qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: qemu-devel@nongnu.org
Cc: fred.konrad@greensocs.com, qemu-stable@nongnu.org, mst@redhat.com
Subject: [Qemu-devel] [PATCH 11/11] virtio-pci: add device_unplugged callback
Date: Fri, 20 Sep 2013 16:58:00 +0200	[thread overview]
Message-ID: <1379689080-32396-12-git-send-email-pbonzini@redhat.com> (raw)
In-Reply-To: <1379689080-32396-1-git-send-email-pbonzini@redhat.com>

This fixes a crash in hot-unplug of virtio-pci devices behind a PCIe
switch.  The crash happens because the ioeventfd is still set whent the
child is destroyed (destruction happens in postorder).  Then the proxy
tries to unset to ioeventfd, but the virtqueue structure that holds the
EventNotifier has been trashed in the meanwhile.  kvm_set_ioeventfd_pio
does not expect failure and aborts.

The fix is simply to move parts of uninitialization to a new
device_unplugged callback, which is called before the child is destroyed.

Cc: qemu-stable@nongnu.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/virtio/virtio-pci.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/hw/virtio/virtio-pci.c b/hw/virtio/virtio-pci.c
index c8d8b99..1a179ea 100644
--- a/hw/virtio/virtio-pci.c
+++ b/hw/virtio/virtio-pci.c
@@ -1001,6 +1001,15 @@ static void virtio_pci_device_plugged(DeviceState *d)
                                                       proxy->host_features);
 }
 
+static void virtio_pci_device_unplugged(DeviceState *d)
+{
+    PCIDevice *pci_dev = PCI_DEVICE(d);
+    VirtIOPCIProxy *proxy = VIRTIO_PCI(d);
+
+    virtio_pci_stop_ioeventfd(proxy);
+    msix_uninit_exclusive_bar(pci_dev);
+}
+
 static int virtio_pci_init(PCIDevice *pci_dev)
 {
     VirtIOPCIProxy *dev = VIRTIO_PCI(pci_dev);
@@ -1015,9 +1024,7 @@ static int virtio_pci_init(PCIDevice *pci_dev)
 static void virtio_pci_exit(PCIDevice *pci_dev)
 {
     VirtIOPCIProxy *proxy = VIRTIO_PCI(pci_dev);
-    virtio_pci_stop_ioeventfd(proxy);
     memory_region_destroy(&proxy->bar);
-    msix_uninit_exclusive_bar(pci_dev);
 }
 
 static void virtio_pci_reset(DeviceState *qdev)
@@ -1552,6 +1559,7 @@ static void virtio_pci_bus_class_init(ObjectClass *klass, void *data)
     k->set_guest_notifiers = virtio_pci_set_guest_notifiers;
     k->vmstate_change = virtio_pci_vmstate_change;
     k->device_plugged = virtio_pci_device_plugged;
+    k->device_unplugged = virtio_pci_device_unplugged;
 }
 
 static const TypeInfo virtio_pci_bus_info = {
-- 
1.8.3.1

  parent reply	other threads:[~2013-09-20 14:58 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-09-20 14:57 [Qemu-devel] [PATCH 00/11] virtio: cleanup and fix hot-unplug Paolo Bonzini
2013-09-20 14:57 ` [Qemu-devel] [PATCH 01/11] virtio-bus: remove vdev field Paolo Bonzini
2013-10-14 16:23   ` [Qemu-devel] [PATCH 12/11] virtio-pci: avoid extra pointer dereferences on fast path Paolo Bonzini
2013-10-15 12:13     ` Frederic Konrad
2013-10-15 12:16       ` Paolo Bonzini
2013-10-15 16:46   ` [Qemu-devel] [PATCH v3 00/12] virtio: cleanup and fix hot-unplug Paolo Bonzini
2013-10-15 16:46     ` [Qemu-devel] [PATCH v3 01/12] virtio-ccw: move virtio_ccw_stop_ioeventfd to virtio_ccw_busdev_unplug Paolo Bonzini
2013-10-15 16:46     ` [Qemu-devel] [PATCH v3 02/12] virtio-bus: remove vdev field Paolo Bonzini
2013-10-15 16:46     ` [Qemu-devel] [PATCH v3 03/12] virtio-ccw: " Paolo Bonzini
2013-10-15 16:46     ` [Qemu-devel] [PATCH v3 04/12] virtio-pci: " Paolo Bonzini
2013-10-15 16:46     ` [Qemu-devel] [PATCH v3 05/12] virtio-bus: cleanup plug/unplug interface Paolo Bonzini
2013-10-15 16:46     ` [Qemu-devel] [PATCH v3 06/12] virtio-blk: switch exit callback to VirtioDeviceClass Paolo Bonzini
2013-10-15 16:46     ` [Qemu-devel] [PATCH v3 07/12] virtio-serial: " Paolo Bonzini
2013-10-15 16:46     ` [Qemu-devel] [PATCH v3 08/12] virtio-net: " Paolo Bonzini
2013-10-15 16:46     ` [Qemu-devel] [PATCH v3 09/12] virtio-scsi: " Paolo Bonzini
2013-10-15 16:46     ` [Qemu-devel] [PATCH v3 10/12] virtio-balloon: " Paolo Bonzini
2013-10-15 16:46     ` [Qemu-devel] [PATCH v3 11/12] virtio-rng: " Paolo Bonzini
2013-10-15 16:46     ` [Qemu-devel] [PATCH v3 12/12] virtio-pci: add device_unplugged callback Paolo Bonzini
2013-09-20 14:57 ` [Qemu-devel] [PATCH 02/11] virtio-pci: remove vdev field Paolo Bonzini
2013-09-20 14:57 ` [Qemu-devel] [PATCH 03/11] virtio-ccw: " Paolo Bonzini
2013-09-20 14:57 ` [Qemu-devel] [PATCH 04/11] virtio-bus: cleanup plug/unplug interface Paolo Bonzini
2013-09-20 14:57 ` [Qemu-devel] [PATCH 05/11] virtio-blk: switch exit callback to VirtioDeviceClass Paolo Bonzini
2013-09-20 14:57 ` [Qemu-devel] [PATCH 06/11] virtio-serial: " Paolo Bonzini
2013-09-20 14:57 ` [Qemu-devel] [PATCH 07/11] virtio-net: " Paolo Bonzini
2013-09-20 14:57 ` [Qemu-devel] [PATCH 08/11] virtio-scsi: " Paolo Bonzini
2013-09-20 14:57 ` [Qemu-devel] [PATCH 09/11] virtio-balloon: " Paolo Bonzini
2013-09-20 14:57 ` [Qemu-devel] [PATCH 10/11] virtio-rng: " Paolo Bonzini
2013-09-20 14:58 ` Paolo Bonzini [this message]
2013-09-21 19:17 ` [Qemu-devel] [PATCH 00/11] virtio: cleanup and fix hot-unplug Michael S. Tsirkin
2013-09-22  8:08   ` Paolo Bonzini
2013-10-08 15:45     ` Paolo Bonzini
2013-10-08 17:02 ` Paolo Bonzini
2013-10-15 12:32   ` Michael S. Tsirkin

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1379689080-32396-12-git-send-email-pbonzini@redhat.com \
    --to=pbonzini@redhat.com \
    --cc=fred.konrad@greensocs.com \
    --cc=mst@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-stable@nongnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).