Blob Blame History Raw
From 8fe39316da28bdff610da708148f87b21e7ba045 Mon Sep 17 00:00:00 2001
From: Amit Shah <amit.shah@redhat.com>
Date: Tue, 6 Apr 2010 23:36:50 +0530
Subject: [PATCH] virtio-console patches

Hey Justin,

Attached are the kernel and qemu patches for the new abi. Rusty is ok
with the kernel patches and Juan and Gerd are ok with the userspace
ones.

The kernel one is big because of some code movement.

The qemu one is big because of code being moved and some fixes being
applied to make the host less vulnerable to malicious guests.

There's also a patch at

http://lkml.org/lkml/2010/4/6/110

that we should apply (affects the virt-console-fix-race.patch in the
repo).

Please let me know if you need any more information!

Thanks,
		Amit

Content-Disposition: attachment; filename=",qemu-virtio-serial-rollup-2.patch"
---
 Makefile.hw            |    1 +
 hw/iov.c               |   70 ++++++++++
 hw/iov.h               |   19 +++
 hw/virtio-balloon.c    |   31 +----
 hw/virtio-console.c    |   11 +-
 hw/virtio-net.c        |   20 +---
 hw/virtio-serial-bus.c |  332 ++++++++++++++++++++++++++++++++++++------------
 hw/virtio-serial.h     |   34 ++++--
 8 files changed, 372 insertions(+), 146 deletions(-)
 create mode 100644 hw/iov.c
 create mode 100644 hw/iov.h

diff --git a/Makefile.hw b/Makefile.hw
index 43ca541..079c5d2 100644
--- a/Makefile.hw
+++ b/Makefile.hw
@@ -13,6 +13,7 @@ QEMU_CFLAGS+=-I.. -I$(SRC_PATH)/fpu

 obj-y =
 obj-y += loader.o
+obj-y += iov.o
 obj-y += virtio.o virtio-console.o
 obj-y += fw_cfg.o
 obj-y += watchdog.o
diff --git a/hw/iov.c b/hw/iov.c
new file mode 100644
index 0000000..588cd04
--- /dev/null
+++ b/hw/iov.c
@@ -0,0 +1,70 @@
+/*
+ * Helpers for getting linearized buffers from iov / filling buffers into iovs
+ *
+ * Copyright IBM, Corp. 2007, 2008
+ * Copyright (C) 2010 Red Hat, Inc.
+ *
+ * Author(s):
+ *  Anthony Liguori <aliguori@us.ibm.com>
+ *  Amit Shah <amit.shah@redhat.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.  See
+ * the COPYING file in the top-level directory.
+ */
+
+#include "iov.h"
+
+size_t iov_from_buf(struct iovec *iov, unsigned int iovcnt,
+                    const void *buf, size_t size)
+{
+    size_t offset;
+    unsigned int i;
+
+    offset = 0;
+    for (i = 0; offset < size && i < iovcnt; i++) {
+        size_t len;
+
+        len = MIN(iov[i].iov_len, size - offset);
+
+        memcpy(iov[i].iov_base, buf + offset, len);
+        offset += len;
+    }
+    return offset;
+}
+
+size_t iov_to_buf(const struct iovec *iov, const unsigned int iovcnt,
+                  void *buf, size_t offset, size_t size)
+{
+    uint8_t *ptr;
+    size_t iov_off, buf_off;
+    unsigned int i;
+
+    ptr = buf;
+    iov_off = 0;
+    buf_off = 0;
+    for (i = 0; i < iovcnt && size; i++) {
+        if (offset < (iov_off + iov[i].iov_len)) {
+            size_t len = MIN((iov_off + iov[i].iov_len) - offset , size);
+
+            memcpy(ptr + buf_off, iov[i].iov_base + (offset - iov_off), len);
+
+            buf_off += len;
+            offset += len;
+            size -= len;
+        }
+        iov_off += iov[i].iov_len;
+    }
+    return buf_off;
+}
+
+size_t iov_size(const struct iovec *iov, const unsigned int iovcnt)
+{
+    size_t len;
+    unsigned int i;
+
+    len = 0;
+    for (i = 0; i < iovcnt; i++) {
+        len += iov[i].iov_len;
+    }
+    return len;
+}
diff --git a/hw/iov.h b/hw/iov.h
new file mode 100644
index 0000000..60a8547
--- /dev/null
+++ b/hw/iov.h
@@ -0,0 +1,19 @@
+/*
+ * Helpers for getting linearized buffers from iov / filling buffers into iovs
+ *
+ * Copyright (C) 2010 Red Hat, Inc.
+ *
+ * Author(s):
+ *  Amit Shah <amit.shah@redhat.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.  See
+ * the COPYING file in the top-level directory.
+ */
+
+#include "qemu-common.h"
+
+size_t iov_from_buf(struct iovec *iov, unsigned int iovcnt,
+                    const void *buf, size_t size);
+size_t iov_to_buf(const struct iovec *iov, const unsigned int iovcnt,
+                  void *buf, size_t offset, size_t size);
+size_t iov_size(const struct iovec *iov, const unsigned int iovcnt);
diff --git a/hw/virtio-balloon.c b/hw/virtio-balloon.c
index 242c6c8..10e2647 100644
--- a/hw/virtio-balloon.c
+++ b/hw/virtio-balloon.c
@@ -11,6 +11,7 @@
  *
  */

+#include "iov.h"
 #include "qemu-common.h"
 #include "virtio.h"
 #include "pc.h"
@@ -47,33 +48,6 @@ static void balloon_page(void *addr, int deflate)
 #endif
 }

-/* FIXME: once we do a virtio refactoring, this will get subsumed into common
- * code */
-static size_t memcpy_from_iovector(void *data, size_t offset, size_t size,
-                                   struct iovec *iov, int iovlen)
-{
-    int i;
-    uint8_t *ptr = data;
-    size_t iov_off = 0;
-    size_t data_off = 0;
-
-    for (i = 0; i < iovlen && size; i++) {
-        if (offset < (iov_off + iov[i].iov_len)) {
-            size_t len = MIN((iov_off + iov[i].iov_len) - offset , size);
-
-            memcpy(ptr + data_off, iov[i].iov_base + (offset - iov_off), len);
-
-            data_off += len;
-            offset += len;
-            size -= len;
-        }
-
-        iov_off += iov[i].iov_len;
-    }
-
-    return data_off;
-}
-
 static void virtio_balloon_handle_output(VirtIODevice *vdev, VirtQueue *vq)
 {
     VirtIOBalloon *s = to_virtio_balloon(vdev);
@@ -83,8 +57,7 @@ static void virtio_balloon_handle_output(VirtIODevice *vdev, VirtQueue *vq)
         size_t offset = 0;
         uint32_t pfn;

-        while (memcpy_from_iovector(&pfn, offset, 4,
-                                    elem.out_sg, elem.out_num) == 4) {
+        while (iov_to_buf(elem.out_sg, elem.out_num, &pfn, offset, 4) == 4) {
             ram_addr_t pa;
             ram_addr_t addr;

diff --git a/hw/virtio-console.c b/hw/virtio-console.c
index bd44ec6..caea11f 100644
--- a/hw/virtio-console.c
+++ b/hw/virtio-console.c
@@ -1,7 +1,7 @@
 /*
  * Virtio Console and Generic Serial Port Devices
  *
- * Copyright Red Hat, Inc. 2009
+ * Copyright Red Hat, Inc. 2009, 2010
  *
  * Authors:
  *  Amit Shah <amit.shah@redhat.com>
@@ -20,14 +20,11 @@ typedef struct VirtConsole {


 /* Callback function that's called when the guest sends us data */
-static size_t flush_buf(VirtIOSerialPort *port, const uint8_t *buf, size_t len)
+static void flush_buf(VirtIOSerialPort *port, const uint8_t *buf, size_t len)
 {
     VirtConsole *vcon = DO_UPCAST(VirtConsole, port, port);
-    ssize_t ret;

-    ret = qemu_chr_write(vcon->chr, buf, len);
-
-    return ret < 0 ? 0 : ret;
+    qemu_chr_write(vcon->chr, buf, len);
 }

 /* Readiness of the guest to accept data on a port */
@@ -99,6 +96,7 @@ static VirtIOSerialPortInfo virtconsole_info = {
     .exit          = virtconsole_exitfn,
     .qdev.props = (Property[]) {
         DEFINE_PROP_UINT8("is_console", VirtConsole, port.is_console, 1),
+        DEFINE_PROP_UINT32("nr", VirtConsole, port.id, VIRTIO_CONSOLE_BAD_ID),
         DEFINE_PROP_CHR("chardev", VirtConsole, chr),
         DEFINE_PROP_STRING("name", VirtConsole, port.name),
         DEFINE_PROP_END_OF_LIST(),
@@ -133,6 +131,7 @@ static VirtIOSerialPortInfo virtserialport_info = {
     .init          = virtserialport_initfn,
     .exit          = virtconsole_exitfn,
     .qdev.props = (Property[]) {
+        DEFINE_PROP_UINT32("nr", VirtConsole, port.id, VIRTIO_CONSOLE_BAD_ID),
         DEFINE_PROP_CHR("chardev", VirtConsole, chr),
         DEFINE_PROP_STRING("name", VirtConsole, port.name),
         DEFINE_PROP_END_OF_LIST(),
diff --git a/hw/virtio-net.c b/hw/virtio-net.c
index f8e228f..320e99f 100644
--- a/hw/virtio-net.c
+++ b/hw/virtio-net.c
@@ -11,6 +11,7 @@
  *
  */

+#include "iov.h"
 #include "virtio.h"
 #include "net.h"
 #include "net/checksum.h"
@@ -436,21 +437,6 @@ static void work_around_broken_dhclient(struct virtio_net_hdr *hdr,
     }
 }

-static int iov_fill(struct iovec *iov, int iovcnt, const void *buf, int count)
-{
-    int offset, i;
-
-    offset = i = 0;
-    while (offset < count && i < iovcnt) {
-        int len = MIN(iov[i].iov_len, count - offset);
-        memcpy(iov[i].iov_base, buf + offset, len);
-        offset += len;
-        i++;
-    }
-
-    return offset;
-}
-
 static int receive_header(VirtIONet *n, struct iovec *iov, int iovcnt,
                           const void *buf, size_t size, size_t hdr_len)
 {
@@ -586,8 +572,8 @@ static ssize_t virtio_net_receive(VLANClientState *nc, const uint8_t *buf, size_
         }

         /* copy in packet.  ugh */
-        len = iov_fill(sg, elem.in_num,
-                       buf + offset, size - offset);
+        len = iov_from_buf(sg, elem.in_num,
+                           buf + offset, size - offset);
         total += len;

         /* signal other side */
diff --git a/hw/virtio-serial-bus.c b/hw/virtio-serial-bus.c
index d0e0219..6245f6e 100644
--- a/hw/virtio-serial-bus.c
+++ b/hw/virtio-serial-bus.c
@@ -1,7 +1,7 @@
 /*
  * A bus for connecting virtio serial and console ports
  *
- * Copyright (C) 2009 Red Hat, Inc.
+ * Copyright (C) 2009, 2010 Red Hat, Inc.
  *
  * Author(s):
  *  Amit Shah <amit.shah@redhat.com>
@@ -15,6 +15,7 @@
  * the COPYING file in the top-level directory.
  */

+#include "iov.h"
 #include "monitor.h"
 #include "qemu-queue.h"
 #include "sysbus.h"
@@ -41,6 +42,10 @@ struct VirtIOSerial {
     VirtIOSerialBus *bus;

     QTAILQ_HEAD(, VirtIOSerialPort) ports;
+
+    /* bitmap for identifying active ports */
+    uint32_t *ports_map;
+
     struct virtio_console_config config;
 };

@@ -48,6 +53,10 @@ static VirtIOSerialPort *find_port_by_id(VirtIOSerial *vser, uint32_t id)
 {
     VirtIOSerialPort *port;

+    if (id == VIRTIO_CONSOLE_BAD_ID) {
+        return NULL;
+    }
+
     QTAILQ_FOREACH(port, &vser->ports, next) {
         if (port->id == id)
             return port;
@@ -76,30 +85,25 @@ static size_t write_to_port(VirtIOSerialPort *port,
 {
     VirtQueueElement elem;
     VirtQueue *vq;
-    size_t offset = 0;
-    size_t len = 0;
+    size_t offset;

     vq = port->ivq;
     if (!virtio_queue_ready(vq)) {
         return 0;
     }
-    if (!size) {
-        return 0;
-    }

+    offset = 0;
     while (offset < size) {
-        int i;
+        size_t len;

         if (!virtqueue_pop(vq, &elem)) {
             break;
         }

-        for (i = 0; offset < size && i < elem.in_num; i++) {
-            len = MIN(elem.in_sg[i].iov_len, size - offset);
+        len = iov_from_buf(elem.in_sg, elem.in_num,
+                           buf + offset, size - offset);
+        offset += len;

-            memcpy(elem.in_sg[i].iov_base, buf + offset, len);
-            offset += len;
-        }
         virtqueue_push(vq, &elem, len);
     }

@@ -107,6 +111,29 @@ static size_t write_to_port(VirtIOSerialPort *port,
     return offset;
 }

+static void flush_queued_data(VirtIOSerialPort *port, bool discard)
+{
+    VirtQueue *vq;
+    VirtQueueElement elem;
+
+    vq = port->ovq;
+    while (virtqueue_pop(vq, &elem)) {
+        uint8_t *buf;
+        size_t ret, buf_size;
+
+        if (!discard) {
+            buf_size = iov_size(elem.out_sg, elem.out_num);
+            buf = qemu_malloc(buf_size);
+            ret = iov_to_buf(elem.out_sg, elem.out_num, buf, 0, buf_size);
+
+            port->info->have_data(port, buf, ret);
+            qemu_free(buf);
+        }
+        virtqueue_push(vq, &elem, 0);
+    }
+    virtio_notify(&port->vser->vdev, vq);
+}
+
 static size_t send_control_msg(VirtIOSerialPort *port, void *buf, size_t len)
 {
     VirtQueueElement elem;
@@ -158,6 +185,13 @@ int virtio_serial_open(VirtIOSerialPort *port)
 int virtio_serial_close(VirtIOSerialPort *port)
 {
     port->host_connected = false;
+    /*
+     * If there's any data the guest sent which the app didn't
+     * consume, discard it and reset the throttling flag.
+     */
+    flush_queued_data(port, true);
+    port->throttled = false;
+
     send_control_event(port, VIRTIO_CONSOLE_PORT_OPEN, 0);

     return 0;
@@ -199,8 +233,23 @@ size_t virtio_serial_guest_ready(VirtIOSerialPort *port)
     return 0;
 }

+void virtio_serial_throttle_port(VirtIOSerialPort *port, bool throttle)
+{
+    if (!port) {
+        return;
+    }
+
+    if (throttle) {
+        port->throttled = true;
+        return;
+    }
+
+    port->throttled = false;
+    flush_queued_data(port, false);
+}
+
 /* Guest wants to notify us of some event */
-static void handle_control_message(VirtIOSerial *vser, void *buf)
+static void handle_control_message(VirtIOSerial *vser, void *buf, size_t len)
 {
     struct VirtIOSerialPort *port;
     struct virtio_console_control cpkt, *gcpkt;
@@ -208,15 +257,41 @@ static void handle_control_message(VirtIOSerial *vser, void *buf)
     size_t buffer_len;

     gcpkt = buf;
-    port = find_port_by_id(vser, ldl_p(&gcpkt->id));
-    if (!port)
+
+    if (len < sizeof(cpkt)) {
+        /* The guest sent an invalid control packet */
         return;
+    }

     cpkt.event = lduw_p(&gcpkt->event);
     cpkt.value = lduw_p(&gcpkt->value);

+    port = find_port_by_id(vser, ldl_p(&gcpkt->id));
+    if (!port && cpkt.event != VIRTIO_CONSOLE_DEVICE_READY)
+        return;
+
     switch(cpkt.event) {
+    case VIRTIO_CONSOLE_DEVICE_READY:
+        if (!cpkt.value) {
+            qemu_error("virtio-serial-bus: Guest failure in adding device %s\n",
+                         vser->bus->qbus.name);
+            break;
+        }
+        /*
+         * The device is up, we can now tell the device about all the
+         * ports we have here.
+         */
+        QTAILQ_FOREACH(port, &vser->ports, next) {
+            send_control_event(port, VIRTIO_CONSOLE_PORT_ADD, 1);
+        }
+        break;
+
     case VIRTIO_CONSOLE_PORT_READY:
+        if (!cpkt.value) {
+            qemu_error("virtio-serial-bus: Guest failure in adding port %u for device %s\n",
+                         port->id, vser->bus->qbus.name);
+            break;
+        }
         /*
          * Now that we know the guest asked for the port name, we're
          * sure the guest has initialised whatever state is necessary
@@ -281,12 +356,35 @@ static void control_out(VirtIODevice *vdev, VirtQueue *vq)
 {
     VirtQueueElement elem;
     VirtIOSerial *vser;
+    uint8_t *buf;
+    size_t len;

