@@ -35,7 +35,7 @@#define virtio_rmb(vq) \do{if((vq)->weak_barriers)smp_rmb();elsermb();}while(0)#define virtio_wmb(vq) \-do{if((vq)->weak_barriers)smp_rmb();elsermb();}while(0)+do{if((vq)->weak_barriers)smp_wmb();elsewmb();}while(0)#else/* We must force memory ordering even if guest is UP since host could be*runningonanotherCPU,butSMPbarriersaredefinedtobarrier()inthat
From: Jason Wang <hidden> Date: 2012-01-20 08:21:18
Use virtio_mb() to make sure the available index to be exposed before
checking the the avail event. Otherwise we may get stale value of
avail event in guest and never kick the host after.
Signed-off-by: Jason Wang <redacted>
---
drivers/virtio/virtio_ring.c | 6 +++---
1 files changed, 3 insertions(+), 3 deletions(-)
@@ -308,9 +308,9 @@ bool virtqueue_kick_prepare(struct virtqueue *_vq)boolneeds_kick;START_USE(vq);-/* Descriptors and available array need to be set before we expose the-*newavailablearrayentries.*/-virtio_wmb(vq);+/* We need expose available array entries before checking avail+*event.*/+virtio_mb(vq);old=vq->vring.avail->idx-vq->num_added;new=vq->vring.avail->idx;
From: "Michael S. Tsirkin" <mst@redhat.com> Date: 2012-01-22 11:45:40
On Fri, Jan 20, 2012 at 04:16:59PM +0800, Jason Wang wrote:
Signed-off-by: Jason Wang <redacted>
Good catch.
Note: this fixes a bug introduced by
7b21e34fd1c272e3a8c3846168f2f6287a4cd72b.
It's probably a good idea to mention
this is the commit log.
Acked-by: Michael S. Tsirkin <mst@redhat.com>
@@ -35,7 +35,7 @@#define virtio_rmb(vq) \do{if((vq)->weak_barriers)smp_rmb();elsermb();}while(0)#define virtio_wmb(vq) \-do{if((vq)->weak_barriers)smp_rmb();elsermb();}while(0)+do{if((vq)->weak_barriers)smp_wmb();elsewmb();}while(0)#else/* We must force memory ordering even if guest is UP since host could be*runningonanotherCPU,butSMPbarriersaredefinedtobarrier()inthat
From: "Michael S. Tsirkin" <mst@redhat.com> Date: 2012-01-22 11:46:13
On Fri, Jan 20, 2012 at 04:17:08PM +0800, Jason Wang wrote:
Use virtio_mb() to make sure the available index to be exposed before
checking the the avail event. Otherwise we may get stale value of
avail event in guest and never kick the host after.
Signed-off-by: Jason Wang <redacted>
Good catch.
Note: this fixes a bug introduced by ee7cd8981e15bcb365fc762afe3fc47b8242f630.
Acked-by: Michael S. Tsirkin <mst@redhat.com>
@@ -308,9 +308,9 @@ bool virtqueue_kick_prepare(struct virtqueue *_vq)boolneeds_kick;START_USE(vq);-/* Descriptors and available array need to be set before we expose the-*newavailablearrayentries.*/-virtio_wmb(vq);+/* We need expose available array entries before checking avail
Nit:
We need expose -> Need to expose
+ * event. */
+ virtio_mb(vq);
old = vq->vring.avail->idx - vq->num_added;
new = vq->vring.avail->idx;
From: Rusty Russell <hidden> Date: 2012-01-22 23:58:34
On Sun, 22 Jan 2012 13:47:57 +0200, "Michael S. Tsirkin" [off-list ref] wrote:
On Fri, Jan 20, 2012 at 04:17:08PM +0800, Jason Wang wrote:
quoted
Use virtio_mb() to make sure the available index to be exposed before
checking the the avail event. Otherwise we may get stale value of
avail event in guest and never kick the host after.
Signed-off-by: Jason Wang <redacted>
Good catch.
Note: this fixes a bug introduced by ee7cd8981e15bcb365fc762afe3fc47b8242f630.
Indeed, great catch. It replaces the missing barrier, and after some
careful thought I've convinced myself that it is necessary for both the
avail index and flag cases.
quoted
START_USE(vq);
- /* Descriptors and available array need to be set before we expose the
- * new available array entries. */
- virtio_wmb(vq);
+ /* We need expose available array entries before checking avail
From: Rusty Russell <hidden> Date: 2012-01-22 23:58:35
On Sun, 22 Jan 2012 13:47:46 +0200, "Michael S. Tsirkin" [off-list ref] wrote:
On Fri, Jan 20, 2012 at 04:16:59PM +0800, Jason Wang wrote:
quoted
Signed-off-by: Jason Wang <redacted>
Good catch.
Note: this fixes a bug introduced by
7b21e34fd1c272e3a8c3846168f2f6287a4cd72b.
It's probably a good idea to mention
this is the commit log.
Acked-by: Michael S. Tsirkin <mst@redhat.com>