On 01/29/2018 11:07 AM, Jianchao Wang wrote:
> nvme_set_host_mem will invoke nvme_alloc_request without NOWAIT
> flag, it is unsafe for nvme_dev_disable. The adminq driver tags
> may have been used up when the previous outstanding adminq requests
> cannot be completed due to some hardware error. We have to depend
> on the timeout path to complete the previous outstanding adminq
> requests and free the tags.
> However, nvme_timeout will invoke nvme_dev_disable and try to
> get the shutdown_lock which is held by another context who is
> sleeping to wait for the tags to be freed by timeout path. A
> deadlock comes up.
>
> To fix it, let nvme_set_host_mem use NOWAIT flag.
In fact, this is the only case about nvme_set_host_mem in
nvme_dev_disable.
Consider the following case:
A adminq request expired.
timeout_work context
nvme_timeout
-> nvme_dev_disable
-> nvme_set_host_mem
if this adminq request expires again, the timeout_work
cannot handle this case, because it is waiting for the
result of nvme_set_host_mem.
Thanks
Jianchao
