thebeanogamer / rpms / qemu-kvm

Forked from rpms/qemu-kvm 5 months ago
Clone
Blob Blame History Raw
From 01ebac2f7d54e80a690c855680d35574d118b072 Mon Sep 17 00:00:00 2001
From: Michael S. Tsirkin <mst@redhat.com>
Date: Wed, 12 Mar 2014 06:52:43 +0100
Subject: [PATCH 06/16] virtio-net: add feature bit for any header s/g

Message-id: <1394169006-8402-1-git-send-email-mst@redhat.com>
Patchwork-id: 58046
O-Subject: [PATCH qemu-kvm RHEL7.0] virtio-net: add feature bit for any header s/g
Bugzilla: 990989
RH-Acked-by: Xiao Wang <jasowang@redhat.com>
RH-Acked-by: Paolo Bonzini <pbonzini@redhat.com>
RH-Acked-by: Laszlo Ersek <lersek@redhat.com>

Old qemu versions required that 1st s/g entry is the header.

Since QEMU 1.5, patchset titled "virtio-net: iovec handling cleanup"
removed this limitation but a feature bit is needed so guests know it's
safe to lay out header differently.

This patch applies on top and adds such a feature bit to QEMU.
It is set by default for virtio-net.
virtio net header inline with the data is beneficial
for latency and small packet bandwidth - guest driver
code utilizing this feature has been acked but missed 3.11
by a narrow margin, it's pending for 3.12.

This feature bit is cleared by default when compatibility with old
machine types is requested.

Other performance-sensitive devices (blk and scsi)
don't yet support arbitrary s/g layouts, so
we only set this bit for virtio-net for now.
There are plans to allow arbitrary layouts there, but
no code has been posted yet.

Cc: Rusty Russell <rusty@rustcorp.com.au>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
(cherry picked from commit 488f069bd1f3aeb6ee748acb02b7581831bcb3f8)

Brew build: http://brewweb.devel.redhat.com/brew/taskinfo?taskID=7155286
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=990989
Tested: on developer's box

Conflicts:
	include/hw/i386/pc.h
---

Oops, the patch got lost somehow.
I was sure our last rebase included it but no ...
Pls review ASAP and we'll discuss setting blocker for this one
separately.

Impossible to fix in zstream since it would affect
migration.

 include/hw/i386/pc.h           | 4 ++++
 include/hw/virtio/virtio-net.h | 1 +
 include/hw/virtio/virtio.h     | 2 ++
 hw/i386/pc_piix.c              | 4 ++++
 4 files changed, 11 insertions(+)

Signed-off-by: Miroslav Rezanina <mrezanin@redhat.com>
---
 hw/i386/pc_piix.c              |    4 ++++
 include/hw/i386/pc.h           |    4 ++++
 include/hw/virtio/virtio-net.h |    1 +
 include/hw/virtio/virtio.h     |    2 ++
 4 files changed, 11 insertions(+), 0 deletions(-)

diff --git a/hw/i386/pc_piix.c b/hw/i386/pc_piix.c
index 697f12f..98a5253 100644
--- a/hw/i386/pc_piix.c
+++ b/hw/i386/pc_piix.c
@@ -904,6 +904,10 @@ static QEMUMachine pc_machine_rhel700 = {
         .driver   = TYPE_USB_DEVICE,\
         .property = "msos-desc",\
         .value    = "no",\
+    },{\
+        .driver   = "virtio-net-pci",\
+        .property = "any_layout",\
+        .value    = "off",\
     }
 
 static void pc_compat_rhel650(QEMUMachineInitArgs *args)
diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h
index f56e9ea..8a69347 100644
--- a/include/hw/i386/pc.h
+++ b/include/hw/i386/pc.h
@@ -274,6 +274,10 @@ int e820_add_entry(uint64_t, uint64_t, uint32_t);
             .driver = TYPE_X86_CPU,\
             .property = "pmu",\
             .value = "on",\
+        },{\
+            .driver   = "virtio-net-pci",\
+            .property = "any_layout",\
+            .value    = "off",\
         }
 
 #define PC_COMPAT_1_4 \
diff --git a/include/hw/virtio/virtio-net.h b/include/hw/virtio/virtio-net.h
index beeead7..75723a8 100644
--- a/include/hw/virtio/virtio-net.h
+++ b/include/hw/virtio/virtio-net.h
@@ -231,6 +231,7 @@ struct virtio_net_ctrl_mq {
 
 #define DEFINE_VIRTIO_NET_FEATURES(_state, _field) \
         DEFINE_VIRTIO_COMMON_FEATURES(_state, _field), \
+        DEFINE_PROP_BIT("any_layout", _state, _field, VIRTIO_F_ANY_LAYOUT, true), \
         DEFINE_PROP_BIT("csum", _state, _field, VIRTIO_NET_F_CSUM, true), \
         DEFINE_PROP_BIT("guest_csum", _state, _field, VIRTIO_NET_F_GUEST_CSUM, true), \
         DEFINE_PROP_BIT("gso", _state, _field, VIRTIO_NET_F_GSO, true), \
diff --git a/include/hw/virtio/virtio.h b/include/hw/virtio/virtio.h
index cd886c3..9e22865 100644
--- a/include/hw/virtio/virtio.h
+++ b/include/hw/virtio/virtio.h
@@ -43,6 +43,8 @@
 /* We notify when the ring is completely used, even if the guest is suppressing
  * callbacks */
 #define VIRTIO_F_NOTIFY_ON_EMPTY        24
+/* Can the device handle any descriptor layout? */
+#define VIRTIO_F_ANY_LAYOUT             27
 /* We support indirect buffer descriptors */
 #define VIRTIO_RING_F_INDIRECT_DESC     28
 /* The Guest publishes the used index for which it expects an interrupt
-- 
1.7.1