     vser = DO_UPCAST(VirtIOSerial, vdev, vdev);

+    len = 0;
+    buf = NULL;
     while (virtqueue_pop(vq, &elem)) {
-        handle_control_message(vser, elem.out_sg[0].iov_base);
-        virtqueue_push(vq, &elem, elem.out_sg[0].iov_len);
+        size_t cur_len, copied;
+
+        cur_len = iov_size(elem.out_sg, elem.out_num);
+        /*
+         * Allocate a new buf only if we didn't have one previously or
+         * if the size of the buf differs
+         */
+        if (cur_len > len) {
+            if (len) {
+                qemu_free(buf);
+            }
+            buf = qemu_malloc(cur_len);
+            len = cur_len;
+        }
+        copied = iov_to_buf(elem.out_sg, elem.out_num, buf, 0, len);
+
+        handle_control_message(vser, buf, copied);
+        virtqueue_push(vq, &elem, 0);
+    }
+    if (len) {
+        qemu_free(buf);
     }
     virtio_notify(vdev, vq);
 }
@@ -295,38 +393,22 @@ static void control_out(VirtIODevice *vdev, VirtQueue *vq)
 static void handle_output(VirtIODevice *vdev, VirtQueue *vq)
 {
     VirtIOSerial *vser;
-    VirtQueueElement elem;
+    VirtIOSerialPort *port;
+    bool discard;

     vser = DO_UPCAST(VirtIOSerial, vdev, vdev);
+    port = find_port_by_vq(vser, vq);

-    while (virtqueue_pop(vq, &elem)) {
-        VirtIOSerialPort *port;
-        size_t ret;
-
-        port = find_port_by_vq(vser, vq);
-        if (!port) {
-            ret = 0;
-            goto next_buf;
-        }
-
-        /*
-         * A port may not have any handler registered for consuming the
-         * data that the guest sends or it may not have a chardev associated
-         * with it. Just ignore the data in that case.
-         */
-        if (!port->info->have_data) {
-            ret = 0;
-            goto next_buf;
-        }
-
-        /* The guest always sends only one sg */
-        ret = port->info->have_data(port, elem.out_sg[0].iov_base,
-                                    elem.out_sg[0].iov_len);
+    discard = false;
+    if (!port || !port->host_connected || !port->info->have_data) {
+        discard = true;
+    }

-    next_buf:
-        virtqueue_push(vq, &elem, ret);
+    if (!discard && port->throttled) {
+        return;
     }
-    virtio_notify(vdev, vq);
+
+    flush_queued_data(port, discard);
 }

 static void handle_input(VirtIODevice *vdev, VirtQueue *vq)
@@ -335,7 +417,10 @@ static void handle_input(VirtIODevice *vdev, VirtQueue *vq)

 static uint32_t get_features(VirtIODevice *vdev, uint32_t features)
 {
-    VirtIOSerial *vser = DO_UPCAST(VirtIOSerial, vdev, vdev);
+    VirtIOSerial *vser;
+
+    vser = DO_UPCAST(VirtIOSerial, vdev, vdev);
+
     if (vser->bus->max_nr_ports > 1) {
         features |= (1 << VIRTIO_CONSOLE_F_MULTIPORT);
     }
@@ -370,14 +455,20 @@ static void virtio_serial_save(QEMUFile *f, void *opaque)
     /* The config space */
     qemu_put_be16s(f, &s->config.cols);
     qemu_put_be16s(f, &s->config.rows);
-    qemu_put_be32s(f, &s->config.nr_ports);

