Commit face9eb2 authored by Dariusz Stojaczyk's avatar Dariusz Stojaczyk Committed by Jim Harris
Browse files

virtio: add mb() before checking notify flag 



The kernel vhost target enables notifying after processing
each interrupt in a following manner.

 * unset NO_NOTIFY flag
 * mb()
 * check avail ring for slipped requests

And we do the following after issuing each I/O.

 * update avail ring
 * check NO_NOTIFY

If NO_NOTIFY check is reordered and read first, we might read
old `true` value, and avail ring might be updated after the
kernel has already done its check. This easily leads to deadlock.

Change-Id: I6bb4490775dfed6fb2987e97c39b713054ae26ad
Signed-off-by: default avatarDariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/396499


Tested-by: default avatarSPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: default avatarJim Harris <james.r.harris@intel.com>
Reviewed-by: default avatarDaniel Verkamp <daniel.verkamp@intel.com>
parent 59266471

lib/virtio/virtio.c

+1 −0
Original line number Diff line number Diff line
@@ -464,6 +464,7 @@ virtqueue_req_flush(struct virtqueue *vq) 
	virtio_wmb();

	vq->vq_ring.avail->idx = vq->vq_avail_idx;



	virtio_mb();

	if (spdk_unlikely(!(vq->vq_ring.used->flags & VRING_USED_F_NO_NOTIFY))) {

		virtio_dev_backend_ops(vq->vdev)->notify_queue(vq->vdev, vq);

		SPDK_DEBUGLOG(SPDK_LOG_VIRTIO_DEV, "Notified backend after xmit\n");