, add to hitmask. */ +- hitmask |= 1 << i; ++ hitmask |= UINT32_C(1) << i; + } + + return hitmask; +@@ -659,47 +662,5 @@ DECLARE_MFEX_FUNC(ip_udp, PROFILE_ETH_IPV4_UDP) + DECLARE_MFEX_FUNC(ip_tcp, PROFILE_ETH_IPV4_TCP) + DECLARE_MFEX_FUNC(dot1q_ip_udp, PROFILE_ETH_VLAN_IPV4_UDP) + DECLARE_MFEX_FUNC(dot1q_ip_tcp, PROFILE_ETH_VLAN_IPV4_TCP) +- +- +-static int32_t +-avx512_isa_probe(uint32_t needs_vbmi) +-{ +- static enum ovs_cpu_isa isa_required[] = { +- OVS_CPU_ISA_X86_AVX512F, +- OVS_CPU_ISA_X86_AVX512BW, +- OVS_CPU_ISA_X86_BMI2, +- }; +- +- int32_t ret = 0; +- for (uint32_t i = 0; i < ARRAY_SIZE(isa_required); i++) { +- if (!cpu_has_isa(isa_required[i])) { +- ret = -ENOTSUP; +- } +- } +- +- if (needs_vbmi) { +- if (!cpu_has_isa(OVS_CPU_ISA_X86_AVX512VBMI)) { +- ret = -ENOTSUP; +- } +- } +- +- return ret; +-} +- +-/* Probe functions to check ISA requirements. */ +-int32_t +-mfex_avx512_probe(void) +-{ +- const uint32_t needs_vbmi = 0; +- return avx512_isa_probe(needs_vbmi); +-} +- +-int32_t +-mfex_avx512_vbmi_probe(void) +-{ +- const uint32_t needs_vbmi = 1; +- return avx512_isa_probe(needs_vbmi); +-} +- + #endif /* __CHECKER__ */ + #endif /* __x86_64__ */ +diff --git a/lib/dpif-netdev-lookup-avx512-gather.c b/lib/dpif-netdev-lookup-avx512-gather.c +index 7bc1e9e9a5..fb2084392a 100644 +--- a/lib/dpif-netdev-lookup-avx512-gather.c ++++ b/lib/dpif-netdev-lookup-avx512-gather.c +@@ -23,7 +23,6 @@ + #include "dpif-netdev-lookup.h" + + #include "cmap.h" +-#include "cpu.h" + #include "flow.h" + #include "pvector.h" + #include "openvswitch/vlog.h" +@@ -396,18 +395,11 @@ dpcls_avx512_gather_mf_any(struct dpcls_subtable *subtable, uint32_t keys_map, + } + + dpcls_subtable_lookup_func +-dpcls_subtable_avx512_gather_probe(uint32_t u0_bits, uint32_t u1_bits) ++dpcls_subtable_avx512_gather_probe__(uint32_t u0_bits, uint32_t u1_bits, ++ bool use_vpop) + { + dpcls_subtable_lookup_func f = NULL; + +- int avx512f_available = cpu_has_isa(OVS_CPU_ISA_X86_AVX512F); +- int bmi2_available = cpu_has_isa(OVS_CPU_ISA_X86_BMI2); +- if (!avx512f_available || !bmi2_available) { +- return NULL; +- } +- +- int use_vpop = cpu_has_isa(OVS_CPU_ISA_X86_VPOPCNTDQ); +- + CHECK_LOOKUP_FUNCTION(9, 4, use_vpop); + CHECK_LOOKUP_FUNCTION(9, 1, use_vpop); + CHECK_LOOKUP_FUNCTION(5, 3, use_vpop); +diff --git a/lib/dpif-netdev-lookup.c b/lib/dpif-netdev-lookup.c +index bd0a99abe7..8612501474 100644 +--- a/lib/dpif-netdev-lookup.c ++++ b/lib/dpif-netdev-lookup.c +@@ -18,9 +18,26 @@ + #include + #include "dpif-netdev-lookup.h" + ++#include "cpu.h" + #include "openvswitch/vlog.h" + + VLOG_DEFINE_THIS_MODULE(dpif_netdev_lookup); ++#define DPCLS_IMPL_AVX512_CHECK (__x86_64__ && HAVE_AVX512F \ ++ && HAVE_LD_AVX512_GOOD && __SSE4_2__) ++ ++#if DPCLS_IMPL_AVX512_CHECK ++static dpcls_subtable_lookup_func ++dpcls_subtable_avx512_gather_probe(uint32_t u0_bits, uint32_t u1_bits) ++{ ++ if (!cpu_has_isa(OVS_CPU_ISA_X86_AVX512F) ++ || !cpu_has_isa(OVS_CPU_ISA_X86_BMI2)) { ++ return NULL; ++ } ++ ++ return dpcls_subtable_avx512_gather_probe__(u0_bits, u1_bits, ++ cpu_has_isa(OVS_CPU_ISA_X86_VPOPCNTDQ)); ++} ++#endif + + /* Actual list of implementations goes here */ + static struct dpcls_subtable_lookup_info_t subtable_lookups[] = { +@@ -43,7 +60,7 @@ static struct dpcls_subtable_lookup_info_t subtable_lookups[] = { + .probe = dpcls_subtable_generic_probe, + .name = "generic", }, + +-#if (__x86_64__ && HAVE_AVX512F && HAVE_LD_AVX512_GOOD && __SSE4_2__) ++#if DPCLS_IMPL_AVX512_CHECK + /* Only available on x86_64 bit builds with SSE 4.2 used for OVS core. */ + { .prio = 0, + .probe = dpcls_subtable_avx512_gather_probe, +diff --git a/lib/dpif-netdev-lookup.h b/lib/dpif-netdev-lookup.h +index 59f51faa0e..5d2d845945 100644 +--- a/lib/dpif-netdev-lookup.h ++++ b/lib/dpif-netdev-lookup.h +@@ -44,7 +44,8 @@ dpcls_subtable_generic_probe(uint32_t u0_bit_count, uint32_t u1_bit_count); + + /* Probe function for AVX-512 gather implementation */ + dpcls_subtable_lookup_func +-dpcls_subtable_avx512_gather_probe(uint32_t u0_bit_cnt, uint32_t u1_bit_cnt); ++dpcls_subtable_avx512_gather_probe__(uint32_t u0_bit_cnt, uint32_t u1_bit_cnt, ++ bool use_vpop); + + + /* Subtable registration and iteration helpers */ +diff --git a/lib/dpif-netdev-private-dpif.c b/lib/dpif-netdev-private-dpif.c +index 84d4ec156e..ef4cee2bad 100644 +--- a/lib/dpif-netdev-private-dpif.c ++++ b/lib/dpif-netdev-private-dpif.c +@@ -22,17 +22,33 @@ + #include + #include + ++#include "cpu.h" + #include "openvswitch/dynamic-string.h" + #include "openvswitch/vlog.h" + #include "util.h" + + VLOG_DEFINE_THIS_MODULE(dpif_netdev_impl); ++#define DPIF_NETDEV_IMPL_AVX512_CHECK (__x86_64__ && HAVE_AVX512F \ ++ && HAVE_LD_AVX512_GOOD && __SSE4_2__) + + enum dpif_netdev_impl_info_idx { + DPIF_NETDEV_IMPL_SCALAR, + DPIF_NETDEV_IMPL_AVX512 + }; + ++#if DPIF_NETDEV_IMPL_AVX512_CHECK ++static int32_t ++dp_netdev_input_outer_avx512_probe(void) ++{ ++ if (!cpu_has_isa(OVS_CPU_ISA_X86_AVX512F) ++ || !cpu_has_isa(OVS_CPU_ISA_X86_BMI2)) { ++ return -ENOTSUP; ++ } ++ ++ return 0; ++} ++#endif ++ + /* Actual list of implementations goes here. */ + static struct dpif_netdev_impl_info_t dpif_impls[] = { + /* The default scalar C code implementation. */ +@@ -40,7 +56,7 @@ static struct dpif_netdev_impl_info_t dpif_impls[] = { + .probe = NULL, + .name = "dpif_scalar", }, + +-#if (__x86_64__ && HAVE_AVX512F && HAVE_LD_AVX512_GOOD && __SSE4_2__) ++#if DPIF_NETDEV_IMPL_AVX512_CHECK + /* Only available on x86_64 bit builds with SSE 4.2 used for OVS core. */ + [DPIF_NETDEV_IMPL_AVX512] = { .input_func = dp_netdev_input_outer_avx512, + .probe = dp_netdev_input_outer_avx512_probe, +@@ -59,7 +75,7 @@ dp_netdev_impl_get_default(void) + int dpif_idx = DPIF_NETDEV_IMPL_SCALAR; + + /* Configure-time overriding to run test suite on all implementations. */ +-#if (__x86_64__ && HAVE_AVX512F && HAVE_LD_AVX512_GOOD && __SSE4_2__) ++#if DPIF_NETDEV_IMPL_AVX512_CHECK + #ifdef DPIF_AVX512_DEFAULT + dp_netdev_input_func_probe probe; + +diff --git a/lib/dpif-netdev-private-dpif.h b/lib/dpif-netdev-private-dpif.h +index 0da639c55a..3e38630f53 100644 +--- a/lib/dpif-netdev-private-dpif.h ++++ b/lib/dpif-netdev-private-dpif.h +@@ -67,10 +67,7 @@ dp_netdev_input(struct dp_netdev_pmd_thread *pmd, + struct dp_packet_batch *packets, + odp_port_t in_port); + +-/* AVX512 enabled DPIF implementation and probe functions. */ +-int32_t +-dp_netdev_input_outer_avx512_probe(void); +- ++/* AVX512 enabled DPIF implementation function. */ + int32_t + dp_netdev_input_outer_avx512(struct dp_netdev_pmd_thread *pmd, + struct dp_packet_batch *packets, +diff --git a/lib/dpif-netdev-private-extract.c b/lib/dpif-netdev-private-extract.c +index a29bdcfa78..e24e5cf0ed 100644 +--- a/lib/dpif-netdev-private-extract.c ++++ b/lib/dpif-netdev-private-extract.c +@@ -19,6 +19,7 @@ + #include + #include + ++#include "cpu.h" + #include "dp-packet.h" + #include "dpif-netdev-private-dpcls.h" + #include "dpif-netdev-private-extract.h" +@@ -33,6 +34,43 @@ VLOG_DEFINE_THIS_MODULE(dpif_netdev_extract); + /* Variable to hold the default MFEX implementation. */ + static ATOMIC(miniflow_extract_func) default_mfex_func; + ++#if MFEX_IMPL_AVX512_CHECK ++static int32_t ++avx512_isa_probe(bool needs_vbmi) ++{ ++ static enum ovs_cpu_isa isa_required[] = { ++ OVS_CPU_ISA_X86_AVX512F, ++ OVS_CPU_ISA_X86_AVX512BW, ++ OVS_CPU_ISA_X86_BMI2, ++ }; ++ ++ for (uint32_t i = 0; i < ARRAY_SIZE(isa_required); i++) { ++ if (!cpu_has_isa(isa_required[i])) { ++ return -ENOTSUP; ++ } ++ } ++ ++ if (needs_vbmi && !cpu_has_isa(OVS_CPU_ISA_X86_AVX512VBMI)) { ++ return -ENOTSUP; ++ } ++ ++ return 0; ++} ++ ++/* Probe functions to check ISA requirements. */ ++static int32_t ++mfex_avx512_probe(void) ++{ ++ return avx512_isa_probe(false); ++} ++ ++static int32_t ++mfex_avx512_vbmi_probe(void) ++{ ++ return avx512_isa_probe(true); ++} ++#endif ++ + /* Implementations of available extract options and + * the implementations are always in order of preference. + */ +@@ -54,7 +92,7 @@ static struct dpif_miniflow_extract_impl mfex_impls[] = { + .name = "study", }, + + /* Compile in implementations only if the compiler ISA checks pass. */ +-#if (__x86_64__ && HAVE_AVX512F && HAVE_LD_AVX512_GOOD && __SSE4_2__) ++#if MFEX_IMPL_AVX512_CHECK + [MFEX_IMPL_VMBI_IPv4_UDP] = { + .probe = mfex_avx512_vbmi_probe, + .extract_func = mfex_avx512_vbmi_ip_udp, +diff --git a/lib/dpif-netdev-private-extract.h b/lib/dpif-netdev-private-extract.h +index f9a757ba41..d92090190c 100644 +--- a/lib/dpif-netdev-private-extract.h ++++ b/lib/dpif-netdev-private-extract.h +@@ -19,6 +19,9 @@ + + #include + ++#define MFEX_IMPL_AVX512_CHECK (__x86_64__ && HAVE_AVX512F \ ++ && HAVE_LD_AVX512_GOOD && __SSE4_2__) ++ + /* Forward declarations. */ + struct dp_packet; + struct miniflow; +@@ -81,7 +84,7 @@ enum dpif_miniflow_extract_impl_idx { + MFEX_IMPL_AUTOVALIDATOR, + MFEX_IMPL_SCALAR, + MFEX_IMPL_STUDY, +-#if (__x86_64__ && HAVE_AVX512F && HAVE_LD_AVX512_GOOD && __SSE4_2__) ++#if MFEX_IMPL_AVX512_CHECK + MFEX_IMPL_VMBI_IPv4_UDP, + MFEX_IMPL_IPv4_UDP, + MFEX_IMPL_VMBI_IPv4_TCP, +@@ -99,7 +102,7 @@ extern struct ovs_mutex dp_netdev_mutex; + /* Define a index which points to the first traffic optimized MFEX + * option from the enum list else holds max value. + */ +-#if (__x86_64__ && HAVE_AVX512F && HAVE_LD_AVX512_GOOD && __SSE4_2__) ++#if MFEX_IMPL_AVX512_CHECK + + #define MFEX_IMPL_START_IDX MFEX_IMPL_VMBI_IPv4_UDP + #else +@@ -176,10 +179,8 @@ mfex_study_traffic(struct dp_packet_batch *packets, + int + mfex_set_study_pkt_cnt(uint32_t pkt_cmp_count, const char *name); + +-/* AVX512 MFEX Probe and Implementations functions. */ ++/* AVX512 MFEX Implementation functions. */ + #ifdef __x86_64__ +-int32_t mfex_avx512_probe(void); +-int32_t mfex_avx512_vbmi_probe(void); + + #define DECLARE_AVX512_MFEX_PROTOTYPE(name) \ + uint32_t \ +diff --git a/lib/dpif-netdev-private-flow.h b/lib/dpif-netdev-private-flow.h +index 66016eb099..7425dd44e7 100644 +--- a/lib/dpif-netdev-private-flow.h ++++ b/lib/dpif-netdev-private-flow.h +@@ -104,6 +104,7 @@ struct dp_netdev_flow { + bool dead; + uint32_t mark; /* Unique flow mark for netdev offloading. */ + uint64_t simple_match_mark; /* Unique flow mark for the simple match. */ ++ odp_port_t orig_in_port; + + /* Statistics. */ + struct dp_netdev_flow_stats stats; +diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c +index 9f35713ef5..3d9d8929f7 100644 +--- a/lib/dpif-netdev.c ++++ b/lib/dpif-netdev.c +@@ -93,7 +93,8 @@ VLOG_DEFINE_THIS_MODULE(dpif_netdev); + /* Auto Load Balancing Defaults */ + #define ALB_IMPROVEMENT_THRESHOLD 25 + #define ALB_LOAD_THRESHOLD 95 +-#define ALB_REBALANCE_INTERVAL 1 /* 1 Min */ ++#define ALB_REBALANCE_INTERVAL 1 /* 1 Min */ ++#define MAX_ALB_REBALANCE_INTERVAL 20000 /* 20000 Min */ + #define MIN_TO_MSEC 60000 + + #define FLOW_DUMP_MAX_BATCH 50 +@@ -1932,13 +1933,13 @@ static void + dp_netdev_free(struct dp_netdev *dp) + OVS_REQUIRES(dp_netdev_mutex) + { +- struct dp_netdev_port *port, *next; ++ struct dp_netdev_port *port; + struct tx_bond *bond; + + shash_find_and_delete(&dp_netdevs, dp->name); + + ovs_rwlock_wrlock(&dp->port_rwlock); +- HMAP_FOR_EACH_SAFE (port, next, node, &dp->ports) { ++ HMAP_FOR_EACH_SAFE (port, node, &dp->ports) { + do_del_port(dp, port); + } + ovs_rwlock_unlock(&dp->port_rwlock); +@@ -3006,7 +3007,7 @@ static void + queue_netdev_flow_put(struct dp_netdev_pmd_thread *pmd, + struct dp_netdev_flow *flow, struct match *match, + const struct nlattr *actions, size_t actions_len, +- odp_port_t orig_in_port, int op) ++ int op) + { + struct dp_offload_thread_item *item; + struct dp_offload_flow_item *flow_offload; +@@ -3021,7 +3022,7 @@ queue_netdev_flow_put(struct dp_netdev_pmd_thread *pmd, + flow_offload->actions = xmalloc(actions_len); + memcpy(flow_offload->actions, actions, actions_len); + flow_offload->actions_len = actions_len; +- flow_offload->orig_in_port = orig_in_port; ++ flow_offload->orig_in_port = flow->orig_in_port; + + item->timestamp = pmd->ctx.now; + dp_netdev_offload_flow_enqueue(item); +@@ -4095,6 +4096,7 @@ dp_netdev_flow_add(struct dp_netdev_pmd_thread *pmd, + flow->dead = false; + flow->batch = NULL; + flow->mark = INVALID_FLOW_MARK; ++ flow->orig_in_port = orig_in_port; + *CONST_CAST(unsigned *, &flow->pmd_id) = pmd->core_id; + *CONST_CAST(struct flow *, &flow->flow) = match->flow; + *CONST_CAST(ovs_u128 *, &flow->ufid) = *ufid; +@@ -4129,7 +4131,7 @@ dp_netdev_flow_add(struct dp_netdev_pmd_thread *pmd, + } + + queue_netdev_flow_put(pmd, flow, match, actions, actions_len, +- orig_in_port, DP_NETDEV_FLOW_OFFLOAD_OP_ADD); ++ DP_NETDEV_FLOW_OFFLOAD_OP_ADD); + log_netdev_flow_change(flow, match, NULL, actions, actions_len); + + return flow; +@@ -4171,7 +4173,7 @@ flow_put_on_pmd(struct dp_netdev_pmd_thread *pmd, + ovsrcu_set(&netdev_flow->actions, new_actions); + + queue_netdev_flow_put(pmd, netdev_flow, match, +- put->actions, put->actions_len, ODPP_NONE, ++ put->actions, put->actions_len, + DP_NETDEV_FLOW_OFFLOAD_OP_MOD); + log_netdev_flow_change(netdev_flow, match, old_actions, + put->actions, put->actions_len); +@@ -4778,8 +4780,8 @@ dpif_netdev_set_config(struct dpif *dpif, const struct smap *other_config) + uint32_t insert_min, cur_min; + uint32_t tx_flush_interval, cur_tx_flush_interval; + uint64_t rebalance_intvl; +- uint8_t rebalance_load, cur_rebalance_load; +- uint8_t rebalance_improve; ++ uint8_t cur_rebalance_load; ++ uint32_t rebalance_load, rebalance_improve; + bool log_autolb = false; + enum sched_assignment_type pmd_rxq_assign_type; + +@@ -4880,8 +4882,12 @@ dpif_netdev_set_config(struct dpif *dpif, const struct smap *other_config) + + struct pmd_auto_lb *pmd_alb = &dp->pmd_alb; + +- rebalance_intvl = smap_get_int(other_config, "pmd-auto-lb-rebal-interval", +- ALB_REBALANCE_INTERVAL); ++ rebalance_intvl = smap_get_ullong(other_config, ++ "pmd-auto-lb-rebal-interval", ++ ALB_REBALANCE_INTERVAL); ++ if (rebalance_intvl > MAX_ALB_REBALANCE_INTERVAL) { ++ rebalance_intvl = ALB_REBALANCE_INTERVAL; ++ } + + /* Input is in min, convert it to msec. */ + rebalance_intvl = +@@ -4894,21 +4900,21 @@ dpif_netdev_set_config(struct dpif *dpif, const struct smap *other_config) + log_autolb = true; + } + +- rebalance_improve = smap_get_int(other_config, +- "pmd-auto-lb-improvement-threshold", +- ALB_IMPROVEMENT_THRESHOLD); ++ rebalance_improve = smap_get_uint(other_config, ++ "pmd-auto-lb-improvement-threshold", ++ ALB_IMPROVEMENT_THRESHOLD); + if (rebalance_improve > 100) { + rebalance_improve = ALB_IMPROVEMENT_THRESHOLD; + } + if (rebalance_improve != pmd_alb->rebalance_improve_thresh) { + pmd_alb->rebalance_improve_thresh = rebalance_improve; + VLOG_INFO("PMD auto load balance improvement threshold set to " +- "%"PRIu8"%%", rebalance_improve); ++ "%"PRIu32"%%", rebalance_improve); + log_autolb = true; + } + +- rebalance_load = smap_get_int(other_config, "pmd-auto-lb-load-threshold", +- ALB_LOAD_THRESHOLD); ++ rebalance_load = smap_get_uint(other_config, "pmd-auto-lb-load-threshold", ++ ALB_LOAD_THRESHOLD); + if (rebalance_load > 100) { + rebalance_load = ALB_LOAD_THRESHOLD; + } +@@ -4916,7 +4922,7 @@ dpif_netdev_set_config(struct dpif *dpif, const struct smap *other_config) + if (rebalance_load != cur_rebalance_load) { + atomic_store_relaxed(&pmd_alb->rebalance_load_thresh, + rebalance_load); +- VLOG_INFO("PMD auto load balance load threshold set to %"PRIu8"%%", ++ VLOG_INFO("PMD auto load balance load threshold set to %"PRIu32"%%", + rebalance_load); + log_autolb = true; + } +@@ -5684,23 +5690,28 @@ sched_numa_list_put_in_place(struct sched_numa_list *numa_list) + } + } + ++/* Returns 'true' if OVS rxq scheduling algorithm assigned any unpinned rxq to ++ * a PMD thread core on a non-local numa node. */ + static bool + sched_numa_list_cross_numa_polling(struct sched_numa_list *numa_list) + { + struct sched_numa *numa; + +- /* For each numa */ + HMAP_FOR_EACH (numa, node, &numa_list->numas) { +- /* For each pmd */ + for (int i = 0; i < numa->n_pmds; i++) { + struct sched_pmd *sched_pmd; + + sched_pmd = &numa->pmds[i]; +- /* For each rxq. */ ++ if (sched_pmd->isolated) { ++ /* All rxqs on this PMD thread core are pinned. */ ++ continue; ++ } + for (unsigned k = 0; k < sched_pmd->n_rxq; k++) { + struct dp_netdev_rxq *rxq = sched_pmd->rxqs[k]; +- +- if (!sched_pmd->isolated && ++ /* Check if the rxq is not pinned to a specific PMD thread core ++ * by the user AND the PMD thread core that OVS assigned is ++ * non-local to the rxq port. */ ++ if (rxq->core_id == OVS_CORE_UNSPEC && + rxq->pmd->numa_id != + netdev_get_numa_id(rxq->port->netdev)) { + return true; +@@ -6000,10 +6011,10 @@ sched_numa_list_schedule(struct sched_numa_list *numa_list, + /* Find any numa with available PMDs. */ + for (int j = 0; j < n_numa; j++) { + numa = sched_numa_list_next(numa_list, last_cross_numa); ++ last_cross_numa = numa; + if (sched_numa_noniso_pmd_count(numa)) { + break; + } +- last_cross_numa = numa; + numa = NULL; + } + } +@@ -6111,7 +6122,7 @@ sched_numa_list_variance(struct sched_numa_list *numa_list) + * pmd_rebalance_dry_run() can be avoided when it is not needed. + */ + static bool +-pmd_reblance_dry_run_needed(struct dp_netdev *dp) ++pmd_rebalance_dry_run_needed(struct dp_netdev *dp) + OVS_REQ_RDLOCK(dp->port_rwlock) + { + struct dp_netdev_pmd_thread *pmd; +@@ -6342,11 +6353,11 @@ pmd_remove_stale_ports(struct dp_netdev *dp, + OVS_EXCLUDED(pmd->port_mutex) + OVS_REQ_RDLOCK(dp->port_rwlock) + { +- struct rxq_poll *poll, *poll_next; +- struct tx_port *tx, *tx_next; ++ struct rxq_poll *poll; ++ struct tx_port *tx; + + ovs_mutex_lock(&pmd->port_mutex); +- HMAP_FOR_EACH_SAFE (poll, poll_next, node, &pmd->poll_list) { ++ HMAP_FOR_EACH_SAFE (poll, node, &pmd->poll_list) { + struct dp_netdev_port *port = poll->rxq->port; + + if (port->need_reconfigure +@@ -6354,7 +6365,7 @@ pmd_remove_stale_ports(struct dp_netdev *dp, + dp_netdev_del_rxq_from_pmd(pmd, poll); + } + } +- HMAP_FOR_EACH_SAFE (tx, tx_next, node, &pmd->tx_ports) { ++ HMAP_FOR_EACH_SAFE (tx, node, &pmd->tx_ports) { + struct dp_netdev_port *port = tx->port; + + if (port->need_reconfigure +@@ -6430,8 +6441,7 @@ reconfigure_datapath(struct dp_netdev *dp) + /* We only reconfigure the ports that we determined above, because they're + * not being used by any pmd thread at the moment. If a port fails to + * reconfigure we remove it from the datapath. */ +- struct dp_netdev_port *next_port; +- HMAP_FOR_EACH_SAFE (port, next_port, node, &dp->ports) { ++ HMAP_FOR_EACH_SAFE (port, node, &dp->ports) { + int err; + + if (!port->need_reconfigure) { +@@ -6487,10 +6497,10 @@ reconfigure_datapath(struct dp_netdev *dp) + } + + CMAP_FOR_EACH (pmd, node, &dp->poll_threads) { +- struct rxq_poll *poll, *poll_next; ++ struct rxq_poll *poll; + + ovs_mutex_lock(&pmd->port_mutex); +- HMAP_FOR_EACH_SAFE (poll, poll_next, node, &pmd->poll_list) { ++ HMAP_FOR_EACH_SAFE (poll, node, &pmd->poll_list) { + if (poll->rxq->pmd != pmd) { + dp_netdev_del_rxq_from_pmd(pmd, poll); + +@@ -6682,7 +6692,7 @@ dpif_netdev_run(struct dpif *dpif) + if (pmd_rebalance && + !dp_netdev_is_reconf_required(dp) && + !ports_require_restart(dp) && +- pmd_reblance_dry_run_needed(dp) && ++ pmd_rebalance_dry_run_needed(dp) && + pmd_rebalance_dry_run(dp)) { + VLOG_INFO("PMD auto load balance dry run. " + "Requesting datapath reconfigure."); +@@ -7364,15 +7374,15 @@ static struct dp_netdev_pmd_thread * + dp_netdev_get_pmd(struct dp_netdev *dp, unsigned core_id) + { + struct dp_netdev_pmd_thread *pmd; +- const struct cmap_node *pnode; + +- pnode = cmap_find(&dp->poll_threads, hash_int(core_id, 0)); +- if (!pnode) { +- return NULL; ++ CMAP_FOR_EACH_WITH_HASH (pmd, node, hash_int(core_id, 0), ++ &dp->poll_threads) { ++ if (pmd->core_id == core_id) { ++ return dp_netdev_pmd_try_ref(pmd) ? pmd : NULL; ++ } + } +- pmd = CONTAINER_OF(pnode, struct dp_netdev_pmd_thread, node); + +- return dp_netdev_pmd_try_ref(pmd) ? pmd : NULL; ++ return NULL; + } + + /* Sets the 'struct dp_netdev_pmd_thread' for non-pmd threads. */ +@@ -7505,6 +7515,7 @@ dp_netdev_destroy_pmd(struct dp_netdev_pmd_thread *pmd) + seq_destroy(pmd->reload_seq); + ovs_mutex_destroy(&pmd->port_mutex); + ovs_mutex_destroy(&pmd->bond_mutex); ++ free(pmd->netdev_input_func_userdata); + free(pmd); + } + +diff --git a/lib/dpif-netlink.c b/lib/dpif-netlink.c +index 71e35ccdda..484545cfb8 100644 +--- a/lib/dpif-netlink.c ++++ b/lib/dpif-netlink.c +@@ -85,7 +85,7 @@ enum { MAX_PORTS = USHRT_MAX }; + #define EPOLLEXCLUSIVE (1u << 28) + #endif + +-#define OVS_DP_F_UNSUPPORTED (1 << 31); ++#define OVS_DP_F_UNSUPPORTED (1u << 31); + + /* This PID is not used by the kernel datapath when using dispatch per CPU, + * but it is required to be set (not zero). */ +@@ -801,14 +801,28 @@ dpif_netlink_set_handler_pids(struct dpif *dpif_, const uint32_t *upcall_pids, + uint32_t n_upcall_pids) + { + struct dpif_netlink *dpif = dpif_netlink_cast(dpif_); ++ int largest_cpu_id = ovs_numa_get_largest_core_id(); + struct dpif_netlink_dp request, reply; + struct ofpbuf *bufp; +- int error; +- int n_cores; + +- n_cores = count_cpu_cores(); +- ovs_assert(n_cores == n_upcall_pids); +- VLOG_DBG("Dispatch mode(per-cpu): Number of CPUs is %d", n_cores); ++ uint32_t *corrected; ++ int error, i, n_cores; ++ ++ if (largest_cpu_id == OVS_NUMA_UNSPEC) { ++ largest_cpu_id = -1; ++ } ++ ++ /* Some systems have non-continuous cpu core ids. count_total_cores() ++ * would return an accurate number, however, this number cannot be used. ++ * e.g. If the largest core_id of a system is cpu9, but the system only ++ * has 4 cpus then the OVS kernel module would throw a "CPU mismatch" ++ * warning. With the MAX() in place in this example we send an array of ++ * size 10 and prevent the warning. This has no bearing on the number of ++ * threads created. ++ */ ++ n_cores = MAX(count_total_cores(), largest_cpu_id + 1); ++ VLOG_DBG("Dispatch mode(per-cpu): Setting up handler PIDs for %d cores", ++ n_cores); + + dpif_netlink_dp_init(&request); + request.cmd = OVS_DP_CMD_SET; +@@ -817,7 +831,12 @@ dpif_netlink_set_handler_pids(struct dpif *dpif_, const uint32_t *upcall_pids, + request.user_features = dpif->user_features | + OVS_DP_F_DISPATCH_UPCALL_PER_CPU; + +- request.upcall_pids = upcall_pids; ++ corrected = xcalloc(n_cores, sizeof *corrected); ++ ++ for (i = 0; i < n_cores; i++) { ++ corrected[i] = upcall_pids[i % n_upcall_pids]; ++ } ++ request.upcall_pids = corrected; + request.n_upcall_pids = n_cores; + + error = dpif_netlink_dp_transact(&request, &reply, &bufp); +@@ -825,9 +844,10 @@ dpif_netlink_set_handler_pids(struct dpif *dpif_, const uint32_t *upcall_pids, + dpif->user_features = reply.user_features; + ofpbuf_delete(bufp); + if (!dpif_netlink_upcall_per_cpu(dpif)) { +- return -EOPNOTSUPP; ++ error = -EOPNOTSUPP; + } + } ++ free(corrected); + return error; + } + +@@ -1074,7 +1094,7 @@ dpif_netlink_port_add_compat(struct dpif_netlink *dpif, struct netdev *netdev, + + ext_ofs = nl_msg_start_nested(&options, OVS_TUNNEL_ATTR_EXTENSION); + for (i = 0; i < 32; i++) { +- if (tnl_cfg->exts & (1 << i)) { ++ if (tnl_cfg->exts & (UINT32_C(1) << i)) { + nl_msg_put_flag(&options, i); + } + } +@@ -2237,8 +2257,6 @@ parse_flow_put(struct dpif_netlink *dpif, struct dpif_flow_put *put) + size_t left; + struct netdev *dev; + struct offload_info info; +- ovs_be16 dst_port = 0; +- uint8_t csum_on = false; + int err; + + info.tc_modify_flow_deleted = false; +@@ -2258,10 +2276,9 @@ parse_flow_put(struct dpif_netlink *dpif, struct dpif_flow_put *put) + return EOPNOTSUPP; + } + +- /* Get tunnel dst port */ ++ /* Check the output port for a tunnel. */ + NL_ATTR_FOR_EACH(nla, left, put->actions, put->actions_len) { + if (nl_attr_type(nla) == OVS_ACTION_ATTR_OUTPUT) { +- const struct netdev_tunnel_config *tnl_cfg; + struct netdev *outdev; + odp_port_t out_port; + +@@ -2271,19 +2288,10 @@ parse_flow_put(struct dpif_netlink *dpif, struct dpif_flow_put *put) + err = EOPNOTSUPP; + goto out; + } +- tnl_cfg = netdev_get_tunnel_config(outdev); +- if (tnl_cfg && tnl_cfg->dst_port != 0) { +- dst_port = tnl_cfg->dst_port; +- } +- if (tnl_cfg) { +- csum_on = tnl_cfg->csum; +- } + netdev_close(outdev); + } + } + +- info.tp_dst_port = dst_port; +- info.tunnel_csum_on = csum_on; + info.recirc_id_shared_with_tc = (dpif->user_features + & OVS_DP_F_TC_RECIRC_SHARING); + err = netdev_flow_put(dev, &match, +@@ -2506,6 +2514,77 @@ dpif_netlink_handler_uninit(struct dpif_handler *handler) + } + #endif + ++/* Returns true if num is a prime number, ++ * otherwise, return false. ++ */ ++static bool ++is_prime(uint32_t num) ++{ ++ if (num == 2) { ++ return true; ++ } ++ ++ if (num < 2) { ++ return false; ++ } ++ ++ if (num % 2 == 0) { ++ return false; ++ } ++ ++ for (uint64_t i = 3; i * i <= num; i += 2) { ++ if (num % i == 0) { ++ return false; ++ } ++ } ++ ++ return true; ++} ++ ++/* Returns start if start is a prime number. Otherwise returns the next ++ * prime greater than start. Search is limited by UINT32_MAX. ++ * ++ * Returns 0 if no prime has been found between start and UINT32_MAX. ++ */ ++static uint32_t ++next_prime(uint32_t start) ++{ ++ if (start <= 2) { ++ return 2; ++ } ++ ++ for (uint32_t i = start; i < UINT32_MAX; i++) { ++ if (is_prime(i)) { ++ return i; ++ } ++ } ++ ++ return 0; ++} ++ ++/* Calculates and returns the number of handler threads needed based ++ * the following formula: ++ * ++ * handlers_n = min(next_prime(active_cores + 1), total_cores) ++ */ ++static uint32_t ++dpif_netlink_calculate_n_handlers(void) ++{ ++ uint32_t total_cores = count_total_cores(); ++ uint32_t n_handlers = count_cpu_cores(); ++ uint32_t next_prime_num; ++ ++ /* If not all cores are available to OVS, create additional handler ++ * threads to ensure more fair distribution of load between them. ++ */ ++ if (n_handlers < total_cores && total_cores > 2) { ++ next_prime_num = next_prime(n_handlers + 1); ++ n_handlers = MIN(next_prime_num, total_cores); ++ } ++ ++ return n_handlers; ++} ++ + static int + dpif_netlink_refresh_handlers_cpu_dispatch(struct dpif_netlink *dpif) + OVS_REQ_WRLOCK(dpif->upcall_lock) +@@ -2515,7 +2594,7 @@ dpif_netlink_refresh_handlers_cpu_dispatch(struct dpif_netlink *dpif) + uint32_t n_handlers; + uint32_t *upcall_pids; + +- n_handlers = count_cpu_cores(); ++ n_handlers = dpif_netlink_calculate_n_handlers(); + if (dpif->n_handlers != n_handlers) { + VLOG_DBG("Dispatch mode(per-cpu): initializing %d handlers", + n_handlers); +@@ -2755,7 +2834,7 @@ dpif_netlink_number_handlers_required(struct dpif *dpif_, uint32_t *n_handlers) + struct dpif_netlink *dpif = dpif_netlink_cast(dpif_); + + if (dpif_netlink_upcall_per_cpu(dpif)) { +- *n_handlers = count_cpu_cores(); ++ *n_handlers = dpif_netlink_calculate_n_handlers(); + return true; + } + +diff --git a/lib/dynamic-string.c b/lib/dynamic-string.c +index fd0127ed17..8e9555a630 100644 +--- a/lib/dynamic-string.c ++++ b/lib/dynamic-string.c +@@ -152,7 +152,10 @@ ds_put_format_valist(struct ds *ds, const char *format, va_list args_) + + va_copy(args, args_); + available = ds->string ? ds->allocated - ds->length + 1 : 0; +- needed = vsnprintf(&ds->string[ds->length], available, format, args); ++ needed = vsnprintf(ds->string ++ ? &ds->string[ds->length] ++ : NULL, ++ available, format, args); + va_end(args); + + if (needed < available) { +@@ -162,7 +165,8 @@ ds_put_format_valist(struct ds *ds, const char *format, va_list args_) + + va_copy(args, args_); + available = ds->allocated - ds->length + 1; +- needed = vsnprintf(&ds->string[ds->length], available, format, args); ++ needed = vsnprintf(&ds->string[ds->length], ++ available, format, args); + va_end(args); + + ovs_assert(needed < available); +@@ -198,10 +202,11 @@ ds_put_strftime_msec(struct ds *ds, const char *template, long long int when, + localtime_msec(when, &tm); + } + ++ ds_reserve(ds, 64); + for (;;) { +- size_t avail = ds->string ? ds->allocated - ds->length + 1 : 0; +- size_t used = strftime_msec(&ds->string[ds->length], avail, template, +- &tm); ++ size_t avail = ds->allocated - ds->length + 1; ++ char *dest = &ds->string[ds->length]; ++ size_t used = strftime_msec(dest, avail, template, &tm); + if (used) { + ds->length += used; + return; +@@ -384,13 +389,9 @@ ds_put_hex(struct ds *ds, const void *buf_, size_t size) + } + } + +-/* Writes the 'size' bytes in 'buf' to 'string' as hex bytes arranged 16 per +- * line. Numeric offsets are also included, starting at 'ofs' for the first +- * byte in 'buf'. If 'ascii' is true then the corresponding ASCII characters +- * are also rendered alongside. */ +-void +-ds_put_hex_dump(struct ds *ds, const void *buf_, size_t size, +- uintptr_t ofs, bool ascii) ++static void ++ds_put_hex_dump__(struct ds *ds, const void *buf_, size_t size, ++ uintptr_t ofs, bool ascii, bool skip_zero_lines) + { + const uint8_t *buf = buf_; + const size_t per_line = 16; /* Maximum bytes per line. */ +@@ -406,6 +407,10 @@ ds_put_hex_dump(struct ds *ds, const void *buf_, size_t size, + end = start + size; + n = end - start; + ++ if (skip_zero_lines && is_all_zeros(&buf[start], n)) { ++ goto next; ++ } ++ + /* Print line. */ + ds_put_format(ds, "%08"PRIxMAX" ", + (uintmax_t) ROUND_DOWN(ofs, per_line)); +@@ -433,13 +438,33 @@ ds_put_hex_dump(struct ds *ds, const void *buf_, size_t size, + ds_chomp(ds, ' '); + } + ds_put_format(ds, "\n"); +- ++next: + ofs += n; + buf += n; + size -= n; + } + } + ++/* Writes the 'size' bytes in 'buf' to 'string' as hex bytes arranged 16 per ++ * line. Numeric offsets are also included, starting at 'ofs' for the first ++ * byte in 'buf'. If 'ascii' is true then the corresponding ASCII characters ++ * are also rendered alongside. */ ++void ++ds_put_hex_dump(struct ds *ds, const void *buf_, size_t size, ++ uintptr_t ofs, bool ascii) ++{ ++ ds_put_hex_dump__(ds, buf_, size, ofs, ascii, false); ++} ++ ++/* Same as 'ds_put_hex_dump', but doesn't print lines that only contains ++ * zero bytes. */ ++void ++ds_put_sparse_hex_dump(struct ds *ds, const void *buf_, size_t size, ++ uintptr_t ofs, bool ascii) ++{ ++ ds_put_hex_dump__(ds, buf_, size, ofs, ascii, true); ++} ++ + int + ds_last(const struct ds *ds) + { +diff --git a/lib/fat-rwlock.c b/lib/fat-rwlock.c +index d913b2088f..771ccc9737 100644 +--- a/lib/fat-rwlock.c ++++ b/lib/fat-rwlock.c +@@ -97,14 +97,14 @@ fat_rwlock_init(struct fat_rwlock *rwlock) + void + fat_rwlock_destroy(struct fat_rwlock *rwlock) + { +- struct fat_rwlock_slot *slot, *next; ++ struct fat_rwlock_slot *slot; + + /* Order is important here. By destroying the thread-specific data first, + * before we destroy the slots, we ensure that the thread-specific + * data destructor can't race with our loop below. */ + ovsthread_key_delete(rwlock->key); + +- LIST_FOR_EACH_SAFE (slot, next, list_node, &rwlock->threads) { ++ LIST_FOR_EACH_SAFE (slot, list_node, &rwlock->threads) { + free_slot(slot); + } + ovs_mutex_destroy(&rwlock->mutex); +diff --git a/lib/flow.c b/lib/flow.c +index dd523c889b..418c3c4f48 100644 +--- a/lib/flow.c ++++ b/lib/flow.c +@@ -1950,7 +1950,7 @@ flow_wildcards_init_for_packet(struct flow_wildcards *wc, + } + + /* IPv4 or IPv6. */ +- WC_MASK_FIELD(wc, nw_frag); ++ WC_MASK_FIELD_MASK(wc, nw_frag, FLOW_NW_FRAG_MASK); + WC_MASK_FIELD(wc, nw_tos); + WC_MASK_FIELD(wc, nw_ttl); + WC_MASK_FIELD(wc, nw_proto); +diff --git a/lib/hindex.h b/lib/hindex.h +index 876c5a9e39..ea7402587e 100644 +--- a/lib/hindex.h ++++ b/lib/hindex.h +@@ -128,18 +128,38 @@ void hindex_remove(struct hindex *, struct hindex_node *); + * Evaluates HASH only once. + */ + #define HINDEX_FOR_EACH_WITH_HASH(NODE, MEMBER, HASH, HINDEX) \ +- for (INIT_CONTAINER(NODE, hindex_node_with_hash(HINDEX, HASH), MEMBER); \ +- NODE != OBJECT_CONTAINING(NULL, NODE, MEMBER); \ +- ASSIGN_CONTAINER(NODE, (NODE)->MEMBER.s, MEMBER)) ++ for (INIT_MULTIVAR(NODE, MEMBER, hindex_node_with_hash(HINDEX, HASH), \ ++ struct hindex_node); \ ++ CONDITION_MULTIVAR(NODE, MEMBER, ITER_VAR(NODE) != NULL); \ ++ UPDATE_MULTIVAR(NODE, ITER_VAR(NODE)->s)) + + /* Safe when NODE may be freed (not needed when NODE may be removed from the + * hash map but its members remain accessible and intact). */ +-#define HINDEX_FOR_EACH_WITH_HASH_SAFE(NODE, NEXT, MEMBER, HASH, HINDEX) \ +- for (INIT_CONTAINER(NODE, hindex_node_with_hash(HINDEX, HASH), MEMBER); \ +- (NODE != OBJECT_CONTAINING(NULL, NODE, MEMBER) \ +- ? INIT_CONTAINER(NEXT, (NODE)->MEMBER.s, MEMBER), 1 \ +- : 0); \ +- (NODE) = (NEXT)) ++#define HINDEX_FOR_EACH_WITH_HASH_SAFE_LONG(NODE, NEXT, MEMBER, HASH, HINDEX) \ ++ for (INIT_MULTIVAR_SAFE_LONG(NODE, NEXT, MEMBER, \ ++ hindex_node_with_hash(HINDEX, HASH), \ ++ struct hindex_node); \ ++ CONDITION_MULTIVAR_SAFE_LONG(NODE, NEXT, MEMBER, \ ++ ITER_VAR(NODE) != NULL, \ ++ ITER_VAR(NEXT) = ITER_VAR(NODE)->s, \ ++ ITER_VAR(NEXT) != NULL); \ ++ UPDATE_MULTIVAR_SAFE_LONG(NODE, NEXT)) ++ ++/* Short version of HINDEX_FOR_EACH_WITH_HASH_SAFE. */ ++#define HINDEX_FOR_EACH_WITH_HASH_SAFE_SHORT(NODE, MEMBER, HASH, HINDEX) \ ++ for (INIT_MULTIVAR_SAFE_SHORT(NODE, MEMBER, \ ++ hindex_node_with_hash(HINDEX, HASH), \ ++ struct hindex_node); \ ++ CONDITION_MULTIVAR_SAFE_SHORT(NODE, MEMBER, \ ++ ITER_VAR(NODE) != NULL, \ ++ ITER_NEXT_VAR(NODE) = ITER_VAR(NODE)->s); \ ++ UPDATE_MULTIVAR_SAFE_SHORT(NODE)) ++ ++#define HINDEX_FOR_EACH_WITH_HASH_SAFE(...) \ ++ OVERLOAD_SAFE_MACRO(HINDEX_FOR_EACH_WITH_HASH_SAFE_LONG, \ ++ HINDEX_FOR_EACH_WITH_HASH_SAFE_SHORT, \ ++ 5, __VA_ARGS__) ++ + + /* Returns the head node in 'hindex' with the given 'hash', or a null pointer + * if no nodes have that hash value. */ +@@ -157,19 +177,36 @@ hindex_node_with_hash(const struct hindex *hindex, size_t hash) + /* Iteration. */ + + /* Iterates through every node in HINDEX. */ +-#define HINDEX_FOR_EACH(NODE, MEMBER, HINDEX) \ +- for (INIT_CONTAINER(NODE, hindex_first(HINDEX), MEMBER); \ +- NODE != OBJECT_CONTAINING(NULL, NODE, MEMBER); \ +- ASSIGN_CONTAINER(NODE, hindex_next(HINDEX, &(NODE)->MEMBER), MEMBER)) ++#define HINDEX_FOR_EACH(NODE, MEMBER, HINDEX) \ ++ for (INIT_MULTIVAR(NODE, MEMBER, hindex_first(HINDEX), \ ++ struct hindex_node); \ ++ CONDITION_MULTIVAR(NODE, MEMBER, ITER_VAR(NODE) != NULL); \ ++ UPDATE_MULTIVAR(NODE, hindex_next(HINDEX, ITER_VAR(NODE)))) + + /* Safe when NODE may be freed (not needed when NODE may be removed from the + * hash index but its members remain accessible and intact). */ +-#define HINDEX_FOR_EACH_SAFE(NODE, NEXT, MEMBER, HINDEX) \ +- for (INIT_CONTAINER(NODE, hindex_first(HINDEX), MEMBER); \ +- (NODE != OBJECT_CONTAINING(NULL, NODE, MEMBER) \ +- ? INIT_CONTAINER(NEXT, hindex_next(HINDEX, &(NODE)->MEMBER), MEMBER), 1 \ +- : 0); \ +- (NODE) = (NEXT)) ++#define HINDEX_FOR_EACH_SAFE_LONG(NODE, NEXT, MEMBER, HINDEX) \ ++ for (INIT_MULTIVAR_SAFE_LONG(NODE, NEXT, MEMBER, hindex_first(HINDEX), \ ++ struct hindex_node); \ ++ CONDITION_MULTIVAR_SAFE_LONG(NODE, NEXT, MEMBER, \ ++ ITER_VAR(NODE) != NULL, \ ++ ITER_VAR(NEXT) = hindex_next(HINDEX, ITER_VAR(NODE)), \ ++ ITER_VAR(NEXT) != NULL); \ ++ UPDATE_MULTIVAR_SAFE_LONG(NODE, NEXT)) ++ ++/* Short version of HINDEX_FOR_EACH_SAFE. */ ++#define HINDEX_FOR_EACH_SAFE_SHORT(NODE, MEMBER, HINDEX) \ ++ for (INIT_MULTIVAR_SAFE_SHORT(NODE, MEMBER, hindex_first(HINDEX), \ ++ struct hindex_node); \ ++ CONDITION_MULTIVAR_SAFE_SHORT(NODE, MEMBER, \ ++ ITER_VAR(NODE) != NULL, \ ++ ITER_NEXT_VAR(NODE) = hindex_next(HINDEX, ITER_VAR(NODE))); \ ++ UPDATE_MULTIVAR_SAFE_SHORT(NODE)) ++ ++#define HINDEX_FOR_EACH_SAFE(...) \ ++ OVERLOAD_SAFE_MACRO(HINDEX_FOR_EACH_SAFE_LONG, \ ++ HINDEX_FOR_EACH_SAFE_SHORT, \ ++ 4, __VA_ARGS__) + + struct hindex_node *hindex_first(const struct hindex *); + struct hindex_node *hindex_next(const struct hindex *, +diff --git a/lib/hmapx.c b/lib/hmapx.c +index eadfe640ac..68192fc2c5 100644 +--- a/lib/hmapx.c ++++ b/lib/hmapx.c +@@ -123,9 +123,9 @@ hmapx_add_assert(struct hmapx *map, void *data) + void + hmapx_clear(struct hmapx *map) + { +- struct hmapx_node *node, *next; ++ struct hmapx_node *node; + +- HMAPX_FOR_EACH_SAFE (node, next, map) { ++ HMAPX_FOR_EACH_SAFE (node, map) { + hmapx_delete(map, node); + } + } +diff --git a/lib/hmapx.h b/lib/hmapx.h +index 06a6bbe67d..0b03411491 100644 +--- a/lib/hmapx.h ++++ b/lib/hmapx.h +@@ -67,10 +67,20 @@ bool hmapx_equals(const struct hmapx *, const struct hmapx *); + + /* Safe when NODE may be freed (not needed when NODE may be removed from the + * hash map but its members remain accessible and intact). */ +-#define HMAPX_FOR_EACH_SAFE(NODE, NEXT, HMAPX) \ +- HMAP_FOR_EACH_SAFE_INIT(NODE, NEXT, hmap_node, &(HMAPX)->map, \ ++#define HMAPX_FOR_EACH_SAFE_SHORT(NODE, HMAPX) \ ++ HMAP_FOR_EACH_SAFE_SHORT_INIT (NODE, hmap_node, &(HMAPX)->map, \ ++ BUILD_ASSERT_TYPE(NODE, struct hmapx_node *), \ ++ BUILD_ASSERT_TYPE(HMAPX, struct hmapx *)) ++ ++#define HMAPX_FOR_EACH_SAFE_LONG(NODE, NEXT, HMAPX) \ ++ HMAP_FOR_EACH_SAFE_LONG_INIT (NODE, NEXT, hmap_node, &(HMAPX)->map, \ + BUILD_ASSERT_TYPE(NODE, struct hmapx_node *), \ + BUILD_ASSERT_TYPE(NEXT, struct hmapx_node *), \ + BUILD_ASSERT_TYPE(HMAPX, struct hmapx *)) + ++#define HMAPX_FOR_EACH_SAFE(...) \ ++ OVERLOAD_SAFE_MACRO(HMAPX_FOR_EACH_SAFE_LONG, \ ++ HMAPX_FOR_EACH_SAFE_SHORT, \ ++ 3, __VA_ARGS__) ++ + #endif /* hmapx.h */ +diff --git a/lib/id-fpool.c b/lib/id-fpool.c +index 15cef5d003..7108c104a4 100644 +--- a/lib/id-fpool.c ++++ b/lib/id-fpool.c +@@ -166,11 +166,10 @@ void + id_fpool_destroy(struct id_fpool *pool) + { + struct id_slab *slab; +- struct id_slab *next; + size_t i; + + id_fpool_lock(&pool->pool_lock); +- LIST_FOR_EACH_SAFE (slab, next, node, &pool->free_slabs) { ++ LIST_FOR_EACH_SAFE (slab, node, &pool->free_slabs) { + free(slab); + } + ovs_list_poison(&pool->free_slabs); +diff --git a/lib/ipf.c b/lib/ipf.c +index 507db2aea2..d452663743 100644 +--- a/lib/ipf.c ++++ b/lib/ipf.c +@@ -1058,9 +1058,9 @@ ipf_send_completed_frags(struct ipf *ipf, struct dp_packet_batch *pb, + } + + ovs_mutex_lock(&ipf->ipf_lock); +- struct ipf_list *ipf_list, *next; ++ struct ipf_list *ipf_list; + +- LIST_FOR_EACH_SAFE (ipf_list, next, list_node, &ipf->frag_complete_list) { ++ LIST_FOR_EACH_SAFE (ipf_list, list_node, &ipf->frag_complete_list) { + if (ipf_send_frags_in_list(ipf, ipf_list, pb, IPF_FRAG_COMPLETED_LIST, + v6, now)) { + ipf_completed_list_clean(&ipf->frag_lists, ipf_list); +@@ -1090,10 +1090,10 @@ ipf_send_expired_frags(struct ipf *ipf, struct dp_packet_batch *pb, + } + + ovs_mutex_lock(&ipf->ipf_lock); +- struct ipf_list *ipf_list, *next; ++ struct ipf_list *ipf_list; + size_t lists_removed = 0; + +- LIST_FOR_EACH_SAFE (ipf_list, next, list_node, &ipf->frag_exp_list) { ++ LIST_FOR_EACH_SAFE (ipf_list, list_node, &ipf->frag_exp_list) { + if (now <= ipf_list->expiration || + lists_removed >= IPF_FRAG_LIST_MAX_EXPIRED) { + break; +@@ -1121,9 +1121,9 @@ ipf_execute_reass_pkts(struct ipf *ipf, struct dp_packet_batch *pb) + } + + ovs_mutex_lock(&ipf->ipf_lock); +- struct reassembled_pkt *rp, *next; ++ struct reassembled_pkt *rp; + +- LIST_FOR_EACH_SAFE (rp, next, rp_list_node, &ipf->reassembled_pkt_list) { ++ LIST_FOR_EACH_SAFE (rp, rp_list_node, &ipf->reassembled_pkt_list) { + if (!rp->list->reass_execute_ctx && + ipf_dp_packet_batch_add(pb, rp->pkt, false)) { + rp->list->reass_execute_ctx = rp->pkt; +@@ -1144,9 +1144,9 @@ ipf_post_execute_reass_pkts(struct ipf *ipf, + } + + ovs_mutex_lock(&ipf->ipf_lock); +- struct reassembled_pkt *rp, *next; ++ struct reassembled_pkt *rp; + +- LIST_FOR_EACH_SAFE (rp, next, rp_list_node, &ipf->reassembled_pkt_list) { ++ LIST_FOR_EACH_SAFE (rp, rp_list_node, &ipf->reassembled_pkt_list) { + const size_t pb_cnt = dp_packet_batch_size(pb); + int pb_idx; + struct dp_packet *pkt; +@@ -1271,15 +1271,15 @@ ipf_clean_thread_main(void *f) + + ovs_mutex_lock(&ipf->ipf_lock); + +- struct ipf_list *ipf_list, *next; +- LIST_FOR_EACH_SAFE (ipf_list, next, list_node, ++ struct ipf_list *ipf_list; ++ LIST_FOR_EACH_SAFE (ipf_list, list_node, + &ipf->frag_exp_list) { + if (ipf_purge_list_check(ipf, ipf_list, now)) { + ipf_expiry_list_clean(&ipf->frag_lists, ipf_list); + } + } + +- LIST_FOR_EACH_SAFE (ipf_list, next, list_node, ++ LIST_FOR_EACH_SAFE (ipf_list, list_node, + &ipf->frag_complete_list) { + if (ipf_purge_list_check(ipf, ipf_list, now)) { + ipf_completed_list_clean(&ipf->frag_lists, ipf_list); +diff --git a/lib/json.c b/lib/json.c +index 720c73d940..042aab83b3 100644 +--- a/lib/json.c ++++ b/lib/json.c +@@ -397,9 +397,9 @@ json_destroy__(struct json *json) + static void + json_destroy_object(struct shash *object) + { +- struct shash_node *node, *next; ++ struct shash_node *node; + +- SHASH_FOR_EACH_SAFE (node, next, object) { ++ SHASH_FOR_EACH_SAFE (node, object) { + struct json *value = node->data; + + json_destroy(value); +diff --git a/lib/lacp.c b/lib/lacp.c +index 89d711225f..3252f17ebf 100644 +--- a/lib/lacp.c ++++ b/lib/lacp.c +@@ -280,10 +280,10 @@ void + lacp_unref(struct lacp *lacp) OVS_EXCLUDED(mutex) + { + if (lacp && ovs_refcount_unref_relaxed(&lacp->ref_cnt) == 1) { +- struct member *member, *next; ++ struct member *member; + + lacp_lock(); +- HMAP_FOR_EACH_SAFE (member, next, node, &lacp->members) { ++ HMAP_FOR_EACH_SAFE (member, node, &lacp->members) { + member_destroy(member); + } + +diff --git a/lib/libopenvswitch.pc.in b/lib/libopenvswitch.pc.in +index 2a3f2ca7bc..44fbb1f9fd 100644 +--- a/lib/libopenvswitch.pc.in ++++ b/lib/libopenvswitch.pc.in +@@ -7,5 +7,5 @@ Name: libopenvswitch + Description: Open vSwitch library + Version: @VERSION@ + Libs: -L${libdir} -lopenvswitch +-Libs.private: @LIBS@ +-Cflags: -I${includedir}/openvswitch ++Libs.private: @LIBS@ @SSL_LIBS@ @CAPNG_LDADD@ @LIBBPF_LDADD@ ++Cflags: -I${includedir} +diff --git a/lib/libsflow.pc.in b/lib/libsflow.pc.in +index e70a2b7048..34bb7e3d69 100644 +--- a/lib/libsflow.pc.in ++++ b/lib/libsflow.pc.in +@@ -8,4 +8,4 @@ Description: sFlow library of Open vSwitch + Version: @VERSION@ + Libs: -L${libdir} -lsflow + Libs.private: @LIBS@ +-Cflags: -I${includedir}/openvswitch ++Cflags: -I${includedir} +diff --git a/lib/lldp/lldpd-structs.c b/lib/lldp/lldpd-structs.c +index 499b441746..a8c7fad098 100644 +--- a/lib/lldp/lldpd-structs.c ++++ b/lib/lldp/lldpd-structs.c +@@ -64,11 +64,11 @@ lldpd_remote_cleanup(struct lldpd_hardware *hw, + struct lldpd_port *), + bool all) + { +- struct lldpd_port *port, *port_next; ++ struct lldpd_port *port; + time_t now = time_now(); + + VLOG_DBG("cleanup remote port on %s", hw->h_ifname); +- LIST_FOR_EACH_SAFE (port, port_next, p_entries, &hw->h_rports) { ++ LIST_FOR_EACH_SAFE (port, p_entries, &hw->h_rports) { + bool del = all; + if (!all && expire && + (now >= port->p_lastupdate + port->p_chassis->c_ttl)) { +@@ -99,11 +99,10 @@ static void + lldpd_aa_maps_cleanup(struct lldpd_port *port) + { + struct lldpd_aa_isid_vlan_maps_tlv *isid_vlan_map = NULL; +- struct lldpd_aa_isid_vlan_maps_tlv *isid_vlan_map_next = NULL; + + if (!ovs_list_is_empty(&port->p_isid_vlan_maps)) { + +- LIST_FOR_EACH_SAFE (isid_vlan_map, isid_vlan_map_next, m_entries, ++ LIST_FOR_EACH_SAFE (isid_vlan_map, m_entries, + &port->p_isid_vlan_maps) { + + ovs_list_remove(&isid_vlan_map->m_entries); +diff --git a/lib/lldp/lldpd.c b/lib/lldp/lldpd.c +index a024dc5e58..4bff7b017f 100644 +--- a/lib/lldp/lldpd.c ++++ b/lib/lldp/lldpd.c +@@ -134,24 +134,20 @@ lldpd_hardware_cleanup(struct lldpd *cfg, struct lldpd_hardware *hardware) + void + lldpd_cleanup(struct lldpd *cfg) + { +- struct lldpd_hardware *hw, *hw_next; +- struct lldpd_chassis *chassis, *chassis_next; ++ struct lldpd_hardware *hw; ++ struct lldpd_chassis *chassis; + + VLOG_DBG("cleanup all ports"); + +- LIST_FOR_EACH_SAFE (hw, hw_next, h_entries, &cfg->g_hardware) { +- if (!hw->h_flags) { +- ovs_list_remove(&hw->h_entries); +- lldpd_remote_cleanup(hw, NULL, true); +- lldpd_hardware_cleanup(cfg, hw); +- } else { +- lldpd_remote_cleanup(hw, NULL, false); +- } ++ LIST_FOR_EACH_SAFE (hw, h_entries, &cfg->g_hardware) { ++ ovs_list_remove(&hw->h_entries); ++ lldpd_remote_cleanup(hw, NULL, true); ++ lldpd_hardware_cleanup(cfg, hw); + } + + VLOG_DBG("cleanup all chassis"); + +- LIST_FOR_EACH_SAFE (chassis, chassis_next, list, &cfg->g_chassis) { ++ LIST_FOR_EACH_SAFE (chassis, list, &cfg->g_chassis) { + if (chassis->c_refcount == 0) { + ovs_list_remove(&chassis->list); + lldpd_chassis_cleanup(chassis, 1); +diff --git a/lib/mac-learning.c b/lib/mac-learning.c +index 3fcd7d9b77..a60794fb26 100644 +--- a/lib/mac-learning.c ++++ b/lib/mac-learning.c +@@ -244,10 +244,10 @@ void + mac_learning_unref(struct mac_learning *ml) + { + if (ml && ovs_refcount_unref(&ml->ref_cnt) == 1) { +- struct mac_entry *e, *next; ++ struct mac_entry *e; + + ovs_rwlock_wrlock(&ml->rwlock); +- HMAP_FOR_EACH_SAFE (e, next, hmap_node, &ml->table) { ++ HMAP_FOR_EACH_SAFE (e, hmap_node, &ml->table) { + mac_learning_expire(ml, e); + } + hmap_destroy(&ml->table); +diff --git a/lib/match.c b/lib/match.c +index 2ad03e044e..0b9dc4278c 100644 +--- a/lib/match.c ++++ b/lib/match.c +@@ -1737,7 +1737,7 @@ match_format(const struct match *match, + format_be32_masked(s, "mpls_lse1", f->mpls_lse[1], wc->masks.mpls_lse[1]); + format_be32_masked(s, "mpls_lse2", f->mpls_lse[2], wc->masks.mpls_lse[2]); + +- switch (wc->masks.nw_frag) { ++ switch (wc->masks.nw_frag & FLOW_NW_FRAG_MASK) { + case FLOW_NW_FRAG_ANY | FLOW_NW_FRAG_LATER: + ds_put_format(s, "%snw_frag=%s%s,", colors.param, colors.end, + f->nw_frag & FLOW_NW_FRAG_ANY +diff --git a/lib/mcast-snooping.c b/lib/mcast-snooping.c +index 6730301b67..029ca28558 100644 +--- a/lib/mcast-snooping.c ++++ b/lib/mcast-snooping.c +@@ -356,11 +356,11 @@ mcast_snooping_prune_expired(struct mcast_snooping *ms, + OVS_REQ_WRLOCK(ms->rwlock) + { + int expired; +- struct mcast_group_bundle *b, *next_b; ++ struct mcast_group_bundle *b; + time_t timenow = time_now(); + + expired = 0; +- LIST_FOR_EACH_SAFE (b, next_b, bundle_node, &grp->bundle_lru) { ++ LIST_FOR_EACH_SAFE (b, bundle_node, &grp->bundle_lru) { + /* This list is sorted on expiration time. */ + if (b->expires > timenow) { + break; +@@ -946,15 +946,15 @@ mcast_snooping_wait(struct mcast_snooping *ms) + void + mcast_snooping_flush_bundle(struct mcast_snooping *ms, void *port) + { +- struct mcast_group *g, *next_g; +- struct mcast_mrouter_bundle *m, *next_m; ++ struct mcast_group *g; ++ struct mcast_mrouter_bundle *m; + + if (!mcast_snooping_enabled(ms)) { + return; + } + + ovs_rwlock_wrlock(&ms->rwlock); +- LIST_FOR_EACH_SAFE (g, next_g, group_node, &ms->group_lru) { ++ LIST_FOR_EACH_SAFE (g, group_node, &ms->group_lru) { + if (mcast_group_delete_bundle(ms, g, port)) { + ms->need_revalidate = true; + +@@ -964,7 +964,7 @@ mcast_snooping_flush_bundle(struct mcast_snooping *ms, void *port) + } + } + +- LIST_FOR_EACH_SAFE (m, next_m, mrouter_node, &ms->mrouter_lru) { ++ LIST_FOR_EACH_SAFE (m, mrouter_node, &ms->mrouter_lru) { + if (m->port == port) { + mcast_snooping_flush_mrouter(m); + ms->need_revalidate = true; +diff --git a/lib/meta-flow.c b/lib/meta-flow.c +index e03cd8d0c5..c576ae6202 100644 +--- a/lib/meta-flow.c ++++ b/lib/meta-flow.c +@@ -3442,7 +3442,9 @@ mf_get_vl_mff(const struct mf_field *mff, + const struct vl_mff_map *vl_mff_map) + { + if (mff && mff->variable_len && vl_mff_map) { +- return &mf_get_vl_mff__(mff->id, vl_mff_map)->mf; ++ struct vl_mf_field *vl_mff = mf_get_vl_mff__(mff->id, vl_mff_map); ++ ++ return vl_mff ? &vl_mff->mf : NULL; + } + + return NULL; +diff --git a/lib/namemap.c b/lib/namemap.c +index 785cda4c27..dd317ea52e 100644 +--- a/lib/namemap.c ++++ b/lib/namemap.c +@@ -90,9 +90,9 @@ void + namemap_destroy(struct namemap *map) + { + if (map) { +- struct namemap_node *node, *next; ++ struct namemap_node *node; + +- HMAP_FOR_EACH_SAFE (node, next, name_node, &map->by_name) { ++ HMAP_FOR_EACH_SAFE (node, name_node, &map->by_name) { + hmap_remove(&map->by_name, &node->name_node); + hmap_remove(&map->by_number, &node->number_node); + free(node->name); +diff --git a/lib/netdev-afxdp.c b/lib/netdev-afxdp.c +index 482400d8d1..ca3f2431ea 100644 +--- a/lib/netdev-afxdp.c ++++ b/lib/netdev-afxdp.c +@@ -235,11 +235,11 @@ netdev_afxdp_cleanup_unused_pool(struct unused_pool *pool) + static void + netdev_afxdp_sweep_unused_pools(void *aux OVS_UNUSED) + { +- struct unused_pool *pool, *next; ++ struct unused_pool *pool; + unsigned int count; + + ovs_mutex_lock(&unused_pools_mutex); +- LIST_FOR_EACH_SAFE (pool, next, list_node, &unused_pools) { ++ LIST_FOR_EACH_SAFE (pool, list_node, &unused_pools) { + + count = umem_pool_count(&pool->umem_info->mpool); + ovs_assert(count + pool->lost_in_rings <= NUM_FRAMES); +diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c +index b6b29c75e3..e28e397d7e 100644 +--- a/lib/netdev-dpdk.c ++++ b/lib/netdev-dpdk.c +@@ -622,9 +622,9 @@ dpdk_mp_full(const struct rte_mempool *mp) OVS_REQUIRES(dpdk_mp_mutex) + static void + dpdk_mp_sweep(void) OVS_REQUIRES(dpdk_mp_mutex) + { +- struct dpdk_mp *dmp, *next; ++ struct dpdk_mp *dmp; + +- LIST_FOR_EACH_SAFE (dmp, next, list_node, &dpdk_mp_list) { ++ LIST_FOR_EACH_SAFE (dmp, list_node, &dpdk_mp_list) { + if (!dmp->refcount && dpdk_mp_full(dmp->mp)) { + VLOG_DBG("Freeing mempool \"%s\"", dmp->mp->name); + ovs_list_remove(&dmp->list_node); +@@ -2561,90 +2561,6 @@ netdev_dpdk_vhost_update_tx_counters(struct netdev_dpdk *dev, + } + } + +-static void +-__netdev_dpdk_vhost_send(struct netdev *netdev, int qid, +- struct dp_packet **pkts, int cnt) +-{ +- struct netdev_dpdk *dev = netdev_dpdk_cast(netdev); +- struct rte_mbuf **cur_pkts = (struct rte_mbuf **) pkts; +- struct netdev_dpdk_sw_stats sw_stats_add; +- unsigned int n_packets_to_free = cnt; +- unsigned int total_packets = cnt; +- int i, retries = 0; +- int max_retries = VHOST_ENQ_RETRY_MIN; +- int vid = netdev_dpdk_get_vid(dev); +- +- qid = dev->tx_q[qid % netdev->n_txq].map; +- +- if (OVS_UNLIKELY(vid < 0 || !dev->vhost_reconfigured || qid < 0 +- || !(dev->flags & NETDEV_UP))) { +- rte_spinlock_lock(&dev->stats_lock); +- dev->stats.tx_dropped+= cnt; +- rte_spinlock_unlock(&dev->stats_lock); +- goto out; +- } +- +- if (OVS_UNLIKELY(!rte_spinlock_trylock(&dev->tx_q[qid].tx_lock))) { +- COVERAGE_INC(vhost_tx_contention); +- rte_spinlock_lock(&dev->tx_q[qid].tx_lock); +- } +- +- sw_stats_add.tx_invalid_hwol_drops = cnt; +- if (userspace_tso_enabled()) { +- cnt = netdev_dpdk_prep_hwol_batch(dev, cur_pkts, cnt); +- } +- +- sw_stats_add.tx_invalid_hwol_drops -= cnt; +- sw_stats_add.tx_mtu_exceeded_drops = cnt; +- cnt = netdev_dpdk_filter_packet_len(dev, cur_pkts, cnt); +- sw_stats_add.tx_mtu_exceeded_drops -= cnt; +- +- /* Check has QoS has been configured for the netdev */ +- sw_stats_add.tx_qos_drops = cnt; +- cnt = netdev_dpdk_qos_run(dev, cur_pkts, cnt, true); +- sw_stats_add.tx_qos_drops -= cnt; +- +- n_packets_to_free = cnt; +- +- do { +- int vhost_qid = qid * VIRTIO_QNUM + VIRTIO_RXQ; +- unsigned int tx_pkts; +- +- tx_pkts = rte_vhost_enqueue_burst(vid, vhost_qid, cur_pkts, cnt); +- if (OVS_LIKELY(tx_pkts)) { +- /* Packets have been sent.*/ +- cnt -= tx_pkts; +- /* Prepare for possible retry.*/ +- cur_pkts = &cur_pkts[tx_pkts]; +- if (OVS_UNLIKELY(cnt && !retries)) { +- /* +- * Read max retries as there are packets not sent +- * and no retries have already occurred. +- */ +- atomic_read_relaxed(&dev->vhost_tx_retries_max, &max_retries); +- } +- } else { +- /* No packets sent - do not retry.*/ +- break; +- } +- } while (cnt && (retries++ < max_retries)); +- +- rte_spinlock_unlock(&dev->tx_q[qid].tx_lock); +- +- sw_stats_add.tx_failure_drops = cnt; +- sw_stats_add.tx_retries = MIN(retries, max_retries); +- +- rte_spinlock_lock(&dev->stats_lock); +- netdev_dpdk_vhost_update_tx_counters(dev, pkts, total_packets, +- &sw_stats_add); +- rte_spinlock_unlock(&dev->stats_lock); +- +-out: +- for (i = 0; i < n_packets_to_free; i++) { +- dp_packet_delete(pkts[i]); +- } +-} +- + static void + netdev_dpdk_extbuf_free(void *addr OVS_UNUSED, void *opaque) + { +@@ -2749,76 +2665,69 @@ dpdk_copy_dp_packet_to_mbuf(struct rte_mempool *mp, struct dp_packet *pkt_orig) + return pkt_dest; + } + +-/* Tx function. Transmit packets indefinitely */ +-static void +-dpdk_do_tx_copy(struct netdev *netdev, int qid, struct dp_packet_batch *batch) +- OVS_NO_THREAD_SAFETY_ANALYSIS +-{ +- const size_t batch_cnt = dp_packet_batch_size(batch); +-#if !defined(__CHECKER__) && !defined(_WIN32) +- const size_t PKT_ARRAY_SIZE = batch_cnt; +-#else +- /* Sparse or MSVC doesn't like variable length array. */ +- enum { PKT_ARRAY_SIZE = NETDEV_MAX_BURST }; +-#endif ++/* Replace packets in a 'batch' with their corresponding copies using ++ * DPDK memory. ++ * ++ * Returns the number of good packets in the batch. */ ++static size_t ++dpdk_copy_batch_to_mbuf(struct netdev *netdev, struct dp_packet_batch *batch) ++{ + struct netdev_dpdk *dev = netdev_dpdk_cast(netdev); +- struct dp_packet *pkts[PKT_ARRAY_SIZE]; +- struct netdev_dpdk_sw_stats *sw_stats = dev->sw_stats; +- uint32_t cnt = batch_cnt; +- uint32_t dropped = 0; +- uint32_t tx_failure = 0; +- uint32_t mtu_drops = 0; +- uint32_t qos_drops = 0; +- +- if (dev->type != DPDK_DEV_VHOST) { +- /* Check if QoS has been configured for this netdev. */ +- cnt = netdev_dpdk_qos_run(dev, (struct rte_mbuf **) batch->packets, +- batch_cnt, false); +- qos_drops = batch_cnt - cnt; +- } +- +- uint32_t txcnt = 0; +- +- for (uint32_t i = 0; i < cnt; i++) { +- struct dp_packet *packet = batch->packets[i]; +- uint32_t size = dp_packet_size(packet); +- +- if (size > dev->max_packet_len +- && !(packet->mbuf.ol_flags & RTE_MBUF_F_TX_TCP_SEG)) { +- VLOG_WARN_RL(&rl, "Too big size %u max_packet_len %d", size, +- dev->max_packet_len); +- mtu_drops++; +- continue; +- } ++ size_t i, size = dp_packet_batch_size(batch); ++ struct dp_packet *packet; + +- pkts[txcnt] = dpdk_copy_dp_packet_to_mbuf(dev->dpdk_mp->mp, packet); +- if (OVS_UNLIKELY(!pkts[txcnt])) { +- dropped = cnt - i; +- break; +- } ++ DP_PACKET_BATCH_REFILL_FOR_EACH (i, size, packet, batch) { ++ if (OVS_UNLIKELY(packet->source == DPBUF_DPDK)) { ++ dp_packet_batch_refill(batch, packet, i); ++ } else { ++ struct dp_packet *pktcopy; + +- txcnt++; +- } ++ pktcopy = dpdk_copy_dp_packet_to_mbuf(dev->dpdk_mp->mp, packet); ++ if (pktcopy) { ++ dp_packet_batch_refill(batch, pktcopy, i); ++ } + +- if (OVS_LIKELY(txcnt)) { +- if (dev->type == DPDK_DEV_VHOST) { +- __netdev_dpdk_vhost_send(netdev, qid, pkts, txcnt); +- } else { +- tx_failure += netdev_dpdk_eth_tx_burst(dev, qid, +- (struct rte_mbuf **)pkts, +- txcnt); ++ dp_packet_delete(packet); + } + } + +- dropped += qos_drops + mtu_drops + tx_failure; +- if (OVS_UNLIKELY(dropped)) { +- rte_spinlock_lock(&dev->stats_lock); +- dev->stats.tx_dropped += dropped; +- sw_stats->tx_failure_drops += tx_failure; +- sw_stats->tx_mtu_exceeded_drops += mtu_drops; +- sw_stats->tx_qos_drops += qos_drops; +- rte_spinlock_unlock(&dev->stats_lock); ++ return dp_packet_batch_size(batch); ++} ++ ++static size_t ++netdev_dpdk_common_send(struct netdev *netdev, struct dp_packet_batch *batch, ++ struct netdev_dpdk_sw_stats *stats) ++{ ++ struct rte_mbuf **pkts = (struct rte_mbuf **) batch->packets; ++ struct netdev_dpdk *dev = netdev_dpdk_cast(netdev); ++ size_t cnt, pkt_cnt = dp_packet_batch_size(batch); ++ ++ memset(stats, 0, sizeof *stats); ++ ++ /* Copy dp-packets to mbufs. */ ++ if (OVS_UNLIKELY(batch->packets[0]->source != DPBUF_DPDK)) { ++ cnt = dpdk_copy_batch_to_mbuf(netdev, batch); ++ stats->tx_failure_drops += pkt_cnt - cnt; ++ pkt_cnt = cnt; + } ++ ++ /* Drop oversized packets. */ ++ cnt = netdev_dpdk_filter_packet_len(dev, pkts, pkt_cnt); ++ stats->tx_mtu_exceeded_drops += pkt_cnt - cnt; ++ pkt_cnt = cnt; ++ ++ /* Prepare each mbuf for hardware offloading. */ ++ if (userspace_tso_enabled()) { ++ cnt = netdev_dpdk_prep_hwol_batch(dev, pkts, pkt_cnt); ++ stats->tx_invalid_hwol_drops += pkt_cnt - cnt; ++ pkt_cnt = cnt; ++ } ++ ++ /* Apply Quality of Service policy. */ ++ cnt = netdev_dpdk_qos_run(dev, pkts, pkt_cnt, true); ++ stats->tx_qos_drops += pkt_cnt - cnt; ++ ++ return cnt; + } + + static int +@@ -2826,25 +2735,92 @@ netdev_dpdk_vhost_send(struct netdev *netdev, int qid, + struct dp_packet_batch *batch, + bool concurrent_txq OVS_UNUSED) + { ++ struct netdev_dpdk *dev = netdev_dpdk_cast(netdev); ++ int max_retries = VHOST_ENQ_RETRY_MIN; ++ int cnt, batch_cnt, vhost_batch_cnt; ++ int vid = netdev_dpdk_get_vid(dev); ++ struct netdev_dpdk_sw_stats stats; ++ struct rte_mbuf **pkts; ++ int retries; + +- if (OVS_UNLIKELY(batch->packets[0]->source != DPBUF_DPDK)) { +- dpdk_do_tx_copy(netdev, qid, batch); ++ batch_cnt = cnt = dp_packet_batch_size(batch); ++ qid = dev->tx_q[qid % netdev->n_txq].map; ++ if (OVS_UNLIKELY(vid < 0 || !dev->vhost_reconfigured || qid < 0 ++ || !(dev->flags & NETDEV_UP))) { ++ rte_spinlock_lock(&dev->stats_lock); ++ dev->stats.tx_dropped += cnt; ++ rte_spinlock_unlock(&dev->stats_lock); + dp_packet_delete_batch(batch, true); +- } else { +- __netdev_dpdk_vhost_send(netdev, qid, batch->packets, +- dp_packet_batch_size(batch)); ++ return 0; ++ } ++ ++ if (OVS_UNLIKELY(!rte_spinlock_trylock(&dev->tx_q[qid].tx_lock))) { ++ COVERAGE_INC(vhost_tx_contention); ++ rte_spinlock_lock(&dev->tx_q[qid].tx_lock); ++ } ++ ++ cnt = netdev_dpdk_common_send(netdev, batch, &stats); ++ ++ pkts = (struct rte_mbuf **) batch->packets; ++ vhost_batch_cnt = cnt; ++ retries = 0; ++ do { ++ int vhost_qid = qid * VIRTIO_QNUM + VIRTIO_RXQ; ++ int tx_pkts; ++ ++ tx_pkts = rte_vhost_enqueue_burst(vid, vhost_qid, pkts, cnt); ++ if (OVS_LIKELY(tx_pkts)) { ++ /* Packets have been sent.*/ ++ cnt -= tx_pkts; ++ /* Prepare for possible retry.*/ ++ pkts = &pkts[tx_pkts]; ++ if (OVS_UNLIKELY(cnt && !retries)) { ++ /* ++ * Read max retries as there are packets not sent ++ * and no retries have already occurred. ++ */ ++ atomic_read_relaxed(&dev->vhost_tx_retries_max, &max_retries); ++ } ++ } else { ++ /* No packets sent - do not retry.*/ ++ break; ++ } ++ } while (cnt && (retries++ < max_retries)); ++ ++ rte_spinlock_unlock(&dev->tx_q[qid].tx_lock); ++ ++ stats.tx_failure_drops += cnt; ++ stats.tx_retries = MIN(retries, max_retries); ++ ++ rte_spinlock_lock(&dev->stats_lock); ++ netdev_dpdk_vhost_update_tx_counters(dev, batch->packets, batch_cnt, ++ &stats); ++ rte_spinlock_unlock(&dev->stats_lock); ++ ++ pkts = (struct rte_mbuf **) batch->packets; ++ for (int i = 0; i < vhost_batch_cnt; i++) { ++ rte_pktmbuf_free(pkts[i]); + } ++ + return 0; + } + +-static inline void +-netdev_dpdk_send__(struct netdev_dpdk *dev, int qid, +- struct dp_packet_batch *batch, +- bool concurrent_txq) ++static int ++netdev_dpdk_eth_send(struct netdev *netdev, int qid, ++ struct dp_packet_batch *batch, bool concurrent_txq) + { ++ struct rte_mbuf **pkts = (struct rte_mbuf **) batch->packets; ++ struct netdev_dpdk *dev = netdev_dpdk_cast(netdev); ++ int batch_cnt = dp_packet_batch_size(batch); ++ struct netdev_dpdk_sw_stats stats; ++ int cnt, dropped; ++ + if (OVS_UNLIKELY(!(dev->flags & NETDEV_UP))) { ++ rte_spinlock_lock(&dev->stats_lock); ++ dev->stats.tx_dropped += dp_packet_batch_size(batch); ++ rte_spinlock_unlock(&dev->stats_lock); + dp_packet_delete_batch(batch, true); +- return; ++ return 0; + } + + if (OVS_UNLIKELY(concurrent_txq)) { +@@ -2852,56 +2828,27 @@ netdev_dpdk_send__(struct netdev_dpdk *dev, int qid, + rte_spinlock_lock(&dev->tx_q[qid].tx_lock); + } + +- if (OVS_UNLIKELY(batch->packets[0]->source != DPBUF_DPDK)) { +- struct netdev *netdev = &dev->up; ++ cnt = netdev_dpdk_common_send(netdev, batch, &stats); + +- dpdk_do_tx_copy(netdev, qid, batch); +- dp_packet_delete_batch(batch, true); +- } else { ++ dropped = batch_cnt - cnt; ++ ++ dropped += netdev_dpdk_eth_tx_burst(dev, qid, pkts, cnt); ++ if (OVS_UNLIKELY(dropped)) { + struct netdev_dpdk_sw_stats *sw_stats = dev->sw_stats; +- int dropped; +- int tx_failure, mtu_drops, qos_drops, hwol_drops; +- int batch_cnt = dp_packet_batch_size(batch); +- struct rte_mbuf **pkts = (struct rte_mbuf **) batch->packets; + +- hwol_drops = batch_cnt; +- if (userspace_tso_enabled()) { +- batch_cnt = netdev_dpdk_prep_hwol_batch(dev, pkts, batch_cnt); +- } +- hwol_drops -= batch_cnt; +- mtu_drops = batch_cnt; +- batch_cnt = netdev_dpdk_filter_packet_len(dev, pkts, batch_cnt); +- mtu_drops -= batch_cnt; +- qos_drops = batch_cnt; +- batch_cnt = netdev_dpdk_qos_run(dev, pkts, batch_cnt, true); +- qos_drops -= batch_cnt; +- +- tx_failure = netdev_dpdk_eth_tx_burst(dev, qid, pkts, batch_cnt); +- +- dropped = tx_failure + mtu_drops + qos_drops + hwol_drops; +- if (OVS_UNLIKELY(dropped)) { +- rte_spinlock_lock(&dev->stats_lock); +- dev->stats.tx_dropped += dropped; +- sw_stats->tx_failure_drops += tx_failure; +- sw_stats->tx_mtu_exceeded_drops += mtu_drops; +- sw_stats->tx_qos_drops += qos_drops; +- sw_stats->tx_invalid_hwol_drops += hwol_drops; +- rte_spinlock_unlock(&dev->stats_lock); +- } ++ rte_spinlock_lock(&dev->stats_lock); ++ dev->stats.tx_dropped += dropped; ++ sw_stats->tx_failure_drops += stats.tx_failure_drops; ++ sw_stats->tx_mtu_exceeded_drops += stats.tx_mtu_exceeded_drops; ++ sw_stats->tx_qos_drops += stats.tx_qos_drops; ++ sw_stats->tx_invalid_hwol_drops += stats.tx_invalid_hwol_drops; ++ rte_spinlock_unlock(&dev->stats_lock); + } + + if (OVS_UNLIKELY(concurrent_txq)) { + rte_spinlock_unlock(&dev->tx_q[qid].tx_lock); + } +-} +- +-static int +-netdev_dpdk_eth_send(struct netdev *netdev, int qid, +- struct dp_packet_batch *batch, bool concurrent_txq) +-{ +- struct netdev_dpdk *dev = netdev_dpdk_cast(netdev); + +- netdev_dpdk_send__(dev, qid, batch, concurrent_txq); + return 0; + } + +@@ -4686,11 +4633,11 @@ trtcm_policer_qos_construct(const struct smap *details, + static void + trtcm_policer_qos_destruct(struct qos_conf *conf) + { +- struct trtcm_policer_queue *queue, *next_queue; ++ struct trtcm_policer_queue *queue; + struct trtcm_policer *policer = CONTAINER_OF(conf, struct trtcm_policer, + qos_conf); + +- HMAP_FOR_EACH_SAFE (queue, next_queue, hmap_node, &policer->queues) { ++ HMAP_FOR_EACH_SAFE (queue, hmap_node, &policer->queues) { + hmap_remove(&policer->queues, &queue->hmap_node); + free(queue); + } +diff --git a/lib/netdev-linux.c b/lib/netdev-linux.c +index 620a451dec..1b9c2874ac 100644 +--- a/lib/netdev-linux.c ++++ b/lib/netdev-linux.c +@@ -247,6 +247,14 @@ enum { + VALID_NUMA_ID = 1 << 8, + }; + ++/* Linux 4.4 introduced the ability to skip the internal stats gathering ++ * that netlink does via an external filter mask that can be passed into ++ * a netlink request. ++ */ ++#ifndef RTEXT_FILTER_SKIP_STATS ++#define RTEXT_FILTER_SKIP_STATS (1 << 3) ++#endif ++ + /* Use one for the packet buffer and another for the aux buffer to receive + * TSO packets. */ + #define IOV_STD_SIZE 1 +@@ -682,7 +690,10 @@ netdev_linux_update_lag(struct rtnetlink_change *change) + return; + } + +- if (is_netdev_linux_class(master_netdev->netdev_class)) { ++ /* If LAG master is not attached to ovs, ingress block on LAG ++ * members shoud not be updated. */ ++ if (!master_netdev->auto_classified && ++ is_netdev_linux_class(master_netdev->netdev_class)) { + block_id = netdev_get_block_id(master_netdev); + if (!block_id) { + netdev_close(master_netdev); +@@ -2623,7 +2634,7 @@ static void + nl_msg_act_police_end_nest(struct ofpbuf *request, size_t offset, + size_t act_offset) + { +- nl_msg_put_u32(request, TCA_POLICE_RESULT, TC_ACT_PIPE); ++ nl_msg_put_u32(request, TCA_POLICE_RESULT, TC_ACT_UNSPEC); + nl_msg_end_nested(request, offset); + nl_msg_end_nested(request, act_offset); + } +@@ -5331,11 +5342,11 @@ static void + hfsc_tc_destroy(struct tc *tc) + { + struct hfsc *hfsc; +- struct hfsc_class *hc, *next; ++ struct hfsc_class *hc; + + hfsc = CONTAINER_OF(tc, struct hfsc, tc); + +- HMAP_FOR_EACH_SAFE (hc, next, tc_queue.hmap_node, &hfsc->tc.queues) { ++ HMAP_FOR_EACH_SAFE (hc, tc_queue.hmap_node, &hfsc->tc.queues) { + hmap_remove(&hfsc->tc.queues, &hc->tc_queue.hmap_node); + free(hc); + } +@@ -6295,7 +6306,14 @@ get_stats_via_netlink(const struct netdev *netdev_, struct netdev_stats *stats) + if (ofpbuf_try_pull(reply, NLMSG_HDRLEN + sizeof(struct ifinfomsg))) { + const struct nlattr *a = nl_attr_find(reply, 0, IFLA_STATS64); + if (a && nl_attr_get_size(a) >= sizeof(struct rtnl_link_stats64)) { +- netdev_stats_from_rtnl_link_stats64(stats, nl_attr_get(a)); ++ const struct rtnl_link_stats64 *lstats = nl_attr_get(a); ++ struct rtnl_link_stats64 aligned_lstats; ++ ++ if (!IS_PTR_ALIGNED(lstats)) { ++ memcpy(&aligned_lstats, lstats, sizeof aligned_lstats); ++ lstats = &aligned_lstats; ++ } ++ netdev_stats_from_rtnl_link_stats64(stats, lstats); + error = 0; + } else { + a = nl_attr_find(reply, 0, IFLA_STATS); +@@ -6411,6 +6429,9 @@ netdev_linux_update_via_netlink(struct netdev_linux *netdev) + if (netdev_linux_netnsid_is_remote(netdev)) { + nl_msg_put_u32(&request, IFLA_IF_NETNSID, netdev->netnsid); + } ++ ++ nl_msg_put_u32(&request, IFLA_EXT_MASK, RTEXT_FILTER_SKIP_STATS); ++ + error = nl_transact(NETLINK_ROUTE, &request, &reply); + ofpbuf_uninit(&request); + if (error) { +diff --git a/lib/netdev-offload-dpdk.c b/lib/netdev-offload-dpdk.c +index 94dc6a9b74..be6adc32ff 100644 +--- a/lib/netdev-offload-dpdk.c ++++ b/lib/netdev-offload-dpdk.c +@@ -18,6 +18,7 @@ + + #include + #include ++#include + #include + #include + +@@ -363,6 +364,8 @@ dump_flow_pattern(struct ds *s, + + ds_put_cstr(s, "eth "); + if (eth_spec) { ++ uint32_t has_vlan_mask; ++ + if (!eth_mask) { + eth_mask = &rte_flow_item_eth_mask; + } +@@ -377,6 +380,9 @@ dump_flow_pattern(struct ds *s, + DUMP_PATTERN_ITEM(eth_mask->type, false, "type", "0x%04"PRIx16, + ntohs(eth_spec->type), + ntohs(eth_mask->type), 0); ++ has_vlan_mask = eth_mask->has_vlan ? UINT32_MAX : 0; ++ DUMP_PATTERN_ITEM(has_vlan_mask, false, "has_vlan", "%d", ++ eth_spec->has_vlan, eth_mask->has_vlan, 0); + } + ds_put_cstr(s, "/ "); + } else if (item->type == RTE_FLOW_ITEM_TYPE_VLAN) { +@@ -1369,6 +1375,7 @@ parse_flow_match(struct netdev *netdev, + struct flow_patterns *patterns, + struct match *match) + { ++ struct rte_flow_item_eth *eth_spec = NULL, *eth_mask = NULL; + struct flow *consumed_masks; + uint8_t proto = 0; + +@@ -1414,6 +1421,11 @@ parse_flow_match(struct netdev *netdev, + memset(&consumed_masks->dl_src, 0, sizeof consumed_masks->dl_src); + consumed_masks->dl_type = 0; + ++ spec->has_vlan = 0; ++ mask->has_vlan = 1; ++ eth_spec = spec; ++ eth_mask = mask; ++ + add_flow_pattern(patterns, RTE_FLOW_ITEM_TYPE_ETH, spec, mask, NULL); + } + +@@ -1427,8 +1439,14 @@ parse_flow_match(struct netdev *netdev, + spec->tci = match->flow.vlans[0].tci & ~htons(VLAN_CFI); + mask->tci = match->wc.masks.vlans[0].tci & ~htons(VLAN_CFI); + +- /* Match any protocols. */ +- mask->inner_type = 0; ++ if (eth_spec && eth_mask) { ++ eth_spec->has_vlan = 1; ++ eth_mask->has_vlan = 1; ++ spec->inner_type = eth_spec->type; ++ mask->inner_type = eth_mask->type; ++ eth_spec->type = match->flow.vlans[0].tpid; ++ eth_mask->type = match->wc.masks.vlans[0].tpid; ++ } + + add_flow_pattern(patterns, RTE_FLOW_ITEM_TYPE_VLAN, spec, mask, NULL); + } +@@ -1697,7 +1715,7 @@ add_flow_mark_rss_actions(struct flow_actions *actions, + .conf = (struct rte_flow_action_rss) { + .func = RTE_ETH_HASH_FUNCTION_DEFAULT, + .level = 0, +- .types = 0, ++ .types = RTE_ETH_RSS_IP | RTE_ETH_RSS_UDP | RTE_ETH_RSS_TCP, + .queue_num = netdev_n_rxq(netdev), + .queue = rss_data->queue, + .key_len = 0, +diff --git a/lib/netdev-offload-tc.c b/lib/netdev-offload-tc.c +index 9845e8d3fe..93321989a9 100644 +--- a/lib/netdev-offload-tc.c ++++ b/lib/netdev-offload-tc.c +@@ -44,6 +44,7 @@ + VLOG_DEFINE_THIS_MODULE(netdev_offload_tc); + + static struct vlog_rate_limit error_rl = VLOG_RATE_LIMIT_INIT(60, 5); ++static struct vlog_rate_limit warn_rl = VLOG_RATE_LIMIT_INIT(10, 2); + + static struct hmap ufid_to_tc = HMAP_INITIALIZER(&ufid_to_tc); + static struct hmap tc_to_ufid = HMAP_INITIALIZER(&tc_to_ufid); +@@ -417,11 +418,11 @@ delete_chains_from_netdev(struct netdev *netdev, struct tcf_id *id) + static int + netdev_tc_flow_flush(struct netdev *netdev) + { +- struct ufid_tc_data *data, *next; ++ struct ufid_tc_data *data; + int err; + + ovs_mutex_lock(&ufid_lock); +- HMAP_FOR_EACH_SAFE (data, next, tc_to_ufid_node, &tc_to_ufid) { ++ HMAP_FOR_EACH_SAFE (data, tc_to_ufid_node, &tc_to_ufid) { + if (data->netdev != netdev) { + continue; + } +@@ -481,10 +482,10 @@ netdev_tc_flow_dump_destroy(struct netdev_flow_dump *dump) + + static void + parse_flower_rewrite_to_netlink_action(struct ofpbuf *buf, +- struct tc_flower *flower) ++ struct tc_action *action) + { +- char *mask = (char *) &flower->rewrite.mask; +- char *data = (char *) &flower->rewrite.key; ++ char *mask = (char *) &action->rewrite.mask; ++ char *data = (char *) &action->rewrite.key; + + for (int type = 0; type < ARRAY_SIZE(set_flower_map); type++) { + char *put = NULL; +@@ -550,30 +551,42 @@ flower_tun_opt_to_match(struct match *match, struct tc_flower *flower) + struct geneve_opt *opt, *opt_mask; + int len, cnt = 0; + ++ /* Options are always in UDPIF format in the 'flower'. */ ++ match->flow.tunnel.flags |= FLOW_TNL_F_UDPIF; ++ match->wc.masks.tunnel.flags |= FLOW_TNL_F_UDPIF; ++ ++ match->flow.tunnel.metadata.present.len = ++ flower->key.tunnel.metadata.present.len; ++ /* In the 'flower' mask len is an actual length, not a mask. But in the ++ * 'match' it is an actual mask, so should be an exact match, because TC ++ * will always match on the exact value. */ ++ match->wc.masks.tunnel.metadata.present.len = 0xff; ++ ++ if (!flower->key.tunnel.metadata.present.len) { ++ /* No options present. */ ++ return; ++ } ++ + memcpy(match->flow.tunnel.metadata.opts.gnv, + flower->key.tunnel.metadata.opts.gnv, + flower->key.tunnel.metadata.present.len); +- match->flow.tunnel.metadata.present.len = +- flower->key.tunnel.metadata.present.len; +- match->flow.tunnel.flags |= FLOW_TNL_F_UDPIF; + memcpy(match->wc.masks.tunnel.metadata.opts.gnv, + flower->mask.tunnel.metadata.opts.gnv, + flower->mask.tunnel.metadata.present.len); + ++ /* Fixing up 'length' fields of particular options, since these are ++ * also not masks, but actual lengths in the 'flower' structure. */ + len = flower->key.tunnel.metadata.present.len; + while (len) { + opt = &match->flow.tunnel.metadata.opts.gnv[cnt]; + opt_mask = &match->wc.masks.tunnel.metadata.opts.gnv[cnt]; + ++ /* "Exact" match as set in tun_metadata_to_geneve_mask__(). */ + opt_mask->length = 0x1f; + + cnt += sizeof(struct geneve_opt) / 4 + opt->length; + len -= sizeof(struct geneve_opt) + opt->length * 4; + } +- +- match->wc.masks.tunnel.metadata.present.len = +- flower->mask.tunnel.metadata.present.len; +- match->wc.masks.tunnel.flags |= FLOW_TNL_F_UDPIF; + } + + static void +@@ -585,8 +598,10 @@ parse_tc_flower_to_stats(struct tc_flower *flower, + } + + memset(stats, 0, sizeof *stats); +- stats->n_packets = get_32aligned_u64(&flower->stats.n_packets); +- stats->n_bytes = get_32aligned_u64(&flower->stats.n_bytes); ++ stats->n_packets = get_32aligned_u64(&flower->stats_sw.n_packets); ++ stats->n_packets += get_32aligned_u64(&flower->stats_hw.n_packets); ++ stats->n_bytes = get_32aligned_u64(&flower->stats_sw.n_bytes); ++ stats->n_bytes += get_32aligned_u64(&flower->stats_hw.n_bytes); + stats->used = flower->lastused; + } + +@@ -616,7 +631,8 @@ parse_tc_flower_terse_to_match(struct tc_flower *flower, + } + + static int +-parse_tc_flower_to_match(struct tc_flower *flower, ++parse_tc_flower_to_match(const struct netdev *netdev, ++ struct tc_flower *flower, + struct match *match, + struct nlattr **actions, + struct dpif_flow_stats *stats, +@@ -803,18 +819,24 @@ parse_tc_flower_to_match(struct tc_flower *flower, + &flower->key.tunnel.ipv6.ipv6_src, + &flower->mask.tunnel.ipv6.ipv6_src); + } +- if (flower->key.tunnel.tos) { ++ if (flower->mask.tunnel.tos) { + match_set_tun_tos_masked(match, flower->key.tunnel.tos, + flower->mask.tunnel.tos); + } +- if (flower->key.tunnel.ttl) { ++ if (flower->mask.tunnel.ttl) { + match_set_tun_ttl_masked(match, flower->key.tunnel.ttl, + flower->mask.tunnel.ttl); + } +- if (flower->key.tunnel.tp_dst) { +- match_set_tun_tp_dst(match, flower->key.tunnel.tp_dst); ++ if (flower->mask.tunnel.tp_src) { ++ match_set_tun_tp_dst_masked(match, flower->key.tunnel.tp_src, ++ flower->mask.tunnel.tp_src); + } +- if (flower->key.tunnel.metadata.present.len) { ++ if (flower->mask.tunnel.tp_dst) { ++ match_set_tun_tp_dst_masked(match, flower->key.tunnel.tp_dst, ++ flower->mask.tunnel.tp_dst); ++ } ++ ++ if (!strcmp(netdev_get_type(netdev), "geneve")) { + flower_tun_opt_to_match(match, flower); + } + } +@@ -877,7 +899,7 @@ parse_tc_flower_to_match(struct tc_flower *flower, + } + break; + case TC_ACT_PEDIT: { +- parse_flower_rewrite_to_netlink_action(buf, flower); ++ parse_flower_rewrite_to_netlink_action(buf, action); + } + break; + case TC_ACT_ENCAP: { +@@ -1054,8 +1076,8 @@ netdev_tc_flow_dump_next(struct netdev_flow_dump *dump, + continue; + } + +- if (parse_tc_flower_to_match(&flower, match, actions, stats, attrs, +- wbuffer, dump->terse)) { ++ if (parse_tc_flower_to_match(netdev, &flower, match, actions, ++ stats, attrs, wbuffer, dump->terse)) { + continue; + } + +@@ -1222,8 +1244,8 @@ parse_put_flow_set_masked_action(struct tc_flower *flower, + uint64_t set_stub[1024 / 8]; + struct ofpbuf set_buf = OFPBUF_STUB_INITIALIZER(set_stub); + char *set_data, *set_mask; +- char *key = (char *) &flower->rewrite.key; +- char *mask = (char *) &flower->rewrite.mask; ++ char *key = (char *) &action->rewrite.key; ++ char *mask = (char *) &action->rewrite.mask; + const struct nlattr *attr; + int i, j, type; + size_t size; +@@ -1265,14 +1287,6 @@ parse_put_flow_set_masked_action(struct tc_flower *flower, + } + } + +- if (!is_all_zeros(&flower->rewrite, sizeof flower->rewrite)) { +- if (flower->rewrite.rewrite == false) { +- flower->rewrite.rewrite = true; +- action->type = TC_ACT_PEDIT; +- flower->action_count++; +- } +- } +- + if (hasmask && !is_all_zeros(set_mask, size)) { + VLOG_DBG_RL(&rl, "unsupported sub attribute of set action type %d", + type); +@@ -1281,6 +1295,8 @@ parse_put_flow_set_masked_action(struct tc_flower *flower, + } + + ofpbuf_uninit(&set_buf); ++ action->type = TC_ACT_PEDIT; ++ flower->action_count++; + return 0; + } + +@@ -1288,6 +1304,7 @@ static int + parse_put_flow_set_action(struct tc_flower *flower, struct tc_action *action, + const struct nlattr *set, size_t set_len) + { ++ static struct vlog_rate_limit rl = VLOG_RATE_LIMIT_INIT(5, 20); + const struct nlattr *tunnel; + const struct nlattr *tun_attr; + size_t tun_left, tunnel_len; +@@ -1306,6 +1323,7 @@ parse_put_flow_set_action(struct tc_flower *flower, struct tc_action *action, + + action->type = TC_ACT_ENCAP; + action->encap.id_present = false; ++ action->encap.no_csum = 1; + flower->action_count++; + NL_ATTR_FOR_EACH_UNSAFE(tun_attr, tun_left, tunnel, tunnel_len) { + switch (nl_attr_type(tun_attr)) { +@@ -1330,6 +1348,18 @@ parse_put_flow_set_action(struct tc_flower *flower, struct tc_action *action, + action->encap.ttl = nl_attr_get_u8(tun_attr); + } + break; ++ case OVS_TUNNEL_KEY_ATTR_DONT_FRAGMENT: { ++ /* XXX: This is wrong! We're ignoring the DF flag configuration ++ * requested by the user. However, TC for now has no way to pass ++ * that flag and it is set by default, meaning tunnel offloading ++ * will not work if 'options:df_default=false' is not set. ++ * Keeping incorrect behavior for now. */ ++ } ++ break; ++ case OVS_TUNNEL_KEY_ATTR_CSUM: { ++ action->encap.no_csum = 0; ++ } ++ break; + case OVS_TUNNEL_KEY_ATTR_IPV6_SRC: { + action->encap.ipv6.ipv6_src = + nl_attr_get_in6_addr(tun_attr); +@@ -1354,12 +1384,31 @@ parse_put_flow_set_action(struct tc_flower *flower, struct tc_action *action, + action->encap.data.present.len = nl_attr_get_size(tun_attr); + } + break; ++ default: ++ VLOG_DBG_RL(&rl, "unsupported tunnel key attribute %d", ++ nl_attr_type(tun_attr)); ++ return EOPNOTSUPP; + } + } + + return 0; + } + ++static bool ++is_ipv6_fragment_and_masked(const struct flow *key, const struct flow *mask) ++{ ++ if (key->dl_type != htons(ETH_P_IPV6)) { ++ return false; ++ } ++ if (mask->nw_proto && key->nw_proto == IPPROTO_FRAGMENT) { ++ return true; ++ } ++ if (key->nw_frag & (mask->nw_frag & FLOW_NW_FRAG_ANY)) { ++ return true; ++ } ++ return false; ++} ++ + static int + test_key_and_mask(struct match *match) + { +@@ -1442,8 +1491,23 @@ test_key_and_mask(struct match *match) + return EOPNOTSUPP; + } + ++ if (is_ipv6_fragment_and_masked(key, mask)) { ++ VLOG_DBG_RL(&rl, "offloading of IPv6 fragments isn't supported"); ++ return EOPNOTSUPP; ++ } ++ + if (!is_all_zeros(mask, sizeof *mask)) { +- VLOG_DBG_RL(&rl, "offloading isn't supported, unknown attribute"); ++ if (!VLOG_DROP_DBG(&rl)) { ++ struct ds ds = DS_EMPTY_INITIALIZER; ++ ++ ds_put_cstr(&ds, ++ "offloading isn't supported, unknown attribute\n" ++ "Unused mask bits:\n"); ++ ds_put_sparse_hex_dump(&ds, mask, sizeof *mask, 0, false); ++ ++ VLOG_DBG("%s", ds_cstr(&ds)); ++ ds_destroy(&ds); ++ } + return EOPNOTSUPP; + } + +@@ -1452,18 +1516,51 @@ test_key_and_mask(struct match *match) + + static void + flower_match_to_tun_opt(struct tc_flower *flower, const struct flow_tnl *tnl, +- const struct flow_tnl *tnl_mask) ++ struct flow_tnl *tnl_mask) + { + struct geneve_opt *opt, *opt_mask; + int len, cnt = 0; + +- memcpy(flower->key.tunnel.metadata.opts.gnv, tnl->metadata.opts.gnv, +- tnl->metadata.present.len); ++ /* 'flower' always has an exact match on tunnel metadata length, so having ++ * it in a wrong format is not acceptable unless it is empty. */ ++ if (!