All of lore.kernel.org
 help / color / mirror / Atom feed
* May nvme_get_ns_from_disk introduce deadlock?
@ 2016-05-03  6:33 Wang Sheng-Hui
  2016-05-03  7:52 ` hch
  0 siblings, 1 reply; 2+ messages in thread
From: Wang Sheng-Hui @ 2016-05-03  6:33 UTC (permalink / raw)


Hi,

I'm reading the code, and found that nvme_get_ns_from_disk *may* 
introduce deadlock by invoking nvme_free_ns in core.c (nested
spin_lock(&dev_list_lock) ). 
But I can't figure out which condition can lead to deadlock.

Will someone help to explain the code that how can it escape from deadlock?

Thanks,
Wang Sheng-Hui

^ permalink raw reply	[flat|nested] 2+ messages in thread

* May nvme_get_ns_from_disk introduce deadlock?
  2016-05-03  6:33 May nvme_get_ns_from_disk introduce deadlock? Wang Sheng-Hui
@ 2016-05-03  7:52 ` hch
  0 siblings, 0 replies; 2+ messages in thread
From: hch @ 2016-05-03  7:52 UTC (permalink / raw)


On Tue, May 03, 2016@02:33:32PM +0800, Wang Sheng-Hui wrote:
> Hi,
> 
> I'm reading the code, and found that nvme_get_ns_from_disk *may* 
> introduce deadlock by invoking nvme_free_ns in core.c (nested
> spin_lock(&dev_list_lock) ). 
> But I can't figure out which condition can lead to deadlock.
> 
> Will someone help to explain the code that how can it escape from deadlock?

Just move the put outside the spinlock..  Can you send a patch?

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2016-05-03  7:52 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2016-05-03  6:33 May nvme_get_ns_from_disk introduce deadlock? Wang Sheng-Hui
2016-05-03  7:52 ` hch

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.