-    /* Items in struct VirtIOSerial */
+    qemu_put_be32s(f, &s->config.max_nr_ports);
+
+    /* The ports map */
+
+    qemu_put_buffer(f, (uint8_t *)s->ports_map,
+                    sizeof(uint32_t) * (s->config.max_nr_ports + 31) / 32);
+
+    /* Ports */

-    /* Do this because we might have hot-unplugged some ports */
     nr_active_ports = 0;
-    QTAILQ_FOREACH(port, &s->ports, next)
+    QTAILQ_FOREACH(port, &s->ports, next) {
         nr_active_ports++;
+    }

     qemu_put_be32s(f, &nr_active_ports);

@@ -385,13 +476,9 @@ static void virtio_serial_save(QEMUFile *f, void *opaque)
      * Items in struct VirtIOSerialPort.
      */
     QTAILQ_FOREACH(port, &s->ports, next) {
-        /*
-         * We put the port number because we may not have an active
-         * port at id 0 that's reserved for a console port, or in case
-         * of ports that might have gotten unplugged
-         */
         qemu_put_be32s(f, &port->id);
         qemu_put_byte(f, port->guest_connected);
+        qemu_put_byte(f, port->host_connected);
     }
 }

@@ -399,7 +486,8 @@ static int virtio_serial_load(QEMUFile *f, void *opaque, int version_id)
 {
     VirtIOSerial *s = opaque;
     VirtIOSerialPort *port;
-    uint32_t nr_active_ports;
+    size_t ports_map_size;
+    uint32_t max_nr_ports, nr_active_ports, *ports_map;
     unsigned int i;

     if (version_id > 2) {
@@ -416,22 +504,50 @@ static int virtio_serial_load(QEMUFile *f, void *opaque, int version_id)
     /* The config space */
     qemu_get_be16s(f, &s->config.cols);
     qemu_get_be16s(f, &s->config.rows);
-    s->config.nr_ports = qemu_get_be32(f);

-    /* Items in struct VirtIOSerial */
+    qemu_get_be32s(f, &max_nr_ports);
+    if (max_nr_ports > s->config.max_nr_ports) {
+        /* Source could have had more ports than us. Fail migration. */
+        return -EINVAL;
+    }
+
+    ports_map_size = sizeof(uint32_t) * (max_nr_ports + 31) / 32;
+    ports_map = qemu_malloc(ports_map_size);
+    qemu_get_buffer(f, (uint8_t *)ports_map, ports_map_size);
+
+    for (i = 0; i < (max_nr_ports + 31) / 32; i++) {
+        if (ports_map[i] != s->ports_map[i]) {
+            /*
+             * Ports active on source and destination don't
+             * match. Fail migration.
+             */
+            qemu_free(ports_map);
+            return -EINVAL;
+        }
+    }
+    qemu_free(ports_map);

     qemu_get_be32s(f, &nr_active_ports);

     /* Items in struct VirtIOSerialPort */
     for (i = 0; i < nr_active_ports; i++) {
         uint32_t id;
+        bool host_connected;

         id = qemu_get_be32(f);
         port = find_port_by_id(s, id);

         port->guest_connected = qemu_get_byte(f);
+        host_connected = qemu_get_byte(f);
+        if (host_connected != port->host_connected) {
+            /*
+             * We have to let the guest know of the host connection
+             * status change
+             */
+            send_control_event(port, VIRTIO_CONSOLE_PORT_OPEN,
+                               port->host_connected);
+        }
     }
-
     return 0;
 }

@@ -466,6 +582,54 @@ static void virtser_bus_dev_print(Monitor *mon, DeviceState *qdev, int indent)
                    indent, "", port->host_connected);
 }