(tnl->flags & FLOW_TNL_F_UDPIF)) { ++ if (tnl->metadata.present.map) { ++ /* XXX: Add non-UDPIF format parsing here? */ ++ VLOG_WARN_RL(&warn_rl, "Tunnel options are in the wrong format."); ++ } else { ++ /* There are no options, that equals for them to be in UDPIF format ++ * with a zero 'len'. Clearing the 'map' mask as consumed. ++ * No need to explicitly set 'len' to zero in the 'flower'. */ ++ tnl_mask->flags &= ~FLOW_TNL_F_UDPIF; ++ memset(&tnl_mask->metadata.present.map, 0, ++ sizeof tnl_mask->metadata.present.map); ++ } ++ return; ++ } ++ ++ tnl_mask->flags &= ~FLOW_TNL_F_UDPIF; ++ + flower->key.tunnel.metadata.present.len = tnl->metadata.present.len; ++ /* Copying from the key and not from the mask, since in the 'flower' ++ * the length for a mask is not a mask, but the actual length. TC ++ * will use an exact match for the length. */ ++ flower->mask.tunnel.metadata.present.len = tnl->metadata.present.len; ++ memset(&tnl_mask->metadata.present.len, 0, ++ sizeof tnl_mask->metadata.present.len); + ++ if (!tnl->metadata.present.len) { ++ return; ++ } ++ ++ memcpy(flower->key.tunnel.metadata.opts.gnv, tnl->metadata.opts.gnv, ++ tnl->metadata.present.len); + memcpy(flower->mask.tunnel.metadata.opts.gnv, tnl_mask->metadata.opts.gnv, + tnl->metadata.present.len); + ++ memset(tnl_mask->metadata.opts.gnv, 0, tnl->metadata.present.len); ++ ++ /* Fixing up 'length' fields of particular options, since these are ++ * also not masks, but actual lengths in the 'flower' structure. */ + len = flower->key.tunnel.metadata.present.len; + while (len) { + opt = &flower->key.tunnel.metadata.opts.gnv[cnt]; +@@ -1474,8 +1571,6 @@ flower_match_to_tun_opt(struct tc_flower *flower, const struct flow_tnl *tnl, + cnt += sizeof(struct geneve_opt) / 4 + opt->length; + len -= sizeof(struct geneve_opt) + opt->length * 4; + } +- +- flower->mask.tunnel.metadata.present.len = tnl->metadata.present.len; + } + + static void +@@ -1541,6 +1636,12 @@ parse_match_ct_state_to_flower(struct tc_flower *flower, struct match *match) + flower->key.ct_state &= ~(TCA_FLOWER_KEY_CT_FLAGS_NEW); + flower->mask.ct_state &= ~(TCA_FLOWER_KEY_CT_FLAGS_NEW); + } ++ ++ if (flower->key.ct_state && ++ !(flower->key.ct_state & TCA_FLOWER_KEY_CT_FLAGS_TRACKED)) { ++ flower->key.ct_state |= TCA_FLOWER_KEY_CT_FLAGS_TRACKED; ++ flower->mask.ct_state |= TCA_FLOWER_KEY_CT_FLAGS_TRACKED; ++ } + } + + if (mask->ct_zone) { +@@ -1574,7 +1675,7 @@ netdev_tc_flow_put(struct netdev *netdev, struct match *match, + const struct flow *key = &match->flow; + struct flow *mask = &match->wc.masks; + const struct flow_tnl *tnl = &match->flow.tunnel; +- const struct flow_tnl *tnl_mask = &mask->tunnel; ++ struct flow_tnl *tnl_mask = &mask->tunnel; + struct tc_action *action; + bool recirc_act = false; + uint32_t block_id = 0; +@@ -1615,17 +1716,49 @@ netdev_tc_flow_put(struct netdev *netdev, struct match *match, + flower.key.tunnel.ttl = tnl->ip_ttl; + flower.key.tunnel.tp_src = tnl->tp_src; + flower.key.tunnel.tp_dst = tnl->tp_dst; ++ + flower.mask.tunnel.ipv4.ipv4_src = tnl_mask->ip_src; + flower.mask.tunnel.ipv4.ipv4_dst = tnl_mask->ip_dst; + flower.mask.tunnel.ipv6.ipv6_src = tnl_mask->ipv6_src; + flower.mask.tunnel.ipv6.ipv6_dst = tnl_mask->ipv6_dst; + flower.mask.tunnel.tos = tnl_mask->ip_tos; + flower.mask.tunnel.ttl = tnl_mask->ip_ttl; ++ flower.mask.tunnel.tp_src = tnl_mask->tp_src; ++ /* XXX: We should be setting the mask from 'tnl_mask->tp_dst' here, but ++ * some hardware drivers (mlx5) doesn't support masked matches and will ++ * refuse to offload such flows keeping them in software path. ++ * Degrading the flow down to exact match for now as a workaround. */ ++ flower.mask.tunnel.tp_dst = OVS_BE16_MAX; + flower.mask.tunnel.id = (tnl->flags & FLOW_TNL_F_KEY) ? tnl_mask->tun_id : 0; +- flower_match_to_tun_opt(&flower, tnl, tnl_mask); ++ ++ memset(&tnl_mask->ip_src, 0, sizeof tnl_mask->ip_src); ++ memset(&tnl_mask->ip_dst, 0, sizeof tnl_mask->ip_dst); ++ memset(&tnl_mask->ipv6_src, 0, sizeof tnl_mask->ipv6_src); ++ memset(&tnl_mask->ipv6_dst, 0, sizeof tnl_mask->ipv6_dst); ++ memset(&tnl_mask->ip_tos, 0, sizeof tnl_mask->ip_tos); ++ memset(&tnl_mask->ip_ttl, 0, sizeof tnl_mask->ip_ttl); ++ memset(&tnl_mask->tp_src, 0, sizeof tnl_mask->tp_src); ++ memset(&tnl_mask->tp_dst, 0, sizeof tnl_mask->tp_dst); ++ ++ memset(&tnl_mask->tun_id, 0, sizeof tnl_mask->tun_id); ++ tnl_mask->flags &= ~FLOW_TNL_F_KEY; ++ ++ /* XXX: This is wrong! We're ignoring DF and CSUM flags configuration ++ * requested by the user. However, TC for now has no way to pass ++ * these flags in a flower key and their masks are set by default, ++ * meaning tunnel offloading will not work at all if not cleared. ++ * Keeping incorrect behavior for now. */ ++ tnl_mask->flags &= ~(FLOW_TNL_F_DONT_FRAGMENT | FLOW_TNL_F_CSUM); ++ ++ if (!strcmp(netdev_get_type(netdev), "geneve")) { ++ flower_match_to_tun_opt(&flower, tnl, tnl_mask); ++ } + flower.tunnel = true; ++ } else { ++ /* There is no tunnel metadata to match on, but there could be some ++ * mask bits set due to flow translation artifacts. Clear them. */ ++ memset(&mask->tunnel, 0, sizeof mask->tunnel); + } +- memset(&mask->tunnel, 0, sizeof mask->tunnel); + + flower.key.eth_type = key->dl_type; + flower.mask.eth_type = mask->dl_type; +@@ -1638,7 +1771,7 @@ netdev_tc_flow_put(struct netdev *netdev, struct match *match, + + if (mask->vlans[0].tpid && eth_type_vlan(key->vlans[0].tpid)) { + flower.key.encap_eth_type[0] = flower.key.eth_type; +- flower.mask.encap_eth_type[0] = flower.mask.eth_type; ++ flower.mask.encap_eth_type[0] = CONSTANT_HTONS(0xffff); + flower.key.eth_type = key->vlans[0].tpid; + flower.mask.eth_type = mask->vlans[0].tpid; + } +@@ -1734,7 +1867,7 @@ netdev_tc_flow_put(struct netdev *netdev, struct match *match, + memset(&mask->arp_tha, 0, sizeof mask->arp_tha); + } + +- if (is_ip_any(key)) { ++ if (is_ip_any(key) && !is_ipv6_fragment_and_masked(key, mask)) { + flower.key.ip_proto = key->nw_proto; + flower.mask.ip_proto = mask->nw_proto; + mask->nw_proto = 0; +@@ -1841,7 +1974,25 @@ netdev_tc_flow_put(struct netdev *netdev, struct match *match, + VLOG_DBG_RL(&rl, "Can't find netdev for output port %d", port); + return ENODEV; + } ++ ++ if (!netdev_flow_api_equals(netdev, outdev)) { ++ VLOG_DBG_RL(&rl, ++ "Flow API provider mismatch between ingress (%s) " ++ "and egress (%s) ports", ++ netdev_get_name(netdev), netdev_get_name(outdev)); ++ netdev_close(outdev); ++ return EOPNOTSUPP; ++ } ++ + action->out.ifindex_out = netdev_get_ifindex(outdev); ++ if (action->out.ifindex_out < 0) { ++ VLOG_DBG_RL(&rl, ++ "Can't find ifindex for output port %s, error %d", ++ netdev_get_name(outdev), action->out.ifindex_out); ++ netdev_close(outdev); ++ return -action->out.ifindex_out; ++ } ++ + action->out.ingress = is_internal_port(netdev_get_type(outdev)); + action->type = TC_ACT_OUTPUT; + flower.action_count++; +@@ -1879,10 +2030,6 @@ netdev_tc_flow_put(struct netdev *netdev, struct match *match, + if (err) { + return err; + } +- if (action->type == TC_ACT_ENCAP) { +- action->encap.tp_dst = info->tp_dst_port; +- action->encap.no_csum = !info->tunnel_csum_on; +- } + } else if (nl_attr_type(nla) == OVS_ACTION_ATTR_SET_MASKED) { + const struct nlattr *set = nl_attr_get(nla); + const size_t set_len = nl_attr_get_size(nla); +@@ -1989,7 +2136,8 @@ netdev_tc_flow_get(struct netdev *netdev, + } + + in_port = netdev_ifindex_to_odp_port(id.ifindex); +- parse_tc_flower_to_match(&flower, match, actions, stats, attrs, buf, false); ++ parse_tc_flower_to_match(netdev, &flower, match, actions, ++ stats, attrs, buf, false); + + match->wc.masks.in_port.odp_port = u32_to_odp(UINT32_MAX); + match->flow.in_port.odp_port = in_port; +@@ -2015,9 +2163,7 @@ netdev_tc_flow_del(struct netdev *netdev OVS_UNUSED, + if (stats) { + memset(stats, 0, sizeof *stats); + if (!tc_get_flower(&id, &flower)) { +- stats->n_packets = get_32aligned_u64(&flower.stats.n_packets); +- stats->n_bytes = get_32aligned_u64(&flower.stats.n_bytes); +- stats->used = flower.lastused; ++ parse_tc_flower_to_stats(&flower, stats); + } + } + +diff --git a/lib/netdev-offload.h b/lib/netdev-offload.h +index 8237a85ddb..93eb2df48b 100644 +--- a/lib/netdev-offload.h ++++ b/lib/netdev-offload.h +@@ -65,9 +65,6 @@ struct netdev_flow_dump { + + /* Flow offloading. */ + struct offload_info { +- ovs_be16 tp_dst_port; /* Destination port for tunnel in SET action */ +- uint8_t tunnel_csum_on; /* Tunnel header with checksum */ +- + bool recirc_id_shared_with_tc; /* Indicates whever tc chains will be in + * sync with datapath recirc ids. */ + +diff --git a/lib/netdev.c b/lib/netdev.c +index 8305f6c427..ce0d4117ac 100644 +--- a/lib/netdev.c ++++ b/lib/netdev.c +@@ -387,25 +387,30 @@ netdev_open(const char *name, const char *type, struct netdev **netdevp) + ovs_mutex_lock(&netdev_mutex); + netdev = shash_find_data(&netdev_shash, name); + +- if (netdev && +- type && type[0] && strcmp(type, netdev->netdev_class->type)) { +- +- if (netdev->auto_classified) { +- /* If this device was first created without a classification type, +- * for example due to routing or tunneling code, and they keep a +- * reference, a "classified" call to open will fail. In this case +- * we remove the classless device, and re-add it below. We remove +- * the netdev from the shash, and change the sequence, so owners of +- * the old classless device can release/cleanup. */ +- if (netdev->node) { +- shash_delete(&netdev_shash, netdev->node); +- netdev->node = NULL; +- netdev_change_seq_changed(netdev); +- } ++ if (netdev && type && type[0]) { ++ if (strcmp(type, netdev->netdev_class->type)) { ++ ++ if (netdev->auto_classified) { ++ /* If this device was first created without a classification ++ * type, for example due to routing or tunneling code, and they ++ * keep a reference, a "classified" call to open will fail. ++ * In this case we remove the classless device, and re-add it ++ * below. We remove the netdev from the shash, and change the ++ * sequence, so owners of the old classless device can ++ * release/cleanup. */ ++ if (netdev->node) { ++ shash_delete(&netdev_shash, netdev->node); ++ netdev->node = NULL; ++ netdev_change_seq_changed(netdev); ++ } + +- netdev = NULL; +- } else { +- error = EEXIST; ++ netdev = NULL; ++ } else { ++ error = EEXIST; ++ } ++ } else if (netdev->auto_classified) { ++ /* If netdev reopened with type "system", clear auto_classified. */ ++ netdev->auto_classified = false; + } + } + +diff --git a/lib/odp-util.c b/lib/odp-util.c +index 9a705cffa3..2d2a6893c6 100644 +--- a/lib/odp-util.c ++++ b/lib/odp-util.c +@@ -3429,16 +3429,16 @@ format_eth(struct ds *ds, const char *name, const struct eth_addr key, + + static void + format_be64(struct ds *ds, const char *name, ovs_be64 key, +- const ovs_be64 *mask, bool verbose) ++ const ovs_32aligned_be64 *mask_, bool verbose) + { +- bool mask_empty = mask && !*mask; ++ ovs_be64 mask = mask_ ? get_32aligned_be64(mask_) : htonll(0); + +- if (verbose || !mask_empty) { +- bool mask_full = !mask || *mask == OVS_BE64_MAX; ++ if (verbose || mask) { ++ bool mask_full = !mask_ || mask == OVS_BE64_MAX; + + ds_put_format(ds, "%s=0x%"PRIx64, name, ntohll(key)); + if (!mask_full) { /* Partially masked. */ +- ds_put_format(ds, "/%#"PRIx64, ntohll(*mask)); ++ ds_put_format(ds, "/%#"PRIx64, ntohll(mask)); + } + ds_put_char(ds, ','); + } +@@ -4630,6 +4630,11 @@ odp_flow_format(const struct nlattr *key, size_t key_len, + ds_put_char(ds, ','); + } + ds_put_cstr(ds, "eth()"); ++ } else if (attr_type == OVS_KEY_ATTR_PACKET_TYPE && is_wildcard) { ++ /* See the above help text, however in the case where the ++ * packet type is not shown, we still need to display the ++ * eth() header if the packets type is wildcarded. */ ++ has_packet_type_key = false; + } + ofpbuf_clear(&ofp); + } +@@ -7161,11 +7166,6 @@ parse_l2_5_onward(const struct nlattr *attrs[OVS_KEY_ATTR_MAX + 1], + } + } + } +- } else if (src_flow->nw_proto == IPPROTO_IGMP +- && src_flow->dl_type == htons(ETH_TYPE_IP)) { +- /* OVS userspace parses the IGMP type, code, and group, but its +- * datapaths do not, so there is always missing information. */ +- return ODP_FIT_TOO_LITTLE; + } + if (is_mask && expected_bit != OVS_KEY_ATTR_UNSPEC) { + if ((flow->tp_src || flow->tp_dst) && flow->nw_proto != 0xff) { +@@ -7188,7 +7188,8 @@ parse_8021q_onward(const struct nlattr *attrs[OVS_KEY_ATTR_MAX + 1], + uint64_t present_attrs, int out_of_range_attr, + uint64_t expected_attrs, struct flow *flow, + const struct nlattr *key, size_t key_len, +- const struct flow *src_flow, char **errorp) ++ const struct flow *src_flow, char **errorp, ++ bool ignore_vlan_limit) + { + static struct vlog_rate_limit rl = VLOG_RATE_LIMIT_INIT(1, 5); + bool is_mask = src_flow != flow; +@@ -7196,9 +7197,11 @@ parse_8021q_onward(const struct nlattr *attrs[OVS_KEY_ATTR_MAX + 1], + const struct nlattr *encap; + enum odp_key_fitness encap_fitness; + enum odp_key_fitness fitness = ODP_FIT_ERROR; ++ int vlan_limit; + int encaps = 0; + +- while (encaps < flow_vlan_limit && ++ vlan_limit = ignore_vlan_limit ? FLOW_MAX_VLAN_HEADERS : flow_vlan_limit; ++ while (encaps < vlan_limit && + (is_mask + ? (src_flow->vlans[encaps].tci & htons(VLAN_CFI)) != 0 + : eth_type_vlan(flow->dl_type))) { +@@ -7259,6 +7262,14 @@ parse_8021q_onward(const struct nlattr *attrs[OVS_KEY_ATTR_MAX + 1], + } + expected_attrs = 0; + ++ /* For OVS to be backward compatible with newer datapath ++ * implementations, we should ignore out of range attributes. */ ++ if (out_of_range_attr) { ++ VLOG_DBG("Flow key decode found unknown OVS_KEY_ATTR, %d", ++ out_of_range_attr); ++ out_of_range_attr = 0; ++ } ++ + if (!parse_ethertype(attrs, present_attrs, &expected_attrs, + flow, src_flow, errorp)) { + return ODP_FIT_ERROR; +@@ -7281,7 +7292,7 @@ parse_8021q_onward(const struct nlattr *attrs[OVS_KEY_ATTR_MAX + 1], + static enum odp_key_fitness + odp_flow_key_to_flow__(const struct nlattr *key, size_t key_len, + struct flow *flow, const struct flow *src_flow, +- char **errorp) ++ char **errorp, bool ignore_vlan_limit) + { + /* New "struct flow" fields that are visible to the datapath (including all + * data fields) should be translated from equivalent datapath flow fields +@@ -7308,6 +7319,14 @@ odp_flow_key_to_flow__(const struct nlattr *key, size_t key_len, + } + expected_attrs = 0; + ++ /* For OVS to be backward compatible with newer datapath implementations, ++ * we should ignore out of range attributes. */ ++ if (out_of_range_attr) { ++ VLOG_DBG("Flow key decode found unknown OVS_KEY_ATTR, %d", ++ out_of_range_attr); ++ out_of_range_attr = 0; ++ } ++ + /* Metadata. */ + if (present_attrs & (UINT64_C(1) << OVS_KEY_ATTR_RECIRC_ID)) { + flow->recirc_id = nl_attr_get_u32(attrs[OVS_KEY_ATTR_RECIRC_ID]); +@@ -7431,7 +7450,7 @@ odp_flow_key_to_flow__(const struct nlattr *key, size_t key_len, + : eth_type_vlan(src_flow->dl_type)) { + fitness = parse_8021q_onward(attrs, present_attrs, out_of_range_attr, + expected_attrs, flow, key, key_len, +- src_flow, errorp); ++ src_flow, errorp, ignore_vlan_limit); + } else { + if (is_mask) { + /* A missing VLAN mask means exact match on vlan_tci 0 (== no +@@ -7497,7 +7516,7 @@ enum odp_key_fitness + odp_flow_key_to_flow(const struct nlattr *key, size_t key_len, + struct flow *flow, char **errorp) + { +- return odp_flow_key_to_flow__(key, key_len, flow, flow, errorp); ++ return odp_flow_key_to_flow__(key, key_len, flow, flow, errorp, false); + } + + /* Converts the 'mask_key_len' bytes of OVS_KEY_ATTR_* attributes in 'mask_key' +@@ -7509,14 +7528,16 @@ odp_flow_key_to_flow(const struct nlattr *key, size_t key_len, + * If 'errorp' is nonnull, this function uses it for detailed error reports: if + * the return value is ODP_FIT_ERROR, it stores a malloc()'d error string in + * '*errorp', otherwise NULL. */ +-enum odp_key_fitness +-odp_flow_key_to_mask(const struct nlattr *mask_key, size_t mask_key_len, +- struct flow_wildcards *mask, const struct flow *src_flow, +- char **errorp) ++static enum odp_key_fitness ++odp_flow_key_to_mask__(const struct nlattr *mask_key, size_t mask_key_len, ++ struct flow_wildcards *mask, ++ const struct flow *src_flow, ++ char **errorp, bool ignore_vlan_limit) + { + if (mask_key_len) { + return odp_flow_key_to_flow__(mask_key, mask_key_len, +- &mask->masks, src_flow, errorp); ++ &mask->masks, src_flow, errorp, ++ ignore_vlan_limit); + } else { + if (errorp) { + *errorp = NULL; +@@ -7530,6 +7551,15 @@ odp_flow_key_to_mask(const struct nlattr *mask_key, size_t mask_key_len, + } + } + ++enum odp_key_fitness ++odp_flow_key_to_mask(const struct nlattr *mask_key, size_t mask_key_len, ++ struct flow_wildcards *mask, ++ const struct flow *src_flow, char **errorp) ++{ ++ return odp_flow_key_to_mask__(mask_key, mask_key_len, mask, src_flow, ++ errorp, false); ++} ++ + /* Converts the netlink formated key/mask to match. + * Fails if odp_flow_key_from_key/mask and odp_flow_key_key/mask + * disagree on the acceptable form of flow */ +@@ -7540,12 +7570,15 @@ parse_key_and_mask_to_match(const struct nlattr *key, size_t key_len, + { + enum odp_key_fitness fitness; + +- fitness = odp_flow_key_to_flow(key, key_len, &match->flow, NULL); ++ fitness = odp_flow_key_to_flow__(key, key_len, &match->flow, &match->flow, ++ NULL, true); + if (fitness) { +- /* This should not happen: it indicates that +- * odp_flow_key_from_flow() and odp_flow_key_to_flow() disagree on +- * the acceptable form of a flow. Log the problem as an error, +- * with enough details to enable debugging. */ ++ /* This will happen when the odp_flow_key_to_flow() function can't ++ * parse the netlink message to a match structure. It will return ++ * ODP_FIT_TOO_LITTLE if there is not enough information to parse the ++ * content successfully, ODP_FIT_TOO_MUCH if there is too much netlink ++ * data and we do not know how to safely ignore it, and ODP_FIT_ERROR ++ * in any other case. */ + static struct vlog_rate_limit rl = VLOG_RATE_LIMIT_INIT(1, 5); + + if (!VLOG_DROP_ERR(&rl)) { +@@ -7553,20 +7586,18 @@ parse_key_and_mask_to_match(const struct nlattr *key, size_t key_len, + + ds_init(&s); + odp_flow_format(key, key_len, NULL, 0, NULL, &s, true); +- VLOG_ERR("internal error parsing flow key %s", ds_cstr(&s)); ++ VLOG_ERR("internal error parsing flow key %s (%s)", ++ ds_cstr(&s), odp_key_fitness_to_string(fitness)); + ds_destroy(&s); + } + + return EINVAL; + } + +- fitness = odp_flow_key_to_mask(mask, mask_len, &match->wc, &match->flow, +- NULL); ++ fitness = odp_flow_key_to_mask__(mask, mask_len, &match->wc, &match->flow, ++ NULL, true); + if (fitness) { +- /* This should not happen: it indicates that +- * odp_flow_key_from_mask() and odp_flow_key_to_mask() +- * disagree on the acceptable form of a mask. Log the problem +- * as an error, with enough details to enable debugging. */ ++ /* This should not happen, see comment above. */ + static struct vlog_rate_limit rl = VLOG_RATE_LIMIT_INIT(1, 5); + + if (!VLOG_DROP_ERR(&rl)) { +diff --git a/lib/ofp-actions.c b/lib/ofp-actions.c +index 006837c2e1..c13f97b5c9 100644 +--- a/lib/ofp-actions.c ++++ b/lib/ofp-actions.c +@@ -853,7 +853,9 @@ decode_NXAST_RAW_CONTROLLER2(const struct ext_action_header *eah, + case NXAC2PT_REASON: { + uint8_t u8; + error = ofpprop_parse_u8(&payload, &u8); +- oc->reason = u8; ++ if (!error) { ++ oc->reason = u8; ++ } + break; + } + +@@ -3200,16 +3202,21 @@ set_field_split_str(char *arg, char **key, char **value, char **delim) + { + char *value_end; + ++ *key = NULL; + *value = arg; +- value_end = strstr(arg, "->"); +- *key = value_end + strlen("->"); + if (delim) { +- *delim = value_end; ++ *delim = NULL; + } + ++ value_end = strstr(arg, "->"); + if (!value_end) { + return xasprintf("%s: missing `->'", arg); + } ++ ++ *key = value_end + strlen("->"); ++ if (delim) { ++ *delim = value_end; ++ } + if (strlen(value_end) <= strlen("->")) { + return xasprintf("%s: missing field name following `->'", arg); + } +diff --git a/lib/ofp-msgs.c b/lib/ofp-msgs.c +index fec54f75f8..93aa812978 100644 +--- a/lib/ofp-msgs.c ++++ b/lib/ofp-msgs.c +@@ -1123,8 +1123,8 @@ ofpmp_partial_error(struct hmap *assembler, struct ofpmp_partial *p, + void + ofpmp_assembler_clear(struct hmap *assembler) + { +- struct ofpmp_partial *p, *next; +- HMAP_FOR_EACH_SAFE (p, next, hmap_node, assembler) { ++ struct ofpmp_partial *p; ++ HMAP_FOR_EACH_SAFE (p, hmap_node, assembler) { + ofpmp_partial_destroy(assembler, p); + } + } +@@ -1290,8 +1290,8 @@ ofpmp_assembler_execute(struct hmap *assembler, struct ofpbuf *msg, + * on either side by parts with 0-byte bodies. We remove the 0-byte + * ones here to simplify processing later. + */ +- struct ofpbuf *b, *next; +- LIST_FOR_EACH_SAFE (b, next, list_node, out) { ++ struct ofpbuf *b; ++ LIST_FOR_EACH_SAFE (b, list_node, out) { + if (b->size <= min_len && !ovs_list_is_short(out)) { + ovs_list_remove(&b->list_node); + ofpbuf_delete(b); +diff --git a/lib/ofp-packet.c b/lib/ofp-packet.c +index 4579548ee1..9485ddfc93 100644 +--- a/lib/ofp-packet.c ++++ b/lib/ofp-packet.c +@@ -133,7 +133,9 @@ decode_nx_packet_in2(const struct ofp_header *oh, bool loose, + case NXPINT_FULL_LEN: { + uint32_t u32; + error = ofpprop_parse_u32(&payload, &u32); +- *total_len = u32; ++ if (!error) { ++ *total_len = u32; ++ } + break; + } + +@@ -152,7 +154,9 @@ decode_nx_packet_in2(const struct ofp_header *oh, bool loose, + case NXPINT_REASON: { + uint8_t reason; + error = ofpprop_parse_u8(&payload, &reason); +- pin->reason = reason; ++ if (!error) { ++ pin->reason = reason; ++ } + break; + } + +@@ -883,7 +887,9 @@ ofputil_decode_packet_in_private(const struct ofp_header *oh, bool loose, + case NXCPT_ODP_PORT: { + uint32_t value; + error = ofpprop_parse_u32(&payload, &value); +- pin->odp_port = u32_to_odp(value); ++ if (!error) { ++ pin->odp_port = u32_to_odp(value); ++ } + break; + } + +diff --git a/lib/ofpbuf.c b/lib/ofpbuf.c +index 271105bdea..879275a7a3 100644 +--- a/lib/ofpbuf.c ++++ b/lib/ofpbuf.c +@@ -426,6 +426,10 @@ void + ofpbuf_reserve(struct ofpbuf *b, size_t size) + { + ovs_assert(!b->size); ++ ++ if (!size) { ++ return; ++ } + ofpbuf_prealloc_tailroom(b, size); + b->data = (char*)b->data + size; + } +@@ -436,6 +440,10 @@ ofpbuf_reserve(struct ofpbuf *b, size_t size) + void * + ofpbuf_push_uninit(struct ofpbuf *b, size_t size) + { ++ if (!size) { ++ return b->data; ++ } ++ + ofpbuf_prealloc_headroom(b, size); + b->data = (char*)b->data - size; + b->size += size; +diff --git a/lib/ovs-lldp.c b/lib/ovs-lldp.c +index 162311fa45..2d13e971ed 100644 +--- a/lib/ovs-lldp.c ++++ b/lib/ovs-lldp.c +@@ -559,9 +559,9 @@ aa_mapping_unregister_mapping(struct lldp *lldp, + struct lldpd_hardware *hw, + struct aa_mapping_internal *m) + { +- struct lldpd_aa_isid_vlan_maps_tlv *lm, *lm_next; ++ struct lldpd_aa_isid_vlan_maps_tlv *lm; + +- LIST_FOR_EACH_SAFE (lm, lm_next, m_entries, ++ LIST_FOR_EACH_SAFE (lm, m_entries, + &hw->h_lport.p_isid_vlan_maps) { + uint32_t isid = lm->isid_vlan_data.isid; + +@@ -738,6 +738,14 @@ lldp_put_packet(struct lldp *lldp, struct dp_packet *packet, + ovs_mutex_unlock(&mutex); + } + ++/* Is LLDP enabled? ++ */ ++bool ++lldp_is_enabled(struct lldp *lldp) ++{ ++ return lldp ? lldp->enabled : false; ++} ++ + /* Configures the LLDP stack. + */ + bool +@@ -953,8 +961,8 @@ lldp_ref(const struct lldp *lldp_) + void + lldp_destroy_dummy(struct lldp *lldp) + { +- struct lldpd_hardware *hw, *hw_next; +- struct lldpd_chassis *chassis, *chassis_next; ++ struct lldpd_hardware *hw; ++ struct lldpd_chassis *chassis; + struct lldpd *cfg; + + if (!lldp) { +@@ -963,13 +971,13 @@ lldp_destroy_dummy(struct lldp *lldp) + + cfg = lldp->lldpd; + +- LIST_FOR_EACH_SAFE (hw, hw_next, h_entries, &cfg->g_hardware) { ++ LIST_FOR_EACH_SAFE (hw, h_entries, &cfg->g_hardware) { + ovs_list_remove(&hw->h_entries); + free(hw->h_lport.p_lastframe); + free(hw); + } + +- LIST_FOR_EACH_SAFE (chassis, chassis_next, list, &cfg->g_chassis) { ++ LIST_FOR_EACH_SAFE (chassis, list, &cfg->g_chassis) { + ovs_list_remove(&chassis->list); + free(chassis); + } +diff --git a/lib/ovs-lldp.h b/lib/ovs-lldp.h +index 0e536e8c27..661ac4e18a 100644 +--- a/lib/ovs-lldp.h ++++ b/lib/ovs-lldp.h +@@ -86,6 +86,7 @@ void lldp_run(struct lldpd *cfg); + bool lldp_should_send_packet(struct lldp *cfg); + bool lldp_should_process_flow(struct lldp *lldp, const struct flow *flow); + bool lldp_configure(struct lldp *lldp, const struct smap *cfg); ++bool lldp_is_enabled(struct lldp *lldp); + void lldp_process_packet(struct lldp *cfg, const struct dp_packet *); + void lldp_put_packet(struct lldp *lldp, struct dp_packet *packet, + const struct eth_addr eth_src); +diff --git a/lib/ovs-numa.c b/lib/ovs-numa.c +index 9e3fa54216..6a197772c1 100644 +--- a/lib/ovs-numa.c ++++ b/lib/ovs-numa.c +@@ -387,6 +387,35 @@ ovs_numa_get_n_cores_on_numa(int numa_id) + return OVS_CORE_UNSPEC; + } + ++/* Returns the largest core_id. ++ * ++ * Return OVS_CORE_UNSPEC, if core_id information is not found. ++ * ++ * Returning OVS_CORE_UNSPEC comes at a caveat. The caller function ++ * must remember to check the return value of this callee function ++ * against OVS_CORE_UNSPEC. OVS_CORE_UNSPEC is a positive integer ++ * INT_MAX, which the caller may interpret it as the largest ++ * core_id if it's not checking for it. ++ */ ++unsigned ++ovs_numa_get_largest_core_id(void) ++{ ++ struct cpu_core *core; ++ unsigned max_id = 0; ++ ++ if (!found_numa_and_core) { ++ return OVS_CORE_UNSPEC; ++ } ++ ++ HMAP_FOR_EACH (core, hmap_node, &all_cpu_cores) { ++ if (core->core_id > max_id) { ++ max_id = core->core_id; ++ } ++ } ++ ++ return max_id; ++} ++ + static struct ovs_numa_dump * + ovs_numa_dump_create(void) + { +diff --git a/lib/ovs-numa.h b/lib/ovs-numa.h +index ecc251a7ff..02c9e84cf5 100644 +--- a/lib/ovs-numa.h ++++ b/lib/ovs-numa.h +@@ -56,6 +56,7 @@ int ovs_numa_get_n_numas(void); + int ovs_numa_get_n_cores(void); + int ovs_numa_get_numa_id(unsigned core_id); + int ovs_numa_get_n_cores_on_numa(int numa_id); ++unsigned ovs_numa_get_largest_core_id(void); + struct ovs_numa_dump *ovs_numa_dump_cores_on_numa(int numa_id); + struct ovs_numa_dump *ovs_numa_dump_cores_with_cmask(const char *cmask); + struct ovs_numa_dump *ovs_numa_dump_n_cores_per_numa(int n); +@@ -68,9 +69,9 @@ void ovs_numa_dump_destroy(struct ovs_numa_dump *); + int ovs_numa_thread_setaffinity_core(unsigned core_id); + + #define FOR_EACH_CORE_ON_DUMP(ITER, DUMP) \ +- HMAP_FOR_EACH((ITER), hmap_node, &(DUMP)->cores) ++ HMAP_FOR_EACH (ITER, hmap_node, &(DUMP)->cores) + + #define FOR_EACH_NUMA_ON_DUMP(ITER, DUMP) \ +- HMAP_FOR_EACH((ITER), hmap_node, &(DUMP)->numas) ++ HMAP_FOR_EACH (ITER, hmap_node, &(DUMP)->numas) + + #endif /* ovs-numa.h */ +diff --git a/lib/ovs-rcu.c b/lib/ovs-rcu.c +index 1866bd3088..946aa04d18 100644 +--- a/lib/ovs-rcu.c ++++ b/lib/ovs-rcu.c +@@ -444,3 +444,40 @@ ovsrcu_init_module(void) + ovsthread_once_done(&once); + } + } ++ ++static void ++ovsrcu_barrier_func(void *seq_) ++{ ++ struct seq *seq = (struct seq *) seq_; ++ seq_change(seq); ++} ++ ++/* Similar to the kernel rcu_barrier, ovsrcu_barrier waits for all outstanding ++ * RCU callbacks to complete. However, unlike the kernel rcu_barrier, which ++ * might return immediately if there are no outstanding RCU callbacks, ++ * this API will at least wait for a grace period. ++ * ++ * Another issue the caller might need to know is that the barrier is just ++ * for "one-shot", i.e. if inside some RCU callbacks, another RCU callback is ++ * registered, this API only guarantees the first round of RCU callbacks have ++ * been executed after it returns. ++ */ ++void ++ovsrcu_barrier(void) ++{ ++ struct seq *seq = seq_create(); ++ /* First let all threads flush their cbsets. */ ++ ovsrcu_synchronize(); ++ ++ /* Then register a new cbset, ensure this cbset ++ * is at the tail of the global list. */ ++ uint64_t seqno = seq_read(seq); ++ ovsrcu_postpone__(ovsrcu_barrier_func, (void *) seq); ++ ++ do { ++ seq_wait(seq, seqno); ++ poll_block(); ++ } while (seqno == seq_read(seq)); ++ ++ seq_destroy(seq); ++} +diff --git a/lib/ovs-rcu.h b/lib/ovs-rcu.h +index ecc4c92010..8b397b7fb0 100644 +--- a/lib/ovs-rcu.h ++++ b/lib/ovs-rcu.h +@@ -155,6 +155,19 @@ + * port_delete(id); + * } + * ++ * Use ovsrcu_barrier() to wait for all the outstanding RCU callbacks to ++ * finish. This is useful when you have to destroy some resources however ++ * these resources are referenced in the outstanding RCU callbacks. ++ * ++ * void rcu_cb(void *A) { ++ * do_something(A); ++ * } ++ * ++ * void destroy_A() { ++ * ovsrcu_postpone(rcu_cb, A); // will use A later ++ * ovsrcu_barrier(); // wait for rcu_cb done ++ * do_destroy_A(); // free A ++ * } + */ + + #include "compiler.h" +@@ -310,4 +323,6 @@ void ovsrcu_synchronize(void); + + void ovsrcu_exit(void); + ++void ovsrcu_barrier(void); ++ + #endif /* ovs-rcu.h */ +diff --git a/lib/ovs-router.c b/lib/ovs-router.c +index 09b81c6e5a..5d0fbd503e 100644 +--- a/lib/ovs-router.c ++++ b/lib/ovs-router.c +@@ -164,9 +164,10 @@ static void rt_init_match(struct match *match, uint32_t mark, + match->flow.pkt_mark = mark; + } + +-static int +-get_src_addr(const struct in6_addr *ip6_dst, +- const char output_bridge[], struct in6_addr *psrc) ++int ++ovs_router_get_netdev_source_address(const struct in6_addr *ip6_dst, ++ const char output_bridge[], ++ struct in6_addr *psrc) + { + struct in6_addr *mask, *addr6; + int err, n_in6, i, max_plen = -1; +@@ -235,9 +236,11 @@ ovs_router_insert__(uint32_t mark, uint8_t priority, bool local, + p->plen = plen; + p->local = local; + p->priority = priority; +- err = get_src_addr(ip6_dst, output_bridge, &p->src_addr); ++ err = ovs_router_get_netdev_source_address(ip6_dst, output_bridge, ++ &p->src_addr); + if (err && ipv6_addr_is_set(gw)) { +- err = get_src_addr(gw, output_bridge, &p->src_addr); ++ err = ovs_router_get_netdev_source_address(gw, output_bridge, ++ &p->src_addr); + } + if (err) { + struct ds ds = DS_EMPTY_INITIALIZER; +diff --git a/lib/ovs-router.h b/lib/ovs-router.h +index 34ea163eef..d8ce3c00de 100644 +--- a/lib/ovs-router.h ++++ b/lib/ovs-router.h +@@ -37,6 +37,10 @@ void ovs_router_flush(void); + + void ovs_router_disable_system_routing_table(void); + ++int ovs_router_get_netdev_source_address(const struct in6_addr *ip6_dst, ++ const char output_bridge[], ++ struct in6_addr *psrc); ++ + #ifdef __cplusplus + } + #endif +diff --git a/lib/ovs-thread.c b/lib/ovs-thread.c +index 805cba6223..78ed3e9707 100644 +--- a/lib/ovs-thread.c ++++ b/lib/ovs-thread.c +@@ -663,6 +663,23 @@ count_cpu_cores(void) + return n_cores > 0 ? n_cores : 0; + } + ++/* Returns the total number of cores on the system, or 0 if the ++ * number cannot be determined. */ ++int ++count_total_cores(void) ++{ ++ long int n_cores; ++ ++#ifndef _WIN32 ++ n_cores = sysconf(_SC_NPROCESSORS_CONF); ++#else ++ n_cores = 0; ++ errno = ENOTSUP; ++#endif ++ ++ return n_cores > 0 ? n_cores : 0; ++} ++ + /* Returns 'true' if current thread is PMD thread. */ + bool + thread_is_pmd(void) +diff --git a/lib/ovs-thread.h b/lib/ovs-thread.h +index 3b444ccdcc..aac5e19c99 100644 +--- a/lib/ovs-thread.h ++++ b/lib/ovs-thread.h +@@ -522,6 +522,7 @@ bool may_fork(void); + /* Useful functions related to threading. */ + + int count_cpu_cores(void); ++int count_total_cores(void); + bool thread_is_pmd(void); + + #endif /* ovs-thread.h */ +diff --git a/lib/ovsdb-cs.c b/lib/ovsdb-cs.c +index dead31275d..9713c7dc7c 100644 +--- a/lib/ovsdb-cs.c ++++ b/lib/ovsdb-cs.c +@@ -900,8 +900,8 @@ ovsdb_cs_db_get_table(struct ovsdb_cs_db *db, const char *table) + static void + ovsdb_cs_db_destroy_tables(struct ovsdb_cs_db *db) + { +- struct ovsdb_cs_db_table *table, *next; +- HMAP_FOR_EACH_SAFE (table, next, hmap_node, &db->tables) { ++ struct ovsdb_cs_db_table *table; ++ HMAP_FOR_EACH_SAFE (table, hmap_node, &db->tables) { + json_destroy(table->ack_cond); + json_destroy(table->req_cond); + json_destroy(table->new_cond); +@@ -1793,8 +1793,8 @@ ovsdb_cs_update_server_row(struct server_row *row, + static void + ovsdb_cs_clear_server_rows(struct ovsdb_cs *cs) + { +- struct server_row *row, *next; +- HMAP_FOR_EACH_SAFE (row, next, hmap_node, &cs->server_rows) { ++ struct server_row *row; ++ HMAP_FOR_EACH_SAFE (row, hmap_node, &cs->server_rows) { + ovsdb_cs_delete_server_row(cs, row); + } + } +@@ -2128,9 +2128,9 @@ void + ovsdb_cs_free_schema(struct shash *schema) + { + if (schema) { +- struct shash_node *node, *next; ++ struct shash_node *node; + +- SHASH_FOR_EACH_SAFE (node, next, schema) { ++ SHASH_FOR_EACH_SAFE (node, schema) { + struct sset *sset = node->data; + sset_destroy(sset); + free(sset); +diff --git a/lib/ovsdb-data.c b/lib/ovsdb-data.c +index 6b1c20ff85..61ad7679a6 100644 +--- a/lib/ovsdb-data.c ++++ b/lib/ovsdb-data.c +@@ -1957,6 +1957,19 @@ ovsdb_datum_add_unsafe(struct ovsdb_datum *datum, + } + } + ++void ++ovsdb_datum_add_from_index_unsafe(struct ovsdb_datum *dst, ++ const struct ovsdb_datum *src, ++ size_t idx, ++ const struct ovsdb_type *type) ++{ ++ const union ovsdb_atom *key = &src->keys[idx]; ++ const union ovsdb_atom *value = type->value.type != OVSDB_TYPE_VOID ++ ? &src->values[idx] ++ : NULL; ++ ovsdb_datum_add_unsafe(dst, key, value, type, NULL); ++} ++ + /* Adds 'n' atoms starting from index 'start_idx' from 'src' to the end of + * 'dst'. 'dst' should have enough memory allocated to hold the additional + * 'n' atoms. Atoms are not cloned, i.e. 'dst' will reference the same data. +@@ -2165,12 +2178,10 @@ ovsdb_datum_added_removed(struct ovsdb_datum *added, + int c = ovsdb_atom_compare_3way(&old->keys[oi], &new->keys[ni], + type->key.type); + if (c < 0) { +- ovsdb_datum_add_unsafe(removed, &old->keys[oi], &old->values[oi], +- type, NULL); ++ ovsdb_datum_add_from_index_unsafe(removed, old, oi, type); + oi++; + } else if (c > 0) { +- ovsdb_datum_add_unsafe(added, &new->keys[ni], &new->values[ni], +- type, NULL); ++ ovsdb_datum_add_from_index_unsafe(added, new, ni, type); + ni++; + } else { + if (type->value.type != OVSDB_TYPE_VOID && +@@ -2186,13 +2197,11 @@ ovsdb_datum_added_removed(struct ovsdb_datum *added, + } + + for (; oi < old->n; oi++) { +- ovsdb_datum_add_unsafe(removed, &old->keys[oi], &old->values[oi], +- type, NULL); ++ ovsdb_datum_add_from_index_unsafe(removed, old, oi, type); + } + + for (; ni < new->n; ni++) { +- ovsdb_datum_add_unsafe(added, &new->keys[ni], &new->values[ni], +- type, NULL); ++ ovsdb_datum_add_from_index_unsafe(added, new, ni, type); + } + } + +@@ -2228,12 +2237,10 @@ ovsdb_datum_diff(struct ovsdb_datum *diff, + int c = ovsdb_atom_compare_3way(&old->keys[oi], &new->keys[ni], + type->key.type); + if (c < 0) { +- ovsdb_datum_add_unsafe(diff, &old->keys[oi], &old->values[oi], +- type, NULL); ++ ovsdb_datum_add_from_index_unsafe(diff, old, oi, type); + oi++; + } else if (c > 0) { +- ovsdb_datum_add_unsafe(diff, &new->keys[ni], &new->values[ni], +- type, NULL); ++ ovsdb_datum_add_from_index_unsafe(diff, new, ni, type); + ni++; + } else { + if (type->value.type != OVSDB_TYPE_VOID && +@@ -2247,13 +2254,11 @@ ovsdb_datum_diff(struct ovsdb_datum *diff, + } + + for (; oi < old->n; oi++) { +- ovsdb_datum_add_unsafe(diff, &old->keys[oi], &old->values[oi], +- type, NULL); ++ ovsdb_datum_add_from_index_unsafe(diff, old, oi, type); + } + + for (; ni < new->n; ni++) { +- ovsdb_datum_add_unsafe(diff, &new->keys[ni], &new->values[ni], +- type, NULL); ++ ovsdb_datum_add_from_index_unsafe(diff, new, ni, type); + } + } + +diff --git a/lib/ovsdb-data.h b/lib/ovsdb-data.h +index 47115a7b85..ba5d179a65 100644 +--- a/lib/ovsdb-data.h ++++ b/lib/ovsdb-data.h +@@ -280,6 +280,10 @@ void ovsdb_datum_add_unsafe(struct ovsdb_datum *, + const union ovsdb_atom *value, + const struct ovsdb_type *, + const union ovsdb_atom *range_end_atom); ++void ovsdb_datum_add_from_index_unsafe(struct ovsdb_datum *dst, ++ const struct ovsdb_datum *src, ++ size_t idx, ++ const struct ovsdb_type *type); + + /* Transactions with named-uuid row names. */ + struct json *ovsdb_datum_to_json_with_row_names(const struct ovsdb_datum *, +diff --git a/lib/ovsdb-idl.c b/lib/ovsdb-idl.c +index c19128d55c..882ede7559 100644 +--- a/lib/ovsdb-idl.c ++++ b/lib/ovsdb-idl.c +@@ -389,25 +389,25 @@ ovsdb_idl_clear(struct ovsdb_idl *db) + */ + for (size_t i = 0; i < db->class_->n_tables; i++) { + struct ovsdb_idl_table *table = &db->tables[i]; +- struct ovsdb_idl_row *row, *next_row; ++ struct ovsdb_idl_row *row; + + if (hmap_is_empty(&table->rows)) { + continue; + } + +- HMAP_FOR_EACH_SAFE (row, next_row, hmap_node, &table->rows) { +- struct ovsdb_idl_arc *arc, *next_arc; ++ HMAP_FOR_EACH_SAFE (row, hmap_node, &table->rows) { ++ struct ovsdb_idl_arc *arc; + + if (!ovsdb_idl_row_is_orphan(row)) { + ovsdb_idl_remove_from_indexes(row); + ovsdb_idl_row_unparse(row); + } +- LIST_FOR_EACH_SAFE (arc, next_arc, src_node, &row->src_arcs) { ++ LIST_FOR_EACH_SAFE (arc, src_node, &row->src_arcs) { + ovs_list_remove(&arc->src_node); + ovs_list_remove(&arc->dst_node); + free(arc); + } +- LIST_FOR_EACH_SAFE (arc, next_arc, dst_node, &row->dst_arcs) { ++ LIST_FOR_EACH_SAFE (arc, dst_node, &row->dst_arcs) { + ovs_list_remove(&arc->src_node); + ovs_list_remove(&arc->dst_node); + free(arc); +@@ -1041,8 +1041,8 @@ ovsdb_idl_condition_destroy(struct ovsdb_idl_condition *cond) + void + ovsdb_idl_condition_clear(struct ovsdb_idl_condition *cond) + { +- struct ovsdb_idl_clause *clause, *next; +- HMAP_FOR_EACH_SAFE (clause, next, hmap_node, &cond->clauses) { ++ struct ovsdb_idl_clause *clause; ++ HMAP_FOR_EACH_SAFE (clause, hmap_node, &cond->clauses) { + hmap_remove(&cond->clauses, &clause->hmap_node); + ovsdb_idl_clause_destroy(clause); + } +@@ -1345,9 +1345,9 @@ ovsdb_idl_track_clear__(struct ovsdb_idl *idl, bool flush_all) + struct ovsdb_idl_table *table = &idl->tables[i]; + + if (!ovs_list_is_empty(&table->track_list)) { +- struct ovsdb_idl_row *row, *next; ++ struct ovsdb_idl_row *row; + +- LIST_FOR_EACH_SAFE(row, next, track_node, &table->track_list) { ++ LIST_FOR_EACH_SAFE (row, track_node, &table->track_list) { + if (row->updated) { + free(row->updated); + row->updated = NULL; +@@ -1480,9 +1480,9 @@ ovsdb_idl_parse_update(struct ovsdb_idl *idl, + static void + ovsdb_idl_reparse_deleted(struct ovsdb_idl *db) + { +- struct ovsdb_idl_row *row, *next; ++ struct ovsdb_idl_row *row; + +- LIST_FOR_EACH_SAFE (row, next, track_node, &db->deleted_untracked_rows) { ++ LIST_FOR_EACH_SAFE (row, track_node, &db->deleted_untracked_rows) { + ovsdb_idl_row_untrack_change(row); + add_tracked_change_for_references(row); + ovsdb_idl_row_reparse_backrefs(row); +@@ -1906,8 +1906,8 @@ ovsdb_idl_index_create2(struct ovsdb_idl *idl, + static void + ovsdb_idl_destroy_indexes(struct ovsdb_idl_table *table) + { +- struct ovsdb_idl_index *index, *next; +- LIST_FOR_EACH_SAFE (index, next, node, &table->indexes) { ++ struct ovsdb_idl_index *index; ++ LIST_FOR_EACH_SAFE (index, node, &table->indexes) { + skiplist_destroy(index->skiplist, NULL); + free(index->columns); + free(index); +@@ -2145,12 +2145,12 @@ ovsdb_idl_row_clear_new(struct ovsdb_idl_row *row) + static void + ovsdb_idl_row_clear_arcs(struct ovsdb_idl_row *row, bool destroy_dsts) + { +- struct ovsdb_idl_arc *arc, *next; ++ struct ovsdb_idl_arc *arc; + + /* Delete all forward arcs. If 'destroy_dsts', destroy any orphaned rows + * that this causes to be unreferenced. + */ +- LIST_FOR_EACH_SAFE (arc, next, src_node, &row->src_arcs) { ++ LIST_FOR_EACH_SAFE (arc, src_node, &row->src_arcs) { + ovs_list_remove(&arc->dst_node); + if (destroy_dsts + && ovsdb_idl_row_is_orphan(arc->dst) +@@ -2166,7 +2166,7 @@ ovsdb_idl_row_clear_arcs(struct ovsdb_idl_row *row, bool destroy_dsts) + static void + ovsdb_idl_row_reparse_backrefs(struct ovsdb_idl_row *row) + { +- struct ovsdb_idl_arc *arc, *next; ++ struct ovsdb_idl_arc *arc; + + /* This is trickier than it looks. ovsdb_idl_row_clear_arcs() will destroy + * 'arc', so we need to use the "safe" variant of list traversal. However, +@@ -2178,7 +2178,7 @@ ovsdb_idl_row_reparse_backrefs(struct ovsdb_idl_row *row) + * (If duplicate arcs were possible then we would need to make sure that + * 'next' didn't also point into 'arc''s destination, but we forbid + * duplicate arcs.) */ +- LIST_FOR_EACH_SAFE (arc, next, dst_node, &row->dst_arcs) { ++ LIST_FOR_EACH_SAFE (arc, dst_node, &row->dst_arcs) { + struct ovsdb_idl_row *ref = arc->src; + + ovsdb_idl_row_unparse(ref); +@@ -2329,9 +2329,9 @@ ovsdb_idl_row_destroy_postprocess(struct ovsdb_idl *idl) + struct ovsdb_idl_table *table = &idl->tables[i]; + + if (!ovs_list_is_empty(&table->track_list)) { +- struct ovsdb_idl_row *row, *next; ++ struct ovsdb_idl_row *row; + +- LIST_FOR_EACH_SAFE(row, next, track_node, &table->track_list) { ++ LIST_FOR_EACH_SAFE (row, track_node, &table->track_list) { + if (!ovsdb_idl_track_is_set(row->table)) { + ovs_list_remove(&row->track_node); + ovsdb_idl_row_unparse(row); +@@ -2729,7 +2729,7 @@ ovsdb_idl_txn_increment(struct ovsdb_idl_txn *txn, + void + ovsdb_idl_txn_destroy(struct ovsdb_idl_txn *txn) + { +- struct ovsdb_idl_txn_insert *insert, *next; ++ struct ovsdb_idl_txn_insert *insert; + + if (txn->status == TXN_INCOMPLETE) { + ovsdb_cs_forget_transaction(txn->idl->cs, txn->request_id); +@@ -2739,7 +2739,7 @@ ovsdb_idl_txn_destroy(struct ovsdb_idl_txn *txn) + ovsdb_idl_txn_abort(txn); + ds_destroy(&txn->comment); + free(txn->error); +- HMAP_FOR_EACH_SAFE (insert, next, hmap_node, &txn->inserted_rows) { ++ HMAP_FOR_EACH_SAFE (insert, hmap_node, &txn->inserted_rows) { + free(insert); + } + hmap_destroy(&txn->inserted_rows); +@@ -2824,7 +2824,7 @@ substitute_uuids(struct json *json, const struct ovsdb_idl_txn *txn) + static void + ovsdb_idl_txn_disassemble(struct ovsdb_idl_txn *txn) + { +- struct ovsdb_idl_row *row, *next; ++ struct ovsdb_idl_row *row; + + /* This must happen early. Otherwise, ovsdb_idl_row_parse() will call an + * ovsdb_idl_column's 'parse' function, which will call +@@ -2832,7 +2832,7 @@ ovsdb_idl_txn_disassemble(struct ovsdb_idl_txn *txn) + * transaction and fail to update the graph. */ + txn->idl->txn = NULL; + +- HMAP_FOR_EACH_SAFE (row, next, txn_node, &txn->txn_rows) { ++ HMAP_FOR_EACH_SAFE (row, txn_node, &txn->txn_rows) { + enum { INSERTED, MODIFIED, DELETED } op + = (!row->new_datum ? DELETED + : !row->old_datum ? INSERTED +diff --git a/lib/ovsdb-map-op.c b/lib/ovsdb-map-op.c +index 7b90ba84f9..795066e8ef 100644 +--- a/lib/ovsdb-map-op.c ++++ b/lib/ovsdb-map-op.c +@@ -91,8 +91,8 @@ map_op_list_create(void) + void + map_op_list_destroy(struct map_op_list *list, const struct ovsdb_type *type) + { +- struct map_op *map_op, *next; +- HMAP_FOR_EACH_SAFE (map_op, next, node, &list->hmap) { ++ struct map_op *map_op; ++ HMAP_FOR_EACH_SAFE (map_op, node, &list->hmap) { + map_op_destroy(map_op, type); + } + hmap_destroy(&list->hmap); +diff --git a/lib/ovsdb-set-op.c b/lib/ovsdb-set-op.c +index 62c4621181..321043282e 100644 +--- a/lib/ovsdb-set-op.c ++++ b/lib/ovsdb-set-op.c +@@ -90,8 +90,8 @@ set_op_list_create(void) + void + set_op_list_destroy(struct set_op_list *list, const struct ovsdb_type *type) + { +- struct set_op *set_op, *next; +- HMAP_FOR_EACH_SAFE (set_op, next, node, &list->hmap) { ++ struct set_op *set_op; ++ HMAP_FOR_EACH_SAFE (set_op, node, &list->hmap) { + set_op_destroy(set_op, type); + } + hmap_destroy(&list->hmap); +diff --git a/lib/packets.c b/lib/packets.c +index d0fba81766..1dcd4a6fcd 100644 +--- a/lib/packets.c ++++ b/lib/packets.c +@@ -427,9 +427,9 @@ add_mpls(struct dp_packet *packet, ovs_be16 ethtype, ovs_be32 lse, + } + + if (!l3_encap) { +- ovs_be32 *header = dp_packet_push_uninit(packet, MPLS_HLEN); ++ struct mpls_hdr *header = dp_packet_push_uninit(packet, MPLS_HLEN); + +- *header = lse; ++ put_16aligned_be32(&header->mpls_lse, lse); + packet->l2_5_ofs = 0; + packet->packet_type = PACKET_TYPE_BE(OFPHTN_ETHERTYPE, + ntohs(ethtype)); +@@ -1153,7 +1153,7 @@ packet_set_ipv4_addr(struct dp_packet *packet, + * + * This function assumes that L3 and L4 offsets are set in the packet. */ + static bool +-packet_rh_present(struct dp_packet *packet, uint8_t *nexthdr) ++packet_rh_present(struct dp_packet *packet, uint8_t *nexthdr, bool *first_frag) + { + const struct ovs_16aligned_ip6_hdr *nh; + size_t len; +@@ -1203,6 +1203,8 @@ packet_rh_present(struct dp_packet *packet, uint8_t *nexthdr) + const struct ovs_16aligned_ip6_frag *frag_hdr + = ALIGNED_CAST(struct ovs_16aligned_ip6_frag *, data); + ++ *first_frag = !(frag_hdr->ip6f_offlg & IP6F_OFF_MASK) && ++ (frag_hdr->ip6f_offlg & IP6F_MORE_FRAG); + *nexthdr = frag_hdr->ip6f_nxt; + len = sizeof *frag_hdr; + } else if (*nexthdr == IPPROTO_ROUTING) { +@@ -1333,18 +1335,20 @@ packet_set_ipv6(struct dp_packet *packet, const struct in6_addr *src, + uint8_t key_hl) + { + struct ovs_16aligned_ip6_hdr *nh = dp_packet_l3(packet); ++ bool recalc_csum = true; + uint8_t proto = 0; + bool rh_present; + +- rh_present = packet_rh_present(packet, &proto); ++ rh_present = packet_rh_present(packet, &proto, &recalc_csum); + + if (memcmp(&nh->ip6_src, src, sizeof(ovs_be32[4]))) { +- packet_set_ipv6_addr(packet, proto, nh->ip6_src.be32, src, true); ++ packet_set_ipv6_addr(packet, proto, nh->ip6_src.be32, ++ src, recalc_csum); + } + + if (memcmp(&nh->ip6_dst, dst, sizeof(ovs_be32[4]))) { + packet_set_ipv6_addr(packet, proto, nh->ip6_dst.be32, dst, +- !rh_present); ++ !rh_present && recalc_csum); + } + + packet_set_ipv6_tc(&nh->ip6_flow, key_tc); +@@ -1705,7 +1709,7 @@ compose_ipv6(struct dp_packet *packet, uint8_t proto, + const struct in6_addr *src, const struct in6_addr *dst, + uint8_t key_tc, ovs_be32 key_fl, uint8_t key_hl, int size) + { +- struct ip6_hdr *nh; ++ struct ovs_16aligned_ip6_hdr *nh; + void *data; + + nh = dp_packet_l3(packet); +@@ -1843,7 +1847,7 @@ packet_put_ra_prefix_opt(struct dp_packet *b, + const ovs_be128 prefix) + { + size_t prev_l4_size = dp_packet_l4_size(b); +- struct ip6_hdr *nh = dp_packet_l3(b); ++ struct ovs_16aligned_ip6_hdr *nh = dp_packet_l3(b); + nh->ip6_plen = htons(prev_l4_size + ND_PREFIX_OPT_LEN); + + struct ovs_nd_prefix_opt *prefix_opt = +diff --git a/lib/pcap-file.c b/lib/pcap-file.c +index 41835f6f4d..3ed7ea4880 100644 +--- a/lib/pcap-file.c ++++ b/lib/pcap-file.c +@@ -344,9 +344,9 @@ tcp_reader_open(void) + void + tcp_reader_close(struct tcp_reader *r) + { +- struct tcp_stream *stream, *next_stream; ++ struct tcp_stream *stream; + +- HMAP_FOR_EACH_SAFE (stream, next_stream, hmap_node, &r->streams) { ++ HMAP_FOR_EACH_SAFE (stream, hmap_node, &r->streams) { + tcp_stream_destroy(r, stream); + } + hmap_destroy(&r->streams); +diff --git a/lib/perf-counter.c b/lib/perf-counter.c +index e4eca58d03..6952fcb594 100644 +--- a/lib/perf-counter.c ++++ b/lib/perf-counter.c +@@ -178,14 +178,14 @@ perf_counters_clear(void) + void + perf_counters_destroy(void) + { +- struct shash_node *node, *next; ++ struct shash_node *node; + + if (fd__ != -1) { + ioctl(fd__, PERF_EVENT_IOC_DISABLE, 0); + close(fd__); + } + +- SHASH_FOR_EACH_SAFE (node, next, &perf_counters) { ++ SHASH_FOR_EACH_SAFE (node, &perf_counters) { + shash_delete(&perf_counters, node); + } + +diff --git a/lib/poll-loop.c b/lib/poll-loop.c +index 4e751ff2c7..70fabeb8a3 100644 +--- a/lib/poll-loop.c ++++ b/lib/poll-loop.c +@@ -298,9 +298,9 @@ log_wakeup(const char *where, const struct pollfd *pollfd, int timeout) + static void + free_poll_nodes(struct poll_loop *loop) + { +- struct poll_node *node, *next; ++ struct poll_node *node; + +- HMAP_FOR_EACH_SAFE (node, next, hmap_node, &loop->poll_nodes) { ++ HMAP_FOR_EACH_SAFE (node, hmap_node, &loop->poll_nodes) { + hmap_remove(&loop->poll_nodes, &node->hmap_node); + #ifdef _WIN32 + if (node->wevent && node->pollfd.fd) { +diff --git a/lib/rculist.h b/lib/rculist.h +index 1072b87af2..c0d77acf94 100644 +--- a/lib/rculist.h ++++ b/lib/rculist.h +@@ -365,35 +365,57 @@ rculist_is_singleton_protected(const struct rculist *list) + return list_next == list->prev && list_next != list; + } + +-#define RCULIST_FOR_EACH(ITER, MEMBER, RCULIST) \ +- for (INIT_CONTAINER(ITER, rculist_next(RCULIST), MEMBER); \ +- &(ITER)->MEMBER != (RCULIST); \ +- ASSIGN_CONTAINER(ITER, rculist_next(&(ITER)->MEMBER), MEMBER)) +-#define RCULIST_FOR_EACH_CONTINUE(ITER, MEMBER, RCULIST) \ +- for (ASSIGN_CONTAINER(ITER, rculist_next(&(ITER)->MEMBER), MEMBER); \ +- &(ITER)->MEMBER != (RCULIST); \ +- ASSIGN_CONTAINER(ITER, rculist_next(&(ITER)->MEMBER), MEMBER)) +- +-#define RCULIST_FOR_EACH_REVERSE_PROTECTED(ITER, MEMBER, RCULIST) \ +- for (INIT_CONTAINER(ITER, (RCULIST)->prev, MEMBER); \ +- &(ITER)->MEMBER != (RCULIST); \ +- ASSIGN_CONTAINER(ITER, (ITER)->MEMBER.prev, MEMBER)) +-#define RCULIST_FOR_EACH_REVERSE_PROTECTED_CONTINUE(ITER, MEMBER, RCULIST) \ +- for (ASSIGN_CONTAINER(ITER, (ITER)->MEMBER.prev, MEMBER); \ +- &(ITER)->MEMBER != (RCULIST); \ +- ASSIGN_CONTAINER(ITER, (ITER)->MEMBER.prev, MEMBER)) +- +-#define RCULIST_FOR_EACH_PROTECTED(ITER, MEMBER, RCULIST) \ +- for (INIT_CONTAINER(ITER, rculist_next_protected(RCULIST), MEMBER); \ +- &(ITER)->MEMBER != (RCULIST); \ +- ASSIGN_CONTAINER(ITER, rculist_next_protected(&(ITER)->MEMBER), \ +- MEMBER)) +- +-#define RCULIST_FOR_EACH_SAFE_PROTECTED(ITER, NEXT, MEMBER, RCULIST) \ +- for (INIT_CONTAINER(ITER, rculist_next_protected(RCULIST), MEMBER); \ +- (&(ITER)->MEMBER != (RCULIST) \ +- ? INIT_CONTAINER(NEXT, rculist_next_protected(&(ITER)->MEMBER), \ +- MEMBER), 1 : 0); \ +- (ITER) = (NEXT)) ++#define RCULIST_FOR_EACH(ITER, MEMBER, RCULIST) \ ++ for (INIT_MULTIVAR(ITER, MEMBER, rculist_next(RCULIST), \ ++ const struct rculist); \ ++ CONDITION_MULTIVAR(ITER, MEMBER, ITER_VAR(ITER) != (RCULIST)); \ ++ UPDATE_MULTIVAR(ITER, rculist_next(ITER_VAR(ITER)))) ++ ++#define RCULIST_FOR_EACH_CONTINUE(ITER, MEMBER, RCULIST) \ ++ for (INIT_MULTIVAR(ITER, MEMBER, rculist_next(&(ITER)->MEMBER), \ ++ const struct rculist); \ ++ CONDITION_MULTIVAR(ITER, MEMBER, ITER_VAR(ITER) != (RCULIST)); \ ++ UPDATE_MULTIVAR(ITER, rculist_next(ITER_VAR(ITER)))) ++ ++#define RCULIST_FOR_EACH_REVERSE_PROTECTED(ITER, MEMBER, RCULIST) \ ++ for (INIT_MULTIVAR(ITER, MEMBER, (RCULIST)->prev, struct rculist); \ ++ CONDITION_MULTIVAR(ITER, MEMBER, ITER_VAR(ITER) != (RCULIST)); \ ++ UPDATE_MULTIVAR(ITER, ITER_VAR(VAR).prev)) ++ ++#define RCULIST_FOR_EACH_REVERSE_PROTECTED_CONTINUE(ITER, MEMBER, RCULIST) \ ++ for (INIT_MULTIVAR(ITER, MEMBER, (ITER)->MEMBER.prev, struct rculist); \ ++ CONDITION_MULTIVAR(ITER, MEMBER, ITER_VAR(ITER) != (RCULIST)); \ ++ UPDATE_MULTIVAR(ITER, ITER_VAR(VAR).prev)) ++ ++#define RCULIST_FOR_EACH_PROTECTED(ITER, MEMBER, RCULIST) \ ++ for (INIT_MULTIVAR(ITER, MEMBER, rculist_next_protected(RCULIST), \ ++ struct rculist); \ ++ CONDITION_MULTIVAR(ITER, MEMBER, ITER_VAR(ITER) != (RCULIST)); \ ++ UPDATE_MULTIVAR(ITER, rculist_next_protected(ITER_VAR(ITER))) \ ++ ++#define RCULIST_FOR_EACH_SAFE_SHORT_PROTECTED(ITER, MEMBER, RCULIST) \ ++ for (INIT_MULTIVAR_SAFE_SHORT(ITER, MEMBER, \ ++ rculist_next_protected(RCULIST), \ ++ struct rculist); \ ++ CONDITION_MULTIVAR_SAFE_SHORT(ITER, MEMBER, \ ++ ITER_VAR(ITER) != (RCULIST), \ ++ ITER_NEXT_VAR(ITER) = rculist_next_protected(ITER_VAR(VAR))); \ ++ UPDATE_MULTIVAR_SHORT(ITER)) ++ ++#define RCULIST_FOR_EACH_SAFE_LONG_PROTECTED(ITER, NEXT, MEMBER, RCULIST) \ ++ for (INIT_MULTIVAR_SAFE_LONG(ITER, NEXT, MEMBER, \ ++ rculist_next_protected(RCULIST) \ ++ struct rculist); \ ++ CONDITION_MULTIVAR_SAFE_LONG(VAR, NEXT, MEMBER \ ++ ITER_VAR(ITER) != (RCULIST), \ ++ ITER_VAR(NEXT) = rculist_next_protected(ITER_VAR(VAR)), \ ++ ITER_VAR(NEXT) != (RCULIST)); \ ++ UPDATE_MULTIVAR_LONG(ITER)) ++ ++#define RCULIST_FOR_EACH_SAFE_PROTECTED(...) \ ++ OVERLOAD_SAFE_MACRO(RCULIST_FOR_EACH_SAFE_LONG_PROTECTED, \ ++ RCULIST_FOR_EACH_SAFE_SHORT_PROTECTED, \ ++ 4, __VA_ARGS__) ++ + + #endif /* rculist.h */ +diff --git a/lib/reconnect.c b/lib/reconnect.c +index a929ddfd2d..89a0bcaf95 100644 +--- a/lib/reconnect.c ++++ b/lib/reconnect.c +@@ -75,7 +75,8 @@ struct reconnect { + + static void reconnect_transition__(struct reconnect *, long long int now, + enum state state); +-static long long int reconnect_deadline__(const struct reconnect *); ++static long long int reconnect_deadline__(const struct reconnect *, ++ long long int now); + static bool reconnect_may_retry(struct reconnect *); + + static const char * +@@ -539,7 +540,7 @@ reconnect_transition__(struct reconnect *fsm, long long int now, + } + + static long long int +-reconnect_deadline__(const struct reconnect *fsm) ++reconnect_deadline__(const struct reconnect *fsm, long long int now) + { + ovs_assert(fsm->state_entered != LLONG_MIN); + switch (fsm->state) { +@@ -557,8 +558,18 @@ reconnect_deadline__(const struct reconnect *fsm) + if (fsm->probe_interval) { + long long int base = MAX(fsm->last_activity, fsm->state_entered); + long long int expiration = base + fsm->probe_interval; +- if (fsm->last_receive_attempt >= expiration) { ++ if (now < expiration || fsm->last_receive_attempt >= expiration) { ++ /* We still have time before the expiration or the time has ++ * already passed and there was no activity. In the first case ++ * we need to wait for the expiration, in the second - we're ++ * already past the deadline. */ + return expiration; ++ } else { ++ /* Time has already passed, but we didn't attempt to receive ++ * anything. We need to wake up and try to receive even if ++ * nothing is pending, so we can update the expiration time or ++ * transition to a different state. */ ++ return now + 1; + } + } + return LLONG_MAX; +@@ -566,8 +577,10 @@ reconnect_deadline__(const struct reconnect *fsm) + case S_IDLE: + if (fsm->probe_interval) { + long long int expiration = fsm->state_entered + fsm->probe_interval; +- if (fsm->last_receive_attempt >= expiration) { ++ if (now < expiration || fsm->last_receive_attempt >= expiration) { + return expiration; ++ } else { ++ return now + 1; + } + } + return LLONG_MAX; +@@ -618,7 +631,7 @@ reconnect_deadline__(const struct reconnect *fsm) + enum reconnect_action + reconnect_run(struct reconnect *fsm, long long int now) + { +- if (now >= reconnect_deadline__(fsm)) { ++ if (now >= reconnect_deadline__(fsm, now)) { + switch (fsm->state) { + case S_VOID: + return 0; +@@ -671,7 +684,7 @@ reconnect_wait(struct reconnect *fsm, long long int now) + int + reconnect_timeout(struct reconnect *fsm, long long int now) + { +- long long int deadline = reconnect_deadline__(fsm); ++ long long int deadline = reconnect_deadline__(fsm, now); + if (deadline != LLONG_MAX) { + long long int remaining = deadline - now; + return MAX(0, MIN(INT_MAX, remaining)); +diff --git a/lib/seq.c b/lib/seq.c +index 6581cb06ba..99e5bf8bd1 100644 +--- a/lib/seq.c ++++ b/lib/seq.c +@@ -297,9 +297,9 @@ static void + seq_thread_woke(struct seq_thread *thread) + OVS_REQUIRES(seq_mutex) + { +- struct seq_waiter *waiter, *next_waiter; ++ struct seq_waiter *waiter; + +- LIST_FOR_EACH_SAFE (waiter, next_waiter, list_node, &thread->waiters) { ++ LIST_FOR_EACH_SAFE (waiter, list_node, &thread->waiters) { + ovs_assert(waiter->thread == thread); + seq_waiter_destroy(waiter); + } +@@ -319,9 +319,9 @@ static void + seq_wake_waiters(struct seq *seq) + OVS_REQUIRES(seq_mutex) + { +- struct seq_waiter *waiter, *next_waiter; ++ struct seq_waiter *waiter; + +- HMAP_FOR_EACH_SAFE (waiter, next_waiter, hmap_node, &seq->waiters) { ++ HMAP_FOR_EACH_SAFE (waiter, hmap_node, &seq->waiters) { + latch_set(&waiter->thread->latch); + seq_waiter_destroy(waiter); + } +diff --git a/lib/shash.c b/lib/shash.c +index a8433629ab..a7b2c64582 100644 +--- a/lib/shash.c ++++ b/lib/shash.c +@@ -68,9 +68,9 @@ shash_moved(struct shash *sh) + void + shash_clear(struct shash *sh) + { +- struct shash_node *node, *next; ++ struct shash_node *node; + +- SHASH_FOR_EACH_SAFE (node, next, sh) { ++ SHASH_FOR_EACH_SAFE (node, sh) { + hmap_remove(&sh->map, &node->node); + free(node->name); + free(node); +@@ -81,9 +81,9 @@ shash_clear(struct shash *sh) + void + shash_clear_free_data(struct shash *sh) + { +- struct shash_node *node, *next; ++ struct shash_node *node; + +- SHASH_FOR_EACH_SAFE (node, next, sh) { ++ SHASH_FOR_EACH_SAFE (node, sh) { + hmap_remove(&sh->map, &node->node); + free(node->data); + free(node->name); +diff --git a/lib/simap.c b/lib/simap.c +index f404ece677..0ee08d74d5 100644 +--- a/lib/simap.c ++++ b/lib/simap.c +@@ -63,9 +63,9 @@ simap_moved(struct simap *simap) + void + simap_clear(struct simap *simap) + { +- struct simap_node *node, *next; ++ struct simap_node *node; + +- SIMAP_FOR_EACH_SAFE (node, next, simap) { ++ SIMAP_FOR_EACH_SAFE (node, simap) { + hmap_remove(&simap->map, &node->node); + free(node->name); + free(node); +diff --git a/lib/simap.h b/lib/simap.h +index 5e646e6607..8db7bea7c9 100644 +--- a/lib/simap.h ++++ b/lib/simap.h +@@ -41,12 +41,22 @@ struct simap_node { + BUILD_ASSERT_TYPE(SIMAP_NODE, struct simap_node *), \ + BUILD_ASSERT_TYPE(SIMAP, struct simap *)) + +-#define SIMAP_FOR_EACH_SAFE(SIMAP_NODE, NEXT, SIMAP) \ +- HMAP_FOR_EACH_SAFE_INIT (SIMAP_NODE, NEXT, node, &(SIMAP)->map, \ ++#define SIMAP_FOR_EACH_SAFE_SHORT(SIMAP_NODE, SIMAP) \ ++ HMAP_FOR_EACH_SAFE_SHORT_INIT (SIMAP_NODE, node, &(SIMAP)->map, \ + BUILD_ASSERT_TYPE(SIMAP_NODE, struct simap_node *), \ +- BUILD_ASSERT_TYPE(NEXT, struct simap_node *), \ + BUILD_ASSERT_TYPE(SIMAP, struct simap *)) + ++#define SIMAP_FOR_EACH_SAFE_LONG(SIMAP_NODE, NEXT, SIMAP) \ ++ HMAP_FOR_EACH_SAFE_LONG_INIT (SIMAP_NODE, NEXT, node, &(SIMAP)->map, \ ++ BUILD_ASSERT_TYPE(SIMAP_NODE, struct simap_node *), \ ++ BUILD_ASSERT_TYPE(NEXT, struct simap_node *), \ ++ BUILD_ASSERT_TYPE(SIMAP, struct simap *)) ++ ++#define SIMAP_FOR_EACH_SAFE(...) \ ++ OVERLOAD_SAFE_MACRO(SIMAP_FOR_EACH_SAFE_LONG, \ ++ SIMAP_FOR_EACH_SAFE_SHORT, \ ++ 3, __VA_ARGS__) ++ + void simap_init(struct simap *); + void simap_destroy(struct simap *); + void simap_swap(struct simap *, struct simap *); +diff --git a/lib/smap.c b/lib/smap.c +index e82261497c..b23eeb52d3 100644 +--- a/lib/smap.c ++++ b/lib/smap.c +@@ -185,9 +185,9 @@ smap_steal(struct smap *smap, struct smap_node *node, + void + smap_clear(struct smap *smap) + { +- struct smap_node *node, *next; ++ struct smap_node *node; + +- SMAP_FOR_EACH_SAFE (node, next, smap) { ++ SMAP_FOR_EACH_SAFE (node, smap) { + smap_remove_node(smap, node); + } + } +diff --git a/lib/smap.h b/lib/smap.h +index a921159667..2fe6c540a7 100644 +--- a/lib/smap.h ++++ b/lib/smap.h +@@ -45,13 +45,24 @@ struct smap_node { + BUILD_ASSERT_TYPE(SMAP_NODE, struct smap_node *), \ + BUILD_ASSERT_TYPE(SMAP, struct smap *)) + +-#define SMAP_FOR_EACH_SAFE(SMAP_NODE, NEXT, SMAP) \ +- HMAP_FOR_EACH_SAFE_INIT ( \ ++#define SMAP_FOR_EACH_SAFE_SHORT(SMAP_NODE, SMAP) \ ++ HMAP_FOR_EACH_SAFE_SHORT_INIT ( \ ++ SMAP_NODE, node, &(SMAP)->map, \ ++ BUILD_ASSERT_TYPE(SMAP_NODE, struct smap_node *), \ ++ BUILD_ASSERT_TYPE(SMAP, struct smap *)) ++ ++#define SMAP_FOR_EACH_SAFE_LONG(SMAP_NODE, NEXT, SMAP) \ ++ HMAP_FOR_EACH_SAFE_LONG_INIT ( \ + SMAP_NODE, NEXT, node, &(SMAP)->map, \ + BUILD_ASSERT_TYPE(SMAP_NODE, struct smap_node *), \ + BUILD_ASSERT_TYPE(NEXT, struct smap_node *), \ + BUILD_ASSERT_TYPE(SMAP, struct smap *)) + ++#define SMAP_FOR_EACH_SAFE(...) \ ++ OVERLOAD_SAFE_MACRO(SMAP_FOR_EACH_SAFE_LONG, \ ++ SMAP_FOR_EACH_SAFE_SHORT, \ ++ 3, __VA_ARGS__) ++ + /* Initializer for an immutable struct smap 'SMAP' that contains one or two + * key-value pairs, e.g. + * +diff --git a/lib/socket-util.c b/lib/socket-util.c +index 4f1ffecf5d..38705cc51e 100644 +--- a/lib/socket-util.c ++++ b/lib/socket-util.c +@@ -62,7 +62,8 @@ static bool parse_sockaddr_components(struct sockaddr_storage *ss, + const char *port_s, + uint16_t default_port, + const char *s, +- bool resolve_host); ++ bool resolve_host, ++ bool *dns_failure); + + /* Sets 'fd' to non-blocking mode. Returns 0 if successful, otherwise a + * positive errno value. */ +@@ -438,7 +439,7 @@ parse_sockaddr_components_dns(struct sockaddr_storage *ss OVS_UNUSED, + dns_resolve(host_s, &tmp_host_s); + if (tmp_host_s != NULL) { + parse_sockaddr_components(ss, tmp_host_s, port_s, +- default_port, s, false); ++ default_port, s, false, NULL); + free(tmp_host_s); + return true; + } +@@ -450,11 +451,15 @@ parse_sockaddr_components(struct sockaddr_storage *ss, + char *host_s, + const char *port_s, uint16_t default_port, + const char *s, +- bool resolve_host) ++ bool resolve_host, bool *dns_failure) + { + struct sockaddr_in *sin = sin_cast(sa_cast(ss)); + int port; + ++ if (dns_failure) { ++ *dns_failure = false; ++ } ++ + if (port_s && port_s[0]) { + if (!str_to_int(port_s, 10, &port) || port < 0 || port > 65535) { + VLOG_ERR("%s: bad port number \"%s\"", s, port_s); +@@ -501,10 +506,15 @@ parse_sockaddr_components(struct sockaddr_storage *ss, + return true; + + resolve: +- if (resolve_host && parse_sockaddr_components_dns(ss, host_s, port_s, +- default_port, s)) { +- return true; +- } else if (!resolve_host) { ++ if (resolve_host) { ++ if (parse_sockaddr_components_dns(ss, host_s, port_s, ++ default_port, s)) { ++ return true; ++ } ++ if (dns_failure) { ++ *dns_failure = true; ++ } ++ } else { + VLOG_ERR("%s: bad IP address \"%s\"", s, host_s); + } + exit: +@@ -521,10 +531,12 @@ exit: + * It resolves the host if 'resolve_host' is true. + * + * On success, returns true and stores the parsed remote address into '*ss'. +- * On failure, logs an error, stores zeros into '*ss', and returns false. */ ++ * On failure, logs an error, stores zeros into '*ss', and returns false, ++ * '*dns_failure' indicates if the host resolution failed. */ + bool + inet_parse_active(const char *target_, int default_port, +- struct sockaddr_storage *ss, bool resolve_host) ++ struct sockaddr_storage *ss, ++ bool resolve_host, bool *dns_failure) + { + char *target = xstrdup(target_); + char *port, *host; +@@ -539,7 +551,7 @@ inet_parse_active(const char *target_, int default_port, + ok = false; + } else { + ok = parse_sockaddr_components(ss, host, port, default_port, +- target_, resolve_host); ++ target_, resolve_host, dns_failure); + } + if (!ok) { + memset(ss, 0, sizeof *ss); +@@ -576,7 +588,7 @@ inet_open_active(int style, const char *target, int default_port, + int error; + + /* Parse. */ +- if (!inet_parse_active(target, default_port, &ss, true)) { ++ if (!inet_parse_active(target, default_port, &ss, true, NULL)) { + error = EAFNOSUPPORT; + goto exit; + } +@@ -660,7 +672,7 @@ inet_parse_passive(const char *target_, int default_port, + ok = false; + } else { + ok = parse_sockaddr_components(ss, host, port, default_port, +- target_, true); ++ target_, true, NULL); + } + if (!ok) { + memset(ss, 0, sizeof *ss); +@@ -783,7 +795,8 @@ inet_parse_address(const char *target_, struct sockaddr_storage *ss) + { + char *target = xstrdup(target_); + char *host = unbracket(target); +- bool ok = parse_sockaddr_components(ss, host, NULL, 0, target_, false); ++ bool ok = parse_sockaddr_components(ss, host, NULL, 0, ++ target_, false, NULL); + if (!ok) { + memset(ss, 0, sizeof *ss); + } +diff --git a/lib/socket-util.h b/lib/socket-util.h +index 9ccb7d4cc4..bf66393df9 100644 +--- a/lib/socket-util.h ++++ b/lib/socket-util.h +@@ -49,7 +49,8 @@ ovs_be32 guess_netmask(ovs_be32 ip); + void inet_parse_host_port_tokens(char *s, char **hostp, char **portp); + void inet_parse_port_host_tokens(char *s, char **portp, char **hostp); + bool inet_parse_active(const char *target, int default_port, +- struct sockaddr_storage *ssp, bool resolve_host); ++ struct sockaddr_storage *ssp, ++ bool resolve_host, bool *dns_failure); + int inet_open_active(int style, const char *target, int default_port, + struct sockaddr_storage *ssp, int *fdp, uint8_t dscp); + +diff --git a/lib/sset.c b/lib/sset.c +index b2e3f43ec9..6fbaa9d60d 100644 +--- a/lib/sset.c ++++ b/lib/sset.c +@@ -212,9 +212,9 @@ sset_add_array(struct sset *set, char **names, size_t n) + void + sset_clear(struct sset *set) + { +- const char *name, *next; ++ const char *name; + +- SSET_FOR_EACH_SAFE (name, next, set) { ++ SSET_FOR_EACH_SAFE (name, set) { + sset_delete(set, SSET_NODE_FROM_NAME(name)); + } + } +@@ -312,7 +312,9 @@ sset_at_position(const struct sset *set, struct sset_position *pos) + struct hmap_node *hmap_node; + + hmap_node = hmap_at_position(&set->map, &pos->pos); +- return SSET_NODE_FROM_HMAP_NODE(hmap_node); ++ return hmap_node ++ ? SSET_NODE_FROM_HMAP_NODE(hmap_node) ++ : NULL; + } + + /* Replaces 'a' by the intersection of 'a' and 'b'. That is, removes from 'a' +@@ -320,9 +322,9 @@ sset_at_position(const struct sset *set, struct sset_position *pos) + void + sset_intersect(struct sset *a, const struct sset *b) + { +- const char *name, *next; ++ const char *name; + +- SSET_FOR_EACH_SAFE (name, next, a) { ++ SSET_FOR_EACH_SAFE (name, a) { + if (!sset_contains(b, name)) { + sset_delete(a, SSET_NODE_FROM_NAME(name)); + } +diff --git a/lib/sset.h b/lib/sset.h +index f0bb8b5344..214d6fb41c 100644 +--- a/lib/sset.h ++++ b/lib/sset.h +@@ -87,13 +87,26 @@ void sset_intersect(struct sset *, const struct sset *); + NAME != NULL; \ + (NAME) = SSET_NEXT(SSET, NAME)) + +-#define SSET_FOR_EACH_SAFE(NAME, NEXT, SSET) \ ++#define SSET_FOR_EACH_SAFE_LONG(NAME, NEXT, SSET) \ + for ((NAME) = SSET_FIRST(SSET); \ + (NAME != NULL \ + ? (NEXT) = SSET_NEXT(SSET, NAME), true \ + : false); \ + (NAME) = (NEXT)) + ++#define SSET_FOR_EACH_SAFE_SHORT(NAME, SSET) \ ++ for (const char * NAME__next = \ ++ ((NAME) = SSET_FIRST(SSET), NULL); \ ++ (NAME != NULL \ ++ ? (NAME__next = SSET_NEXT(SSET, NAME), true) \ ++ : (NAME__next = NULL, false)); \ ++ (NAME) = NAME__next) ++ ++#define SSET_FOR_EACH_SAFE(...) \ ++ OVERLOAD_SAFE_MACRO(SSET_FOR_EACH_SAFE_LONG, \ ++ SSET_FOR_EACH_SAFE_SHORT, \ ++ 3, __VA_ARGS__) ++ + const char **sset_array(const struct sset *); + const char **sset_sort(const struct sset *); + +diff --git a/lib/stopwatch.c b/lib/stopwatch.c +index 1c71df1a12..ec567603b1 100644 +--- a/lib/stopwatch.c ++++ b/lib/stopwatch.c +@@ -464,7 +464,7 @@ stopwatch_thread(void *ign OVS_UNUSED) + static void + stopwatch_exit(void) + { +- struct shash_node *node, *node_next; ++ struct shash_node *node; + struct stopwatch_packet *pkt = stopwatch_packet_create(OP_SHUTDOWN); + stopwatch_packet_write(pkt); + xpthread_join(stopwatch_thread_id, NULL); +@@ -473,7 +473,7 @@ stopwatch_exit(void) + * other competing thread. We are now the sole owners + * of all data in the file. + */ +- SHASH_FOR_EACH_SAFE (node, node_next, &stopwatches) { ++ SHASH_FOR_EACH_SAFE (node, &stopwatches) { + struct stopwatch *sw = node->data; + shash_delete(&stopwatches, node); + free(sw); +diff --git a/lib/stream.c b/lib/stream.c +index fcaddf10ad..71039e24f1 100644 +--- a/lib/stream.c ++++ b/lib/stream.c +@@ -788,7 +788,7 @@ stream_parse_target_with_default_port(const char *target, int default_port, + struct sockaddr_storage *ss) + { + return ((!strncmp(target, "tcp:", 4) || !strncmp(target, "ssl:", 4)) +- && inet_parse_active(target + 4, default_port, ss, true)); ++ && inet_parse_active(target + 4, default_port, ss, true, NULL)); + } + + /* Attempts to guess the content type of a stream whose first few bytes were +diff --git a/lib/tc.c b/lib/tc.c +index adb2d3182a..276cc54b35 100644 +--- a/lib/tc.c ++++ b/lib/tc.c +@@ -395,8 +395,14 @@ static const struct nl_policy tca_flower_policy[] = { + [TCA_FLOWER_KEY_ENC_IPV6_DST_MASK] = { .type = NL_A_UNSPEC, + .min_len = sizeof(struct in6_addr), + .optional = true, }, ++ [TCA_FLOWER_KEY_ENC_UDP_SRC_PORT] = { .type = NL_A_U16, ++ .optional = true, }, ++ [TCA_FLOWER_KEY_ENC_UDP_SRC_PORT_MASK] = { .type = NL_A_U16, ++ .optional = true, }, + [TCA_FLOWER_KEY_ENC_UDP_DST_PORT] = { .type = NL_A_U16, + .optional = true, }, ++ [TCA_FLOWER_KEY_ENC_UDP_DST_PORT_MASK] = { .type = NL_A_U16, ++ .optional = true, }, + [TCA_FLOWER_KEY_FLAGS] = { .type = NL_A_BE32, .optional = true, }, + [TCA_FLOWER_KEY_FLAGS_MASK] = { .type = NL_A_BE32, .optional = true, }, + [TCA_FLOWER_KEY_IP_TTL] = { .type = NL_A_U8, +@@ -568,16 +574,17 @@ nl_parse_flower_vlan(struct nlattr **attrs, struct tc_flower *flower) + + flower->key.encap_eth_type[0] = + nl_attr_get_be16(attrs[TCA_FLOWER_KEY_ETH_TYPE]); ++ flower->mask.encap_eth_type[0] = CONSTANT_HTONS(0xffff); + + if (attrs[TCA_FLOWER_KEY_VLAN_ID]) { + flower->key.vlan_id[0] = + nl_attr_get_u16(attrs[TCA_FLOWER_KEY_VLAN_ID]); +- flower->mask.vlan_id[0] = 0xffff; ++ flower->mask.vlan_id[0] = VLAN_VID_MASK >> VLAN_VID_SHIFT; + } + if (attrs[TCA_FLOWER_KEY_VLAN_PRIO]) { + flower->key.vlan_prio[0] = + nl_attr_get_u8(attrs[TCA_FLOWER_KEY_VLAN_PRIO]); +- flower->mask.vlan_prio[0] = 0xff; ++ flower->mask.vlan_prio[0] = VLAN_PCP_MASK >> VLAN_PCP_SHIFT; + } + + if (!attrs[TCA_FLOWER_KEY_VLAN_ETH_TYPE]) { +@@ -590,17 +597,18 @@ nl_parse_flower_vlan(struct nlattr **attrs, struct tc_flower *flower) + } + + flower->key.encap_eth_type[1] = flower->key.encap_eth_type[0]; ++ flower->mask.encap_eth_type[1] = CONSTANT_HTONS(0xffff); + flower->key.encap_eth_type[0] = encap_ethtype; + + if (attrs[TCA_FLOWER_KEY_CVLAN_ID]) { + flower->key.vlan_id[1] = + nl_attr_get_u16(attrs[TCA_FLOWER_KEY_CVLAN_ID]); +- flower->mask.vlan_id[1] = 0xffff; ++ flower->mask.vlan_id[1] = VLAN_VID_MASK >> VLAN_VID_SHIFT; + } + if (attrs[TCA_FLOWER_KEY_CVLAN_PRIO]) { + flower->key.vlan_prio[1] = + nl_attr_get_u8(attrs[TCA_FLOWER_KEY_CVLAN_PRIO]); +- flower->mask.vlan_prio[1] = 0xff; ++ flower->mask.vlan_prio[1] = VLAN_PCP_MASK >> VLAN_PCP_SHIFT; + } + } + +@@ -705,15 +713,17 @@ flower_tun_geneve_opt_check_len(struct tun_metadata *key, + const struct geneve_opt *opt, *opt_mask; + int len, cnt = 0; + ++ if (key->present.len != mask->present.len) { ++ goto bad_length; ++ } ++ + len = key->present.len; + while (len) { + opt = &key->opts.gnv[cnt]; + opt_mask = &mask->opts.gnv[cnt]; + + if (opt->length != opt_mask->length) { +- VLOG_ERR_RL(&error_rl, +- "failed to parse tun options; key/mask length differ"); +- return EINVAL; ++ goto bad_length; + } + + cnt += sizeof(struct geneve_opt) / 4 + opt->length; +@@ -721,6 +731,11 @@ flower_tun_geneve_opt_check_len(struct tun_metadata *key, + } + + return 0; ++ ++bad_length: ++ VLOG_ERR_RL(&error_rl, ++ "failed to parse tun options; key/mask length differ"); ++ return EINVAL; + } + + static int +@@ -758,7 +773,15 @@ nl_parse_flower_tunnel(struct nlattr **attrs, struct tc_flower *flower) + flower->key.tunnel.ipv6.ipv6_dst = + nl_attr_get_in6_addr(attrs[TCA_FLOWER_KEY_ENC_IPV6_DST]); + } +- if (attrs[TCA_FLOWER_KEY_ENC_UDP_DST_PORT]) { ++ if (attrs[TCA_FLOWER_KEY_ENC_UDP_SRC_PORT_MASK]) { ++ flower->mask.tunnel.tp_src = ++ nl_attr_get_be16(attrs[TCA_FLOWER_KEY_ENC_UDP_SRC_PORT_MASK]); ++ flower->key.tunnel.tp_src = ++ nl_attr_get_be16(attrs[TCA_FLOWER_KEY_ENC_UDP_SRC_PORT]); ++ } ++ if (attrs[TCA_FLOWER_KEY_ENC_UDP_DST_PORT_MASK]) { ++ flower->mask.tunnel.tp_dst = ++ nl_attr_get_be16(attrs[TCA_FLOWER_KEY_ENC_UDP_DST_PORT_MASK]); + flower->key.tunnel.tp_dst = + nl_attr_get_be16(attrs[TCA_FLOWER_KEY_ENC_UDP_DST_PORT]); + } +@@ -937,24 +960,21 @@ nl_parse_flower_ip(struct nlattr **attrs, struct tc_flower *flower) { + key->icmp_code = + nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV4_CODE]); + mask->icmp_code = +- nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV4_CODE]); ++ nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV4_CODE_MASK]); + } + if (attrs[TCA_FLOWER_KEY_ICMPV4_TYPE_MASK]) { +- key->icmp_type = +- nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV4_TYPE_MASK]); ++ key->icmp_type = nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV4_TYPE]); + mask->icmp_type = + nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV4_TYPE_MASK]); + } + } else if (ip_proto == IPPROTO_ICMPV6) { + if (attrs[TCA_FLOWER_KEY_ICMPV6_CODE_MASK]) { +- key->icmp_code = +- nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV6_CODE]); ++ key->icmp_code = nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV6_CODE]); + mask->icmp_code = +- nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV6_CODE]); ++ nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV6_CODE_MASK]); + } + if (attrs[TCA_FLOWER_KEY_ICMPV6_TYPE_MASK]) { +- key->icmp_type = +- nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV6_TYPE_MASK]); ++ key->icmp_type = nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV6_TYPE]); + mask->icmp_type = + nl_attr_get_u8(attrs[TCA_FLOWER_KEY_ICMPV6_TYPE_MASK]); + } +@@ -1006,14 +1026,14 @@ static const struct nl_policy pedit_policy[] = { + static int + nl_parse_act_pedit(struct nlattr *options, struct tc_flower *flower) + { +- struct tc_action *action; ++ struct tc_action *action = &flower->actions[flower->action_count++]; + struct nlattr *pe_attrs[ARRAY_SIZE(pedit_policy)]; + const struct tc_pedit *pe; + const struct tc_pedit_key *keys; + const struct nlattr *nla, *keys_ex, *ex_type; + const void *keys_attr; +- char *rewrite_key = (void *) &flower->rewrite.key; +- char *rewrite_mask = (void *) &flower->rewrite.mask; ++ char *rewrite_key = (void *) &action->rewrite.key; ++ char *rewrite_mask = (void *) &action->rewrite.mask; + size_t keys_ex_size, left; + int type, i = 0, err; + +@@ -1092,7 +1112,6 @@ nl_parse_act_pedit(struct nlattr *options, struct tc_flower *flower) + i++; + } + +- action = &flower->actions[flower->action_count++]; + action->type = TC_ACT_PEDIT; + + return 0; +@@ -1314,8 +1333,8 @@ nl_parse_act_gact(struct nlattr *options, struct tc_flower *flower) + struct nlattr *gact_attrs[ARRAY_SIZE(gact_policy)]; + const struct tc_gact *p; + struct nlattr *gact_parms; +- const struct tcf_t *tm; + struct tc_action *action; ++ struct tcf_t tm; + + if (!nl_parse_nested(options, gact_policy, gact_attrs, + ARRAY_SIZE(gact_policy))) { +@@ -1335,8 +1354,9 @@ nl_parse_act_gact(struct nlattr *options, struct tc_flower *flower) + return EINVAL; + } + +- tm = nl_attr_get_unspec(gact_attrs[TCA_GACT_TM], sizeof *tm); +- nl_parse_tcf(tm, flower); ++ memcpy(&tm, nl_attr_get_unspec(gact_attrs[TCA_GACT_TM], sizeof tm), ++ sizeof tm); ++ nl_parse_tcf(&tm, flower); + + return 0; + } +@@ -1357,9 +1377,9 @@ nl_parse_act_mirred(struct nlattr *options, struct tc_flower *flower) + struct nlattr *mirred_attrs[ARRAY_SIZE(mirred_policy)]; + const struct tc_mirred *m; + const struct nlattr *mirred_parms; +- const struct tcf_t *tm; + struct nlattr *mirred_tm; + struct tc_action *action; ++ struct tcf_t tm; + + if (!nl_parse_nested(options, mirred_policy, mirred_attrs, + ARRAY_SIZE(mirred_policy))) { +@@ -1387,8 +1407,8 @@ nl_parse_act_mirred(struct nlattr *options, struct tc_flower *flower) + action->type = TC_ACT_OUTPUT; + + mirred_tm = mirred_attrs[TCA_MIRRED_TM]; +- tm = nl_attr_get_unspec(mirred_tm, sizeof *tm); +- nl_parse_tcf(tm, flower); ++ memcpy(&tm, nl_attr_get_unspec(mirred_tm, sizeof tm), sizeof tm); ++ nl_parse_tcf(&tm, flower); + + return 0; + } +@@ -1487,7 +1507,9 @@ nl_parse_act_ct(struct nlattr *options, struct tc_flower *flower) + if (ipv4_max) { + ovs_be32 addr = nl_attr_get_be32(ipv4_max); + +- action->ct.range.ipv4.max = addr; ++ if (action->ct.range.ipv4.min != addr) { ++ action->ct.range.ipv4.max = addr; ++ } + } + } else if (ipv6_min) { + action->ct.range.ip_family = AF_INET6; +@@ -1496,7 +1518,9 @@ nl_parse_act_ct(struct nlattr *options, struct tc_flower *flower) + if (ipv6_max) { + struct in6_addr addr = nl_attr_get_in6_addr(ipv6_max); + +- action->ct.range.ipv6.max = addr; ++ if (!ipv6_addr_equals(&action->ct.range.ipv6.min, &addr)) { ++ action->ct.range.ipv6.max = addr; ++ } + } + } + +@@ -1504,6 +1528,10 @@ nl_parse_act_ct(struct nlattr *options, struct tc_flower *flower) + action->ct.range.port.min = nl_attr_get_be16(port_min); + if (port_max) { + action->ct.range.port.max = nl_attr_get_be16(port_max); ++ if (action->ct.range.port.min == ++ action->ct.range.port.max) { ++ action->ct.range.port.max = 0; ++ } + } + } + } +@@ -1702,6 +1730,9 @@ static const struct nl_policy stats_policy[] = { + [TCA_STATS_BASIC] = { .type = NL_A_UNSPEC, + .min_len = sizeof(struct gnet_stats_basic), + .optional = false, }, ++ [TCA_STATS_BASIC_HW] = { .type = NL_A_UNSPEC, ++ .min_len = sizeof(struct gnet_stats_basic), ++ .optional = true, }, + }; + + static int +@@ -1714,8 +1745,9 @@ nl_parse_single_action(struct nlattr *action, struct tc_flower *flower, + const char *act_kind; + struct nlattr *action_attrs[ARRAY_SIZE(act_policy)]; + struct nlattr *stats_attrs[ARRAY_SIZE(stats_policy)]; +- struct ovs_flow_stats *stats = &flower->stats; +- const struct gnet_stats_basic *bs; ++ struct ovs_flow_stats *stats_sw = &flower->stats_sw; ++ struct ovs_flow_stats *stats_hw = &flower->stats_hw; ++ struct gnet_stats_basic bs_all, bs_hw, bs_sw; + int err = 0; + + if (!nl_parse_nested(action, act_policy, action_attrs, +@@ -1771,10 +1803,30 @@ nl_parse_single_action(struct nlattr *action, struct tc_flower *flower, + return EPROTO; + } + +- bs = nl_attr_get_unspec(stats_attrs[TCA_STATS_BASIC], sizeof *bs); +- if (bs->packets) { +- put_32aligned_u64(&stats->n_packets, bs->packets); +- put_32aligned_u64(&stats->n_bytes, bs->bytes); ++ memcpy(&bs_all, ++ nl_attr_get_unspec(stats_attrs[TCA_STATS_BASIC], sizeof bs_all), ++ sizeof bs_all); ++ if (stats_attrs[TCA_STATS_BASIC_HW]) { ++ memcpy(&bs_hw, nl_attr_get_unspec(stats_attrs[TCA_STATS_BASIC_HW], ++ sizeof bs_hw), ++ sizeof bs_hw); ++ ++ bs_sw.packets = bs_all.packets - bs_hw.packets; ++ bs_sw.bytes = bs_all.bytes - bs_hw.bytes; ++ } else { ++ bs_sw.packets = bs_all.packets; ++ bs_sw.bytes = bs_all.bytes; ++ } ++ ++ if (bs_sw.packets > get_32aligned_u64(&stats_sw->n_packets)) { ++ put_32aligned_u64(&stats_sw->n_packets, bs_sw.packets); ++ put_32aligned_u64(&stats_sw->n_bytes, bs_sw.bytes); ++ } ++ ++ if (stats_attrs[TCA_STATS_BASIC_HW] ++ && bs_hw.packets > get_32aligned_u64(&stats_hw->n_packets)) { ++ put_32aligned_u64(&stats_hw->n_packets, bs_hw.packets); ++ put_32aligned_u64(&stats_hw->n_bytes, bs_hw.bytes); + } + + return 0; +@@ -2399,14 +2451,14 @@ nl_msg_put_act_flags(struct ofpbuf *request) { + * first_word_mask/last_word_mask - the mask to use for the first/last read + * (as we read entire words). */ + static void +-calc_offsets(struct tc_flower *flower, struct flower_key_to_pedit *m, ++calc_offsets(struct tc_action *action, struct flower_key_to_pedit *m, + int *cur_offset, int *cnt, ovs_be32 *last_word_mask, + ovs_be32 *first_word_mask, ovs_be32 **mask, ovs_be32 **data) + { + int start_offset, max_offset, total_size; + int diff, right_zero_bits, left_zero_bits; +- char *rewrite_key = (void *) &flower->rewrite.key; +- char *rewrite_mask = (void *) &flower->rewrite.mask; ++ char *rewrite_key = (void *) &action->rewrite.key; ++ char *rewrite_mask = (void *) &action->rewrite.mask; + + max_offset = m->offset + m->size; + start_offset = ROUND_DOWN(m->offset, 4); +@@ -2473,7 +2525,8 @@ csum_update_flag(struct tc_flower *flower, + + static int + nl_msg_put_flower_rewrite_pedits(struct ofpbuf *request, +- struct tc_flower *flower) ++ struct tc_flower *flower, ++ struct tc_action *action) + { + struct { + struct tc_pedit sel; +@@ -2497,12 +2550,12 @@ nl_msg_put_flower_rewrite_pedits(struct ofpbuf *request, + continue; + } + +- calc_offsets(flower, m, &cur_offset, &cnt, &last_word_mask, ++ calc_offsets(action, m, &cur_offset, &cnt, &last_word_mask, + &first_word_mask, &mask, &data); + + for (j = 0; j < cnt; j++, mask++, data++, cur_offset += 4) { +- ovs_be32 mask_word = *mask; +- ovs_be32 data_word = *data; ++ ovs_be32 mask_word = get_unaligned_be32(mask); ++ ovs_be32 data_word = get_unaligned_be32(data); + + if (j == 0) { + mask_word &= first_word_mask; +@@ -2556,6 +2609,29 @@ nl_msg_put_flower_acts_release(struct ofpbuf *request, uint16_t act_index) + nl_msg_end_nested(request, act_offset); + } + ++/* Aggregates all previous successive pedit actions csum_update_flags ++ * to flower->csum_update_flags. Only append one csum action to the ++ * last pedit action. */ ++static void ++nl_msg_put_csum_act(struct ofpbuf *request, struct tc_flower *flower, ++ uint16_t *act_index) ++{ ++ size_t act_offset; ++ ++ /* No pedit actions or processed already. */ ++ if (!flower->csum_update_flags) { ++ return; ++ } ++ ++ act_offset = nl_msg_start_nested(request, (*act_index)++); ++ nl_msg_put_act_csum(request, flower->csum_update_flags); ++ nl_msg_put_act_flags(request); ++ nl_msg_end_nested(request, act_offset); ++ ++ /* Clear it. So we can have another series of pedit actions. */ ++ flower->csum_update_flags = 0; ++} ++ + static int + nl_msg_put_flower_acts(struct ofpbuf *request, struct tc_flower *flower) + { +@@ -2572,20 +2648,22 @@ nl_msg_put_flower_acts(struct ofpbuf *request, struct tc_flower *flower) + + action = flower->actions; + for (i = 0; i < flower->action_count; i++, action++) { ++ if (action->type != TC_ACT_PEDIT) { ++ nl_msg_put_csum_act(request, flower, &act_index); ++ } + switch (action->type) { + case TC_ACT_PEDIT: { + act_offset = nl_msg_start_nested(request, act_index++); +- error = nl_msg_put_flower_rewrite_pedits(request, flower); ++ error = nl_msg_put_flower_rewrite_pedits(request, flower, ++ action); + if (error) { + return error; + } + nl_msg_end_nested(request, act_offset); + +- if (flower->csum_update_flags) { +- act_offset = nl_msg_start_nested(request, act_index++); +- nl_msg_put_act_csum(request, flower->csum_update_flags); +- nl_msg_put_act_flags(request); +- nl_msg_end_nested(request, act_offset); ++ if (i == flower->action_count - 1) { ++ /* If this is the last action check csum calc again. */ ++ nl_msg_put_csum_act(request, flower, &act_index); + } + } + break; +@@ -2792,13 +2870,16 @@ nl_msg_put_flower_tunnel(struct ofpbuf *request, struct tc_flower *flower) + struct in6_addr *ipv6_dst_mask = &flower->mask.tunnel.ipv6.ipv6_dst; + struct in6_addr *ipv6_src = &flower->key.tunnel.ipv6.ipv6_src; + struct in6_addr *ipv6_dst = &flower->key.tunnel.ipv6.ipv6_dst; +- ovs_be16 tp_dst = flower->key.tunnel.tp_dst; + ovs_be32 id = be64_to_be32(flower->key.tunnel.id); ++ ovs_be16 tp_src = flower->key.tunnel.tp_src; ++ ovs_be16 tp_dst = flower->key.tunnel.tp_dst; + uint8_t tos = flower->key.tunnel.tos; + uint8_t ttl = flower->key.tunnel.ttl; + uint8_t tos_mask = flower->mask.tunnel.tos; + uint8_t ttl_mask = flower->mask.tunnel.ttl; + ovs_be64 id_mask = flower->mask.tunnel.id; ++ ovs_be16 tp_src_mask = flower->mask.tunnel.tp_src; ++ ovs_be16 tp_dst_mask = flower->mask.tunnel.tp_dst; + + if (ipv4_dst_mask || ipv4_src_mask) { + nl_msg_put_be32(request, TCA_FLOWER_KEY_ENC_IPV4_DST_MASK, +@@ -2824,8 +2905,15 @@ nl_msg_put_flower_tunnel(struct ofpbuf *request, struct tc_flower *flower) + nl_msg_put_u8(request, TCA_FLOWER_KEY_ENC_IP_TTL, ttl); + nl_msg_put_u8(request, TCA_FLOWER_KEY_ENC_IP_TTL_MASK, ttl_mask); + } +- if (tp_dst) { ++ if (tp_src_mask) { ++ nl_msg_put_be16(request, TCA_FLOWER_KEY_ENC_UDP_SRC_PORT, tp_src); ++ nl_msg_put_be16(request, TCA_FLOWER_KEY_ENC_UDP_SRC_PORT_MASK, ++ tp_src_mask); ++ } ++ if (tp_dst_mask) { + nl_msg_put_be16(request, TCA_FLOWER_KEY_ENC_UDP_DST_PORT, tp_dst); ++ nl_msg_put_be16(request, TCA_FLOWER_KEY_ENC_UDP_DST_PORT_MASK, ++ tp_dst_mask); + } + if (id_mask) { + nl_msg_put_be32(request, TCA_FLOWER_KEY_ENC_KEY_ID, id); +@@ -2914,13 +3002,13 @@ nl_msg_put_flower_options(struct ofpbuf *request, struct tc_flower *flower) + FLOWER_PUT_MASKED_VALUE(icmp_code, TCA_FLOWER_KEY_ICMPV6_CODE); + FLOWER_PUT_MASKED_VALUE(icmp_type, TCA_FLOWER_KEY_ICMPV6_TYPE); + } +- +- FLOWER_PUT_MASKED_VALUE(ct_state, TCA_FLOWER_KEY_CT_STATE); +- FLOWER_PUT_MASKED_VALUE(ct_zone, TCA_FLOWER_KEY_CT_ZONE); +- FLOWER_PUT_MASKED_VALUE(ct_mark, TCA_FLOWER_KEY_CT_MARK); +- FLOWER_PUT_MASKED_VALUE(ct_label, TCA_FLOWER_KEY_CT_LABELS); + } + ++ FLOWER_PUT_MASKED_VALUE(ct_state, TCA_FLOWER_KEY_CT_STATE); ++ FLOWER_PUT_MASKED_VALUE(ct_zone, TCA_FLOWER_KEY_CT_ZONE); ++ FLOWER_PUT_MASKED_VALUE(ct_mark, TCA_FLOWER_KEY_CT_MARK); ++ FLOWER_PUT_MASKED_VALUE(ct_label, TCA_FLOWER_KEY_CT_LABELS); ++ + if (host_eth_type == ETH_P_IP) { + FLOWER_PUT_MASKED_VALUE(ipv4.ipv4_src, TCA_FLOWER_KEY_IPV4_SRC); + FLOWER_PUT_MASKED_VALUE(ipv4.ipv4_dst, TCA_FLOWER_KEY_IPV4_DST); +@@ -2993,12 +3081,79 @@ nl_msg_put_flower_options(struct ofpbuf *request, struct tc_flower *flower) + return 0; + } + ++static void ++log_tc_flower_match(const char *msg, ++ const struct tc_flower *a, ++ const struct tc_flower *b) ++{ ++ uint8_t key_a[sizeof(struct tc_flower_key)]; ++ uint8_t key_b[sizeof(struct tc_flower_key)]; ++ struct ds s = DS_EMPTY_INITIALIZER; ++ ++ for (int i = 0; i < sizeof a->key; i++) { ++ uint8_t mask_a = ((uint8_t *) &a->mask)[i]; ++ uint8_t mask_b = ((uint8_t *) &b->mask)[i]; ++ ++ key_a[i] = ((uint8_t *) &a->key)[i] & mask_a; ++ key_b[i] = ((uint8_t *) &b->key)[i] & mask_b; ++ } ++ ds_put_cstr(&s, "\nExpected Mask:\n"); ++ ds_put_sparse_hex_dump(&s, &a->mask, sizeof a->mask, 0, false); ++ ds_put_cstr(&s, "\nReceived Mask:\n"); ++ ds_put_sparse_hex_dump(&s, &b->mask, sizeof b->mask, 0, false); ++ ds_put_cstr(&s, "\nExpected Key:\n"); ++ ds_put_sparse_hex_dump(&s, &a->key, sizeof a->key, 0, false); ++ ds_put_cstr(&s, "\nReceived Key:\n"); ++ ds_put_sparse_hex_dump(&s, &b->key, sizeof b->key, 0, false); ++ ds_put_cstr(&s, "\nExpected Masked Key:\n"); ++ ds_put_sparse_hex_dump(&s, key_a, sizeof key_a, 0, false); ++ ds_put_cstr(&s, "\nReceived Masked Key:\n"); ++ ds_put_sparse_hex_dump(&s, key_b, sizeof key_b, 0, false); ++ ++ if (a->action_count != b->action_count) { ++ /* If action count is not equal, we print all actions to see which ++ * ones are missing. */ ++ const struct tc_action *action; ++ int i; ++ ++ ds_put_cstr(&s, "\nExpected Actions:\n"); ++ for (i = 0, action = a->actions; i < a->action_count; i++, action++) { ++ ds_put_cstr(&s, " - "); ++ ds_put_hex(&s, action, sizeof *action); ++ ds_put_cstr(&s, "\n"); ++ } ++ ds_put_cstr(&s, "Received Actions:\n"); ++ for (i = 0, action = b->actions; i < b->action_count; i++, action++) { ++ ds_put_cstr(&s, " - "); ++ ds_put_hex(&s, action, sizeof *action); ++ ds_put_cstr(&s, "\n"); ++ } ++ } else { ++ /* Only dump the delta in actions. */ ++ const struct tc_action *action_a = a->actions; ++ const struct tc_action *action_b = b->actions; ++ ++ for (int i = 0; i < a->action_count; i++, action_a++, action_b++) { ++ if (memcmp(action_a, action_b, sizeof *action_a)) { ++ ds_put_format(&s, ++ "\nAction %d mismatch:\n - Expected Action: ", ++ i); ++ ds_put_hex(&s, action_a, sizeof *action_a); ++ ds_put_cstr(&s, "\n - Received Action: "); ++ ds_put_hex(&s, action_b, sizeof *action_b); ++ } ++ } ++ } ++ VLOG_DBG_RL(&error_rl, "%s%s", msg, ds_cstr(&s)); ++ ds_destroy(&s); ++} ++ + static bool + cmp_tc_flower_match_action(const struct tc_flower *a, + const struct tc_flower *b) + { + if (memcmp(&a->mask, &b->mask, sizeof a->mask)) { +- VLOG_DBG_RL(&error_rl, "tc flower compare failed mask compare"); ++ log_tc_flower_match("tc flower compare failed mask compare:", a, b); + return false; + } + +@@ -3011,8 +3166,8 @@ cmp_tc_flower_match_action(const struct tc_flower *a, + uint8_t key_b = ((uint8_t *)&b->key)[i] & mask; + + if (key_a != key_b) { +- VLOG_DBG_RL(&error_rl, "tc flower compare failed key compare at " +- "%d", i); ++ log_tc_flower_match("tc flower compare failed masked key compare:", ++ a, b); + return false; + } + } +@@ -3022,14 +3177,15 @@ cmp_tc_flower_match_action(const struct tc_flower *a, + const struct tc_action *action_b = b->actions; + + if (a->action_count != b->action_count) { +- VLOG_DBG_RL(&error_rl, "tc flower compare failed action length check"); ++ log_tc_flower_match("tc flower compare failed action length check", ++ a, b); + return false; + } + + for (int i = 0; i < a->action_count; i++, action_a++, action_b++) { + if (memcmp(action_a, action_b, sizeof *action_a)) { +- VLOG_DBG_RL(&error_rl, "tc flower compare failed action compare " +- "for %d", i); ++ log_tc_flower_match("tc flower compare failed action compare", ++ a, b); + return false; + } + } +diff --git a/lib/tc.h b/lib/tc.h +index a147ca461d..d6cdddd169 100644 +--- a/lib/tc.h ++++ b/lib/tc.h +@@ -256,11 +256,23 @@ struct tc_action { + bool force; + bool commit; + } ct; ++ ++ struct { ++ struct tc_flower_key key; ++ struct tc_flower_key mask; ++ } rewrite; + }; + + enum tc_action_type type; + }; + ++/* assert that if we overflow with a masked write of uint32_t to the last byte ++ * of action.rewrite we overflow inside struct tc_action. ++ * shouldn't happen unless someone moves rewrite to the end of action */ ++BUILD_ASSERT_DECL(offsetof(struct tc_action, rewrite) ++ + MEMBER_SIZEOF(struct tc_action, rewrite) ++ + sizeof(uint32_t) - 2 < sizeof(struct tc_action)); ++ + enum tc_offloaded_state { + TC_OFFLOADED_STATE_UNDEFINED, + TC_OFFLOADED_STATE_IN_HW, +@@ -330,15 +342,10 @@ struct tc_flower { + int action_count; + struct tc_action actions[TCA_ACT_MAX_NUM]; + +- struct ovs_flow_stats stats; ++ struct ovs_flow_stats stats_sw; ++ struct ovs_flow_stats stats_hw; + uint64_t lastused; + +- struct { +- bool rewrite; +- struct tc_flower_key key; +- struct tc_flower_key mask; +- } rewrite; +- + uint32_t csum_update_flags; + + bool tunnel; +@@ -352,13 +359,6 @@ struct tc_flower { + enum tc_offload_policy tc_policy; + }; + +-/* assert that if we overflow with a masked write of uint32_t to the last byte +- * of flower.rewrite we overflow inside struct flower. +- * shouldn't happen unless someone moves rewrite to the end of flower */ +-BUILD_ASSERT_DECL(offsetof(struct tc_flower, rewrite) +- + MEMBER_SIZEOF(struct tc_flower, rewrite) +- + sizeof(uint32_t) - 2 < sizeof(struct tc_flower)); +- + int tc_replace_flower(struct tcf_id *id, struct tc_flower *flower); + int tc_del_filter(struct tcf_id *id); + int tc_get_flower(struct tcf_id *id, struct tc_flower *flower); +diff --git a/lib/tnl-ports.c b/lib/tnl-ports.c +index 58269d3b16..050eafa6b8 100644 +--- a/lib/tnl-ports.c ++++ b/lib/tnl-ports.c +@@ -71,7 +71,7 @@ tnl_port_cast(const struct cls_rule *cr) + { + BUILD_ASSERT_DECL(offsetof(struct tnl_port_in, cr) == 0); + +- return CONTAINER_OF(cr, struct tnl_port_in, cr); ++ return cr ? CONTAINER_OF(cr, struct tnl_port_in, cr) : NULL; + } + + static void +@@ -259,14 +259,14 @@ ipdev_map_delete(struct ip_device *ip_dev, ovs_be16 tp_port, uint8_t nw_proto) + void + tnl_port_map_delete(odp_port_t port, const char type[]) + { +- struct tnl_port *p, *next; ++ struct tnl_port *p; + struct ip_device *ip_dev; + uint8_t nw_proto; + + nw_proto = tnl_type_to_nw_proto(type); + + ovs_mutex_lock(&mutex); +- LIST_FOR_EACH_SAFE(p, next, node, &port_list) { ++ LIST_FOR_EACH_SAFE (p, node, &port_list) { + if (p->port == port && p->nw_proto == nw_proto && + ovs_refcount_unref_relaxed(&p->ref_cnt) == 1) { + ovs_list_remove(&p->node); +@@ -444,11 +444,11 @@ delete_ipdev(struct ip_device *ip_dev) + void + tnl_port_map_insert_ipdev(const char dev_name[]) + { +- struct ip_device *ip_dev, *next; ++ struct ip_device *ip_dev; + + ovs_mutex_lock(&mutex); + +- LIST_FOR_EACH_SAFE(ip_dev, next, node, &addr_list) { ++ LIST_FOR_EACH_SAFE (ip_dev, node, &addr_list) { + if (!strcmp(netdev_get_name(ip_dev->dev), dev_name)) { + if (ip_dev->change_seq == netdev_get_change_seq(ip_dev->dev)) { + goto out; +@@ -466,10 +466,10 @@ out: + void + tnl_port_map_delete_ipdev(const char dev_name[]) + { +- struct ip_device *ip_dev, *next; ++ struct ip_device *ip_dev; + + ovs_mutex_lock(&mutex); +- LIST_FOR_EACH_SAFE(ip_dev, next, node, &addr_list) { ++ LIST_FOR_EACH_SAFE (ip_dev, node, &addr_list) { + if (!strcmp(netdev_get_name(ip_dev->dev), dev_name)) { + delete_ipdev(ip_dev); + } +@@ -480,10 +480,10 @@ tnl_port_map_delete_ipdev(const char dev_name[]) + void + tnl_port_map_run(void) + { +- struct ip_device *ip_dev, *next; ++ struct ip_device *ip_dev; + + ovs_mutex_lock(&mutex); +- LIST_FOR_EACH_SAFE(ip_dev, next, node, &addr_list) { ++ LIST_FOR_EACH_SAFE (ip_dev, node, &addr_list) { + char dev_name[IFNAMSIZ]; + + if (ip_dev->change_seq == netdev_get_change_seq(ip_dev->dev)) { +diff --git a/lib/unixctl.c b/lib/unixctl.c +index 69aed6722c..103357ee91 100644 +--- a/lib/unixctl.c ++++ b/lib/unixctl.c +@@ -390,8 +390,8 @@ unixctl_server_run(struct unixctl_server *server) + } + } + +- struct unixctl_conn *conn, *next; +- LIST_FOR_EACH_SAFE (conn, next, node, &server->conns) { ++ struct unixctl_conn *conn; ++ LIST_FOR_EACH_SAFE (conn, node, &server->conns) { + int error = run_connection(conn); + if (error && error != EAGAIN) { + kill_connection(conn); +@@ -422,9 +422,9 @@ void + unixctl_server_destroy(struct unixctl_server *server) + { + if (server) { +- struct unixctl_conn *conn, *next; ++ struct unixctl_conn *conn; + +- LIST_FOR_EACH_SAFE (conn, next, node, &server->conns) { ++ LIST_FOR_EACH_SAFE (conn, node, &server->conns) { + kill_connection(conn); + } + +diff --git a/lib/vconn.c b/lib/vconn.c +index 7415e6291f..b556762277 100644 +--- a/lib/vconn.c ++++ b/lib/vconn.c +@@ -960,8 +960,8 @@ vconn_transact_multipart(struct vconn *vconn, + ovs_list_init(replies); + + /* Send all the requests. */ +- struct ofpbuf *b, *next; +- LIST_FOR_EACH_SAFE (b, next, list_node, requests) { ++ struct ofpbuf *b; ++ LIST_FOR_EACH_SAFE (b, list_node, requests) { + ovs_list_remove(&b->list_node); + int error = vconn_send_block(vconn, b); + if (error) { +diff --git a/m4/ax_func_posix_memalign.m4 b/m4/ax_func_posix_memalign.m4 +index bd60adcbc8..2442ceca74 100644 +--- a/m4/ax_func_posix_memalign.m4 ++++ b/m4/ax_func_posix_memalign.m4 +@@ -1,5 +1,5 @@ + # =========================================================================== +-# http://www.gnu.org/software/autoconf-archive/ax_func_posix_memalign.html ++# https://www.gnu.org/software/autoconf-archive/ax_func_posix_memalign.html + # =========================================================================== + # + # SYNOPSIS +@@ -22,12 +22,12 @@ + # and this notice are preserved. This file is offered as-is, without any + # warranty. + +-#serial 7 ++#serial 9 + + AC_DEFUN([AX_FUNC_POSIX_MEMALIGN], + [AC_CACHE_CHECK([for working posix_memalign], + [ax_cv_func_posix_memalign_works], +- [AC_TRY_RUN([ ++ [AC_RUN_IFELSE([AC_LANG_SOURCE([[ + #include + + int +@@ -39,7 +39,7 @@ main () + * the size word. */ + exit (posix_memalign (&buffer, sizeof(void *), 123) != 0); + } +- ], ++ ]])], + [ax_cv_func_posix_memalign_works=yes], + [ax_cv_func_posix_memalign_works=no], + [ax_cv_func_posix_memalign_works=no])]) +diff --git a/m4/openvswitch.m4 b/m4/openvswitch.m4 +index 4c3bace6ef..21808483e9 100644 +--- a/m4/openvswitch.m4 ++++ b/m4/openvswitch.m4 +@@ -21,7 +21,7 @@ AC_DEFUN([OVS_CHECK_COVERAGE], + [AC_REQUIRE([AC_PROG_CC]) + AC_ARG_ENABLE( + [coverage], +- [AC_HELP_STRING([--enable-coverage], ++ [AS_HELP_STRING([--enable-coverage], + [Enable gcov coverage tool.])], + [case "${enableval}" in + (yes) coverage=true ;; +@@ -50,7 +50,7 @@ dnl Checks for --enable-ndebug and defines NDEBUG if it is specified. + AC_DEFUN([OVS_CHECK_NDEBUG], + [AC_ARG_ENABLE( + [ndebug], +- [AC_HELP_STRING([--enable-ndebug], ++ [AS_HELP_STRING([--enable-ndebug], + [Disable debugging features for max performance])], + [case "${enableval}" in + (yes) ndebug=true ;; +@@ -64,7 +64,7 @@ dnl Checks for --enable-usdt-probes and defines HAVE_USDT if it is specified. + AC_DEFUN([OVS_CHECK_USDT], [ + AC_ARG_ENABLE( + [usdt-probes], +- [AC_HELP_STRING([--enable-usdt-probes], ++ [AS_HELP_STRING([--enable-usdt-probes], + [Enable User Statically Defined Tracing (USDT) probes])], + [case "${enableval}" in + (yes) usdt=true ;; +@@ -227,7 +227,7 @@ dnl Checks for libcap-ng. + AC_DEFUN([OVS_CHECK_LIBCAPNG], + [AC_ARG_ENABLE( + [libcapng], +- [AC_HELP_STRING([--disable-libcapng], [Disable Linux capability support])], ++ [AS_HELP_STRING([--disable-libcapng], [Disable Linux capability support])], + [case "${enableval}" in + (yes) libcapng=true ;; + (no) libcapng=false ;; +@@ -263,7 +263,7 @@ dnl Checks for OpenSSL. + AC_DEFUN([OVS_CHECK_OPENSSL], + [AC_ARG_ENABLE( + [ssl], +- [AC_HELP_STRING([--disable-ssl], [Disable OpenSSL support])], ++ [AS_HELP_STRING([--disable-ssl], [Disable OpenSSL support])], + [case "${enableval}" in + (yes) ssl=true ;; + (no) ssl=false ;; +@@ -320,7 +320,7 @@ dnl Checks for the directory in which to store the PKI. + AC_DEFUN([OVS_CHECK_PKIDIR], + [AC_ARG_WITH( + [pkidir], +- AC_HELP_STRING([--with-pkidir=DIR], ++ AS_HELP_STRING([--with-pkidir=DIR], + [PKI hierarchy directory [[LOCALSTATEDIR/lib/openvswitch/pki]]]), + [PKIDIR=$withval], + [PKIDIR='${localstatedir}/lib/openvswitch/pki']) +@@ -330,7 +330,7 @@ dnl Checks for the directory in which to store pidfiles. + AC_DEFUN([OVS_CHECK_RUNDIR], + [AC_ARG_WITH( + [rundir], +- AC_HELP_STRING([--with-rundir=DIR], ++ AS_HELP_STRING([--with-rundir=DIR], + [directory used for pidfiles + [[LOCALSTATEDIR/run/openvswitch]]]), + [RUNDIR=$withval], +@@ -341,7 +341,7 @@ dnl Checks for the directory in which to store logs. + AC_DEFUN([OVS_CHECK_LOGDIR], + [AC_ARG_WITH( + [logdir], +- AC_HELP_STRING([--with-logdir=DIR], ++ AS_HELP_STRING([--with-logdir=DIR], + [directory used for logs [[LOCALSTATEDIR/log/PACKAGE]]]), + [LOGDIR=$withval], + [LOGDIR='${localstatedir}/log/${PACKAGE}']) +@@ -351,7 +351,7 @@ dnl Checks for the directory in which to store the Open vSwitch database. + AC_DEFUN([OVS_CHECK_DBDIR], + [AC_ARG_WITH( + [dbdir], +- AC_HELP_STRING([--with-dbdir=DIR], ++ AS_HELP_STRING([--with-dbdir=DIR], + [directory used for conf.db [[SYSCONFDIR/PACKAGE]]]), + [DBDIR=$withval], + [DBDIR='${sysconfdir}/${PACKAGE}']) +diff --git a/ofproto/bond.c b/ofproto/bond.c +index cdfdf0b9d8..845f69e21d 100644 +--- a/ofproto/bond.c ++++ b/ofproto/bond.c +@@ -338,7 +338,7 @@ static void + update_recirc_rules__(struct bond *bond) + { + struct match match; +- struct bond_pr_rule_op *pr_op, *next_op; ++ struct bond_pr_rule_op *pr_op; + uint64_t ofpacts_stub[128 / 8]; + struct ofpbuf ofpacts; + int i; +@@ -372,7 +372,7 @@ update_recirc_rules__(struct bond *bond) + + ofpbuf_use_stub(&ofpacts, ofpacts_stub, sizeof ofpacts_stub); + +- HMAP_FOR_EACH_SAFE(pr_op, next_op, hmap_node, &bond->pr_rule_ops) { ++ HMAP_FOR_EACH_SAFE (pr_op, hmap_node, &bond->pr_rule_ops) { + int error; + switch (pr_op->op) { + case ADD: +@@ -1258,7 +1258,7 @@ insert_bal(struct ovs_list *bals, struct bond_member *member) + break; + } + } +- ovs_list_insert(&pos->bal_node, &member->bal_node); ++ ovs_list_insert(pos ? &pos->bal_node : bals, &member->bal_node); + } + + /* Removes 'member' from its current list and then inserts it into 'bals' so +diff --git a/ofproto/connmgr.c b/ofproto/connmgr.c +index fa8f6cd0e8..172a58cfb7 100644 +--- a/ofproto/connmgr.c ++++ b/ofproto/connmgr.c +@@ -310,8 +310,8 @@ connmgr_destroy(struct connmgr *mgr) + return; + } + +- struct ofservice *ofservice, *next_ofservice; +- HMAP_FOR_EACH_SAFE (ofservice, next_ofservice, hmap_node, &mgr->services) { ++ struct ofservice *ofservice; ++ HMAP_FOR_EACH_SAFE (ofservice, hmap_node, &mgr->services) { + ofservice_destroy(ofservice); + } + hmap_destroy(&mgr->services); +@@ -351,8 +351,8 @@ connmgr_run(struct connmgr *mgr, + } + } + +- struct ofconn *ofconn, *next_ofconn; +- LIST_FOR_EACH_SAFE (ofconn, next_ofconn, connmgr_node, &mgr->conns) { ++ struct ofconn *ofconn; ++ LIST_FOR_EACH_SAFE (ofconn, connmgr_node, &mgr->conns) { + ofconn_run(ofconn, handle_openflow); + } + ofmonitor_run(mgr); +@@ -592,8 +592,8 @@ connmgr_set_controllers(struct connmgr *mgr, struct shash *controllers) + + /* Delete services that are no longer configured. + * Update configuration of all now-existing services. */ +- struct ofservice *ofservice, *next_ofservice; +- HMAP_FOR_EACH_SAFE (ofservice, next_ofservice, hmap_node, &mgr->services) { ++ struct ofservice *ofservice; ++ HMAP_FOR_EACH_SAFE (ofservice, hmap_node, &mgr->services) { + const char *target = ofservice->target; + struct ofproto_controller *c = shash_find_data(controllers, target); + if (!c) { +@@ -1137,9 +1137,9 @@ ofconn_remove_bundle(struct ofconn *ofconn, struct ofp_bundle *bundle) + static void + bundle_remove_all(struct ofconn *ofconn) + { +- struct ofp_bundle *b, *next; ++ struct ofp_bundle *b; + +- HMAP_FOR_EACH_SAFE (b, next, node, &ofconn->bundles) { ++ HMAP_FOR_EACH_SAFE (b, node, &ofconn->bundles) { + ofp_bundle_remove__(ofconn, b); + } + } +@@ -1149,8 +1149,8 @@ bundle_remove_expired(struct ofconn *ofconn, long long int now) + { + long long int limit = now - bundle_idle_timeout; + +- struct ofp_bundle *b, *next; +- HMAP_FOR_EACH_SAFE (b, next, node, &ofconn->bundles) { ++ struct ofp_bundle *b; ++ HMAP_FOR_EACH_SAFE (b, node, &ofconn->bundles) { + if (b->used <= limit) { + ofconn_send_error(ofconn, b->msg, OFPERR_OFPBFC_TIMEOUT); + ofp_bundle_remove__(ofconn, b); +@@ -1247,8 +1247,8 @@ ofconn_destroy(struct ofconn *ofconn) + + free(ofconn->async_cfg); + +- struct ofmonitor *monitor, *next_monitor; +- HMAP_FOR_EACH_SAFE (monitor, next_monitor, ofconn_node, ++ struct ofmonitor *monitor; ++ HMAP_FOR_EACH_SAFE (monitor, ofconn_node, + &ofconn->monitors) { + ofmonitor_destroy(monitor); + } +@@ -1953,8 +1953,8 @@ static void + ofservice_close_all(struct ofservice *ofservice) + OVS_REQUIRES(ofproto_mutex) + { +- struct ofconn *ofconn, *next; +- LIST_FOR_EACH_SAFE (ofconn, next, ofservice_node, &ofservice->conns) { ++ struct ofconn *ofconn; ++ LIST_FOR_EACH_SAFE (ofconn, ofservice_node, &ofservice->conns) { + ofconn_destroy(ofconn); + } + } +diff --git a/ofproto/in-band.c b/ofproto/in-band.c +index 82d8dfa147..3992251f5f 100644 +--- a/ofproto/in-band.c ++++ b/ofproto/in-band.c +@@ -377,7 +377,7 @@ in_band_run(struct in_band *ib) + uint64_t ofpacts_stub[128 / 8]; + struct ofpbuf ofpacts; + +- struct in_band_rule *rule, *next; ++ struct in_band_rule *rule; + + ofpbuf_use_stub(&ofpacts, ofpacts_stub, sizeof ofpacts_stub); + +@@ -391,7 +391,7 @@ in_band_run(struct in_band *ib) + + update_rules(ib); + +- HMAP_FOR_EACH_SAFE (rule, next, hmap_node, &ib->rules) { ++ HMAP_FOR_EACH_SAFE (rule, hmap_node, &ib->rules) { + switch (rule->op) { + case ADD: + ofproto_add_flow(ib->ofproto, &rule->match, rule->priority, +diff --git a/ofproto/libofproto.pc.in b/ofproto/libofproto.pc.in +index 2740712505..49894fb695 100644 +--- a/ofproto/libofproto.pc.in ++++ b/ofproto/libofproto.pc.in +@@ -8,4 +8,4 @@ Description: OpenFlow library of Open vSwitch + Version: @VERSION@ + Libs: -L${libdir} -lofproto + Libs.private: @LIBS@ +-Cflags: -I${includedir}/openvswitch ++Cflags: -I${includedir} +diff --git a/ofproto/netflow.c b/ofproto/netflow.c +index ed58de17de..aad9f9c77a 100644 +--- a/ofproto/netflow.c ++++ b/ofproto/netflow.c +@@ -299,7 +299,7 @@ static void + netflow_run__(struct netflow *nf) OVS_REQUIRES(mutex) + { + long long int now = time_msec(); +- struct netflow_flow *nf_flow, *next; ++ struct netflow_flow *nf_flow; + + if (nf->packet.size) { + collectors_send(nf->collectors, nf->packet.data, nf->packet.size); +@@ -312,7 +312,7 @@ netflow_run__(struct netflow *nf) OVS_REQUIRES(mutex) + + nf->next_timeout = now + 1000; + +- HMAP_FOR_EACH_SAFE (nf_flow, next, hmap_node, &nf->flows) { ++ HMAP_FOR_EACH_SAFE (nf_flow, hmap_node, &nf->flows) { + if (now > nf_flow->last_expired + nf->active_timeout) { + bool idle = nf_flow->used < nf_flow->last_expired; + netflow_expire__(nf, nf_flow); +@@ -416,8 +416,8 @@ netflow_unref(struct netflow *nf) + collectors_destroy(nf->collectors); + ofpbuf_uninit(&nf->packet); + +- struct netflow_flow *nf_flow, *next; +- HMAP_FOR_EACH_SAFE (nf_flow, next, hmap_node, &nf->flows) { ++ struct netflow_flow *nf_flow; ++ HMAP_FOR_EACH_SAFE (nf_flow, hmap_node, &nf->flows) { + hmap_remove(&nf->flows, &nf_flow->hmap_node); + free(nf_flow); + } +diff --git a/ofproto/ofproto-dpif-ipfix.c b/ofproto/ofproto-dpif-ipfix.c +index 9280e008ea..742eed3998 100644 +--- a/ofproto/ofproto-dpif-ipfix.c ++++ b/ofproto/ofproto-dpif-ipfix.c +@@ -926,17 +926,21 @@ dpif_ipfix_bridge_exporter_destroy(struct dpif_ipfix_bridge_exporter *exporter) + static void + dpif_ipfix_bridge_exporter_set_options( + struct dpif_ipfix_bridge_exporter *exporter, +- const struct ofproto_ipfix_bridge_exporter_options *options) ++ const struct ofproto_ipfix_bridge_exporter_options *options, ++ bool *options_changed) + { +- bool options_changed; +- + if (!options || sset_is_empty(&options->targets)) { + /* No point in doing any work if there are no targets. */ +- dpif_ipfix_bridge_exporter_clear(exporter); ++ if (exporter->options) { ++ dpif_ipfix_bridge_exporter_clear(exporter); ++ *options_changed = true; ++ } else { ++ *options_changed = false; ++ } + return; + } + +- options_changed = ( ++ *options_changed = ( + !exporter->options + || !ofproto_ipfix_bridge_exporter_options_equal( + options, exporter->options)); +@@ -945,7 +949,7 @@ dpif_ipfix_bridge_exporter_set_options( + * shortchanged in collectors (which indicates that opening one or + * more of the configured collectors failed, so that we should + * retry). */ +- if (options_changed ++ if (*options_changed + || collectors_count(exporter->exporter.collectors) + < sset_count(&options->targets)) { + if (!dpif_ipfix_exporter_set_options( +@@ -957,7 +961,7 @@ dpif_ipfix_bridge_exporter_set_options( + } + + /* Avoid reconfiguring if options didn't change. */ +- if (!options_changed) { ++ if (!*options_changed) { + return; + } + +@@ -1015,17 +1019,21 @@ dpif_ipfix_flow_exporter_destroy(struct dpif_ipfix_flow_exporter *exporter) + static bool + dpif_ipfix_flow_exporter_set_options( + struct dpif_ipfix_flow_exporter *exporter, +- const struct ofproto_ipfix_flow_exporter_options *options) ++ const struct ofproto_ipfix_flow_exporter_options *options, ++ bool *options_changed) + { +- bool options_changed; +- + if (sset_is_empty(&options->targets)) { + /* No point in doing any work if there are no targets. */ +- dpif_ipfix_flow_exporter_clear(exporter); ++ if (exporter->options) { ++ dpif_ipfix_flow_exporter_clear(exporter); ++ *options_changed = true; ++ } else { ++ *options_changed = false; ++ } + return true; + } + +- options_changed = ( ++ *options_changed = ( + !exporter->options + || !ofproto_ipfix_flow_exporter_options_equal( + options, exporter->options)); +@@ -1034,7 +1042,7 @@ dpif_ipfix_flow_exporter_set_options( + * shortchanged in collectors (which indicates that opening one or + * more of the configured collectors failed, so that we should + * retry). */ +- if (options_changed ++ if (*options_changed + || collectors_count(exporter->exporter.collectors) + < sset_count(&options->targets)) { + if (!dpif_ipfix_exporter_set_options( +@@ -1046,7 +1054,7 @@ dpif_ipfix_flow_exporter_set_options( + } + + /* Avoid reconfiguring if options didn't change. */ +- if (!options_changed) { ++ if (!*options_changed) { + return true; + } + +@@ -1069,7 +1077,7 @@ remove_flow_exporter(struct dpif_ipfix *di, + free(node); + } + +-void ++bool + dpif_ipfix_set_options( + struct dpif_ipfix *di, + const struct ofproto_ipfix_bridge_exporter_options *bridge_exporter_options, +@@ -1077,16 +1085,19 @@ dpif_ipfix_set_options( + size_t n_flow_exporters_options) OVS_EXCLUDED(mutex) + { + int i; ++ bool beo_changed, feo_changed, entry_changed; + struct ofproto_ipfix_flow_exporter_options *options; +- struct dpif_ipfix_flow_exporter_map_node *node, *next; ++ struct dpif_ipfix_flow_exporter_map_node *node; + + ovs_mutex_lock(&mutex); + dpif_ipfix_bridge_exporter_set_options(&di->bridge_exporter, +- bridge_exporter_options); ++ bridge_exporter_options, ++ &beo_changed); + + /* Add new flow exporters and update current flow exporters. */ + options = (struct ofproto_ipfix_flow_exporter_options *) + flow_exporters_options; ++ feo_changed = false; + for (i = 0; i < n_flow_exporters_options; i++) { + node = dpif_ipfix_find_flow_exporter_map_node( + di, options->collector_set_id); +@@ -1095,15 +1106,19 @@ dpif_ipfix_set_options( + dpif_ipfix_flow_exporter_init(&node->exporter); + hmap_insert(&di->flow_exporter_map, &node->node, + hash_int(options->collector_set_id, 0)); ++ feo_changed = true; + } +- if (!dpif_ipfix_flow_exporter_set_options(&node->exporter, options)) { ++ if (!dpif_ipfix_flow_exporter_set_options(&node->exporter, ++ options, ++ &entry_changed)) { + remove_flow_exporter(di, node); + } ++ feo_changed = entry_changed ? true : feo_changed; + options++; + } + + /* Remove dropped flow exporters, if any needs to be removed. */ +- HMAP_FOR_EACH_SAFE (node, next, node, &di->flow_exporter_map) { ++ HMAP_FOR_EACH_SAFE (node, node, &di->flow_exporter_map) { + /* This is slow but doesn't take any extra memory, and + * this table is not supposed to contain many rows anyway. */ + options = (struct ofproto_ipfix_flow_exporter_options *) +@@ -1117,10 +1132,12 @@ dpif_ipfix_set_options( + } + if (i == n_flow_exporters_options) { /* Not found. */ + remove_flow_exporter(di, node); ++ feo_changed = true; + } + } + + ovs_mutex_unlock(&mutex); ++ return beo_changed || feo_changed; + } + + struct dpif_ipfix * +@@ -1215,7 +1232,7 @@ static void + dpif_ipfix_clear(struct dpif_ipfix *di) OVS_REQUIRES(mutex) + { + struct dpif_ipfix_flow_exporter_map_node *exp_node; +- struct dpif_ipfix_port *dip, *next; ++ struct dpif_ipfix_port *dip; + + dpif_ipfix_bridge_exporter_clear(&di->bridge_exporter); + +@@ -1224,7 +1241,7 @@ dpif_ipfix_clear(struct dpif_ipfix *di) OVS_REQUIRES(mutex) + free(exp_node); + } + +- HMAP_FOR_EACH_SAFE (dip, next, hmap_node, &di->ports) { ++ HMAP_FOR_EACH_SAFE (dip, hmap_node, &di->ports) { + dpif_ipfix_del_port__(di, dip); + } + } +@@ -2799,7 +2816,7 @@ dpif_ipfix_cache_expire(struct dpif_ipfix_exporter *exporter, + bool forced_end, const uint64_t export_time_usec, + const uint32_t export_time_sec) + { +- struct ipfix_flow_cache_entry *entry, *next_entry; ++ struct ipfix_flow_cache_entry *entry; + uint64_t max_flow_start_timestamp_usec; + bool template_msg_sent = false; + enum ipfix_flow_end_reason flow_end_reason; +@@ -2811,7 +2828,7 @@ dpif_ipfix_cache_expire(struct dpif_ipfix_exporter *exporter, + max_flow_start_timestamp_usec = export_time_usec - + 1000000LL * exporter->cache_active_timeout; + +- LIST_FOR_EACH_SAFE (entry, next_entry, cache_flow_start_timestamp_list_node, ++ LIST_FOR_EACH_SAFE (entry, cache_flow_start_timestamp_list_node, + &exporter->cache_flow_start_timestamp_list) { + if (forced_end) { + flow_end_reason = FORCED_END; +diff --git a/ofproto/ofproto-dpif-ipfix.h b/ofproto/ofproto-dpif-ipfix.h +index 1f42cd5275..75c0ab81ac 100644 +--- a/ofproto/ofproto-dpif-ipfix.h ++++ b/ofproto/ofproto-dpif-ipfix.h +@@ -48,7 +48,7 @@ bool dpif_ipfix_get_bridge_exporter_output_sampling(const struct dpif_ipfix *); + bool dpif_ipfix_get_flow_exporter_tunnel_sampling(const struct dpif_ipfix *, + const uint32_t); + bool dpif_ipfix_is_tunnel_port(const struct dpif_ipfix *, odp_port_t); +-void dpif_ipfix_set_options( ++bool dpif_ipfix_set_options( + struct dpif_ipfix *, + const struct ofproto_ipfix_bridge_exporter_options *, + const struct ofproto_ipfix_flow_exporter_options *, size_t); +diff --git a/ofproto/ofproto-dpif-sflow.c b/ofproto/ofproto-dpif-sflow.c +index 30e7caf54a..e8e1de920b 100644 +--- a/ofproto/ofproto-dpif-sflow.c ++++ b/ofproto/ofproto-dpif-sflow.c +@@ -468,7 +468,8 @@ sflow_choose_agent_address(const char *agent_device, + const char *target; + SSET_FOR_EACH (target, targets) { + struct sockaddr_storage ss; +- if (inet_parse_active(target, SFL_DEFAULT_COLLECTOR_PORT, &ss, true)) { ++ if (inet_parse_active(target, SFL_DEFAULT_COLLECTOR_PORT, ++ &ss, true, NULL)) { + /* sFlow only supports target in default routing table with + * packet mark zero. + */ +@@ -590,10 +591,10 @@ void + dpif_sflow_unref(struct dpif_sflow *ds) OVS_EXCLUDED(mutex) + { + if (ds && ovs_refcount_unref_relaxed(&ds->ref_cnt) == 1) { +- struct dpif_sflow_port *dsp, *next; ++ struct dpif_sflow_port *dsp; + + dpif_sflow_clear(ds); +- HMAP_FOR_EACH_SAFE (dsp, next, hmap_node, &ds->ports) { ++ HMAP_FOR_EACH_SAFE (dsp, hmap_node, &ds->ports) { + dpif_sflow_del_port__(ds, dsp); + } + hmap_destroy(&ds->ports); +diff --git a/ofproto/ofproto-dpif-trace.c b/ofproto/ofproto-dpif-trace.c +index 78a54c715d..109940ad2a 100644 +--- a/ofproto/ofproto-dpif-trace.c ++++ b/ofproto/ofproto-dpif-trace.c +@@ -65,8 +65,8 @@ static void + oftrace_node_list_destroy(struct ovs_list *nodes) + { + if (nodes) { +- struct oftrace_node *node, *next; +- LIST_FOR_EACH_SAFE (node, next, node, nodes) { ++ struct oftrace_node *node; ++ LIST_FOR_EACH_SAFE (node, node, nodes) { + ovs_list_remove(&node->node); + oftrace_node_destroy(node); + } +diff --git a/ofproto/ofproto-dpif-xlate-cache.c b/ofproto/ofproto-dpif-xlate-cache.c +index dcc91cb380..9224ee2e6d 100644 +--- a/ofproto/ofproto-dpif-xlate-cache.c ++++ b/ofproto/ofproto-dpif-xlate-cache.c +@@ -209,6 +209,7 @@ xlate_cache_clear_entry(struct xc_entry *entry) + { + switch (entry->type) { + case XC_TABLE: ++ ofproto_unref(&(entry->table.ofproto->up)); + break; + case XC_RULE: + ofproto_rule_unref(&entry->rule->up); +@@ -231,6 +232,7 @@ xlate_cache_clear_entry(struct xc_entry *entry) + free(entry->learn.ofm); + break; + case XC_NORMAL: ++ ofproto_unref(&(entry->normal.ofproto->up)); + break; + case XC_FIN_TIMEOUT: + /* 'u.fin.rule' is always already held as a XC_RULE, which +diff --git a/ofproto/ofproto-dpif-xlate-cache.h b/ofproto/ofproto-dpif-xlate-cache.h +index 114aff8ea3..0fc6d2ea60 100644 +--- a/ofproto/ofproto-dpif-xlate-cache.h ++++ b/ofproto/ofproto-dpif-xlate-cache.h +@@ -61,9 +61,8 @@ enum xc_type { + * that a flow relates to, although they may be used for other effects as well + * (for instance, refreshing hard timeouts for learned flows). + * +- * An explicit reference is taken to all pointers other than the ones for +- * struct ofproto_dpif. ofproto_dpif pointers are explicitly protected by +- * destroying all xlate caches before the ofproto is destroyed. */ ++ * An explicit reference is taken to all pointers. ++ */ + struct xc_entry { + enum xc_type type; + union { +diff --git a/ofproto/ofproto-dpif-xlate.c b/ofproto/ofproto-dpif-xlate.c +index 578cbfe581..a13cf7bc65 100644 +--- a/ofproto/ofproto-dpif-xlate.c ++++ b/ofproto/ofproto-dpif-xlate.c +@@ -865,7 +865,7 @@ xlate_xbridge_init(struct xlate_cfg *xcfg, struct xbridge *xbridge) + ovs_list_init(&xbridge->xbundles); + hmap_init(&xbridge->xports); + hmap_insert(&xcfg->xbridges, &xbridge->hmap_node, +- hash_pointer(xbridge->ofproto, 0)); ++ uuid_hash(&xbridge->ofproto->uuid)); + } + + static void +@@ -1222,13 +1222,13 @@ xlate_txn_start(void) + static void + xlate_xcfg_free(struct xlate_cfg *xcfg) + { +- struct xbridge *xbridge, *next_xbridge; ++ struct xbridge *xbridge; + + if (!xcfg) { + return; + } + +- HMAP_FOR_EACH_SAFE (xbridge, next_xbridge, hmap_node, &xcfg->xbridges) { ++ HMAP_FOR_EACH_SAFE (xbridge, hmap_node, &xcfg->xbridges) { + xlate_xbridge_remove(xcfg, xbridge); + } + +@@ -1282,18 +1282,18 @@ xlate_ofproto_set(struct ofproto_dpif *ofproto, const char *name, + static void + xlate_xbridge_remove(struct xlate_cfg *xcfg, struct xbridge *xbridge) + { +- struct xbundle *xbundle, *next_xbundle; +- struct xport *xport, *next_xport; ++ struct xbundle *xbundle; ++ struct xport *xport; + + if (!xbridge) { + return; + } + +- HMAP_FOR_EACH_SAFE (xport, next_xport, ofp_node, &xbridge->xports) { ++ HMAP_FOR_EACH_SAFE (xport, ofp_node, &xbridge->xports) { + xlate_xport_remove(xcfg, xport); + } + +- LIST_FOR_EACH_SAFE (xbundle, next_xbundle, list_node, &xbridge->xbundles) { ++ LIST_FOR_EACH_SAFE (xbundle, list_node, &xbridge->xbundles) { + xlate_xbundle_remove(xcfg, xbundle); + } + +@@ -1515,7 +1515,7 @@ xlate_lookup_ofproto_(const struct dpif_backer *backer, + if (OVS_UNLIKELY(!recirc_id_node)) { + if (errorp) { + *errorp = xasprintf("no recirculation data for recirc_id " +- "%"PRIu32, flow->recirc_id); ++ "%#"PRIx32, flow->recirc_id); + } + return NULL; + } +@@ -1556,8 +1556,8 @@ xlate_lookup_ofproto_(const struct dpif_backer *backer, + if (errorp) { + *errorp = (tnl_port_should_receive(flow) + ? xstrdup("no OpenFlow tunnel port for this packet") +- : xasprintf("no OpenFlow tunnel port for datapath " +- "port %"PRIu32, flow->in_port.odp_port)); ++ : xasprintf("no OpenFlow port for datapath port " ++ "%"PRIu32, flow->in_port.odp_port)); + } + return NULL; + } +@@ -1639,7 +1639,7 @@ xbridge_lookup(struct xlate_cfg *xcfg, const struct ofproto_dpif *ofproto) + + xbridges = &xcfg->xbridges; + +- HMAP_FOR_EACH_IN_BUCKET (xbridge, hmap_node, hash_pointer(ofproto, 0), ++ HMAP_FOR_EACH_IN_BUCKET (xbridge, hmap_node, uuid_hash(&ofproto->uuid), + xbridges) { + if (xbridge->ofproto == ofproto) { + return xbridge; +@@ -1661,6 +1661,23 @@ xbridge_lookup_by_uuid(struct xlate_cfg *xcfg, const struct uuid *uuid) + return NULL; + } + ++struct ofproto_dpif * ++xlate_ofproto_lookup(const struct uuid *uuid) ++{ ++ struct xlate_cfg *xcfg = ovsrcu_get(struct xlate_cfg *, &xcfgp); ++ struct xbridge *xbridge; ++ ++ if (!xcfg) { ++ return NULL; ++ } ++ ++ xbridge = xbridge_lookup_by_uuid(xcfg, uuid); ++ if (xbridge != NULL) { ++ return xbridge->ofproto; ++ } ++ return NULL; ++} ++ + static struct xbundle * + xbundle_lookup(struct xlate_cfg *xcfg, const struct ofbundle *ofbundle) + { +@@ -2125,9 +2142,14 @@ mirror_packet(struct xlate_ctx *ctx, struct xbundle *xbundle, + int snaplen; + + /* Get the details of the mirror represented by the rightmost 1-bit. */ +- ovs_assert(mirror_get(xbridge->mbridge, raw_ctz(mirrors), +- &vlans, &dup_mirrors, +- &out, &snaplen, &out_vlan)); ++ if (OVS_UNLIKELY(!mirror_get(xbridge->mbridge, raw_ctz(mirrors), ++ &vlans, &dup_mirrors, ++ &out, &snaplen, &out_vlan))) { ++ /* The mirror got reconfigured before we got to read it's ++ * configuration. */ ++ mirrors = zero_rightmost_1bit(mirrors); ++ continue; ++ } + + + /* If this mirror selects on the basis of VLAN, and it does not select +@@ -3015,7 +3037,7 @@ xlate_normal(struct xlate_ctx *ctx) + bool is_grat_arp = is_gratuitous_arp(flow, wc); + if (ctx->xin->allow_side_effects + && flow->packet_type == htonl(PT_ETH) +- && in_port->pt_mode != NETDEV_PT_LEGACY_L3 ++ && in_port && in_port->pt_mode != NETDEV_PT_LEGACY_L3 + ) { + update_learning_table(ctx, in_xbundle, flow->dl_src, vlan, + is_grat_arp); +@@ -3024,12 +3046,14 @@ xlate_normal(struct xlate_ctx *ctx) + struct xc_entry *entry; + + /* Save just enough info to update mac learning table later. */ +- entry = xlate_cache_add_entry(ctx->xin->xcache, XC_NORMAL); +- entry->normal.ofproto = ctx->xbridge->ofproto; +- entry->normal.in_port = flow->in_port.ofp_port; +- entry->normal.dl_src = flow->dl_src; +- entry->normal.vlan = vlan; +- entry->normal.is_gratuitous_arp = is_grat_arp; ++ if (ofproto_try_ref(&ctx->xbridge->ofproto->up)) { ++ entry = xlate_cache_add_entry(ctx->xin->xcache, XC_NORMAL); ++ entry->normal.ofproto = ctx->xbridge->ofproto; ++ entry->normal.in_port = flow->in_port.ofp_port; ++ entry->normal.dl_src = flow->dl_src; ++ entry->normal.vlan = vlan; ++ entry->normal.is_gratuitous_arp = is_grat_arp; ++ } + } + + /* Determine output bundle. */ +@@ -3048,7 +3072,6 @@ xlate_normal(struct xlate_ctx *ctx) + */ + ctx->xout->slow |= SLOW_ACTION; + +- memset(&wc->masks.tp_src, 0xff, sizeof wc->masks.tp_src); + if (mcast_snooping_is_membership(flow->tp_src) || + mcast_snooping_is_query(flow->tp_src)) { + if (ctx->xin->allow_side_effects && ctx->xin->packet) { +@@ -3523,6 +3546,9 @@ propagate_tunnel_data_to_flow__(struct flow *dst_flow, + dst_flow->dl_dst = dmac; + dst_flow->dl_src = smac; + ++ /* Clear VLAN entries which do not apply for tunnel flows. */ ++ memset(dst_flow->vlans, 0, sizeof dst_flow->vlans); ++ + dst_flow->packet_type = htonl(PT_ETH); + dst_flow->nw_dst = src_flow->tunnel.ip_dst; + dst_flow->nw_src = src_flow->tunnel.ip_src; +@@ -3654,14 +3680,27 @@ native_tunnel_output(struct xlate_ctx *ctx, const struct xport *xport, + + err = tnl_neigh_lookup(out_dev->xbridge->name, &d_ip6, &dmac); + if (err) { ++ struct in6_addr nh_s_ip6 = in6addr_any; ++ + xlate_report(ctx, OFT_DETAIL, + "neighbor cache miss for %s on bridge %s, " + "sending %s request", + buf_dip6, out_dev->xbridge->name, d_ip ? "ARP" : "ND"); ++ ++ err = ovs_router_get_netdev_source_address(&d_ip6, ++ out_dev->xbridge->name, ++ &nh_s_ip6); ++ if (err) { ++ nh_s_ip6 = s_ip6; ++ } ++ + if (d_ip) { +- tnl_send_arp_request(ctx, out_dev, smac, s_ip, d_ip); ++ ovs_be32 nh_s_ip; ++ ++ nh_s_ip = in6_addr_get_mapped_ipv4(&nh_s_ip6); ++ tnl_send_arp_request(ctx, out_dev, smac, nh_s_ip, d_ip); + } else { +- tnl_send_nd_request(ctx, out_dev, smac, &s_ip6, &d_ip6); ++ tnl_send_nd_request(ctx, out_dev, smac, &nh_s_ip6, &d_ip6); + } + return err; + } +@@ -4176,6 +4215,10 @@ compose_output_action__(struct xlate_ctx *ctx, ofp_port_t ofp_port, + if (xport->pt_mode == NETDEV_PT_LEGACY_L3) { + flow->packet_type = PACKET_TYPE_BE(OFPHTN_ETHERTYPE, + ntohs(flow->dl_type)); ++ if (ctx->pending_encap) { ++ /* The Ethernet header was not actually added yet. */ ++ ctx->pending_encap = false; ++ } + } + } + +@@ -5622,7 +5665,8 @@ xlate_sample_action(struct xlate_ctx *ctx, + + /* Scale the probability from 16-bit to 32-bit while representing + * the same percentage. */ +- uint32_t probability = (os->probability << 16) | os->probability; ++ uint32_t probability = ++ ((uint32_t) os->probability << 16) | os->probability; + + /* If ofp_port in flow sample action is equel to ofp_port, + * this sample action is a input port action. */ +@@ -7609,6 +7653,10 @@ xlate_wc_finish(struct xlate_ctx *ctx) + ctx->wc->masks.vlans[i].tci = 0; + } + } ++ /* Clear tunnel wc bits if original packet is non-tunnel. */ ++ if (!flow_tnl_dst_is_set(&ctx->xin->upcall_flow->tunnel)) { ++ memset(&ctx->wc->masks.tunnel, 0, sizeof ctx->wc->masks.tunnel); ++ } + } + + /* Translates the flow, actions, or rule in 'xin' into datapath actions in +@@ -7784,6 +7832,12 @@ xlate_actions(struct xlate_in *xin, struct xlate_out *xout) + goto exit; + } + ++ if (!xin->frozen_state ++ && xin->flow.ct_state ++ && xin->flow.ct_state & CS_TRACKED) { ++ ctx.conntracked = true; ++ } ++ + /* Tunnel metadata in udpif format must be normalized before translation. */ + if (flow->tunnel.flags & FLOW_TNL_F_UDPIF) { + const struct tun_table *tun_tab = ofproto_get_tun_tab( +diff --git a/ofproto/ofproto-dpif-xlate.h b/ofproto/ofproto-dpif-xlate.h +index 851088d794..2ba90e999c 100644 +--- a/ofproto/ofproto-dpif-xlate.h ++++ b/ofproto/ofproto-dpif-xlate.h +@@ -176,6 +176,7 @@ void xlate_ofproto_set(struct ofproto_dpif *, const char *name, struct dpif *, + bool forward_bpdu, bool has_in_band, + const struct dpif_backer_support *support); + void xlate_remove_ofproto(struct ofproto_dpif *); ++struct ofproto_dpif *xlate_ofproto_lookup(const struct uuid *uuid); + + void xlate_bundle_set(struct ofproto_dpif *, struct ofbundle *, + const char *name, enum port_vlan_mode, +diff --git a/ofproto/ofproto-dpif.c b/ofproto/ofproto-dpif.c +index 8143dd965f..f9562dee87 100644 +--- a/ofproto/ofproto-dpif.c ++++ b/ofproto/ofproto-dpif.c +@@ -215,10 +215,6 @@ struct shash all_dpif_backers = SHASH_INITIALIZER(&all_dpif_backers); + static struct hmap all_ofproto_dpifs_by_name = + HMAP_INITIALIZER(&all_ofproto_dpifs_by_name); + +-/* All existing ofproto_dpif instances, indexed by ->uuid. */ +-static struct hmap all_ofproto_dpifs_by_uuid = +- HMAP_INITIALIZER(&all_ofproto_dpifs_by_uuid); +- + static bool ofproto_use_tnl_push_pop = true; + static void ofproto_unixctl_init(void); + static void ct_zone_config_init(struct dpif_backer *backer); +@@ -1663,7 +1659,7 @@ static int + construct(struct ofproto *ofproto_) + { + struct ofproto_dpif *ofproto = ofproto_dpif_cast(ofproto_); +- struct shash_node *node, *next; ++ struct shash_node *node; + int error; + + /* Tunnel module can get used right after the udpif threads are running. */ +@@ -1701,7 +1697,7 @@ construct(struct ofproto *ofproto_) + ofproto->ams_seqno = seq_read(ofproto->ams_seq); + + +- SHASH_FOR_EACH_SAFE (node, next, &init_ofp_ports) { ++ SHASH_FOR_EACH_SAFE (node, &init_ofp_ports) { + struct iface_hint *iface_hint = node->data; + + if (!strcmp(iface_hint->br_name, ofproto->up.name)) { +@@ -1720,9 +1716,6 @@ construct(struct ofproto *ofproto_) + hmap_insert(&all_ofproto_dpifs_by_name, + &ofproto->all_ofproto_dpifs_by_name_node, + hash_string(ofproto->up.name, 0)); +- hmap_insert(&all_ofproto_dpifs_by_uuid, +- &ofproto->all_ofproto_dpifs_by_uuid_node, +- uuid_hash(&ofproto->uuid)); + memset(&ofproto->stats, 0, sizeof ofproto->stats); + + ofproto_init_tables(ofproto_, N_TABLES); +@@ -1820,8 +1813,6 @@ destruct(struct ofproto *ofproto_, bool del) + + hmap_remove(&all_ofproto_dpifs_by_name, + &ofproto->all_ofproto_dpifs_by_name_node); +- hmap_remove(&all_ofproto_dpifs_by_uuid, +- &ofproto->all_ofproto_dpifs_by_uuid_node); + + OFPROTO_FOR_EACH_TABLE (table, &ofproto->up) { + CLS_FOR_EACH (rule, up.cr, &table->cls) { +@@ -1857,6 +1848,8 @@ destruct(struct ofproto *ofproto_, bool del) + + seq_destroy(ofproto->ams_seq); + ++ /* Wait for all the meter destroy work to finish. */ ++ ovsrcu_barrier(); + close_dpif_backer(ofproto->backer, del); + } + +@@ -1945,7 +1938,7 @@ run(struct ofproto *ofproto_) + + new_dump_seq = seq_read(udpif_dump_seq(ofproto->backer->udpif)); + if (ofproto->dump_seq != new_dump_seq) { +- struct rule *rule, *next_rule; ++ struct rule *rule; + long long now = time_msec(); + + /* We know stats are relatively fresh, so now is a good time to do some +@@ -1955,7 +1948,7 @@ run(struct ofproto *ofproto_) + /* Expire OpenFlow flows whose idle_timeout or hard_timeout + * has passed. */ + ovs_mutex_lock(&ofproto_mutex); +- LIST_FOR_EACH_SAFE (rule, next_rule, expirable, ++ LIST_FOR_EACH_SAFE (rule, expirable, + &ofproto->up.expirable) { + rule_expire(rule_dpif_cast(rule), now); + } +@@ -2346,6 +2339,7 @@ set_ipfix( + struct dpif_ipfix *di = ofproto->ipfix; + bool has_options = bridge_exporter_options || flow_exporters_options; + bool new_di = false; ++ bool options_changed = false; + + if (has_options && !di) { + di = ofproto->ipfix = dpif_ipfix_create(); +@@ -2355,7 +2349,7 @@ set_ipfix( + if (di) { + /* Call set_options in any case to cleanly flush the flow + * caches in the last exporters that are to be destroyed. */ +- dpif_ipfix_set_options( ++ options_changed = dpif_ipfix_set_options( + di, bridge_exporter_options, flow_exporters_options, + n_flow_exporters_options); + +@@ -2371,6 +2365,10 @@ set_ipfix( + dpif_ipfix_unref(di); + ofproto->ipfix = NULL; + } ++ ++ if (new_di || options_changed) { ++ ofproto->backer->need_revalidate = REV_RECONFIGURE; ++ } + } + + return 0; +@@ -2493,11 +2491,11 @@ set_lldp(struct ofport *ofport_, + { + struct ofport_dpif *ofport = ofport_dpif_cast(ofport_); + struct ofproto_dpif *ofproto = ofproto_dpif_cast(ofport->up.ofproto); ++ bool old_enable = lldp_is_enabled(ofport->lldp); + int error = 0; + +- if (cfg) { ++ if (cfg && !smap_is_empty(cfg)) { + if (!ofport->lldp) { +- ofproto->backer->need_revalidate = REV_RECONFIGURE; + ofport->lldp = lldp_create(ofport->up.netdev, ofport_->mtu, cfg); + } + +@@ -2509,6 +2507,9 @@ set_lldp(struct ofport *ofport_, + } else if (ofport->lldp) { + lldp_unref(ofport->lldp); + ofport->lldp = NULL; ++ } ++ ++ if (lldp_is_enabled(ofport->lldp) != old_enable) { + ofproto->backer->need_revalidate = REV_RECONFIGURE; + } + +@@ -3106,11 +3107,11 @@ bundle_flush_macs(struct ofbundle *bundle, bool all_ofprotos) + { + struct ofproto_dpif *ofproto = bundle->ofproto; + struct mac_learning *ml = ofproto->ml; +- struct mac_entry *mac, *next_mac; ++ struct mac_entry *mac; + + ofproto->backer->need_revalidate = REV_RECONFIGURE; + ovs_rwlock_wrlock(&ml->rwlock); +- LIST_FOR_EACH_SAFE (mac, next_mac, lru_node, &ml->lrus) { ++ LIST_FOR_EACH_SAFE (mac, lru_node, &ml->lrus) { + if (mac_entry_get_port(ml, mac) == bundle) { + if (all_ofprotos) { + struct ofproto_dpif *o; +@@ -3141,13 +3142,13 @@ bundle_move(struct ofbundle *old, struct ofbundle *new) + { + struct ofproto_dpif *ofproto = old->ofproto; + struct mac_learning *ml = ofproto->ml; +- struct mac_entry *mac, *next_mac; ++ struct mac_entry *mac; + + ovs_assert(new->ofproto == old->ofproto); + + ofproto->backer->need_revalidate = REV_RECONFIGURE; + ovs_rwlock_wrlock(&ml->rwlock); +- LIST_FOR_EACH_SAFE (mac, next_mac, lru_node, &ml->lrus) { ++ LIST_FOR_EACH_SAFE (mac, lru_node, &ml->lrus) { + if (mac_entry_get_port(ml, mac) == old) { + mac_entry_set_port(ml, mac, new); + } +@@ -3244,7 +3245,7 @@ static void + bundle_destroy(struct ofbundle *bundle) + { + struct ofproto_dpif *ofproto; +- struct ofport_dpif *port, *next_port; ++ struct ofport_dpif *port; + + if (!bundle) { + return; +@@ -3257,7 +3258,7 @@ bundle_destroy(struct ofbundle *bundle) + xlate_bundle_remove(bundle); + xlate_txn_commit(); + +- LIST_FOR_EACH_SAFE (port, next_port, bundle_node, &bundle->ports) { ++ LIST_FOR_EACH_SAFE (port, bundle_node, &bundle->ports) { + bundle_del_port(port); + } + +@@ -3347,9 +3348,7 @@ bundle_set(struct ofproto *ofproto_, void *aux, + } + } + if (!ok || ovs_list_size(&bundle->ports) != s->n_members) { +- struct ofport_dpif *next_port; +- +- LIST_FOR_EACH_SAFE (port, next_port, bundle_node, &bundle->ports) { ++ LIST_FOR_EACH_SAFE (port, bundle_node, &bundle->ports) { + for (i = 0; i < s->n_members; i++) { + if (s->members[i] == port->up.ofp_port) { + goto found; +@@ -3963,6 +3962,10 @@ port_add(struct ofproto *ofproto_, struct netdev *netdev) + simap_put(&ofproto->backer->tnl_backers, + dp_port_name, odp_to_u32(port_no)); + } ++ } else { ++ struct dpif *dpif = ofproto->backer->dpif; ++ const char *dpif_type_str = dpif_normalize_type(dpif_type(dpif)); ++ netdev_set_dpif_type(netdev, dpif_type_str); + } + + if (netdev_get_tunnel_config(netdev)) { +@@ -4471,12 +4474,14 @@ rule_dpif_lookup_from_table(struct ofproto_dpif *ofproto, + atomic_add_relaxed(&tbl->n_matched, stats->n_packets, &orig); + } + if (xcache) { +- struct xc_entry *entry; ++ if (ofproto_try_ref(&ofproto->up)) { ++ struct xc_entry *entry; + +- entry = xlate_cache_add_entry(xcache, XC_TABLE); +- entry->table.ofproto = ofproto; +- entry->table.id = *table_id; +- entry->table.match = true; ++ entry = xlate_cache_add_entry(xcache, XC_TABLE); ++ entry->table.ofproto = ofproto; ++ entry->table.id = *table_id; ++ entry->table.match = true; ++ } + } + return rule; + } +@@ -4507,12 +4512,14 @@ rule_dpif_lookup_from_table(struct ofproto_dpif *ofproto, + stats->n_packets, &orig); + } + if (xcache) { +- struct xc_entry *entry; ++ if (ofproto_try_ref(&ofproto->up)) { ++ struct xc_entry *entry; + +- entry = xlate_cache_add_entry(xcache, XC_TABLE); +- entry->table.ofproto = ofproto; +- entry->table.id = next_id; +- entry->table.match = (rule != NULL); ++ entry = xlate_cache_add_entry(xcache, XC_TABLE); ++ entry->table.ofproto = ofproto; ++ entry->table.id = next_id; ++ entry->table.match = (rule != NULL); ++ } + } + if (rule) { + goto out; /* Match. */ +@@ -5550,9 +5557,9 @@ ct_zone_timeout_policy_sweep(struct dpif_backer *backer) + { + if (!ovs_list_is_empty(&backer->ct_tp_kill_list) + && time_msec() >= timeout_policy_cleanup_timer) { +- struct ct_timeout_policy *ct_tp, *next; ++ struct ct_timeout_policy *ct_tp; + +- LIST_FOR_EACH_SAFE (ct_tp, next, list_node, &backer->ct_tp_kill_list) { ++ LIST_FOR_EACH_SAFE (ct_tp, list_node, &backer->ct_tp_kill_list) { + if (!ct_dpif_del_timeout_policy(backer->dpif, ct_tp->tp_id)) { + ovs_list_remove(&ct_tp->list_node); + ct_timeout_policy_destroy(ct_tp, backer->tp_ids); +@@ -5594,6 +5601,7 @@ ct_set_zone_timeout_policy(const char *datapath_type, uint16_t zone_id, + ct_timeout_policy_unref(backer, ct_zone->ct_tp); + ct_zone->ct_tp = ct_tp; + ct_tp->ref_count++; ++ backer->need_revalidate = REV_RECONFIGURE; + } + } else { + struct ct_zone *new_ct_zone = ct_zone_alloc(zone_id); +@@ -5601,6 +5609,7 @@ ct_set_zone_timeout_policy(const char *datapath_type, uint16_t zone_id, + cmap_insert(&backer->ct_zones, &new_ct_zone->node, + hash_int(zone_id, 0)); + ct_tp->ref_count++; ++ backer->need_revalidate = REV_RECONFIGURE; + } + } + +@@ -5617,6 +5626,7 @@ ct_del_zone_timeout_policy(const char *datapath_type, uint16_t zone_id) + if (ct_zone) { + ct_timeout_policy_unref(backer, ct_zone->ct_tp); + ct_zone_remove_and_destroy(backer, ct_zone); ++ backer->need_revalidate = REV_RECONFIGURE; + } + } + +@@ -5818,15 +5828,7 @@ ofproto_dpif_lookup_by_name(const char *name) + struct ofproto_dpif * + ofproto_dpif_lookup_by_uuid(const struct uuid *uuid) + { +- struct ofproto_dpif *ofproto; +- +- HMAP_FOR_EACH_WITH_HASH (ofproto, all_ofproto_dpifs_by_uuid_node, +- uuid_hash(uuid), &all_ofproto_dpifs_by_uuid) { +- if (uuid_equals(&ofproto->uuid, uuid)) { +- return ofproto; +- } +- } +- return NULL; ++ return xlate_ofproto_lookup(uuid); + } + + static void +diff --git a/ofproto/ofproto-provider.h b/ofproto/ofproto-provider.h +index 14b909973d..47e96e62e1 100644 +--- a/ofproto/ofproto-provider.h ++++ b/ofproto/ofproto-provider.h +@@ -143,6 +143,8 @@ struct ofproto { + /* Variable length mf_field mapping. Stores all configured variable length + * meta-flow fields (struct mf_field) in a switch. */ + struct vl_mff_map vl_mff_map; ++ /* refcount to this ofproto, held by rule/group/xlate_caches */ ++ struct ovs_refcount refcount; + }; + + void ofproto_init_tables(struct ofproto *, int n_tables); +diff --git a/ofproto/ofproto.c b/ofproto/ofproto.c +index 56aeac7209..933f7de2dc 100644 +--- a/ofproto/ofproto.c ++++ b/ofproto/ofproto.c +@@ -549,6 +549,7 @@ ofproto_create(const char *datapath_name, const char *datapath_type, + + ovs_mutex_init(&ofproto->vl_mff_map.mutex); + cmap_init(&ofproto->vl_mff_map.cmap); ++ ovs_refcount_init(&ofproto->refcount); + + error = ofproto->ofproto_class->construct(ofproto); + if (error) { +@@ -1695,9 +1696,33 @@ ofproto_destroy__(struct ofproto *ofproto) + ofproto->ofproto_class->dealloc(ofproto); + } + +-/* Destroying rules is doubly deferred, must have 'ofproto' around for them. +- * - 1st we defer the removal of the rules from the classifier +- * - 2nd we defer the actual destruction of the rules. */ ++/* ++ * Rule destruction requires ofproto to remain accessible. ++ * Depending on the rule destruction call (shown in below), it can take several ++ * RCU grace periods before the ofproto reference is not needed anymore. ++ * The ofproto destruction callback is thus protected by a refcount, ++ * and such destruction is itself deferred. ++ * ++ * remove_rules_postponed (one grace period) ++ * -> remove_rule_rcu ++ * -> remove_rule_rcu__ ++ * -> ofproto_rule_unref -> ref count != 1 ++ * -> ... more grace periods. ++ * -> rule_destroy_cb (> 2 grace periods) ++ * -> free ++ * ++ * NOTE: The original ofproto destruction is only deferred by two grace ++ * periods to keep ofproto accessible. By using refcount together the ++ * destruction can be deferred for longer time. Now ofproto has 3 states: ++ * ++ * state 1: alive, with refcount >= 1 ++ * state 2: dying, with refcount == 0, however pointer is valid ++ * state 3: died, memory freed, pointer might be dangling. ++ * ++ * We only need to add refcount to certain objects whose destruction can ++ * take several RCU grace periods (rule, group, xlate_cache). Other ++ * references to ofproto must be cleared before the 2 RCU grace periods. ++ */ + static void + ofproto_destroy_defer__(struct ofproto *ofproto) + OVS_EXCLUDED(ofproto_mutex) +@@ -1705,11 +1730,31 @@ ofproto_destroy_defer__(struct ofproto *ofproto) + ovsrcu_postpone(ofproto_destroy__, ofproto); + } + ++void ++ofproto_ref(struct ofproto *ofproto) ++{ ++ ovs_refcount_ref(&ofproto->refcount); ++} ++ ++bool ++ofproto_try_ref(struct ofproto *ofproto) ++{ ++ return ovs_refcount_try_ref_rcu(&ofproto->refcount); ++} ++ ++void ++ofproto_unref(struct ofproto *ofproto) ++{ ++ if (ofproto && ovs_refcount_unref(&ofproto->refcount) == 1) { ++ ovsrcu_postpone(ofproto_destroy_defer__, ofproto); ++ } ++} ++ + void + ofproto_destroy(struct ofproto *p, bool del) + OVS_EXCLUDED(ofproto_mutex) + { +- struct ofport *ofport, *next_ofport; ++ struct ofport *ofport; + struct ofport_usage *usage; + + if (!p) { +@@ -1717,7 +1762,7 @@ ofproto_destroy(struct ofproto *p, bool del) + } + + ofproto_flush__(p, del); +- HMAP_FOR_EACH_SAFE (ofport, next_ofport, hmap_node, &p->ports) { ++ HMAP_FOR_EACH_SAFE (ofport, hmap_node, &p->ports) { + ofport_destroy(ofport, del); + } + +@@ -1736,8 +1781,7 @@ ofproto_destroy(struct ofproto *p, bool del) + p->connmgr = NULL; + ovs_mutex_unlock(&ofproto_mutex); + +- /* Destroying rules is deferred, must have 'ofproto' around for them. */ +- ovsrcu_postpone(ofproto_destroy_defer__, p); ++ ofproto_unref(p); + } + + /* Destroys the datapath with the respective 'name' and 'type'. With the Linux +@@ -2782,7 +2826,7 @@ init_ports(struct ofproto *p) + { + struct ofproto_port_dump dump; + struct ofproto_port ofproto_port; +- struct shash_node *node, *next; ++ struct shash_node *node; + + OFPROTO_PORT_FOR_EACH (&ofproto_port, &dump, p) { + const char *name = ofproto_port.name; +@@ -2813,7 +2857,7 @@ init_ports(struct ofproto *p) + } + } + +- SHASH_FOR_EACH_SAFE(node, next, &init_ofp_ports) { ++ SHASH_FOR_EACH_SAFE (node, &init_ofp_ports) { + struct iface_hint *iface_hint = node->data; + + if (!strcmp(iface_hint->br_name, p->name)) { +@@ -2929,6 +2973,9 @@ ofproto_rule_destroy__(struct rule *rule) + cls_rule_destroy(CONST_CAST(struct cls_rule *, &rule->cr)); + rule_actions_destroy(rule_get_actions(rule)); + ovs_mutex_destroy(&rule->mutex); ++ /* ofproto_unref() must be called first. It is possible because ofproto ++ * destruction is deferred by an RCU grace period. */ ++ ofproto_unref(rule->ofproto); + rule->ofproto->ofproto_class->rule_dealloc(rule); + } + +@@ -3069,6 +3116,9 @@ group_destroy_cb(struct ofgroup *group) + &group->props)); + ofputil_bucket_list_destroy(CONST_CAST(struct ovs_list *, + &group->buckets)); ++ /* ofproto_unref() must be called first. It is possible because ofproto ++ * destruction is deferred by an RCU grace period. */ ++ ofproto_unref(group->ofproto); + group->ofproto->ofproto_class->group_dealloc(group); + } + +@@ -5271,10 +5321,15 @@ ofproto_rule_create(struct ofproto *ofproto, struct cls_rule *cr, + struct rule *rule; + enum ofperr error; + ++ if (!ofproto_try_ref(ofproto)) { ++ return OFPERR_OFPFMFC_UNKNOWN; ++ } ++ + /* Allocate new rule. */ + rule = ofproto->ofproto_class->rule_alloc(); + if (!rule) { + cls_rule_destroy(cr); ++ ofproto_unref(ofproto); + VLOG_WARN_RL(&rl, "%s: failed to allocate a rule.", ofproto->name); + return OFPERR_OFPFMFC_UNKNOWN; + } +@@ -6797,9 +6852,9 @@ static void + meter_delete_all(struct ofproto *ofproto) + OVS_REQUIRES(ofproto_mutex) + { +- struct meter *meter, *next; ++ struct meter *meter; + +- HMAP_FOR_EACH_SAFE (meter, next, node, &ofproto->meters) { ++ HMAP_FOR_EACH_SAFE (meter, node, &ofproto->meters) { + hmap_remove(&ofproto->meters, &meter->node); + meter_destroy(ofproto, meter); + } +@@ -7339,8 +7394,13 @@ init_group(struct ofproto *ofproto, const struct ofputil_group_mod *gm, + return OFPERR_OFPGMFC_BAD_TYPE; + } + ++ if (!ofproto_try_ref(ofproto)) { ++ return OFPERR_OFPFMFC_UNKNOWN; ++ } ++ + *ofgroup = ofproto->ofproto_class->group_alloc(); + if (!*ofgroup) { ++ ofproto_unref(ofproto); + VLOG_WARN_RL(&rl, "%s: failed to allocate group", ofproto->name); + return OFPERR_OFPGMFC_OUT_OF_GROUPS; + } +@@ -7377,6 +7437,7 @@ init_group(struct ofproto *ofproto, const struct ofputil_group_mod *gm, + &(*ofgroup)->props)); + ofputil_bucket_list_destroy(CONST_CAST(struct ovs_list *, + &(*ofgroup)->buckets)); ++ ofproto_unref(ofproto); + ofproto->ofproto_class->group_dealloc(*ofgroup); + } + return error; +@@ -8902,7 +8963,7 @@ eviction_group_hash_rule(struct rule *rule) + hash = table->eviction_group_id_basis; + miniflow_expand(rule->cr.match.flow, &flow); + for (sf = table->eviction_fields; +- sf < &table->eviction_fields[table->n_eviction_fields]; ++ sf && sf < &table->eviction_fields[table->n_eviction_fields]; + sf++) + { + if (mf_are_prereqs_ok(sf->field, &flow, NULL)) { +@@ -9138,8 +9199,8 @@ oftable_configure_eviction(struct oftable *table, unsigned int eviction, + + /* Destroy existing eviction groups, then destroy and recreate data + * structures to recover memory. */ +- struct eviction_group *evg, *next; +- HMAP_FOR_EACH_SAFE (evg, next, id_node, &table->eviction_groups_by_id) { ++ struct eviction_group *evg; ++ HMAP_FOR_EACH_SAFE (evg, id_node, &table->eviction_groups_by_id) { + eviction_group_destroy(table, evg); + } + hmap_destroy(&table->eviction_groups_by_id); +diff --git a/ofproto/ofproto.h b/ofproto/ofproto.h +index b0262da2df..4e15167ab7 100644 +--- a/ofproto/ofproto.h ++++ b/ofproto/ofproto.h +@@ -563,6 +563,10 @@ int ofproto_port_get_cfm_status(const struct ofproto *, + enum ofputil_table_miss ofproto_table_get_miss_config(const struct ofproto *, + uint8_t table_id); + ++void ofproto_ref(struct ofproto *); ++void ofproto_unref(struct ofproto *); ++bool ofproto_try_ref(struct ofproto *); ++ + #ifdef __cplusplus + } + #endif +diff --git a/ovsdb/.gitignore b/ovsdb/.gitignore +index fbcefafc6e..a4f9d38f11 100644 +--- a/ovsdb/.gitignore ++++ b/ovsdb/.gitignore +@@ -1,5 +1,7 @@ + /_server.ovsschema.inc + /_server.ovsschema.stamp ++/local-config.ovsschema.stamp ++/ovsdb.local-config.5 + /ovsdb-client + /ovsdb-client.1 + /ovsdb-doc +diff --git a/ovsdb/automake.mk b/ovsdb/automake.mk +index 62cc02686f..eba713bb6d 100644 +--- a/ovsdb/automake.mk ++++ b/ovsdb/automake.mk +@@ -148,4 +148,25 @@ ovsdb/ovsdb-server.5: \ + $(srcdir)/ovsdb/_server.xml > $@.tmp && \ + mv $@.tmp $@ + ++EXTRA_DIST += ovsdb/local-config.ovsschema ++pkgdata_DATA += ovsdb/local-config.ovsschema ++ ++# Version checking for local-config.ovsschema. ++ALL_LOCAL += ovsdb/local-config.ovsschema.stamp ++ovsdb/local-config.ovsschema.stamp: ovsdb/local-config.ovsschema ++ $(srcdir)/build-aux/cksum-schema-check $? $@ ++CLEANFILES += ovsdb/local-config.ovsschema.stamp ++ ++# Local_Config schema documentation ++EXTRA_DIST += ovsdb/local-config.xml ++CLEANFILES += ovsdb/ovsdb.local-config.5 ++man_MANS += ovsdb/ovsdb.local-config.5 ++ovsdb/ovsdb.local-config.5: \ ++ ovsdb/ovsdb-doc ovsdb/local-config.xml ovsdb/local-config.ovsschema ++ $(AM_V_GEN)$(OVSDB_DOC) \ ++ --version=$(VERSION) \ ++ $(srcdir)/ovsdb/local-config.ovsschema \ ++ $(srcdir)/ovsdb/local-config.xml > $@.tmp && \ ++ mv $@.tmp $@ ++ + EXTRA_DIST += ovsdb/TODO.rst +diff --git a/ovsdb/condition.c b/ovsdb/condition.c +index 388dd54a16..9aa3788dbb 100644 +--- a/ovsdb/condition.c ++++ b/ovsdb/condition.c +@@ -220,13 +220,13 @@ ovsdb_condition_optimize(struct ovsdb_condition *cnd) + static void + ovsdb_condition_optimize_destroy(struct ovsdb_condition *cnd) + { +- struct shash_node *node, *next; ++ struct shash_node *node; + +- SHASH_FOR_EACH_SAFE (node, next, &cnd->o_columns) { ++ SHASH_FOR_EACH_SAFE (node, &cnd->o_columns) { + struct ovsdb_o_column *o_column = node->data; +- struct ovsdb_o_clause *c, *c_next; ++ struct ovsdb_o_clause *c; + +- HMAP_FOR_EACH_SAFE(c, c_next, hmap_node, &o_column->o_clauses) { ++ HMAP_FOR_EACH_SAFE (c, hmap_node, &o_column->o_clauses) { + hmap_remove(&o_column->o_clauses, &c->hmap_node); + free(c); + } +diff --git a/ovsdb/file.c b/ovsdb/file.c +index 9f44007d97..ca80c28235 100644 +--- a/ovsdb/file.c ++++ b/ovsdb/file.c +@@ -524,6 +524,7 @@ ovsdb_file_read__(const char *filename, bool rw, + + error = ovsdb_txn_replay_commit(txn); + if (error) { ++ ovsdb_error_destroy(error); + ovsdb_storage_unread(storage); + break; + } +diff --git a/ovsdb/jsonrpc-server.c b/ovsdb/jsonrpc-server.c +index 351c39d8aa..916a1f414e 100644 +--- a/ovsdb/jsonrpc-server.c ++++ b/ovsdb/jsonrpc-server.c +@@ -197,9 +197,9 @@ ovsdb_jsonrpc_server_remove_db(struct ovsdb_jsonrpc_server *svr, + void + ovsdb_jsonrpc_server_destroy(struct ovsdb_jsonrpc_server *svr) + { +- struct shash_node *node, *next; ++ struct shash_node *node; + +- SHASH_FOR_EACH_SAFE (node, next, &svr->remotes) { ++ SHASH_FOR_EACH_SAFE (node, &svr->remotes) { + ovsdb_jsonrpc_server_del_remote(node); + } + shash_destroy(&svr->remotes); +@@ -227,9 +227,9 @@ void + ovsdb_jsonrpc_server_set_remotes(struct ovsdb_jsonrpc_server *svr, + const struct shash *new_remotes) + { +- struct shash_node *node, *next; ++ struct shash_node *node; + +- SHASH_FOR_EACH_SAFE (node, next, &svr->remotes) { ++ SHASH_FOR_EACH_SAFE (node, &svr->remotes) { + struct ovsdb_jsonrpc_remote *remote = node->data; + struct ovsdb_jsonrpc_options *options + = shash_find_data(new_remotes, node->name); +@@ -585,9 +585,9 @@ ovsdb_jsonrpc_session_set_options(struct ovsdb_jsonrpc_session *session, + static void + ovsdb_jsonrpc_session_run_all(struct ovsdb_jsonrpc_remote *remote) + { +- struct ovsdb_jsonrpc_session *s, *next; ++ struct ovsdb_jsonrpc_session *s; + +- LIST_FOR_EACH_SAFE (s, next, node, &remote->sessions) { ++ LIST_FOR_EACH_SAFE (s, node, &remote->sessions) { + int error = ovsdb_jsonrpc_session_run(s); + if (error) { + ovsdb_jsonrpc_session_close(s); +@@ -642,9 +642,9 @@ ovsdb_jsonrpc_session_get_memory_usage_all( + static void + ovsdb_jsonrpc_session_close_all(struct ovsdb_jsonrpc_remote *remote) + { +- struct ovsdb_jsonrpc_session *s, *next; ++ struct ovsdb_jsonrpc_session *s; + +- LIST_FOR_EACH_SAFE (s, next, node, &remote->sessions) { ++ LIST_FOR_EACH_SAFE (s, node, &remote->sessions) { + ovsdb_jsonrpc_session_close(s); + } + } +@@ -660,9 +660,9 @@ static void + ovsdb_jsonrpc_session_reconnect_all(struct ovsdb_jsonrpc_remote *remote, + bool force, const char *comment) + { +- struct ovsdb_jsonrpc_session *s, *next; ++ struct ovsdb_jsonrpc_session *s; + +- LIST_FOR_EACH_SAFE (s, next, node, &remote->sessions) { ++ LIST_FOR_EACH_SAFE (s, node, &remote->sessions) { + if (force || !s->db_change_aware) { + jsonrpc_session_force_reconnect(s->js); + if (comment && jsonrpc_session_is_connected(s->js)) { +@@ -909,9 +909,9 @@ error: + static void + ovsdb_jsonrpc_session_unlock_all(struct ovsdb_jsonrpc_session *s) + { +- struct ovsdb_lock_waiter *waiter, *next; ++ struct ovsdb_lock_waiter *waiter; + +- HMAP_FOR_EACH_SAFE (waiter, next, session_node, &s->up.waiters) { ++ HMAP_FOR_EACH_SAFE (waiter, session_node, &s->up.waiters) { + ovsdb_jsonrpc_session_unlock__(waiter); + } + } +@@ -1198,8 +1198,8 @@ static void + ovsdb_jsonrpc_trigger_remove__(struct ovsdb_jsonrpc_session *s, + struct ovsdb *db) + { +- struct ovsdb_jsonrpc_trigger *t, *next; +- HMAP_FOR_EACH_SAFE (t, next, hmap_node, &s->triggers) { ++ struct ovsdb_jsonrpc_trigger *t; ++ HMAP_FOR_EACH_SAFE (t, hmap_node, &s->triggers) { + if (!db || t->trigger.db == db) { + ovsdb_jsonrpc_trigger_complete(t); + } +@@ -1226,8 +1226,8 @@ ovsdb_jsonrpc_trigger_complete_all(struct ovsdb_jsonrpc_session *s) + static void + ovsdb_jsonrpc_trigger_complete_done(struct ovsdb_jsonrpc_session *s) + { +- struct ovsdb_jsonrpc_trigger *trigger, *next; +- LIST_FOR_EACH_SAFE (trigger, next, trigger.node, &s->up.completions) { ++ struct ovsdb_jsonrpc_trigger *trigger; ++ LIST_FOR_EACH_SAFE (trigger, trigger.node, &s->up.completions) { + ovsdb_jsonrpc_trigger_complete(trigger); + } + } +@@ -1688,8 +1688,8 @@ ovsdb_jsonrpc_monitor_preremove_db(struct ovsdb_jsonrpc_session *s, + { + ovs_assert(db); + +- struct ovsdb_jsonrpc_monitor *m, *next; +- HMAP_FOR_EACH_SAFE (m, next, node, &s->monitors) { ++ struct ovsdb_jsonrpc_monitor *m; ++ HMAP_FOR_EACH_SAFE (m, node, &s->monitors) { + if (m->db == db) { + ovsdb_jsonrpc_monitor_destroy(m, true); + } +@@ -1700,9 +1700,9 @@ ovsdb_jsonrpc_monitor_preremove_db(struct ovsdb_jsonrpc_session *s, + static void + ovsdb_jsonrpc_monitor_remove_all(struct ovsdb_jsonrpc_session *s) + { +- struct ovsdb_jsonrpc_monitor *m, *next; ++ struct ovsdb_jsonrpc_monitor *m; + +- HMAP_FOR_EACH_SAFE (m, next, node, &s->monitors) { ++ HMAP_FOR_EACH_SAFE (m, node, &s->monitors) { + ovsdb_jsonrpc_monitor_destroy(m, false); + } + } +diff --git a/ovsdb/libovsdb.pc.in b/ovsdb/libovsdb.pc.in +index fe367ea7b1..54c9039cfd 100644 +--- a/ovsdb/libovsdb.pc.in ++++ b/ovsdb/libovsdb.pc.in +@@ -8,4 +8,4 @@ Description: OVSDB library of Open vSwitch + Version: @VERSION@ + Libs: -L${libdir} -lovsdb + Libs.private: @LIBS@ +-Cflags: -I${includedir}/openvswitch ++Cflags: -I${includedir} +diff --git a/ovsdb/local-config.ovsschema b/ovsdb/local-config.ovsschema +new file mode 100644 +index 0000000000..bd86d0f4f6 +--- /dev/null ++++ b/ovsdb/local-config.ovsschema +@@ -0,0 +1,43 @@ ++{ ++ "name": "Local_Config", ++ "version": "1.0.0", ++ "cksum": "2048726482 1858", ++ "tables": { ++ "Config": { ++ "columns": { ++ "connections": { ++ "type": {"key": {"type": "uuid", ++ "refTable": "Connection"}, ++ "min": 0, ++ "max": "unlimited"}}}, ++ "maxRows": 1, ++ "isRoot": true}, ++ "Connection": { ++ "columns": { ++ "target": {"type": "string"}, ++ "max_backoff": {"type": {"key": {"type": "integer", ++ "minInteger": 1000}, ++ "min": 0, ++ "max": 1}}, ++ "inactivity_probe": {"type": {"key": "integer", ++ "min": 0, ++ "max": 1}}, ++ "read_only": {"type": "boolean"}, ++ "role": {"type": "string"}, ++ "other_config": {"type": {"key": "string", ++ "value": "string", ++ "min": 0, ++ "max": "unlimited"}}, ++ "external_ids": {"type": {"key": "string", ++ "value": "string", ++ "min": 0, ++ "max": "unlimited"}}, ++ "is_connected": {"type": "boolean", "ephemeral": true}, ++ "status": {"type": {"key": "string", ++ "value": "string", ++ "min": 0, ++ "max": "unlimited"}, ++ "ephemeral": true}}, ++ "indexes": [["target"]]} ++ } ++} +diff --git a/ovsdb/local-config.xml b/ovsdb/local-config.xml +new file mode 100644 +index 0000000000..b502aea4cf +--- /dev/null ++++ b/ovsdb/local-config.xml +@@ -0,0 +1,296 @@ ++ ++ ++

