diff mbox

[3/3] dm mpath: delay the requeue of blk-mq requests while all paths down

Message ID 1473782495-44128-4-git-send-email-snitzer@redhat.com (mailing list archive)
State New, archived
Headers show

Commit Message

Mike Snitzer Sept. 13, 2016, 4:01 p.m. UTC 
Return DM_MAPIO_DELAY_REQUEUE from .clone_and_map_rq.  Also, return
false from .busy, if all paths are down, so that blk-mq requests get
mapped via .clone_and_map_rq -- which results in DM_MAPIO_DELAY_REQUEUE
being returned to dm-rq.

This change allows for a noticeable reduction in cpu utilization
(reduced kworker load) while all paths are down, e.g.:

system CPU idleness (as measured by fio's --idle-prof=system):
before: system: 87.53%
after:  system: 97.07%

Signed-off-by: Mike Snitzer <snitzer@redhat.com>
---
 drivers/md/dm-mpath.c | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)

Comments

Hannes Reinecke Sept. 14, 2016, 6:25 a.m. UTC | #1 
On 09/13/2016 06:01 PM, Mike Snitzer wrote:
> Return DM_MAPIO_DELAY_REQUEUE from .clone_and_map_rq.  Also, return
> false from .busy, if all paths are down, so that blk-mq requests get
> mapped via .clone_and_map_rq -- which results in DM_MAPIO_DELAY_REQUEUE
> being returned to dm-rq.
> 
> This change allows for a noticeable reduction in cpu utilization
> (reduced kworker load) while all paths are down, e.g.:
> 
> system CPU idleness (as measured by fio's --idle-prof=system):
> before: system: 87.53%
> after:  system: 97.07%
> 
> Signed-off-by: Mike Snitzer <snitzer@redhat.com>
> ---
>  drivers/md/dm-mpath.c | 15 +++++++++------
>  1 file changed, 9 insertions(+), 6 deletions(-)
> 
Reviewed-by: Hannes Reinecke <hare@suse.com>

Cheers,

Hannes
diff mbox

Patch

diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c
index c777d38..7613d64 100644
--- a/drivers/md/dm-mpath.c
+++ b/drivers/md/dm-mpath.c
@@ -550,9 +550,9 @@  static int __multipath_map(struct dm_target *ti, struct request *clone,
 		pgpath = choose_pgpath(m, nr_bytes);
 
 	if (!pgpath) {
-		if (!must_push_back_rq(m))
-			r = -EIO;	/* Failed */
-		return r;
+		if (must_push_back_rq(m))
+			return DM_MAPIO_DELAY_REQUEUE;
+		return -EIO;	/* Failed */
 	} else if (test_bit(MPATHF_QUEUE_IO, &m->flags) ||
 		   test_bit(MPATHF_PG_INIT_REQUIRED, &m->flags)) {
 		pg_init_all_paths(m);
@@ -1990,11 +1990,14 @@  static int multipath_busy(struct dm_target *ti)
 	struct priority_group *pg, *next_pg;
 	struct pgpath *pgpath;
 
-	/* pg_init in progress or no paths available */
-	if (atomic_read(&m->pg_init_in_progress) ||
-	    (!atomic_read(&m->nr_valid_paths) && test_bit(MPATHF_QUEUE_IF_NO_PATH, &m->flags)))
+	/* pg_init in progress */
+	if (atomic_read(&m->pg_init_in_progress))
 		return true;
 
+	/* no paths available, for blk-mq: rely on IO mapping to delay requeue */
+	if (!atomic_read(&m->nr_valid_paths) && test_bit(MPATHF_QUEUE_IF_NO_PATH, &m->flags))
+		return (m->queue_mode != DM_TYPE_MQ_REQUEST_BASED);
+
 	/* Guess which priority_group will be used at next mapping time */
 	pg = lockless_dereference(m->current_pg);
 	next_pg = lockless_dereference(m->next_pg);