On Tue, 2018-03-13 at 22:32 +0800, Ming Lei wrote: > On Tue, Mar 13, 2018 at 02:08:23PM +0100, Martin Steigerwald wrote: > > Ming and Bart, I added you to cc, cause I had to do with you about another > > blk-mq report, please feel free to adapt. > > Looks RIP points to scsi_times_out+0x17/0x1d0, maybe a SCSI regression? I think that it's much more likely that this is a block layer regression. See e.g. "[PATCH v2] blk-mq: Fix race between resetting the timer and completion handling" (https://www.mail-archive.com/linux-block@xxxxxxxxxxxxxxx/msg18338.html). Bart.