++ This database is for local configuration of an ovsdb-server. The ++ database is meant to be unique, even among multiple clustered db ++ servers, so that configuration that is local to that server can ++ be configured separately (e.g. Connection information). ++

++ ++ ++

++ The root local configuration table for an ovsdb-server. This table ++ must have exactly one row. ++

++ ++ ++ Database clients to which the Open vSwitch database server should ++ connect or on which it should listen, along with options for how these ++ connections should be configured. See the ++ table for more information. ++ ++ ++
++ ++ ++

++ Configuration for a database connection to an Open vSwitch database ++ (OVSDB) client. ++

++ ++

++ This table primarily configures the Open vSwitch database server ++ (ovsdb-server). ++

++ ++

++ The Open vSwitch database server can initiate and maintain active ++ connections to remote clients. It can also listen for database ++ connections. ++

++ ++ ++ ++

Connection methods for clients.

++

++ The following connection methods are currently supported: ++

++
++
++ ssl:host[:port] ++
++
++

++ The specified SSL port on the host at the given ++ host, which can either be a DNS name (if built with ++ unbound library) or an IP address. A valid SSL configuration must ++ be provided when this form is used, this configuration can be ++ specified via command-line options or the ++ table. ++

++

++ If port is not specified, it defaults to 6640. ++

++

++ SSL support is an optional feature that is not always ++ built as part of Open vSwitch. ++

++
++ ++
++ tcp:host[:port] ++
++
++

++ The specified TCP port on the host at the given ++ host, which can either be a DNS name (if built with ++ unbound library) or an IP address. If host is an IPv6 ++ address, wrap it in square brackets, e.g. ++ tcp:[::1]:6640. ++

++

++ If port is not specified, it defaults to 6640. ++

++
++
++ pssl:[port][:host] ++
++
++

++ Listens for SSL connections on the specified TCP port. ++ Specify 0 for port to have the kernel automatically ++ choose an available port. If host, which can either ++ be a DNS name (if built with unbound library) or an IP address, ++ is specified, then connections are restricted to the resolved or ++ specified local IPaddress (either IPv4 or IPv6 address). If ++ host is an IPv6 address, wrap in square brackets, ++ e.g. pssl:6640:[::1]. If host is not ++ specified then it listens only on IPv4 (but not IPv6) addresses. ++ A valid SSL configuration must be provided when this form is ++ used, this can be specified either via command-line options or ++ the table. ++

++

++ If port is not specified, it defaults to 6640. ++

++

++ SSL support is an optional feature that is not always built as ++ part of Open vSwitch. ++

++
++
++ ptcp:[port][:host] ++
++
++

++ Listens for connections on the specified TCP port. ++ Specify 0 for port to have the kernel automatically ++ choose an available port. If host, which can either ++ be a DNS name (if built with unbound library) or an IP address, ++ is specified, then connections are restricted to the resolved or ++ specified local IP address (either IPv4 or IPv6 address). If ++ host is an IPv6 address, wrap it in square brackets, ++ e.g. ptcp:6640:[::1]. If host is not ++ specified then it listens only on IPv4 addresses. ++

++

++ If port is not specified, it defaults to 6640. ++

++
++
++

When multiple clients are configured, the ++ values must be unique. Duplicate values yield ++ unspecified results.

++ ++ ++ ++ true to restrict these connections to read-only ++ transactions, false to allow them to modify the database. ++ ++ ++ ++ String containing role name for this connection entry. ++ ++ ++ ++ ++ ++ Maximum number of milliseconds to wait between connection attempts. ++ Default is implementation-specific. ++ ++ ++ ++ Maximum number of milliseconds of idle time on connection to the client ++ before sending an inactivity probe message. If Open vSwitch does not ++ communicate with the client for the specified number of seconds, it ++ will send a probe. If a response is not received for the same ++ additional amount of time, Open vSwitch assumes the connection has been ++ broken and attempts to reconnect. Default is implementation-specific. ++ A value of 0 disables inactivity probes. ++ ++ ++ ++ ++

++ Key-value pair of is always updated. ++ Other key-value pairs in the status columns may be updated depends ++ on the type. ++

++ ++

++ When specifies a connection method that ++ listens for inbound connections (e.g. ptcp: or ++ punix:), both and ++ may also be updated while the ++ remaining key-value pairs are omitted. ++

++ ++

++ On the other hand, when specifies an ++ outbound connection, all key-value pairs may be updated, except ++ the above-mentioned two key-value pairs associated with inbound ++ connection targets. They are omitted. ++

++ ++ ++ true if currently connected to this client, ++ false otherwise. ++ ++ ++ ++ A human-readable description of the last error on the connection ++ to the manager; i.e. strerror(errno). This key ++ will exist only if an error has occurred. ++ ++ ++ ++

++ The state of the connection to the manager: ++

++
++
VOID
++
Connection is disabled.
++ ++
BACKOFF
++
Attempting to reconnect at an increasing period.
++ ++
CONNECTING
++
Attempting to connect.
++ ++
ACTIVE
++
Connected, remote host responsive.
++ ++
IDLE
++
Connection is idle. Waiting for response to keep-alive.
++
++

++ These values may change in the future. They are provided only for ++ human consumption. ++

++ ++ ++ ++ The amount of time since this client last successfully connected ++ to the database (in seconds). Value is empty if client has never ++ successfully been connected. ++ ++ ++ ++ The amount of time since this client last disconnected from the ++ database (in seconds). Value is empty if client has never ++ disconnected. ++ ++ ++ ++ Space-separated list of the names of OVSDB locks that the connection ++ holds. Omitted if the connection does not hold any locks. ++ ++ ++ ++ Space-separated list of the names of OVSDB locks that the connection is ++ currently waiting to acquire. Omitted if the connection is not waiting ++ for any locks. ++ ++ ++ ++ Space-separated list of the names of OVSDB locks that the connection ++ has had stolen by another OVSDB client. Omitted if no locks have been ++ stolen from this connection. ++ ++ ++ ++ When specifies a connection method that ++ listens for inbound connections (e.g. ptcp: or ++ pssl:) and more than one connection is actually active, ++ the value is the number of active connections. Otherwise, this ++ key-value pair is omitted. ++ ++ ++ ++ When is ptcp: or ++ pssl:, this is the TCP port on which the OVSDB server is ++ listening. (This is particularly useful when specifies a port of 0, allowing the kernel to ++ choose any available port.) ++ ++ ++ ++ ++ ++ The Differentiated Service Code Point (DSCP) is specified using 6 bits ++ in the Type of Service (TOS) field in the IP header. DSCP provides a ++ mechanism to classify the network traffic and provide Quality of ++ Service (QoS) on IP networks. ++ ++ The DSCP value specified here is used when establishing the connection ++ between the manager and the Open vSwitch. If no value is specified, a ++ default value of 48 is chosen. Valid DSCP values must be in the range ++ 0 to 63. ++ ++ ++ ++ ++ External configuration options ++ ++ ++ External client-defined key-value pairs ++ ++ ++
++ +diff --git a/ovsdb/monitor.c b/ovsdb/monitor.c +index 0f222cc992..952fa902e4 100644 +--- a/ovsdb/monitor.c ++++ b/ovsdb/monitor.c +@@ -638,14 +638,14 @@ ovsdb_monitor_change_set_destroy(struct ovsdb_monitor_change_set *mcs) + { + ovs_list_remove(&mcs->list_node); + +- struct ovsdb_monitor_change_set_for_table *mcst, *next_mcst; +- LIST_FOR_EACH_SAFE (mcst, next_mcst, list_in_change_set, ++ struct ovsdb_monitor_change_set_for_table *mcst; ++ LIST_FOR_EACH_SAFE (mcst, list_in_change_set, + &mcs->change_set_for_tables) { + ovs_list_remove(&mcst->list_in_change_set); + ovs_list_remove(&mcst->list_in_mt); + +- struct ovsdb_monitor_row *row, *next; +- HMAP_FOR_EACH_SAFE (row, next, hmap_node, &mcst->rows) { ++ struct ovsdb_monitor_row *row; ++ HMAP_FOR_EACH_SAFE (row, hmap_node, &mcst->rows) { + hmap_remove(&mcst->rows, &row->hmap_node); + ovsdb_monitor_row_destroy(mcst->mt, row, mcst->n_columns); + } +@@ -700,13 +700,13 @@ void + ovsdb_monitor_session_condition_destroy( + struct ovsdb_monitor_session_condition *condition) + { +- struct shash_node *node, *next; ++ struct shash_node *node; + + if (!condition) { + return; + } + +- SHASH_FOR_EACH_SAFE (node, next, &condition->tables) { ++ SHASH_FOR_EACH_SAFE (node, &condition->tables) { + struct ovsdb_monitor_table_condition *mtc = node->data; + + ovsdb_condition_destroy(&mtc->new_condition); +@@ -1122,11 +1122,11 @@ ovsdb_monitor_compose_update( + json = NULL; + struct ovsdb_monitor_change_set_for_table *mcst; + LIST_FOR_EACH (mcst, list_in_change_set, &mcs->change_set_for_tables) { +- struct ovsdb_monitor_row *row, *next; ++ struct ovsdb_monitor_row *row; + struct json *table_json = NULL; + struct ovsdb_monitor_table *mt = mcst->mt; + +- HMAP_FOR_EACH_SAFE (row, next, hmap_node, &mcst->rows) { ++ HMAP_FOR_EACH_SAFE (row, hmap_node, &mcst->rows) { + struct json *row_json; + row_json = (*row_update)(mt, condition, OVSDB_MONITOR_ROW, row, + initial, changed, mcst->n_columns); +@@ -1711,8 +1711,8 @@ ovsdb_monitor_destroy(struct ovsdb_monitor *dbmon) + ovsdb_monitor_json_cache_flush(dbmon); + hmap_destroy(&dbmon->json_cache); + +- struct ovsdb_monitor_change_set *cs, *cs_next; +- LIST_FOR_EACH_SAFE (cs, cs_next, list_node, &dbmon->change_sets) { ++ struct ovsdb_monitor_change_set *cs; ++ LIST_FOR_EACH_SAFE (cs, list_node, &dbmon->change_sets) { + ovsdb_monitor_change_set_destroy(cs); + } + +@@ -1760,14 +1760,14 @@ ovsdb_monitors_commit(struct ovsdb *db, const struct ovsdb_txn *txn) + void + ovsdb_monitors_remove(struct ovsdb *db) + { +- struct ovsdb_monitor *m, *next_m; ++ struct ovsdb_monitor *m; + +- LIST_FOR_EACH_SAFE (m, next_m, list_node, &db->monitors) { +- struct jsonrpc_monitor_node *jm, *next_jm; ++ LIST_FOR_EACH_SAFE (m, list_node, &db->monitors) { ++ struct jsonrpc_monitor_node *jm; + + /* Delete all front-end monitors. Removing the last front-end monitor + * will also destroy the corresponding ovsdb_monitor. */ +- LIST_FOR_EACH_SAFE (jm, next_jm, node, &m->jsonrpc_monitors) { ++ LIST_FOR_EACH_SAFE (jm, node, &m->jsonrpc_monitors) { + ovsdb_jsonrpc_monitor_destroy(jm->jsonrpc_monitor, false); + } + } +@@ -1789,14 +1789,14 @@ ovsdb_monitor_get_memory_usage(struct simap *usage) + void + ovsdb_monitor_prereplace_db(struct ovsdb *db) + { +- struct ovsdb_monitor *m, *next_m; ++ struct ovsdb_monitor *m; + +- LIST_FOR_EACH_SAFE (m, next_m, list_node, &db->monitors) { +- struct jsonrpc_monitor_node *jm, *next_jm; ++ LIST_FOR_EACH_SAFE (m, list_node, &db->monitors) { ++ struct jsonrpc_monitor_node *jm; + + /* Delete all front-end monitors. Removing the last front-end monitor + * will also destroy the corresponding ovsdb_monitor. */ +- LIST_FOR_EACH_SAFE (jm, next_jm, node, &m->jsonrpc_monitors) { ++ LIST_FOR_EACH_SAFE (jm, node, &m->jsonrpc_monitors) { + ovsdb_jsonrpc_monitor_destroy(jm->jsonrpc_monitor, true); + } + } +diff --git a/ovsdb/ovsdb-idlc.in b/ovsdb/ovsdb-idlc.in +index 10a70ae26f..13c5359395 100755 +--- a/ovsdb/ovsdb-idlc.in ++++ b/ovsdb/ovsdb-idlc.in +@@ -251,10 +251,18 @@ const struct %(s)s *%(s)s_table_first(const struct %(s)s_table *); + for ((ROW) = %(s)s_table_first(TABLE); \\ + (ROW); \\ + (ROW) = %(s)s_next(ROW)) +-#define %(S)s_TABLE_FOR_EACH_SAFE(ROW, NEXT, TABLE) \\ ++#define %(S)s_TABLE_FOR_EACH_SAFE_LONG(ROW, NEXT, TABLE) \\ + for ((ROW) = %(s)s_table_first(TABLE); \\ + (ROW) ? ((NEXT) = %(s)s_next(ROW), 1) : 0; \\ + (ROW) = (NEXT)) ++#define %(S)s_TABLE_FOR_EACH_SAFE_SHORT(ROW, TABLE) \\ ++ for (const struct %(s)s * ROW__next = ((ROW) = %(s)s_table_first(TABLE), NULL); \\ ++ (ROW) ? (ROW__next = %(s)s_next(ROW), 1) : (ROW__next = NULL, 0); \\ ++ (ROW) = ROW__next) ++#define %(S)s_TABLE_FOR_EACH_SAFE(...) \\ ++ OVERLOAD_SAFE_MACRO(%(S)s_TABLE_FOR_EACH_SAFE_LONG, \\ ++ %(S)s_TABLE_FOR_EACH_SAFE_SHORT, 3, __VA_ARGS__) ++ + + const struct %(s)s *%(s)s_get_for_uuid(const struct ovsdb_idl *, const struct uuid *); + const struct %(s)s *%(s)s_table_get_for_uuid(const struct %(s)s_table *, const struct uuid *); +@@ -264,10 +272,17 @@ const struct %(s)s *%(s)s_next(const struct %(s)s *); + for ((ROW) = %(s)s_first(IDL); \\ + (ROW); \\ + (ROW) = %(s)s_next(ROW)) +-#define %(S)s_FOR_EACH_SAFE(ROW, NEXT, IDL) \\ ++#define %(S)s_FOR_EACH_SAFE_LONG(ROW, NEXT, IDL) \\ + for ((ROW) = %(s)s_first(IDL); \\ + (ROW) ? ((NEXT) = %(s)s_next(ROW), 1) : 0; \\ + (ROW) = (NEXT)) ++#define %(S)s_FOR_EACH_SAFE_SHORT(ROW, IDL) \\ ++ for (const struct %(s)s * ROW__next = ((ROW) = %(s)s_first(IDL), NULL); \\ ++ (ROW) ? (ROW__next = %(s)s_next(ROW), 1) : (ROW__next = NULL, 0); \\ ++ (ROW) = ROW__next) ++#define %(S)s_FOR_EACH_SAFE(...) \\ ++ OVERLOAD_SAFE_MACRO(%(S)s_FOR_EACH_SAFE_LONG, \\ ++ %(S)s_FOR_EACH_SAFE_SHORT, 3, __VA_ARGS__) + + unsigned int %(s)s_get_seqno(const struct ovsdb_idl *); + unsigned int %(s)s_row_get_seqno(const struct %(s)s *row, enum ovsdb_idl_change change); +diff --git a/ovsdb/ovsdb-server.c b/ovsdb/ovsdb-server.c +index 9fe90592ea..774416fc7a 100644 +--- a/ovsdb/ovsdb-server.c ++++ b/ovsdb/ovsdb-server.c +@@ -26,6 +26,7 @@ + #include "command-line.h" + #include "daemon.h" + #include "dirs.h" ++#include "dns-resolve.h" + #include "openvswitch/dynamic-string.h" + #include "fatal-signal.h" + #include "file.h" +@@ -228,8 +229,7 @@ main_loop(struct server_config *config, + + ovsdb_relay_run(); + +- struct shash_node *next; +- SHASH_FOR_EACH_SAFE (node, next, all_dbs) { ++ SHASH_FOR_EACH_SAFE (node, all_dbs) { + struct db *db = node->data; + ovsdb_txn_history_run(db->db); + ovsdb_storage_run(db->db->storage); +@@ -321,7 +321,7 @@ main(int argc, char *argv[]) + FILE *config_tmpfile; + struct server_config server_config; + struct shash all_dbs; +- struct shash_node *node, *next; ++ struct shash_node *node; + int replication_probe_interval = REPLICATION_DEFAULT_PROBE_INTERVAL; + + ovs_cmdl_proctitle_init(argc, argv); +@@ -329,6 +329,7 @@ main(int argc, char *argv[]) + service_start(&argc, &argv); + fatal_ignore_sigpipe(); + process_init(); ++ dns_resolve_init(true); + + bool active = false; + parse_options(argc, argv, &db_filenames, &remotes, &unixctl_path, +@@ -490,7 +491,7 @@ main(int argc, char *argv[]) + main_loop(&server_config, jsonrpc, &all_dbs, unixctl, &remotes, + run_process, &exiting, &is_backup); + +- SHASH_FOR_EACH_SAFE(node, next, &all_dbs) { ++ SHASH_FOR_EACH_SAFE (node, &all_dbs) { + struct db *db = node->data; + close_db(&server_config, db, NULL); + shash_delete(&all_dbs, node); +@@ -511,6 +512,7 @@ main(int argc, char *argv[]) + run_command, process_status_msg(status)); + } + } ++ dns_resolve_destroy(); + perf_counters_destroy(); + service_stop(); + return 0; +@@ -1240,8 +1242,8 @@ update_server_status(struct shash *all_dbs) + + /* Update rows for databases that still exist. + * Delete rows for databases that no longer exist. */ +- const struct ovsdb_row *row, *next_row; +- HMAP_FOR_EACH_SAFE (row, next_row, hmap_node, &database_table->rows) { ++ const struct ovsdb_row *row; ++ HMAP_FOR_EACH_SAFE (row, hmap_node, &database_table->rows) { + const char *name; + ovsdb_util_read_string_column(row, "name", &name); + struct db *db = shash_find_data(all_dbs, name); +diff --git a/ovsdb/ovsdb-tool.c b/ovsdb/ovsdb-tool.c +index d4a9e34cc4..df2e373c3c 100644 +--- a/ovsdb/ovsdb-tool.c ++++ b/ovsdb/ovsdb-tool.c +@@ -1579,15 +1579,14 @@ do_check_cluster(struct ovs_cmdl_context *ctx) + } + free(c.servers); + +- struct commit *next_commit; +- HMAP_FOR_EACH_SAFE (commit, next_commit, hmap_node, &c.commits) { ++ HMAP_FOR_EACH_SAFE (commit, hmap_node, &c.commits) { + hmap_remove(&c.commits, &commit->hmap_node); + free(commit); + } + hmap_destroy(&c.commits); + +- struct leader *leader, *next_leader; +- HMAP_FOR_EACH_SAFE (leader, next_leader, hmap_node, &c.leaders) { ++ struct leader *leader; ++ HMAP_FOR_EACH_SAFE (leader, hmap_node, &c.leaders) { + hmap_remove(&c.leaders, &leader->hmap_node); + free(leader); + } +diff --git a/ovsdb/ovsdb.c b/ovsdb/ovsdb.c +index e6d866182c..91b4a01af8 100644 +--- a/ovsdb/ovsdb.c ++++ b/ovsdb/ovsdb.c +@@ -571,8 +571,8 @@ ovsdb_replace(struct ovsdb *dst, struct ovsdb *src) + ovsdb_monitor_prereplace_db(dst); + + /* Cancel triggers. */ +- struct ovsdb_trigger *trigger, *next; +- LIST_FOR_EACH_SAFE (trigger, next, node, &dst->triggers) { ++ struct ovsdb_trigger *trigger; ++ LIST_FOR_EACH_SAFE (trigger, node, &dst->triggers) { + ovsdb_trigger_prereplace_db(trigger); + } + +diff --git a/ovsdb/query.c b/ovsdb/query.c +index de74519989..eebe564127 100644 +--- a/ovsdb/query.c ++++ b/ovsdb/query.c +@@ -40,9 +40,9 @@ ovsdb_query(struct ovsdb_table *table, const struct ovsdb_condition *cnd, + } + } else { + /* Linear scan. */ +- const struct ovsdb_row *row, *next; ++ const struct ovsdb_row *row; + +- HMAP_FOR_EACH_SAFE (row, next, hmap_node, &table->rows) { ++ HMAP_FOR_EACH_SAFE (row, hmap_node, &table->rows) { + if (ovsdb_condition_match_every_clause(row, cnd) && + !output_row(row, aux)) { + break; +diff --git a/ovsdb/raft-private.c b/ovsdb/raft-private.c +index 30760233ee..e685c8103b 100644 +--- a/ovsdb/raft-private.c ++++ b/ovsdb/raft-private.c +@@ -36,7 +36,10 @@ raft_address_validate(const char *address) + return NULL; + } else if (!strncmp(address, "ssl:", 4) || !strncmp(address, "tcp:", 4)) { + struct sockaddr_storage ss; +- if (!inet_parse_active(address + 4, -1, &ss, true)) { ++ bool dns_failure = false; ++ ++ if (!inet_parse_active(address + 4, -1, &ss, true, &dns_failure) ++ && !dns_failure) { + return ovsdb_error(NULL, "%s: syntax error in address", address); + } + return NULL; +@@ -147,8 +150,8 @@ raft_server_destroy(struct raft_server *s) + void + raft_servers_destroy(struct hmap *servers) + { +- struct raft_server *s, *next; +- HMAP_FOR_EACH_SAFE (s, next, hmap_node, servers) { ++ struct raft_server *s; ++ HMAP_FOR_EACH_SAFE (s, hmap_node, servers) { + hmap_remove(servers, &s->hmap_node); + raft_server_destroy(s); + } +diff --git a/ovsdb/raft.c b/ovsdb/raft.c +index 1a3447a8dd..cf9edf35c6 100644 +--- a/ovsdb/raft.c ++++ b/ovsdb/raft.c +@@ -74,9 +74,12 @@ enum raft_failure_test { + FT_CRASH_BEFORE_SEND_EXEC_REQ, + FT_CRASH_AFTER_SEND_EXEC_REQ, + FT_CRASH_AFTER_RECV_APPEND_REQ_UPDATE, ++ FT_CRASH_BEFORE_SEND_SNAPSHOT_REP, + FT_DELAY_ELECTION, + FT_DONT_SEND_VOTE_REQUEST, + FT_STOP_RAFT_RPC, ++ FT_TRANSFER_LEADERSHIP, ++ FT_TRANSFER_LEADERSHIP_AFTER_SEND_APPEND_REQ, + }; + static enum raft_failure_test failure_test; + +@@ -379,12 +382,19 @@ static bool raft_handle_write_error(struct raft *, struct ovsdb_error *); + static void raft_run_reconfigure(struct raft *); + + static void raft_set_leader(struct raft *, const struct uuid *sid); ++ + static struct raft_server * + raft_find_server(const struct raft *raft, const struct uuid *sid) + { + return raft_server_find(&raft->servers, sid); + } + ++static struct raft_server * ++raft_find_new_server(struct raft *raft, const struct uuid *uuid) ++{ ++ return raft_server_find(&raft->add_servers, uuid); ++} ++ + static char * + raft_make_address_passive(const char *address_) + { +@@ -692,8 +702,8 @@ static void + raft_set_servers(struct raft *raft, const struct hmap *new_servers, + enum vlog_level level) + { +- struct raft_server *s, *next; +- HMAP_FOR_EACH_SAFE (s, next, hmap_node, &raft->servers) { ++ struct raft_server *s; ++ HMAP_FOR_EACH_SAFE (s, hmap_node, &raft->servers) { + if (!raft_server_find(new_servers, &s->sid)) { + ovs_assert(s != raft->remove_server); + +@@ -703,7 +713,7 @@ raft_set_servers(struct raft *raft, const struct hmap *new_servers, + } + } + +- HMAP_FOR_EACH_SAFE (s, next, hmap_node, new_servers) { ++ HMAP_FOR_EACH_SAFE (s, hmap_node, new_servers) { + if (!raft_find_server(raft, &s->sid)) { + VLOG(level, "server %s added to configuration", s->nickname); + +@@ -1376,8 +1386,8 @@ raft_close__(struct raft *raft) + raft->remove_server = NULL; + } + +- struct raft_conn *conn, *next; +- LIST_FOR_EACH_SAFE (conn, next, list_node, &raft->conns) { ++ struct raft_conn *conn; ++ LIST_FOR_EACH_SAFE (conn, list_node, &raft->conns) { + raft_conn_close(conn); + } + } +@@ -1713,8 +1723,8 @@ raft_waiters_run(struct raft *raft) + } + + uint64_t cur = ovsdb_log_commit_progress(raft->log); +- struct raft_waiter *w, *next; +- LIST_FOR_EACH_SAFE (w, next, list_node, &raft->waiters) { ++ struct raft_waiter *w; ++ LIST_FOR_EACH_SAFE (w, list_node, &raft->waiters) { + if (cur < w->commit_ticket) { + break; + } +@@ -1736,8 +1746,8 @@ raft_waiters_wait(struct raft *raft) + static void + raft_waiters_destroy(struct raft *raft) + { +- struct raft_waiter *w, *next; +- LIST_FOR_EACH_SAFE (w, next, list_node, &raft->waiters) { ++ struct raft_waiter *w; ++ LIST_FOR_EACH_SAFE (w, list_node, &raft->waiters) { + raft_waiter_destroy(w); + } + } +@@ -1867,6 +1877,8 @@ raft_open_conn(struct raft *raft, const char *address, const struct uuid *sid) + static void + raft_conn_close(struct raft_conn *conn) + { ++ VLOG_DBG("closing connection to server %s (%s)", ++ conn->nickname, jsonrpc_session_get_name(conn->js)); + jsonrpc_session_close(conn->js); + ovs_list_remove(&conn->list_node); + free(conn->nickname); +@@ -1921,6 +1933,13 @@ raft_run(struct raft *raft) + return; + } + ++ if (failure_test == FT_TRANSFER_LEADERSHIP) { ++ /* Using this function as it conveniently implements all we need and ++ * snapshotting is the main test scenario for leadership transfer. */ ++ raft_notify_snapshot_recommended(raft); ++ failure_test = FT_NO_TEST; ++ } ++ + raft_waiters_run(raft); + + if (!raft->listener && time_msec() >= raft->listen_backoff) { +@@ -1957,16 +1976,29 @@ raft_run(struct raft *raft) + } + + /* Close unneeded sessions. */ +- struct raft_conn *next; +- LIST_FOR_EACH_SAFE (conn, next, list_node, &raft->conns) { ++ struct raft_server *server; ++ LIST_FOR_EACH_SAFE (conn, list_node, &raft->conns) { + if (!raft_conn_should_stay_open(raft, conn)) { ++ server = raft_find_new_server(raft, &conn->sid); ++ if (server) { ++ /* We only have one incoming connection from joining servers, ++ * so if it's closed, we need to destroy the record about the ++ * server. This way the process can be started over on the ++ * next join request. */ ++ static struct vlog_rate_limit rl = VLOG_RATE_LIMIT_INIT(5, 5); ++ VLOG_INFO_RL(&rl, "cluster "CID_FMT": server %s (%s) " ++ "disconnected while joining", ++ CID_ARGS(&raft->cid), ++ server->nickname, server->address); ++ hmap_remove(&raft->add_servers, &server->hmap_node); ++ raft_server_destroy(server); ++ } + raft->n_disconnections++; + raft_conn_close(conn); + } + } + + /* Open needed sessions. */ +- struct raft_server *server; + HMAP_FOR_EACH (server, hmap_node, &raft->servers) { + raft_open_conn(raft, server->address, &server->sid); + } +@@ -2039,11 +2071,18 @@ raft_run(struct raft *raft) + * commands becomes new leader: the pending commands can still complete + * if the crashed leader has replicated the transactions to majority of + * followers before it crashed. */ +- struct raft_command *cmd, *next_cmd; +- HMAP_FOR_EACH_SAFE (cmd, next_cmd, hmap_node, &raft->commands) { ++ struct raft_command *cmd; ++ HMAP_FOR_EACH_SAFE (cmd, hmap_node, &raft->commands) { + if (cmd->timestamp + && now - cmd->timestamp > raft->election_timer * 2) { +- raft_command_complete(raft, cmd, RAFT_CMD_TIMEOUT); ++ if (cmd->index && raft->role != RAFT_LEADER) { ++ /* This server lost leadership and command didn't complete ++ * in time. Likely, it wasn't replicated to the majority ++ * of servers before losing the leadership. */ ++ raft_command_complete(raft, cmd, RAFT_CMD_LOST_LEADERSHIP); ++ } else { ++ raft_command_complete(raft, cmd, RAFT_CMD_TIMEOUT); ++ } + } + } + raft_reset_ping_timer(raft); +@@ -2235,6 +2274,9 @@ raft_command_initiate(struct raft *raft, + if (failure_test == FT_CRASH_AFTER_SEND_APPEND_REQ) { + ovs_fatal(0, "Raft test: crash after sending append_request."); + } ++ if (failure_test == FT_TRANSFER_LEADERSHIP_AFTER_SEND_APPEND_REQ) { ++ failure_test = FT_TRANSFER_LEADERSHIP; ++ } + raft_reset_ping_timer(raft); + + return cmd; +@@ -2243,8 +2285,12 @@ raft_command_initiate(struct raft *raft, + static void + log_all_commands(struct raft *raft) + { +- struct raft_command *cmd, *next; +- HMAP_FOR_EACH_SAFE (cmd, next, hmap_node, &raft->commands) { ++ if (!VLOG_IS_DBG_ENABLED()) { ++ return; ++ } ++ ++ struct raft_command *cmd; ++ HMAP_FOR_EACH (cmd, hmap_node, &raft->commands) { + VLOG_DBG("raft command eid: "UUID_FMT, UUID_ARGS(&cmd->eid)); + } + } +@@ -2398,8 +2444,8 @@ raft_command_complete(struct raft *raft, + static void + raft_complete_all_commands(struct raft *raft, enum raft_command_status status) + { +- struct raft_command *cmd, *next; +- HMAP_FOR_EACH_SAFE (cmd, next, hmap_node, &raft->commands) { ++ struct raft_command *cmd; ++ HMAP_FOR_EACH_SAFE (cmd, hmap_node, &raft->commands) { + raft_command_complete(raft, cmd, status); + } + } +@@ -2601,7 +2647,13 @@ raft_become_follower(struct raft *raft) + * configuration is already part of the log. Possibly the configuration + * log entry will not be committed, but until we know that we must use the + * new configuration. Our AppendEntries processing will properly update +- * the server configuration later, if necessary. */ ++ * the server configuration later, if necessary. ++ * ++ * Also we do not complete commands here, as they can still be completed ++ * if their log entries have already been replicated to other servers. ++ * If the entries were actually committed according to the new leader, our ++ * AppendEntries processing will complete the corresponding commands. ++ */ + struct raft_server *s; + HMAP_FOR_EACH (s, hmap_node, &raft->add_servers) { + raft_send_add_server_reply__(raft, &s->sid, s->address, false, +@@ -2615,8 +2667,6 @@ raft_become_follower(struct raft *raft) + raft_server_destroy(raft->remove_server); + raft->remove_server = NULL; + } +- +- raft_complete_all_commands(raft, RAFT_CMD_LOST_LEADERSHIP); + } + + static void +@@ -2865,61 +2915,56 @@ raft_update_commit_index(struct raft *raft, uint64_t new_commit_index) + return false; + } + +- if (raft->role == RAFT_LEADER) { +- while (raft->commit_index < new_commit_index) { +- uint64_t index = ++raft->commit_index; +- const struct raft_entry *e = raft_get_entry(raft, index); +- if (raft_entry_has_data(e)) { +- struct raft_command *cmd +- = raft_find_command_by_eid(raft, &e->eid); +- if (cmd) { +- if (!cmd->index) { +- VLOG_DBG("Command completed after role change from" +- " follower to leader "UUID_FMT, +- UUID_ARGS(&e->eid)); +- cmd->index = index; +- } +- raft_command_complete(raft, cmd, RAFT_CMD_SUCCESS); ++ while (raft->commit_index < new_commit_index) { ++ uint64_t index = ++raft->commit_index; ++ const struct raft_entry *e = raft_get_entry(raft, index); ++ ++ if (raft_entry_has_data(e)) { ++ struct raft_command *cmd = raft_find_command_by_eid(raft, &e->eid); ++ static struct vlog_rate_limit rl = VLOG_RATE_LIMIT_INIT(5, 5); ++ ++ if (cmd) { ++ if (!cmd->index && raft->role == RAFT_LEADER) { ++ VLOG_INFO_RL(&rl, ++ "command completed after role change from " ++ "follower to leader (eid: "UUID_FMT", " ++ "commit index: %"PRIu64")", UUID_ARGS(&e->eid), index); ++ } else if (!cmd->index && raft->role != RAFT_LEADER) { ++ /* This can happen when leader fail-over before sending ++ * execute_command_reply. */ ++ VLOG_INFO_RL(&rl, ++ "command completed without reply (eid: "UUID_FMT", " ++ "commit index: %"PRIu64")", UUID_ARGS(&e->eid), index); ++ } else if (cmd->index && raft->role != RAFT_LEADER) { ++ /* This can happen if current server lost leadership after ++ * sending append requests to the majority of servers, but ++ * before receiving majority of append replies. */ ++ VLOG_INFO_RL(&rl, ++ "command completed after role change from " ++ "leader to follower (eid: "UUID_FMT", " ++ "commit index: %"PRIu64")", UUID_ARGS(&e->eid), index); ++ /* Clearing 'sid' to avoid sending cmd execution reply. */ ++ cmd->sid = UUID_ZERO; ++ } else { ++ /* (cmd->index && raft->role == RAFT_LEADER) ++ * Normal command completion on a leader. */ + } +- } +- if (e->election_timer) { +- VLOG_INFO("Election timer changed from %"PRIu64" to %"PRIu64, +- raft->election_timer, e->election_timer); +- raft->election_timer = e->election_timer; +- raft->election_timer_new = 0; +- raft_update_probe_intervals(raft); +- } +- if (e->servers) { +- /* raft_run_reconfigure() can write a new Raft entry, which can +- * reallocate raft->entries, which would invalidate 'e', so +- * this case must be last, after the one for 'e->data'. */ +- raft_run_reconfigure(raft); ++ cmd->index = index; ++ raft_command_complete(raft, cmd, RAFT_CMD_SUCCESS); + } + } +- } else { +- while (raft->commit_index < new_commit_index) { +- uint64_t index = ++raft->commit_index; +- const struct raft_entry *e = raft_get_entry(raft, index); +- if (e->election_timer) { +- VLOG_INFO("Election timer changed from %"PRIu64" to %"PRIu64, +- raft->election_timer, e->election_timer); +- raft->election_timer = e->election_timer; +- raft_update_probe_intervals(raft); +- } ++ if (e->election_timer) { ++ VLOG_INFO("Election timer changed from %"PRIu64" to %"PRIu64, ++ raft->election_timer, e->election_timer); ++ raft->election_timer = e->election_timer; ++ raft->election_timer_new = 0; ++ raft_update_probe_intervals(raft); + } +- /* Check if any pending command can be completed, and complete it. +- * This can happen when leader fail-over before sending +- * execute_command_reply. */ +- const struct uuid *eid = raft_get_eid(raft, new_commit_index); +- struct raft_command *cmd = raft_find_command_by_eid(raft, eid); +- if (cmd) { +- static struct vlog_rate_limit rl = VLOG_RATE_LIMIT_INIT(5, 5); +- VLOG_INFO_RL(&rl, +- "Command completed without reply (eid: "UUID_FMT", " +- "commit index: %"PRIu64")", +- UUID_ARGS(eid), new_commit_index); +- cmd->index = new_commit_index; +- raft_command_complete(raft, cmd, RAFT_CMD_SUCCESS); ++ if (e->servers && raft->role == RAFT_LEADER) { ++ /* raft_run_reconfigure() can write a new Raft entry, which can ++ * reallocate raft->entries, which would invalidate 'e', so ++ * this case must be last, after the one for 'e->data'. */ ++ raft_run_reconfigure(raft); + } + } + +@@ -3354,12 +3399,6 @@ raft_find_peer(struct raft *raft, const struct uuid *uuid) + return s && !uuid_equals(&raft->sid, &s->sid) ? s : NULL; + } + +-static struct raft_server * +-raft_find_new_server(struct raft *raft, const struct uuid *uuid) +-{ +- return raft_server_find(&raft->add_servers, uuid); +-} +- + /* Figure 3.1: "If there exists an N such that N > commitIndex, a + * majority of matchIndex[i] >= N, and log[N].term == currentTerm, set + * commitIndex = N (sections 3.5 and 3.6)." */ +@@ -4000,7 +4039,7 @@ raft_write_snapshot(struct raft *raft, struct ovsdb_log *log, + if (error) { + return error; + } +- ovsdb_log_mark_base(raft->log); ++ ovsdb_log_mark_base(log); + + /* Write log records. */ + for (uint64_t index = new_log_start; index < raft->log_end; index++) { +@@ -4142,6 +4181,10 @@ static void + raft_handle_install_snapshot_request( + struct raft *raft, const struct raft_install_snapshot_request *rq) + { ++ if (failure_test == FT_CRASH_BEFORE_SEND_SNAPSHOT_REP) { ++ ovs_fatal(0, "Raft test: crash before sending install_snapshot_reply"); ++ } ++ + if (raft_handle_install_snapshot_request__(raft, rq)) { + union raft_rpc rpy = { + .install_snapshot_reply = { +@@ -4940,6 +4983,8 @@ raft_unixctl_failure_test(struct unixctl_conn *conn OVS_UNUSED, + failure_test = FT_CRASH_AFTER_SEND_EXEC_REQ; + } else if (!strcmp(test, "crash-after-receiving-append-request-update")) { + failure_test = FT_CRASH_AFTER_RECV_APPEND_REQ_UPDATE; ++ } else if (!strcmp(test, "crash-before-sending-install-snapshot-reply")) { ++ failure_test = FT_CRASH_BEFORE_SEND_SNAPSHOT_REP; + } else if (!strcmp(test, "delay-election")) { + failure_test = FT_DELAY_ELECTION; + struct raft *raft; +@@ -4952,6 +4997,11 @@ raft_unixctl_failure_test(struct unixctl_conn *conn OVS_UNUSED, + failure_test = FT_DONT_SEND_VOTE_REQUEST; + } else if (!strcmp(test, "stop-raft-rpc")) { + failure_test = FT_STOP_RAFT_RPC; ++ } else if (!strcmp(test, ++ "transfer-leadership-after-sending-append-request")) { ++ failure_test = FT_TRANSFER_LEADERSHIP_AFTER_SEND_APPEND_REQ; ++ } else if (!strcmp(test, "transfer-leadership")) { ++ failure_test = FT_TRANSFER_LEADERSHIP; + } else if (!strcmp(test, "clear")) { + failure_test = FT_NO_TEST; + unixctl_command_reply(conn, "test dismissed"); +diff --git a/ovsdb/relay.c b/ovsdb/relay.c +index ef0e44d340..122ee8c52f 100644 +--- a/ovsdb/relay.c ++++ b/ovsdb/relay.c +@@ -269,9 +269,9 @@ ovsdb_relay_clear(struct ovsdb *db) + + SHASH_FOR_EACH (table_node, &db->tables) { + struct ovsdb_table *table = table_node->data; +- struct ovsdb_row *row, *next; ++ struct ovsdb_row *row; + +- HMAP_FOR_EACH_SAFE (row, next, hmap_node, &table->rows) { ++ HMAP_FOR_EACH_SAFE (row, hmap_node, &table->rows) { + ovsdb_txn_row_delete(txn, row); + } + } +diff --git a/ovsdb/replication.c b/ovsdb/replication.c +index d8b56d8131..477c69d701 100644 +--- a/ovsdb/replication.c ++++ b/ovsdb/replication.c +@@ -549,8 +549,8 @@ reset_database(struct ovsdb *db) + /* Delete all rows if the table is not excluded. */ + if (!excluded_tables_find(db->schema->name, table_node->name)) { + struct ovsdb_table *table = table_node->data; +- struct ovsdb_row *row, *next; +- HMAP_FOR_EACH_SAFE (row, next, hmap_node, &table->rows) { ++ struct ovsdb_row *row; ++ HMAP_FOR_EACH_SAFE (row, hmap_node, &table->rows) { + ovsdb_txn_row_delete(txn, row); + } + } +@@ -769,9 +769,9 @@ replication_dbs_destroy(void) + return; + } + +- struct shash_node *node, *next; ++ struct shash_node *node; + +- SHASH_FOR_EACH_SAFE (node, next, replication_dbs) { ++ SHASH_FOR_EACH_SAFE (node, replication_dbs) { + hmap_remove(&replication_dbs->map, &node->node); + struct replication_db *rdb = node->data; + if (rdb->active_db_schema) { +diff --git a/ovsdb/table.c b/ovsdb/table.c +index 455a3663fe..2184701ec1 100644 +--- a/ovsdb/table.c ++++ b/ovsdb/table.c +@@ -309,10 +309,10 @@ void + ovsdb_table_destroy(struct ovsdb_table *table) + { + if (table) { +- struct ovsdb_row *row, *next; ++ struct ovsdb_row *row; + size_t i; + +- HMAP_FOR_EACH_SAFE (row, next, hmap_node, &table->rows) { ++ HMAP_FOR_EACH_SAFE (row, hmap_node, &table->rows) { + ovsdb_row_destroy(row); + } + hmap_destroy(&table->rows); +diff --git a/ovsdb/transaction-forward.c b/ovsdb/transaction-forward.c +index d15f2f1d6d..963e937957 100644 +--- a/ovsdb/transaction-forward.c ++++ b/ovsdb/transaction-forward.c +@@ -126,10 +126,10 @@ ovsdb_txn_forward_steal_reply(struct ovsdb_txn_forward *txn_fwd) + void + ovsdb_txn_forward_run(struct ovsdb *db, struct ovsdb_cs *cs) + { +- struct ovsdb_txn_forward *t, *next; ++ struct ovsdb_txn_forward *t; + + /* Send all transactions that needs to be forwarded. */ +- LIST_FOR_EACH_SAFE (t, next, new_node, &db->txn_forward_new) { ++ LIST_FOR_EACH_SAFE (t, new_node, &db->txn_forward_new) { + if (!ovsdb_cs_may_send_transaction(cs)) { + break; + } +@@ -167,9 +167,9 @@ ovsdb_txn_forward_cancel(struct ovsdb *db, struct ovsdb_txn_forward *txn_fwd) + void + ovsdb_txn_forward_cancel_all(struct ovsdb *db, bool sent_only) + { +- struct ovsdb_txn_forward *t, *next; ++ struct ovsdb_txn_forward *t; + +- HMAP_FOR_EACH_SAFE (t, next, sent_node, &db->txn_forward_sent) { ++ HMAP_FOR_EACH_SAFE (t, sent_node, &db->txn_forward_sent) { + ovsdb_txn_forward_cancel(db, t); + } + +@@ -177,7 +177,7 @@ ovsdb_txn_forward_cancel_all(struct ovsdb *db, bool sent_only) + return; + } + +- LIST_FOR_EACH_SAFE (t, next, new_node, &db->txn_forward_new) { ++ LIST_FOR_EACH_SAFE (t, new_node, &db->txn_forward_new) { + ovsdb_txn_forward_cancel(db, t); + } + } +diff --git a/ovsdb/transaction.c b/ovsdb/transaction.c +index db86d847c3..3a6ddfa1df 100644 +--- a/ovsdb/transaction.c ++++ b/ovsdb/transaction.c +@@ -159,15 +159,15 @@ ovsdb_txn_row_abort(struct ovsdb_txn *txn OVS_UNUSED, + hmap_replace(&new->table->rows, &new->hmap_node, &old->hmap_node); + } + +- struct ovsdb_weak_ref *weak, *next; +- LIST_FOR_EACH_SAFE (weak, next, src_node, &txn_row->deleted_refs) { ++ struct ovsdb_weak_ref *weak; ++ LIST_FOR_EACH_SAFE (weak, src_node, &txn_row->deleted_refs) { + ovs_list_remove(&weak->src_node); + ovs_list_init(&weak->src_node); + if (hmap_node_is_null(&weak->dst_node)) { + ovsdb_weak_ref_destroy(weak); + } + } +- LIST_FOR_EACH_SAFE (weak, next, src_node, &txn_row->added_refs) { ++ LIST_FOR_EACH_SAFE (weak, src_node, &txn_row->added_refs) { + ovs_list_remove(&weak->src_node); + ovs_list_init(&weak->src_node); + if (hmap_node_is_null(&weak->dst_node)) { +@@ -508,11 +508,11 @@ static struct ovsdb_error * + ovsdb_txn_update_weak_refs(struct ovsdb_txn *txn OVS_UNUSED, + struct ovsdb_txn_row *txn_row) + { +- struct ovsdb_weak_ref *weak, *next, *dst_weak; ++ struct ovsdb_weak_ref *weak, *dst_weak; + struct ovsdb_row *dst_row; + + /* Find and clean up deleted references from destination rows. */ +- LIST_FOR_EACH_SAFE (weak, next, src_node, &txn_row->deleted_refs) { ++ LIST_FOR_EACH_SAFE (weak, src_node, &txn_row->deleted_refs) { + dst_row = CONST_CAST(struct ovsdb_row *, + ovsdb_table_get_row(weak->dst_table, &weak->dst)); + if (dst_row) { +@@ -529,7 +529,7 @@ ovsdb_txn_update_weak_refs(struct ovsdb_txn *txn OVS_UNUSED, + } + + /* Insert the weak references added in the new version of the row. */ +- LIST_FOR_EACH_SAFE (weak, next, src_node, &txn_row->added_refs) { ++ LIST_FOR_EACH_SAFE (weak, src_node, &txn_row->added_refs) { + dst_row = CONST_CAST(struct ovsdb_row *, + ovsdb_table_get_row(weak->dst_table, &weak->dst)); + +@@ -597,7 +597,7 @@ find_and_add_weak_ref(struct ovsdb_txn_row *txn_row, + static struct ovsdb_error * OVS_WARN_UNUSED_RESULT + assess_weak_refs(struct ovsdb_txn *txn, struct ovsdb_txn_row *txn_row) + { +- struct ovsdb_weak_ref *weak, *next; ++ struct ovsdb_weak_ref *weak; + struct ovsdb_table *table; + struct shash_node *node; + +@@ -642,7 +642,7 @@ assess_weak_refs(struct ovsdb_txn *txn, struct ovsdb_txn_row *txn_row) + + /* Collecting all key-value pairs that references deleted rows. */ + ovsdb_datum_init_empty(&deleted_refs); +- LIST_FOR_EACH_SAFE (weak, next, src_node, &txn_row->deleted_refs) { ++ LIST_FOR_EACH_SAFE (weak, src_node, &txn_row->deleted_refs) { + if (column->index == weak->column_idx) { + ovsdb_datum_add_unsafe(&deleted_refs, &weak->key, &weak->value, + &column->type, NULL); +@@ -1094,10 +1094,10 @@ static void + ovsdb_txn_destroy_cloned(struct ovsdb_txn *txn) + { + ovs_assert(!txn->db); +- struct ovsdb_txn_table *t, *next_txn_table; +- LIST_FOR_EACH_SAFE (t, next_txn_table, node, &txn->txn_tables) { +- struct ovsdb_txn_row *r, *next_txn_row; +- HMAP_FOR_EACH_SAFE (r, next_txn_row, hmap_node, &t->txn_rows) { ++ struct ovsdb_txn_table *t; ++ LIST_FOR_EACH_SAFE (t, node, &txn->txn_tables) { ++ struct ovsdb_txn_row *r; ++ HMAP_FOR_EACH_SAFE (r, hmap_node, &t->txn_rows) { + if (r->old) { + ovsdb_row_destroy(r->old); + } +@@ -1549,19 +1549,19 @@ for_each_txn_row(struct ovsdb_txn *txn, + serial++; + + do { +- struct ovsdb_txn_table *t, *next_txn_table; ++ struct ovsdb_txn_table *t; + + any_work = false; +- LIST_FOR_EACH_SAFE (t, next_txn_table, node, &txn->txn_tables) { ++ LIST_FOR_EACH_SAFE (t, node, &txn->txn_tables) { + if (t->serial != serial) { + t->serial = serial; + t->n_processed = 0; + } + + while (t->n_processed < hmap_count(&t->txn_rows)) { +- struct ovsdb_txn_row *r, *next_txn_row; ++ struct ovsdb_txn_row *r; + +- HMAP_FOR_EACH_SAFE (r, next_txn_row, hmap_node, &t->txn_rows) { ++ HMAP_FOR_EACH_SAFE (r, hmap_node, &t->txn_rows) { + if (r->serial != serial) { + struct ovsdb_error *error; + +@@ -1629,8 +1629,8 @@ ovsdb_txn_history_destroy(struct ovsdb *db) + return; + } + +- struct ovsdb_txn_history_node *txn_h_node, *next; +- LIST_FOR_EACH_SAFE (txn_h_node, next, node, &db->txn_history) { ++ struct ovsdb_txn_history_node *txn_h_node; ++ LIST_FOR_EACH_SAFE (txn_h_node, node, &db->txn_history) { + ovs_list_remove(&txn_h_node->node); + ovsdb_txn_destroy_cloned(txn_h_node->txn); + free(txn_h_node); +diff --git a/ovsdb/trigger.c b/ovsdb/trigger.c +index 726c138bf0..7d3003bca3 100644 +--- a/ovsdb/trigger.c ++++ b/ovsdb/trigger.c +@@ -146,14 +146,14 @@ ovsdb_trigger_prereplace_db(struct ovsdb_trigger *trigger) + bool + ovsdb_trigger_run(struct ovsdb *db, long long int now) + { +- struct ovsdb_trigger *t, *next; ++ struct ovsdb_trigger *t; + + bool run_triggers = db->run_triggers; + db->run_triggers_now = db->run_triggers = false; + + bool disconnect_all = false; + +- LIST_FOR_EACH_SAFE (t, next, node, &db->triggers) { ++ LIST_FOR_EACH_SAFE (t, node, &db->triggers) { + if (run_triggers + || now - t->created >= t->timeout_msec + || t->progress || t->txn_forward) { +diff --git a/python/ovs/_json.c b/python/ovs/_json.c +index ef7bb4b8ee..c36a140a8e 100644 +--- a/python/ovs/_json.c ++++ b/python/ovs/_json.c +@@ -2,10 +2,6 @@ + #include + #include "structmember.h" + +-#if PY_MAJOR_VERSION >= 3 +-#define IS_PY3K +-#endif +- + typedef struct { + PyObject_HEAD + struct json_parser *_parser; +@@ -54,7 +50,7 @@ Parser_feed(json_ParserObject * self, PyObject * args) + Py_ssize_t input_sz; + PyObject *input; + size_t rd; +- char *input_str; ++ const char *input_str; + + if (self->_parser == NULL) { + return NULL; +@@ -63,21 +59,13 @@ Parser_feed(json_ParserObject * self, PyObject * args) + if (!PyArg_UnpackTuple(args, "input", 1, 1, &input)) { + return NULL; + } +-#ifdef IS_PY3K + if ((input_str = PyUnicode_AsUTF8AndSize(input, &input_sz)) == NULL) { +-#else +- if (PyString_AsStringAndSize(input, &input_str, &input_sz) < 0) { +-#endif + return NULL; + } + + rd = json_parser_feed(self->_parser, input_str, (size_t) input_sz); + +-#ifdef IS_PY3K + return PyLong_FromSize_t(rd); +-#else +- return PyInt_FromSize_t(rd); +-#endif + } + + static PyObject * +@@ -123,7 +111,7 @@ json_to_python(struct json *json) + return dict; + } + case JSON_ARRAY:{ +- int i; ++ size_t i; + PyObject *arr = PyList_New(json->array.n); + + if (arr == NULL) { +@@ -144,11 +132,7 @@ json_to_python(struct json *json) + return PyFloat_FromDouble(json->real); + } /* fall through to treat 0 as int */ + case JSON_INTEGER: +-#ifdef IS_PY3K + return PyLong_FromLong((long) json->integer); +-#else +- return PyInt_FromLong((long) json->integer); +-#endif + + case JSON_STRING: + return PyUnicode_FromString(json->string); +@@ -225,7 +209,6 @@ static PyTypeObject json_ParserType = { + Parser_new, /* tp_new */ + }; + +-#ifdef IS_PY3K + static struct PyModuleDef moduledef = { + PyModuleDef_HEAD_INIT, + "ovs._json", /* m_name */ +@@ -238,32 +221,25 @@ static struct PyModuleDef moduledef = { + 0, /* m_free */ + }; + +-#define INITERROR return NULL +-#else /* !IS_PY3K */ +-#define INITERROR return +-#endif +- + PyMODINIT_FUNC +-#ifdef IS_PY3K + PyInit__json(void) +-#else +-init_json(void) +-#endif + { + PyObject *m; + + if (PyType_Ready(&json_ParserType) < 0) { +- INITERROR; ++ return NULL; + } +-#ifdef IS_PY3K ++ + m = PyModule_Create(&moduledef); +-#else +- m = Py_InitModule3("ovs._json", NULL, "OVS JSON Parser module"); +-#endif ++ if (!m) { ++ return NULL; ++ } + + Py_INCREF(&json_ParserType); +- PyModule_AddObject(m, "Parser", (PyObject *) & json_ParserType); +-#ifdef IS_PY3K ++ if (PyModule_AddObject(m, "Parser", (PyObject *) &json_ParserType) < 0) { ++ Py_DECREF(&json_ParserType); ++ Py_DECREF(m); ++ return NULL; ++ } + return m; +-#endif + } +diff --git a/python/ovs/db/idl.py b/python/ovs/db/idl.py +index 4ecdcaa197..8f13d1f55a 100644 +--- a/python/ovs/db/idl.py ++++ b/python/ovs/db/idl.py +@@ -140,6 +140,47 @@ class ConditionState(object): + return False + + ++class IdlTable(object): ++ def __init__(self, idl, table): ++ assert isinstance(table, ovs.db.schema.TableSchema) ++ self._table = table ++ self.need_table = False ++ self.rows = custom_index.IndexedRows(self) ++ self.idl = idl ++ self._condition_state = ConditionState() ++ self.columns = {k: IdlColumn(v) for k, v in table.columns.items()} ++ ++ def __getattr__(self, attr): ++ return getattr(self._table, attr) ++ ++ @property ++ def condition_state(self): ++ # read-only, no setter ++ return self._condition_state ++ ++ @property ++ def condition(self): ++ return self.condition_state.latest ++ ++ @condition.setter ++ def condition(self, condition): ++ assert isinstance(condition, list) ++ self.idl.cond_change(self.name, condition) ++ ++ @classmethod ++ def schema_tables(cls, idl, schema): ++ return {k: cls(idl, v) for k, v in schema.tables.items()} ++ ++ ++class IdlColumn(object): ++ def __init__(self, column): ++ self._column = column ++ self.alert = True ++ ++ def __getattr__(self, attr): ++ return getattr(self._column, attr) ++ ++ + class Idl(object): + """Open vSwitch Database Interface Definition Language (OVSDB IDL). + +@@ -241,7 +282,7 @@ class Idl(object): + assert isinstance(schema_helper, SchemaHelper) + schema = schema_helper.get_idl_schema() + +- self.tables = schema.tables ++ self.tables = IdlTable.schema_tables(self, schema) + self.readonly = schema.readonly + self._db = schema + remotes = self._parse_remotes(remote) +@@ -282,15 +323,6 @@ class Idl(object): + self.cond_changed = False + self.cond_seqno = 0 + +- for table in schema.tables.values(): +- for column in table.columns.values(): +- if not hasattr(column, 'alert'): +- column.alert = True +- table.need_table = False +- table.rows = custom_index.IndexedRows(table) +- table.idl = self +- table.condition = ConditionState() +- + def _parse_remotes(self, remote): + # If remote is - + # "tcp:10.0.0.1:6641,unix:/tmp/db.sock,t,s,tcp:10.0.0.2:6642" +@@ -330,7 +362,7 @@ class Idl(object): + def ack_conditions(self): + """Mark all requested table conditions as acked""" + for table in self.tables.values(): +- table.condition.ack() ++ table.condition_state.ack() + + def sync_conditions(self): + """Synchronize condition state when the FSM is restarted +@@ -356,14 +388,17 @@ class Idl(object): + flushing the local cached DB contents. + """ + ack_all = self.last_id == str(uuid.UUID(int=0)) ++ if ack_all: ++ self.cond_changed = False ++ + for table in self.tables.values(): + if ack_all: +- table.condition.request() +- table.condition.ack() ++ table.condition_state.request() ++ table.condition_state.ack() + else: +- if table.condition.reset(): ++ if table.condition_state.reset(): + self.last_id = str(uuid.UUID(int=0)) +- self.cond_changed = True ++ self.cond_changed = True + + def restart_fsm(self): + # Resync data DB table conditions to avoid missing updated due to +@@ -482,7 +517,7 @@ class Idl(object): + sh.register_table(self._server_db_table) + schema = sh.get_idl_schema() + self._server_db = schema +- self.server_tables = schema.tables ++ self.server_tables = IdlTable.schema_tables(self, schema) + self.__send_server_monitor_request() + except error.Error as e: + vlog.err("%s: error receiving server schema: %s" +@@ -588,10 +623,10 @@ class Idl(object): + for table in self.tables.values(): + # Always use the most recent conditions set by the IDL client when + # requesting monitor_cond_change +- if table.condition.new is not None: ++ if table.condition_state.new is not None: + change_requests[table.name] = [ +- {"where": table.condition.new}] +- table.condition.request() ++ {"where": table.condition_state.new}] ++ table.condition_state.request() + + if not change_requests: + return +@@ -627,19 +662,20 @@ class Idl(object): + cond = [False] + + # Compare the new condition to the last known condition +- if table.condition.latest != cond: +- table.condition.init(cond) ++ if table.condition_state.latest != cond: ++ table.condition_state.init(cond) + self.cond_changed = True + + # New condition will be sent out after all already requested ones + # are acked. +- if table.condition.new: +- any_reqs = any(t.condition.request for t in self.tables.values()) ++ if table.condition_state.new: ++ any_reqs = any(t.condition_state.request ++ for t in self.tables.values()) + return self.cond_seqno + int(any_reqs) + 1 + + # Already requested conditions should be up to date at + # self.cond_seqno + 1 while acked conditions are already up to date +- return self.cond_seqno + int(bool(table.condition.requested)) ++ return self.cond_seqno + int(bool(table.condition_state.requested)) + + def wait(self, poller): + """Arranges for poller.block() to wake up when self.run() has something +@@ -811,8 +847,8 @@ class Idl(object): + columns.append(column) + monitor_request = {"columns": columns} + if method in ("monitor_cond", "monitor_cond_since") and ( +- not ConditionState.is_true(table.condition.acked)): +- monitor_request["where"] = table.condition.acked ++ not ConditionState.is_true(table.condition_state.acked)): ++ monitor_request["where"] = table.condition_state.acked + monitor_requests[table.name] = [monitor_request] + + args = [self._db.name, str(self.uuid), monitor_requests] +@@ -1148,13 +1184,6 @@ class Idl(object): + return True + + +-def _uuid_to_row(atom, base): +- if base.ref_table: +- return base.ref_table.rows.get(atom) +- else: +- return atom +- +- + def _row_to_uuid(value): + if isinstance(value, Row): + return value.uuid +@@ -1268,6 +1297,17 @@ class Row(object): + data=", ".join("{col}={val}".format(col=c, val=getattr(self, c)) + for c in sorted(self._table.columns))) + ++ def _uuid_to_row(self, atom, base): ++ if base.ref_table: ++ try: ++ table = self._idl.tables[base.ref_table.name] ++ except KeyError as e: ++ msg = "Table {} is not registered".format(base.ref_table.name) ++ raise AttributeError(msg) from e ++ return table.rows.get(atom) ++ else: ++ return atom ++ + def __getattr__(self, column_name): + assert self._changes is not None + assert self._mutations is not None +@@ -1309,7 +1349,7 @@ class Row(object): + datum = data.Datum.from_python(column.type, dlist, + _row_to_uuid) + elif column.type.is_map(): +- dmap = datum.to_python(_uuid_to_row) ++ dmap = datum.to_python(self._uuid_to_row) + if inserts is not None: + dmap.update(inserts) + if removes is not None: +@@ -1326,7 +1366,7 @@ class Row(object): + else: + datum = inserts - return datum.to_python(_uuid_to_row) + return datum.to_python(self._uuid_to_row) - def __setattr__(self, column_name, value): - assert self._changes is not None -@@ -1410,7 +1450,7 @@ class Row(object): - if value: - try: - old_value = data.Datum.to_python(self._data[column_name], -- _uuid_to_row) -+ self._uuid_to_row) - except error.Error: - return - if key not in old_value: -diff --git a/python/ovs/reconnect.py b/python/ovs/reconnect.py -index c4c6c87e9f..6b0d023ae3 100644 ---- a/python/ovs/reconnect.py -+++ b/python/ovs/reconnect.py -@@ -44,7 +44,7 @@ class Reconnect(object): - is_connected = False + def __setattr__(self, column_name, value): + assert self._changes is not None +@@ -1410,7 +1450,7 @@ class Row(object): + if value: + try: + old_value = data.Datum.to_python(self._data[column_name], +- _uuid_to_row) ++ self._uuid_to_row) + except error.Error: + return + if key not in old_value: +diff --git a/python/ovs/reconnect.py b/python/ovs/reconnect.py +index c4c6c87e9f..6b0d023ae3 100644 +--- a/python/ovs/reconnect.py ++++ b/python/ovs/reconnect.py +@@ -44,7 +44,7 @@ class Reconnect(object): + is_connected = False + + @staticmethod +- def deadline(fsm): ++ def deadline(fsm, now): + return None + + @staticmethod +@@ -56,7 +56,7 @@ class Reconnect(object): + is_connected = False + + @staticmethod +- def deadline(fsm): ++ def deadline(fsm, now): + return None + + @staticmethod +@@ -68,7 +68,7 @@ class Reconnect(object): + is_connected = False + + @staticmethod +- def deadline(fsm): ++ def deadline(fsm, now): + return fsm.state_entered + fsm.backoff + + @staticmethod +@@ -80,7 +80,7 @@ class Reconnect(object): + is_connected = False + + @staticmethod +- def deadline(fsm): ++ def deadline(fsm, now): + return fsm.state_entered + max(1000, fsm.backoff) + + @staticmethod +@@ -92,13 +92,24 @@ class Reconnect(object): + is_connected = True + + @staticmethod +- def deadline(fsm): ++ def deadline(fsm, now): + if fsm.probe_interval: + base = max(fsm.last_activity, fsm.state_entered) + expiration = base + fsm.probe_interval +- if (fsm.last_receive_attempt is None or ++ if (now < expiration or ++ fsm.last_receive_attempt is None or + fsm.last_receive_attempt >= expiration): ++ # We still have time before the expiration or the time has ++ # already passed and there was no activity. In the first ++ # case we need to wait for the expiration, in the second - ++ # we're already past the deadline. */ + return expiration ++ else: ++ # Time has already passed, but we didn't attempt to receive ++ # anything. We need to wake up and try to receive even if ++ # nothing is pending, so we can update the expiration time ++ # or transition to a different state. ++ return now + 1 + return None + + @staticmethod +@@ -114,12 +125,15 @@ class Reconnect(object): + is_connected = True + + @staticmethod +- def deadline(fsm): ++ def deadline(fsm, now): + if fsm.probe_interval: + expiration = fsm.state_entered + fsm.probe_interval +- if (fsm.last_receive_attempt is None or ++ if (now < expiration or ++ fsm.last_receive_attempt is None or + fsm.last_receive_attempt >= expiration): + return expiration ++ else: ++ return now + 1 + return None + + @staticmethod +@@ -134,7 +148,7 @@ class Reconnect(object): + is_connected = False + + @staticmethod +- def deadline(fsm): ++ def deadline(fsm, now): + return fsm.state_entered + + @staticmethod +@@ -545,7 +559,7 @@ class Reconnect(object): + returned if the "probe interval" is nonzero--see + self.set_probe_interval()).""" + +- deadline = self.state.deadline(self) ++ deadline = self.state.deadline(self, now) + if deadline is not None and now >= deadline: + return self.state.run(self, now) + else: +@@ -562,7 +576,7 @@ class Reconnect(object): + """Returns the number of milliseconds after which self.run() should be + called if nothing else notable happens in the meantime, or None if this + is currently unnecessary.""" +- deadline = self.state.deadline(self) ++ deadline = self.state.deadline(self, now) + if deadline is not None: + remaining = deadline - now + return max(0, remaining) +diff --git a/python/ovs/socket_util.py b/python/ovs/socket_util.py +index 651012bf06..7b41dc44bf 100644 +--- a/python/ovs/socket_util.py ++++ b/python/ovs/socket_util.py +@@ -23,6 +23,11 @@ import ovs.fatal_signal + import ovs.poller + import ovs.vlog + ++try: ++ import ssl ++except ImportError: ++ ssl = None ++ + if sys.platform == 'win32': + import ovs.winutils as winutils + import win32file +@@ -178,7 +183,12 @@ def check_connection_completion(sock): + if revents & ovs.poller.POLLERR or revents & ovs.poller.POLLHUP: + try: + # The following should raise an exception. +- sock.send("\0".encode(), socket.MSG_DONTWAIT) ++ if ssl and isinstance(sock, ssl.SSLSocket): ++ # SSL wrapped socket does not allow ++ # non-zero optional flag. ++ sock.send("\0".encode()) ++ else: ++ sock.send("\0".encode(), socket.MSG_DONTWAIT) + + # (Here's where we end up if it didn't.) + # XXX rate-limit +diff --git a/python/setup.py b/python/setup.py +index cfe01763f3..36ced65089 100644 +--- a/python/setup.py ++++ b/python/setup.py +@@ -12,9 +12,13 @@ + + import sys + +-from distutils.command.build_ext import build_ext +-from distutils.errors import CCompilerError, DistutilsExecError, \ +- DistutilsPlatformError ++from setuptools.command.build_ext import build_ext ++try: ++ from setuptools.errors import CCompilerError, ExecError, PlatformError ++except ImportError: # Needed for setuptools < 59.0 ++ from distutils.errors import CCompilerError ++ from distutils.errors import DistutilsExecError as ExecError ++ from distutils.errors import DistutilsPlatformError as PlatformError + + import setuptools + +@@ -37,7 +41,7 @@ except IOError: + file=sys.stderr) + sys.exit(-1) + +-ext_errors = (CCompilerError, DistutilsExecError, DistutilsPlatformError) ++ext_errors = (CCompilerError, ExecError, PlatformError) + if sys.platform == 'win32': + ext_errors += (IOError, ValueError) + +@@ -53,7 +57,7 @@ class try_build_ext(build_ext): + def run(self): + try: + build_ext.run(self) +- except DistutilsPlatformError: ++ except PlatformError: + raise BuildFailed() + + def build_extension(self, ext): +@@ -102,6 +106,6 @@ except BuildFailed: + print("Retrying the build without the C extension.") + print("*" * 75) + +- del(setup_args['cmdclass']) +- del(setup_args['ext_modules']) ++ del setup_args['cmdclass'] ++ del setup_args['ext_modules'] + setuptools.setup(**setup_args) +diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in +index 16ef1ac3ab..ddde4f7d12 100644 +--- a/rhel/openvswitch-fedora.spec.in ++++ b/rhel/openvswitch-fedora.spec.in +@@ -199,20 +199,6 @@ make install DESTDIR=$RPM_BUILD_ROOT + install -d -m 0755 $RPM_BUILD_ROOT%{_rundir}/openvswitch + install -d -m 0750 $RPM_BUILD_ROOT%{_localstatedir}/log/openvswitch + install -d -m 0755 $RPM_BUILD_ROOT%{_sysconfdir}/openvswitch +-copy_headers() { +- src=$1 +- dst=$RPM_BUILD_ROOT/$2 +- install -d -m 0755 $dst +- install -m 0644 $src/*.h $dst +-} +-copy_headers include %{_includedir}/openvswitch +-copy_headers include/openflow %{_includedir}/openvswitch/openflow +-copy_headers include/openvswitch %{_includedir}/openvswitch/openvswitch +-copy_headers include/sparse %{_includedir}/openvswitch/sparse +-copy_headers include/sparse/arpa %{_includedir}/openvswitch/sparse/arpa +-copy_headers include/sparse/netinet %{_includedir}/openvswitch/sparse/netinet +-copy_headers include/sparse/sys %{_includedir}/openvswitch/sparse/sys +-copy_headers lib %{_includedir}/openvswitch/lib + + %if %{with dpdk} + install -p -D -m 0644 rhel/usr_lib_udev_rules.d_91-vfio.rules \ +@@ -455,6 +441,7 @@ fi + %{_datadir}/openvswitch/scripts/ovs-ctl + %{_datadir}/openvswitch/scripts/ovs-kmod-ctl + %{_datadir}/openvswitch/scripts/ovs-systemd-reload ++%config %{_datadir}/openvswitch/local-config.ovsschema + %config %{_datadir}/openvswitch/vswitch.ovsschema + %config %{_datadir}/openvswitch/vtep.ovsschema + %{_bindir}/ovs-appctl +@@ -476,6 +463,7 @@ fi + %{_mandir}/man1/ovsdb-server.1* + %{_mandir}/man1/ovsdb-tool.1* + %{_mandir}/man5/ovsdb-server.5* ++%{_mandir}/man5/ovsdb.local-config.5* + %{_mandir}/man5/ovs-vswitchd.conf.db.5* + %{_mandir}/man5/ovsdb.5* + %{_mandir}/man5/vtep.5* +diff --git a/rhel/openvswitch.spec.in b/rhel/openvswitch.spec.in +index 220e5c7472..8ee8a99c22 100644 +--- a/rhel/openvswitch.spec.in ++++ b/rhel/openvswitch.spec.in +@@ -110,24 +110,6 @@ install -d -m 0755 $RPM_BUILD_ROOT%{_rundir}/openvswitch + install -d -m 0755 $RPM_BUILD_ROOT%{_localstatedir}/log/openvswitch + install -d -m 0755 $RPM_BUILD_ROOT/var/lib/openvswitch + +-copy_headers() { +- src=$1 +- dst=$RPM_BUILD_ROOT/$2 +- install -d -m 0755 $dst +- install -m 0644 $src/*.h $dst +-} +-copy_headers include %{_includedir}/openvswitch +-copy_headers include/openflow %{_includedir}/openvswitch/openflow +-copy_headers include/openvswitch %{_includedir}/openvswitch/openvswitch +-copy_headers include/sparse %{_includedir}/openvswitch/sparse +-copy_headers include/sparse/arpa %{_includedir}/openvswitch/sparse/arpa +-copy_headers include/sparse/netinet %{_includedir}/openvswitch/sparse/netinet +-copy_headers include/sparse/sys %{_includedir}/openvswitch/sparse/sys +-copy_headers lib %{_includedir}/openvswitch/lib +- +-install -D -m 0644 lib/.libs/libopenvswitch.a \ +- $RPM_BUILD_ROOT/%{_libdir}/libopenvswitch.a +- + %check + %if %{with check} + if make check TESTSUITEFLAGS='%{_smp_mflags}' RECHECK=yes; then :; +@@ -229,6 +211,7 @@ exit 0 + /usr/share/man/man1/ovsdb-client.1.gz + /usr/share/man/man1/ovsdb-server.1.gz + /usr/share/man/man1/ovsdb-tool.1.gz ++/usr/share/man/man5/ovsdb.local-config.5.gz + /usr/share/man/man5/ovsdb-server.5.gz + /usr/share/man/man5/ovs-vswitchd.conf.db.5.gz + %{_mandir}/man5/ovsdb.5* +@@ -262,6 +245,7 @@ exit 0 + /usr/share/openvswitch/scripts/ovs-vtep + /usr/share/openvswitch/scripts/sysconfig.template + /usr/share/openvswitch/scripts/ovs-monitor-ipsec ++/usr/share/openvswitch/local-config.ovsschema + /usr/share/openvswitch/vswitch.ovsschema + /usr/share/openvswitch/vtep.ovsschema + %doc NOTICE README.rst NEWS rhel/README.RHEL.rst +diff --git a/tests/alb.at b/tests/alb.at +index 2bef06f39c..922185d61d 100644 +--- a/tests/alb.at ++++ b/tests/alb.at +@@ -96,6 +96,52 @@ OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "PMD auto load balance + OVS_VSWITCHD_STOP + AT_CLEANUP + ++AT_SETUP([ALB - cross-numa]) ++OVS_VSWITCHD_START([add-port br0 p0 \ ++ -- set Interface p0 type=dummy-pmd options:n_rxq=4 \ ++ -- set Interface p0 options:numa_id=0 \ ++ -- set Open_vSwitch . other_config:pmd-cpu-mask=0x3 \ ++ -- set open_vswitch . other_config:pmd-rxq-assign=group \ ++ -- set open_vswitch . other_config:pmd-rxq-isolate=false \ ++ -- set open_vswitch . other_config:pmd-auto-lb="true" \ ++ -- set open_vswitch . other_config:pmd-auto-lb-load-threshold=0], ++ [], [], [--dummy-numa 1,2,1,2]) ++OVS_WAIT_UNTIL([grep "PMD auto load balance is enabled" ovs-vswitchd.log]) ++AT_CHECK([ovs-appctl vlog/set dpif_netdev:dbg]) ++ ++# no pinned rxqs - cross-numa pmd could change ++get_log_next_line_num ++ovs-appctl time/warp 600000 10000 ++OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "PMD auto load balance performing dry run."]) ++OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "PMD auto load balance detected cross-numa polling"]) ++ ++# all pinned rxqs - cross-numa pmd will not change ++AT_CHECK([ovs-vsctl set Interface p0 other_config:pmd-rxq-affinity='0:0,1:0,2:1,3:1']) ++get_log_next_line_num ++ovs-appctl time/warp 600000 10000 ++OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "PMD auto load balance performing dry run."]) ++OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "Variance improvement 0%."]) ++ ++# mix of pinned (non-isolated) and non-pinned rxqs - cross-numa pmd could change ++AT_CHECK([ovs-vsctl remove Interface p0 other_config pmd-rxq-affinity]) ++AT_CHECK([ovs-vsctl set Interface p0 other_config:pmd-rxq-affinity='0:0,1:0,2:1']) ++get_log_next_line_num ++ovs-appctl time/warp 600000 10000 ++OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "PMD auto load balance performing dry run."]) ++OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "PMD auto load balance detected cross-numa polling"]) ++ ++# mix of pinned (isolated) and non-pinned rxqs - cross-numa pmd could change ++AT_CHECK([ovs-vsctl set Open_vSwitch . other_config:pmd-cpu-mask=0xf]) ++AT_CHECK([ovs-vsctl set Interface p0 options:n_rxq=6]) ++AT_CHECK([ovs-vsctl set Open_vSwitch . other_config:pmd-rxq-isolate=true]) ++get_log_next_line_num ++ovs-appctl time/warp 600000 10000 ++OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "PMD auto load balance performing dry run."]) ++OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "PMD auto load balance detected cross-numa polling"]) ++ ++OVS_VSWITCHD_STOP ++AT_CLEANUP ++ + AT_SETUP([ALB - PMD/RxQ assignment type]) + OVS_VSWITCHD_START([add-port br0 p0 \ + -- set Interface p0 type=dummy-pmd options:n_rxq=3 \ +@@ -197,7 +243,25 @@ get_log_next_line_num + AT_CHECK([ovs-vsctl set open_vswitch . other_config:pmd-auto-lb-rebal-interval="0"]) + CHECK_ALB_PARAM([interval], [1 mins], [+$LINENUM]) + +-# No check for above max as it is only a documented max value and not a hard limit ++# Set new value ++get_log_next_line_num ++AT_CHECK([ovs-vsctl set open_vswitch . other_config:pmd-auto-lb-rebal-interval="100"]) ++CHECK_ALB_PARAM([interval], [100 mins], [+$LINENUM]) ++ ++# Set above max value ++get_log_next_line_num ++AT_CHECK([ovs-vsctl set open_vswitch . other_config:pmd-auto-lb-rebal-interval="20001"]) ++CHECK_ALB_PARAM([interval], [1 mins], [+$LINENUM]) ++ ++# Set new value ++get_log_next_line_num ++AT_CHECK([ovs-vsctl set open_vswitch . other_config:pmd-auto-lb-rebal-interval="1000"]) ++CHECK_ALB_PARAM([interval], [1000 mins], [+$LINENUM]) ++ ++# Set Negative value ++get_log_next_line_num ++AT_CHECK([ovs-vsctl set open_vswitch . other_config:pmd-auto-lb-rebal-interval="-1"]) ++CHECK_ALB_PARAM([interval], [1 mins], [+$LINENUM]) + + OVS_VSWITCHD_STOP + AT_CLEANUP +diff --git a/tests/classifier.at b/tests/classifier.at +index cdcd72c156..f652b59837 100644 +--- a/tests/classifier.at ++++ b/tests/classifier.at +@@ -129,6 +129,31 @@ Datapath actions: 3 + OVS_VSWITCHD_STOP(["/'prefixes' with incompatible field: ipv6_label/d"]) + AT_CLEANUP + ++AT_SETUP([flow classifier - ipv6 ND dependency]) ++OVS_VSWITCHD_START ++add_of_ports br0 1 2 ++AT_DATA([flows.txt], [dnl ++ table=0,priority=100,ipv6,ipv6_src=1000::/10 actions=resubmit(,1) ++ table=0,priority=0 actions=NORMAL ++ table=1,priority=110,ipv6,ipv6_dst=1000::3 actions=resubmit(,2) ++ table=1,priority=100,ipv6,ipv6_dst=1000::4 actions=resubmit(,2) ++ table=1,priority=0 actions=NORMAL ++ table=2,priority=120,icmp6,nw_ttl=255,icmp_type=135,icmp_code=0,nd_target=1000::1 actions=NORMAL ++ table=2,priority=100,tcp actions=NORMAL ++ table=2,priority=100,icmp6 actions=NORMAL ++ table=2,priority=0 actions=NORMAL ++]) ++AT_CHECK([ovs-ofctl add-flows br0 flows.txt]) ++ ++# test ICMPv6 echo request (which should have no nd_target field) ++AT_CHECK([ovs-appctl ofproto/trace br0 "in_port=1,eth_src=f6:d2:b0:19:5e:7b,eth_dst=d2:49:19:91:78:fe,dl_type=0x86dd,ipv6_src=1000::3,ipv6_dst=1000::4,nw_proto=58,icmpv6_type=128,icmpv6_code=0"], [0], [stdout]) ++AT_CHECK([tail -2 stdout], [0], ++ [Megaflow: recirc_id=0,eth,icmp6,in_port=1,dl_src=f6:d2:b0:19:5e:7b,dl_dst=d2:49:19:91:78:fe,ipv6_src=1000::/10,ipv6_dst=1000::4,nw_ttl=0,nw_frag=no ++Datapath actions: 100,2 ++]) ++OVS_VSWITCHD_STOP ++AT_CLEANUP ++ + AT_BANNER([conjunctive match]) + + AT_SETUP([single conjunctive match]) +diff --git a/tests/completion.at b/tests/completion.at +index 00e3a46b8b..b6155af253 100644 +--- a/tests/completion.at ++++ b/tests/completion.at +@@ -351,22 +351,22 @@ OVS_VSWITCHD_START + TMP="$(ovs-vsctl --commands | cut -d',' -f1-2 | tr -d ',[[]]' | tr -s ' ' '\n') + $(ovs-vsctl --options | grep -- '--' | sed -e 's/=.*$/=/g')" + MATCH="$(PREPARE_MATCH_NOSPACE(${TMP}))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "--db=unix:$OVS_RUNDIR/db.sock "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "--db=unix:$OVS_RUNDIR/db.sock "], + [0], [dnl + ${MATCH} + ]) + # complete ovs-vsctl [TAB] +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test ""], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test ""], + [0], [dnl + ${MATCH} + ]) + + # complete on global options. +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "--dry-run "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "--dry-run "], + [0], [dnl + ${MATCH} + ]) +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "--dry-run --pretty "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "--dry-run --pretty "], + [0], [dnl + ${MATCH} + ]) +@@ -374,7 +374,7 @@ ${MATCH} + # complete on local options. + TMP="$(ovs-vsctl --commands | grep -- '--may-exist' | cut -d',' -f1-2 | tr -d ',[[]]' | tr -s ' ' '\n' | grep -v -- '--may-exist')" + MATCH="$(PREPARE_MATCH_SPACE(${TMP}))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "--may-exist "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "--may-exist "], + [0], [dnl + ${MATCH} + ]) +@@ -385,37 +385,37 @@ ${MATCH} + # test !. no following arguments are expanded. + TMP="$(ovsdb-client --no-heading list-tables)" + MATCH="$(PREPARE_MATCH_SPACE(${TMP}))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set "], + [0], [dnl + ${MATCH} + ]) + # test ?. will show completions for both current and following arguments. + ovs-vsctl br-set-external-id br0 bridge-id br0 + MATCH="$(PREPARE_MATCH_SPACE(bridge-id --))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "br-get-external-id br0 "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "br-get-external-id br0 "], + [0], [dnl + ${MATCH} + ]) + # test *. argument with this prefix could be completed for zero or more times. + TMP="$(ovs-vsctl --no-heading --columns=_uuid,name list Bridge | tr -d '\"')" + MATCH="$(PREPARE_MATCH_SPACE(${TMP} --))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "destroy Bridge "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "destroy Bridge "], + [0], [dnl + ${MATCH} + ]) +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "destroy Bridge br0 "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "destroy Bridge br0 "], + [0], [dnl + ${MATCH} + ]) + # test +. the first time, an argument is required, after that, it becomes '*'. + TMP="$(ovsdb-client --no-heading list-columns Open_vSwitch Bridge | awk '/key.*value/ { print $1":"; next } { print $1; next }')" + MATCH="$(PREPARE_MATCH_NOSPACE(${TMP}))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set Bridge br0 "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set Bridge br0 "], + [0], [dnl + ${MATCH} + ]) + MATCH="$(PREPARE_MATCH_NOSPACE(${TMP} --))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set Bridge br0 other_config:random_key=123 "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set Bridge br0 other_config:random_key=123 "], + [0], [dnl + ${MATCH} + ]) +@@ -453,12 +453,12 @@ OVS_VSWITCHD_START( + # + TMP="$(ovsdb-client --no-heading list-tables)" + MATCH="$(PREPARE_MATCH_SPACE(${TMP}))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set "], + [0], [dnl + ${MATCH} + ]) + MATCH="$(PREPARE_MATCH_SPACE(Open_vSwitch))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set Open"], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set Open"], + [0], [dnl + ${MATCH} + ]) +@@ -469,13 +469,13 @@ ${MATCH} + # + TMP="$(ovs-vsctl --no-heading --columns=_uuid list Open_vSwitch | tr -d '\"')" + MATCH="$(PREPARE_MATCH_SPACE(${TMP}))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set Open_vSwitch "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set Open_vSwitch "], + [0], [dnl + ${MATCH} + ]) + TMP="$(ovs-vsctl --no-heading --columns=_uuid,name list Bridge | tr -d '\"')" + MATCH="$(PREPARE_MATCH_SPACE(${TMP}))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set Bridge "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set Bridge "], + [0], [dnl + ${MATCH} + ]) +@@ -486,13 +486,13 @@ ${MATCH} + # + TMP="$(ovs-vsctl list-br)" + MATCH="$(PREPARE_MATCH_SPACE(${TMP}))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "br-to-vlan "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "br-to-vlan "], + [0], [dnl + ${MATCH} + ]) + # this also helps check the '_ovs_vsctl_check_startswith_string'. + MATCH="$(PREPARE_MATCH_SPACE(--weird-br_name))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "br-to-vlan --"], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "br-to-vlan --"], + [0], [dnl + ${MATCH} + ]) +@@ -503,14 +503,14 @@ ${MATCH} + # + TMP="$(ovs-vsctl --no-heading --columns=name list Port | tr -d '\"')" + MATCH="$(PREPARE_MATCH_SPACE(${TMP}))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "port-to-br "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "port-to-br "], + [0], [dnl + ${MATCH} + ]) + # complete on ports in particular bridge. + TMP="$(ovs-vsctl list-ports br0)" + MATCH="$(PREPARE_MATCH_SPACE(${TMP}))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "del-port br0 "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "del-port br0 "], + [0], [dnl + ${MATCH} + ]) +@@ -523,7 +523,7 @@ for br in `ovs-vsctl list-br`; do + TMP="${TMP} $(ovs-vsctl list-ifaces $br)" + done + MATCH="$(PREPARE_MATCH_SPACE(${TMP}))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "iface-to-br "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "iface-to-br "], + [0], [dnl + ${MATCH} + ]) +@@ -533,7 +533,7 @@ ${MATCH} + # test: _ovs_vsctl_complete_bridge_fail_mode + # + MATCH="$(PREPARE_MATCH_SPACE(standalone secure))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set-fail-mode br0 "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set-fail-mode br0 "], + [0], [dnl + ${MATCH} + ]) +@@ -542,25 +542,25 @@ ${MATCH} + # + # test: _ovs_vsctl_complete_key + # +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "br-set-external-id br0 "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "br-set-external-id br0 "], + [0], [dnl + + ]) + # since there is no key added yet, we will only get our own input. + MATCH="$(PREPARE_MATCH_SPACE(test_key))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "br-set-external-id br0 test_key"], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "br-set-external-id br0 test_key"], + [0], [dnl + ${MATCH} + ]) + # now add a key, as we should see it. + ovs-vsctl br-set-external-id br0 bridge-id br0 + MATCH="$(PREPARE_MATCH_SPACE(bridge-id))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "br-set-external-id br0 "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "br-set-external-id br0 "], + [0], [dnl + ${MATCH} + ]) + MATCH="$(PREPARE_MATCH_SPACE(bridge-id --))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "br-get-external-id br0 "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "br-get-external-id br0 "], + [0], [dnl + ${MATCH} + ]) +@@ -571,7 +571,7 @@ ${MATCH} + # + # should just return the user input. + MATCH="$(PREPARE_MATCH_SPACE(test_value --))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "br-set-external-id br0 bridge-id test_value"], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "br-set-external-id br0 bridge-id test_value"], + [0], [dnl + ${MATCH} + ]) +@@ -583,13 +583,13 @@ ${MATCH} + TMP="$(ovsdb-client --no-heading list-columns Open_vSwitch Open_vSwitch | tr -d ':' | cut -d' ' -f1)" + UUID="$(ovs-vsctl --no-heading --columns=_uuid list Open_vSwitch | tr -d ' ')" + MATCH="$(PREPARE_MATCH_SPACE(${TMP}))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "clear Open_vSwitch $UUID "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "clear Open_vSwitch $UUID "], + [0], [dnl + ${MATCH} + ]) + TMP="$(ovsdb-client --no-heading list-columns Open_vSwitch Bridge | tr -d ':' | cut -d' ' -f1)" + MATCH="$(PREPARE_MATCH_SPACE(${TMP}))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "clear Bridge br0 "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "clear Bridge br0 "], + [0], [dnl + ${MATCH} + ]) +@@ -597,7 +597,7 @@ ${MATCH} + # so, with one specified COLUMN 'other_config', it should still complete on + # COLUMNs, plus '--'. + MATCH="$(PREPARE_MATCH_SPACE(${TMP} --))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "clear Bridge br0 other_config "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "clear Bridge br0 other_config "], + [0], [dnl + ${MATCH} + ]) +@@ -608,19 +608,19 @@ ${MATCH} + # + # with no key available, should always get user input. + MATCH="$(PREPARE_MATCH_NOSPACE(random_key))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "add Bridge br0 other_config random_key"], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "add Bridge br0 other_config random_key"], + [0], [dnl + ${MATCH} + ]) + MATCH="$(PREPARE_MATCH_NOSPACE(abc))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "add Bridge br0 other_config random_key=abc"], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "add Bridge br0 other_config random_key=abc"], + [0], [dnl + ${MATCH} + ]) + # now add two random keys. + ovs-vsctl set Bridge br0 other_config:random_key1=abc other_config:random_val1=xyz + MATCH="$(PREPARE_MATCH_NOSPACE(random_key1= random_val1=))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "add Bridge br0 other_config ran"], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "add Bridge br0 other_config ran"], + [0], [dnl + ${MATCH} + ]) +@@ -632,25 +632,25 @@ ${MATCH} + # at first, we should complete on column. + TMP="$(ovsdb-client --no-heading list-columns Open_vSwitch Bridge | awk '/key.*value/ { print $1":"; next } { print $1; next }')" + MATCH="$(PREPARE_MATCH_NOSPACE(${TMP}))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set Bridge br0 "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set Bridge br0 "], + [0], [dnl + ${MATCH} + ]) + MATCH="$(PREPARE_MATCH_NOSPACE(other_config:))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set Bridge br0 other"], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set Bridge br0 other"], + [0], [dnl + ${MATCH} + ]) + # then, with the ':' we should complete on key. + TMP="$(ovs-vsctl --no-heading --columns=other_config list Bridge br0 | tr -d '{\"}' | tr -s ', ' '\n' | cut -d'=' -f1)" + MATCH="$(PREPARE_MATCH_NOSPACE(${TMP}))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set Bridge br0 other_config:"], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set Bridge br0 other_config:"], + [0], [dnl + ${MATCH} + ]) + # finally, if user fill in some value, we should just complete on user input. + MATCH="$(PREPARE_MATCH_NOSPACE(random_val1))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set Bridge br0 other_config:random_val1=12345"], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set Bridge br0 other_config:random_val1=12345"], + [0], [dnl + ${MATCH} + ]) +@@ -661,12 +661,12 @@ ${MATCH} + # + touch private_key certificate + MATCH="$(PREPARE_MATCH_SPACE(private_key))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set-ssl priva"], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set-ssl priva"], + [0], [dnl + ${MATCH} + ]) + MATCH="$(PREPARE_MATCH_SPACE(certificate))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set-ssl private_key cer"], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set-ssl private_key cer"], + [0], [dnl + ${MATCH} + ]) +@@ -676,20 +676,20 @@ ${MATCH} + # test: _ovs_vsctl_complete_target + # + MATCH="$(PREPARE_MATCH_NOSPACE(pssl: ptcp: punix: ssl: tcp: unix:))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set-manager "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set-manager "], + [0], [dnl + ${MATCH} + ]) + # filename completion on unix, punix. + MATCH="$(PREPARE_MATCH_NOSPACE(testsuite.log))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set-manager unix:test"], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set-manager unix:test"], + [0], [dnl + ${MATCH} + ]) + # no completion on other type, just return available types. + # in real environment, bash will not complete on anything. + MATCH="$(PREPARE_MATCH_NOSPACE(pssl: ptcp: punix: tcp: unix:))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set-manager ssl:something"], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set-manager ssl:something"], + [0], [dnl + ${MATCH} + ]) +@@ -699,14 +699,14 @@ ${MATCH} + # test: _ovs_vsctl_complete_new + # + # test 'add-br' +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "add-br "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "add-br "], + [0], [dnl + --- BEGIN MESSAGE + Enter a new bridge: + > ovs-vsctl add-br --- END MESSAGE + ]) + # user input does not change the output. +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "add-br new-br"], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "add-br new-br"], + [0], [dnl + --- BEGIN MESSAGE + Enter a new bridge: +@@ -715,7 +715,7 @@ Enter a new bridge: + # after specifying the new bridge name, we should complete on parent bridge. + TMP="$(ovs-vsctl list-br)" + MATCH="$(PREPARE_MATCH_SPACE(${TMP}))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "add-br new-br "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "add-br new-br "], + [0], [dnl + ${MATCH} + ]) +@@ -724,7 +724,7 @@ ${MATCH} + # of '*COLUMN?:KEY=VALUE'. + TMP="$(ovsdb-client --no-heading list-columns Open_vSwitch Port | awk '/key.*value/ { print $1":"; next } { print $1; next }')" + MATCH="$(PREPARE_MATCH_NOSPACE(${TMP} --))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "add-port br0 new-port "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "add-port br0 new-port "], + [0], [dnl + ${MATCH} + ]) +@@ -736,13 +736,13 @@ ${MATCH} + # after '--', there should be no global options available for completion. + TMP="$(ovs-vsctl --commands | cut -d',' -f1-2 | tr -d ',[[]]' | tr -s ' ' '\n')" + MATCH="$(PREPARE_MATCH_NOSPACE(${TMP}))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "init -- "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "init -- "], + [0], [dnl + ${MATCH} + ]) + TMP="$(ovs-vsctl --no-heading --columns=name,_uuid list Port | tr -d '\"')" + MATCH="$(PREPARE_MATCH_SPACE(${TMP} newp1 newp2))" +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "add-port br0 newp1 -- add-port br1 newp2 -- set Port "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "add-port br0 newp1 -- add-port br1 newp2 -- set Port "], + [0], [dnl + ${MATCH} + ]) +@@ -757,25 +757,25 @@ AT_SKIP_IF([eval 'test ${BASH_VERSINFO[[0]]} -lt 4']) + OVS_VSWITCHD_START + + # complete non-matching command. +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "invalid"], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "invalid"], + [0], [dnl - @staticmethod -- def deadline(fsm): -+ def deadline(fsm, now): - return None + ]) - @staticmethod -@@ -56,7 +56,7 @@ class Reconnect(object): - is_connected = False + # complete after invalid command. +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "invalid argu"], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "invalid argu"], + [0], [dnl - @staticmethod -- def deadline(fsm): -+ def deadline(fsm, now): - return None + ]) - @staticmethod -@@ -68,7 +68,7 @@ class Reconnect(object): - is_connected = False + # complete non-matching end argument. +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set INVALID_"], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set INVALID_"], + [0], [dnl - @staticmethod -- def deadline(fsm): -+ def deadline(fsm, now): - return fsm.state_entered + fsm.backoff + ]) - @staticmethod -@@ -80,7 +80,7 @@ class Reconnect(object): - is_connected = False + # complete after invalid intermediate argument. +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set INVALID_TBL "], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set INVALID_TBL "], + [1], [dnl + --- BEGIN MESSAGE + Cannot complete 'INVALID_TBL' at index 3: +@@ -783,12 +783,12 @@ Cannot complete 'INVALID_TBL' at index 3: - @staticmethod -- def deadline(fsm): -+ def deadline(fsm, now): - return fsm.state_entered + max(1000, fsm.backoff) + # complete ovs-vsctl --db=wrongdb [TAB] + # should return 1 and show nothing. +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "--db=wrongdb"], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "--db=wrongdb"], + [1], []) - @staticmethod -@@ -92,13 +92,24 @@ class Reconnect(object): - is_connected = True + OVS_VSWITCHD_STOP + # delete ovsdb-server and try again. +-AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test ""], ++AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test ""], + [1], []) - @staticmethod -- def deadline(fsm): -+ def deadline(fsm, now): - if fsm.probe_interval: - base = max(fsm.last_activity, fsm.state_entered) - expiration = base + fsm.probe_interval -- if (fsm.last_receive_attempt is None or -+ if (now < expiration or -+ fsm.last_receive_attempt is None or - fsm.last_receive_attempt >= expiration): -+ # We still have time before the expiration or the time has -+ # already passed and there was no activity. In the first -+ # case we need to wait for the expiration, in the second - -+ # we're already past the deadline. */ - return expiration -+ else: -+ # Time has already passed, but we didn't attempt to receive -+ # anything. We need to wake up and try to receive even if -+ # nothing is pending, so we can update the expiration time -+ # or transition to a different state. -+ return now + 1 - return None + AT_CLEANUP +diff --git a/tests/dpif-netdev.at b/tests/dpif-netdev.at +index a79ebdb618..3179e1645d 100644 +--- a/tests/dpif-netdev.at ++++ b/tests/dpif-netdev.at +@@ -439,7 +439,7 @@ recirc_id(0),in_port(1),packet_type(ns=0,id=0),eth_type(0x0800),ipv4(frag=no), a + + # Check for succesfull packet matching with installed offloaded flow. + AT_CHECK([filter_hw_packet_netdev_dummy < ovs-vswitchd.log | strip_xout], [0], [dnl +-p1: packet: ip,vlan_tci=0x0000,dl_src=00:06:07:08:09:0a,dl_dst=00:01:02:03:04:05,nw_src=127.0.0.1,nw_dst=127.0.0.1,nw_proto=0,nw_tos=0,nw_ecn=0,nw_ttl=64 matches with flow: recirc_id=0,eth,ip,vlan_tci=0x0000/0x1fff,nw_frag=no with mark: 1 ++p1: packet: ip,vlan_tci=0x0000,dl_src=00:06:07:08:09:0a,dl_dst=00:01:02:03:04:05,nw_src=127.0.0.1,nw_dst=127.0.0.1,nw_proto=0,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no matches with flow: recirc_id=0,eth,ip,vlan_tci=0x0000/0x1fff,nw_frag=no with mark: 1 + ]) - @staticmethod -@@ -114,12 +125,15 @@ class Reconnect(object): - is_connected = True + ovs-appctl revalidator/wait +@@ -506,7 +506,7 @@ recirc_id(0),in_port(1),packet_type(ns=0,id=0),eth_type(0x8100),vlan(vid=99,pcp= - @staticmethod -- def deadline(fsm): -+ def deadline(fsm, now): - if fsm.probe_interval: - expiration = fsm.state_entered + fsm.probe_interval -- if (fsm.last_receive_attempt is None or -+ if (now < expiration or -+ fsm.last_receive_attempt is None or - fsm.last_receive_attempt >= expiration): - return expiration -+ else: -+ return now + 1 - return None + # Check for succesfull packet matching with installed offloaded flow. + AT_CHECK([filter_hw_packet_netdev_dummy < ovs-vswitchd.log | strip_xout], [0], [dnl +-p1: packet: udp,dl_vlan=99,dl_vlan_pcp=7,vlan_tci1=0x0000,dl_src=00:06:07:08:09:0a,dl_dst=00:01:02:03:04:05,nw_src=127.0.0.1,nw_dst=127.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=81,tp_dst=82 dnl ++p1: packet: udp,dl_vlan=99,dl_vlan_pcp=7,vlan_tci1=0x0000,dl_src=00:06:07:08:09:0a,dl_dst=00:01:02:03:04:05,nw_src=127.0.0.1,nw_dst=127.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=81,tp_dst=82 dnl + matches with flow: recirc_id=0,eth,udp,dl_vlan=99,nw_src=127.0.0.1,nw_frag=no,tp_dst=82 with mark: 1 + ]) - @staticmethod -@@ -134,7 +148,7 @@ class Reconnect(object): - is_connected = False +@@ -530,10 +530,10 @@ p1: flow del: mark: 1 + + # Check that ip address and udp port were correctly modified in output packets. + AT_CHECK([ovs-ofctl parse-pcap p1.pcap], [0], [dnl +-udp,in_port=ANY,dl_vlan=99,dl_vlan_pcp=7,vlan_tci1=0x0000,dl_src=00:06:07:08:09:0a,dl_dst=00:01:02:03:04:05,nw_src=127.0.0.1,nw_dst=127.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=81,tp_dst=82 +-udp,in_port=ANY,dl_vlan=99,dl_vlan_pcp=7,vlan_tci1=0x0000,dl_src=00:06:07:08:09:0a,dl_dst=00:01:02:03:04:05,nw_src=192.168.0.7,nw_dst=127.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=81,tp_dst=3773 +-udp,in_port=ANY,dl_vlan=99,dl_vlan_pcp=7,vlan_tci1=0x0000,dl_src=00:06:07:08:09:0a,dl_dst=00:01:02:03:04:05,nw_src=127.0.0.1,nw_dst=127.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=81,tp_dst=82 +-udp,in_port=ANY,dl_vlan=99,dl_vlan_pcp=7,vlan_tci1=0x0000,dl_src=00:06:07:08:09:0a,dl_dst=00:01:02:03:04:05,nw_src=192.168.0.7,nw_dst=127.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=81,tp_dst=3773 ++udp,in_port=ANY,dl_vlan=99,dl_vlan_pcp=7,vlan_tci1=0x0000,dl_src=00:06:07:08:09:0a,dl_dst=00:01:02:03:04:05,nw_src=127.0.0.1,nw_dst=127.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=81,tp_dst=82 ++udp,in_port=ANY,dl_vlan=99,dl_vlan_pcp=7,vlan_tci1=0x0000,dl_src=00:06:07:08:09:0a,dl_dst=00:01:02:03:04:05,nw_src=192.168.0.7,nw_dst=127.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=81,tp_dst=3773 ++udp,in_port=ANY,dl_vlan=99,dl_vlan_pcp=7,vlan_tci1=0x0000,dl_src=00:06:07:08:09:0a,dl_dst=00:01:02:03:04:05,nw_src=127.0.0.1,nw_dst=127.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=81,tp_dst=82 ++udp,in_port=ANY,dl_vlan=99,dl_vlan_pcp=7,vlan_tci1=0x0000,dl_src=00:06:07:08:09:0a,dl_dst=00:01:02:03:04:05,nw_src=192.168.0.7,nw_dst=127.