From: Ming Lei <ming.lei@redhat.com>
To: Bart Van Assche <bvanassche@acm.org>
Cc: Jens Axboe <axboe@kernel.dk>,
linux-block@vger.kernel.org, Christoph Hellwig <hch@lst.de>,
Christoph Hellwig <hch@infradead.org>,
Hannes Reinecke <hare@suse.com>,
Johannes Thumshirn <jth@kernel.org>,
syzbot+d44e1b26ce5c3e77458d@syzkaller.appspotmail.com
Subject: Re: [PATCH 2/5] blk-mq: Keep set->nr_hw_queues and set->map[].nr_queues in sync
Date: Tue, 18 Feb 2020 11:16:43 +0800 [thread overview]
Message-ID: <20200218031643.GB30750@ming.t460p> (raw)
In-Reply-To: <20200217210839.28535-3-bvanassche@acm.org>
On Mon, Feb 17, 2020 at 01:08:36PM -0800, Bart Van Assche wrote:
> This patch fixes the following kernel warning:
>
> WARNING: CPU: 0 PID: 2501 at include/linux/cpumask.h:137
> Call Trace:
> blk_mq_run_hw_queue+0x19d/0x350 block/blk-mq.c:1508
> blk_mq_run_hw_queues+0x112/0x1a0 block/blk-mq.c:1525
> blk_mq_requeue_work+0x502/0x780 block/blk-mq.c:775
> process_one_work+0x9af/0x1740 kernel/workqueue.c:2269
> worker_thread+0x98/0xe40 kernel/workqueue.c:2415
> kthread+0x361/0x430 kernel/kthread.c:255
>
> Cc: Christoph Hellwig <hch@infradead.org>
> Cc: Ming Lei <ming.lei@redhat.com>
> Cc: Hannes Reinecke <hare@suse.com>
> Cc: Johannes Thumshirn <jth@kernel.org>
> Reported-by: syzbot+d44e1b26ce5c3e77458d@syzkaller.appspotmail.com
> Fixes: ed76e329d74a ("blk-mq: abstract out queue map") # v5.0
> Signed-off-by: Bart Van Assche <bvanassche@acm.org>
> ---
> block/blk-mq.c | 16 +++++++++++++---
> 1 file changed, 13 insertions(+), 3 deletions(-)
>
> diff --git a/block/blk-mq.c b/block/blk-mq.c
> index f298500e6dda..2b9f490f5a64 100644
> --- a/block/blk-mq.c
> +++ b/block/blk-mq.c
> @@ -3050,6 +3050,16 @@ static int blk_mq_update_queue_map(struct blk_mq_tag_set *set)
> }
> }
>
> +static void blk_mq_set_nr_hw_queues(struct blk_mq_tag_set *set,
> + int new_nr_hw_queues)
> +{
> + int i;
> +
> + set->nr_hw_queues = new_nr_hw_queues;
> + for (i = 0; i < set->nr_maps; i++)
> + set->map[i].nr_queues = new_nr_hw_queues;
> +}
> +
> static int blk_mq_realloc_tag_set_tags(struct blk_mq_tag_set *set,
> int cur_nr_hw_queues, int new_nr_hw_queues)
> {
> @@ -3068,7 +3078,7 @@ static int blk_mq_realloc_tag_set_tags(struct blk_mq_tag_set *set,
> sizeof(*set->tags));
> kfree(set->tags);
> set->tags = new_tags;
> - set->nr_hw_queues = new_nr_hw_queues;
> + blk_mq_set_nr_hw_queues(set, new_nr_hw_queues);
>
> return 0;
> }
> @@ -3330,7 +3340,7 @@ static void __blk_mq_update_nr_hw_queues(struct blk_mq_tag_set *set,
> goto reregister;
>
> prev_nr_hw_queues = set->nr_hw_queues;
> - set->nr_hw_queues = nr_hw_queues;
> + blk_mq_set_nr_hw_queues(set, nr_hw_queues);
> blk_mq_update_queue_map(set);
> fallback:
> list_for_each_entry(q, &set->tag_list, tag_set_list) {
> @@ -3338,7 +3348,7 @@ static void __blk_mq_update_nr_hw_queues(struct blk_mq_tag_set *set,
> if (q->nr_hw_queues != set->nr_hw_queues) {
> pr_warn("Increasing nr_hw_queues to %d fails, fallback to %d\n",
> nr_hw_queues, prev_nr_hw_queues);
> - set->nr_hw_queues = prev_nr_hw_queues;
> + blk_mq_set_nr_hw_queues(set, prev_nr_hw_queues);
> blk_mq_map_queues(&set->map[HCTX_TYPE_DEFAULT]);
> goto fallback;
> }
>
I guess the issue can be fixed by the following change, and we do not
need to touch each queue map:
diff --git a/block/blk-mq.c b/block/blk-mq.c
index 5f5c43ae3792..f7340afb89ec 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -3046,6 +3046,7 @@ static int blk_mq_update_queue_map(struct blk_mq_tag_set *set)
return set->ops->map_queues(set);
} else {
BUG_ON(set->nr_maps > 1);
+ set->map[HCTX_TYPE_DEFAULT].nr_queues = set->nr_hw_queues;
return blk_mq_map_queues(&set->map[HCTX_TYPE_DEFAULT]);
}
}
Thanks,
Ming
next prev parent reply other threads:[~2020-02-18 3:17 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-02-17 21:08 [PATCH 0/5] Five patches related to changing the number of hardware queues Bart Van Assche
2020-02-17 21:08 ` [PATCH 1/5] blk-mq: Fix a comment in include/linux/blk-mq.h Bart Van Assche
2020-02-18 3:04 ` Ming Lei
2020-02-17 21:08 ` [PATCH 2/5] blk-mq: Keep set->nr_hw_queues and set->map[].nr_queues in sync Bart Van Assche
2020-02-18 3:16 ` Ming Lei [this message]
2020-02-19 4:24 ` Bart Van Assche
2020-02-17 21:08 ` [PATCH 3/5] blk-mq: Fix a recently introduced regression in blk_mq_realloc_hw_ctxs() Bart Van Assche
2020-02-18 3:45 ` Ming Lei
2020-02-17 21:08 ` [PATCH 4/5] null_blk: Suppress an UBSAN complaint triggered when setting 'memory_backed' Bart Van Assche
2020-02-17 21:08 ` [PATCH 5/5] null_blk: Fix changing the number of hardware queues Bart Van Assche
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200218031643.GB30750@ming.t460p \
--to=ming.lei@redhat.com \
--cc=axboe@kernel.dk \
--cc=bvanassche@acm.org \
--cc=hare@suse.com \
--cc=hch@infradead.org \
--cc=hch@lst.de \
--cc=jth@kernel.org \
--cc=linux-block@vger.kernel.org \
--cc=syzbot+d44e1b26ce5c3e77458d@syzkaller.appspotmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).