vhost: fix barrier pairing
Subsystems (5)
Michael S. Tsirkin
maintainerRFC PATCH 1/1 vhost: fix barrier pairing
According to memory-barriers.txt, an smp memory barrier should always be paired with another smp memory barrier, and I quote "a lack of appropriate pairing is almost certainly an error".
In case of vhost, failure to flush out used index update before looking at the interrupt disable flag could result in missed interrupts, resulting in networking hang under stress.
This might happen when flags read bypasses used index write. So we see interrupts disabled and do not interrupt, at the same time guest writes flags value to enable interrupt, reads an old used index value, thinks that
used ring is empty and waits for interrupt.
Note: the barrier we pair with here is indrivers/virtio/virtio_ring.c, functionvring_enable_cb.
---
Dave, I think this is needed in 2.6.34, I'll send a pull request after doing some more testing.
Rusty, Juan, could you take a look as well please? Thanks!
drivers/vhost/vhost.c | 5 ++++- 1 files changed, 4 insertions(+), 1 deletions(-)
drivers/vhost/vhost.c +4 -1
@@ -1035,7 +1035,10 @@ int vhost_add_used(struct vhost_virtqueue *vq, unsigned int head, int len)/* This actually signals the guest, using eventfd. */ void vhost_signal (struct vhost_dev *dev,struct vhost_virtqueue *vq) { -__u16 flags =0 ; +__u16 flags; +/* Flush out used index updates. */ +smp_mb (); +if (get_user (flags, &vq->avail->flags)) {vq_err (vq,"Failed to get flags" );return ;
Signature · 1 line
1.7.1.12.g42b7f
Quoted 20 lines · Michael S. Tsirkin
According to memory-barriers.txt, an smp memory barrier should always be paired with another smp memory barrier, and I quote "a lack of appropriate pairing is almost certainly an error".
In case of vhost, failure to flush out used index update before looking at the interrupt disable flag could result in missed interrupts, resulting in networking hang under stress.
This might happen when flags read bypasses used index write. So we see interrupts disabled and do not interrupt, at the same time guest writes flags value to enable interrupt, reads an old used index value, thinks that
used ring is empty and waits for interrupt.Note: the barrier we pair with here is in
drivers/virtio/virtio_ring.c, functionvring_enable_cb.
Quoted 8 lines
Signed-off-by Michael S. Tsirkin <[email protected]>---
Dave, I think this is needed in 2.6.34, I'll send a pull request after doing some more testing.
Rusty, Juan, could you take a look as well please? Thanks!
I would have prefered to put it:
void vhost_add_used_and_signal(struct vhost_dev *dev,
struct vhost_virtqueue *vq, unsigned int head, int len)
{
vhost_add_used(vq, head, len);Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
smp_mb();
vhost_signal(dev, vq); }
Because it looks strange to have a barrier as the 1st instruction of a function. And this way it is clearer (at least to me) what we are protecting.
But on the other hand, we would have to put a comment explainingthat all users of vhost_signal() have to put that smp_mb() so .....
Perhaps just improving the commet stating that the corresponding barrier is there?
Quoted 3 lines
Note: the barrier we pair with here is in
drivers/virtio/virtio_ring.c, functionvring_enable_cb.
Good catch.
Later, Juan.