Since nobody else is going to restart our hw_queue for us, the
blk_mq_start_stopped_hw_queues() is in virtblk_done() is not sufficient
necessarily sufficient to ensure that the queue will get started again.
In case of global resource outage (-ENOMEM because mapping failure,
because of swiotlb full) our virtqueue may be empty and we can get
stuck with a stopped hw_queue.
Let us not stop the queue on arbitrary errors, but only on -EONSPC which
indicates a full virtqueue, where the hw_queue is guaranteed to get
started by virtblk_done() before when it makes sense to carry on
submitting requests. Let us also remove a stale comment.
Signed-off-by: Halil Pasic <pasic@linux.ibm.com>
Cc: Jens Axboe <axboe@kernel.dk>
Fixes: e24adaf39551 ("virtio_ring: fix return code on DMA mapping fails")
Link: https://lore.kernel.org/r/20200213123728.61216-2-pasic@linux.ibm.com
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
err = virtblk_add_req(vblk->vqs[qid].vq, vbr, vbr->sg, num);
if (err) {
virtqueue_kick(vblk->vqs[qid].vq);
- blk_mq_stop_hw_queue(hctx);
+ /* Don't stop the queue if -ENOMEM: we may have failed to
+ * bounce the buffer due to global resource outage.
+ */
+ if (err == -ENOSPC)
+ blk_mq_stop_hw_queue(hctx);
spin_unlock_irqrestore(&vblk->vqs[qid].lock, flags);
- /* Out of mem doesn't actually happen, since we fall back
- * to direct descriptors */
if (err == -ENOMEM || err == -ENOSPC)
return BLK_STS_DEV_RESOURCE;
return BLK_STS_IOERR;