From: Bart Van Assche <bvanassche@acm.org>
To: Christoph Hellwig <hch@lst.de>, axboe@kernel.dk
Cc: linux-block@vger.kernel.org
Subject: Re: [PATCH 3/4] blk-mq: remove a pointless queue enter pair in blk_mq_alloc_request_hctx
Date: Sat, 16 May 2020 09:28:22 -0700 [thread overview]
Message-ID: <87513e5c-270c-41cf-51d8-9106351449b5@acm.org> (raw)
In-Reply-To: <20200516153430.294324-4-hch@lst.de>
On 2020-05-16 08:34, Christoph Hellwig wrote:
> No need for two queue references. Also reduce the q_usage_counter
> critical section to just the actual request allocation.
>
> Signed-off-by: Christoph Hellwig <hch@lst.de>
> ---
> block/blk-mq.c | 14 ++++----------
> 1 file changed, 4 insertions(+), 10 deletions(-)
>
> diff --git a/block/blk-mq.c b/block/blk-mq.c
> index d96d3931f33e6..69e58cc4244c0 100644
> --- a/block/blk-mq.c
> +++ b/block/blk-mq.c
> @@ -439,26 +439,20 @@ struct request *blk_mq_alloc_request_hctx(struct request_queue *q,
> if (hctx_idx >= q->nr_hw_queues)
> return ERR_PTR(-EIO);
>
> - ret = blk_queue_enter(q, flags);
> - if (ret)
> - return ERR_PTR(ret);
> -
> /*
> * Check if the hardware context is actually mapped to anything.
> * If not tell the caller that it should skip this queue.
> */
> alloc_data.hctx = q->queue_hw_ctx[hctx_idx];
> - if (!blk_mq_hw_queue_mapped(alloc_data.hctx)) {
> - blk_queue_exit(q);
> + if (!blk_mq_hw_queue_mapped(alloc_data.hctx))
> return ERR_PTR(-EXDEV);
> - }
> cpu = cpumask_first_and(alloc_data.hctx->cpumask, cpu_online_mask);
> alloc_data.ctx = __blk_mq_get_ctx(q, cpu);
>
> - blk_queue_enter_live(q);
> + ret = blk_queue_enter(q, flags);
> + if (ret)
> + return ERR_PTR(ret);
> rq = blk_mq_get_request(q, NULL, &alloc_data);
> - blk_queue_exit(q);
> -
> if (!rq) {
> blk_queue_exit(q);
> return ERR_PTR(-EWOULDBLOCK);
This change looks wrong to me. blk_mq_update_nr_hw_queues() modifies
q->queue_hw_ctx so q_usage_counter must be incremented before that
pointer is dereferenced.
Bart.
next prev parent reply other threads:[~2020-05-16 16:28 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-05-16 15:34 avoid a few q_usage_counter roundtrips v2 Christoph Hellwig
2020-05-16 15:34 ` [PATCH 1/4] blk-mq: move the call to blk_queue_enter_live out of blk_mq_get_request Christoph Hellwig
2020-05-16 16:24 ` Bart Van Assche
2020-05-16 15:34 ` [PATCH 2/4] blk-mq: remove a pointless queue enter pair in blk_mq_alloc_request Christoph Hellwig
2020-05-16 16:25 ` Bart Van Assche
2020-05-16 15:34 ` [PATCH 3/4] blk-mq: remove a pointless queue enter pair in blk_mq_alloc_request_hctx Christoph Hellwig
2020-05-16 16:28 ` Bart Van Assche [this message]
2020-05-16 17:31 ` Christoph Hellwig
2020-05-16 15:34 ` [PATCH 4/4] blk-mq: allow blk_mq_make_request to consume the q_usage_counter reference Christoph Hellwig
2020-05-16 18:27 avoid a few q_usage_counter roundtrips v3 Christoph Hellwig
2020-05-16 18:28 ` [PATCH 3/4] blk-mq: remove a pointless queue enter pair in blk_mq_alloc_request_hctx Christoph Hellwig
2020-05-16 18:44 ` 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=87513e5c-270c-41cf-51d8-9106351449b5@acm.org \
--to=bvanassche@acm.org \
--cc=axboe@kernel.dk \
--cc=hch@lst.de \
--cc=linux-block@vger.kernel.org \
/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).