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=81,tp_dst=3773 + ]) - @staticmethod -- def deadline(fsm): -+ def deadline(fsm, now): - return fsm.state_entered + OVS_VSWITCHD_STOP +diff --git a/tests/drop-stats.at b/tests/drop-stats.at +index f3e19cd83b..1d3af98dab 100644 +--- a/tests/drop-stats.at ++++ b/tests/drop-stats.at +@@ -83,6 +83,9 @@ AT_CHECK([ + ovs-ofctl -Oopenflow13 add-flows br0 flows.txt + ovs-ofctl -Oopenflow13 dump-flows br0 | ofctl_strip | sort | grep actions ], [0], [ignore]) - @staticmethod -@@ -545,7 +559,7 @@ class Reconnect(object): - returned if the "probe interval" is nonzero--see - self.set_probe_interval()).""" ++ovs-appctl time/warp 15000 ++AT_CHECK([ovs-appctl revalidator/wait]) ++ + AT_CHECK([ + ovs-appctl netdev-dummy/receive p1 'in_port(1),packet_type(ns=0,id=0),eth(src=3a:6d:d2:09:9c:ab,dst=1e:2c:e9:2a:66:9e),ipv4(src=192.168.10.10,dst=192.168.10.30,proto=1,tos=0,ttl=64,frag=no),icmp(type=8,code=0)' + ], [0], [ignore]) +diff --git a/tests/library.at b/tests/library.at +index db4997d8f0..6489be2c15 100644 +--- a/tests/library.at ++++ b/tests/library.at +@@ -252,7 +252,7 @@ AT_CHECK([ovstest test-barrier], [0], []) + AT_CLEANUP -- deadline = self.state.deadline(self) -+ deadline = self.state.deadline(self, now) - if deadline is not None and now >= deadline: - return self.state.run(self, now) - else: -@@ -562,7 +576,7 @@ class Reconnect(object): - """Returns the number of milliseconds after which self.run() should be - called if nothing else notable happens in the meantime, or None if this - is currently unnecessary.""" -- deadline = self.state.deadline(self) -+ deadline = self.state.deadline(self, now) - if deadline is not None: - remaining = deadline - now - return max(0, remaining) -diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in -index 16ef1ac3ab..d0ae78e4ed 100644 ---- a/rhel/openvswitch-fedora.spec.in -+++ b/rhel/openvswitch-fedora.spec.in -@@ -455,6 +455,7 @@ fi - %{_datadir}/openvswitch/scripts/ovs-ctl - %{_datadir}/openvswitch/scripts/ovs-kmod-ctl - %{_datadir}/openvswitch/scripts/ovs-systemd-reload -+%config %{_datadir}/openvswitch/local-config.ovsschema - %config %{_datadir}/openvswitch/vswitch.ovsschema - %config %{_datadir}/openvswitch/vtep.ovsschema - %{_bindir}/ovs-appctl -@@ -476,6 +477,7 @@ fi - %{_mandir}/man1/ovsdb-server.1* - %{_mandir}/man1/ovsdb-tool.1* - %{_mandir}/man5/ovsdb-server.5* -+%{_mandir}/man5/ovsdb.local-config.5* - %{_mandir}/man5/ovs-vswitchd.conf.db.5* - %{_mandir}/man5/ovsdb.5* - %{_mandir}/man5/vtep.5* -diff --git a/rhel/openvswitch.spec.in b/rhel/openvswitch.spec.in -index 220e5c7472..2d8ff18bb0 100644 ---- a/rhel/openvswitch.spec.in -+++ b/rhel/openvswitch.spec.in -@@ -229,6 +229,7 @@ exit 0 - /usr/share/man/man1/ovsdb-client.1.gz - /usr/share/man/man1/ovsdb-server.1.gz - /usr/share/man/man1/ovsdb-tool.1.gz -+/usr/share/man/man5/ovsdb.local-config.5.gz - /usr/share/man/man5/ovsdb-server.5.gz - /usr/share/man/man5/ovs-vswitchd.conf.db.5.gz - %{_mandir}/man5/ovsdb.5* -@@ -262,6 +263,7 @@ exit 0 - /usr/share/openvswitch/scripts/ovs-vtep - /usr/share/openvswitch/scripts/sysconfig.template - /usr/share/openvswitch/scripts/ovs-monitor-ipsec -+/usr/share/openvswitch/local-config.ovsschema - /usr/share/openvswitch/vswitch.ovsschema - /usr/share/openvswitch/vtep.ovsschema - %doc NOTICE README.rst NEWS rhel/README.RHEL.rst -diff --git a/tests/alb.at b/tests/alb.at -index 2bef06f39c..922185d61d 100644 ---- a/tests/alb.at -+++ b/tests/alb.at -@@ -96,6 +96,52 @@ OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "PMD auto load balance - OVS_VSWITCHD_STOP + AT_SETUP([rcu]) +-AT_CHECK([ovstest test-rcu-quiesce], [0], []) ++AT_CHECK([ovstest test-rcu], [0], []) AT_CLEANUP -+AT_SETUP([ALB - cross-numa]) -+OVS_VSWITCHD_START([add-port br0 p0 \ -+ -- set Interface p0 type=dummy-pmd options:n_rxq=4 \ -+ -- set Interface p0 options:numa_id=0 \ -+ -- set Open_vSwitch . other_config:pmd-cpu-mask=0x3 \ -+ -- set open_vswitch . other_config:pmd-rxq-assign=group \ -+ -- set open_vswitch . other_config:pmd-rxq-isolate=false \ -+ -- set open_vswitch . other_config:pmd-auto-lb="true" \ -+ -- set open_vswitch . other_config:pmd-auto-lb-load-threshold=0], -+ [], [], [--dummy-numa 1,2,1,2]) -+OVS_WAIT_UNTIL([grep "PMD auto load balance is enabled" ovs-vswitchd.log]) -+AT_CHECK([ovs-appctl vlog/set dpif_netdev:dbg]) + AT_SETUP([stopwatch module]) +diff --git a/tests/mcast-snooping.at b/tests/mcast-snooping.at +index 757cf7186e..fe475e7b38 100644 +--- a/tests/mcast-snooping.at ++++ b/tests/mcast-snooping.at +@@ -216,3 +216,70 @@ AT_CHECK([ovs-appctl mdb/show br0], [0], [dnl + ]) + + AT_CLEANUP + -+# no pinned rxqs - cross-numa pmd could change -+get_log_next_line_num -+ovs-appctl time/warp 600000 10000 -+OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "PMD auto load balance performing dry run."]) -+OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "PMD auto load balance detected cross-numa polling"]) + -+# all pinned rxqs - cross-numa pmd will not change -+AT_CHECK([ovs-vsctl set Interface p0 other_config:pmd-rxq-affinity='0:0,1:0,2:1,3:1']) -+get_log_next_line_num -+ovs-appctl time/warp 600000 10000 -+OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "PMD auto load balance performing dry run."]) -+OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "Variance improvement 0%."]) ++AT_SETUP([mcast - igmp flood for non-snoop enabled]) ++OVS_VSWITCHD_START([]) + -+# mix of pinned (non-isolated) and non-pinned rxqs - cross-numa pmd could change -+AT_CHECK([ovs-vsctl remove Interface p0 other_config pmd-rxq-affinity]) -+AT_CHECK([ovs-vsctl set Interface p0 other_config:pmd-rxq-affinity='0:0,1:0,2:1']) -+get_log_next_line_num -+ovs-appctl time/warp 600000 10000 -+OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "PMD auto load balance performing dry run."]) -+OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "PMD auto load balance detected cross-numa polling"]) ++AT_CHECK([ ++ ovs-vsctl set bridge br0 \ ++ datapath_type=dummy], [0]) + -+# mix of pinned (isolated) and non-pinned rxqs - cross-numa pmd could change -+AT_CHECK([ovs-vsctl set Open_vSwitch . other_config:pmd-cpu-mask=0xf]) -+AT_CHECK([ovs-vsctl set Interface p0 options:n_rxq=6]) -+AT_CHECK([ovs-vsctl set Open_vSwitch . other_config:pmd-rxq-isolate=true]) -+get_log_next_line_num -+ovs-appctl time/warp 600000 10000 -+OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "PMD auto load balance performing dry run."]) -+OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "PMD auto load balance detected cross-numa polling"]) ++add_of_ports br0 1 2 + -+OVS_VSWITCHD_STOP -+AT_CLEANUP ++AT_CHECK([ovs-ofctl add-flow br0 action=normal]) + - AT_SETUP([ALB - PMD/RxQ assignment type]) - OVS_VSWITCHD_START([add-port br0 p0 \ - -- set Interface p0 type=dummy-pmd options:n_rxq=3 \ -@@ -197,7 +243,25 @@ get_log_next_line_num - AT_CHECK([ovs-vsctl set open_vswitch . other_config:pmd-auto-lb-rebal-interval="0"]) - CHECK_ALB_PARAM([interval], [1 mins], [+$LINENUM]) - --# No check for above max as it is only a documented max value and not a hard limit -+# Set new value -+get_log_next_line_num -+AT_CHECK([ovs-vsctl set open_vswitch . other_config:pmd-auto-lb-rebal-interval="100"]) -+CHECK_ALB_PARAM([interval], [100 mins], [+$LINENUM]) ++ovs-appctl time/stop + -+# Set above max value -+get_log_next_line_num -+AT_CHECK([ovs-vsctl set open_vswitch . other_config:pmd-auto-lb-rebal-interval="20001"]) -+CHECK_ALB_PARAM([interval], [1 mins], [+$LINENUM]) ++dnl Basic scenario - needs to flood for IGMP followed by unicast ICMP ++dnl in reverse direction ++AT_CHECK([ovs-appctl netdev-dummy/receive p1 \ ++ '0101000c29a0aa55aa550001080046c00028000040000102d3494565eb4ae0000016940400002200f9020000000104000000e00000fb000000000000']) ++AT_CHECK([ovs-appctl netdev-dummy/receive p2 \ ++ 'aa55aa5500010101000c29a008004500001c00010000400164dc0a0101010a0101020800f7ffffffffff']) + -+# Set new value -+get_log_next_line_num -+AT_CHECK([ovs-vsctl set open_vswitch . other_config:pmd-auto-lb-rebal-interval="1000"]) -+CHECK_ALB_PARAM([interval], [1000 mins], [+$LINENUM]) + -+# Set Negative value -+get_log_next_line_num -+AT_CHECK([ovs-vsctl set open_vswitch . other_config:pmd-auto-lb-rebal-interval="-1"]) -+CHECK_ALB_PARAM([interval], [1 mins], [+$LINENUM]) - - OVS_VSWITCHD_STOP - AT_CLEANUP -diff --git a/tests/classifier.at b/tests/classifier.at -index cdcd72c156..f652b59837 100644 ---- a/tests/classifier.at -+++ b/tests/classifier.at -@@ -129,6 +129,31 @@ Datapath actions: 3 - OVS_VSWITCHD_STOP(["/'prefixes' with incompatible field: ipv6_label/d"]) - AT_CLEANUP - -+AT_SETUP([flow classifier - ipv6 ND dependency]) -+OVS_VSWITCHD_START -+add_of_ports br0 1 2 ++AT_CHECK([ovs-appctl dpctl/dump-flows | grep -e .*ipv4 | sort | dnl ++ strip_stats | strip_used | strip_recirc | dnl ++ sed -e 's/,packet_type(ns=[[0-9]]*,id=[[0-9]]*),/,/'], ++ [0], [dnl ++recirc_id(),in_port(1),eth(src=aa:55:aa:55:00:01,dst=01:01:00:0c:29:a0),eth_type(0x0800),ipv4(frag=no), packets:0, bytes:0, used:never, actions:100,2 ++recirc_id(),in_port(2),eth(src=01:01:00:0c:29:a0,dst=aa:55:aa:55:00:01),eth_type(0x0800),ipv4(frag=no), packets:0, bytes:0, used:never, actions:1 ++]) ++ ++ovs-appctl time/warp 100000 ++ ++dnl Next we should clear the flows and install a complex case ++AT_CHECK([ovs-ofctl del-flows br0]) ++ +AT_DATA([flows.txt], [dnl -+ table=0,priority=100,ipv6,ipv6_src=1000::/10 actions=resubmit(,1) -+ table=0,priority=0 actions=NORMAL -+ table=1,priority=110,ipv6,ipv6_dst=1000::3 actions=resubmit(,2) -+ table=1,priority=100,ipv6,ipv6_dst=1000::4 actions=resubmit(,2) -+ table=1,priority=0 actions=NORMAL -+ table=2,priority=120,icmp6,nw_ttl=255,icmp_type=135,icmp_code=0,nd_target=1000::1 actions=NORMAL -+ table=2,priority=100,tcp actions=NORMAL -+ table=2,priority=100,icmp6 actions=NORMAL -+ table=2,priority=0 actions=NORMAL ++table=0, arp actions=NORMAL ++table=0, ip,in_port=1 actions=ct(table=1,zone=64000) ++table=0, in_port=2 actions=output:1 ++table=1, ip,ct_state=+trk+inv actions=drop ++table=1 ip,in_port=1,icmp,ct_state=+trk+new actions=output:2 ++table=1, in_port=1,ip,ct_state=+trk+new actions=controller(userdata=00.de.ad.be.ef.ca.fe.01) ++table=1, in_port=1,ip,ct_state=+trk+est actions=output:2 +]) +AT_CHECK([ovs-ofctl add-flows br0 flows.txt]) + -+# test ICMPv6 echo request (which should have no nd_target field) -+AT_CHECK([ovs-appctl ofproto/trace br0 "in_port=1,eth_src=f6:d2:b0:19:5e:7b,eth_dst=d2:49:19:91:78:fe,dl_type=0x86dd,ipv6_src=1000::3,ipv6_dst=1000::4,nw_proto=58,icmpv6_type=128,icmpv6_code=0"], [0], [stdout]) -+AT_CHECK([tail -2 stdout], [0], -+ [Megaflow: recirc_id=0,eth,icmp6,in_port=1,dl_src=f6:d2:b0:19:5e:7b,dl_dst=d2:49:19:91:78:fe,ipv6_src=1000::/10,ipv6_dst=1000::4,nw_ttl=0,nw_frag=no -+Datapath actions: 100,2 ++ovs-appctl time/warp 100000 ++ ++dnl Send the IGMP, followed by a unicast ICMP - ensure we won't black hole ++AT_CHECK([ovs-appctl netdev-dummy/receive p1 \ ++ '0101000c29a0aa55aa550001080046c00028000040000102d3494565eb4ae0000016940400002200f9020000000104000000e00000fb000000000000']) ++AT_CHECK([ovs-appctl netdev-dummy/receive p1 \ ++ 'aa55aa550001aa55aa55000208004500001c00010000400164dc0a0101010a0101020800f7ffffffffff']) ++ ++ ++AT_CHECK([ovs-appctl dpctl/dump-flows | grep -e .*ipv4 | sort | dnl ++ strip_stats | strip_used | strip_recirc | dnl ++ sed 's/pid=[[0-9]]*,// ++ s/,packet_type(ns=[[0-9]]*,id=[[0-9]]*),/,/'], ++ [0], [dnl ++ct_state(+new-inv+trk),recirc_id(),in_port(1),eth_type(0x0800),ipv4(proto=1,frag=no), packets:0, bytes:0, used:never, actions:2 ++ct_state(+new-inv+trk),recirc_id(),in_port(1),eth_type(0x0800),ipv4(proto=2,frag=no), packets:0, bytes:0, used:never, actions:userspace(controller(reason=1,dont_send=0,continuation=0,recirc_id=,rule_cookie=0,controller_id=0,max_len=65535)) ++recirc_id(),in_port(1),eth_type(0x0800),ipv4(frag=no), packets:0, bytes:0, used:0.0s, actions:ct(zone=64000),recirc() +]) -+OVS_VSWITCHD_STOP -+AT_CLEANUP + - AT_BANNER([conjunctive match]) - - AT_SETUP([single conjunctive match]) -diff --git a/tests/completion.at b/tests/completion.at -index 00e3a46b8b..b6155af253 100644 ---- a/tests/completion.at -+++ b/tests/completion.at -@@ -351,22 +351,22 @@ OVS_VSWITCHD_START - TMP="$(ovs-vsctl --commands | cut -d',' -f1-2 | tr -d ',[[]]' | tr -s ' ' '\n') - $(ovs-vsctl --options | grep -- '--' | sed -e 's/=.*$/=/g')" - MATCH="$(PREPARE_MATCH_NOSPACE(${TMP}))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "--db=unix:$OVS_RUNDIR/db.sock "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "--db=unix:$OVS_RUNDIR/db.sock "], - [0], [dnl - ${MATCH} ++AT_CLEANUP +diff --git a/tests/nsh.at b/tests/nsh.at +index 4d49f12017..91ded1445d 100644 +--- a/tests/nsh.at ++++ b/tests/nsh.at +@@ -27,7 +27,7 @@ AT_CHECK([ + AT_CHECK([ + ovs-appctl ofproto/trace br0 'in_port=1,dl_type=0x894f,nsh_ttl=63,nsh_mdtype=1,nsh_np=3,nsh_spi=0x123456,nsh_si=255,nsh_c1=0x11223344,nsh_c2=0x55667788,nsh_c3=0x99aabbcc,nsh_c4=0xddeeff00' + ], [0], [dnl +-Flow: in_port=1,vlan_tci=0x0000,dl_src=00:00:00:00:00:00,dl_dst=00:00:00:00:00:00,dl_type=0x894f,nsh_flags=0,nsh_ttl=63,nsh_mdtype=1,nsh_np=3,nsh_spi=0x123456,nsh_si=255,nsh_c1=0x11223344,nsh_c2=0x55667788,nsh_c3=0x99aabbcc,nsh_c4=0xddeeff00,nw_proto=0,nw_tos=0,nw_ecn=0,nw_ttl=0 ++Flow: in_port=1,vlan_tci=0x0000,dl_src=00:00:00:00:00:00,dl_dst=00:00:00:00:00:00,dl_type=0x894f,nsh_flags=0,nsh_ttl=63,nsh_mdtype=1,nsh_np=3,nsh_spi=0x123456,nsh_si=255,nsh_c1=0x11223344,nsh_c2=0x55667788,nsh_c3=0x99aabbcc,nsh_c4=0xddeeff00,nw_proto=0,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no + + bridge("br0") + ------------- +@@ -37,7 +37,7 @@ bridge("br0") + set_field:0x44332211->nsh_c1 + output:2 + +-Final flow: in_port=1,vlan_tci=0x0000,dl_src=00:00:00:00:00:00,dl_dst=00:00:00:00:00:00,dl_type=0x894f,nsh_flags=2,nsh_ttl=63,nsh_mdtype=1,nsh_np=3,nsh_spi=0x123456,nsh_si=254,nsh_c1=0x44332211,nsh_c2=0x55667788,nsh_c3=0x99aabbcc,nsh_c4=0xddeeff00,nw_proto=0,nw_tos=0,nw_ecn=0,nw_ttl=0 ++Final flow: in_port=1,vlan_tci=0x0000,dl_src=00:00:00:00:00:00,dl_dst=00:00:00:00:00:00,dl_type=0x894f,nsh_flags=2,nsh_ttl=63,nsh_mdtype=1,nsh_np=3,nsh_spi=0x123456,nsh_si=254,nsh_c1=0x44332211,nsh_c2=0x55667788,nsh_c3=0x99aabbcc,nsh_c4=0xddeeff00,nw_proto=0,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no + Megaflow: recirc_id=0,eth,in_port=1,dl_type=0x894f,nsh_flags=0,nsh_ttl=63,nsh_mdtype=1,nsh_np=3,nsh_spi=0x123456,nsh_si=255,nsh_c1=0x11223344 + Datapath actions: set(nsh(flags=2,ttl=63,spi=0x123456,si=254,c1=0x44332211)),2 ]) - # complete ovs-vsctl [TAB] --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test ""], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test ""], - [0], [dnl - ${MATCH} +@@ -85,7 +85,7 @@ AT_CHECK([ + AT_CHECK([ + ovs-appctl ofproto/trace br0 'in_port=1,icmp,dl_src=00:11:22:33:44:55,dl_dst=66:77:88:99:aa:bb,nw_dst=10.10.10.10,nw_src=20.20.20.20' + ], [0], [dnl +-Flow: icmp,in_port=1,vlan_tci=0x0000,dl_src=00:11:22:33:44:55,dl_dst=66:77:88:99:aa:bb,nw_src=20.20.20.20,nw_dst=10.10.10.10,nw_tos=0,nw_ecn=0,nw_ttl=0,icmp_type=0,icmp_code=0 ++Flow: icmp,in_port=1,vlan_tci=0x0000,dl_src=00:11:22:33:44:55,dl_dst=66:77:88:99:aa:bb,nw_src=20.20.20.20,nw_dst=10.10.10.10,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,icmp_type=0,icmp_code=0 + + bridge("br0") + ------------- +@@ -103,7 +103,7 @@ bridge("br0") + decap() + decap() + +-Final flow: in_port=1,vlan_tci=0x0000,dl_src=00:00:00:00:00:00,dl_dst=11:22:33:44:55:66,dl_type=0x894f,nsh_flags=0,nsh_ttl=63,nsh_mdtype=1,nsh_np=3,nsh_spi=0x1234,nsh_si=255,nsh_c1=0x11223344,nsh_c2=0x0,nsh_c3=0x0,nsh_c4=0x0,nw_proto=0,nw_tos=0,nw_ecn=0,nw_ttl=0 ++Final flow: in_port=1,vlan_tci=0x0000,dl_src=00:00:00:00:00:00,dl_dst=11:22:33:44:55:66,dl_type=0x894f,nsh_flags=0,nsh_ttl=63,nsh_mdtype=1,nsh_np=3,nsh_spi=0x1234,nsh_si=255,nsh_c1=0x11223344,nsh_c2=0x0,nsh_c3=0x0,nsh_c4=0x0,nw_proto=0,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no + Megaflow: recirc_id=0,eth,ip,in_port=1,dl_dst=66:77:88:99:aa:bb,nw_frag=no + Datapath actions: push_nsh(flags=0,ttl=63,mdtype=1,np=3,spi=0x1234,si=255,c1=0x11223344,c2=0x0,c3=0x0,c4=0x0),push_eth(src=00:00:00:00:00:00,dst=11:22:33:44:55:66),pop_eth,pop_nsh(),recirc(0x1) + ]) +@@ -111,7 +111,7 @@ Datapath actions: push_nsh(flags=0,ttl=63,mdtype=1,np=3,spi=0x1234,si=255,c1=0x1 + AT_CHECK([ + ovs-appctl ofproto/trace br0 'in_port=4,dl_type=0x894f,nsh_mdtype=1,nsh_np=3,nsh_spi=0x1234,nsh_c1=0x11223344' + ], [0], [dnl +-Flow: in_port=4,vlan_tci=0x0000,dl_src=00:00:00:00:00:00,dl_dst=00:00:00:00:00:00,dl_type=0x894f,nsh_flags=0,nsh_ttl=0,nsh_mdtype=1,nsh_np=3,nsh_spi=0x1234,nsh_si=0,nsh_c1=0x11223344,nsh_c2=0x0,nsh_c3=0x0,nsh_c4=0x0,nw_proto=0,nw_tos=0,nw_ecn=0,nw_ttl=0 ++Flow: in_port=4,vlan_tci=0x0000,dl_src=00:00:00:00:00:00,dl_dst=00:00:00:00:00:00,dl_type=0x894f,nsh_flags=0,nsh_ttl=0,nsh_mdtype=1,nsh_np=3,nsh_spi=0x1234,nsh_si=0,nsh_c1=0x11223344,nsh_c2=0x0,nsh_c3=0x0,nsh_c4=0x0,nw_proto=0,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no + + bridge("br0") + ------------- +@@ -213,7 +213,7 @@ AT_CHECK([ + AT_CHECK([ + ovs-appctl ofproto/trace br0 'in_port=1,icmp,dl_src=00:11:22:33:44:55,dl_dst=66:77:88:99:aa:bb,nw_dst=10.10.10.10,nw_src=20.20.20.20' + ], [0], [dnl +-Flow: icmp,in_port=1,vlan_tci=0x0000,dl_src=00:11:22:33:44:55,dl_dst=66:77:88:99:aa:bb,nw_src=20.20.20.20,nw_dst=10.10.10.10,nw_tos=0,nw_ecn=0,nw_ttl=0,icmp_type=0,icmp_code=0 ++Flow: icmp,in_port=1,vlan_tci=0x0000,dl_src=00:11:22:33:44:55,dl_dst=66:77:88:99:aa:bb,nw_src=20.20.20.20,nw_dst=10.10.10.10,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,icmp_type=0,icmp_code=0 + + bridge("br0") + ------------- +@@ -230,7 +230,7 @@ bridge("br0") + decap() + decap() + +-Final flow: in_port=1,vlan_tci=0x0000,dl_src=00:00:00:00:00:00,dl_dst=11:22:33:44:55:66,dl_type=0x894f,nsh_flags=0,nsh_ttl=63,nsh_mdtype=2,nsh_np=3,nsh_spi=0x1234,nsh_si=255,nw_proto=0,nw_tos=0,nw_ecn=0,nw_ttl=0 ++Final flow: in_port=1,vlan_tci=0x0000,dl_src=00:00:00:00:00:00,dl_dst=11:22:33:44:55:66,dl_type=0x894f,nsh_flags=0,nsh_ttl=63,nsh_mdtype=2,nsh_np=3,nsh_spi=0x1234,nsh_si=255,nw_proto=0,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no + Megaflow: recirc_id=0,eth,ip,in_port=1,dl_dst=66:77:88:99:aa:bb,nw_frag=no + Datapath actions: push_nsh(flags=0,ttl=63,mdtype=2,np=3,spi=0x1234,si=255,md2=0x10000a041234567820001408fedcba9876543210),push_eth(src=00:00:00:00:00:00,dst=11:22:33:44:55:66),pop_eth,pop_nsh(),recirc(0x1) + ]) +@@ -238,7 +238,7 @@ Datapath actions: push_nsh(flags=0,ttl=63,mdtype=2,np=3,spi=0x1234,si=255,md2=0x + AT_CHECK([ + ovs-appctl ofproto/trace br0 'in_port=4,dl_type=0x894f,nsh_mdtype=2,nsh_np=3,nsh_spi=0x1234' + ], [0], [dnl +-Flow: in_port=4,vlan_tci=0x0000,dl_src=00:00:00:00:00:00,dl_dst=00:00:00:00:00:00,dl_type=0x894f,nsh_flags=0,nsh_ttl=0,nsh_mdtype=2,nsh_np=3,nsh_spi=0x1234,nsh_si=0,nw_proto=0,nw_tos=0,nw_ecn=0,nw_ttl=0 ++Flow: in_port=4,vlan_tci=0x0000,dl_src=00:00:00:00:00:00,dl_dst=00:00:00:00:00:00,dl_type=0x894f,nsh_flags=0,nsh_ttl=0,nsh_mdtype=2,nsh_np=3,nsh_spi=0x1234,nsh_si=0,nw_proto=0,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no + + bridge("br0") + ------------- +@@ -325,7 +325,7 @@ AT_CHECK([ + AT_CHECK([ + ovs-appctl ofproto/trace br0 'in_port=1,icmp,dl_src=00:11:22:33:44:55,dl_dst=66:77:88:99:aa:bb,nw_dst=10.10.10.10,nw_src=20.20.20.20' + ], [0], [dnl +-Flow: icmp,in_port=1,vlan_tci=0x0000,dl_src=00:11:22:33:44:55,dl_dst=66:77:88:99:aa:bb,nw_src=20.20.20.20,nw_dst=10.10.10.10,nw_tos=0,nw_ecn=0,nw_ttl=0,icmp_type=0,icmp_code=0 ++Flow: icmp,in_port=1,vlan_tci=0x0000,dl_src=00:11:22:33:44:55,dl_dst=66:77:88:99:aa:bb,nw_src=20.20.20.20,nw_dst=10.10.10.10,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,icmp_type=0,icmp_code=0 + + bridge("br0") + ------------- +@@ -381,7 +381,7 @@ Datapath actions: pop_nsh(),recirc(0x2) + AT_CHECK([ + ovs-appctl ofproto/trace br0 'recirc_id=2,in_port=4,ip' + ], [0], [dnl +-Flow: recirc_id=0x2,eth,ip,in_port=4,vlan_tci=0x0000,dl_src=00:00:00:00:00:00,dl_dst=00:00:00:00:00:00,nw_src=0.0.0.0,nw_dst=0.0.0.0,nw_proto=0,nw_tos=0,nw_ecn=0,nw_ttl=0 ++Flow: recirc_id=0x2,eth,ip,in_port=4,vlan_tci=0x0000,dl_src=00:00:00:00:00:00,dl_dst=00:00:00:00:00:00,nw_src=0.0.0.0,nw_dst=0.0.0.0,nw_proto=0,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no + + bridge("br0") + ------------- +diff --git a/tests/ofp-print.at b/tests/ofp-print.at +index 2c7e163bd6..7be6628c34 100644 +--- a/tests/ofp-print.at ++++ b/tests/ofp-print.at +@@ -471,7 +471,7 @@ c0 a8 00 02 27 2f 00 00 78 50 cc 5b 57 af 42 1e \ + 50 02 02 00 26 e8 00 00 00 00 00 00 00 00 \ + "], [0], [dnl + OFPT_PACKET_IN (xid=0x0): total_len=60 in_port=3 (via no_match) data_len=60 buffer=0x00000111 +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:06,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=10031,tp_dst=0,tcp_flags=syn tcp_csum:26e8 ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:06,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=10031,tp_dst=0,tcp_flags=syn tcp_csum:26e8 ]) + AT_CLEANUP - # complete on global options. --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "--dry-run "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "--dry-run "], - [0], [dnl - ${MATCH} +@@ -485,7 +485,7 @@ c0 a8 00 02 27 2f 00 00 78 50 cc 5b 57 af 42 1e \ + 50 10 02 00 26 e8 00 00 00 00 00 00 00 00 \ + " 3], [0], [dnl + OFPT_PACKET_IN (xid=0x0): total_len=60 in_port=3 (via no_match) data_len=60 buffer=0x00000111 +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:06,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=10031,tp_dst=0,tcp_flags=ack tcp_csum:26e8 ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:06,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=10031,tp_dst=0,tcp_flags=ack tcp_csum:26e8 + 00000000 50 54 00 00 00 06 50 54-00 00 00 05 08 00 45 00 + 00000010 00 28 bd 12 00 00 40 06-3c 6a c0 a8 00 01 c0 a8 + 00000020 00 02 27 2f 00 00 78 50-cc 5b 57 af 42 1e 50 10 +@@ -504,7 +504,7 @@ c0 a8 00 02 27 2f 00 00 78 50 cc 5b 57 af 42 1e \ + 50 02 02 00 26 e8 00 00 00 00 00 00 00 00 \ + "], [0], [dnl + OFPT_PACKET_IN (OF1.1) (xid=0x0): total_len=60 in_port=3 (via no_match) data_len=60 buffer=0x00000111 +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:06,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=10031,tp_dst=0,tcp_flags=syn tcp_csum:26e8 ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:06,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=10031,tp_dst=0,tcp_flags=syn tcp_csum:26e8 ]) --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "--dry-run --pretty "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "--dry-run --pretty "], - [0], [dnl - ${MATCH} + AT_CLEANUP + +@@ -736,7 +736,7 @@ b9 7c c0 a8 00 02 c0 a8 00 01 00 00 2b 60 00 00 \ + 00 00 00 00 \ + "], [0], [dnl + OFPT_PACKET_OUT (xid=0x0): in_port=1 actions=output:3 data_len=60 +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:06,dl_dst=50:54:00:00:00:05,nw_src=192.168.0.2,nw_dst=192.168.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=0,tp_dst=11104,tcp_flags=rst|ack tcp_csum:6d75 ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:06,dl_dst=50:54:00:00:00:05,nw_src=192.168.0.2,nw_dst=192.168.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=0,tp_dst=11104,tcp_flags=rst|ack tcp_csum:6d75 ]) -@@ -374,7 +374,7 @@ ${MATCH} - # complete on local options. - TMP="$(ovs-vsctl --commands | grep -- '--may-exist' | cut -d',' -f1-2 | tr -d ',[[]]' | tr -s ' ' '\n' | grep -v -- '--may-exist')" - MATCH="$(PREPARE_MATCH_SPACE(${TMP}))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "--may-exist "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "--may-exist "], - [0], [dnl - ${MATCH} + AT_CLEANUP + +@@ -751,7 +751,7 @@ b9 7c c0 a8 00 02 c0 a8 00 01 00 00 2b 60 00 00 \ + 00 00 00 00 \ + " 3], [0], [dnl + OFPT_PACKET_OUT (xid=0x0): in_port=1 actions=output:3 data_len=60 +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:06,dl_dst=50:54:00:00:00:05,nw_src=192.168.0.2,nw_dst=192.168.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=0,tp_dst=11104,tcp_flags=rst|ack tcp_csum:6d75 ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:06,dl_dst=50:54:00:00:00:05,nw_src=192.168.0.2,nw_dst=192.168.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=0,tp_dst=11104,tcp_flags=rst|ack tcp_csum:6d75 + 00000000 50 54 00 00 00 05 50 54-00 00 00 06 08 00 45 00 + 00000010 00 28 00 00 40 00 40 06-b9 7c c0 a8 00 02 c0 a8 + 00000020 00 01 00 00 2b 60 00 00-00 00 6a 4f 2b 58 50 14 +@@ -782,7 +782,7 @@ b9 7c c0 a8 00 02 c0 a8 00 01 00 00 2b 60 00 00 \ + 00 00 00 00 \ + "], [0], [dnl + OFPT_PACKET_OUT (OF1.2) (xid=0x8858dfc5): in_port=LOCAL actions=FLOOD data_len=60 +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:06,dl_dst=50:54:00:00:00:05,nw_src=192.168.0.2,nw_dst=192.168.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=0,tp_dst=11104,tcp_flags=rst|ack tcp_csum:6d75 ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:06,dl_dst=50:54:00:00:00:05,nw_src=192.168.0.2,nw_dst=192.168.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=0,tp_dst=11104,tcp_flags=rst|ack tcp_csum:6d75 ]) -@@ -385,37 +385,37 @@ ${MATCH} - # test !. no following arguments are expanded. - TMP="$(ovsdb-client --no-heading list-tables)" - MATCH="$(PREPARE_MATCH_SPACE(${TMP}))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set "], - [0], [dnl - ${MATCH} + AT_CLEANUP + +@@ -850,7 +850,7 @@ b9 7c c0 a8 00 02 c0 a8 00 01 00 00 2b 60 00 00 \ + 00 00 00 00 + "], [0], [dnl + OFPT_PACKET_OUT (OF1.5) (xid=0x11223344): metadata=0x3,in_port=1 actions=FLOOD data_len=60 +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:06,dl_dst=50:54:00:00:00:05,nw_src=192.168.0.2,nw_dst=192.168.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=0,tp_dst=11104,tcp_flags=rst|ack tcp_csum:6d75 ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:06,dl_dst=50:54:00:00:00:05,nw_src=192.168.0.2,nw_dst=192.168.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=0,tp_dst=11104,tcp_flags=rst|ack tcp_csum:6d75 ]) - # test ?. will show completions for both current and following arguments. - ovs-vsctl br-set-external-id br0 bridge-id br0 - MATCH="$(PREPARE_MATCH_SPACE(bridge-id --))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "br-get-external-id br0 "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "br-get-external-id br0 "], - [0], [dnl - ${MATCH} + AT_CLEANUP + +@@ -3103,7 +3103,7 @@ ff ff ff ff ff ff 00 00 00 00 82 82 82 82 82 82 \ + 31 6d 00 00 00 00 00 00 00 00 \ + "], [0], [dnl + NXT_PACKET_IN (xid=0x0): table_id=7 cookie=0x9 total_len=64 reg0=0x1,reg1=0x2,reg2=0x3,reg3=0x4,reg4=0x5,tun_id=0x6,metadata=0x5a5a5a5a5a5a5a5a,in_port=1 (via action) data_len=64 (unbuffered) +-tcp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=85,tp_dst=86,tcp_flags=syn tcp_csum:316d ++tcp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=85,tp_dst=86,tcp_flags=syn tcp_csum:316d ]) - # test *. argument with this prefix could be completed for zero or more times. - TMP="$(ovs-vsctl --no-heading --columns=_uuid,name list Bridge | tr -d '\"')" - MATCH="$(PREPARE_MATCH_SPACE(${TMP} --))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "destroy Bridge "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "destroy Bridge "], - [0], [dnl - ${MATCH} + AT_CLEANUP + +@@ -3124,7 +3124,7 @@ ff ff ff ff ff ff 00 00 00 00 82 82 82 82 82 82 \ + 31 6d 00 00 00 00 00 00 00 00 \ + " 3], [0], [dnl + NXT_PACKET_IN (xid=0x0): table_id=7 cookie=0x9 total_len=64 reg0=0x1,reg1=0x2,reg2=0x3,reg3=0x4,reg4=0x5,tun_id=0x6,metadata=0x5a5a5a5a5a5a5a5a,in_port=1 (via action) data_len=64 (unbuffered) +-tcp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=85,tp_dst=86,tcp_flags=fin tcp_csum:316d ++tcp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=85,tp_dst=86,tcp_flags=fin tcp_csum:316d + 00000000 82 82 82 82 82 82 80 81-81 81 81 81 81 00 00 50 + 00000010 08 00 45 00 00 28 00 00-00 00 00 06 32 05 53 53 + 00000020 53 53 54 54 54 54 00 55-00 56 00 00 00 00 00 00 +@@ -3151,7 +3151,7 @@ AT_CHECK([ovs-ofctl ofp-print " + ], [0], [dnl + NXT_PACKET_IN2 (xid=0x0): table_id=7 cookie=0xfedcba9876543210 total_len=64 metadata=0x5a5a5a5a5a5a5a5a (via action) data_len=48 buffer=0x00000114 + userdata=01.02.03.04.05 +-ip,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=0.0.0.0,nw_dst=0.0.0.0,nw_proto=0,nw_tos=0,nw_ecn=0,nw_ttl=0 ++ip,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=0.0.0.0,nw_dst=0.0.0.0,nw_proto=0,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no ]) --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "destroy Bridge br0 "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "destroy Bridge br0 "], - [0], [dnl - ${MATCH} + AT_CLEANUP + +@@ -3869,7 +3869,7 @@ b9 7c c0 a8 00 02 c0 a8 00 01 00 00 2b 60 00 00 \ + OFPT_BUNDLE_ADD_MESSAGE (OF1.4) (xid=0x3): + bundle_id=0x1 flags=atomic + OFPT_PACKET_OUT (OF1.4) (xid=0x3): in_port=LOCAL actions=FLOOD data_len=60 +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:06,dl_dst=50:54:00:00:00:05,nw_src=192.168.0.2,nw_dst=192.168.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=0,tp_dst=11104,tcp_flags=rst|ack tcp_csum:6d75 ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:06,dl_dst=50:54:00:00:00:05,nw_src=192.168.0.2,nw_dst=192.168.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=0,tp_dst=11104,tcp_flags=rst|ack tcp_csum:6d75 ]) - # test +. the first time, an argument is required, after that, it becomes '*'. - TMP="$(ovsdb-client --no-heading list-columns Open_vSwitch Bridge | awk '/key.*value/ { print $1":"; next } { print $1; next }')" - MATCH="$(PREPARE_MATCH_NOSPACE(${TMP}))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set Bridge br0 "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set Bridge br0 "], - [0], [dnl - ${MATCH} + AT_CLEANUP + +diff --git a/tests/ofproto-dpif.at b/tests/ofproto-dpif.at +index 7c2edeb9d4..aee2af9fad 100644 +--- a/tests/ofproto-dpif.at ++++ b/tests/ofproto-dpif.at +@@ -29,6 +29,39 @@ AT_CHECK([ovs-appctl revalidator/wait]) + OVS_VSWITCHD_STOP + AT_CLEANUP + ++AT_SETUP([ofproto-dpif - lldp revalidator event(REV_RECONFIGURE)]) ++OVS_VSWITCHD_START( ++ [add-port br0 p1 -- set interface p1 ofport_request=1 type=dummy] ++) ++dnl first revalidation triggered by add interface ++AT_CHECK([ovs-appctl coverage/read-counter rev_reconfigure], [0], [dnl ++1 ++]) ++ ++dnl enable lldp ++AT_CHECK([ovs-vsctl set interface p1 lldp:enable=true]) ++AT_CHECK([ovs-appctl revalidator/wait]) ++AT_CHECK([ovs-appctl coverage/read-counter rev_reconfigure], [0], [dnl ++2 ++]) ++ ++dnl disable lldp ++AT_CHECK([ovs-vsctl set interface p1 lldp:enable=false]) ++AT_CHECK([ovs-appctl revalidator/wait]) ++AT_CHECK([ovs-appctl coverage/read-counter rev_reconfigure], [0], [dnl ++3 ++]) ++ ++dnl remove lldp, no revalidation as lldp was disabled ++AT_CHECK([ovs-vsctl remove interface p1 lldp enable]) ++AT_CHECK([ovs-appctl revalidator/wait]) ++AT_CHECK([ovs-appctl coverage/read-counter rev_reconfigure], [0], [dnl ++3 ++]) ++ ++OVS_VSWITCHD_STOP ++AT_CLEANUP ++ + AT_SETUP([ofproto-dpif - active-backup bonding (with primary)]) + + dnl Create br0 with members p1, p2 and p7, creating bond0 with p1 and +@@ -81,11 +114,12 @@ recirc_id(0),in_port(4),packet_type(ns=0,id=0),eth(src=50:54:00:00:00:0b,dst=ff: + + ovs-appctl netdev-dummy/set-admin-state p1 up + ovs-appctl time/warp 100 +-OVS_WAIT_UNTIL([ovs-appctl bond/show | STRIP_RECIRC_ID | STRIP_ACTIVE_MEMBER_MAC], [0], [dnl ++OVS_WAIT_UNTIL_EQUAL([ovs-appctl bond/show | STRIP_RECIRC_ID | STRIP_ACTIVE_MEMBER_MAC], [dnl + ---- bond0 ---- + bond_mode: active-backup + bond may use recirculation: no, + bond-hash-basis: 0 ++lb_output action: disabled, bond-id: -1 + updelay: 0 ms + downdelay: 0 ms + lacp_status: off +@@ -99,7 +133,6 @@ member p1: enabled + + member p2: enabled + may_enable: true +- ]) - MATCH="$(PREPARE_MATCH_NOSPACE(${TMP} --))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set Bridge br0 other_config:random_key=123 "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set Bridge br0 other_config:random_key=123 "], - [0], [dnl - ${MATCH} + + OVS_VSWITCHD_STOP +@@ -126,14 +159,15 @@ dnl bring the primary back and verify that we switched back to the + dnl primary. + ovs-appctl netdev-dummy/set-admin-state p1 down + ovs-appctl time/warp 100 +-OVS_WAIT_UNTIL([test -n "`ovs-appctl bond/show | fgrep 'member p1: disabled'`"]) ++OVS_WAIT_UNTIL([test -n "`ovs-appctl bond/show | grep -F 'member p1: disabled'`"]) + ovs-appctl netdev-dummy/set-admin-state p1 up + ovs-appctl time/warp 100 +-OVS_WAIT_UNTIL([ovs-appctl bond/show | STRIP_RECIRC_ID | STRIP_ACTIVE_MEMBER_MAC], [0], [dnl ++OVS_WAIT_UNTIL_EQUAL([ovs-appctl bond/show | STRIP_RECIRC_ID | STRIP_ACTIVE_MEMBER_MAC], [dnl + ---- bond0 ---- + bond_mode: active-backup + bond may use recirculation: no, + bond-hash-basis: 0 ++lb_output action: disabled, bond-id: -1 + updelay: 0 ms + downdelay: 0 ms + lacp_status: off +@@ -150,14 +184,13 @@ member p2: enabled + + member p3: enabled + may_enable: true +- ]) -@@ -453,12 +453,12 @@ OVS_VSWITCHD_START( - # - TMP="$(ovsdb-client --no-heading list-tables)" - MATCH="$(PREPARE_MATCH_SPACE(${TMP}))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set "], - [0], [dnl - ${MATCH} + + dnl Now delete the primary and verify that the output shows that the + dnl primary is no longer an member + ovs-vsctl --id=@p1 get Interface p1 -- remove Port bond0 interfaces @p1 + ovs-appctl time/warp 100 +-OVS_WAIT_UNTIL([test -n "`ovs-appctl bond/show | fgrep 'active-backup primary: p1 (no such member)'`"]) ++OVS_WAIT_UNTIL([test -n "`ovs-appctl bond/show | grep -F 'active-backup primary: p1 (no such member)'`"]) + + dnl Now re-add the primary and verify that the output shows that the + dnl primary is available again. +@@ -171,11 +204,12 @@ ovs-vsctl \ + --id=@p1 create Interface name=p1 type=dummy options:pstream=punix:$OVS_RUNDIR/p1.sock ofport_request=1 -- \ + set Port bond0 interfaces="$uuids, @p1]" + ovs-appctl time/warp 100 +-OVS_WAIT_UNTIL([ovs-appctl bond/show | STRIP_RECIRC_ID | STRIP_ACTIVE_MEMBER_MAC], [0], [dnl ++OVS_WAIT_UNTIL_EQUAL([ovs-appctl bond/show | STRIP_RECIRC_ID | STRIP_ACTIVE_MEMBER_MAC], [dnl + ---- bond0 ---- + bond_mode: active-backup + bond may use recirculation: no, + bond-hash-basis: 0 ++lb_output action: disabled, bond-id: -1 + updelay: 0 ms + downdelay: 0 ms + lacp_status: off +@@ -192,17 +226,17 @@ member p2: enabled + + member p3: enabled + may_enable: true +- ]) - MATCH="$(PREPARE_MATCH_SPACE(Open_vSwitch))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set Open"], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set Open"], - [0], [dnl - ${MATCH} + + dnl Switch to another primary + ovs-vsctl set port bond0 other_config:bond-primary=p2 + ovs-appctl time/warp 100 +-OVS_WAIT_UNTIL([ovs-appctl bond/show | STRIP_RECIRC_ID | STRIP_ACTIVE_MEMBER_MAC], [0], [dnl ++OVS_WAIT_UNTIL_EQUAL([ovs-appctl bond/show | STRIP_RECIRC_ID | STRIP_ACTIVE_MEMBER_MAC], [dnl + ---- bond0 ---- + bond_mode: active-backup + bond may use recirculation: no, + bond-hash-basis: 0 ++lb_output action: disabled, bond-id: -1 + updelay: 0 ms + downdelay: 0 ms + lacp_status: off +@@ -211,25 +245,25 @@ active-backup primary: p2 + + + member p1: enabled +- active member + may_enable: true + + member p2: enabled ++ active member + may_enable: true + + member p3: enabled + may_enable: true +- ]) -@@ -469,13 +469,13 @@ ${MATCH} - # - TMP="$(ovs-vsctl --no-heading --columns=_uuid list Open_vSwitch | tr -d '\"')" - MATCH="$(PREPARE_MATCH_SPACE(${TMP}))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set Open_vSwitch "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set Open_vSwitch "], - [0], [dnl - ${MATCH} + + dnl Remove the "bond-primary" config directive from the bond. + AT_CHECK([ovs-vsctl remove Port bond0 other_config bond-primary]) + ovs-appctl time/warp 100 +-OVS_WAIT_UNTIL([ovs-appctl bond/show | STRIP_RECIRC_ID | STRIP_ACTIVE_MEMBER_MAC], [0], [dnl ++OVS_WAIT_UNTIL_EQUAL([ovs-appctl bond/show | STRIP_RECIRC_ID | STRIP_ACTIVE_MEMBER_MAC], [dnl + ---- bond0 ---- + bond_mode: active-backup + bond may use recirculation: no, + bond-hash-basis: 0 ++lb_output action: disabled, bond-id: -1 + updelay: 0 ms + downdelay: 0 ms + lacp_status: off +@@ -238,15 +272,14 @@ active-backup primary: + + + member p1: enabled +- active member + may_enable: true + + member p2: enabled ++ active member + may_enable: true + + member p3: enabled + may_enable: true +- ]) - TMP="$(ovs-vsctl --no-heading --columns=_uuid,name list Bridge | tr -d '\"')" - MATCH="$(PREPARE_MATCH_SPACE(${TMP}))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set Bridge "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set Bridge "], - [0], [dnl - ${MATCH} + + OVS_VSWITCHD_STOP +@@ -336,9 +369,9 @@ ovs-appctl time/warp 100 + AT_CHECK([ovs-appctl dpif/dump-flows br1 > br1_flows.txt]) + # Make sure there is resonable distribution to all three ports. + # We don't want to make this check precise, in case hash function changes. +-AT_CHECK([test `egrep 'in_port\(4\)' br1_flows.txt |wc -l` -gt 3]) +-AT_CHECK([test `egrep 'in_port\(5\)' br1_flows.txt |wc -l` -gt 3]) +-AT_CHECK([test `egrep 'in_port\(6\)' br1_flows.txt |wc -l` -gt 3]) ++AT_CHECK([test `grep -E 'in_port\(4\)' br1_flows.txt |wc -l` -gt 3]) ++AT_CHECK([test `grep -E 'in_port\(5\)' br1_flows.txt |wc -l` -gt 3]) ++AT_CHECK([test `grep -E 'in_port\(6\)' br1_flows.txt |wc -l` -gt 3]) + OVS_VSWITCHD_STOP + AT_CLEANUP + +@@ -570,7 +603,7 @@ table=1 in_port=2 priority=1500 icmp actions=output(17),resubmit(,2) + table=1 in_port=3 priority=1500 icmp actions=output(14),resubmit(,2) ]) -@@ -486,13 +486,13 @@ ${MATCH} - # - TMP="$(ovs-vsctl list-br)" - MATCH="$(PREPARE_MATCH_SPACE(${TMP}))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "br-to-vlan "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "br-to-vlan "], - [0], [dnl - ${MATCH} + AT_CHECK([ovs-ofctl add-flows br0 flows.txt]) +-AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=p1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=1,nw_tos=0,nw_ttl=128,icmp_type=8,icmp_code=0'], [0], [stdout]) ++AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=p1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=1,nw_tos=0,nw_ttl=128,nw_frag=no,icmp_type=8,icmp_code=0'], [0], [stdout]) + AT_CHECK([tail -1 stdout], [0], + [Datapath actions: 10,11,12,13,14,15,16,17,18,19,20,21 ]) - # this also helps check the '_ovs_vsctl_check_startswith_string'. - MATCH="$(PREPARE_MATCH_SPACE(--weird-br_name))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "br-to-vlan --"], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "br-to-vlan --"], - [0], [dnl - ${MATCH} +@@ -584,7 +617,7 @@ echo "table=0 in_port=1 actions=output(10),goto_table(1)" > flows.txt + for i in `seq 1 63`; do echo "table=$i actions=goto_table($(($i+1)))"; done >> flows.txt + echo "table=64 actions=output(11)" >> flows.txt + AT_CHECK([ovs-ofctl -O OpenFlow12 add-flows br0 flows.txt]) +-AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=1,nw_tos=0,nw_ttl=128,icmp_type=8,icmp_code=0'], [0], [stdout]) ++AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=1,nw_tos=0,nw_ttl=128,nw_frag=no,icmp_type=8,icmp_code=0'], [0], [stdout]) + AT_CHECK([tail -1 stdout], [0], + [Datapath actions: 10,11 ]) -@@ -503,14 +503,14 @@ ${MATCH} - # - TMP="$(ovs-vsctl --no-heading --columns=name list Port | tr -d '\"')" - MATCH="$(PREPARE_MATCH_SPACE(${TMP}))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "port-to-br "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "port-to-br "], - [0], [dnl - ${MATCH} +@@ -600,7 +633,7 @@ table=1 ip actions=write_actions(output(13)),goto_table(2) + table=2 ip actions=set_field:192.168.3.91->ip_src,output(11) ]) - # complete on ports in particular bridge. - TMP="$(ovs-vsctl list-ports br0)" - MATCH="$(PREPARE_MATCH_SPACE(${TMP}))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "del-port br0 "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "del-port br0 "], - [0], [dnl - ${MATCH} + AT_CHECK([ovs-ofctl -O OpenFlow12 add-flows br0 flows.txt]) +-AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=1,nw_tos=0,nw_ttl=128,icmp_type=8,icmp_code=0'], [0], [stdout]) ++AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=1,nw_tos=0,nw_ttl=128,nw_frag=no,icmp_type=8,icmp_code=0'], [0], [stdout]) + AT_CHECK([tail -2 stdout], [0], + [Megaflow: recirc_id=0,eth,ip,in_port=1,nw_src=192.168.0.1,nw_frag=no + Datapath actions: 10,set(ipv4(src=192.168.3.91)),11,set(ipv4(src=192.168.3.90)),13 +@@ -617,7 +650,7 @@ table=1 icmp6 actions=write_actions(output(13)),goto_table(2) + table=2 in_port=1,icmp6,icmpv6_type=135 actions=set_field:fe80::4->nd_target,set_field:cc:cc:cc:cc:cc:cc->nd_sll,output(11) ]) -@@ -523,7 +523,7 @@ for br in `ovs-vsctl list-br`; do - TMP="${TMP} $(ovs-vsctl list-ifaces $br)" - done - MATCH="$(PREPARE_MATCH_SPACE(${TMP}))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "iface-to-br "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "iface-to-br "], - [0], [dnl - ${MATCH} + AT_CHECK([ovs-ofctl -O OpenFlow12 add-flows br0 flows.txt]) +-AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,icmp6,ipv6_src=fe80::1,ipv6_dst=fe80::2,nw_tos=0,nw_ttl=128,icmpv6_type=135,nd_target=fe80::2020,nd_sll=66:55:44:33:22:11'], [0], [stdout]) ++AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,icmp6,ipv6_src=fe80::1,ipv6_dst=fe80::2,nw_tos=0,nw_ttl=128,nw_frag=no,icmpv6_type=135,nd_target=fe80::2020,nd_sll=66:55:44:33:22:11'], [0], [stdout]) + AT_CHECK([tail -4 stdout], [0], + [Megaflow: recirc_id=0,eth,icmp6,in_port=1,nw_frag=no,icmp_type=0x87/0xff,icmp_code=0x0/0xff,nd_target=fe80::2020,nd_sll=66:55:44:33:22:11 + Datapath actions: 10,set(nd(target=fe80::4,sll=cc:cc:cc:cc:cc:cc)),11,set(nd(target=fe80::3,sll=aa:aa:aa:aa:aa:aa)),13 +@@ -635,7 +668,7 @@ table=0 in_port=1,ip actions=output(10),write_actions(set_field:192.168.3.90->ip + table=1 tcp actions=set_field:91->tp_src,output(11),clear_actions ]) -@@ -533,7 +533,7 @@ ${MATCH} - # test: _ovs_vsctl_complete_bridge_fail_mode - # - MATCH="$(PREPARE_MATCH_SPACE(standalone secure))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set-fail-mode br0 "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set-fail-mode br0 "], - [0], [dnl - ${MATCH} + AT_CHECK([ovs-ofctl -O OpenFlow12 add-flows br0 flows.txt]) +-AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=6,nw_tos=0,nw_ttl=128,tp_src=8,tp_dst=9'], [0], [stdout]) ++AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=6,nw_tos=0,nw_ttl=128,nw_frag=no,tp_src=8,tp_dst=9'], [0], [stdout]) + AT_CHECK([tail -2 stdout], [0], + [Megaflow: recirc_id=0,eth,tcp,in_port=1,nw_frag=no,tp_src=8 + Datapath actions: 10,set(tcp(src=91)),11 +@@ -649,7 +682,7 @@ add_of_ports br0 1 10 11 + AT_CHECK([ovs-ofctl -O OpenFlow12 add-group br0 'group_id=1234,type=all,bucket=set_field:192.168.3.90->ip_src,group:123,bucket=output:11']) + AT_CHECK([ovs-ofctl -O OpenFlow12 add-group br0 'group_id=123,type=all,bucket=output:10']) + AT_CHECK([ovs-ofctl -O OpenFlow12 add-flow br0 'ip actions=group:1234']) +-AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=1,nw_tos=0,nw_ttl=128,icmp_type=8,icmp_code=0'], [0], [stdout]) ++AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=1,nw_tos=0,nw_ttl=128,nw_frag=no,icmp_type=8,icmp_code=0'], [0], [stdout]) + AT_CHECK([tail -1 stdout], [0], + [Datapath actions: set(ipv4(src=192.168.3.90)),10,set(ipv4(src=192.168.0.1)),11 ]) -@@ -542,25 +542,25 @@ ${MATCH} - # - # test: _ovs_vsctl_complete_key - # --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "br-set-external-id br0 "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "br-set-external-id br0 "], - [0], [dnl - +@@ -661,7 +694,7 @@ OVS_VSWITCHD_START + add_of_ports br0 1 10 11 + AT_CHECK([ovs-ofctl -O OpenFlow12 add-group br0 'group_id=1234,type=all,bucket=output:10,set_field:192.168.3.90->ip_src,bucket=output:11']) + AT_CHECK([ovs-ofctl -O OpenFlow12 add-flow br0 'ip actions=group:1234']) +-AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=1,nw_tos=0,nw_ttl=128,icmp_type=8,icmp_code=0'], [0], [stdout]) ++AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=1,nw_tos=0,nw_ttl=128,nw_frag=no,icmp_type=8,icmp_code=0'], [0], [stdout]) + # Must match on the source address to be able to restore it's value for + # the second bucket + AT_CHECK([tail -2 stdout], [0], +@@ -676,7 +709,7 @@ OVS_VSWITCHD_START + add_of_ports br0 1 10 + AT_CHECK([ovs-ofctl -O OpenFlow12 add-group br0 group_id=1234,type=indirect,bucket=output:10]) + AT_CHECK([ovs-ofctl -O OpenFlow12 add-flow br0 'ip actions=group:1234']) +-AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=1,nw_tos=0,nw_ttl=128,icmp_type=8,icmp_code=0'], [0], [stdout]) ++AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=1,nw_tos=0,nw_ttl=128,nw_frag=no,icmp_type=8,icmp_code=0'], [0], [stdout]) + AT_CHECK([tail -1 stdout], [0], + [Datapath actions: 10 ]) - # since there is no key added yet, we will only get our own input. - MATCH="$(PREPARE_MATCH_SPACE(test_key))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "br-set-external-id br0 test_key"], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "br-set-external-id br0 test_key"], - [0], [dnl - ${MATCH} +@@ -708,7 +741,7 @@ OVS_VSWITCHD_START + add_of_ports br0 1 10 11 + AT_CHECK([ovs-ofctl -O OpenFlow12 add-group br0 'group_id=1234,type=all,bucket=output:10,set_field:192.168.3.90->ip_src,bucket=output:11']) + AT_CHECK([ovs-ofctl -O OpenFlow12 add-flow br0 'ip actions=write_actions(group:1234)']) +-AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=1,nw_tos=0,nw_ttl=128,icmp_type=8,icmp_code=0'], [0], [stdout]) ++AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=1,nw_tos=0,nw_ttl=128,nw_frag=no,icmp_type=8,icmp_code=0'], [0], [stdout]) + # Must match on the source address to be able to restore it's value for + # the third bucket + AT_CHECK([tail -2 stdout], [0], +@@ -723,7 +756,7 @@ OVS_VSWITCHD_START + add_of_ports br0 1 10 + AT_CHECK([ovs-ofctl -O OpenFlow12 add-group br0 group_id=1234,type=indirect,bucket=output:10]) + AT_CHECK([ovs-ofctl -O OpenFlow12 add-flow br0 'ip actions=write_actions(group:1234)']) +-AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=1,nw_tos=0,nw_ttl=128,icmp_type=8,icmp_code=0'], [0], [stdout]) ++AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=1,nw_tos=0,nw_ttl=128,nw_frag=no,icmp_type=8,icmp_code=0'], [0], [stdout]) + AT_CHECK([tail -1 stdout], [0], + [Datapath actions: 10 ]) - # now add a key, as we should see it. - ovs-vsctl br-set-external-id br0 bridge-id br0 - MATCH="$(PREPARE_MATCH_SPACE(bridge-id))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "br-set-external-id br0 "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "br-set-external-id br0 "], - [0], [dnl - ${MATCH} +@@ -743,11 +776,11 @@ add_of_ports br0 1 + add_of_ports br1 2 + AT_CHECK([ovs-ofctl -O OpenFlow12 add-flow br1 'ip actions=write_actions(pop_vlan,output:2)']) + AT_CHECK([ovs-ofctl -O OpenFlow12 add-flow br0 'ip actions=output:10']) +-AT_CHECK([ovs-appctl ofproto/trace br1 'in_port=20,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_vlan=100,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=1,nw_tos=0,nw_ttl=128,icmp_type=8,icmp_code=0'], [0], [stdout]) ++AT_CHECK([ovs-appctl ofproto/trace br1 'in_port=20,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_vlan=100,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=1,nw_tos=0,nw_ttl=128,nw_frag=no,icmp_type=8,icmp_code=0'], [0], [stdout]) + AT_CHECK([tail -1 stdout], [0], + [Datapath actions: pop_vlan,2 ]) - MATCH="$(PREPARE_MATCH_SPACE(bridge-id --))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "br-get-external-id br0 "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "br-get-external-id br0 "], - [0], [dnl - ${MATCH} +-AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_vlan=100,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=1,nw_tos=0,nw_ttl=128,icmp_type=8,icmp_code=0'], [0], [stdout]) ++AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_vlan=100,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=1,nw_tos=0,nw_ttl=128,nw_frag=no,icmp_type=8,icmp_code=0'], [0], [stdout]) + AT_CHECK([tail -1 stdout], [0], + [Datapath actions: pop_vlan,2 ]) -@@ -571,7 +571,7 @@ ${MATCH} - # - # should just return the user input. - MATCH="$(PREPARE_MATCH_SPACE(test_value --))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "br-set-external-id br0 bridge-id test_value"], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "br-set-external-id br0 bridge-id test_value"], - [0], [dnl - ${MATCH} +@@ -1011,7 +1044,7 @@ OVS_VSWITCHD_START + add_of_ports br0 1 10 11 + AT_CHECK([ovs-ofctl -O OpenFlow12 add-group br0 'group_id=1234,type=ff,bucket=watch_port:10,output:10,bucket=watch_port:11,output:11']) + AT_CHECK([ovs-ofctl -O OpenFlow12 add-flow br0 'ip actions=write_actions(group:1234)']) +-AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=1,nw_tos=0,nw_ttl=128,icmp_type=8,icmp_code=0'], [0], [stdout]) ++AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=1,nw_tos=0,nw_ttl=128,nw_frag=no,icmp_type=8,icmp_code=0'], [0], [stdout]) + AT_CHECK([tail -1 stdout], [0], + [Datapath actions: 10 ]) -@@ -583,13 +583,13 @@ ${MATCH} - TMP="$(ovsdb-client --no-heading list-columns Open_vSwitch Open_vSwitch | tr -d ':' | cut -d' ' -f1)" - UUID="$(ovs-vsctl --no-heading --columns=_uuid list Open_vSwitch | tr -d ' ')" - MATCH="$(PREPARE_MATCH_SPACE(${TMP}))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "clear Open_vSwitch $UUID "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "clear Open_vSwitch $UUID "], - [0], [dnl - ${MATCH} +@@ -1142,7 +1175,7 @@ OVS_VSWITCHD_START + add_of_ports br0 1 10 11 + AT_CHECK([ovs-ofctl -O OpenFlow12 add-group br0 'group_id=1234,type=all,bucket=output:10,move:NXM_NX_REG1[[]]->NXM_OF_IP_SRC[[]],bucket=output:11']) + AT_CHECK([ovs-ofctl -O OpenFlow12 add-flow br0 'ip actions=write_actions(load:0xffffffff->NXM_NX_REG1[[]],move:NXM_NX_REG1[[]]->NXM_NX_REG2[[]],group:1234)']) +-AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=1,nw_tos=0,nw_ttl=128,icmp_type=8,icmp_code=0'], [0], [stdout]) ++AT_CHECK([ovs-appctl ofproto/trace br0 'in_port=1,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,dl_type=0x0800,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=1,nw_tos=0,nw_ttl=128,nw_frag=no,icmp_type=8,icmp_code=0'], [0], [stdout]) + AT_CHECK([tail -2 stdout], [0], + [Megaflow: recirc_id=0,eth,ip,in_port=1,nw_src=192.168.0.1,nw_frag=no + Datapath actions: set(ipv4(src=255.255.255.255)),10,set(ipv4(src=192.168.0.1)),11 +@@ -1288,7 +1321,7 @@ table=1 in_port=1 action=dec_ttl,output:3 + AT_CHECK([ovs-ofctl add-flows br0 flows.txt]) + AT_CHECK([ovs-appctl ofproto/trace ovs-dummy 'in_port(1),eth(src=50:54:00:00:00:05,dst=50:54:00:00:00:07),eth_type(0x0800),ipv4(src=192.168.0.1,dst=192.168.0.2,proto=111,tos=0,ttl=2,frag=no)' -generate], [0], [stdout]) + AT_CHECK([tail -4 stdout], [0], [ +-Final flow: ip,in_port=1,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=111,nw_tos=0,nw_ecn=0,nw_ttl=1 ++Final flow: ip,in_port=1,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=111,nw_tos=0,nw_ecn=0,nw_ttl=1,nw_frag=no + Megaflow: recirc_id=0,eth,ip,in_port=1,nw_ttl=2,nw_frag=no + Datapath actions: set(ipv4(ttl=1)),2,userspace(pid=0,controller(reason=2,dont_send=0,continuation=0,recirc_id=1,rule_cookie=0,controller_id=0,max_len=65535)),4 ]) - TMP="$(ovsdb-client --no-heading list-columns Open_vSwitch Bridge | tr -d ':' | cut -d' ' -f1)" - MATCH="$(PREPARE_MATCH_SPACE(${TMP}))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "clear Bridge br0 "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "clear Bridge br0 "], - [0], [dnl - ${MATCH} +@@ -1311,7 +1344,7 @@ ovs-appctl netdev-dummy/receive p1 'in_port(1),eth(src=50:54:00:00:00:05,dst=50: + OVS_APP_EXIT_AND_WAIT([ovs-ofctl]) + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x0 total_len=34 in_port=1 (via invalid_ttl) data_len=34 (unbuffered) +-ip,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=111,nw_tos=0,nw_ecn=0,nw_ttl=1 ++ip,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=111,nw_tos=0,nw_ecn=0,nw_ttl=1,nw_frag=no ]) -@@ -597,7 +597,7 @@ ${MATCH} - # so, with one specified COLUMN 'other_config', it should still complete on - # COLUMNs, plus '--'. - MATCH="$(PREPARE_MATCH_SPACE(${TMP} --))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "clear Bridge br0 other_config "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "clear Bridge br0 other_config "], - [0], [dnl - ${MATCH} + OVS_VSWITCHD_STOP + AT_CLEANUP +@@ -1497,13 +1530,13 @@ OVS_WAIT_UNTIL([test `wc -l < ofctl_monitor.log` -ge 6]) + OVS_APP_EXIT_AND_WAIT([ovs-ofctl]) + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0x0 total_len=54 in_port=1 (via no_match) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=9,tcp_flags=syn tcp_csum:2e7e ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=9,tcp_flags=syn tcp_csum:2e7e + dnl + NXT_PACKET_IN (xid=0x0): cookie=0x0 total_len=54 in_port=1 (via no_match) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=9,tcp_flags=syn tcp_csum:2e7e ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=9,tcp_flags=syn tcp_csum:2e7e + dnl + NXT_PACKET_IN (xid=0x0): cookie=0x0 total_len=54 in_port=1 (via no_match) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=9,tcp_flags=syn tcp_csum:2e7e ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=9,tcp_flags=syn tcp_csum:2e7e ]) -@@ -608,19 +608,19 @@ ${MATCH} - # - # with no key available, should always get user input. - MATCH="$(PREPARE_MATCH_NOSPACE(random_key))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "add Bridge br0 other_config random_key"], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "add Bridge br0 other_config random_key"], - [0], [dnl - ${MATCH} + + AT_CHECK([ovs-appctl time/warp 5000], [0], [ignore]) +@@ -1560,13 +1593,13 @@ OVS_WAIT_UNTIL([test `wc -l < ofctl_monitor.log` -ge 6]) + OVS_APP_EXIT_AND_WAIT([ovs-ofctl]) + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + OFPT_PACKET_IN (xid=0x0): total_len=54 in_port=1 (via no_match) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=9,tcp_flags=rst|urg tcp_csum:2e5c ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=9,tcp_flags=rst|urg tcp_csum:2e5c + dnl + OFPT_PACKET_IN (xid=0x0): total_len=54 in_port=1 (via no_match) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=9,tcp_flags=rst|urg tcp_csum:2e5c ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=9,tcp_flags=rst|urg tcp_csum:2e5c + dnl + OFPT_PACKET_IN (xid=0x0): total_len=54 in_port=1 (via no_match) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=9,tcp_flags=rst|urg tcp_csum:2e5c ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=9,tcp_flags=rst|urg tcp_csum:2e5c ]) - MATCH="$(PREPARE_MATCH_NOSPACE(abc))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "add Bridge br0 other_config random_key=abc"], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "add Bridge br0 other_config random_key=abc"], - [0], [dnl - ${MATCH} + + AT_CHECK([ovs-appctl revalidator/purge], [0]) +@@ -1627,13 +1660,13 @@ OVS_APP_EXIT_AND_WAIT([ovs-ofctl]) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x0 total_len=118 in_port=1 (via action) data_len=118 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c + dnl + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x0 total_len=118 in_port=1 (via action) data_len=118 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c + dnl + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x0 total_len=118 in_port=1 (via action) data_len=118 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c ]) - # now add two random keys. - ovs-vsctl set Bridge br0 other_config:random_key1=abc other_config:random_val1=xyz - MATCH="$(PREPARE_MATCH_NOSPACE(random_key1= random_val1=))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "add Bridge br0 other_config ran"], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "add Bridge br0 other_config ran"], - [0], [dnl - ${MATCH} + + dnl Hit table 0, Miss all other tables, sent to controller +@@ -1647,13 +1680,13 @@ OVS_APP_EXIT_AND_WAIT([ovs-ofctl]) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): table_id=253 cookie=0x0 total_len=118 in_port=1 (via no_match) data_len=118 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c + dnl + NXT_PACKET_IN (xid=0x0): table_id=253 cookie=0x0 total_len=118 in_port=1 (via no_match) data_len=118 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c + dnl + NXT_PACKET_IN (xid=0x0): table_id=253 cookie=0x0 total_len=118 in_port=1 (via no_match) data_len=118 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c ]) -@@ -632,25 +632,25 @@ ${MATCH} - # at first, we should complete on column. - TMP="$(ovsdb-client --no-heading list-columns Open_vSwitch Bridge | awk '/key.*value/ { print $1":"; next } { print $1; next }')" - MATCH="$(PREPARE_MATCH_NOSPACE(${TMP}))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set Bridge br0 "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set Bridge br0 "], - [0], [dnl - ${MATCH} + + AT_CHECK([ovs-appctl revalidator/purge], [0]) +@@ -1690,13 +1723,13 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): table_id=2 cookie=0x0 total_len=118 in_port=1 (via action) data_len=118 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c + dnl + NXT_PACKET_IN (xid=0x0): table_id=2 cookie=0x0 total_len=118 in_port=1 (via action) data_len=118 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c + dnl + NXT_PACKET_IN (xid=0x0): table_id=2 cookie=0x0 total_len=118 in_port=1 (via action) data_len=118 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c ]) - MATCH="$(PREPARE_MATCH_NOSPACE(other_config:))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set Bridge br0 other"], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set Bridge br0 other"], - [0], [dnl - ${MATCH} + + dnl Hit table 1, Miss all other tables, sent to controller +@@ -1710,13 +1743,13 @@ OVS_APP_EXIT_AND_WAIT([ovs-ofctl]) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): table_id=253 cookie=0x0 total_len=118 in_port=1 (via no_match) data_len=118 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c + dnl + NXT_PACKET_IN (xid=0x0): table_id=253 cookie=0x0 total_len=118 in_port=1 (via no_match) data_len=118 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c + dnl + NXT_PACKET_IN (xid=0x0): table_id=253 cookie=0x0 total_len=118 in_port=1 (via no_match) data_len=118 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c ]) - # then, with the ':' we should complete on key. - TMP="$(ovs-vsctl --no-heading --columns=other_config list Bridge br0 | tr -d '{\"}' | tr -s ', ' '\n' | cut -d'=' -f1)" - MATCH="$(PREPARE_MATCH_NOSPACE(${TMP}))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set Bridge br0 other_config:"], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set Bridge br0 other_config:"], - [0], [dnl - ${MATCH} + + AT_CHECK([ovs-appctl revalidator/purge], [0]) +@@ -1900,13 +1933,13 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + OFPT_PACKET_IN (xid=0x0): total_len=118 in_port=1 (via no_match) data_len=118 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c + dnl + OFPT_PACKET_IN (xid=0x0): total_len=118 in_port=1 (via no_match) data_len=118 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c + dnl + OFPT_PACKET_IN (xid=0x0): total_len=118 in_port=1 (via no_match) data_len=118 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c ]) - # finally, if user fill in some value, we should just complete on user input. - MATCH="$(PREPARE_MATCH_NOSPACE(random_val1))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set Bridge br0 other_config:random_val1=12345"], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set Bridge br0 other_config:random_val1=12345"], - [0], [dnl - ${MATCH} + + dnl Singleton controller action. +@@ -1920,11 +1953,11 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + OFPT_PACKET_IN (xid=0x0): total_len=54 in_port=1 (via action) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + OFPT_PACKET_IN (xid=0x0): total_len=54 in_port=1 (via action) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + OFPT_PACKET_IN (xid=0x0): total_len=54 in_port=1 (via action) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ]) -@@ -661,12 +661,12 @@ ${MATCH} - # - touch private_key certificate - MATCH="$(PREPARE_MATCH_SPACE(private_key))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set-ssl priva"], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set-ssl priva"], - [0], [dnl - ${MATCH} + + dnl Modified controller action. +@@ -1938,13 +1971,13 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + OFPT_PACKET_IN (xid=0x0): total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,dl_vlan=15,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=30:33:33:33:33:33,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=fin tcp_csum:2e7e ++tcp,dl_vlan=15,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=30:33:33:33:33:33,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=fin tcp_csum:2e7e + dnl + OFPT_PACKET_IN (xid=0x0): total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,dl_vlan=15,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=30:33:33:33:33:33,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=fin tcp_csum:2e7e ++tcp,dl_vlan=15,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=30:33:33:33:33:33,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=fin tcp_csum:2e7e + dnl + OFPT_PACKET_IN (xid=0x0): total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,dl_vlan=15,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=30:33:33:33:33:33,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=fin tcp_csum:2e7e ++tcp,dl_vlan=15,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=30:33:33:33:33:33,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=fin tcp_csum:2e7e ]) - MATCH="$(PREPARE_MATCH_SPACE(certificate))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set-ssl private_key cer"], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set-ssl private_key cer"], - [0], [dnl - ${MATCH} + + dnl Modified VLAN controller action. +@@ -1958,13 +1991,13 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0xa total_len=38 in_port=1 (via action) data_len=38 (unbuffered) +-ip,dl_vlan=99,dl_vlan_pcp=1,vlan_tci1=0x0000,dl_src=40:44:44:44:44:41,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=16,nw_tos=0,nw_ecn=0,nw_ttl=64 ++ip,dl_vlan=99,dl_vlan_pcp=1,vlan_tci1=0x0000,dl_src=40:44:44:44:44:41,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=16,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xa total_len=38 in_port=1 (via action) data_len=38 (unbuffered) +-ip,dl_vlan=99,dl_vlan_pcp=1,vlan_tci1=0x0000,dl_src=40:44:44:44:44:41,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=16,nw_tos=0,nw_ecn=0,nw_ttl=64 ++ip,dl_vlan=99,dl_vlan_pcp=1,vlan_tci1=0x0000,dl_src=40:44:44:44:44:41,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=16,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xa total_len=38 in_port=1 (via action) data_len=38 (unbuffered) +-ip,dl_vlan=99,dl_vlan_pcp=1,vlan_tci1=0x0000,dl_src=40:44:44:44:44:41,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=16,nw_tos=0,nw_ecn=0,nw_ttl=64 ++ip,dl_vlan=99,dl_vlan_pcp=1,vlan_tci1=0x0000,dl_src=40:44:44:44:44:41,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=16,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no ]) -@@ -676,20 +676,20 @@ ${MATCH} - # test: _ovs_vsctl_complete_target - # - MATCH="$(PREPARE_MATCH_NOSPACE(pssl: ptcp: punix: ssl: tcp: unix:))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set-manager "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set-manager "], - [0], [dnl - ${MATCH} + + dnl Checksum TCP. +@@ -1978,31 +2011,31 @@ OVS_APP_EXIT_AND_WAIT([ovs-ofctl]) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0x1 total_len=54 in_port=1 (via action) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=20:22:22:22:22:22,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=11,tcp_flags=fin tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=20:22:22:22:22:22,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=11,tcp_flags=fin tcp_csum:2e7d + dnl + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x3 total_len=58 reg0=0x1,in_port=1 (via action) data_len=58 (unbuffered) +-tcp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=20:22:22:22:22:22,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=11,tcp_flags=fin tcp_csum:2e7d ++tcp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=20:22:22:22:22:22,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=11,tcp_flags=fin tcp_csum:2e7d + dnl + NXT_PACKET_IN (xid=0x0): table_id=2 cookie=0x4 total_len=58 reg0=0x1,reg1=0x2,in_port=1 (via action) data_len=58 (unbuffered) +-tcp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=11,tcp_flags=fin tcp_csum:2e7d ++tcp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=11,tcp_flags=fin tcp_csum:2e7d + dnl + NXT_PACKET_IN (xid=0x0): table_id=3 cookie=0x5 total_len=58 reg0=0x1,reg1=0x2,reg2=0x3,in_port=1 (via action) data_len=58 (unbuffered) +-tcp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=11,tcp_flags=fin tcp_csum:2e7d ++tcp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=11,tcp_flags=fin tcp_csum:2e7d + dnl + NXT_PACKET_IN (xid=0x0): table_id=4 cookie=0x6 total_len=58 reg0=0x1,reg1=0x2,reg2=0x3,reg3=0x4,in_port=1 (via action) data_len=58 (unbuffered) +-tcp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=11,tcp_flags=fin tcp_csum:4880 ++tcp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=11,tcp_flags=fin tcp_csum:4880 + dnl + NXT_PACKET_IN (xid=0x0): table_id=5 cookie=0x7 total_len=58 reg0=0x1,reg1=0x2,reg2=0x3,reg3=0x4,reg4=0x5,tun_id=0x6,in_port=1 (via action) data_len=58 (unbuffered) +-tcp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=11,tcp_flags=fin tcp_csum:6082 ++tcp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=11,tcp_flags=fin tcp_csum:6082 + dnl + NXT_PACKET_IN (xid=0x0): table_id=6 cookie=0x8 total_len=58 reg0=0x1,reg1=0x2,reg2=0x3,reg3=0x4,reg4=0x5,tun_id=0x6,in_port=1 (via action) data_len=58 (unbuffered) +-tcp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=85,tp_dst=11,tcp_flags=fin tcp_csum:6035 ++tcp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=85,tp_dst=11,tcp_flags=fin tcp_csum:6035 + dnl + NXT_PACKET_IN (xid=0x0): table_id=7 cookie=0x9 total_len=58 reg0=0x1,reg1=0x2,reg2=0x3,reg3=0x4,reg4=0x5,tun_id=0x6,in_port=1 (via action) data_len=58 (unbuffered) +-tcp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=85,tp_dst=86,tcp_flags=fin tcp_csum:5fea ++tcp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=85,tp_dst=86,tcp_flags=fin tcp_csum:5fea + dnl + NXT_PACKET_IN (xid=0x0): table_id=7 cookie=0x9 total_len=58 reg0=0x1,reg1=0x2,reg2=0x3,reg3=0x4,reg4=0x5,tun_id=0x6,in_port=1 (via action) data_len=58 (unbuffered) +-tcp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=85,tp_dst=86,tcp_flags=fin tcp_csum:5fea ++tcp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=85,tp_dst=86,tcp_flags=fin tcp_csum:5fea ]) - # filename completion on unix, punix. - MATCH="$(PREPARE_MATCH_NOSPACE(testsuite.log))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set-manager unix:test"], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set-manager unix:test"], - [0], [dnl - ${MATCH} + + dnl Checksum UDP. +@@ -2016,31 +2049,31 @@ OVS_APP_EXIT_AND_WAIT([ovs-ofctl]) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0x1 total_len=60 in_port=1 (via action) data_len=60 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=20:22:22:22:22:22,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=8,tp_dst=11 udp_csum:1234 ++udp,vlan_tci=0x0000,dl_src=20:22:22:22:22:22,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=8,tp_dst=11 udp_csum:1234 + dnl + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x3 total_len=64 reg0=0x1,in_port=1 (via action) data_len=64 (unbuffered) +-udp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=20:22:22:22:22:22,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=8,tp_dst=11 udp_csum:1234 ++udp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=20:22:22:22:22:22,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=8,tp_dst=11 udp_csum:1234 + dnl + NXT_PACKET_IN (xid=0x0): table_id=2 cookie=0x4 total_len=64 reg0=0x1,reg1=0x2,in_port=1 (via action) data_len=64 (unbuffered) +-udp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=8,tp_dst=11 udp_csum:1234 ++udp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=8,tp_dst=11 udp_csum:1234 + dnl + NXT_PACKET_IN (xid=0x0): table_id=3 cookie=0x5 total_len=64 reg0=0x1,reg1=0x2,reg2=0x3,in_port=1 (via action) data_len=64 (unbuffered) +-udp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=8,tp_dst=11 udp_csum:1234 ++udp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=8,tp_dst=11 udp_csum:1234 + dnl + NXT_PACKET_IN (xid=0x0): table_id=4 cookie=0x6 total_len=64 reg0=0x1,reg1=0x2,reg2=0x3,reg3=0x4,in_port=1 (via action) data_len=64 (unbuffered) +-udp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=8,tp_dst=11 udp_csum:2c37 ++udp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=8,tp_dst=11 udp_csum:2c37 + dnl + NXT_PACKET_IN (xid=0x0): table_id=5 cookie=0x7 total_len=64 reg0=0x1,reg1=0x2,reg2=0x3,reg3=0x4,reg4=0x5,tun_id=0x6,in_port=1 (via action) data_len=64 (unbuffered) +-udp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=8,tp_dst=11 udp_csum:4439 ++udp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=8,tp_dst=11 udp_csum:4439 + dnl + NXT_PACKET_IN (xid=0x0): table_id=6 cookie=0x8 total_len=64 reg0=0x1,reg1=0x2,reg2=0x3,reg3=0x4,reg4=0x5,tun_id=0x6,in_port=1 (via action) data_len=64 (unbuffered) +-udp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=85,tp_dst=11 udp_csum:43ec ++udp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=85,tp_dst=11 udp_csum:43ec + dnl + NXT_PACKET_IN (xid=0x0): table_id=7 cookie=0x9 total_len=64 reg0=0x1,reg1=0x2,reg2=0x3,reg3=0x4,reg4=0x5,tun_id=0x6,in_port=1 (via action) data_len=64 (unbuffered) +-udp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=85,tp_dst=86 udp_csum:43a1 ++udp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=85,tp_dst=86 udp_csum:43a1 + dnl + NXT_PACKET_IN (xid=0x0): table_id=7 cookie=0x9 total_len=64 reg0=0x1,reg1=0x2,reg2=0x3,reg3=0x4,reg4=0x5,tun_id=0x6,in_port=1 (via action) data_len=64 (unbuffered) +-udp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=85,tp_dst=86 udp_csum:43a1 ++udp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=85,tp_dst=86 udp_csum:43a1 ]) - # no completion on other type, just return available types. - # in real environment, bash will not complete on anything. - MATCH="$(PREPARE_MATCH_NOSPACE(pssl: ptcp: punix: tcp: unix:))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set-manager ssl:something"], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set-manager ssl:something"], - [0], [dnl - ${MATCH} + + dnl Modified ARP controller action. +@@ -2087,31 +2120,31 @@ OVS_WAIT_UNTIL([test `wc -l < ofctl_monitor.log` -ge 18]) + OVS_APP_EXIT_AND_WAIT([ovs-ofctl]) + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0x1 total_len=98 in_port=1 (via action) data_len=98 (unbuffered) +-sctp,vlan_tci=0x0000,dl_src=20:22:22:22:22:22,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=1112,tp_dst=2223 sctp_csum:d9d79157 ++sctp,vlan_tci=0x0000,dl_src=20:22:22:22:22:22,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=1112,tp_dst=2223 sctp_csum:d9d79157 + dnl + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x3 total_len=102 reg0=0x1,in_port=1 (via action) data_len=102 (unbuffered) +-sctp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=20:22:22:22:22:22,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=1112,tp_dst=2223 sctp_csum:d9d79157 ++sctp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=20:22:22:22:22:22,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=1112,tp_dst=2223 sctp_csum:d9d79157 + dnl + NXT_PACKET_IN (xid=0x0): table_id=2 cookie=0x4 total_len=102 reg0=0x1,reg1=0x2,in_port=1 (via action) data_len=102 (unbuffered) +-sctp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=1112,tp_dst=2223 sctp_csum:d9d79157 ++sctp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=1112,tp_dst=2223 sctp_csum:d9d79157 + dnl + NXT_PACKET_IN (xid=0x0): table_id=3 cookie=0x5 total_len=102 reg0=0x1,reg1=0x2,reg2=0x3,in_port=1 (via action) data_len=102 (unbuffered) +-sctp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=1112,tp_dst=2223 sctp_csum:d9d79157 ++sctp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=1112,tp_dst=2223 sctp_csum:d9d79157 + dnl + NXT_PACKET_IN (xid=0x0): table_id=4 cookie=0x6 total_len=102 reg0=0x1,reg1=0x2,reg2=0x3,reg3=0x4,in_port=1 (via action) data_len=102 (unbuffered) +-sctp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=1112,tp_dst=2223 sctp_csum:d9d79157 ++sctp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=1112,tp_dst=2223 sctp_csum:d9d79157 + dnl + NXT_PACKET_IN (xid=0x0): table_id=5 cookie=0x7 total_len=102 reg0=0x1,reg1=0x2,reg2=0x3,reg3=0x4,reg4=0x5,tun_id=0x6,in_port=1 (via action) data_len=102 (unbuffered) +-sctp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=1112,tp_dst=2223 sctp_csum:d9d79157 ++sctp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=1112,tp_dst=2223 sctp_csum:d9d79157 + dnl + NXT_PACKET_IN (xid=0x0): table_id=6 cookie=0x8 total_len=102 reg0=0x1,reg1=0x2,reg2=0x3,reg3=0x4,reg4=0x5,tun_id=0x6,in_port=1 (via action) data_len=102 (unbuffered) +-sctp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=85,tp_dst=2223 sctp_csum:dd778f5f ++sctp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=85,tp_dst=2223 sctp_csum:dd778f5f + dnl + NXT_PACKET_IN (xid=0x0): table_id=7 cookie=0x9 total_len=102 reg0=0x1,reg1=0x2,reg2=0x3,reg3=0x4,reg4=0x5,tun_id=0x6,in_port=1 (via action) data_len=102 (unbuffered) +-sctp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=85,tp_dst=86 sctp_csum:62051f56 ++sctp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=85,tp_dst=86 sctp_csum:62051f56 + dnl + NXT_PACKET_IN (xid=0x0): table_id=7 cookie=0x9 total_len=102 reg0=0x1,reg1=0x2,reg2=0x3,reg3=0x4,reg4=0x5,tun_id=0x6,in_port=1 (via action) data_len=102 (unbuffered) +-sctp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=85,tp_dst=86 sctp_csum:62051f56 ++sctp,dl_vlan=80,dl_vlan_pcp=0,vlan_tci1=0x0000,dl_src=80:81:81:81:81:81,dl_dst=82:82:82:82:82:82,nw_src=83.83.83.83,nw_dst=84.84.84.84,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=85,tp_dst=86 sctp_csum:62051f56 ]) -@@ -699,14 +699,14 @@ ${MATCH} - # test: _ovs_vsctl_complete_new - # - # test 'add-br' --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "add-br "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "add-br "], - [0], [dnl - --- BEGIN MESSAGE - Enter a new bridge: - > ovs-vsctl add-br --- END MESSAGE + + AT_CHECK([ovs-ofctl dump-flows br0 | ofctl_strip | sort], [0], [dnl +@@ -2151,13 +2184,13 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + OFPT_PACKET_IN (xid=0x0): total_len=118 in_port=1 (via action) data_len=118 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c + dnl + OFPT_PACKET_IN (xid=0x0): total_len=118 in_port=1 (via action) data_len=118 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c + dnl + OFPT_PACKET_IN (xid=0x0): total_len=118 in_port=1 (via action) data_len=118 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c ++tcp,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=9,tcp_flags=ack tcp_csum:4a2c ]) - # user input does not change the output. --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "add-br new-br"], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "add-br new-br"], - [0], [dnl - --- BEGIN MESSAGE - Enter a new bridge: -@@ -715,7 +715,7 @@ Enter a new bridge: - # after specifying the new bridge name, we should complete on parent bridge. - TMP="$(ovs-vsctl list-br)" - MATCH="$(PREPARE_MATCH_SPACE(${TMP}))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "add-br new-br "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "add-br new-br "], - [0], [dnl - ${MATCH} + + OVS_VSWITCHD_STOP +@@ -2325,13 +2358,13 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0xa total_len=34 in_port=1 (via action) data_len=34 (unbuffered) +-ip,vlan_tci=0x0000,dl_src=41:44:44:44:44:42,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=16,nw_tos=0,nw_ecn=0,nw_ttl=64 ++ip,vlan_tci=0x0000,dl_src=41:44:44:44:44:42,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=16,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xa total_len=34 in_port=1 (via action) data_len=34 (unbuffered) +-ip,vlan_tci=0x0000,dl_src=41:44:44:44:44:42,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=16,nw_tos=0,nw_ecn=0,nw_ttl=64 ++ip,vlan_tci=0x0000,dl_src=41:44:44:44:44:42,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=16,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xa total_len=34 in_port=1 (via action) data_len=34 (unbuffered) +-ip,vlan_tci=0x0000,dl_src=41:44:44:44:44:42,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=16,nw_tos=0,nw_ecn=0,nw_ttl=64 ++ip,vlan_tci=0x0000,dl_src=41:44:44:44:44:42,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_proto=16,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no ]) -@@ -724,7 +724,7 @@ ${MATCH} - # of '*COLUMN?:KEY=VALUE'. - TMP="$(ovsdb-client --no-heading list-columns Open_vSwitch Port | awk '/key.*value/ { print $1":"; next } { print $1; next }')" - MATCH="$(PREPARE_MATCH_NOSPACE(${TMP} --))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "add-port br0 new-port "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "add-port br0 new-port "], - [0], [dnl - ${MATCH} + + dnl Modified MPLS controller action. +@@ -2539,13 +2572,13 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:66:66,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:66:66,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:66:66,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:66:66,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:66:66,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:66:66,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ]) -@@ -736,13 +736,13 @@ ${MATCH} - # after '--', there should be no global options available for completion. - TMP="$(ovs-vsctl --commands | cut -d',' -f1-2 | tr -d ',[[]]' | tr -s ' ' '\n')" - MATCH="$(PREPARE_MATCH_NOSPACE(${TMP}))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "init -- "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "init -- "], - [0], [dnl - ${MATCH} + + AT_CHECK([ovs-appctl time/warp 5000], [0], [ignore]) +@@ -2565,13 +2598,13 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) + + AT_CHECK([strip_metadata < ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:01,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=254,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:01,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=254,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:01,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=254,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:01,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=254,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:01,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=254,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:01,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=254,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ]) - TMP="$(ovs-vsctl --no-heading --columns=name,_uuid list Port | tr -d '\"')" - MATCH="$(PREPARE_MATCH_SPACE(${TMP} newp1 newp2))" --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "add-port br0 newp1 -- add-port br1 newp2 -- set Port "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "add-port br0 newp1 -- add-port br1 newp2 -- set Port "], - [0], [dnl - ${MATCH} + + AT_CHECK([ovs-appctl time/warp 5000], [0], [ignore]) +@@ -2591,13 +2624,13 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) + + AT_CHECK([strip_metadata < ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:02,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:2dee ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:02,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:2dee + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:02,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:2dee ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:02,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:2dee + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:02,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:2dee ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:02,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:2dee ]) -@@ -757,25 +757,25 @@ AT_SKIP_IF([eval 'test ${BASH_VERSINFO[[0]]} -lt 4']) - OVS_VSWITCHD_START - # complete non-matching command. --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "invalid"], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "invalid"], - [0], [dnl + AT_CHECK([ovs-appctl time/warp 5000], [0], [ignore]) +@@ -2617,13 +2650,13 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) + AT_CHECK([strip_metadata < ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:03,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.2,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7743 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:03,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.2,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7743 + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:03,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.2,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7743 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:03,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.2,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7743 + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:03,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.2,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7743 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:03,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.2,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7743 ]) - # complete after invalid command. --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "invalid argu"], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "invalid argu"], - [0], [dnl + AT_CHECK([ovs-appctl time/warp 5000], [0], [ignore]) +@@ -2643,13 +2676,13 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) + AT_CHECK([strip_metadata < ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:04,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.2,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7743 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:04,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.2,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7743 + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:04,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.2,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7743 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:04,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.2,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7743 + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:04,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.2,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7743 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:04,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.2,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7743 ]) - # complete non-matching end argument. --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set INVALID_"], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set INVALID_"], - [0], [dnl + AT_CHECK([ovs-appctl time/warp 5000], [0], [ignore]) +@@ -2669,13 +2702,13 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) + AT_CHECK([strip_metadata < ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.0,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7745 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.0,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7745 + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.0,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7745 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.0,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7745 + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.0,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7745 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.0,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7745 ]) - # complete after invalid intermediate argument. --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "set INVALID_TBL "], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "set INVALID_TBL "], - [1], [dnl - --- BEGIN MESSAGE - Cannot complete 'INVALID_TBL' at index 3: -@@ -783,12 +783,12 @@ Cannot complete 'INVALID_TBL' at index 3: + AT_CHECK([ovs-appctl time/warp 5000], [0], [ignore]) +@@ -2695,13 +2728,13 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) - # complete ovs-vsctl --db=wrongdb [TAB] - # should return 1 and show nothing. --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test "--db=wrongdb"], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test "--db=wrongdb"], - [1], []) + AT_CHECK([strip_metadata < ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:06,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:06,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:06,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:06,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:06,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:06,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 + ]) - OVS_VSWITCHD_STOP - # delete ovsdb-server and try again. --AT_CHECK_UNQUOTED([ovs-vsctl-bashcomp.bash test ""], -+AT_CHECK_UNQUOTED([bash ovs-vsctl-bashcomp.bash test ""], - [1], []) + AT_CHECK([ovs-appctl time/warp 5000], [0], [ignore]) +@@ -2721,13 +2754,13 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) - AT_CLEANUP -diff --git a/tests/drop-stats.at b/tests/drop-stats.at -index f3e19cd83b..1d3af98dab 100644 ---- a/tests/drop-stats.at -+++ b/tests/drop-stats.at -@@ -83,6 +83,9 @@ AT_CHECK([ - ovs-ofctl -Oopenflow13 add-flows br0 flows.txt - ovs-ofctl -Oopenflow13 dump-flows br0 | ofctl_strip | sort | grep actions ], [0], [ignore]) + AT_CHECK([strip_metadata < ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:07,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:07,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:07,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:07,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:07,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:07,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 + ]) -+ovs-appctl time/warp 15000 -+AT_CHECK([ovs-appctl revalidator/wait]) -+ - AT_CHECK([ - ovs-appctl netdev-dummy/receive p1 'in_port(1),packet_type(ns=0,id=0),eth(src=3a:6d:d2:09:9c:ab,dst=1e:2c:e9:2a:66:9e),ipv4(src=192.168.10.10,dst=192.168.10.30,proto=1,tos=0,ttl=64,frag=no),icmp(type=8,code=0)' - ], [0], [ignore]) -diff --git a/tests/library.at b/tests/library.at -index db4997d8f0..6489be2c15 100644 ---- a/tests/library.at -+++ b/tests/library.at -@@ -252,7 +252,7 @@ AT_CHECK([ovstest test-barrier], [0], []) - AT_CLEANUP + AT_CHECK([ovs-appctl time/warp 5000], [0], [ignore]) +@@ -2771,13 +2804,13 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) - AT_SETUP([rcu]) --AT_CHECK([ovstest test-rcu-quiesce], [0], []) -+AT_CHECK([ovstest test-rcu], [0], []) - AT_CLEANUP + AT_CHECK([strip_metadata < ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:09,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=48,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:09,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=48,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:09,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=48,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:09,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=48,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:09,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=48,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:09,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=48,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 + ]) - AT_SETUP([stopwatch module]) -diff --git a/tests/mcast-snooping.at b/tests/mcast-snooping.at -index 757cf7186e..fe475e7b38 100644 ---- a/tests/mcast-snooping.at -+++ b/tests/mcast-snooping.at -@@ -216,3 +216,70 @@ AT_CHECK([ovs-appctl mdb/show br0], [0], [dnl + AT_CHECK([ovs-appctl time/warp 5000], [0], [ignore]) +@@ -2797,13 +2830,13 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) + + AT_CHECK([strip_metadata < ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:0a,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:2dee ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:0a,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:2dee + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:0a,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:2dee ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:0a,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:2dee + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:0a,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:2dee ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:0a,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:2dee ]) - AT_CLEANUP -+ -+ -+AT_SETUP([mcast - igmp flood for non-snoop enabled]) -+OVS_VSWITCHD_START([]) -+ -+AT_CHECK([ -+ ovs-vsctl set bridge br0 \ -+ datapath_type=dummy], [0]) -+ -+add_of_ports br0 1 2 -+ -+AT_CHECK([ovs-ofctl add-flow br0 action=normal]) -+ -+ovs-appctl time/stop -+ -+dnl Basic scenario - needs to flood for IGMP followed by unicast ICMP -+dnl in reverse direction -+AT_CHECK([ovs-appctl netdev-dummy/receive p1 \ -+ '0101000c29a0aa55aa550001080046c00028000040000102d3494565eb4ae0000016940400002200f9020000000104000000e00000fb000000000000']) -+AT_CHECK([ovs-appctl netdev-dummy/receive p2 \ -+ 'aa55aa5500010101000c29a008004500001c00010000400164dc0a0101010a0101020800f7ffffffffff']) -+ -+ -+AT_CHECK([ovs-appctl dpctl/dump-flows | grep -e .