+/* This function is only used if a port id is not provided by the user */
+static uint32_t find_free_port_id(VirtIOSerial *vser)
+{
+    unsigned int i;
+
+    for (i = 0; i < (vser->config.max_nr_ports + 31) / 32; i++) {
+        uint32_t map, bit;
+
+        map = vser->ports_map[i];
+        bit = ffs(~map);
+        if (bit) {
+            return (bit - 1) + i * 32;
+        }
+    }
+    return VIRTIO_CONSOLE_BAD_ID;
+}
+
+static void mark_port_added(VirtIOSerial *vser, uint32_t port_id)
+{
+    unsigned int i;
+
+    i = port_id / 32;
+    vser->ports_map[i] |= 1U << (port_id % 32);
+}
+
+static void add_port(VirtIOSerial *vser, uint32_t port_id)
+{
+    mark_port_added(vser, port_id);
+
+    send_control_event(find_port_by_id(vser, port_id),
+                       VIRTIO_CONSOLE_PORT_ADD, 1);
+}
+
+static void remove_port(VirtIOSerial *vser, uint32_t port_id)
+{
+    VirtIOSerialPort *port;
+    unsigned int i;
+
+    i = port_id / 32;
+    vser->ports_map[i] &= ~(1U << (port_id % 32));
+
+    port = find_port_by_id(vser, port_id);
+    /* Flush out any unconsumed buffers first */
+    flush_queued_data(port, true);
+
+    send_control_event(port, VIRTIO_CONSOLE_PORT_REMOVE, 1);
+}
+
 static int virtser_port_qdev_init(DeviceState *qdev, DeviceInfo *base)
 {
     VirtIOSerialDevice *dev = DO_UPCAST(VirtIOSerialDevice, qdev, qdev);
@@ -484,19 +648,36 @@ static int virtser_port_qdev_init(DeviceState *qdev, DeviceInfo *base)
      */
     plugging_port0 = port->is_console && !find_port_by_id(port->vser, 0);

-    if (port->vser->config.nr_ports == bus->max_nr_ports && !plugging_port0) {
-        qemu_error("virtio-serial-bus: Maximum device limit reached\n");
+    if (find_port_by_id(port->vser, port->id)) {
+        qemu_error("virtio-serial-bus: A port already exists at id %u\n",
+                     port->id);
         return -1;
     }
-    dev->info = info;

+    if (port->id == VIRTIO_CONSOLE_BAD_ID) {
+        if (plugging_port0) {
+            port->id = 0;
+        } else {
+            port->id = find_free_port_id(port->vser);
+            if (port->id == VIRTIO_CONSOLE_BAD_ID) {
+                qemu_error("virtio-serial-bus: Maximum port limit for this device reached\n");
+                return -1;
+            }
+        }
+    }
+
+    if (port->id >= port->vser->config.max_nr_ports) {
+        qemu_error("virtio-serial-bus: Out-of-range port id specified, max. allowed: %u\n",
+                     port->vser->config.max_nr_ports - 1);
+        return -1;
+    }
+
+    dev->info = info;
     ret = info->init(dev);
     if (ret) {
         return ret;
     }

-    port->id = plugging_port0 ? 0 : port->vser->config.nr_ports++;
-
     if (!use_multiport(port->vser)) {
         /*
          * Allow writes to guest in this case; we have no way of
@@ -509,6 +690,8 @@ static int virtser_port_qdev_init(DeviceState *qdev, DeviceInfo *base)
     port->ivq = port->vser->ivqs[port->id];
     port->ovq = port->vser->ovqs[port->id];

+    add_port(port->vser, port->id);
+
     /* Send an update to the guest about this new port added */
     virtio_notify_config(&port->vser->vdev);

@@ -521,26 +704,8 @@ static int virtser_port_qdev_exit(DeviceState *qdev)
     VirtIOSerialPort *port = DO_UPCAST(VirtIOSerialPort, dev, &dev->qdev);
     VirtIOSerial *vser = port->vser;

-    send_control_event(port, VIRTIO_CONSOLE_PORT_REMOVE, 1);
+    remove_port(port->vser, port->id);

-    /*
-     * Don't decrement nr_ports here; thus we keep a linearly
-     * increasing port id. Not utilising an id again saves us a couple
-     * of complications:
-     *
-     * - Not having to bother about sending the port id to the guest
-     *   kernel on hotplug or on addition of new ports; the guest can
-     *   also linearly increment the port number. This is preferable
-     *   because the config space won't have the need to store a
-     *   ports_map.
-     *
-     * - Extra state to be stored for all the "holes" that got created
-     *   so that we keep filling in the ids from the least available
-     *   index.
-     *
-     * When such a functionality is desired, a control message to add
-     * a port can be introduced.
-     */
     QTAILQ_REMOVE(&vser->ports, port, next);

