From f5f0d9ed92f4ba1ea746529ccc945cf60d4973c8 Mon Sep 17 00:00:00 2001 From: Ladi Prosek Date: Wed, 5 Oct 2016 17:22:28 +0200 Subject: [PATCH 6/8] virtio: zero vq->inuse in virtio_reset() RH-Author: Ladi Prosek Message-id: <1475666548-9186-7-git-send-email-lprosek@redhat.com> Patchwork-id: 72485 O-Subject: [RHEL-7.4 qemu-kvm v2 PATCH 6/6] virtio: zero vq->inuse in virtio_reset() Bugzilla: 1377968 RH-Acked-by: Paolo Bonzini RH-Acked-by: Michael S. Tsirkin RH-Acked-by: Stefan Hajnoczi From: Stefan Hajnoczi vq->inuse must be zeroed upon device reset like most other virtqueue fields. In theory, virtio_reset() just needs assert(vq->inuse == 0) since devices must clean up in-flight requests during reset (requests cannot not be leaked!). In practice, it is difficult to achieve vq->inuse == 0 across reset because balloon, blk, 9p, etc implement various different strategies for cleaning up requests. Most devices call g_free(elem) directly without telling virtio.c that the VirtQueueElement is cleaned up. Therefore vq->inuse is not decremented during reset. This patch zeroes vq->inuse and trusts that devices are not leaking VirtQueueElements across reset. I will send a follow-up series that refactors request life-cycle across all devices and converts vq->inuse = 0 into assert(vq->inuse == 0) but this more invasive approach is not appropriate for stable trees. Signed-off-by: Stefan Hajnoczi Cc: qemu-stable Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin Reviewed-by: Ladi Prosek (cherry picked from commit 4b7f91ed0270a371e1933efa21ba600b6da23ab9) Signed-off-by: Ladi Prosek Signed-off-by: Miroslav Rezanina --- hw/virtio/virtio.c | 1 + 1 file changed, 1 insertion(+) diff --git a/hw/virtio/virtio.c b/hw/virtio/virtio.c index 87a7639..cdb21b1 100644 --- a/hw/virtio/virtio.c +++ b/hw/virtio/virtio.c @@ -585,6 +585,7 @@ void virtio_reset(void *opaque) vdev->vq[i].signalled_used = 0; vdev->vq[i].signalled_used_valid = false; vdev->vq[i].notification = true; + vdev->vq[i].inuse = 0; } } -- 1.8.3.1