*ipv4 | sort | dnl -+ strip_stats | strip_used | strip_recirc | dnl -+ sed -e 's/,packet_type(ns=[[0-9]]*,id=[[0-9]]*),/,/'], -+ [0], [dnl -+recirc_id(),in_port(1),eth(src=aa:55:aa:55:00:01,dst=01:01:00:0c:29:a0),eth_type(0x0800),ipv4(frag=no), packets:0, bytes:0, used:never, actions:100,2 -+recirc_id(),in_port(2),eth(src=01:01:00:0c:29:a0,dst=aa:55:aa:55:00:01),eth_type(0x0800),ipv4(frag=no), packets:0, bytes:0, used:never, actions:1 -+]) -+ -+ovs-appctl time/warp 100000 -+ -+dnl Next we should clear the flows and install a complex case -+AT_CHECK([ovs-ofctl del-flows br0]) -+ -+AT_DATA([flows.txt], [dnl -+table=0, arp actions=NORMAL -+table=0, ip,in_port=1 actions=ct(table=1,zone=64000) -+table=0, in_port=2 actions=output:1 -+table=1, ip,ct_state=+trk+inv actions=drop -+table=1 ip,in_port=1,icmp,ct_state=+trk+new actions=output:2 -+table=1, in_port=1,ip,ct_state=+trk+new actions=controller(userdata=00.de.ad.be.ef.ca.fe.01) -+table=1, in_port=1,ip,ct_state=+trk+est actions=output:2 -+]) -+AT_CHECK([ovs-ofctl add-flows br0 flows.txt]) -+ -+ovs-appctl time/warp 100000 -+ -+dnl Send the IGMP, followed by a unicast ICMP - ensure we won't black hole -+AT_CHECK([ovs-appctl netdev-dummy/receive p1 \ -+ '0101000c29a0aa55aa550001080046c00028000040000102d3494565eb4ae0000016940400002200f9020000000104000000e00000fb000000000000']) -+AT_CHECK([ovs-appctl netdev-dummy/receive p1 \ -+ 'aa55aa550001aa55aa55000208004500001c00010000400164dc0a0101010a0101020800f7ffffffffff']) -+ -+ -+AT_CHECK([ovs-appctl dpctl/dump-flows | grep -e .*ipv4 | sort | dnl -+ strip_stats | strip_used | strip_recirc | dnl -+ sed 's/pid=[[0-9]]*,// -+ s/,packet_type(ns=[[0-9]]*,id=[[0-9]]*),/,/'], -+ [0], [dnl -+ct_state(+new-inv+trk),recirc_id(),in_port(1),eth_type(0x0800),ipv4(proto=1,frag=no), packets:0, bytes:0, used:never, actions:2 -+ct_state(+new-inv+trk),recirc_id(),in_port(1),eth_type(0x0800),ipv4(proto=2,frag=no), packets:0, bytes:0, used:never, actions:userspace(controller(reason=1,dont_send=0,continuation=0,recirc_id=,rule_cookie=0,controller_id=0,max_len=65535)) -+recirc_id(),in_port(1),eth_type(0x0800),ipv4(frag=no), packets:0, bytes:0, used:0.0s, actions:ct(zone=64000),recirc() -+]) -+ -+AT_CLEANUP -diff --git a/tests/ofproto-dpif.at b/tests/ofproto-dpif.at -index 7c2edeb9d4..c923ed6606 100644 ---- a/tests/ofproto-dpif.at -+++ b/tests/ofproto-dpif.at -@@ -29,6 +29,39 @@ AT_CHECK([ovs-appctl revalidator/wait]) - OVS_VSWITCHD_STOP - AT_CLEANUP + AT_CHECK([ovs-appctl time/warp 5000], [0], [ignore]) +@@ -2823,13 +2856,13 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) -+AT_SETUP([ofproto-dpif - lldp revalidator event(REV_RECONFIGURE)]) -+OVS_VSWITCHD_START( -+ [add-port br0 p1 -- set interface p1 ofport_request=1 type=dummy] -+) -+dnl first revalidation triggered by add interface -+AT_CHECK([ovs-appctl coverage/read-counter rev_reconfigure], [0], [dnl -+1 -+]) -+ -+dnl enable lldp -+AT_CHECK([ovs-vsctl set interface p1 lldp:enable=true]) -+AT_CHECK([ovs-appctl revalidator/wait]) -+AT_CHECK([ovs-appctl coverage/read-counter rev_reconfigure], [0], [dnl -+2 -+]) -+ -+dnl disable lldp -+AT_CHECK([ovs-vsctl set interface p1 lldp:enable=false]) -+AT_CHECK([ovs-appctl revalidator/wait]) -+AT_CHECK([ovs-appctl coverage/read-counter rev_reconfigure], [0], [dnl -+3 -+]) -+ -+dnl remove lldp, no revalidation as lldp was disabled -+AT_CHECK([ovs-vsctl remove interface p1 lldp enable]) -+AT_CHECK([ovs-appctl revalidator/wait]) -+AT_CHECK([ovs-appctl coverage/read-counter rev_reconfigure], [0], [dnl -+3 -+]) -+ -+OVS_VSWITCHD_STOP -+AT_CLEANUP -+ - AT_SETUP([ofproto-dpif - active-backup bonding (with primary)]) + AT_CHECK([strip_metadata < ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:0b,dl_dst=50:54:00:00:00:07,nw_src=10.0.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:2ded ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:0b,dl_dst=50:54:00:00:00:07,nw_src=10.0.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:2ded + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:0b,dl_dst=50:54:00:00:00:07,nw_src=10.0.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:2ded ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:0b,dl_dst=50:54:00:00:00:07,nw_src=10.0.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:2ded + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:0b,dl_dst=50:54:00:00:00:07,nw_src=10.0.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:2ded ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:0b,dl_dst=50:54:00:00:00:07,nw_src=10.0.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:2ded + ]) - dnl Create br0 with members p1, p2 and p7, creating bond0 with p1 and -@@ -81,11 +114,12 @@ recirc_id(0),in_port(4),packet_type(ns=0,id=0),eth(src=50:54:00:00:00:0b,dst=ff: + AT_CHECK([ovs-appctl time/warp 5000], [0], [ignore]) +@@ -2931,13 +2964,13 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) - ovs-appctl netdev-dummy/set-admin-state p1 up - ovs-appctl time/warp 100 --OVS_WAIT_UNTIL([ovs-appctl bond/show | STRIP_RECIRC_ID | STRIP_ACTIVE_MEMBER_MAC], [0], [dnl -+OVS_WAIT_UNTIL_EQUAL([ovs-appctl bond/show | STRIP_RECIRC_ID | STRIP_ACTIVE_MEMBER_MAC], [dnl - ---- bond0 ---- - bond_mode: active-backup - bond may use recirculation: no, - bond-hash-basis: 0 -+lb_output action: disabled, bond-id: -1 - updelay: 0 ms - downdelay: 0 ms - lacp_status: off -@@ -99,7 +133,6 @@ member p1: enabled + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:02:00,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:02:00,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:02:00,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:02:00,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xd total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:02:00,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:02:00,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 + ]) - member p2: enabled - may_enable: true -- + AT_CHECK([ovs-appctl time/warp 5000], [0], [ignore]) +@@ -2959,13 +2992,13 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) + + AT_CHECK([strip_metadata < ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0xe total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:02:01,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=254,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:02:01,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=254,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xe total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:02:01,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=254,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:02:01,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=254,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xe total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:02:01,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=254,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:02:01,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=254,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ]) - OVS_VSWITCHD_STOP -@@ -129,11 +162,12 @@ ovs-appctl time/warp 100 - OVS_WAIT_UNTIL([test -n "`ovs-appctl bond/show | fgrep 'member p1: disabled'`"]) - ovs-appctl netdev-dummy/set-admin-state p1 up - ovs-appctl time/warp 100 --OVS_WAIT_UNTIL([ovs-appctl bond/show | STRIP_RECIRC_ID | STRIP_ACTIVE_MEMBER_MAC], [0], [dnl -+OVS_WAIT_UNTIL_EQUAL([ovs-appctl bond/show | STRIP_RECIRC_ID | STRIP_ACTIVE_MEMBER_MAC], [dnl - ---- bond0 ---- - bond_mode: active-backup - bond may use recirculation: no, - bond-hash-basis: 0 -+lb_output action: disabled, bond-id: -1 - updelay: 0 ms - downdelay: 0 ms - lacp_status: off -@@ -150,7 +184,6 @@ member p2: enabled + AT_CHECK([ovs-appctl time/warp 5000], [0], [ignore]) +@@ -2986,13 +3019,13 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) - member p3: enabled - may_enable: true -- + AT_CHECK([strip_metadata < ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0xe total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:02:10,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=254,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:02:10,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=254,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xe total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:02:10,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=254,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:02:10,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=254,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 + dnl + NXT_PACKET_IN (xid=0x0): cookie=0xe total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:02:10,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=254,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:02:10,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=254,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ]) - dnl Now delete the primary and verify that the output shows that the -@@ -171,11 +204,12 @@ ovs-vsctl \ - --id=@p1 create Interface name=p1 type=dummy options:pstream=punix:$OVS_RUNDIR/p1.sock ofport_request=1 -- \ - set Port bond0 interfaces="$uuids, @p1]" - ovs-appctl time/warp 100 --OVS_WAIT_UNTIL([ovs-appctl bond/show | STRIP_RECIRC_ID | STRIP_ACTIVE_MEMBER_MAC], [0], [dnl -+OVS_WAIT_UNTIL_EQUAL([ovs-appctl bond/show | STRIP_RECIRC_ID | STRIP_ACTIVE_MEMBER_MAC], [dnl - ---- bond0 ---- - bond_mode: active-backup - bond may use recirculation: no, - bond-hash-basis: 0 -+lb_output action: disabled, bond-id: -1 - updelay: 0 ms - downdelay: 0 ms - lacp_status: off -@@ -192,17 +226,17 @@ member p2: enabled + AT_CHECK([ovs-appctl time/warp 5000], [0], [ignore]) +@@ -3314,13 +3347,13 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) - member p3: enabled - may_enable: true -- + AT_CHECK([strip_metadata < ofctl_monitor.log], [0], [dnl + OFPT_PACKET_IN (OF1.2) (xid=0x0): table_id=1 total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:08,dl_dst=50:54:00:00:00:01,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=32,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:08,dl_dst=50:54:00:00:00:01,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=32,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 + dnl + OFPT_PACKET_IN (OF1.2) (xid=0x0): table_id=1 total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:08,dl_dst=50:54:00:00:00:01,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=32,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:08,dl_dst=50:54:00:00:00:01,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=32,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 + dnl + OFPT_PACKET_IN (OF1.2) (xid=0x0): table_id=1 total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:08,dl_dst=50:54:00:00:00:01,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=32,nw_ecn=0,nw_ttl=255,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:08,dl_dst=50:54:00:00:00:01,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=32,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ]) - dnl Switch to another primary - ovs-vsctl set port bond0 other_config:bond-primary=p2 - ovs-appctl time/warp 100 --OVS_WAIT_UNTIL([ovs-appctl bond/show | STRIP_RECIRC_ID | STRIP_ACTIVE_MEMBER_MAC], [0], [dnl -+OVS_WAIT_UNTIL_EQUAL([ovs-appctl bond/show | STRIP_RECIRC_ID | STRIP_ACTIVE_MEMBER_MAC], [dnl - ---- bond0 ---- - bond_mode: active-backup - bond may use recirculation: no, - bond-hash-basis: 0 -+lb_output action: disabled, bond-id: -1 - updelay: 0 ms - downdelay: 0 ms - lacp_status: off -@@ -211,25 +245,25 @@ active-backup primary: p2 - + AT_CHECK([ovs-appctl revalidator/purge], [0]) +@@ -3363,13 +3396,13 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) - member p1: enabled -- active member - may_enable: true + AT_CHECK([strip_metadata < ofctl_monitor.log], [0], [dnl + OFPT_PACKET_IN (OF1.2) (xid=0x0): total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:08,dl_dst=50:54:00:00:00:01,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=32,nw_ecn=0,nw_ttl=254,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:08,dl_dst=50:54:00:00:00:01,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=32,nw_ecn=0,nw_ttl=254,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 + dnl + OFPT_PACKET_IN (OF1.2) (xid=0x0): total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:08,dl_dst=50:54:00:00:00:01,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=32,nw_ecn=0,nw_ttl=254,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:08,dl_dst=50:54:00:00:00:01,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=32,nw_ecn=0,nw_ttl=254,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 + dnl + OFPT_PACKET_IN (OF1.2) (xid=0x0): total_len=58 in_port=1 (via action) data_len=58 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:08,dl_dst=50:54:00:00:00:01,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=32,nw_ecn=0,nw_ttl=254,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 ++tcp,vlan_tci=0x0000,dl_src=60:66:66:66:00:08,dl_dst=50:54:00:00:00:01,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=32,nw_ecn=0,nw_ttl=254,nw_frag=no,tp_src=80,tp_dst=0,tcp_flags=0 tcp_csum:7744 + ]) - member p2: enabled -+ active member - may_enable: true + AT_CHECK([ovs-appctl revalidator/purge], [0]) +@@ -3402,13 +3435,13 @@ OVS_APP_EXIT_AND_WAIT([ovs-ofctl]) - member p3: enabled - may_enable: true -- + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + OFPT_PACKET_IN (xid=0x0): total_len=54 in_port=1 (via action) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (xid=0x0): total_len=54 in_port=1 (via action) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (xid=0x0): total_len=54 in_port=1 (via action) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ]) - dnl Remove the "bond-primary" config directive from the bond. - AT_CHECK([ovs-vsctl remove Port bond0 other_config bond-primary]) - ovs-appctl time/warp 100 --OVS_WAIT_UNTIL([ovs-appctl bond/show | STRIP_RECIRC_ID | STRIP_ACTIVE_MEMBER_MAC], [0], [dnl -+OVS_WAIT_UNTIL_EQUAL([ovs-appctl bond/show | STRIP_RECIRC_ID | STRIP_ACTIVE_MEMBER_MAC], [dnl - ---- bond0 ---- - bond_mode: active-backup - bond may use recirculation: no, - bond-hash-basis: 0 -+lb_output action: disabled, bond-id: -1 - updelay: 0 ms - downdelay: 0 ms - lacp_status: off -@@ -238,15 +272,14 @@ active-backup primary: - + AT_CHECK([ovs-appctl revalidator/purge], [0]) +@@ -3444,13 +3477,13 @@ AT_CHECK([ovs-appctl time/warp 5000], [0], [ignore]) - member p1: enabled -- active member - may_enable: true + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + OFPT_PACKET_IN (OF1.3) (xid=0x0): cookie=0x0 total_len=54 in_port=1 (via no_match) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (OF1.3) (xid=0x0): cookie=0x0 total_len=54 in_port=1 (via no_match) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (OF1.3) (xid=0x0): cookie=0x0 total_len=54 in_port=1 (via no_match) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + ]) - member p2: enabled -+ active member - may_enable: true + AT_CHECK([ovs-appctl revalidator/purge], [0]) +@@ -3516,13 +3549,13 @@ send: OFPT_SET_ASYNC (OF1.3) (xid=0x2): + REQUESTFORWARD: (off) + dnl + OFPT_PACKET_IN (OF1.3) (xid=0x0): cookie=0x0 total_len=54 in_port=1 (via no_match) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (OF1.3) (xid=0x0): cookie=0x0 total_len=54 in_port=1 (via no_match) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (OF1.3) (xid=0x0): cookie=0x0 total_len=54 in_port=1 (via no_match) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + ]) - member p3: enabled - may_enable: true -- + AT_CHECK([ovs-appctl time/warp 5000], [0], [ignore]) +@@ -3558,13 +3591,13 @@ OVS_APP_EXIT_AND_WAIT([ovs-ofctl]) + AT_CHECK([ovs-appctl revalidator/purge], [0]) + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + OFPT_PACKET_IN (OF1.4) (xid=0x0): cookie=0x0 total_len=54 in_port=1 (via no_match) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (OF1.4) (xid=0x0): cookie=0x0 total_len=54 in_port=1 (via no_match) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (OF1.4) (xid=0x0): cookie=0x0 total_len=54 in_port=1 (via no_match) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + ]) + + AT_CHECK([ovs-appctl time/warp 5000], [0], [ignore]) +@@ -3607,34 +3640,34 @@ AT_CHECK([ovs-appctl time/warp 5000], [0], [ignore]) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + OFPT_PACKET_IN (OF1.3) (xid=0x0): table_id=1 cookie=0x0 total_len=54 in_port=1 (via action) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (OF1.3) (xid=0x0): table_id=2 cookie=0x0 total_len=54 in_port=1 (via action) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (OF1.3) (xid=0x0): cookie=0x0 total_len=54 in_port=1 (via action) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (OF1.3) (xid=0x0): table_id=1 cookie=0x0 total_len=54 in_port=1 (via action) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (OF1.3) (xid=0x0): table_id=2 cookie=0x0 total_len=54 in_port=1 (via action) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (OF1.3) (xid=0x0): cookie=0x0 total_len=54 in_port=1 (via action) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (OF1.3) (xid=0x0): table_id=1 cookie=0x0 total_len=54 in_port=1 (via action) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (OF1.3) (xid=0x0): table_id=2 cookie=0x0 total_len=54 in_port=1 (via action) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (OF1.3) (xid=0x0): cookie=0x0 total_len=54 in_port=1 (via action) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (OF1.3) (xid=0x0): total_len=54 in_port=ANY (via action) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + ]) + + AT_CHECK([ovs-appctl time/warp 5000], [0], [ignore]) +@@ -3679,34 +3712,34 @@ AT_CHECK([ovs-appctl time/warp 5000], [0], [ignore]) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + OFPT_PACKET_IN (OF1.4) (xid=0x0): table_id=1 cookie=0x0 total_len=54 in_port=1 (via action) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (OF1.4) (xid=0x0): table_id=2 cookie=0x0 total_len=54 in_port=1 (via group) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (OF1.4) (xid=0x0): cookie=0x0 total_len=54 in_port=1 (via action_set) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (OF1.4) (xid=0x0): table_id=1 cookie=0x0 total_len=54 in_port=1 (via action) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (OF1.4) (xid=0x0): table_id=2 cookie=0x0 total_len=54 in_port=1 (via group) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (OF1.4) (xid=0x0): cookie=0x0 total_len=54 in_port=1 (via action_set) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (OF1.4) (xid=0x0): table_id=1 cookie=0x0 total_len=54 in_port=1 (via action) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (OF1.4) (xid=0x0): table_id=2 cookie=0x0 total_len=54 in_port=1 (via group) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (OF1.4) (xid=0x0): cookie=0x0 total_len=54 in_port=1 (via action_set) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + dnl + OFPT_PACKET_IN (OF1.4) (xid=0x0): total_len=54 in_port=ANY (via packet_out) data_len=54 (unbuffered) +-tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d ++tcp,vlan_tci=0x0000,dl_src=10:11:11:11:11:11,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=8,tp_dst=10,tcp_flags=syn tcp_csum:2e7d + ]) + + AT_CHECK([ovs-appctl time/warp 5000], [0], [ignore]) +@@ -3751,10 +3784,10 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + OFPT_PACKET_IN (OF1.3) (xid=0x0): table_id=43 cookie=0x0 total_len=98 metadata=0x67871d4d000000,in_port=1 (via action) data_len=98 (unbuffered) +-icmp,vlan_tci=0x0000,dl_src=3a:6d:d2:09:9c:ab,dl_dst=1e:2c:e9:2a:66:9e,nw_src=192.168.10.10,nw_dst=192.168.10.30,nw_tos=0,nw_ecn=0,nw_ttl=64,icmp_type=8,icmp_code=0 icmp_csum:6f20 ++icmp,vlan_tci=0x0000,dl_src=3a:6d:d2:09:9c:ab,dl_dst=1e:2c:e9:2a:66:9e,nw_src=192.168.10.10,nw_dst=192.168.10.30,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,icmp_type=8,icmp_code=0 icmp_csum:6f20 + dnl + OFPT_PACKET_IN (OF1.3) (xid=0x0): table_id=50 cookie=0x0 total_len=98 metadata=0x67871d4d000000,in_port=1 (via no_match) data_len=98 (unbuffered) +-icmp,vlan_tci=0x0000,dl_src=3a:6d:d2:09:9c:ab,dl_dst=1e:2c:e9:2a:66:9e,nw_src=192.168.10.10,nw_dst=192.168.10.30,nw_tos=0,nw_ecn=0,nw_ttl=64,icmp_type=8,icmp_code=0 icmp_csum:6f20 ++icmp,vlan_tci=0x0000,dl_src=3a:6d:d2:09:9c:ab,dl_dst=1e:2c:e9:2a:66:9e,nw_src=192.168.10.10,nw_dst=192.168.10.30,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,icmp_type=8,icmp_code=0 icmp_csum:6f20 + ]) + + OVS_VSWITCHD_STOP +@@ -5007,7 +5040,7 @@ ovs-vsctl \ + AT_CHECK([ovs-ofctl add-flow br0 action=output:1]) + + # "in_port" defaults to OFPP_NONE if it's not specified. +-flow="icmp,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_ttl=128,icmp_type=8,icmp_code=0" ++flow="icmp,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_ttl=128,nw_frag=no,icmp_type=8,icmp_code=0" + AT_CHECK([ovs-appctl ofproto/trace br0 "$flow"], [0], [stdout]) + AT_CHECK_UNQUOTED([tail -1 stdout], [0], + [Datapath actions: 1,2 +@@ -5317,7 +5350,7 @@ ovs-vsctl \ + AT_CHECK([ovs-ofctl add-flow br0 action=output:1]) + + # "in_port" defaults to OFPP_NONE if it's not specified. +-flow="icmp,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_ttl=128,icmp_type=8,icmp_code=0" ++flow="icmp,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_ttl=128,nw_frag=no,icmp_type=8,icmp_code=0" + AT_CHECK([ovs-appctl ofproto/trace br0 "$flow"], [0], [stdout]) + AT_CHECK_UNQUOTED([tail -1 stdout], [0], + [Datapath actions: 1,trunc(100),2 +@@ -5464,7 +5497,7 @@ ovs-vsctl \ + + flow="in_port=1" + AT_CHECK([ovs-appctl ofproto/trace br0 "$flow"], [0], [stdout]) +-AT_CHECK([tail -1 stdout | egrep "trunc\(200\),2,trunc\(300\),3,100|trunc\(300\),3,trunc\(200\),2,100"], [0], [stdout]) ++AT_CHECK([tail -1 stdout | grep -E "trunc\(200\),2,trunc\(300\),3,100|trunc\(300\),3,trunc\(200\),2,100"], [0], [stdout]) + + OVS_VSWITCHD_STOP + AT_CLEANUP +@@ -5512,11 +5545,11 @@ AT_CHECK([ovs-ofctl add-flows br0 flows.txt]) + + flow="in_port(1),eth(src=50:54:00:00:00:05,dst=50:54:00:00:00:07),eth_type(0x0800),ipv4(src=192.168.0.1,dst=192.168.0.2,proto=1,tos=0,ttl=128,frag=no),icmp(type=8,code=0)" + AT_CHECK([ovs-appctl ofproto/trace ovs-dummy "$flow" -generate], [0], [stdout]) +-AT_CHECK([grep "Final flow:" stdout], [0], [Final flow: icmp,tun_id=0x6,in_port=1,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=128,icmp_type=8,icmp_code=0 ++AT_CHECK([grep "Final flow:" stdout], [0], [Final flow: icmp,tun_id=0x6,in_port=1,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=128,nw_frag=no,icmp_type=8,icmp_code=0 + ]) + + AT_CHECK([ovs-appctl ofproto/trace ovs-dummy "$flow,recirc_id(1)" -generate], [0], [stdout]) +-AT_CHECK([grep "Final flow:" stdout], [0], [Final flow: recirc_id=0x1,eth,icmp,tun_id=0x6,in_port=1,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=128,icmp_type=8,icmp_code=0 ++AT_CHECK([grep "Final flow:" stdout], [0], [Final flow: recirc_id=0x1,eth,icmp,tun_id=0x6,in_port=1,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=128,nw_frag=no,icmp_type=8,icmp_code=0 ]) OVS_VSWITCHD_STOP @@ -9248,7 +61418,37 @@ index 7c2edeb9d4..c923ed6606 100644 OVS_VSWITCHD_STOP AT_CLEANUP -@@ -7600,13 +7662,28 @@ dnl configure bridge IPFIX and ensure that sample action generation works at the +@@ -6202,6 +6264,20 @@ AT_CHECK([tail -2 stderr], [0], [dnl + ovs-appctl: ovs-vswitchd: server returned an error + ]) + ++# Test incorrect command: ofproto/trace with nonexistent port number ++AT_CHECK([ovs-appctl ofproto/trace ovs-dummy "in_port(42)" ], [2], [stdout], [stderr]) ++AT_CHECK([tail -2 stderr], [0], [dnl ++no OpenFlow port for datapath port 42 ++ovs-appctl: ovs-vswitchd: server returned an error ++]) ++ ++# Test incorrect command: ofproto/trace with nonexistent recirc_id ++AT_CHECK([ovs-appctl ofproto/trace ovs-dummy "recirc_id(0x42)" ], [2], [stdout], [stderr]) ++AT_CHECK([tail -2 stderr], [0], [dnl ++no recirculation data for recirc_id 0x42 ++ovs-appctl: ovs-vswitchd: server returned an error ++]) ++ + OVS_VSWITCHD_STOP + AT_CLEANUP + +@@ -7031,7 +7107,7 @@ dnl An 170 byte packet + AT_CHECK([ovs-appctl netdev-dummy/receive p1 '000c29c8a0a4005056c0000808004500009cb4a6000040019003c0a8da01c0a8da640800cb5fa762000556f431ad0009388e08090a0b0c0d0e0f101112131415161718191a1b1c1d1e1f202122232425262728292a2b2c2d2e2f303132333435363738393a3b3c3d3e3f404142434445464748494a4b4c4d4e4f505152535455565758595a5b5c5d5e5f606162636465666768696a6b6c6d6e6f707172737475767778797a7b7c7d7e7f']) + + AT_CHECK([ovs-ofctl parse-pcap p1.pcap], [0], [dnl +-icmp,in_port=ANY,vlan_tci=0x0000,dl_src=00:50:56:c0:00:08,dl_dst=00:0c:29:c8:a0:a4,nw_src=192.168.218.1,nw_dst=192.168.218.100,nw_tos=0,nw_ecn=0,nw_ttl=64,icmp_type=8,icmp_code=0 ++icmp,in_port=ANY,vlan_tci=0x0000,dl_src=00:50:56:c0:00:08,dl_dst=00:0c:29:c8:a0:a4,nw_src=192.168.218.1,nw_dst=192.168.218.100,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,icmp_type=8,icmp_code=0 + ]) + + AT_CHECK([ovs-appctl revalidator/purge], [0]) +@@ -7600,13 +7676,28 @@ dnl configure bridge IPFIX and ensure that sample action generation works at the dnl datapath level. AT_SETUP([ofproto-dpif - Bridge IPFIX sanity check]) OVS_VSWITCHD_START @@ -9278,6 +61478,238 @@ index 7c2edeb9d4..c923ed6606 100644 dnl Send some packets that should be sampled. for i in `seq 1 3`; do AT_CHECK([ovs-appctl netdev-dummy/receive p1 'in_port(1),eth(src=50:54:00:00:00:09,dst=50:54:00:00:00:0a),eth_type(0x0800)']) +@@ -8666,7 +8757,7 @@ recirc_id(0),in_port(100),packet_type(ns=0,id=0),eth_type(0x0800),ipv4(src=192.1 + ]) + + AT_CHECK([grep -e '|ofproto_dpif_xlate|WARN|' ovs-vswitchd.log | sed "s/^.*|WARN|//"], [0], [dnl +-stack underflow on bridge br1 while processing icmp,in_port=LOCAL,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,icmp_type=8,icmp_code=0 ++stack underflow on bridge br1 while processing icmp,in_port=LOCAL,vlan_tci=0x0000,dl_src=50:54:00:00:00:05,dl_dst=50:54:00:00:00:07,nw_src=192.168.0.1,nw_dst=192.168.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,icmp_type=8,icmp_code=0 + ]) + + OVS_VSWITCHD_STOP(["/stack underflow/d"]) +@@ -9855,7 +9946,7 @@ OVS_APP_EXIT_AND_WAIT([ovs-ofctl]) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0x0 total_len=86 in_port=1 (via no_match) data_len=86 (unbuffered) +-icmp6,vlan_tci=0x0000,dl_src=00:00:86:05:80:da,dl_dst=00:60:97:07:69:ea,ipv6_src=fe80::200:86ff:fe05:80da,ipv6_dst=fe80::260:97ff:fe07:69ea,ipv6_label=0x00000,nw_tos=0,nw_ecn=0,nw_ttl=255,icmp_type=135,icmp_code=0,nd_target=fe80::260:97ff:fe07:69ea,nd_sll=00:00:86:05:80:da,nd_tll=00:00:00:00:00:00 icmp6_csum:68bd ++icmp6,vlan_tci=0x0000,dl_src=00:00:86:05:80:da,dl_dst=00:60:97:07:69:ea,ipv6_src=fe80::200:86ff:fe05:80da,ipv6_dst=fe80::260:97ff:fe07:69ea,ipv6_label=0x00000,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,icmp_type=135,icmp_code=0,nd_target=fe80::260:97ff:fe07:69ea,nd_sll=00:00:86:05:80:da,nd_tll=00:00:00:00:00:00 icmp6_csum:68bd + ]) + + OVS_VSWITCHD_STOP +@@ -9906,7 +9997,7 @@ OVS_APP_EXIT_AND_WAIT([ovs-ofctl]) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0x0 total_len=86 in_port=1 (via action) data_len=86 (unbuffered) +-icmp6,vlan_tci=0x0000,dl_src=00:00:86:05:80:da,dl_dst=00:60:97:07:69:ea,ipv6_src=fe80::200:86ff:fe05:80da,ipv6_dst=fe80::260:97ff:fe07:69ea,ipv6_label=0x00000,nw_tos=0,nw_ecn=0,nw_ttl=255,icmp_type=135,icmp_code=0,nd_target=fe80::1,nd_sll=32:21:14:86:11:74,nd_tll=00:00:00:00:00:00 icmp6_csum:19d3 ++icmp6,vlan_tci=0x0000,dl_src=00:00:86:05:80:da,dl_dst=00:60:97:07:69:ea,ipv6_src=fe80::200:86ff:fe05:80da,ipv6_dst=fe80::260:97ff:fe07:69ea,ipv6_label=0x00000,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,icmp_type=135,icmp_code=0,nd_target=fe80::1,nd_sll=32:21:14:86:11:74,nd_tll=00:00:00:00:00:00 icmp6_csum:19d3 + ]) + + OVS_VSWITCHD_STOP +@@ -10166,10 +10257,10 @@ OVS_WAIT_UNTIL([ovs-appctl -t ovs-ofctl exit]) + dnl Check this output. We only see the latter two packets, not the first. + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): table_id=6 cookie=0x0 total_len=106 reg0=0x1,reg1=0x4d2,reg2=0x1,reg3=0x1,reg4=0x1,in_port=1 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=1,tp_dst=2 udp_csum:553 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=1,tp_dst=2 udp_csum:553 + dnl + NXT_PACKET_IN (xid=0x0): table_id=6 cookie=0x0 total_len=106 ct_state=est|rpl|trk,ct_zone=1,ct_mark=0x1,ct_label=0x4d2000000000000000000000000,ct_nw_src=10.1.1.1,ct_nw_dst=10.1.1.2,ct_nw_proto=17,ct_tp_src=1,ct_tp_dst=2,ip,reg0=0x1,reg1=0x4d2,reg2=0x1,reg3=0x2,reg4=0x1,in_port=2 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=2,tp_dst=1 udp_csum:553 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=2,tp_dst=1 udp_csum:553 + ]) + + AT_CHECK([ovs-ofctl monitor br0 65534 invalid_ttl -P nxt_packet_in --detach --no-chdir --pidfile 2> ofctl_monitor.log]) +@@ -10187,10 +10278,10 @@ OVS_WAIT_UNTIL([ovs-appctl -t ovs-ofctl exit]) + dnl Check this output. We should see both packets + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): table_id=6 cookie=0x0 total_len=106 reg0=0x1,reg1=0x4d2,reg2=0x1,reg3=0x1,reg4=0x1,in_port=1 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=3,tp_dst=2 udp_csum:551 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=3,tp_dst=2 udp_csum:551 + dnl + NXT_PACKET_IN (xid=0x0): table_id=6 cookie=0x0 total_len=106 ct_state=est|rpl|trk,ct_zone=1,ct_mark=0x1,ct_label=0x4d2000000000000000000000000,ct_nw_src=10.1.1.1,ct_nw_dst=10.1.1.2,ct_nw_proto=17,ct_tp_src=3,ct_tp_dst=2,ip,reg0=0x1,reg1=0x4d2,reg2=0x1,reg3=0x2,reg4=0x1,in_port=2 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=2,tp_dst=3 udp_csum:551 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=2,tp_dst=3 udp_csum:551 + ]) + + OVS_VSWITCHD_STOP +@@ -10239,10 +10330,10 @@ dnl Note that the first packet doesn't have the ct_state bits set. This + dnl happens because the ct_state field is available only after recirc. + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0x0 total_len=106 in_port=1 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=1,tp_dst=2 udp_csum:553 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=1,tp_dst=2 udp_csum:553 + dnl + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x0 total_len=106 in_port=2 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=2,tp_dst=1 udp_csum:553 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=2,tp_dst=1 udp_csum:553 + ]) + + AT_CHECK([ovs-ofctl monitor br0 65534 invalid_ttl -P nxt_packet_in --detach --no-chdir --pidfile 2> ofctl_monitor.log]) +@@ -10261,10 +10352,10 @@ dnl Note that the first packet doesn't have the ct_state bits set. This + dnl happens because the ct_state field is available only after recirc. + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0x0 total_len=106 in_port=1 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=3,tp_dst=4 udp_csum:54f ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=3,tp_dst=4 udp_csum:54f + dnl + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x0 total_len=106 in_port=2 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=4,tp_dst=3 udp_csum:54f ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=4,tp_dst=3 udp_csum:54f + ]) + + dnl +@@ -10320,9 +10411,9 @@ dnl Note that the first packet doesn't have the ct_state bits set. This + dnl happens because the ct_state field is available only after recirc. + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0x0 total_len=126 in_port=1 (via action) data_len=126 (unbuffered) +-udp6,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,ipv6_src=2001:db8::1,ipv6_dst=2001:db8::2,ipv6_label=0x00000,nw_tos=112,nw_ecn=0,nw_ttl=128,tp_src=1,tp_dst=2 udp_csum:bfe2 ++udp6,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,ipv6_src=2001:db8::1,ipv6_dst=2001:db8::2,ipv6_label=0x00000,nw_tos=112,nw_ecn=0,nw_ttl=128,nw_frag=no,tp_src=1,tp_dst=2 udp_csum:bfe2 + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x0 total_len=126 ct_state=est|rpl|trk,ct_ipv6_src=2001:db8::1,ct_ipv6_dst=2001:db8::2,ct_nw_proto=17,ct_tp_src=1,ct_tp_dst=2,ipv6,in_port=2 (via action) data_len=126 (unbuffered) +-udp6,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,ipv6_src=2001:db8::2,ipv6_dst=2001:db8::1,ipv6_label=0x00000,nw_tos=112,nw_ecn=0,nw_ttl=128,tp_src=2,tp_dst=1 udp_csum:bfe2 ++udp6,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,ipv6_src=2001:db8::2,ipv6_dst=2001:db8::1,ipv6_label=0x00000,nw_tos=112,nw_ecn=0,nw_ttl=128,nw_frag=no,tp_src=2,tp_dst=1 udp_csum:bfe2 + ]) + + OVS_VSWITCHD_STOP +@@ -10433,7 +10524,7 @@ OVS_WAIT_UNTIL([ovs-appctl -t ovs-ofctl exit]) + dnl Check this output. Only one reply must be there + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x0 total_len=106 ct_state=est|rpl|trk,ct_nw_src=10.1.1.1,ct_nw_dst=10.1.1.2,ct_nw_proto=17,ct_tp_src=1,ct_tp_dst=2,ip,in_port=2 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=2,tp_dst=1 udp_csum:553 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=2,tp_dst=1 udp_csum:553 + dnl + OFPT_ECHO_REQUEST (xid=0x0): 0 bytes of payload + ]) +@@ -10467,7 +10558,7 @@ OVS_WAIT_UNTIL([ovs-appctl -t ovs-ofctl exit]) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0x0 total_len=86 ct_state=inv|trk,ipv6,in_port=2 (via action) data_len=86 (unbuffered) +-icmp6,vlan_tci=0x0000,dl_src=00:00:86:05:80:da,dl_dst=00:60:97:07:69:ea,ipv6_src=fe80::200:86ff:fe05:80da,ipv6_dst=fe80::260:97ff:fe07:69ea,ipv6_label=0x00000,nw_tos=0,nw_ecn=0,nw_ttl=255,icmp_type=135,icmp_code=0,nd_target=fe80::260:97ff:fe07:69ea,nd_sll=00:00:86:05:80:da,nd_tll=00:00:00:00:00:00 icmp6_csum:68bd ++icmp6,vlan_tci=0x0000,dl_src=00:00:86:05:80:da,dl_dst=00:60:97:07:69:ea,ipv6_src=fe80::200:86ff:fe05:80da,ipv6_dst=fe80::260:97ff:fe07:69ea,ipv6_label=0x00000,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,icmp_type=135,icmp_code=0,nd_target=fe80::260:97ff:fe07:69ea,nd_sll=00:00:86:05:80:da,nd_tll=00:00:00:00:00:00 icmp6_csum:68bd + ]) + + OVS_VSWITCHD_STOP +@@ -10523,16 +10614,16 @@ OVS_WAIT_UNTIL([ovs-appctl -t ovs-ofctl exit]) + dnl Check this output. We only see the latter two packets (for each zone), not the first. + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0x0 total_len=106 in_port=1 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=1,tp_dst=2 udp_csum:553 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=1,tp_dst=2 udp_csum:553 + dnl + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x0 total_len=106 ct_state=est|rpl|trk,ct_nw_src=10.1.1.1,ct_nw_dst=10.1.1.2,ct_nw_proto=17,ct_tp_src=1,ct_tp_dst=2,ip,in_port=2 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=2,tp_dst=1 udp_csum:553 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=2,tp_dst=1 udp_csum:553 + dnl + NXT_PACKET_IN (xid=0x0): cookie=0x0 total_len=106 in_port=3 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=1,tp_dst=2 udp_csum:553 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=1,tp_dst=2 udp_csum:553 + dnl + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x0 total_len=106 ct_state=est|rpl|trk,ct_zone=1,ct_nw_src=10.1.1.1,ct_nw_dst=10.1.1.2,ct_nw_proto=17,ct_tp_src=1,ct_tp_dst=2,ip,in_port=4 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=2,tp_dst=1 udp_csum:553 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=2,tp_dst=1 udp_csum:553 + ]) + + OVS_VSWITCHD_STOP +@@ -10579,10 +10670,10 @@ OVS_WAIT_UNTIL([ovs-appctl -t ovs-ofctl exit]) + dnl Check this output. We only see the latter two packets, not the first. + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x0 total_len=106 in_port=1 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=1,tp_dst=2 udp_csum:553 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=1,tp_dst=2 udp_csum:553 + dnl + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x0 total_len=106 ct_state=est|rpl|trk,ct_nw_src=10.1.1.1,ct_nw_dst=10.1.1.2,ct_nw_proto=17,ct_tp_src=1,ct_tp_dst=2,ip,in_port=2 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=2,tp_dst=1 udp_csum:553 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=2,tp_dst=1 udp_csum:553 + ]) + + OVS_VSWITCHD_STOP +@@ -10629,10 +10720,10 @@ OVS_WAIT_UNTIL([ovs-appctl -t ovs-ofctl exit]) + dnl Check this output. We only see the first and the last packet + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x0 total_len=47 ct_state=new|trk,ct_nw_src=172.16.0.1,ct_nw_dst=172.16.0.2,ct_nw_proto=17,ct_tp_src=41614,ct_tp_dst=5555,ip,in_port=1 (via action) data_len=47 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=e6:4c:47:35:28:c9,dl_dst=c6:f9:4e:cb:72:db,nw_src=172.16.0.1,nw_dst=172.16.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=41614,tp_dst=5555 udp_csum:2096 ++udp,vlan_tci=0x0000,dl_src=e6:4c:47:35:28:c9,dl_dst=c6:f9:4e:cb:72:db,nw_src=172.16.0.1,nw_dst=172.16.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=41614,tp_dst=5555 udp_csum:2096 + dnl + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x0 total_len=75 ct_state=rel|rpl|trk,ct_nw_src=172.16.0.1,ct_nw_dst=172.16.0.2,ct_nw_proto=17,ct_tp_src=41614,ct_tp_dst=5555,ip,in_port=2 (via action) data_len=75 (unbuffered) +-icmp,vlan_tci=0x0000,dl_src=c6:f9:4e:cb:72:db,dl_dst=e6:4c:47:35:28:c9,nw_src=172.16.0.2,nw_dst=172.16.0.1,nw_tos=192,nw_ecn=0,nw_ttl=64,icmp_type=3,icmp_code=3 icmp_csum:553f ++icmp,vlan_tci=0x0000,dl_src=c6:f9:4e:cb:72:db,dl_dst=e6:4c:47:35:28:c9,nw_src=172.16.0.2,nw_dst=172.16.0.1,nw_tos=192,nw_ecn=0,nw_ttl=64,nw_frag=no,icmp_type=3,icmp_code=3 icmp_csum:553f + ]) + + OVS_VSWITCHD_STOP +@@ -10681,19 +10772,19 @@ OVS_WAIT_UNTIL([ovs-appctl -t ovs-ofctl exit]) + dnl Check this output. + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0x0 total_len=106 in_port=1 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=1,tp_dst=2 udp_csum:553 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=1,tp_dst=2 udp_csum:553 + dnl + NXT_PACKET_IN (xid=0x0): cookie=0x0 total_len=106 in_port=1 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=3,tp_dst=4 udp_csum:54f ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=3,tp_dst=4 udp_csum:54f + dnl + NXT_PACKET_IN (xid=0x0): cookie=0x0 total_len=106 in_port=1 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=5,tp_dst=6 udp_csum:54b ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=5,tp_dst=6 udp_csum:54b + dnl + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x0 total_len=106 ct_state=est|rpl|trk,ct_mark=0x1,ct_nw_src=10.1.1.1,ct_nw_dst=10.1.1.2,ct_nw_proto=17,ct_tp_src=1,ct_tp_dst=2,ip,in_port=2 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=2,tp_dst=1 udp_csum:553 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=2,tp_dst=1 udp_csum:553 + dnl + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x0 total_len=106 ct_state=est|rpl|trk,ct_mark=0x3,ct_nw_src=10.1.1.1,ct_nw_dst=10.1.1.2,ct_nw_proto=17,ct_tp_src=3,ct_tp_dst=4,ip,in_port=2 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=4,tp_dst=3 udp_csum:54f ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=4,tp_dst=3 udp_csum:54f + ]) + + OVS_VSWITCHD_STOP +@@ -10738,10 +10829,10 @@ OVS_WAIT_UNTIL([ovs-appctl -t ovs-ofctl exit]) + dnl Check this output. + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x0 total_len=106 ct_state=est|rpl|trk,ct_label=0x1,ct_nw_src=10.1.1.1,ct_nw_dst=10.1.1.2,ct_nw_proto=17,ct_tp_src=1,ct_tp_dst=2,ip,in_port=2 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=2,tp_dst=1 udp_csum:553 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=2,tp_dst=1 udp_csum:553 + dnl + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x0 total_len=106 ct_state=est|rpl|trk,ct_label=0x2,ct_nw_src=10.1.1.1,ct_nw_dst=10.1.1.2,ct_nw_proto=17,ct_tp_src=3,ct_tp_dst=4,ip,in_port=2 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=4,tp_dst=3 udp_csum:54f ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=4,tp_dst=3 udp_csum:54f + ]) + + OVS_VSWITCHD_STOP +@@ -11152,16 +11243,16 @@ dnl Note that the first packet doesn't have the ct_state bits set. This + dnl happens because the ct_state field is available only after recirc. + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): cookie=0x0 total_len=106 in_port=1 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=1,tp_dst=2 udp_csum:553 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=1,tp_dst=2 udp_csum:553 + dnl + NXT_PACKET_IN (xid=0x0): cookie=0x0 total_len=106 in_port=1 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=1,tp_dst=2 udp_csum:553 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=1,tp_dst=2 udp_csum:553 + dnl + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x0 total_len=106 ct_state=est|rpl|trk,ct_nw_src=10.1.1.1,ct_nw_dst=10.1.1.2,ct_nw_proto=17,ct_tp_src=1,ct_tp_dst=2,ip,in_port=2 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=2,tp_dst=1 udp_csum:553 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=2,tp_dst=1 udp_csum:553 + dnl + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x0 total_len=106 in_port=2 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=2,tp_dst=1 udp_csum:553 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=2,tp_dst=1 udp_csum:553 + ]) + + dnl The next test verifies that ct_clear at the datapath only gets executed +@@ -11235,13 +11326,13 @@ OVS_WAIT_UNTIL([ovs-appctl -t ovs-ofctl exit]) + dnl Check this output. + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x0 total_len=106 ct_state=est|rpl|trk,ct_nw_src=10.1.2.100,ct_nw_dst=10.1.2.200,ct_nw_proto=17,ct_tp_src=6,ct_tp_dst=6,ip,in_port=2 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.2.200,nw_dst=10.1.2.100,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=6,tp_dst=6 udp_csum:221 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.2.200,nw_dst=10.1.2.100,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=6,tp_dst=6 udp_csum:221 + dnl + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x0 total_len=126 ct_state=est|rpl|trk,ct_ipv6_src=2001:db8::1,ct_ipv6_dst=2001:db8::2,ct_nw_proto=17,ct_tp_src=1,ct_tp_dst=2,ipv6,in_port=2 (via action) data_len=126 (unbuffered) +-udp6,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,ipv6_src=2001:db8::2,ipv6_dst=2001:db8::1,ipv6_label=0x00000,nw_tos=112,nw_ecn=0,nw_ttl=128,tp_src=2,tp_dst=1 udp_csum:bfe2 ++udp6,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,ipv6_src=2001:db8::2,ipv6_dst=2001:db8::1,ipv6_label=0x00000,nw_tos=112,nw_ecn=0,nw_ttl=128,nw_frag=no,tp_src=2,tp_dst=1 udp_csum:bfe2 + dnl + NXT_PACKET_IN (xid=0x0): table_id=1 cookie=0x0 total_len=106 ct_state=est|rpl|trk,ct_nw_src=10.1.1.1,ct_nw_dst=10.1.1.2,ct_nw_proto=17,ct_tp_src=1,ct_tp_dst=2,ip,in_port=2 (via action) data_len=106 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=2,tp_dst=1 udp_csum:553 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:0a,dl_dst=50:54:00:00:00:09,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=2,tp_dst=1 udp_csum:553 + ]) + + OVS_VSWITCHD_STOP diff --git a/tests/ofproto-macros.at b/tests/ofproto-macros.at index 736d9809cb..b18f0fbc1e 100644 --- a/tests/ofproto-macros.at @@ -9320,10 +61752,68 @@ index 736d9809cb..b18f0fbc1e 100644 /|WARN|/p /|ERR|/p /|EMER|/p" ${logs} +diff --git a/tests/ofproto.at b/tests/ofproto.at +index 156d3e058c..39c3b04704 100644 +--- a/tests/ofproto.at ++++ b/tests/ofproto.at +@@ -3108,7 +3108,7 @@ vlan_tci=0x0000,dl_src=00:10:20:30:40:50,dl_dst=00:01:02:03:04:05,dl_type=0x1234 + ovs-ofctl packet-out br0 "in_port=controller packet=002583dfb4000026b98cb0f908004500003eb7e200000011339bac11370dac100002d7730035002b8f6d86fb0100000100000000000006626c702d7873066e696369726103636f6d00000f00 actions=dec_ttl" + if test X"$1" = X"OFPR_INVALID_TTL"; then shift; + echo >>expout "OFPT_PACKET_IN: total_len=76 in_port=CONTROLLER (via invalid_ttl) data_len=76 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=00:26:b9:8c:b0:f9,dl_dst=00:25:83:df:b4:00,nw_src=172.17.55.13,nw_dst=172.16.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=55155,tp_dst=53 udp_csum:8f6d" ++udp,vlan_tci=0x0000,dl_src=00:26:b9:8c:b0:f9,dl_dst=00:25:83:df:b4:00,nw_src=172.17.55.13,nw_dst=172.16.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=55155,tp_dst=53 udp_csum:8f6d" + fi + + # OFPT_PORT_STATUS, OFPPR_ADD +@@ -3211,7 +3211,7 @@ vlan_tci=0x0000,dl_src=00:10:20:30:40:50,dl_dst=00:01:02:03:04:05,dl_type=0x1234 + ovs-ofctl -O OpenFlow12 packet-out br0 none dec_ttl '002583dfb4000026b98cb0f908004500003eb7e200000011339bac11370dac100002d7730035002b8f6d86fb0100000100000000000006626c702d7873066e696369726103636f6d00000f00' + if test X"$1" = X"OFPR_INVALID_TTL"; then shift; + echo >>expout "OFPT_PACKET_IN (OF1.2): total_len=76 in_port=ANY (via invalid_ttl) data_len=76 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=00:26:b9:8c:b0:f9,dl_dst=00:25:83:df:b4:00,nw_src=172.17.55.13,nw_dst=172.16.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=55155,tp_dst=53 udp_csum:8f6d" ++udp,vlan_tci=0x0000,dl_src=00:26:b9:8c:b0:f9,dl_dst=00:25:83:df:b4:00,nw_src=172.17.55.13,nw_dst=172.16.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=55155,tp_dst=53 udp_csum:8f6d" + fi + + # OFPT_PORT_STATUS, OFPPR_ADD +@@ -3325,7 +3325,7 @@ vlan_tci=0x0000,dl_src=00:10:20:30:40:50,dl_dst=00:01:02:03:04:05,dl_type=0x1234 + ovs-ofctl -O OpenFlow13 packet-out br0 none dec_ttl '002583dfb4000026b98cb0f908004500003eb7e200000011339bac11370dac100002d7730035002b8f6d86fb0100000100000000000006626c702d7873066e696369726103636f6d00000f00' + if test X"$1" = X"OFPR_INVALID_TTL"; then shift; + echo >>expout "OFPT_PACKET_IN (OF1.3): total_len=76 in_port=ANY (via invalid_ttl) data_len=76 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=00:26:b9:8c:b0:f9,dl_dst=00:25:83:df:b4:00,nw_src=172.17.55.13,nw_dst=172.16.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=55155,tp_dst=53 udp_csum:8f6d" ++udp,vlan_tci=0x0000,dl_src=00:26:b9:8c:b0:f9,dl_dst=00:25:83:df:b4:00,nw_src=172.17.55.13,nw_dst=172.16.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=55155,tp_dst=53 udp_csum:8f6d" + fi + + # OFPT_PORT_STATUS, OFPPR_ADD +@@ -3459,7 +3459,7 @@ vlan_tci=0x0000,dl_src=00:10:20:30:40:50,dl_dst=00:01:02:03:04:05,dl_type=0x1234 + ovs_ofctl -O OpenFlow14 packet-out br0 none dec_ttl '002583dfb4000026b98cb0f908004500003eb7e200000011339bac11370dac100002d7730035002b8f6d86fb0100000100000000000006626c702d7873066e696369726103636f6d00000f00' + if test X"$1" = X"OFPR_INVALID_TTL"; then shift; + echo >>expout "OFPT_PACKET_IN (OF1.4): total_len=76 in_port=ANY (via invalid_ttl) data_len=76 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=00:26:b9:8c:b0:f9,dl_dst=00:25:83:df:b4:00,nw_src=172.17.55.13,nw_dst=172.16.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=55155,tp_dst=53 udp_csum:8f6d" ++udp,vlan_tci=0x0000,dl_src=00:26:b9:8c:b0:f9,dl_dst=00:25:83:df:b4:00,nw_src=172.17.55.13,nw_dst=172.16.0.2,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=55155,tp_dst=53 udp_csum:8f6d" + fi + + # OFPT_PORT_STATUS, OFPPR_ADD +@@ -4434,7 +4434,7 @@ ovs-appctl -t ovs-ofctl exit + + AT_CHECK([sed 's/ (xid=0x[[0-9a-fA-F]]*)//' monitor.log], [0], [dnl + OFPT_PACKET_IN (OF1.3): total_len=32 packet_type=(1,0x800),metadata=0xfafafafa5a5a5a5a,in_port=CONTROLLER (via action) data_len=32 (unbuffered) +-packet_type=(1,0x800),nw_src=10.0.0.20,nw_dst=10.0.0.30,nw_proto=17,nw_tos=0,nw_ecn=0,nw_ttl=255,tp_src=100,tp_dst=200 udp_csum:ea78 ++packet_type=(1,0x800),nw_src=10.0.0.20,nw_dst=10.0.0.30,nw_proto=17,nw_tos=0,nw_ecn=0,nw_ttl=255,nw_frag=no,tp_src=100,tp_dst=200 udp_csum:ea78 + OFPT_BARRIER_REPLY (OF1.3): + ]) + diff --git a/tests/ovs-macros.at b/tests/ovs-macros.at -index 66545da572..e6c5bc6e94 100644 +index 66545da572..d09dbb4cd5 100644 --- a/tests/ovs-macros.at +++ b/tests/ovs-macros.at +@@ -134,7 +134,7 @@ parent_pid () { + # e.g. Alpine Linux) is noncompliant, so we use a Linux-specific approach + # when it's available. We check the format of the status file to avoid + # the NetBSD file with the same name but different contents. +- if egrep '^PPid:[[:space:]]*[0-9]*$' /proc/$1/status > /dev/null 2>&1; then ++ if grep -E '^PPid:[[:space:]]*[0-9]*$' /proc/$1/status > /dev/null 2>&1; then + sed -n 's/^PPid: \([0-9]*\)/\1/p' /proc/$1/status + else + ps -o ppid= -p $1 @@ -259,7 +259,20 @@ dnl Executes shell COMMAND in a loop until it returns zero. If COMMAND does dnl not return zero within a reasonable time limit, executes the commands dnl in IF-FAILED (if provided) and fails the test. @@ -9358,6 +61848,23 @@ index 66545da572..e6c5bc6e94 100644 [AT_LINE], [while $1])]) dnl OVS_APP_EXIT_AND_WAIT(DAEMON) +diff --git a/tests/ovs-ofctl.at b/tests/ovs-ofctl.at +index 267711bfa4..c9c67f2b1e 100644 +--- a/tests/ovs-ofctl.at ++++ b/tests/ovs-ofctl.at +@@ -3243,9 +3243,9 @@ AT_CHECK([ovs-testcontroller -vsyslog:off --detach --no-chdir --pidfile punix:te + OVS_WAIT_UNTIL([test -e testcontroller]) + + dnl check for some of the initial handshake messages +-OVS_WAIT_UNTIL([egrep "OFPT_FEATURES_REQUEST" snoopbr0.txt >/dev/null 2>&1]) +-OVS_WAIT_UNTIL([egrep "OFPT_FEATURES_REPLY" snoopbr0.txt >/dev/null 2>&1]) +-OVS_WAIT_UNTIL([egrep "OFPT_SET_CONFIG" snoopbr0.txt >/dev/null 2>&1]) ++OVS_WAIT_UNTIL([grep -E "OFPT_FEATURES_REQUEST" snoopbr0.txt >/dev/null 2>&1]) ++OVS_WAIT_UNTIL([grep -E "OFPT_FEATURES_REPLY" snoopbr0.txt >/dev/null 2>&1]) ++OVS_WAIT_UNTIL([grep -E "OFPT_SET_CONFIG" snoopbr0.txt >/dev/null 2>&1]) + + dnl need to suppress the 'connection failed' WARN message in ovs-vswitchd + dnl because we need ovs-vswitchd to have the controller config before starting diff --git a/tests/ovs-vswitchd.at b/tests/ovs-vswitchd.at index bba4fea2bc..977b2eba1f 100644 --- a/tests/ovs-vswitchd.at @@ -9990,7 +62497,7 @@ index 876cb836cd..e672c13b27 100644 db.tmp], [0], [stdout], [stderr]) PARSE_LISTENING_PORT([listener.log], [BAD_TCP_PORT]) diff --git a/tests/pmd.at b/tests/pmd.at -index a2f9d34a2a..3962dd2bd9 100644 +index a2f9d34a2a..a2832b544c 100644 --- a/tests/pmd.at +++ b/tests/pmd.at @@ -199,7 +199,7 @@ pmd thread numa_id core_id : @@ -10047,6 +62554,108 @@ index a2f9d34a2a..3962dd2bd9 100644 AT_SETUP([PMD - stats]) OVS_VSWITCHD_START([add-port br0 p0 -- set Interface p0 ofport_request=7 type=dummy-pmd options:n_rxq=4], [], [], [DUMMY_NUMA]) +@@ -460,7 +498,7 @@ OVS_WAIT_UNTIL([ovs-appctl -t ovs-ofctl exit]) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN2 (xid=0x0): cookie=0x0 total_len=106 in_port=1 (via action) data_len=106 (unbuffered) +-icmp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.0.0.2,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,icmp_type=8,icmp_code=0 icmp_csum:13fc ++icmp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.0.0.2,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,icmp_type=8,icmp_code=0 icmp_csum:13fc + ]) + + AT_CHECK([ovs-vsctl set interface p1 options:n_rxq=4]) +@@ -482,7 +520,7 @@ OVS_WAIT_UNTIL([ovs-appctl -t ovs-ofctl exit]) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN2 (xid=0x0): cookie=0x0 total_len=106 in_port=1 (via action) data_len=106 (unbuffered) +-icmp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.0.0.2,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,icmp_type=8,icmp_code=0 icmp_csum:13fc ++icmp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.0.0.2,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,icmp_type=8,icmp_code=0 icmp_csum:13fc + ]) + + dnl Check resetting to default number of rx queues after removal from the db. +@@ -533,9 +571,9 @@ AT_CHECK([ovs-appctl dpctl/dump-flows | flow_dump_prepend_pmd], [0], [dnl + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN2 (xid=0x0): cookie=0x0 total_len=106 in_port=1 (via action) data_len=106 (unbuffered) +-icmp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.0.0.2,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,icmp_type=8,icmp_code=0 icmp_csum:13fc ++icmp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.0.0.2,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,icmp_type=8,icmp_code=0 icmp_csum:13fc + NXT_PACKET_IN2 (xid=0x0): cookie=0x0 total_len=106 in_port=1 (via action) data_len=106 (unbuffered) +-icmp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.0.0.2,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,icmp_type=8,icmp_code=0 icmp_csum:13fc ++icmp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.0.0.2,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,icmp_type=8,icmp_code=0 icmp_csum:13fc + ]) + + OVS_VSWITCHD_STOP +@@ -568,7 +606,7 @@ OVS_WAIT_UNTIL([ovs-appctl -t ovs-ofctl exit]) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN2 (xid=0x0): cookie=0x0 total_len=106 in_port=1 (via action) data_len=106 (unbuffered) +-icmp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.0.0.2,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,icmp_type=8,icmp_code=0 icmp_csum:13fc ++icmp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.0.0.2,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,icmp_type=8,icmp_code=0 icmp_csum:13fc + ]) + + AT_CHECK([ovs-ofctl monitor br0 65534 invalid_ttl --detach --no-chdir --pidfile 2> ofctl_monitor.log]) +@@ -580,7 +618,7 @@ OVS_WAIT_UNTIL([ovs-appctl -t ovs-ofctl exit]) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN2 (xid=0x0): cookie=0x0 total_len=106 in_port=2 (via action) data_len=106 (unbuffered) +-icmp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.0.0.2,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,icmp_type=8,icmp_code=0 icmp_csum:13fc ++icmp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.0.0.2,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,icmp_type=8,icmp_code=0 icmp_csum:13fc + ]) + + AT_CHECK([ovs-vsctl set Interface p2 options:numa_id=1]) +@@ -601,7 +639,7 @@ OVS_WAIT_UNTIL([ovs-appctl -t ovs-ofctl exit]) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN2 (xid=0x0): cookie=0x0 total_len=106 in_port=1 (via action) data_len=106 (unbuffered) +-icmp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.0.0.2,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,icmp_type=8,icmp_code=0 icmp_csum:13fc ++icmp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.0.0.2,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,icmp_type=8,icmp_code=0 icmp_csum:13fc + ]) + + AT_CHECK([ovs-ofctl monitor br0 65534 invalid_ttl --detach --no-chdir --pidfile 2> ofctl_monitor.log]) +@@ -613,7 +651,7 @@ OVS_WAIT_UNTIL([ovs-appctl -t ovs-ofctl exit]) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN2 (xid=0x0): cookie=0x0 total_len=106 in_port=2 (via action) data_len=106 (unbuffered) +-icmp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.0.0.2,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,icmp_type=8,icmp_code=0 icmp_csum:13fc ++icmp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.0.0.2,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,icmp_type=8,icmp_code=0 icmp_csum:13fc + ]) + + AT_CHECK([ovs-vsctl set Interface p1 options:numa_id=8]) +@@ -634,7 +672,7 @@ OVS_WAIT_UNTIL([ovs-appctl -t ovs-ofctl exit]) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN2 (xid=0x0): cookie=0x0 total_len=106 in_port=1 (via action) data_len=106 (unbuffered) +-icmp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.0.0.2,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,icmp_type=8,icmp_code=0 icmp_csum:13fc ++icmp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.0.0.2,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,icmp_type=8,icmp_code=0 icmp_csum:13fc + ]) + + AT_CHECK([ovs-ofctl monitor br0 65534 invalid_ttl --detach --no-chdir --pidfile 2> ofctl_monitor.log]) +@@ -646,7 +684,7 @@ OVS_WAIT_UNTIL([ovs-appctl -t ovs-ofctl exit]) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN2 (xid=0x0): cookie=0x0 total_len=106 in_port=2 (via action) data_len=106 (unbuffered) +-icmp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.0.0.2,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,icmp_type=8,icmp_code=0 icmp_csum:13fc ++icmp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.0.0.2,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,icmp_type=8,icmp_code=0 icmp_csum:13fc + ]) + + +@@ -726,7 +764,7 @@ OVS_WAIT_UNTIL([ovs-appctl -t ovs-ofctl exit]) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN2 (xid=0x0): cookie=0x0 total_len=106 in_port=1 (via action) data_len=106 (unbuffered) +-icmp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.0.0.2,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,icmp_type=8,icmp_code=0 icmp_csum:13fc ++icmp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.0.0.2,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,icmp_type=8,icmp_code=0 icmp_csum:13fc + ]) + + AT_CHECK([ovs-vsctl del-port br0 p1]) +@@ -741,7 +779,7 @@ OVS_WAIT_UNTIL([ovs-appctl -t ovs-ofctl exit]) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN2 (xid=0x0): cookie=0x0 total_len=106 in_port=1 (via action) data_len=106 (unbuffered) +-icmp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.0.0.2,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,icmp_type=8,icmp_code=0 icmp_csum:13fc ++icmp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.0.0.2,nw_dst=10.0.0.1,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,icmp_type=8,icmp_code=0 icmp_csum:13fc + ]) + + OVS_VSWITCHD_STOP @@ -1075,15 +1113,15 @@ AT_SETUP([PMD - dpif configuration]) OVS_VSWITCHD_START([], [], [], [--dummy-numa 0,0]) AT_CHECK([ovs-vsctl add-port br0 p1 -- set Interface p1 type=dummy-pmd]) @@ -10288,6 +62897,19 @@ index 19a0b125b9..8b9f5c7525 100644 # OVS_CHECK_VXLAN() # # Do basic check for vxlan functionality, skip the test if it's not there. +diff --git a/tests/system-dpdk-macros.at b/tests/system-dpdk-macros.at +index ef0e84e939..2579098a04 100644 +--- a/tests/system-dpdk-macros.at ++++ b/tests/system-dpdk-macros.at +@@ -6,7 +6,7 @@ + m4_define([OVS_DPDK_PRE_CHECK], + [dnl Check Hugepages + AT_CHECK([cat /proc/meminfo], [], [stdout]) +- AT_SKIP_IF([egrep 'HugePages_Free: *0' stdout], [], [stdout]) ++ AT_SKIP_IF([grep -E 'HugePages_Free: *0' stdout], [], [stdout]) + AT_CHECK([mount], [], [stdout]) + AT_CHECK([grep 'hugetlbfs' stdout], [], [stdout], []) + diff --git a/tests/system-dpdk.at b/tests/system-dpdk.at index c3ee6990ca..7d2715c4a7 100644 --- a/tests/system-dpdk.at @@ -10314,6 +62936,46 @@ index c3ee6990ca..7d2715c4a7 100644 AT_CHECK([ovs-appctl dpif-netdev/miniflow-parser-set autovalidator], [0], [dnl Miniflow extract implementation set to autovalidator. ]) +diff --git a/tests/system-kmod-macros.at b/tests/system-kmod-macros.at +index 86d633ac4f..f0aaae63eb 100644 +--- a/tests/system-kmod-macros.at ++++ b/tests/system-kmod-macros.at +@@ -200,6 +200,13 @@ m4_define([OVS_CHECK_KERNEL_EXCL], + AT_SKIP_IF([ ! ( test $version -lt $1 || ( test $version -eq $1 && test $sublevel -lt $2 ) || test $version -gt $3 || ( test $version -eq $3 && test $sublevel -gt $4 ) ) ]) + ]) + ++# CHECK_LATER_IPV6_FRAGMENTS() ++# ++# Upstream kernels beetween 4.20 and 5.19 are not parsing IPv6 fragments ++# correctly. The issue was also backported in some older distribution ++# kernels, so kernels below 4.20 are not reliable. ++m4_define([CHECK_LATER_IPV6_FRAGMENTS], [OVS_CHECK_MIN_KERNEL(5, 19)]) ++ + # VSCTL_ADD_DATAPATH_TABLE() + # + # Create system datapath table "system" for kernel tests in ovsdb +diff --git a/tests/system-offloads-traffic.at b/tests/system-offloads-traffic.at +index 80bc1dd5c3..14a332f5ed 100644 +--- a/tests/system-offloads-traffic.at ++++ b/tests/system-offloads-traffic.at +@@ -90,7 +90,7 @@ AT_CHECK([tc -o -s -d filter show dev ovs-p0 ingress | + rate 100Kbit burst 1280b + ]) + AT_CHECK([tc -s -d filter show dev ovs-p0 ingress | +- egrep "basic|matchall" > /dev/null], [0]) ++ grep -E "basic|matchall" > /dev/null], [0]) + OVS_TRAFFIC_VSWITCHD_STOP + AT_CLEANUP + +@@ -139,7 +139,7 @@ AT_CHECK([tc -o -s -d filter show dev ovs-p0 ingress | + pkts_rate 100000 pkts_burst 10000 + ]) + AT_CHECK([tc -s -d filter show dev ovs-p0 ingress | +- egrep "basic|matchall" > /dev/null], [0]) ++ grep -E "basic|matchall" > /dev/null], [0]) + OVS_TRAFFIC_VSWITCHD_STOP + AT_CLEANUP + diff --git a/tests/system-route.at b/tests/system-route.at index 1714273e35..270956d13f 100644 --- a/tests/system-route.at @@ -10333,10 +62995,57 @@ index 1714273e35..270956d13f 100644 dnl Delete ip address. AT_CHECK([ip addr del 10.0.0.17/24 dev p1-route], [0], [stdout]) diff --git a/tests/system-traffic.at b/tests/system-traffic.at -index f22d86e466..36e10aa4a8 100644 +index f22d86e466..69de604fa0 100644 --- a/tests/system-traffic.at +++ b/tests/system-traffic.at -@@ -218,6 +218,7 @@ OVS_TRAFFIC_VSWITCHD_STOP +@@ -192,6 +192,46 @@ NS_CHECK_EXEC([at_ns0], [ping6 -s 3200 -q -c 3 -i 0.3 -w 2 fc00:1::2 | FORMAT_PI + OVS_TRAFFIC_VSWITCHD_STOP + AT_CLEANUP + ++AT_SETUP([datapath - ping6 between two ports with header modify]) ++OVS_TRAFFIC_VSWITCHD_START() ++CHECK_LATER_IPV6_FRAGMENTS() ++ ++AT_CHECK([ovs-ofctl add-flow br0 "actions=normal"]) ++ ++ADD_NAMESPACES(at_ns0, at_ns1) ++ ++ADD_VETH(p0, at_ns0, br0, "fc00::1/96", e4:11:22:33:44:55) ++ADD_VETH(p1, at_ns1, br0, "fc00::2/96", e4:11:22:33:44:54) ++NS_CHECK_EXEC([at_ns0], [ip -6 neigh add fc00::3 lladdr e4:11:22:33:44:54 dev p0]) ++ ++dnl Linux seems to take a little time to get its IPv6 stack in order. Without ++dnl waiting, we get occasional failures due to the following error: ++dnl "connect: Cannot assign requested address" ++OVS_WAIT_UNTIL([ip netns exec at_ns0 ping6 -c 1 fc00::2]) ++OVS_WAIT_UNTIL([ip netns exec at_ns1 ping6 -c 1 fc00::1]) ++ ++AT_DATA([flows.txt], [dnl ++priority=100,in_port=ovs-p0,ipv6,ipv6_src=fc00::1,ipv6_dst=fc00::3,actions=set_field:fc00::2->ipv6_dst,ovs-p1 ++priority=100,in_port=ovs-p1,ipv6,ipv6_src=fc00::2,ipv6_dst=fc00::1,actions=set_field:fc00::3->ipv6_src,ovs-p0 ++priority=0,actions=NORMAL ++]) ++ ++AT_CHECK([ovs-ofctl del-flows br0]) ++AT_CHECK([ovs-ofctl add-flows br0 flows.txt]) ++ ++NS_CHECK_EXEC([at_ns0], [ping6 -q -c 3 -i 0.3 -w 2 fc00::3 | FORMAT_PING], [0], [dnl ++3 packets transmitted, 3 received, 0% packet loss, time 0ms ++]) ++NS_CHECK_EXEC([at_ns0], [ping6 -s 1600 -q -c 3 -i 0.3 -w 2 fc00::3 | FORMAT_PING], [0], [dnl ++3 packets transmitted, 3 received, 0% packet loss, time 0ms ++]) ++NS_CHECK_EXEC([at_ns0], [ping6 -s 3200 -q -c 3 -i 0.3 -w 2 fc00::3 | FORMAT_PING], [0], [dnl ++3 packets transmitted, 3 received, 0% packet loss, time 0ms ++]) ++ ++OVS_TRAFFIC_VSWITCHD_STOP ++AT_CLEANUP ++ + AT_SETUP([datapath - ping over bond]) + OVS_TRAFFIC_VSWITCHD_START() + +@@ -218,6 +258,7 @@ OVS_TRAFFIC_VSWITCHD_STOP AT_CLEANUP AT_SETUP([datapath - ping over vxlan tunnel]) @@ -10344,7 +63053,7 @@ index f22d86e466..36e10aa4a8 100644 OVS_CHECK_VXLAN() OVS_TRAFFIC_VSWITCHD_START() -@@ -258,7 +259,55 @@ NS_CHECK_EXEC([at_ns0], [ping -s 3200 -q -c 3 -i 0.3 -w 2 10.1.1.100 | FORMAT_PI +@@ -258,7 +299,55 @@ NS_CHECK_EXEC([at_ns0], [ping -s 3200 -q -c 3 -i 0.3 -w 2 10.1.1.100 | FORMAT_PI OVS_TRAFFIC_VSWITCHD_STOP AT_CLEANUP @@ -10400,7 +63109,7 @@ index f22d86e466..36e10aa4a8 100644 OVS_CHECK_VXLAN_UDP6ZEROCSUM() OVS_TRAFFIC_VSWITCHD_START() -@@ -302,6 +351,7 @@ OVS_TRAFFIC_VSWITCHD_STOP +@@ -302,6 +391,7 @@ OVS_TRAFFIC_VSWITCHD_STOP AT_CLEANUP AT_SETUP([datapath - ping over gre tunnel]) @@ -10408,7 +63117,7 @@ index f22d86e466..36e10aa4a8 100644 OVS_CHECK_KERNEL_EXCL(3, 10, 4, 15) OVS_CHECK_GRE() -@@ -343,6 +393,7 @@ OVS_TRAFFIC_VSWITCHD_STOP +@@ -343,6 +433,7 @@ OVS_TRAFFIC_VSWITCHD_STOP AT_CLEANUP AT_SETUP([datapath - ping over ip6gre L2 tunnel]) @@ -10416,7 +63125,7 @@ index f22d86e466..36e10aa4a8 100644 OVS_CHECK_KERNEL_EXCL(3, 10, 4, 15) OVS_CHECK_GRE() OVS_CHECK_ERSPAN() -@@ -383,6 +434,7 @@ AT_CLEANUP +@@ -383,6 +474,7 @@ AT_CLEANUP AT_SETUP([datapath - ping over erspan v1 tunnel]) @@ -10424,7 +63133,7 @@ index f22d86e466..36e10aa4a8 100644 OVS_CHECK_KERNEL_EXCL(3, 10, 4, 15) OVS_CHECK_GRE() OVS_CHECK_ERSPAN() -@@ -419,6 +471,7 @@ OVS_TRAFFIC_VSWITCHD_STOP +@@ -419,6 +511,7 @@ OVS_TRAFFIC_VSWITCHD_STOP AT_CLEANUP AT_SETUP([datapath - ping over erspan v2 tunnel]) @@ -10432,7 +63141,7 @@ index f22d86e466..36e10aa4a8 100644 OVS_CHECK_KERNEL_EXCL(3, 10, 4, 15) OVS_CHECK_GRE() OVS_CHECK_ERSPAN() -@@ -455,6 +508,7 @@ OVS_TRAFFIC_VSWITCHD_STOP +@@ -455,6 +548,7 @@ OVS_TRAFFIC_VSWITCHD_STOP AT_CLEANUP AT_SETUP([datapath - ping over ip6erspan v1 tunnel]) @@ -10440,7 +63149,7 @@ index f22d86e466..36e10aa4a8 100644 OVS_CHECK_KERNEL_EXCL(3, 10, 4, 15) OVS_CHECK_GRE() OVS_CHECK_ERSPAN() -@@ -494,6 +548,7 @@ OVS_TRAFFIC_VSWITCHD_STOP +@@ -494,6 +588,7 @@ OVS_TRAFFIC_VSWITCHD_STOP AT_CLEANUP AT_SETUP([datapath - ping over ip6erspan v2 tunnel]) @@ -10448,7 +63157,7 @@ index f22d86e466..36e10aa4a8 100644 OVS_CHECK_KERNEL_EXCL(3, 10, 4, 15) OVS_CHECK_GRE() OVS_CHECK_ERSPAN() -@@ -534,6 +589,7 @@ OVS_TRAFFIC_VSWITCHD_STOP +@@ -534,6 +629,7 @@ OVS_TRAFFIC_VSWITCHD_STOP AT_CLEANUP AT_SETUP([datapath - ping over geneve tunnel]) @@ -10456,7 +63165,7 @@ index f22d86e466..36e10aa4a8 100644 OVS_CHECK_GENEVE() OVS_TRAFFIC_VSWITCHD_START() -@@ -575,6 +631,7 @@ OVS_TRAFFIC_VSWITCHD_STOP +@@ -575,6 +671,7 @@ OVS_TRAFFIC_VSWITCHD_STOP AT_CLEANUP AT_SETUP([datapath - ping over geneve tunnel, delete flow regression]) @@ -10464,7 +63173,7 @@ index f22d86e466..36e10aa4a8 100644 OVS_CHECK_GENEVE() OVS_TRAFFIC_VSWITCHD_START() -@@ -629,6 +686,7 @@ OVS_TRAFFIC_VSWITCHD_STOP(["/|ERR|/d +@@ -629,6 +726,7 @@ OVS_TRAFFIC_VSWITCHD_STOP(["/|ERR|/d AT_CLEANUP AT_SETUP([datapath - flow resume with geneve tun_metadata]) @@ -10472,7 +63181,16 @@ index f22d86e466..36e10aa4a8 100644 OVS_CHECK_GENEVE() OVS_TRAFFIC_VSWITCHD_START() -@@ -680,6 +738,7 @@ OVS_TRAFFIC_VSWITCHD_STOP +@@ -666,7 +764,7 @@ NS_CHECK_EXEC([at_ns0], [ping -q -c 3 10.1.1.100 | FORMAT_PING], [0], [dnl + ]) + + dnl Test OVS handles TLV map modifictions properly when restores frozen state. +-NS_CHECK_EXEC([at_ns0], [ping 10.1.1.100 > /dev/null &]) ++NETNS_DAEMONIZE([at_ns0], [ping 10.1.1.100 > /dev/null], [ping0.pid]) + + AT_CHECK([ovs-ofctl add-tlv-map br0 "{class=0xffff,type=0x88,len=4}->tun_metadata1"]) + sleep 1 +@@ -680,6 +778,7 @@ OVS_TRAFFIC_VSWITCHD_STOP AT_CLEANUP AT_SETUP([datapath - ping over geneve6 tunnel]) @@ -10480,7 +63198,7 @@ index f22d86e466..36e10aa4a8 100644 OVS_CHECK_GENEVE_UDP6ZEROCSUM() OVS_TRAFFIC_VSWITCHD_START() -@@ -723,6 +782,7 @@ OVS_TRAFFIC_VSWITCHD_STOP +@@ -723,6 +822,7 @@ OVS_TRAFFIC_VSWITCHD_STOP AT_CLEANUP AT_SETUP([datapath - ping over gre tunnel by simulated packets]) @@ -10488,7 +63206,19 @@ index f22d86e466..36e10aa4a8 100644 OVS_CHECK_MIN_KERNEL(3, 10) OVS_TRAFFIC_VSWITCHD_START() -@@ -769,6 +829,7 @@ OVS_TRAFFIC_VSWITCHD_STOP +@@ -759,16 +859,17 @@ dnl ADD_NATIVE_TUNNEL([gretap], [ns_gre0], [at_ns0], [172.31.1.100], [10.1.1.1/2 + dnl Now, check the overlay by sending out raw arp and icmp packets. + ovs-ofctl -O OpenFlow13 packet-out br-underlay "in_port=1 packet=f2ff00000002f2ff00000003080045000042ec2c4000402ff3bcac1f0101ac1f016400006558fffffffffffff2ff0000000408060001080006040001f2ff000000040a0101010000000000000a010164 actions=NORMAL" + +-OVS_WAIT_UNTIL([cat p0.pcap | egrep "IP 172.31.1.100 > 172.31.1.1: GREv0, length 46: ARP, Reply 10.1.1.100 is-at f2:ff:00:00:00:01.* length 28" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p0.pcap | grep -E "IP 172.31.1.100 > 172.31.1.1: GREv0, length 46: ARP, Reply 10.1.1.100 is-at f2:ff:00:00:00:01.* length 28" 2>&1 1>/dev/null]) + + ovs-ofctl -O OpenFlow13 packet-out br-underlay "in_port=1 packet=f2ff00000002f2ff0000000308004500007aec8e4000402ff322ac1f0101ac1f016400006558f2ff00000001f2ff00000004080045000054548f40004001cfb30a0101010a0101640800e6e829270003e1a3435b00000000ff1a050000000000101112131415161718191a1b1c1d1e1f202122232425262728292a2b2c2d2e2f3031323334353637 actions=NORMAL" + +-OVS_WAIT_UNTIL([cat p0.pcap | egrep "IP 172.31.1.100 > 172.31.1.1: GREv0, length 102: IP 10.1.1.100 > 10.1.1.1: ICMP echo reply,.* length 64$" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p0.pcap | grep -E "IP 172.31.1.100 > 172.31.1.1: GREv0, length 102: IP 10.1.1.100 > 10.1.1.1: ICMP echo reply,.* length 64$" 2>&1 1>/dev/null]) + + OVS_TRAFFIC_VSWITCHD_STOP AT_CLEANUP AT_SETUP([datapath - ping over erspan v1 tunnel by simulated packets]) @@ -10496,7 +63226,25 @@ index f22d86e466..36e10aa4a8 100644 OVS_CHECK_MIN_KERNEL(3, 10) OVS_TRAFFIC_VSWITCHD_START() -@@ -817,6 +878,7 @@ OVS_TRAFFIC_VSWITCHD_STOP +@@ -803,20 +904,21 @@ dnl Okay, now send out an arp request from 10.1.1.1 for 10.1.1.100 in erspan. + ovs-ofctl -O OpenFlow13 packet-out br-underlay "in_port=1 packet=f2ff00000002f2ff0000000308004500004e151d4000402fcac0ac1f0101ac1f0164100088be000000061000000100000007fffffffffffff2ff0000000408060001080006040001f2ff000000040a0101010000000000000a010164 actions=normal" + + dnl 0002 is arp reply, followed by mac address of 10.1.1.100. +-OVS_WAIT_UNTIL([cat p0.pcap | egrep "0x0030: 0806 0001 0800 0604 0002 f2ff 0000 0001" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p0.pcap | egrep "0x0040: 0a01 0164 f2ff 0000 0004 0a01 0101" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p0.pcap | grep -E "0x0030: 0806 0001 0800 0604 0002 f2ff 0000 0001" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p0.pcap | grep -E "0x0040: 0a01 0164 f2ff 0000 0004 0a01 0101" 2>&1 1>/dev/null]) + + dnl Okay, now check the overlay with raw icmp packets. +-AT_FAIL_IF([cat p0.pcap | egrep "IP 172.31.1.100 > 172.31.1.1: GREv0,.* length 122" 2>&1 1>/dev/null]) ++AT_FAIL_IF([cat p0.pcap | grep -E "IP 172.31.1.100 > 172.31.1.1: GREv0,.* length 122" 2>&1 1>/dev/null]) + + ovs-ofctl -O OpenFlow13 packet-out br-underlay "in_port=1 packet=f2ff00000002f2ff0000000308004500008e70cb4000402f6ed2ac1f0101ac1f0164100088be000000051000000100000007f2ff00000001f2ff0000000408004500005c4a3340004001da070a0101010a010164080084f238fb0001f36a6b5b0000000021870e0000000000101112131415161718191a1b1c1d1e1f202122232425262728292a2b2c2d2e2f303132333435363738393a3b3c3d3e3f actions=normal" + +-OVS_WAIT_UNTIL([cat p0.pcap | egrep "IP 172.31.1.100 > 172.31.1.1: GREv0,.* length 122" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p0.pcap | grep -E "IP 172.31.1.100 > 172.31.1.1: GREv0,.* length 122" 2>&1 1>/dev/null]) + + OVS_TRAFFIC_VSWITCHD_STOP AT_CLEANUP AT_SETUP([datapath - ping over erspan v2 tunnel by simulated packets]) @@ -10504,7 +63252,28 @@ index f22d86e466..36e10aa4a8 100644 OVS_CHECK_MIN_KERNEL(3, 10) OVS_TRAFFIC_VSWITCHD_START() -@@ -870,6 +932,7 @@ OVS_TRAFFIC_VSWITCHD_STOP +@@ -854,22 +956,23 @@ NS_CHECK_EXEC([at_ns0], [ping -q -c 3 -i 0.3 -w 2 172.31.1.100 | FORMAT_PING], [ + dnl Okay, send raw arp request and icmp echo request. + ovs-ofctl -O OpenFlow13 packet-out br-underlay "in_port=1 packet=f2ff00000002f2ff00000003080045000052373d4000402fa89cac1f0101ac1f0164100088be00000006200000016f54b41700008078fffffffffffff2ff0000000408060001080006040001f2ff000000040a0101010000000000000a010164 actions=normal" + +-OVS_WAIT_UNTIL([cat p0.pcap | egrep "0x0030: 0000 0001 0806 0001 0800 0604 0002 f2ff" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p0.pcap | egrep "0x0040: 0000 0001 0a01 0164 f2ff 0000 0004 0a01" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p0.pcap | egrep "0x0050: 0101" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p0.pcap | grep -E "0x0030: 0000 0001 0806 0001 0800 0604 0002 f2ff" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p0.pcap | grep -E "0x0040: 0000 0001 0a01 0164 f2ff 0000 0004 0a01" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p0.pcap | grep -E "0x0050: 0101" 2>&1 1>/dev/null]) + + dnl Because tcpdump might not be able to parse erspan headers, we check icmp echo reply + dnl by packet length. +-AT_FAIL_IF([cat p0.pcap | egrep "IP 172.31.1.100 > 172.31.1.1: GREv0,.* length 126" 2>&1 1>/dev/null]) ++AT_FAIL_IF([cat p0.pcap | grep -E "IP 172.31.1.100 > 172.31.1.1: GREv0,.* length 126" 2>&1 1>/dev/null]) + + ovs-ofctl -O OpenFlow13 packet-out br-underlay "in_port=1 packet=f2ff00000002f2ff0000000308004500009287e14000402f57b8ac1f0101ac1f0164100088be0000000520000001144cd5a400008078f2ff00000001f2ff0000000408004500005c38d640004001eb640a0101010a01016408005e57585f0001df6c6b5b0000000045bc050000000000101112131415161718191a1b1c1d1e1f202122232425262728292a2b2c2d2e2f303132333435363738393a3b3c3d3e3f actions=normal" + +-OVS_WAIT_UNTIL([cat p0.pcap | egrep "IP 172.31.1.100 > 172.31.1.1: GREv0,.* length 126" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p0.pcap | grep -E "IP 172.31.1.100 > 172.31.1.1: GREv0,.* length 126" 2>&1 1>/dev/null]) + + OVS_TRAFFIC_VSWITCHD_STOP AT_CLEANUP AT_SETUP([datapath - ping over ip6erspan v1 tunnel by simulated packets]) @@ -10512,7 +63281,26 @@ index f22d86e466..36e10aa4a8 100644 OVS_CHECK_MIN_KERNEL(3, 10) OVS_TRAFFIC_VSWITCHD_START() -@@ -925,6 +988,7 @@ OVS_TRAFFIC_VSWITCHD_STOP +@@ -911,20 +1014,21 @@ dnl Okay, now send raw arp request and icmp echo request. + ovs-ofctl -O OpenFlow13 packet-out br-underlay "in_port=1 packet=f2ff00000002f2ff0000000386dd60008531003a2f40fc000100000000000000000000000001fc000100000000000000000000000100100088be000000051000007b00000007fffffffffffff2ff0000000408060001080006040001f2ff000000040a0101010000000000000a010164 actions=normal" + + dnl Check arp reply. +-OVS_WAIT_UNTIL([cat p0.pcap | egrep "0x0040: 0000 0001 0806 0001 0800 0604 0002 f2ff" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p0.pcap | egrep "0x0050: 0000 0001 0a01 0164 f2ff 0000 0004 0a01" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p0.pcap | egrep "0x0060: 0101" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p0.pcap | grep -E "0x0040: 0000 0001 0806 0001 0800 0604 0002 f2ff" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p0.pcap | grep -E "0x0050: 0000 0001 0a01 0164 f2ff 0000 0004 0a01" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p0.pcap | grep -E "0x0060: 0101" 2>&1 1>/dev/null]) + +-AT_FAIL_IF([cat p0.pcap | egrep "IP6 fc00:100::100 > fc00:100::1: GREv0,.* length 114" 2>&1 1>/dev/null]) ++AT_FAIL_IF([cat p0.pcap | grep -E "IP6 fc00:100::100 > fc00:100::1: GREv0,.* length 114" 2>&1 1>/dev/null]) + + ovs-ofctl -O OpenFlow13 packet-out br-underlay "in_port=1 packet=f2ff00000002f2ff0000000386dd60008531007a3c40fc000100000000000000000000000001fc0001000000000000000000000001002f00040104010100100088be000000061000407b00000007f2ff00000001f2ff0000000408004500005429b640004001fa8c0a0101010a01016408005c2c7526000118d3685b00000000e4aa020000000000101112131415161718191a1b1c1d1e1f202122232425262728292a2b2c2d2e2f3031323334353637 actions=normal" + +-OVS_WAIT_UNTIL([cat p0.pcap | egrep "IP6 fc00:100::100 > fc00:100::1: GREv0,.* length 114" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p0.pcap | grep -E "IP6 fc00:100::100 > fc00:100::1: GREv0,.* length 114" 2>&1 1>/dev/null]) + + OVS_TRAFFIC_VSWITCHD_STOP AT_CLEANUP AT_SETUP([datapath - ping over ip6erspan v2 tunnel by simulated packets]) @@ -10520,7 +63308,210 @@ index f22d86e466..36e10aa4a8 100644 OVS_CHECK_MIN_KERNEL(3, 10) OVS_TRAFFIC_VSWITCHD_START() -@@ -4100,15 +4164,15 @@ action=normal +@@ -965,15 +1069,15 @@ NS_CHECK_EXEC([at_ns0], [ping6 -q -c 3 -i 0.3 -w 2 fc00:100::100 | FORMAT_PING], + dnl Okay, now send raw arp request and icmp echo request. + ovs-ofctl -O OpenFlow13 packet-out br-underlay "in_port=1 packet=f2ff00000002f2ff0000000386dd60008531003e2f40fc000100000000000000000000000001fc000100000000000000000000000100100088be0000000620000079af514f9900008070fffffffffffff2ff0000000408060001080006040001f2ff000000040a0101010000000000000a010164 actions=normal" + +-OVS_WAIT_UNTIL([cat p0.pcap | egrep "0x0040: 0004 f2ff 0000 0001 0806 0001 0800 0604" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p0.pcap | egrep "0x0050: 0002 f2ff 0000 0001 0a01 0164 f2ff 0000" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p0.pcap | egrep "0x0060: 0004 0a01 0101" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p0.pcap | grep -E "0x0040: 0004 f2ff 0000 0001 0806 0001 0800 0604" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p0.pcap | grep -E "0x0050: 0002 f2ff 0000 0001 0a01 0164 f2ff 0000" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p0.pcap | grep -E "0x0060: 0004 0a01 0101" 2>&1 1>/dev/null]) + +-AT_FAIL_IF([cat p0.pcap | egrep "IP6 fc00:100::100 > fc00:100::1: GREv0, .* length 118" 2>&1 1>/dev/null]) ++AT_FAIL_IF([cat p0.pcap | grep -E "IP6 fc00:100::100 > fc00:100::1: GREv0, .* length 118" 2>&1 1>/dev/null]) + + ovs-ofctl -O OpenFlow13 packet-out br-underlay "in_port=1 packet=f2ff00000002f2ff0000000386dd60008531007e3c40fc000100000000000000000000000001fc0001000000000000000000000001002f00040104010100100088be0000000720004079af514f9b00008070f2ff00000001f2ff00000004080045000054ffcb4000400124770a0101010a0101640800419e23ac000112d7685b000000004caf0c0000000000101112131415161718191a1b1c1d1e1f202122232425262728292a2b2c2d2e2f3031323334353637 actions=normal" + +-OVS_WAIT_UNTIL([cat p0.pcap | egrep "IP6 fc00:100::100 > fc00:100::1: GREv0, .* length 118" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p0.pcap | grep -E "IP6 fc00:100::100 > fc00:100::1: GREv0, .