Commit 504db087 authored by Anton Eidelman's avatar Anton Eidelman Committed by Jens Axboe

nvme-multipath: fix possible I/O hang when paths are updated

nvme_state_set_live() making a path available triggers requeue_work
in order to resubmit requests that ended up on requeue_list when no
paths were available.

This requeue_work may race with concurrent nvme_ns_head_make_request()
that do not observe the live path yet.
Such concurrent requests may by made by either:
- New IO submission.
- Requeue_work triggered by nvme_failover_req() or another ana_work.

A race may cause requeue_work capture the state of requeue_list before
more requests get onto the list. These requests will stay on the list
forever unless requeue_work is triggered again.

In order to prevent such race, nvme_state_set_live() should
synchronize_srcu(&head->srcu) before triggering the requeue_work and
prevent nvme_ns_head_make_request referencing an old snapshot of the
path list.
Reviewed-by: default avatarChristoph Hellwig <hch@lst.de>
Signed-off-by: default avatarAnton Eidelman <anton@lightbitslabs.com>
Signed-off-by: default avatarSagi Grimberg <sagi@grimberg.me>
Signed-off-by: default avatarJens Axboe <axboe@kernel.dk>
parent 500f9fba
...@@ -428,6 +428,7 @@ static void nvme_mpath_set_live(struct nvme_ns *ns) ...@@ -428,6 +428,7 @@ static void nvme_mpath_set_live(struct nvme_ns *ns)
srcu_read_unlock(&head->srcu, srcu_idx); srcu_read_unlock(&head->srcu, srcu_idx);
} }
synchronize_srcu(&ns->head->srcu);
kblockd_schedule_work(&ns->head->requeue_work); kblockd_schedule_work(&ns->head->requeue_work);
} }
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment