From a9e323776f72ea5982334690bdbc7e5a648b589d Mon Sep 17 00:00:00 2001 Message-Id: From: "Michael S. Tsirkin" Date: Mon, 23 Apr 2012 13:58:49 +0200 Subject: [PATCH] virtio: add missing mb() on enable notification RH-Author: Michael S. Tsirkin Message-id: Patchwork-id: 39378 O-Subject: [PATCH RHEL6.3 1/2] virtio: add missing mb() on enable notification Bugzilla: 804578 RH-Acked-by: Xiao Wang RH-Acked-by: Amos Kong RH-Acked-by: Paolo Bonzini This fixes an issue dual to the one fixed by patch 'virtio: add missing mb() on notification' and applies on top. In this case, to enable vq kick to exit to host, qemu writes out used flag then reads the avail index. if these are reordered we get a race: host avail index read: ring is empty guest avail index write guest flag read: exit disabled host used flag write: enable exit which results in a lost exit: host will never be notified about the avail index update. Again, happens in the field but only seems to trigger on some specific hardware. Insert an smp_mb barrier operation to ensure the correct ordering. Signed-off-by: Michael S. Tsirkin Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=804578 Brew build: http://brewweb.devel.redhat.com/brew/taskinfo?taskID=4327871 Tested: on x86_64 with virtio-net Upstream status: posted --- hw/virtio.c | 4 ++++ 1 files changed, 4 insertions(+), 0 deletions(-) Signed-off-by: Michal Novotny --- hw/virtio.c | 4 ++++ 1 files changed, 4 insertions(+), 0 deletions(-) diff --git a/hw/virtio.c b/hw/virtio.c index 2db9b68..08e64d9 100644 --- a/hw/virtio.c +++ b/hw/virtio.c @@ -218,6 +218,10 @@ void virtio_queue_set_notification(VirtQueue *vq, int enable) } else { vring_used_flags_set_bit(vq, VRING_USED_F_NO_NOTIFY); } + if (enable) { + /* We need to expose avail event/flags before checking avail idx. */ + mb(); + } } int virtio_queue_ready(VirtQueue *vq) -- 1.7.7.6