2021-09-13 15:19:20

by John Garry

Subject: [PATCH RESEND v3 03/13] blk-mq: Relocate shared sbitmap resize in blk_mq_update_nr_requests()

For shared sbitmap, if the call to blk_mq_tag_update_depth() was
successful for any hctx when hctx->sched_tags is not set, then it would be
successful for all (due to nature in which blk_mq_tag_update_depth()
fails).

As such, there is no need to call blk_mq_tag_resize_shared_sbitmap() for
each hctx. So relocate the call until after the hctx iteration under the
!q->elevator check, which is equivalent (to !hctx->sched_tags).

Signed-off-by: John Garry <[email protected]>
Reviewed-by: Ming Lei <[email protected]>
---
block/blk-mq.c | 12 +++++++-----
1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/block/blk-mq.c b/block/blk-mq.c
index 2316ff27c1f5..1a4bb2db30e5 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -3616,8 +3616,6 @@ int blk_mq_update_nr_requests(struct request_queue *q, unsigned int nr)
if (!hctx->sched_tags) {
ret = blk_mq_tag_update_depth(hctx, &hctx->tags, nr,
false);
- if (!ret && blk_mq_is_sbitmap_shared(set->flags))
- blk_mq_tag_resize_shared_sbitmap(set, nr);
} else {
ret = blk_mq_tag_update_depth(hctx, &hctx->sched_tags,
nr, true);
@@ -3635,9 +3633,13 @@ int blk_mq_update_nr_requests(struct request_queue *q, unsigned int nr)
}
if (!ret) {
q->nr_requests = nr;
- if (q->elevator && blk_mq_is_sbitmap_shared(set->flags))
- sbitmap_queue_resize(&q->sched_bitmap_tags,
- nr - set->reserved_tags);
+ if (blk_mq_is_sbitmap_shared(set->flags)) {
+ if (q->elevator)
+ sbitmap_queue_resize(&q->sched_bitmap_tags,
+ nr - set->reserved_tags);
+ else
+ blk_mq_tag_resize_shared_sbitmap(set, nr);
+ }
}

blk_mq_unquiesce_queue(q);
--
2.26.2

2021-09-14 05:39:52

by Hannes Reinecke

Subject: Re: [PATCH RESEND v3 03/13] blk-mq: Relocate shared sbitmap resize in blk_mq_update_nr_requests()

On 9/13/21 5:12 PM, John Garry wrote:
> For shared sbitmap, if the call to blk_mq_tag_update_depth() was
> successful for any hctx when hctx->sched_tags is not set, then it would be
> successful for all (due to nature in which blk_mq_tag_update_depth()
> fails).
>
> As such, there is no need to call blk_mq_tag_resize_shared_sbitmap() for
> each hctx. So relocate the call until after the hctx iteration under the
> !q->elevator check, which is equivalent (to !hctx->sched_tags).
>
> Signed-off-by: John Garry <[email protected]>
> Reviewed-by: Ming Lei <[email protected]>
> ---
> block/blk-mq.c | 12 +++++++-----
> 1 file changed, 7 insertions(+), 5 deletions(-)
>
Reviewed-by: Hannes Reinecke <[email protected]>

Cheers,

Hannes
--
Dr. Hannes Reinecke Kernel Storage Architect
[email protected] +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Felix Imendörffer