* length 118" 2>&1 1>/dev/null]) + + OVS_TRAFFIC_VSWITCHD_STOP + AT_CLEANUP +@@ -1004,9 +1108,9 @@ NS_CHECK_EXEC([at_ns0], [ping -q -c 3 -i 0.3 -w 2 10.1.1.2 | FORMAT_PING], [0], + OVS_APP_EXIT_AND_WAIT([ovs-ofctl]) + + AT_CHECK([cat ofctl_monitor.log | STRIP_MONITOR_CSUM], [0], [dnl +-icmp,vlan_tci=0x0000,dl_src=ae:c6:7e:54:8d:4d,dl_dst=50:54:00:00:00:0b,nw_src=10.1.1.2,nw_dst=192.168.4.4,nw_tos=0,nw_ecn=0,nw_ttl=64,icmp_type=0,icmp_code=0 icmp_csum: +-icmp,vlan_tci=0x0000,dl_src=ae:c6:7e:54:8d:4d,dl_dst=50:54:00:00:00:0b,nw_src=10.1.1.2,nw_dst=192.168.4.4,nw_tos=0,nw_ecn=0,nw_ttl=64,icmp_type=0,icmp_code=0 icmp_csum: +-icmp,vlan_tci=0x0000,dl_src=ae:c6:7e:54:8d:4d,dl_dst=50:54:00:00:00:0b,nw_src=10.1.1.2,nw_dst=192.168.4.4,nw_tos=0,nw_ecn=0,nw_ttl=64,icmp_type=0,icmp_code=0 icmp_csum: ++icmp,vlan_tci=0x0000,dl_src=ae:c6:7e:54:8d:4d,dl_dst=50:54:00:00:00:0b,nw_src=10.1.1.2,nw_dst=192.168.4.4,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,icmp_type=0,icmp_code=0 icmp_csum: ++icmp,vlan_tci=0x0000,dl_src=ae:c6:7e:54:8d:4d,dl_dst=50:54:00:00:00:0b,nw_src=10.1.1.2,nw_dst=192.168.4.4,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,icmp_type=0,icmp_code=0 icmp_csum: ++icmp,vlan_tci=0x0000,dl_src=ae:c6:7e:54:8d:4d,dl_dst=50:54:00:00:00:0b,nw_src=10.1.1.2,nw_dst=192.168.4.4,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,icmp_type=0,icmp_code=0 icmp_csum: + ]) + + OVS_TRAFFIC_VSWITCHD_STOP +@@ -1765,10 +1869,10 @@ dnl p1(at_ns1) interface + NS_CHECK_EXEC([at_ns0], [$PYTHON3 $srcdir/sendpkt.py p0 36 b1 ee 7c 01 02 36 b1 ee 7c 01 03 08 00 45 00 00 54 03 44 40 00 40 01 21 61 0a 01 01 01 0a 01 01 02 08 00 ef ac 7c e4 00 03 5b 2c 1f 61 00 00 00 00 50 0b 02 00 00 00 00 00 10 11 12 13 14 15 16 17 18 19 1a 1b 1c 1d 1e 1f 20 21 22 23 24 25 26 27 28 29 2a 2b 2c 2d 2e 2f 30 31 32 33 34 35 36 37 > /dev/null]) + + dnl Check the expected mpls encapsulated packet on the egress interface +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0000: *0000 *0000 *0002 *0000 *0000 *0001 *8847 *0000" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0010: *2140 *36b1 *ee7c *0102 *36b1 *ee7c *0103 *0800" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0020: *4500 *0054 *0344 *4000 *4001 *2161 *0a01 *0101" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0030: *0a01 *0102 *0800 *efac *7ce4 *0003 *5b2c *1f61" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0000: *0000 *0000 *0002 *0000 *0000 *0001 *8847 *0000" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0010: *2140 *36b1 *ee7c *0102 *36b1 *ee7c *0103 *0800" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0020: *4500 *0054 *0344 *4000 *4001 *2161 *0a01 *0101" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0030: *0a01 *0102 *0800 *efac *7ce4 *0003 *5b2c *1f61" 2>&1 1>/dev/null]) + + OVS_TRAFFIC_VSWITCHD_STOP + AT_CLEANUP +@@ -1797,10 +1901,10 @@ dnl p1(at_ns1) interface + NS_CHECK_EXEC([at_ns0], [$PYTHON3 $srcdir/sendpkt.py p0 36 b1 ee 7c 01 02 36 b1 ee 7c 01 03 08 00 45 00 00 54 03 44 40 00 40 01 21 61 0a 01 01 01 0a 01 01 02 08 00 ef ac 7c e4 00 03 5b 2c 1f 61 00 00 00 00 50 0b 02 00 00 00 00 00 10 11 12 13 14 15 16 17 18 19 1a 1b 1c 1d 1e 1f 20 21 22 23 24 25 26 27 28 29 2a 2b 2c 2d 2e 2f 30 31 32 33 34 35 36 37 > /dev/null]) + + dnl Check the expected mpls encapsulated packet on the egress interface +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0000: *0000 *0000 *0002 *0000 *0000 *0001 *8847 *0000" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0010: *2140 *36b1 *ee7c *0102 *36b1 *ee7c *0103 *0800" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0020: *4500 *0054 *0344 *4000 *4001 *2161 *0a01 *0101" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0030: *0a01 *0102 *0800 *efac *7ce4 *0003 *5b2c *1f61" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0000: *0000 *0000 *0002 *0000 *0000 *0001 *8847 *0000" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0010: *2140 *36b1 *ee7c *0102 *36b1 *ee7c *0103 *0800" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0020: *4500 *0054 *0344 *4000 *4001 *2161 *0a01 *0101" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0030: *0a01 *0102 *0800 *efac *7ce4 *0003 *5b2c *1f61" 2>&1 1>/dev/null]) + + OVS_TRAFFIC_VSWITCHD_STOP + AT_CLEANUP +@@ -1830,10 +1934,10 @@ dnl p1(at_ns1) interface + NS_CHECK_EXEC([at_ns0], [$PYTHON3 $srcdir/sendpkt.py p0 36 b1 ee 7c 01 02 36 b1 ee 7c 01 03 08 00 45 00 00 54 03 44 40 00 40 01 21 61 0a 01 01 01 0a 01 01 02 08 00 ef ac 7c e4 00 03 5b 2c 1f 61 00 00 00 00 50 0b 02 00 00 00 00 00 10 11 12 13 14 15 16 17 18 19 1a 1b 1c 1d 1e 1f 20 21 22 23 24 25 26 27 28 29 2a 2b 2c 2d 2e 2f 30 31 32 33 34 35 36 37 > /dev/null]) + + dnl Check the expected mpls encapsulated packet on the egress interface +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0000: *0000 *0000 *0002 *0000 *0000 *0001 *8848 *0000" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0010: *2140 *36b1 *ee7c *0102 *36b1 *ee7c *0103 *0800" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0020: *4500 *0054 *0344 *4000 *4001 *2161 *0a01 *0101" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0030: *0a01 *0102 *0800 *efac *7ce4 *0003 *5b2c *1f61" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0000: *0000 *0000 *0002 *0000 *0000 *0001 *8848 *0000" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0010: *2140 *36b1 *ee7c *0102 *36b1 *ee7c *0103 *0800" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0020: *4500 *0054 *0344 *4000 *4001 *2161 *0a01 *0101" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0030: *0a01 *0102 *0800 *efac *7ce4 *0003 *5b2c *1f61" 2>&1 1>/dev/null]) + + OVS_TRAFFIC_VSWITCHD_STOP + AT_CLEANUP +@@ -1862,10 +1966,10 @@ dnl p1(at_ns1) interface + NS_CHECK_EXEC([at_ns0], [$PYTHON3 $srcdir/sendpkt.py p0 36 b1 ee 7c 01 02 36 b1 ee 7c 01 03 08 00 45 00 00 54 03 44 40 00 40 01 21 61 0a 01 01 01 0a 01 01 02 08 00 ef ac 7c e4 00 03 5b 2c 1f 61 00 00 00 00 50 0b 02 00 00 00 00 00 10 11 12 13 14 15 16 17 18 19 1a 1b 1c 1d 1e 1f 20 21 22 23 24 25 26 27 28 29 2a 2b 2c 2d 2e 2f 30 31 32 33 34 35 36 37 > /dev/null]) + + dnl Check the expected mpls encapsulated packet on the egress interface +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0000: *0000 *0000 *0002 *0000 *0000 *0001 *8848 *0000" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0010: *2140 *36b1 *ee7c *0102 *36b1 *ee7c *0103 *0800" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0020: *4500 *0054 *0344 *4000 *4001 *2161 *0a01 *0101" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0030: *0a01 *0102 *0800 *efac *7ce4 *0003 *5b2c *1f61" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0000: *0000 *0000 *0002 *0000 *0000 *0001 *8848 *0000" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0010: *2140 *36b1 *ee7c *0102 *36b1 *ee7c *0103 *0800" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0020: *4500 *0054 *0344 *4000 *4001 *2161 *0a01 *0101" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0030: *0a01 *0102 *0800 *efac *7ce4 *0003 *5b2c *1f61" 2>&1 1>/dev/null]) + + OVS_TRAFFIC_VSWITCHD_STOP + AT_CLEANUP +@@ -1896,13 +2000,13 @@ dnl p1(at_ns1) interface + NS_CHECK_EXEC([at_ns0], [$PYTHON3 $srcdir/sendpkt.py p0 00 00 00 00 00 02 00 00 00 00 00 01 88 47 00 00 21 40 36 b1 ee 7c 01 02 36 b1 ee 7c 01 03 08 00 45 00 00 54 03 44 40 00 40 01 21 61 0a 01 01 01 0a 01 01 02 08 00 ef ac 7c e4 00 03 5b 2c 1f 61 00 00 00 00 50 0b 02 00 00 00 00 00 10 11 12 13 14 15 16 17 18 19 1a 1b 1c 1d 1e 1f 20 21 22 23 24 25 26 27 28 29 2a 2b 2c 2d 2e 2f 30 31 32 33 34 35 36 37 > /dev/null]) + + dnl Check the expected decapsulated on the egress interface +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0000: *36b1 *ee7c *0102 *36b1 *ee7c *0103 *0800 *4500" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0010: *0054 *0344 *4000 *4001 *2161 *0a01 *0101 *0a01" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0020: *0102 *0800 *efac *7ce4 *0003 *5b2c *1f61 *0000" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0030: *0000 *500b *0200 *0000 *0000 *1011 *1213 *1415" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0040: *1617 *1819 *1a1b *1c1d *1e1f *2021 *2223 *2425" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0050: *2627 *2829 *2a2b *2c2d *2e2f *3031 *3233 *3435" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0060: *3637" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0000: *36b1 *ee7c *0102 *36b1 *ee7c *0103 *0800 *4500" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0010: *0054 *0344 *4000 *4001 *2161 *0a01 *0101 *0a01" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0020: *0102 *0800 *efac *7ce4 *0003 *5b2c *1f61 *0000" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0030: *0000 *500b *0200 *0000 *0000 *1011 *1213 *1415" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0040: *1617 *1819 *1a1b *1c1d *1e1f *2021 *2223 *2425" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0050: *2627 *2829 *2a2b *2c2d *2e2f *3031 *3233 *3435" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0060: *3637" 2>&1 1>/dev/null]) + + + OVS_TRAFFIC_VSWITCHD_STOP +@@ -1933,13 +2037,13 @@ dnl p1(at_ns1) interface + NS_CHECK_EXEC([at_ns0], [$PYTHON3 $srcdir/sendpkt.py p0 00 00 00 00 00 02 00 00 00 00 00 01 88 47 00 00 21 40 36 b1 ee 7c 01 02 36 b1 ee 7c 01 03 08 00 45 00 00 54 03 44 40 00 40 01 21 61 0a 01 01 01 0a 01 01 02 08 00 ef ac 7c e4 00 03 5b 2c 1f 61 00 00 00 00 50 0b 02 00 00 00 00 00 10 11 12 13 14 15 16 17 18 19 1a 1b 1c 1d 1e 1f 20 21 22 23 24 25 26 27 28 29 2a 2b 2c 2d 2e 2f 30 31 32 33 34 35 36 37 > /dev/null]) + + dnl Check the expected decapsulated on the egress interface +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0000: *36b1 *ee7c *0102 *36b1 *ee7c *0103 *0800 *4500" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0010: *0054 *0344 *4000 *4001 *2161 *0a01 *0101 *0a01" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0020: *0102 *0800 *efac *7ce4 *0003 *5b2c *1f61 *0000" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0030: *0000 *500b *0200 *0000 *0000 *1011 *1213 *1415" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0040: *1617 *1819 *1a1b *1c1d *1e1f *2021 *2223 *2425" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0050: *2627 *2829 *2a2b *2c2d *2e2f *3031 *3233 *3435" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0060: *3637" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0000: *36b1 *ee7c *0102 *36b1 *ee7c *0103 *0800 *4500" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0010: *0054 *0344 *4000 *4001 *2161 *0a01 *0101 *0a01" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0020: *0102 *0800 *efac *7ce4 *0003 *5b2c *1f61 *0000" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0030: *0000 *500b *0200 *0000 *0000 *1011 *1213 *1415" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0040: *1617 *1819 *1a1b *1c1d *1e1f *2021 *2223 *2425" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0050: *2627 *2829 *2a2b *2c2d *2e2f *3031 *3233 *3435" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0060: *3637" 2>&1 1>/dev/null]) + + + OVS_TRAFFIC_VSWITCHD_STOP +@@ -1985,9 +2089,9 @@ OVS_APP_EXIT_AND_WAIT([ovs-ofctl]) + dnl Check this output. We only see the latter two packets, not the first. + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN2 (xid=0x0): total_len=42 in_port=1 (via action) data_len=42 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=1,tp_dst=2 udp_csum:0 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=1,tp_dst=2 udp_csum:0 + NXT_PACKET_IN2 (xid=0x0): cookie=0x0 total_len=42 ct_state=est|rpl|trk,ct_nw_src=10.1.1.1,ct_nw_dst=10.1.1.2,ct_nw_proto=17,ct_tp_src=1,ct_tp_dst=2,ip,in_port=2 (via action) data_len=42 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=2,tp_dst=1 udp_csum:0 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=2,tp_dst=1 udp_csum:0 + ]) + + OVS_TRAFFIC_VSWITCHD_STOP +@@ -2033,9 +2137,9 @@ OVS_APP_EXIT_AND_WAIT([ovs-ofctl]) + dnl Check this output. We only see the latter two packets, not the first. + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN2 (xid=0x0): cookie=0x0 total_len=42 in_port=1 (via action) data_len=42 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=1,tp_dst=2 udp_csum:0 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.1,nw_dst=10.1.1.2,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=1,tp_dst=2 udp_csum:0 + NXT_PACKET_IN2 (xid=0x0): table_id=1 cookie=0x0 total_len=42 ct_state=new|trk,ct_nw_src=10.1.1.2,ct_nw_dst=10.1.1.1,ct_nw_proto=17,ct_tp_src=2,ct_tp_dst=1,ip,in_port=2 (via action) data_len=42 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=0,tp_src=2,tp_dst=1 udp_csum:0 ++udp,vlan_tci=0x0000,dl_src=50:54:00:00:00:09,dl_dst=50:54:00:00:00:0a,nw_src=10.1.1.2,nw_dst=10.1.1.1,nw_tos=0,nw_ecn=0,nw_ttl=0,nw_frag=no,tp_src=2,tp_dst=1 udp_csum:0 + ]) + + dnl +@@ -2980,6 +3084,15 @@ NXST_FLOW reply: + table=1, priority=100,ct_state=+est+trk,in_port=1 actions=output:2 + ]) + ++dnl Send a 3rd UDP packet on port 1 ++AT_CHECK([ovs-ofctl -O OpenFlow13 packet-out br0 "in_port=1 packet=50540000000a50540000000908004500001c000000000011a4cd0a0101010a0101020001000200080000 actions=resubmit(,0)"]) ++ ++dnl There still should not be any packet that matches the established ct_state. ++AT_CHECK([ovs-ofctl dump-flows br0 "table=1 in_port=1,ct_state=+trk+est" | ofctl_strip], [0], [dnl ++NXST_FLOW reply: ++ table=1, priority=100,ct_state=+est+trk,in_port=1 actions=output:2 ++]) ++ + OVS_TRAFFIC_VSWITCHD_STOP + AT_CLEANUP + +@@ -3140,11 +3253,11 @@ OVS_APP_EXIT_AND_WAIT([ovs-ofctl]) + dnl Check this output. We only see the latter two packets, not the first. + AT_CHECK([cat ofctl_monitor.log | grep -v ff02 | grep -v fe80 | grep -v no_match], [0], [dnl + NXT_PACKET_IN2 (xid=0x0): table_id=1 cookie=0x0 total_len=75 ct_state=inv|trk,ip,in_port=2 (via action) data_len=75 (unbuffered) +-icmp,vlan_tci=0x0000,dl_src=c6:f5:4e:cb:72:db,dl_dst=f6:4c:47:35:28:c9,nw_src=172.16.0.4,nw_dst=172.16.0.3,nw_tos=192,nw_ecn=0,nw_ttl=64,icmp_type=3,icmp_code=3 icmp_csum:da49 ++icmp,vlan_tci=0x0000,dl_src=c6:f5:4e:cb:72:db,dl_dst=f6:4c:47:35:28:c9,nw_src=172.16.0.4,nw_dst=172.16.0.3,nw_tos=192,nw_ecn=0,nw_ttl=64,nw_frag=no,icmp_type=3,icmp_code=3 icmp_csum:da49 + NXT_PACKET_IN2 (xid=0x0): table_id=1 cookie=0x0 total_len=47 ct_state=new|trk,ct_nw_src=172.16.0.1,ct_nw_dst=172.16.0.2,ct_nw_proto=17,ct_tp_src=41614,ct_tp_dst=5555,ip,in_port=1 (via action) data_len=47 (unbuffered) +-udp,vlan_tci=0x0000,dl_src=e6:4c:47:35:28:c9,dl_dst=c6:f9:4e:cb:72:db,nw_src=172.16.0.1,nw_dst=172.16.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,tp_src=41614,tp_dst=5555 udp_csum:2096 ++udp,vlan_tci=0x0000,dl_src=e6:4c:47:35:28:c9,dl_dst=c6:f9:4e:cb:72:db,nw_src=172.16.0.1,nw_dst=172.16.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,tp_src=41614,tp_dst=5555 udp_csum:2096 + NXT_PACKET_IN2 (xid=0x0): table_id=1 cookie=0x0 total_len=75 ct_state=rel|rpl|trk,ct_nw_src=172.16.0.1,ct_nw_dst=172.16.0.2,ct_nw_proto=17,ct_tp_src=41614,ct_tp_dst=5555,ip,in_port=2 (via action) data_len=75 (unbuffered) +-icmp,vlan_tci=0x0000,dl_src=c6:f9:4e:cb:72:db,dl_dst=e6:4c:47:35:28:c9,nw_src=172.16.0.3,nw_dst=172.16.0.1,nw_tos=192,nw_ecn=0,nw_ttl=64,icmp_type=3,icmp_code=3 icmp_csum:553f ++icmp,vlan_tci=0x0000,dl_src=c6:f9:4e:cb:72:db,dl_dst=e6:4c:47:35:28:c9,nw_src=172.16.0.3,nw_dst=172.16.0.1,nw_tos=192,nw_ecn=0,nw_ttl=64,nw_frag=no,icmp_type=3,icmp_code=3 icmp_csum:553f + ]) + + AT_CHECK([ovs-appctl dpctl/dump-conntrack | FORMAT_CT(172.16.0.1)], [0], [dnl +@@ -3345,6 +3458,11 @@ AT_CHECK([ovs-ofctl bundle br0 bundle.txt]) + AT_CHECK([ovs-appctl dpctl/dump-conntrack | FORMAT_CT(10.1.1.2)], [0], [dnl + ]) + ++dnl Send the second fragment in order to avoid keeping the first fragment ++dnl in the queue until the expiration occurs. Fragments already queued, if resent, ++dnl may lead to failures on the kernel datapath. ++AT_CHECK([ovs-ofctl -O OpenFlow13 packet-out br0 "in_port=1, packet=50540000000a505400000009080045000030000100320011a4860a0101010a01010200010002000800000010203040506070809000010203040506070809, actions=ct(commit)"]) ++ + OVS_TRAFFIC_VSWITCHD_STOP + AT_CLEANUP + +@@ -4100,15 +4218,15 @@ action=normal AT_CHECK([ovs-ofctl --bundle add-flows br0 flows.txt]) @@ -10539,7 +63530,25 @@ index f22d86e466..36e10aa4a8 100644 "1616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161616161610a, actions=ct(table=1)"]) AT_CHECK([ovs-appctl dpctl/dump-flows | head -2 | tail -1 | grep -q -e ["]udp[(]src=5001["]]) -@@ -6454,7 +6518,7 @@ on_exit 'ovs-appctl revalidator/purge' +@@ -5384,7 +5502,7 @@ AT_CHECK([ovs-appctl dpctl/dump-conntrack | FORMAT_CT(10.1.1.2) | sed -e 's/dst= + udp,orig=(src=10.1.1.1,dst=10.1.1.2,sport=,dport=),reply=(src=10.1.1.2,dst=10.1.1.2XX,sport=,dport=),mark=1 + ]) + +-AT_CHECK([tcpdump -v "icmp" -r p0.pcap 2>/dev/null | egrep 'wrong|bad'], [1], [ignore-nolog]) ++AT_CHECK([tcpdump -v "icmp" -r p0.pcap 2>/dev/null | grep -E 'wrong|bad'], [1], [ignore-nolog]) + + OVS_TRAFFIC_VSWITCHD_STOP + AT_CLEANUP +@@ -6134,7 +6252,7 @@ sleep 1 + dnl UDP packets from ns0->ns1 should solicit "destination unreachable" response. + NS_CHECK_EXEC([at_ns0], [bash -c "echo a | nc -6 $NC_EOF_OPT -u fc00::2 1"]) + +-AT_CHECK([tcpdump -v "icmp6" -r p0.pcap 2>/dev/null | egrep 'wrong|bad'], [1], [ignore-nolog]) ++AT_CHECK([tcpdump -v "icmp6" -r p0.pcap 2>/dev/null | grep -E 'wrong|bad'], [1], [ignore-nolog]) + + AT_CHECK([ovs-appctl dpctl/dump-conntrack | FORMAT_CT(fc00::2)], [0], [dnl + udp,orig=(src=fc00::1,dst=fc00::2,sport=,dport=),reply=(src=fc00::2,dst=fc00::240,sport=,dport=) +@@ -6454,7 +6572,7 @@ on_exit 'ovs-appctl revalidator/purge' on_exit 'ovs-appctl dpif/dump-flows br0' dnl Should work with the virtual IP address through NAT @@ -10548,7 +63557,7 @@ index f22d86e466..36e10aa4a8 100644 echo Request $i NS_CHECK_EXEC([at_ns1], [wget 10.1.1.64 -t 5 -T 1 --retry-connrefused -v -o wget$i.log]) done -@@ -6743,6 +6807,132 @@ AT_CHECK([ovs-ofctl dump-flows br0 | grep table=2, | OFPROTO_CLEAR_DURATION_IDLE +@@ -6743,6 +6861,132 @@ AT_CHECK([ovs-ofctl dump-flows br0 | grep table=2, | OFPROTO_CLEAR_DURATION_IDLE OVS_TRAFFIC_VSWITCHD_STOP AT_CLEANUP @@ -10681,6 +63690,95 @@ index f22d86e466..36e10aa4a8 100644 AT_BANNER([802.1ad]) AT_SETUP([802.1ad - vlan_limit]) +@@ -7007,12 +7251,12 @@ dnl p1(at_ns1) interface + NS_CHECK_EXEC([at_ns0], [$PYTHON3 $srcdir/sendpkt.py p0 f2 00 00 00 00 02 f2 00 00 00 00 01 08 00 45 00 00 28 00 01 00 00 40 06 b0 13 c0 a8 00 0a 0a 00 00 0a 04 00 08 00 00 00 00 c8 00 00 00 00 50 02 20 00 b8 5e 00 00 > /dev/null]) + + dnl Check the expected nsh encapsulated packet on the egress interface +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0000: *f2ff *0000 *0002 *f2ff *0000 *0001 *894f *0fc6" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0010: *0103 *0012 *34ff *1122 *3344 *0000 *0000 *0000" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0020: *0000 *0000 *0000 *f200 *0000 *0002 *f200 *0000" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0030: *0001 *0800 *4500 *0028 *0001 *0000 *4006 *b013" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0040: *c0a8 *000a *0a00 *000a *0400 *0800 *0000 *00c8" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0050: *0000 *0000 *5002 *2000 *b85e *0000" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0000: *f2ff *0000 *0002 *f2ff *0000 *0001 *894f *0fc6" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0010: *0103 *0012 *34ff *1122 *3344 *0000 *0000 *0000" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0020: *0000 *0000 *0000 *f200 *0000 *0002 *f200 *0000" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0030: *0001 *0800 *4500 *0028 *0001 *0000 *4006 *b013" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0040: *c0a8 *000a *0a00 *000a *0400 *0800 *0000 *00c8" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0050: *0000 *0000 *5002 *2000 *b85e *0000" 2>&1 1>/dev/null]) + + OVS_TRAFFIC_VSWITCHD_STOP + AT_CLEANUP +@@ -7039,10 +7283,10 @@ dnl p1(at_ns1) interface + NS_CHECK_EXEC([at_ns0], [$PYTHON3 $srcdir/sendpkt.py p0 f2 ff 00 00 00 02 f2 ff 00 00 00 01 89 4f 02 06 01 03 00 00 64 03 01 02 03 04 05 06 07 08 09 0a 0b 0c 0d 0e 0f 10 f2 00 00 00 00 02 f2 00 00 00 00 01 08 00 45 00 00 28 00 01 00 00 40 06 b0 13 c0 a8 00 0a 0a 00 00 0a 04 00 08 00 00 00 00 c8 00 00 00 00 50 02 20 00 b8 5e 00 00 > /dev/null]) + + dnl Check the expected de-capsulated TCP packet on the egress interface +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0000: *f200 *0000 *0002 *f200 *0000 *0001 *0800 *4500" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0010: *0028 *0001 *0000 *4006 *b013 *c0a8 *000a *0a00" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0020: *000a *0400 *0800 *0000 *00c8 *0000 *0000 *5002" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0030: *2000 *b85e *0000" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0000: *f200 *0000 *0002 *f200 *0000 *0001 *0800 *4500" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0010: *0028 *0001 *0000 *4006 *b013 *c0a8 *000a *0a00" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0020: *000a *0400 *0800 *0000 *00c8 *0000 *0000 *5002" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0030: *2000 *b85e *0000" 2>&1 1>/dev/null]) + + OVS_TRAFFIC_VSWITCHD_STOP + AT_CLEANUP +@@ -7072,12 +7316,12 @@ dnl p1(at_ns1) interface + NS_CHECK_EXEC([at_ns0], [$PYTHON3 $srcdir/sendpkt.py p0 f2 ff 00 00 00 02 f2 ff 00 00 00 01 89 4f 02 06 01 03 00 01 00 03 01 02 03 04 05 06 07 08 09 0a 0b 0c 0d 0e 0f 10 f2 00 00 00 00 02 f2 00 00 00 00 01 08 00 45 00 00 28 00 01 00 00 40 06 b0 13 c0 a8 00 0a 0a 00 00 0a 04 00 08 00 00 00 00 c8 00 00 00 00 50 02 20 00 b8 5e 00 00 > /dev/null]) + + dnl Check the expected NSH packet with new fields in the header +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0000: *f2ff *0000 *0002 *f2ff *0000* 0001 *894f *01c6" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0010: *0103 *0001 *0104 *100f *0e0d *0c0b *0a09 *0807" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0020: *0605 *0403 *0201 *f200 *0000 *0002 *f200 *0000" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0030: *0001 *0800 *4500 *0028 *0001 *0000 *4006 *b013" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0040: *c0a8 *000a *0a00 *000a *0400 *0800 *0000 *00c8" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0050: *0000 *0000 *5002 *2000 *b85e *0000" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0000: *f2ff *0000 *0002 *f2ff *0000* 0001 *894f *01c6" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0010: *0103 *0001 *0104 *100f *0e0d *0c0b *0a09 *0807" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0020: *0605 *0403 *0201 *f200 *0000 *0002 *f200 *0000" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0030: *0001 *0800 *4500 *0028 *0001 *0000 *4006 *b013" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0040: *c0a8 *000a *0a00 *000a *0400 *0800 *0000 *00c8" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0050: *0000 *0000 *5002 *2000 *b85e *0000" 2>&1 1>/dev/null]) + + OVS_TRAFFIC_VSWITCHD_STOP + AT_CLEANUP +@@ -7106,23 +7350,23 @@ dnl First send packet from at_ns0 --> OVS with SPI=0x100 and SI=2 + NS_CHECK_EXEC([at_ns0], [$PYTHON3 $srcdir/sendpkt.py p0 f2 ff 00 00 00 02 f2 ff 00 00 00 01 89 4f 02 06 01 03 00 01 00 02 01 02 03 04 05 06 07 08 09 0a 0b 0c 0d 0e 0f 10 f2 00 00 00 00 02 f2 00 00 00 00 01 08 00 45 00 00 28 00 01 00 00 40 06 b0 13 c0 a8 00 0a 0a 00 00 0a 04 00 08 00 00 00 00 c8 00 00 00 00 50 02 20 00 b8 5e 00 00 > /dev/null]) + + dnl Check for the above packet on p1 interface +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0000: *f2ff *0000 *0002 *f2ff *0000 *0001 *894f *0206" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0010: *0103 *0001 *0002 *0102 *0304 *0506 *0708 *090a" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0020: *0b0c *0d0e *0f10 *f200 *0000 *0002 *f200 *0000" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0030: *0001 *0800 *4500 *0028 *0001 *0000 *4006 *b013" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0040: *c0a8 *000a *0a00 *000a *0400 *0800 *0000 *00c8" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p1.pcap | egrep "0x0050: *0000 *0000 *5002 *2000 *b85e *0000" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0000: *f2ff *0000 *0002 *f2ff *0000 *0001 *894f *0206" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0010: *0103 *0001 *0002 *0102 *0304 *0506 *0708 *090a" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0020: *0b0c *0d0e *0f10 *f200 *0000 *0002 *f200 *0000" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0030: *0001 *0800 *4500 *0028 *0001 *0000 *4006 *b013" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0040: *c0a8 *000a *0a00 *000a *0400 *0800 *0000 *00c8" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p1.pcap | grep -E "0x0050: *0000 *0000 *5002 *2000 *b85e *0000" 2>&1 1>/dev/null]) + + dnl Send the second packet from at_ns1 --> OVS with SPI=0x100 and SI=1 + NS_CHECK_EXEC([at_ns1], [$PYTHON3 $srcdir/sendpkt.py p1 f2 ff 00 00 00 02 f2 ff 00 00 00 01 89 4f 01 c6 01 03 00 01 00 01 01 02 03 04 05 06 07 08 09 0a 0b 0c 0d 0e 0f 10 f2 00 00 00 00 02 f2 00 00 00 00 01 08 00 45 00 00 28 00 01 00 00 40 06 b0 13 c0 a8 00 0a 0a 00 00 0a 04 00 08 00 00 00 00 c8 00 00 00 00 50 02 20 00 b8 5e 00 00 > /dev/null]) + + dnl Check for the above packet on p2 interface +-OVS_WAIT_UNTIL([cat p2.pcap | egrep "0x0000: *f2ff *0000 *0002 *f2ff *0000 *0001 *894f *01c6" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p2.pcap | egrep "0x0010: *0103 *0001 *0001 *0102 *0304 *0506 *0708 *090a" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p2.pcap | egrep "0x0020: *0b0c *0d0e *0f10 *f200 *0000 *0002 *f200 *0000" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p2.pcap | egrep "0x0030: *0001 *0800 *4500 *0028 *0001 *0000 *4006 *b013" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p2.pcap | egrep "0x0040: *c0a8 *000a *0a00 *000a *0400 *0800 *0000 *00c8" 2>&1 1>/dev/null]) +-OVS_WAIT_UNTIL([cat p2.pcap | egrep "0x0050: *0000 *0000 *5002 *2000 *b85e *0000" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p2.pcap | grep -E "0x0000: *f2ff *0000 *0002 *f2ff *0000 *0001 *894f *01c6" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p2.pcap | grep -E "0x0010: *0103 *0001 *0001 *0102 *0304 *0506 *0708 *090a" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p2.pcap | grep -E "0x0020: *0b0c *0d0e *0f10 *f200 *0000 *0002 *f200 *0000" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p2.pcap | grep -E "0x0030: *0001 *0800 *4500 *0028 *0001 *0000 *4006 *b013" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p2.pcap | grep -E "0x0040: *c0a8 *000a *0a00 *000a *0400 *0800 *0000 *00c8" 2>&1 1>/dev/null]) ++OVS_WAIT_UNTIL([cat p2.pcap | grep -E "0x0050: *0000 *0000 *5002 *2000 *b85e *0000" 2>&1 1>/dev/null]) + + OVS_TRAFFIC_VSWITCHD_STOP + AT_CLEANUP diff --git a/tests/system-tso-macros.at b/tests/system-tso-macros.at index 406334f3e0..1a80047619 100644 --- a/tests/system-tso-macros.at @@ -10691,6 +63789,22 @@ index 406334f3e0..1a80047619 100644 ) + +m4_define([CHECK_SYSTEM_TSO], []) +diff --git a/tests/system-userspace-macros.at b/tests/system-userspace-macros.at +index f639ba53a2..da3a4caca9 100644 +--- a/tests/system-userspace-macros.at ++++ b/tests/system-userspace-macros.at +@@ -299,6 +299,11 @@ m4_define([OVS_CHECK_KERNEL_EXCL], + AT_SKIP_IF([:]) + ]) + ++# CHECK_LATER_IPV6_FRAGMENTS() ++# ++# Userspace is parsing later IPv6 fragments correctly. ++m4_define([CHECK_LATER_IPV6_FRAGMENTS], []) ++ + # VSCTL_ADD_DATAPATH_TABLE() + # + # Create datapath table "netdev" for userspace tests in ovsdb diff --git a/tests/test-cmap.c b/tests/test-cmap.c index 0705475606..588a5dea63 100644 --- a/tests/test-cmap.c @@ -10962,6 +64076,264 @@ index 6f1fb059bc..2c6c444488 100644 } } } +diff --git a/tests/test-ovsdb.c b/tests/test-ovsdb.c +index ca4e87b811..3194f50f2d 100644 +--- a/tests/test-ovsdb.c ++++ b/tests/test-ovsdb.c +@@ -294,11 +294,24 @@ print_and_free_ovsdb_error(struct ovsdb_error *error) + free(string); + } + ++static struct json **json_to_destroy; ++ ++static void ++destroy_on_ovsdb_error(struct json **json) ++{ ++ json_to_destroy = json; ++} ++ + static void + check_ovsdb_error(struct ovsdb_error *error) + { + if (error) { + char *s = ovsdb_error_to_string_free(error); ++ ++ if (json_to_destroy) { ++ json_destroy(*json_to_destroy); ++ json_to_destroy = NULL; ++ } + ovs_fatal(0, "%s", s); + } + } +@@ -481,6 +494,8 @@ do_diff_data(struct ovs_cmdl_context *ctx) + struct json *json; + struct ovsdb_datum new, old, diff, reincarnation; + ++ destroy_on_ovsdb_error(&json); ++ + json = unbox_json(parse_json(ctx->argv[1])); + check_ovsdb_error(ovsdb_type_from_json(&type, json)); + json_destroy(json); +@@ -556,6 +571,8 @@ do_parse_atomic_type(struct ovs_cmdl_context *ctx) + enum ovsdb_atomic_type type; + struct json *json; + ++ destroy_on_ovsdb_error(&json); ++ + json = unbox_json(parse_json(ctx->argv[1])); + check_ovsdb_error(ovsdb_atomic_type_from_json(&type, json)); + json_destroy(json); +@@ -568,6 +585,8 @@ do_parse_base_type(struct ovs_cmdl_context *ctx) + struct ovsdb_base_type base; + struct json *json; + ++ destroy_on_ovsdb_error(&json); ++ + json = unbox_json(parse_json(ctx->argv[1])); + check_ovsdb_error(ovsdb_base_type_from_json(&base, json)); + json_destroy(json); +@@ -581,6 +600,8 @@ do_parse_type(struct ovs_cmdl_context *ctx) + struct ovsdb_type type; + struct json *json; + ++ destroy_on_ovsdb_error(&json); ++ + json = unbox_json(parse_json(ctx->argv[1])); + check_ovsdb_error(ovsdb_type_from_json(&type, json)); + json_destroy(json); +@@ -595,6 +616,8 @@ do_parse_atoms(struct ovs_cmdl_context *ctx) + struct json *json; + int i; + ++ destroy_on_ovsdb_error(&json); ++ + json = unbox_json(parse_json(ctx->argv[1])); + check_ovsdb_error(ovsdb_base_type_from_json(&base, json)); + json_destroy(json); +@@ -624,6 +647,8 @@ do_parse_atom_strings(struct ovs_cmdl_context *ctx) + struct json *json; + int i; + ++ destroy_on_ovsdb_error(&json); ++ + json = unbox_json(parse_json(ctx->argv[1])); + check_ovsdb_error(ovsdb_base_type_from_json(&base, json)); + json_destroy(json); +@@ -669,6 +694,8 @@ do_parse_data__(int argc, char *argv[], + struct json *json; + int i; + ++ destroy_on_ovsdb_error(&json); ++ + json = unbox_json(parse_json(argv[1])); + check_ovsdb_error(ovsdb_type_from_json(&type, json)); + json_destroy(json); +@@ -700,6 +727,8 @@ do_parse_data_strings(struct ovs_cmdl_context *ctx) + struct json *json; + int i; + ++ destroy_on_ovsdb_error(&json); ++ + json = unbox_json(parse_json(ctx->argv[1])); + check_ovsdb_error(ovsdb_type_from_json(&type, json)); + json_destroy(json); +@@ -740,6 +769,8 @@ do_sort_atoms(struct ovs_cmdl_context *ctx) + size_t n_atoms; + int i; + ++ destroy_on_ovsdb_error(&json); ++ + json = unbox_json(parse_json(ctx->argv[1])); + check_ovsdb_error(ovsdb_base_type_from_json(&base, json)); + json_destroy(json); +@@ -779,6 +810,8 @@ do_parse_column(struct ovs_cmdl_context *ctx) + struct ovsdb_column *column; + struct json *json; + ++ destroy_on_ovsdb_error(&json); ++ + json = parse_json(ctx->argv[2]); + check_ovsdb_error(ovsdb_column_from_json(json, ctx->argv[1], &column)); + json_destroy(json); +@@ -795,6 +828,8 @@ do_parse_table(struct ovs_cmdl_context *ctx) + + default_is_root = ctx->argc > 3 && !strcmp(ctx->argv[3], "true"); + ++ destroy_on_ovsdb_error(&json); ++ + json = parse_json(ctx->argv[2]); + check_ovsdb_error(ovsdb_table_schema_from_json(json, ctx->argv[1], &ts)); + json_destroy(json); +@@ -811,6 +846,8 @@ do_parse_rows(struct ovs_cmdl_context *ctx) + struct json *json; + int i; + ++ destroy_on_ovsdb_error(&json); ++ + json = unbox_json(parse_json(ctx->argv[1])); + check_ovsdb_error(ovsdb_table_schema_from_json(json, "mytable", &ts)); + json_destroy(json); +@@ -870,6 +907,8 @@ do_compare_rows(struct ovs_cmdl_context *ctx) + int n_rows; + int i, j; + ++ destroy_on_ovsdb_error(&json); ++ + json = unbox_json(parse_json(ctx->argv[1])); + check_ovsdb_error(ovsdb_table_schema_from_json(json, "mytable", &ts)); + json_destroy(json); +@@ -929,6 +968,8 @@ do_parse_conditions(struct ovs_cmdl_context *ctx) + int exit_code = 0; + int i; + ++ destroy_on_ovsdb_error(&json); ++ + json = unbox_json(parse_json(ctx->argv[1])); + check_ovsdb_error(ovsdb_table_schema_from_json(json, "mytable", &ts)); + json_destroy(json); +@@ -971,6 +1012,8 @@ do_evaluate_condition__(struct ovs_cmdl_context *ctx, int mode) + struct json *json; + size_t i, j; + ++ destroy_on_ovsdb_error(&json); ++ + /* Parse table schema, create table. */ + json = unbox_json(parse_json(ctx->argv[1])); + check_ovsdb_error(ovsdb_table_schema_from_json(json, "mytable", &ts)); +@@ -1058,6 +1101,8 @@ do_compare_conditions(struct ovs_cmdl_context *ctx) + struct json *json; + size_t i; + ++ destroy_on_ovsdb_error(&json); ++ + /* Parse table schema, create table. */ + json = unbox_json(parse_json(ctx->argv[1])); + check_ovsdb_error(ovsdb_table_schema_from_json(json, "mytable", &ts)); +@@ -1099,6 +1144,8 @@ do_parse_mutations(struct ovs_cmdl_context *ctx) + int exit_code = 0; + int i; + ++ destroy_on_ovsdb_error(&json); ++ + json = unbox_json(parse_json(ctx->argv[1])); + check_ovsdb_error(ovsdb_table_schema_from_json(json, "mytable", &ts)); + json_destroy(json); +@@ -1138,6 +1185,8 @@ do_execute_mutations(struct ovs_cmdl_context *ctx) + struct json *json; + size_t i, j; + ++ destroy_on_ovsdb_error(&json); ++ + /* Parse table schema, create table. */ + json = unbox_json(parse_json(ctx->argv[1])); + check_ovsdb_error(ovsdb_table_schema_from_json(json, "mytable", &ts)); +@@ -1262,6 +1311,8 @@ do_query(struct ovs_cmdl_context *ctx) + int exit_code = 0; + size_t i; + ++ destroy_on_ovsdb_error(&json); ++ + /* Parse table schema, create table. */ + json = unbox_json(parse_json(ctx->argv[1])); + check_ovsdb_error(ovsdb_table_schema_from_json(json, "mytable", &ts)); +@@ -1356,6 +1407,8 @@ do_query_distinct(struct ovs_cmdl_context *ctx) + int exit_code = 0; + size_t i; + ++ destroy_on_ovsdb_error(&json); ++ + /* Parse table schema, create table. */ + json = unbox_json(parse_json(ctx->argv[1])); + check_ovsdb_error(ovsdb_table_schema_from_json(json, "mytable", &ts)); +@@ -1483,6 +1536,8 @@ do_parse_schema(struct ovs_cmdl_context *ctx) + struct ovsdb_schema *schema; + struct json *json; + ++ destroy_on_ovsdb_error(&json); ++ + json = parse_json(ctx->argv[1]); + check_ovsdb_error(ovsdb_schema_from_json(json, &schema)); + json_destroy(json); +@@ -1498,6 +1553,8 @@ do_execute__(struct ovs_cmdl_context *ctx, bool ro) + struct ovsdb *db; + int i; + ++ destroy_on_ovsdb_error(&json); ++ + /* Create database. */ + json = parse_json(ctx->argv[1]); + check_ovsdb_error(ovsdb_schema_from_json(json, &schema)); +@@ -1564,6 +1621,8 @@ do_trigger(struct ovs_cmdl_context *ctx) + int number; + int i; + ++ destroy_on_ovsdb_error(&json); ++ + /* Create database. */ + json = parse_json(ctx->argv[1]); + check_ovsdb_error(ovsdb_schema_from_json(json, &schema)); +@@ -1789,6 +1848,8 @@ do_transact(struct ovs_cmdl_context *ctx) + struct json *json; + int i; + ++ destroy_on_ovsdb_error(&json); ++ + /* Create table. */ + json = parse_json("{\"name\": \"testdb\", " + " \"tables\": " +diff --git a/tests/test-ovsdb.py b/tests/test-ovsdb.py +index 853264f22b..402cacbe9d 100644 +--- a/tests/test-ovsdb.py ++++ b/tests/test-ovsdb.py +@@ -620,7 +620,7 @@ def update_condition(idl, commands): + commands = commands[len("condition "):].split(";") + for command in commands: + command = command.split(" ") +- if(len(command) != 2): ++ if len(command) != 2: + sys.stderr.write("Error parsing condition %s\n" % command) + sys.exit(1) + diff --git a/tests/test-rcu.c b/tests/test-rcu.c index 965f3c49f3..bb17092bf0 100644 --- a/tests/test-rcu.c @@ -11042,8 +64414,21 @@ index f0fd042108..7d899fbbfd 100644 /* Check maximum x such that f(x) == n. */ check_ctz32(UINT32_MAX << n, n); +diff --git a/tests/tunnel-push-pop-ipv6.at b/tests/tunnel-push-pop-ipv6.at +index 3f58e3e8fd..8c5af459e9 100644 +--- a/tests/tunnel-push-pop-ipv6.at ++++ b/tests/tunnel-push-pop-ipv6.at +@@ -452,7 +452,7 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN2 (xid=0x0): cookie=0x0 total_len=98 tun_id=0x7b,tun_ipv6_src=2001:cafe::92,tun_ipv6_dst=2001:cafe::88,tun_metadata0=0xa,in_port=5 (via action) data_len=98 (unbuffered) +-icmp,vlan_tci=0x0000,dl_src=be:b6:f4:e1:49:4a,dl_dst=fe:71:d8:83:72:4f,nw_src=30.0.0.1,nw_dst=30.0.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,icmp_type=0,icmp_code=0 icmp_csum:4227 ++icmp,vlan_tci=0x0000,dl_src=be:b6:f4:e1:49:4a,dl_dst=fe:71:d8:83:72:4f,nw_src=30.0.0.1,nw_dst=30.0.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,icmp_type=0,icmp_code=0 icmp_csum:4227 + ]) + + AT_CHECK([ovs-ofctl dump-ports int-br | grep 'port 5'], [0], [dnl diff --git a/tests/tunnel-push-pop.at b/tests/tunnel-push-pop.at -index 57589758f4..c63344196b 100644 +index 57589758f4..c9a04c76bd 100644 --- a/tests/tunnel-push-pop.at +++ b/tests/tunnel-push-pop.at @@ -546,6 +546,28 @@ AT_CHECK([ovs-ofctl dump-ports int-br | grep 'port [[37]]' | sort], [0], [dnl @@ -11075,7 +64460,14 @@ index 57589758f4..c63344196b 100644 dnl Check decapsulation of Geneve packet with options AT_CAPTURE_FILE([ofctl_monitor.log]) AT_CHECK([ovs-ofctl monitor int-br 65534 --detach --no-chdir --pidfile 2> ofctl_monitor.log]) -@@ -565,8 +587,8 @@ icmp,vlan_tci=0x0000,dl_src=be:b6:f4:e1:49:4a,dl_dst=fe:71:d8:83:72:4f,nw_src=30 +@@ -559,14 +581,14 @@ OVS_APP_EXIT_AND_WAIT(ovs-ofctl) + + AT_CHECK([cat ofctl_monitor.log], [0], [dnl + NXT_PACKET_IN2 (xid=0x0): cookie=0x0 total_len=98 tun_id=0x7b,tun_src=1.1.2.92,tun_dst=1.1.2.88,tun_metadata0=0xa,in_port=5 (via action) data_len=98 (unbuffered) +-icmp,vlan_tci=0x0000,dl_src=be:b6:f4:e1:49:4a,dl_dst=fe:71:d8:83:72:4f,nw_src=30.0.0.1,nw_dst=30.0.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,icmp_type=0,icmp_code=0 icmp_csum:4227 ++icmp,vlan_tci=0x0000,dl_src=be:b6:f4:e1:49:4a,dl_dst=fe:71:d8:83:72:4f,nw_src=30.0.0.1,nw_dst=30.0.0.2,nw_tos=0,nw_ecn=0,nw_ttl=64,nw_frag=no,icmp_type=0,icmp_code=0 icmp_csum:4227 + ]) + AT_CHECK([ovs-ofctl dump-ports int-br | grep 'port 5'], [0], [dnl port 5: rx pkts=1, bytes=98, drop=?, errs=?, frame=?, over=?, crc=? ]) @@ -11086,6 +64478,24 @@ index 57589758f4..c63344196b 100644 ]) dnl Receive VXLAN with different MAC and verify that the neigh cache gets updated +@@ -718,14 +740,14 @@ dnl Output to tunnel from a int-br internal port. + dnl Checking that the packet arrived and it was correctly encapsulated. + AT_CHECK([ovs-ofctl add-flow int-br "in_port=LOCAL,actions=debug_slow,output:2"]) + AT_CHECK([ovs-appctl netdev-dummy/receive int-br "${packet}4"]) +-OVS_WAIT_UNTIL([test `ovs-pcap p0.pcap | egrep "${encap}${packet}4" | wc -l` -ge 1]) ++OVS_WAIT_UNTIL([test `ovs-pcap p0.pcap | grep -E "${encap}${packet}4" | wc -l` -ge 1]) + dnl Sending again to exercise the non-miss upcall path. + AT_CHECK([ovs-appctl netdev-dummy/receive int-br "${packet}4"]) +-OVS_WAIT_UNTIL([test `ovs-pcap p0.pcap | egrep "${encap}${packet}4" | wc -l` -ge 2]) ++OVS_WAIT_UNTIL([test `ovs-pcap p0.pcap | grep -E "${encap}${packet}4" | wc -l` -ge 2]) + + dnl Output to tunnel from the controller. + AT_CHECK([ovs-ofctl -O OpenFlow13 packet-out int-br CONTROLLER "debug_slow,output:2" "${packet}5"]) +-OVS_WAIT_UNTIL([test `ovs-pcap p0.pcap | egrep "${encap}${packet}5" | wc -l` -ge 1]) ++OVS_WAIT_UNTIL([test `ovs-pcap p0.pcap | grep -E "${encap}${packet}5" | wc -l` -ge 1]) + + dnl Datapath actions should not have tunnel push action. + AT_CHECK([ovs-appctl dpctl/dump-flows | grep -q tnl_push], [1]) @@ -842,3 +864,54 @@ Datapath actions: 7 OVS_VSWITCHD_STOP @@ -11154,6 +64564,19 @@ index b8ae7caa9b..fd482aa872 100644 ]) OVS_VSWITCHD_STOP +diff --git a/utilities/bugtool/ovs-bugtool.in b/utilities/bugtool/ovs-bugtool.in +index fa62cbe949..fee0de8532 100755 +--- a/utilities/bugtool/ovs-bugtool.in ++++ b/utilities/bugtool/ovs-bugtool.in +@@ -956,7 +956,7 @@ def load_plugins(just_capabilities=False, filter=None): + filters = [] + else: + filters = filters_tmp.split(',') +- if not(filter is None or filter in filters): ++ if not (filter is None or filter in filters): + continue + if el.tagName == "files": + newest_first = getBoolAttr(el, 'newest_first') diff --git a/utilities/gdb/ovs_gdb.py b/utilities/gdb/ovs_gdb.py index 62928d50fc..763ece2a78 100644 --- a/utilities/gdb/ovs_gdb.py @@ -11191,7 +64614,7 @@ index ede7f1e61a..6771973ae9 100644 ovs_be32 error_xid = error_oh->xid; enum ofperr ofperr; diff --git a/utilities/ovs-save b/utilities/ovs-save -index fb2025b765..a190902f4d 100755 +index fb2025b765..67092ecf7e 100755 --- a/utilities/ovs-save +++ b/utilities/ovs-save @@ -102,7 +102,7 @@ save_interfaces () { @@ -11203,6 +64626,15 @@ index fb2025b765..a190902f4d 100755 } save_flows () { +@@ -127,7 +127,7 @@ save_flows () { + # Get the highest enabled OpenFlow version + ofp_version=$(get_highest_ofp_version "$bridge") + +- printf "%s" "ovs-ofctl add-tlv-map ${bridge} '" ++ printf "%s" "ovs-ofctl -O $ofp_version add-tlv-map ${bridge} '" + ovs-ofctl dump-tlv-map ${bridge} -O $ofp_version | \ + awk '/^ *0x/ {if (cnt != 0) printf ","; \ + cnt++;printf "{class="$1",type="$2",len="$3"}->"$4}' diff --git a/utilities/ovs-tcpdump.in b/utilities/ovs-tcpdump.in index 82d1bedfa6..7fd26e4055 100755 --- a/utilities/ovs-tcpdump.in diff --git a/SPECS/openvswitch2.17.spec b/SPECS/openvswitch2.17.spec index 3af1e7f..9461bff 100644 --- a/SPECS/openvswitch2.17.spec +++ b/SPECS/openvswitch2.17.spec @@ -57,7 +57,7 @@ Summary: Open vSwitch Group: System Environment/Daemons daemon/database/utilities URL: http://www.openvswitch.org/ Version: 2.17.0 -Release: 31%{?dist} +Release: 44%{?dist} # Nearly all of openvswitch is ASL 2.0. The bugtool is LGPLv2+, and the # lib/sflow*.[ch] files are SISSL @@ -748,6 +748,862 @@ exit 0 %endif %changelog +* Wed Sep 21 2022 Timothy Redaelli - 2.17.0-44 +- vhost: fix virtqueue use after free on NUMA reallocation [RH git: 68652b2a12] + [ upstream commit 0b2a2ca35037d6a5168f0832c11d9858b8ae946a ] + + translate_ring_addresses (via numa_realloc) may change a virtio device and + virtio queue. + The virtqueue object must be refreshed before accessing the lock. + + Fixes: 04c27cb673b9 ("vhost: fix unsafe vring addresses modifications") + + Signed-off-by: David Marchand + Reviewed-by: Maxime Coquelin + + +* Wed Sep 14 2022 Open vSwitch CI - 2.17.0-43 +- Merging upstream branch-2.17 [RH git: e91381d8a7] + Commit list: + 5cbed27c87 tests: Fix tests with GNU grep 3.8. + + +* Mon Sep 12 2022 Open vSwitch CI - 2.17.0-42 +- Merging upstream branch-2.17 [RH git: 72e5d18fde] + Commit list: + a5cd60db0f cirrus: Upgrade to FreeBSD 13.1 image. + 43ece36f31 netdev-linux: Skip some internal kernel stats gathering. + 846d6a0c51 ofproto-dpif-xlate: Fix error messages for nonexistent ports/recirc_ids. + e8814c9b88 ofproto-dpif-xlate: Clear tunnel wc bits if original packet is non-tunnel. + + +* Tue Sep 06 2022 Timothy Redaelli - 2.17.0-41 +- Merging 7bcd45ce82 version: 21.11.2 [RH git: c132ae122f] + Commit list: + 7bcd45ce82 version: 21.11.2 + e12d415556 vhost: fix header spanned across more than two descriptors + f167022606 vhost: discard too small descriptor chains + 25c01bd323 net/mlx5: fix Rx queue recovery mechanism + 125a65cb03 examples/performance-thread: fix build with GCC 12 + 2a55c38e27 test/crypto: skip oop test for raw api + a561d44985 net/vhost: fix null pointer dereference + 0f80c13b4d version: 21.11.2-rc1 + 84b2018842 app/testpmd: fix GTP PSC raw processing + 9e7d93ae2f net/iavf: fix GTP-U extension flow + b11e955370 vdpa/sfc: resolve race between vhost lib and device conf + 06b246ead6 vdpa/ifc/base: fix null pointer dereference + dbe68f0958 vdpa/mlx5: fix leak on event thread creation + 9224015451 examples/link_status_interrupt: fix stats refresh rate + b941165a00 examples/vhost: fix retry logic on Rx path + f169902058 avoid AltiVec keyword vector + 20ee5fbe91 app/regex: fix mbuf size for multi-segment buffer + 81a0919f6a app/regex: avoid division by zero + e1c3685b21 dma/idxd: fix null dereference in PCI remove + 5b7a2b5672 dma/idxd: fix partial freeing in PCI close + 1fd4a985e4 dma/idxd: fix memory leak in PCI close + 008e1abc82 net/mlx5: reject negative integrity item configuration + fa1d93b8c4 common/mlx5: fix non-expandable global MR cache + 03a6a9f751 common/cnxk: allow changing PTP mode on CN10K + 8fb51606a7 gro: fix identifying fragmented packets + e46eb5a8dc service: fix lingering active status + 399cbc736c net/igc: support multi-process + 97e75c37b0 net/iavf: fix VF reset + bbc9dcaefa common/cnxk: fix GRE tunnel parsing + 36f4c8e67f net/virtio-user: fix Rx interrupts with multi-queue + c353b1de9f vhost: restore device information in log messages + 1f963ee7a3 vhost: add some trailing newline in log messages + 4ef6a79d4a vdpa/sfc: fix sync between QEMU and vhost-user + 5991d25b74 net/vhost: fix deadlock on vring state change + a3ff1d7e9c doc: fix readability in vhost guide + 19457a68ea net/virtio-user: fix socket non-blocking mode + 3287afa5e8 net/netvsc: fix vmbus device reference in multi-process + b6e4963255 app/testpmd: fix supported RSS offload display + 49cc0b73a0 eventdev/eth_tx: fix queue delete + a1564274cd doc: fix grammar and parameters in l2fwd-crypto guide + a8b87a7063 doc: fix grammar and formatting in compressdev guide + 8017591016 crypto/qat: fix DOCSIS crash + f2a62f854e examples/fips_validation: handle empty payload + 390e956b6d test/crypto: fix SNOW3G vector IV format + 35b1acf851 test/crypto: fix ZUC vector IV format + 6706a66aaa test/crypto: fix authentication IV for ZUC SGL + f95b184d1e doc: add more instructions for running as non-root + a74fd43471 net/bnxt: fix check for autoneg enablement in the PHY FW + 7a91bb4238 net/bnxt: cleanup MTU setting + 1ab0afa450 net/bnxt: disallow MTU change when device is started + 8185654d05 net/bnxt: fix setting forced speed + e798345849 net/bnxt: allow Tx only or Rx only + cce3a4048e net/bnxt: fix switch domain allocation + 17d26c7fa0 examples/distributor: fix distributor on Rx core + 8bbab0b5a5 net/hns3: delete unused code + 32535f69cd net/hns3: fix descriptors check with SVE + 5a05333308 net/hns3: fix statistics locking + 6420d2f828 net/hns3: fix PTP interrupt logging + bdabb55ddc net/hns3: support backplane media type + a650bf5cfe net/hns3: fix link status capability query from VF + 44dad33c76 app/testpmd: fix GTP PSC raw processing + 5dee226e46 net: fix GTP PSC headers + 95b87a5314 app/testpmd: fix flex parser destroy command + 2a4ad9bb84 app/testpmd: cleanup port resources after implicit close + bbf31ae0fc test: check memory allocation for CRC + ed8a477487 app/procinfo: show all non-owned ports + 6fce2b8067 test/hash: fix out of bound access + 7181c621fb rib: fix references for IPv6 implementation + 1b31f49983 dma/idxd: fix non-AVX builds with old compilers + d6e109f8aa dma/idxd: fix AVX2 in non-datapath functions + b72fa6fd7a raw/ioat: fix build when ioat dmadev enabled + d028271a0a raw/ioat: fix build missing errno include + a3d0dbcf03 config: fix C++ cross compiler for Arm and PPC + 3a9c3000f4 vdpa/mlx5: fix maximum number of virtqs + 45150fc78b vdpa/mlx5: workaround var offset within page + 0de69e279d doc: fix flow integrity hardware support in mlx5 guide + 02017fcad3 net/mlx5: fix stack buffer overflow in drop action + f8b370bbb4 net/mlx5: fix metering on E-Switch Manager + aa8fb4afda net/mlx5: add limitation for E-Switch Manager match + fac54fde60 net/mlx5: fix RSS expansion for patterns with ICMP item + de9fa7b453 net/mlx5: fix build with clang 14 + e4939398df net/qede: fix build with GCC 12 + cca0819d48 net/ice/base: fix build with GCC 12 + f361d278e7 net/ice: fix race condition in Rx timestamp + f294a3dbb0 net/qede: fix build with GCC 13 + 760f94b15a common/cnxk: handle ROC model init failure + 3a66cbb695 common/cnxk: fix decrypt packet count register update + 614cd42ac0 net/octeontx: fix port close + bee8c21938 malloc: fix allocation of almost hugepage size + 7b610e0a8d net/virtio: unmap PCI device in secondary process + d6e4e0f46e vhost/crypto: fix descriptor processing + f69a61bde0 vhost/crypto: fix build with GCC 12 + 361723acef vhost: fix missing enqueue pseudo-header calculation + 76556a3128 app/testpmd: revert MAC update in checksum forwarding + 1901dc5492 net/ngbe: add more packet statistics + a92e31d35d net/txgbe: fix register polling + f48795dea1 app/testpmd: fix bonding slave devices not released + b3cfb3db85 app/testpmd: add help messages for multi-process + bfaaf994a5 net/hns3: fix TM capability + 35582af08c net/hns3: fix crash from secondary process + ba4aa140b3 net/hns3: fix return value for unsupported tuple + 332e5fca03 net/hns3: fix code check warning + fc61bd5d37 net/hns3: remove duplicate definition + e885f508d9 net/hns3: fix an unreasonable memset + 8854374c9e test/bonding: fix RSS test when disable RSS + 2b71d44b80 net/bonding: fix RSS inconsistency between ports + bd9ffc1961 eventdev/eth_tx: fix adapter creation + 19591ad643 event/dlb2: fix advertized capabilities + f7b34f357c event/cnxk: fix Tx adapter enqueue return for CN10K + 93b1138ccd event/cnxk: fix QoS parameter handling + e6f569043c event/dlb2: fix check of QID in-flight + 7c0439f319 event/dlb2: rework queue drain handling + d2c3d326d0 event/octeontx: fix SSO fast path + bf7aa26ddd net/nfp: fix initialization + 1c770fda6f net/nfp: make sure MTU is never larger than mbuf size + fe2cddeb08 net/nfp: update how max MTU is read + 095d2af061 crypto/cnxk: swap zuc-256 iv + 38eabfdd0d common/cnxk: swap zuc-256 key + d1e2bd80a6 test/ipsec: fix performance test + 387d7f2a33 test/crypto: fix cipher offset for ZUC + bf03e0341c crypto/scheduler: fix queue pair in scheduler failover + 9445fcf138 test/ipsec: fix build with GCC 12 + e0bff8480f crypto/cnxk: fix build with GCC 12 + 978835ed87 common/cpt: fix build with GCC 12 + a9485fd00b examples/ipsec-secgw: fix ESN setting + 362a219f40 net/iavf: fix NAT-T payload length + be3beb946e examples/ipsec-secgw: fix NAT-T header fields + d6a5fb4092 ipsec: fix NAT-T ports and length + fd2d725ae5 baseband/acc100: add protection for some negative scenario + 4184a99adb baseband/acc100: update companion PF configure function + 4fb5429816 eal/x86: drop export of internal alignment macro + bba01c7ab8 sched: remove unnecessary floating point + 03b38f5281 test: drop reference to removed tests + b55b2820d2 trace: fix init with long file prefix + 17615c81fe trace: fix crash when exiting + 64fdce75b6 net/mlx5: fix RSS hash types adjustment + 1fb92a1f45 net/bnxt: fix tunnel stateless offloads + e10c862914 net/iavf: fix segfaults when calling API after VF reset failed + 8e8886a0e9 dma/hisilicon: fix includes in header file + e027f40cd2 dma/skeleton: fix index returned when no memcpy completed + 675b5bdf2c app/flow-perf: fix build with GCC 12 + f85d0fc397 vdpa/ifc: fix build with GCC 12 + ec6a2fa05c net/ice: fix build with GCC 12 + ac8e3a7546 net/enetfec: fix build with GCC 12 + 9c1822f59f net/ena: fix build with GCC 12 + c86456efc9 crypto/ipsec_mb: fix build with GCC 12 + 4cfe560401 kni: use dedicated function to set MAC address + 9b7982b986 kni: use dedicated function to set random MAC address + e731132bca net/tap: fix device freeing + 63bb35c3f3 net/failsafe: fix device freeing + a9062fa2fc app/testpmd: fix multicast address pool leak + c18ad5cc3e app/testpmd: fix packet segment allocation + 8bb9213bbc dma/idxd: fix error code for PCI device commands + 5215fd05ab doc: fix formatting and link in BPF library guide + 7133eadc9c bus/fslmc: fix VFIO setup + d2d91f50f7 raw/ifpga: unregister interrupt on close + 56e6acc152 raw/ifpga: remove virtual devices on close + e06a55362a eal/ppc: fix compilation for musl + c3a48df3d5 dma/hisilicon: enhance CQ scan robustness + 543121b53e dma/hisilicon: fix index returned when no DMA completed + d1461844a6 examples/dma: fix Tx drop statistics + 6564af3d39 examples/dma: fix MTU configuration + 5d71b3d9fa common/mlx5: remove unused lcore check + 94b9525189 net/iavf: remove dead code + f0c897ea5a net/iavf: increase reset complete wait count + 341d13b08b net/iavf: fix device stop + eab5e035ce net/iavf: fix device initialization without inline crypto + 78cf4cbe62 doc: update matching versions in i40e guide + d124639aee net/iavf: fix Rx queue interrupt setting + 4a42ee9346 net/iavf: fix mbuf release in multi-process + e1a84de6a9 net/iavf: fix queue start exception handling + 6730951205 net/i40e: fix max frame size config at port level + cfa67fc84b net/ice: fix MTU info for DCF + 356142f8a2 net/ice/base: fix direction of flow that matches any + 745563ca3c net/ice/base: fix getting sched node from ID type + 97f8a95696 net/ixgbe: add option for link up check on pin SDP3 + 111417a49b net/iavf: fix data path selection + c8868b3c5c kni: fix build + 3aeeea257f kni: fix build with Linux 5.18 + 301300a86e net/mlx5: fix statistics read on Linux + 83abe945a6 net/mlx5: fix Tx recovery + f06feb0822 examples/vhost: fix crash when no VMDq + a7c72e3e6c vhost: fix deadlock when message handling failed + e156da31dd doc: fix vhost multi-queue reconnection + 17a0ef7be3 vhost: fix async access + c8c6eeda37 net/bnxt: fix ULP parser to ignore segment offset + ca961550e9 net/bnxt: fix compatibility with some old firmwares + ee3b68b408 ethdev: fix port close in secondary process + da7caee013 common/sfc_efx/base: convert EFX PCIe INTF to MCDI value + 959cd86178 net/vmxnet3: fix Rx data ring initialization + 8c381b1157 app/testpmd: fix help of create meter command + 237d93b36c net/nfp: fix disabling VLAN stripping + b3ef192fec net/txgbe: fix max number of queues for SR-IOV + 7b5339d563 net/txgbe: fix SGMII mode to link up + 8bf4f37ede net/ngbe: fix PCIe related operations with bus API + 512f325928 net/ngbe: fix reading PHY ID + ba78db53ee net/ngbe: fix link speed check + 8e23b06316 ethdev: fix port state when stop + 54cb103e7b net/memif: fix overwriting of head segment + fde361696c net/bonding: fix mbuf fast free usage + ce5917f846 app/testpmd: do not poll stopped queues + 8b28d584d8 app/testpmd: fix use of indirect action after port close + 28e88ef39e ethdev: prohibit polling stopped queue + 011122b9e0 app/testpmd: fix metering and policing command for RFC4115 + e154ece049 app/testpmd: replace hardcoded min mbuf number with macro + f7638851b4 net/cnxk: fix possible null dereference in telemetry + c05dd44f52 ethdev: fix possible null pointer access + e5177f3853 ethdev: fix memory leak in xstats telemetry + 2104014dbd net/axgbe: fix xstats get return if xstats is null + 3422f4b58c net/mvpp2: fix xstats get return if xstats is null + 7a1086a9d5 net/ipn3ke: fix xstats get return if xstats is null + bae6c70cd3 net/hns3: fix xstats get return if xstats is null + 0c48dafbdf app/testpmd: remove useless pointer checks + 8378498b96 app/testpmd: perform SW IP checksum for GRO/GSO packets + 65bff89f9a app/testpmd: fix port status of bonding slave device + 4f9c7fb5af doc: add missing auth algo for IPsec example + 2d0ec22be8 test/crypto: fix driver name for DPAA raw API test + 082148b6a5 drivers/crypto: fix warnings for OpenSSL version + 48dda925a7 test/crypto: fix null check for ZUC authentication + c195ec01df examples/ipsec-secgw: fix promiscuous mode option + 9c33903649 examples/ipsec-secgw: fix uninitialized memory access + ea0ab8e686 pcapng: fix timestamp wrapping in output files + 412da85334 pipeline: fix emit instruction for invalid headers + ce7b8e673a devtools: fix null test for NUMA systems + 84eb565954 doc: fix API index Markdown syntax + d55a70f874 mbuf: dump outer VLAN + 2fcd1cc163 rib: fix traversal with /32 route + 8c5ab722fb acl: fix rules with 8-byte field size + 5ffee1e906 test: avoid hang if queues are full and Tx fails + c1f49d47a9 eal/freebsd: fix use of newer cpuset macros + 214462a05b devargs: fix leak on hotplug failure + 29fa5a6eaf eal/x86: fix unaligned access for small memcpy + eeaeb58d56 event/cnxk: fix out of bounds access in test + 09d859555f eventdev/eth_rx: fix telemetry Rx stats reset + ec08dcaf4b doc: fix build with sphinx 4.5 + 671e8fa0c8 net/mlx5: fix no-green metering with RSS + 6857653625 net/bnxt: fix freeing VNIC filters + fa1a893ff7 net/bnxt: recheck FW readiness if in reset process + 2ff3768d20 net/bnxt: fix link status when port is stopped + 71ab79d3a7 net/bnxt: force PHY update on certain configurations + db239d7290 net/bnxt: fix speed autonegotiation + ce36a5d910 net/bnxt: avoid unnecessary endianness conversion + 8c464cf618 net/bnxt: handle queue stop during RSS flow create + c25b1d545e net/bnxt: check duplicate queue IDs + 3f9914a7b3 net/bnxt: fix ring group on Rx restart + 821dd9cd43 net/bnxt: fix RSS action + 3774986bdd net/bnxt: fix Rx configuration + d620238a97 net/bnxt: remove unused macro + 2f66d10615 net/bnxt: fix device capability reporting + b174adfcae net/bnxt: fix reordering in NEON Rx + 7d9f5b3b33 net/cnxk: add barrier after meta batch free in scalar + 8790891a6d common/cnxk: fix SQ flush sequence + 08d2d8868e net/cnxk: fix uninitialized variables + a10e2ec8ee common/cnxk: fix null pointer dereference + d13786763a common/cnxk: fix unaligned access to device memory + b117088323 net/cnxk: add message on flow parsing failure + caf428f0a0 app/testpmd: fix MTU verification + 5e1545b730 app/testpmd: check statistics query before printing + 66b7e330d6 net/hns3: remove unnecessary RSS switch + 15b794b152 ethdev: fix RSS update when RSS is disabled + 0cec1c9477 net/hns3: remove redundant RSS tuple field + 5a6fb3a977 net/hns3: fix rollback on RSS hash update + 8c193c0b4c net/hns3: fix RSS disable + 8b00917c08 net/hns3: fix mbuf free on Tx done cleanup + 74089f471a net/hns3: fix pseudo-sharing between threads + 827f72e8ce net/hns3: fix MAC and queues HW statistics overflow + d96ee7bac0 net/hns3: fix order of clearing imissed register in PF + aa2c6d3f69 ethdev: fix build with vtune option + b839853e83 net/tap: fix interrupt handler freeing + 41c0ba64de net/bonding: fix slave stop and remove on port close + 5a8afc69af net/bonding: fix stopping non-active slaves + e856fe9aa6 doc: update matching versions in ice guide + 82ccc27de5 net/dpaa: fix event queue detach + d24d6395d6 vdpa/mlx5: fix dead loop when process interrupted + 879fb64517 vdpa/mlx5: fix interrupt trash that leads to crash + 78414da84e vhost: fix missing virtqueue lock protection + e3036fbd0a net/vhost: fix TSO feature default disablement + 4852da727c net/virtio: restore some optimisations with AVX512 + 58d1b856be net/vhost: fix access to freed memory + 24dabb9d25 net/cxgbe: fix Tx queue stuck with mbuf chain coalescing + 6627ee48b5 net/cxgbe: fix port ID in Rx mbuf + 8cf194f699 net/bonding: fix RSS key config with extended key length + 3192737d10 net/nfp: remove unneeded header inclusion + 8ab93b06bc net/netvsc: fix hot adding multiple VF PCI devices + 03e1864411 test/mem: disable ASan when accessing unallocated memory + e9b46ab763 net/mlx5: fix LRO configuration in drop Rx queue + d5fdf0a2ba net/mlx5: fix LRO validation in Rx setup + 28ecf49a60 examples/l2fwd-crypto: fix stats refresh rate + aeca5959dd common/dpaax: fix short MAC-I IV calculation for ZUC + 5a9af71a6d crypto/dpaa2_sec: fix operation status for simple FD + 5e3a3f48d1 crypto/dpaa2_sec: fix crypto operation pointer + 4644779034 crypto/dpaa_sec: fix secondary process probing + 15a3ae1a5f crypto/dpaa2_sec: fix chained FD length in raw datapath + bee2c296c8 crypto/dpaa_sec: fix chained FD length in raw datapath + 86ba4e206e crypto/dpaa2_sec: fix buffer pool ID check + f72e482fec crypto/dpaa2_sec: fix fle buffer leak + 8bad3a05f1 crypto/mlx5: fix login cleanup + be6637f158 security: fix SA lifetime comments + bb386a9f91 crypto/dpaa_sec: fix digest size + f343d3b4ed eal: fix C++ include for device event and DMA + de48c79f3b malloc: fix ASan handling for unmapped memory + 804b2e64eb mem: skip attaching external memory in secondary process + 65855b2d37 test/table: fix buffer overflow on lpm entry + c7e0471948 net/mlx5: fix Rx/Tx stats concurrency + b0e6a9c183 net/mlx5: fix GTP handling in header modify action + b3896dba13 net/mlx5: restrict Rx queue array access to boundary + c08c6247f2 net/mlx5: fix counter in non-termination meter + 99ba358268 net/mlx5: fix probing with secondary bonding member + 1430ccb1db net/mlx5: fix Tx when inlining is impossible + 72691359fa common/mlx5: fix memory region range calculation + 550f0d8288 net/netvsc: fix calculation of checksums based on mbuf flag + 21edf23c6d net/ice: fix raw flow input pattern parsing + aedf24edbb net/ice: refactor parser usage + bb6683a89f net/ice: add missing Tx burst mode name + 91355ad5b2 net/i40e: populate error in flow director parser + 8ae457cbf5 net/ice: improve performance of Rx timestamp offload + a25197930d test/bpf: skip test if libpcap is unavailable + 6da5f268db examples/bond: fix invalid use of trylock + 9c267cbd10 net/dpaa2: fix dpdmux default interface + d8898f0763 eal/windows: add missing C++ include guards + fad1dbc0c5 eal/windows: fix data race when creating threads + 95e04d4866 doc: fix release note typo + 592c7bf714 net/af_xdp: make compatible with libbpf >= 0.7.0 + 6721fb14eb net/af_xdp: use libxdp if available + fcd039e466 version: 21.11.1 + 2130012318 net/cnxk: fix build with optimization + 9518bcf700 net/mlx5: fix flex item availability + 05aa560efc version: 21.11.1-rc1 + b68dbab7c8 Revert "net/mlx5: fix flex item availability" + bb5ce0625c crypto/ipsec_mb: fix GMAC parameters setting + cef6bb00ce crypto/ipsec_mb: fix length and offset settings + be2edca509 Revert "crypto/ipsec_mb: fix length and offset settings" + dec4b1b89e raw/ifpga: fix build with optimization + 4586b6b8c2 doc: fix telemetry example in cryptodev guide + 2740b29e48 doc: fix missing note on UIO module in Linux guide + 713a4bc48c doc: replace characters for (R) symbol in Linux guide + a50b228d2d net/mlx5: fix CPU socket ID for Rx queue creation + 8db2867c79 net/mlx5: fix port matching in sample flow rule + 7c12be128c eventdev: fix clang C++ include + 4f263532d0 cryptodev: fix clang C++ include + ec8a6dc2e6 compressdev: fix missing space in log macro + 37232971b0 eal/freebsd: add missing C++ include guards + 8320df4804 examples/l3fwd: fix buffer overflow in Tx + 3313fe0301 app/testpmd: fix flow rule with flex input link + 39d09d7155 app/testpmd: fix GTP header parsing in checksum engine + 514668e230 app/testpmd: fix show RSS RETA on Windows + 3a3d4d3332 app/regex: fix number of matches + f2a457c605 bpf: fix build with some libpcap version on FreeBSD + e84b43b5eb crypto/ipsec_mb: fix GCM requested digest length + 8c7bebaa38 net/af_xdp: fix custom program loading with multiple queues + 88dbe7c555 net/qede: fix maximum Rx packet length + 09891782a4 net/qede: fix Rx bulk + 506f3198ab net/qede: fix Tx completion + 268985d32e doc: fix modify field action description for mlx5 + 59a419a416 net/mlx5: fix implicit tag insertion with sample action + 42cf1850e2 net/mlx5: forbid multiple ASO actions in a single rule + dd859e1797 net/mlx5: fix sample flow action on trusted device + 7680d1d321 net/mlx5: fix VLAN push action validation + 691ff0b6db net/mlx5: fix NIC egress flow mismatch in switchdev mode + 6cb68162e4 vhost: fix FD leak with inflight messages + 4c40d30d2b vhost: fix queue number check when setting inflight FD + 6ae8ba6b7a build: suppress rte_crypto_asym_op abi check + efd091d541 devtools: fix symbols check + 026470bafa build: hide local symbols in shared libraries + 89f14be564 common/mlx5: consider local functions as internal + 6e7f8939f2 regexdev: fix section attribute of symbols + 6472c2d476 net/iavf: fix potential out-of-bounds access + 67191a9cb3 net/sfc: reduce log level of tunnel restore info error + ee836190a1 net/mlx5: fix meter creation default state + a17cea76b7 net/mlx5: fix configuration without Rx queue + d31463e0b2 net/mlx5: fix MPLS/GRE Verbs spec ordering + 48fe9efaf2 net/mlx5: fix flex item availability + 3bd5cf393d net/mlx5: fix meter policy creation assert + c77572d2a1 net/mlx5: remove unused reference counter + 0036f3941e net/mlx5: fix modify port action validation + eebfb74c51 net/mlx5: fix shared RSS destroy + 5d3ade99bd net/mlx5: fix next protocol RSS expansion + 4500ec704f net/mlx5: fix inet IPIP protocol type + 9bdcba122b net/bnxt: fix null dereference in session cleanup + 4aadf56c66 ethdev: fix doxygen comments for device info struct + 0c7cbe52f7 build: fix build on FreeBSD with Meson 0.61.1 + dab4a96be2 devtools: remove event/dlb exception in ABI check + 4fa43b7bff vhost: fix physical address mapping + f03f4b98c9 net/cnxk: fix Rx/Tx function update + 3a5e1aaee4 net/mlx5: fix initial link status detection + 295f5022f6 net/mlx5: fix link status change detection + be828a8eaf common/mlx5: add Netlink event helpers + 7214354c52 examples/kni: add missing trailing newline in log + f5ba75eb9a examples/l3fwd: make Rx and Tx queue size configurable + ef48f23bfd examples/l3fwd: share queue size variables + 6bf720d7d7 examples/flow_classify: fix failure message + 2719708908 examples/distributor: reduce Tx queue number to 1 + 7aa3bbafd6 app/dumpcap: check for failure to set promiscuous + 955a6afc6f test/bpf: skip dump if conversion fails + e71f3dc931 pcapng: handle failure of link status query + b8222349eb app/pdump: abort on multi-core capture limit + 8adbf6df92 raw/ifpga: fix monitor thread + 69da51b405 raw/ifpga: fix interrupt handle allocation + d4536cf86a raw/ifpga: fix variable initialization in probing + 186250df3d gpu/cuda: fix dependency loading path + af8ffbba79 sched: remove useless malloc in PIE data init + 8ebcaf23cb eal/linux: fix device monitor stop return + 89d84883b0 examples/vhost: fix launch with physical port + 7a5659dd94 vhost: fix linker script syntax + b7f396be62 net/ice: fix Tx offload path choice + 28acfe550d common/cnxk: fix mbuf data offset for VF + 51af57d005 common/cnxk: fix bitmap usage for TM + 43dec151be net/iavf: fix AES-GMAC IV size + f314e6acfb net/mlx5: fix flex item header length translation + 1926a8d8c5 net/mlx5: fix matcher priority with ICMP or ICMPv6 + 1f5aede9bd net/mlx5: reduce flex item flow handle size + 279cc42d3b net/mlx5: fix GRE item translation in Verbs + 39cba36e63 doc: fix typos and punctuation in flow API guide + 41510092eb net/kni: fix config initialization + 6090ee620d net/txgbe: fix queue statistics mapping + 8a301f166c net/mlx5: fix check in count action validation + c46eaf6f4c net/mlx5: fix shared counter flag in flow validation + de3ad851ca net/mlx5: fix destroying empty matchers list + 6468addfe3 net/mlx5: fix indexed pool fetch overlap + ae071e1851 net/iavf: fix function pointer in multi-process + b82b6ed613 net/iavf: support NAT-T / UDP encapsulation + 5f275a0312 net/ixgbe: fix FSP check for X550EM devices + aa6f865e7e net/hns3: increase time waiting for PF reset completion + 94420985c7 net/hns3: fix VF RSS TC mode entry + dc3cb423f5 net/hns3: fix RSS TC mode entry + 772292049b net/hns3: remove duplicate macro definition + 24939fcc13 compressdev: fix socket ID type + 30fea0f0a6 app/compress-perf: fix number of queue pairs to setup + b2b15ab556 app/compress-perf: fix socket ID type during init + 8ace98122a compress/mlx5: support out-of-space status + d386e37612 app/compress-perf: optimize operations pool allocation + c65e648405 app/compress-perf: fix cycle count operations allocation + 9bb7a3f9df event/dlb2: add shift value check in sparse dequeue + d2b19d6346 event/cnxk: fix Rx adapter config check + dd8c73295c event/cnxk: fix sub-event clearing mask length + 170c124998 kni: fix freeing order in device release + 0617d94900 bus/pci: assign driver pointer before mapping + 099aba7265 devargs: fix crash with uninitialized parsing + dcf545fce1 eal/linux: fix illegal memory access in uevent handler + 38c59b06b0 distributor: fix potential overflow + 77b6873f73 efd: fix uninitialized structure + b017e1159f test/efd: fix sockets mask size + e9100a0196 doc: add CUDA driver features + 9703132099 app/testpmd: fix build without drivers + 158012beee app/testpmd: fix raw encap of GENEVE option + 8c4ce4d7ff net/i40e: fix unintentional integer overflow + 3334722c21 net/cnxk: fix RSS RETA table update + b8bfbcd1a0 net/cnxk: fix build with GCC 12 + c957e1063b net/cnxk: fix inline IPsec security error handling + ee97d867e7 net/cnxk: register callback early to handle initial packets + c5124d0ea8 net/cnxk: fix inline device RQ tag mask + 283f54ba9d mempool/cnxk: fix batch allocation failure path + ba9d00afac doc: correct name of BlueField-2 in mlx5 guide + 9385e97741 doc: replace broken links in mlx guides + 239796f3dd doc: remove obsolete vector Tx explanations from mlx5 guide + 2007577b29 net/mlx5: fix E-Switch manager vport ID + a600672d1a net/mlx5: fix entry in shared Rx queues list + 7f982e1320 net/mlx5: fix meter sub-policy creation + 7b5ea7efc3 net/mlx5: remove unused function + aff5b2ee60 net/mlx5: set flow error for hash list create + d2e99680f9 common/mlx5: fix queue pair ack timeout configuration + 233c5aa3e7 net/ena: fix checksum flag for L4 + f5eff853e4 net/ena: check memory BAR before initializing LLQ + 70c3e891d0 net/ena: fix meta descriptor DF flag setup + 867dd857f4 net/ena: fix reset reason being overwritten + c443512e3d net/ena: skip timer if reset is triggered + 4e9e9e29c4 net/ena: remove unused offload variables + e63e5c79ce net/ena: remove unused enumeration + 399b489328 net/txgbe: fix debug logs + a8be311dd1 net/ngbe: fix debug logs + c1cf1a9735 app/testpmd: fix GENEVE parsing in checksum mode + 036993974f net/mlx5: fix errno update in shared context creation + b6b1c3ad5d net/mlx5: fix ASO CT object release + 49257a9394 net/mlx5: fix ineffective metadata argument adjustment + 50f3a03f75 net/mlx5: fix sibling device config check + e68285796c net/i40e: enable maximum frame size at port level + 15ff989ca8 net/iavf: fix segmentation offload buffer size + dbb1c53725 net/iavf: fix segmentation offload condition + d75be6c28d net/ice: fix overwriting of LSE bit by DCF + a628e2bf19 net/af_xdp: ensure socket is deleted on Rx queue setup error + ae2f030ad1 net/sfc: fix memory allocation size for cache + ea21c6bf4e net/sfc: fix flow tunnel support detection + a58ae9af98 common/sfc_efx/base: add missing handler for 1-byte fields + 4874f1d005 common/sfc_efx/base: fix recirculation ID set in outer rules + e4b43ee28c net/cnxk: fix uninitialized local variable + f0cfb0e3d1 common/cnxk: fix uninitialized pointer read + 2f61027cda common/cnxk fix unintended sign extension + 7eeb8d37ed common/cnxk: add missing checks of return values + dd1851c1de net/af_xdp: add missing trailing newline in logs + 6a9b64907e common/cnxk: fix NPC key extraction validation + 87b639b4ed vhost: fix unsafe vring addresses modifications + 01e3dee29c vhost: fix field naming in guest page struct + e09a0094a6 common/cnxk: fix base rule merge + 1751e87f51 common/cnxk: fix log level during MCAM allocation + d91869302f common/cnxk: fix flow deletion + 450ee57e5f app/testpmd: check starting port is not in bonding + 387187932f net/bonding: fix slaves initializing on MTU setting + c93302dd4f net/cnxk: fix mbuf data length + 116bfaa14e ethdev: fix MAC address in telemetry device info + a42a874599 net/iavf: reset security context pointer on stop + 496747d389 net/txgbe: reset security context pointer on close + 223010f1da net/ixgbe: reset security context pointer on close + 967cb49748 net/nfb: fix multicast/promiscuous mode switching + afe8e58fed net/nfb: fix array indexes in deinit functions + daf06c45e8 crypto/ipsec_mb: fix length and offset settings + cfa7703c8e crypto/ipsec_mb: fix ZUC operation overwrite + 1170e24b20 crypto/ipsec_mb: fix ZUC authentication verify + bbc596578a crypto/ipsec_mb: check missing operation types + 9c67637c8c crypto/virtio: fix out-of-bounds access + 301ee2f378 baseband/acc100: avoid out-of-bounds access + 79247ddc0d examples/l2fwd-crypto: fix port mask overflow + 5772c7b32e doc: fix FIPS guide + ad76dc4e91 examples/ipsec-secgw: fix buffer freeing in vector mode + f092922c36 cryptodev: fix RSA key type name + c8bcbe8b68 crypto/ipsec_mb: remove useless check + fec66e64e5 event/cnxk: fix uninitialized local variables + 52d824d106 event/cnxk: fix variables casting + 9a552423fd event/dlb2: poll HW CQ inflights before mapping queue + 720fb431b4 event/dlb2: update rolling mask used for dequeue + 7d7a9f161d eventdev/eth_rx: fix queue config query + 529f3a735e eventdev/eth_rx: fix parameters parsing memory leak + d33bb6bd28 examples/qos_sched: fix core mask overflow + 9970eab8c4 doc: improve configuration examples in idxd guide + b254386fad dma/idxd: configure maximum batch size to high value + ebc0188ccb test/dma: fix missing checks for device capacity + c6aea57d99 dma/hisilicon: use common PCI device naming + 56d6e5b091 ethdev: fix cast for C++ compatibility + 5d75eb0924 cryptodev: add missing C++ guards + c02f5bcfe9 bpf: add missing C++ guards + 362921a8e3 vhost: add missing C++ guards + fb37e2b3ae kni: add missing C++ guards + e4dbb6873a eventdev: add missing C++ guards + 3d1746c9ac compressdev: add missing C++ guards + 6fca954338 acl: add missing C++ guards + 447210e07c metrics: add missing C++ guards + b99a45df05 ethdev: add missing C++ guards + e7291176c2 telemetry: add missing C++ guards + b2f85a808a eal: add missing C++ guards + 81c40b01d0 dmadev: add missing header include + 692ae335d6 eventdev/eth_tx: fix queue add error code + adfebc59b5 pipeline: fix table state memory allocation + 1e8aa23aba pipeline: fix annotation checks + 13ddcf9dee raw/ntb: clear all valid doorbell bits on init + 0627e93c26 crypto/dpaax_sec: fix auth/cipher xform chain checks + 0fd24703c6 crypto/cnxk: fix update of number of descriptors + 2630bff5a4 compress/octeontx: fix null pointer dereference + 6f9d8df3d1 crypto/qat: fix GEN4 AEAD job in raw data path + 98ec92641b crypto/ipsec_mb: fix buffer overrun + 13aab9f493 crypto/ipsec_mb: fix premature dereference + 98ece68514 test/crypto: fix out-of-place SGL in raw datapath + 4d5d4d7abc examples/ipsec-secgw: fix offload flag used for TSO IPv6 + abfad6b59c net/txgbe: fix KR auto-negotiation + 42960ce408 net/txgbe: fix link up and down + 59691181a3 net/ngbe: fix packet statistics + b9c20ea8f0 net/ngbe: fix Tx hang on queue disable + 3698c17f42 net/ngbe: fix missed link interrupt + cacbd7e4f7 net/ngbe: fix Rx by initializing packet buffer early + 863d787942 net/bnxt: fix ring calculation for representors + e53da2ffbe net/bnxt: set HW coalescing parameters + 6c8ff52958 net/mlx5: fix inline length for multi-segment TSO + 3831da6c7b net/mlx5: fix meter capabilities reporting + 6022babd1d net/mlx5: fix committed bucket size + c9a140e15b net/mlx5: fix metadata endianness in modify field action + 8f821b1135 vdpa/sfc: fix null dereference during removal + 5dfd488d82 vdpa/sfc: fix null dereference during config + 7537c99618 net/ice: fix build with 16-byte Rx descriptor + 607d564355 net/ice: fix pattern check in flow director + d422a9cdae net/ice/base: add profile validation on switch filter + 5dc74f1348 net/iavf: count continuous DD bits for Arm in flex Rx + 68522027e3 net/iavf: count continuous DD bits for Arm + f746bb72ba net/iavf: fix null pointer dereference + b832a197fa net/sfc: demand Tx fast free offload on EF10 simple datapath + 915b0b0b9f net/sfc: do not push fast free offload to default TxQ config + 6fdd1953b0 ethdev: remove unnecessary null check + 851b597291 net: fix L2TPv2 common header + d594afc792 net/memif: remove pointer deference before null check + 273bacf2a8 config: align mempool elements to 128 bytes on CN10K + e183e43e2b vfio: cleanup the multiprocess sync handle + c32322e508 ipc: end multiprocess thread during cleanup + 6e1bc26cde test/mbuf: fix mbuf data content check + 472f790f95 app/fib: fix division by zero + 1058b2c369 mem: check allocation in dynamic hugepage init + c59904ed03 vhost: fix C++ include + 6afaa0f3d7 table: fix C++ include + 91b9d6cd34 ipsec: fix C++ include + 4f328f8e2b graph: fix C++ include + 3668e54828 eventdev: fix C++ include + 1fdfd87f14 eal: fix C++ include + 72334ceaf1 config/arm: add values for native armv7 + 02a96ad251 stack: fix stubs header export + 6b06137c98 regex/mlx5: fix memory allocation check + 72487940d1 net/virtio: fix slots number when indirect feature on + 160769f648 vhost: fix guest to host physical address mapping + eaf935f63c net/sfc: fix lock releases + ce413e1922 app/testpmd: fix stack overflow for EEPROM display + ac180f4d26 net/tap: fix to populate FDs in secondary process + 721d0bbd16 ethdev: add internal function to device struct from name + 605d1de0d3 app/testpmd: fix bonding mode set + 7b71bc2d00 net/bonding: fix reference count on mbufs + 26f2cc6490 net/bonding: fix promiscuous and allmulticast state + 30dcde8467 net/ixgbe: check filter init failure + e72696baa4 net/hns3: delete duplicated RSS type + 2ae91ac660 net/hns3: fix operating queue when TCAM table is invalid + 2b7587ea99 net/hns3: fix double decrement of secondary count + 10342b22ae net/hns3: fix insecure way to query MAC statistics + 9b1f69f906 net/hns3: fix RSS key with null + d7033074e0 net/hns3: fix max packet size rollback in PF + 2c27da1e51 net/bonding: fix MTU set for slaves + 9ac1343c4d net/dpaa2: fix null pointer dereference + 90386f428c net/enic: fix dereference before null check + 57b2aa0265 test/mem: fix error check + 32cb4f09ff eal/windows: fix error code for not supported API + 5a9f8c2ba4 ring: fix overflow in memory size calculation + 8b45a1dea3 ring: fix error code when creating ring + 63cb4ae54f doc: fix KNI PMD name typo + 4f140c9a9a build: remove deprecated Meson functions + aa8ad3e48c build: fix warnings when running external commands + 1e770ae599 pflock: fix header file installation + 86f7ed09ae doc: update matching versions in ice guide + 37d27abc59 net/mlx5: reject jump to root table + bc3452d45d common/mlx5: fix probing failure code + 99f5cd0dc3 net/mlx5: fix mark enabling for Rx + d157628041 common/mlx5: fix MR lookup for non-contiguous mempool + 4c4c0cf459 net/virtio: fix uninitialized RSS key + 0d2ddde419 net/virtio-user: check FD flags getting failure + 4210bb89d8 net/virtio-user: fix resource leak on probing failure + efc7ea9dd7 vdpa/ifc: fix log info mismatch + 7c58dbf159 net/virtio: fix Tx queue 0 overriden by queue 128 + f05bbce185 vdpa/mlx5: workaround queue stop with traffic + ad51b31a30 net/hns3: fix using enum as boolean + 9d6db3c3ad net/nfp: free HW ring memzone on queue release + 961922eb71 net/bonding: fix RSS with early configure + 6492c9875d net/hns3: fix vector Rx/Tx when PTP enabled + 9c10b251a1 net/hns3: fix mailbox wait time + e073f410fb net/hns3: fix Rx/Tx functions update + 581e547a6f net/memif: remove unnecessary Rx interrupt stub + 5de680a494 raw/ifpga/base: fix port feature ID + 0f8f337740 net/bnxt: fix VF resource allocation strategy + f70203b5c0 net/bnxt: fix memzone allocation per VNIC + e44c18821c net/bnxt: handle ring cleanup in case of error + a04034b131 net/bnxt: fix check for autoneg enablement + 72db0cca69 raw/ifpga: fix thread closing + 7c682d5c05 net/ice: fix link up when starting device + b38f8855d6 net/ice: fix mbuf offload flag for Rx timestamp + 81597d6e20 raw/ifpga/base: fix SPI transaction + 4599a6179a net/sfc: validate queue span when parsing flow action RSS + c935f2719d ethdev: fix Rx queue telemetry memory leak on failure + 3fd3c3b3b0 common/cnxk: fix error checking + 2253ed93c3 common/cnxk: fix uninitialized variables + 38f3a00894 common/cnxk: fix null pointer dereferences + 1349f9e568 common/cnxk: always use single interrupt ID with NIX + cfcdf00068 common/cnxk: reset stale values on error debug registers + 285183e606 common/cnxk: fix byte order of frag sizes and infos + 5deff57b9d common/cnxk: fix shift offset for TL3 length disable + 41569f9deb net/nfp: remove useless range checks + 23c2f68598 net/nfp: remove duplicated check when setting MAC address + 85d9e45c32 net/mlx5: fix MPRQ WQE size assertion + a9bc2a46c2 net/mlx5: fix maximum packet headers size for TSO + 70211750cc net/bnxt: restore dependency on kernel modules + 279f0d75c0 net/dpaa2: fix timestamping for IEEE1588 + c96ea2bf1c net/dpaa2: fix unregistering interrupt handler + 699c30f853 net/cxgbe: fix dangling pointer by mailbox access rework + ef94549efe app/testpmd: fix external buffer allocation + 6d5f3984dc app/testpmd: fix dereference before null check + 83774f8a67 net/bonding: fix mode type mismatch + 40a4d0544b net/af_xdp: fix build with -Wunused-function + 181ddedb1a net/axgbe: use PCI root complex device to distinguish device + bb1854bc69 app/testpmd: fix Tx scheduling interval + 96b92d045f net/bonding: fix offloading configuration + 661587eaae net/cnxk: fix promiscuous mode in multicast enable flow + 26a3e3e7d3 net/bnxt: check VF representor pointer before access + 2f9df1413c net/bnxt: fix xstats query + b74a60df81 net/bnxt: fix crash by validating pointer + 2e31b779e8 net/bnxt: fix PAM4 mask setting + 6c57090c01 net/bnxt: fix ring teardown + f2c08d53a1 net/bnxt: fix handling of VF configuration change + 377a9a8197 net/bnxt: get maximum supported multicast filters count + b0fe5e2fa9 net/bnxt: fix flow create when RSS is disabled + 6b722d7b37 net/bnxt: add null check for mark table + a31a8b6a97 net/bnxt: set fast-path pointers only if recovery succeeds + 6b7c0ce0ce net/bnxt: cap maximum number of unicast MAC addresses + a9ea24c80c net/bnxt: fix restoring VLAN filtering after recovery + 56f92b77e7 net/bnxt: restore RSS configuration after reset recovery + 880ed79159 net/bnxt: fix queue stop operation + f4d1e64dec net/bnxt: fix multicast MAC restore during reset recovery + 5e35fae222 net/bnxt: fix multicast address set + c59f883c24 net/bnxt: fix xstats names query overrun + 9d1da3652a net/mlx5: relax headroom assertion + ab06c7bf9b net/mlx5: fix GCC uninitialized variable warning + 379079d6cc net/mlx5: fix GRE protocol type translation for Verbs + d8d54171bd net/mlx5: fix RSS expansion with explicit next protocol + 032e27c0b2 net/mlx5: fix assertion on flags set in packet mbuf + 59f8d27b5f common/mlx5: fix missing validation in devargs parsing + 407b3ae746 net/mlx5: fix memory socket selection in ASO management + 751bca90b1 common/mlx5: fix error handling in multi-class probe + 0832935bf5 net/ixgbe: add vector Rx parameter check + 1eef1cf7d3 net/ice: fix Tx checksum offload + 76a729e7ed net/ice: track DCF state of PF + 85e84c5930 net/iavf: remove git residue symbol + 3380c428a5 net/ice: fix Tx checksum offload capability + a9ff22fc59 net/ice: fix pattern check for flow director parser + f1339fd8ec net/qede: fix redundant condition in debug code + 9ff875ecff common/cnxk: fix nibble parsing order when dumping MCAM + 23b8e0a337 net/mlx5: fix MPRQ stride devargs adjustment + c58aaabede net/mlx5: improve stride parameter names + f873364dfe common/mlx5: add minimum WQE size for striding RQ + 0422d79548 net/mlx5: fix modify field MAC address offset + d021a2f9b7 dma/cnxk: fix installing internal headers + bb6b3ec4e9 devtools: fix comment detection in forbidden token check + 19aefaf2cb examples/ipsec-secgw: fix default flow rule creation + 8c4f0e9e73 examples/ipsec-secgw: fix eventdev start sequence + 5831db3bab crypto/ipsec_mb: fix tainted data for session + 4b8475c98e crypto/ipsec_mb: fix queue cleanup null pointer dereference + 07ee507051 crypto/ipsec_mb: fix queue setup null pointer dereference + 7823f35581 crypto/cnxk: fix extend tail calculation + 5977020bc3 crypto/cnxk: fix inflight count calculation + 99d6741b64 crypto/cnxk: enable allocated queues only + de6b483f38 common/cnxk: fix reset of fields + 7ee503d33f common/cnxk: add workaround for vWQE flush + e3b9a8c32f event/cnxk: fix QoS devargs parsing + 87646d04a1 examples/l3fwd: fix Rx burst size for event mode + 8d0ffec0cf eventdev/eth_rx: fix missing internal port checks + 393d0580db doc: fix dlb2 guide + 19c6e95cd6 eal/linux: log hugepage create errors with filename + 24e496918b config: add arch define for Arm + a8dd54379d gpu/cuda: fix memory list cleanup + 8e8fe373c0 dma/idxd: fix wrap-around in burst capacity calculation + e0f7faeba1 dma/idxd: fix paths to driver sysfs directory + cb7d9a39cb dma/idxd: fix burst capacity calculation + 572305874a bus/ifpga: remove useless check while browsing devices + c30f1ec97d doc: remove dependency on findutils on FreeBSD + cb2e09a4da buildtools: fix AVX512 check for Python 3.5 + 3b511fdf21 maintainers: update for stable branches + b3122779a5 doc: replace deprecated distutils version parsing + 2be1e5158e fix spelling in comments and strings + + +* Thu Sep 01 2022 Open vSwitch CI - 2.17.0-40 +- Merging upstream branch-2.17 [RH git: 696a65df68] + Commit list: + dfc3e65c81 raft: Fix unnecessary periodic compactions. + 6f322ccf8a netdev-offload-tc: Parse tunnel options only for geneve ports. + + +* Tue Aug 30 2022 Open vSwitch CI - 2.17.0-39 +- Merging upstream branch-2.17 [RH git: 18a8544580] + Commit list: + a9f10a2bdc netdev-offload-tc: Add missing handling of the tunnel source port. + ec2e967c1d netdev-offload-tc: Fix ignoring unknown tunnel keys. + 686984d9a0 netdev-offload-tc: Use masks instead of keys while parsing tunnel attributes. + 92c072d944 netdev-offload-tc: Explicitly handle mask for the tunnel destination port. + 87f191a3a3 netdev-offload-tc: Fix the mask for tunnel metadata length. + cadcea6fea releases: Mark 2.17 as a new LTS release. + 8a1b734480 handlers: Fix handlers mapping. + 713072fdac handlers: Create additional handler threads when using CPU isolation. + 84a8910ffe packets: Fix misaligned access to ip6_hdr. + fe27e0c884 python: Do not send non-zero flag for a SSL socket. (#2115035) + + +* Wed Aug 10 2022 Open vSwitch CI - 2.17.0-38 +- Merging upstream branch-2.17 [RH git: 3803095f76] + Commit list: + 729a872f19 dpif-netdev: Simplify AVX512 build time checks to enhance readability. + + +* Wed Aug 10 2022 Timothy Redaelli - 2.17.0-37 +- pkgtool: keep %{?dist} before added bz string [RH git: b129ea83d9] + Signed-off-by: Timothy Redaelli + + +* Tue Aug 09 2022 Open vSwitch CI - 2.17.0-36 +- Merging upstream branch-2.17 [RH git: 66778aea49] + Commit list: + 1b566f8b80 github: Move CI to ubuntu 20.04 base image. + + +* Mon Aug 08 2022 Open vSwitch CI - 2.17.0-35 +- Merging upstream branch-2.17 [RH git: 2b96d7c354] + Commit list: + 86725abe10 netdev-offload-tc: Disable offload of IPv6 fragments. + 2276daf889 ovs-save: Use right OpenFlow version for add-tlv-map. + + +* Mon Aug 08 2022 Open vSwitch CI - 2.17.0-34 +- Merging upstream branch-2.17 [RH git: b2f86a115d] + Commit list: + c353e757d7 system-traffic: Fix IPv4 fragmentation test sequence for check-kernel. + 6f54dc134a system-traffic: Fix incorrect neigh entry in ipv6 header modification test. + 7848ae6ffb system-traffic: Don't run IPv6 header modification test on kernels < 5.19. + 399185865e netdev-linux: set correct action for packets that passed policer + + +* Thu Aug 04 2022 Open vSwitch CI - 2.17.0-33 +- Merging upstream branch-2.17 [RH git: 3e07847983] + Commit list: + cda60c8558 python: Fix E275 missing whitespace after keyword. + 3678fb544d tc: Use sparse hex dump while printing inconsistencies. + 03a0ec82b7 netdev-offload-tc: Print unused mask bits on failure. + 5b8453a44e dynamic-string: Add function for a sparse hex dump. + 8d7cb1daf4 dpif-netlink: Fix incorrect bit shift in compat mode. + d1cec26869 python: Use setuptools instead of distutils. + 8d6ecb2592 packets: Re-calculate IPv6 checksum only for first frag upon modify. + 26dbc822d3 test-ovsdb: Fix false-positive leaks from LeakSanitizer. + 6eab10cf2c m4: Update ax_func_posix_memalign to the latest version. + 2f51bfd23b m4: Replace obsolete AC_HELP_STRING with AS_HELP_STRING. + 8ad325aab5 libopenvswitch.pc: Add missing libs for a static build. + b64ff3f480 rhel: Stop installing internal headers. + b63bbf2dba python-c-ext: Handle initialization failures. + 4ad02ad047 netdev-linux: Do not touch LAG members if master is not attached to OVS. + e6dcd07bc2 netdev: Clear auto_classified if netdev reopened with the type specified. + + +* Mon Jul 25 2022 Open vSwitch CI - 2.17.0-32 +- Merging upstream branch-2.17 [RH git: 04dbaa3629] + Commit list: + 1eedf45e81 system-traffic: Properly stop dangling ping after geneve test. + fb8e34bdba conntrack: Fix conntrack multiple new state. + af37f41188 python-c-ext: Fix a couple of build warnings. + b7d9f76100 python-c-ext: Remove Python 2 support. + 02fb4bfb87 netdev-offload-dpdk: Setting RSS hash types in RSS action. + 8e8fcf7bda lib: Print nw_frag in flow key. + 29d8ce1adc ovsdb: Remove extra make target dependency for local-config.5. + + * Thu Jul 14 2022 Open vSwitch CI - 2.17.0-31 - Merging upstream branch-2.17 [RH git: bdc6c6696c] Commit list: