On 1/30/18 1:30 PM, Keith Busch wrote:

On Tue, Jan 30, 2018 at 08:57:49AM -0700, Jens Axboe wrote:

quoted

Looking at the disassembly, 'n' is 2 and 'segments' is 0xffff.

Is this still a problem if you don't use an IO scheduler? With deadline,
I'm not finding any path to bio_attempt_discard_merge which is where the
nr_phys_segments is supposed to get it set to 2. Not sure how it could
becmoe 0xffff, though.

blk_mq_make_request() -> blk_mq_sched_bio_merge() -> __blk_mq_sched_bio_merge()
	-> blk_mq_attempt_merge() -> bio_attempt_discard_merge()

Doesn't matter what IO scheduler you use.

I don't know if it triggers without a scheduler. I've been running this code
continually on the laptop (always do), and haven't seen it before today.


-- 
Jens Axboe