     if (port->info->exit)
@@ -600,11 +765,12 @@ VirtIODevice *virtio_serial_init(DeviceState *dev, uint32_t max_nr_ports)
     }

     vser->config.max_nr_ports = max_nr_ports;
+    vser->ports_map = qemu_mallocz((max_nr_ports + 31) / 32);
     /*
      * Reserve location 0 for a console port for backward compat
      * (old kernel, new qemu)
      */
-    vser->config.nr_ports = 1;
+    mark_port_added(vser, 0);

     vser->vdev.get_features = get_features;
     vser->vdev.get_config = get_config;
diff --git a/hw/virtio-serial.h b/hw/virtio-serial.h
index f297b00..a93b545 100644
--- a/hw/virtio-serial.h
+++ b/hw/virtio-serial.h
@@ -2,7 +2,7 @@
  * Virtio Serial / Console Support
  *
  * Copyright IBM, Corp. 2008
- * Copyright Red Hat, Inc. 2009
+ * Copyright Red Hat, Inc. 2009, 2010
  *
  * Authors:
  *  Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com>
@@ -27,6 +27,8 @@
 /* Features supported */
 #define VIRTIO_CONSOLE_F_MULTIPORT	1

+#define VIRTIO_CONSOLE_BAD_ID           (~(uint32_t)0)
+
 struct virtio_console_config {
     /*
      * These two fields are used by VIRTIO_CONSOLE_F_SIZE which
@@ -36,7 +38,6 @@ struct virtio_console_config {
     uint16_t rows;

     uint32_t max_nr_ports;
-    uint32_t nr_ports;
 } __attribute__((packed));

 struct virtio_console_control {
@@ -46,12 +47,14 @@ struct virtio_console_control {
 };

 /* Some events for the internal messages (control packets) */
-#define VIRTIO_CONSOLE_PORT_READY	0
-#define VIRTIO_CONSOLE_CONSOLE_PORT	1
-#define VIRTIO_CONSOLE_RESIZE		2
-#define VIRTIO_CONSOLE_PORT_OPEN	3
-#define VIRTIO_CONSOLE_PORT_NAME	4
-#define VIRTIO_CONSOLE_PORT_REMOVE	5
+#define VIRTIO_CONSOLE_DEVICE_READY	0
+#define VIRTIO_CONSOLE_PORT_ADD		1
+#define VIRTIO_CONSOLE_PORT_REMOVE	2
+#define VIRTIO_CONSOLE_PORT_READY	3
+#define VIRTIO_CONSOLE_CONSOLE_PORT	4
+#define VIRTIO_CONSOLE_RESIZE		5
+#define VIRTIO_CONSOLE_PORT_OPEN	6
+#define VIRTIO_CONSOLE_PORT_NAME	7

 /* == In-qemu interface == */

@@ -107,6 +110,8 @@ struct VirtIOSerialPort {
     bool guest_connected;
     /* Is this device open for IO on the host? */
     bool host_connected;
+    /* Do apps not want to receive data? */
+    bool throttled;
 };

 struct VirtIOSerialPortInfo {
@@ -133,10 +138,10 @@ struct VirtIOSerialPortInfo {

     /*
      * Guest wrote some data to the port. This data is handed over to
-     * the app via this callback. The app should return the number of
-     * bytes it successfully consumed.
+     * the app via this callback.  The app is supposed to consume all
+     * the data that is presented to it.
      */
-    size_t (*have_data)(VirtIOSerialPort *port, const uint8_t *buf, size_t len);
+    void (*have_data)(VirtIOSerialPort *port, const uint8_t *buf, size_t len);
 };

 /* Interface to the virtio-serial bus */
@@ -170,4 +175,11 @@ ssize_t virtio_serial_write(VirtIOSerialPort *port, const uint8_t *buf,
  */
 size_t virtio_serial_guest_ready(VirtIOSerialPort *port);

+/*
+ * Flow control: Ports can signal to the virtio-serial core to stop
+ * sending data or re-start sending data, depending on the 'throttle'
+ * value here.
+ */
+void virtio_serial_throttle_port(VirtIOSerialPort *port, bool throttle);
+
 #endif
-- 
1.6.6.1