Commit 8e6e83d7 authored by Kundan Kumar's avatar Kundan Kumar Committed by Jens Axboe
Browse files

block: skip start/end time stamping for passthrough IO 



commit 41fa7222 ("blk-mq: do not include passthrough requests in I/O
accounting")' disables I/O accounting for passthrough requests. Since tools
like 'iostat' do not show anything useful for passthrough I/O, it's
wasteful to do start/end time-stamping. So do away with that.

Avoiding the time-stamping improves the I/O performance by ~7%

Signed-off-by: default avatarKundan Kumar <kundan.kumar@samsung.com>
Signed-off-by: default avatarKanchan Joshi <joshi.k@samsung.com>
Link: https://lore.kernel.org/r/20231222101707.6921-1-kundan.kumar@samsung.com


Signed-off-by: default avatarJens Axboe <axboe@kernel.dk>
parent f70a4792

include/linux/blk-mq.h

+6 −0
Original line number Diff line number Diff line
@@ -830,6 +830,12 @@ void blk_mq_end_request_batch(struct io_comp_batch *ib); 
 */

static inline bool blk_mq_need_time_stamp(struct request *rq)

{

	/*

	 * passthrough io doesn't use iostat accounting, cgroup stats

	 * and io scheduler functionalities.

	 */

	if (blk_rq_is_passthrough(rq))

		return false;

	return (rq->rq_flags & (RQF_IO_STAT | RQF_STATS | RQF_USE_SCHED));

}