From: Bart Van Assche <bart.vanassche@wdc.com>
To: Jens Axboe <axboe@kernel.dk>
Cc: linux-block@vger.kernel.org, Christoph Hellwig <hch@lst.de>,
Bart Van Assche <bart.vanassche@wdc.com>,
Omar Sandoval <osandov@fb.com>, Hannes Reinecke <hare@suse.de>,
Johannes Thumshirn <jthumshirn@suse.de>
Subject: [PATCH 3/7] blk-mq: Make blk_mq_mark_tag_wait() easier to read
Date: Thu, 30 Nov 2017 16:08:44 -0800 [thread overview]
Message-ID: <20171201000848.2656-4-bart.vanassche@wdc.com> (raw)
In-Reply-To: <20171201000848.2656-1-bart.vanassche@wdc.com>
Reduce the number of return statements from three to one. Reduce the
number of spin_unlock(&this_hctx->lock) calls from two to one. Fix
a misleading comment: since blk-mq-tag.c always uses wake_up_all()
other waiters are woken up whether or not the current hctx is removed
from the wait list. This patch does not change any functionality.
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Omar Sandoval <osandov@fb.com>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Johannes Thumshirn <jthumshirn@suse.de>
---
block/blk-mq.c | 41 +++++++++++++++--------------------------
1 file changed, 15 insertions(+), 26 deletions(-)
diff --git a/block/blk-mq.c b/block/blk-mq.c
index 7f290a91a612..26fec4dfa40f 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -1039,6 +1039,11 @@ static bool blk_mq_mark_tag_wait(struct blk_mq_hw_ctx **hctx,
if (!shared_tags) {
if (!test_bit(BLK_MQ_S_SCHED_RESTART, &this_hctx->state))
set_bit(BLK_MQ_S_SCHED_RESTART, &this_hctx->state);
+ ret = blk_mq_get_driver_tag(rq, hctx, false);
+ /*
+ * Don't clear RESTART here, someone else could have set it.
+ * At most this will cost an extra queue run.
+ */
} else {
wait = &this_hctx->dispatch_wait;
if (!list_empty_careful(&wait->entry))
@@ -1052,37 +1057,21 @@ static bool blk_mq_mark_tag_wait(struct blk_mq_hw_ctx **hctx,
ws = bt_wait_ptr(&this_hctx->tags->bitmap_tags, this_hctx);
add_wait_queue(&ws->wait, wait);
- }
-
- /*
- * It's possible that a tag was freed in the window between the
- * allocation failure and adding the hardware queue to the wait
- * queue.
- */
- ret = blk_mq_get_driver_tag(rq, hctx, false);
-
- if (!shared_tags) {
/*
- * Don't clear RESTART here, someone else could have set it.
- * At most this will cost an extra queue run.
+ * It's possible that a tag was freed in the window between the
+ * allocation failure and adding the hardware queue to the wait
+ * queue.
*/
- return ret;
- } else {
- if (!ret) {
- spin_unlock(&this_hctx->lock);
- return false;
+ ret = blk_mq_get_driver_tag(rq, hctx, false);
+ /* If we got a tag remove ourselves from the wait queue. */
+ if (ret) {
+ spin_lock_irq(&ws->wait.lock);
+ list_del_init(&wait->entry);
+ spin_unlock_irq(&ws->wait.lock);
}
-
- /*
- * We got a tag, remove ourselves from the wait queue to ensure
- * someone else gets the wakeup.
- */
- spin_lock_irq(&ws->wait.lock);
- list_del_init(&wait->entry);
- spin_unlock_irq(&ws->wait.lock);
spin_unlock(&this_hctx->lock);
- return true;
}
+ return ret;
}
bool blk_mq_dispatch_rq_list(struct request_queue *q, struct list_head *list,
--
2.15.0
next prev parent reply other threads:[~2017-12-01 0:08 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-12-01 0:08 [PATCH 0/7] blk-mq: Queue running fixes Bart Van Assche
2017-12-01 0:08 ` [PATCH 1/7] blk-mq: Fix spelling in a source code comment Bart Van Assche
2017-12-01 0:08 ` [PATCH 2/7] block: Document more locking requirements Bart Van Assche
2017-12-01 3:03 ` Jens Axboe
2017-12-01 17:05 ` Bart Van Assche
2017-12-01 0:08 ` Bart Van Assche [this message]
2017-12-01 0:08 ` [PATCH 4/7] blk-mq: Avoid that request processing stalls when sharing tags Bart Van Assche
2017-12-01 2:58 ` Ming Lei
2017-12-01 19:52 ` Bart Van Assche
2017-12-02 0:36 ` Ming Lei
2017-12-02 0:48 ` Bart Van Assche
2017-12-02 1:00 ` Ming Lei
2017-12-02 1:05 ` Bart Van Assche
2017-12-02 1:14 ` Ming Lei
2017-12-02 1:20 ` Ming Lei
2017-12-01 0:08 ` [PATCH 5/7] blk-mq: Avoid that blk_mq_mark_tag_wait() triggers a queue stall Bart Van Assche
2017-12-01 3:51 ` Ming Lei
2017-12-01 0:08 ` [PATCH 6/7] blk-mq: Rerun hardware queues after having called .put_budget() Bart Van Assche
2017-12-01 3:23 ` Ming Lei
2017-12-01 0:08 ` [PATCH 7/7] blk-mq: Fix another queue stall Bart Van Assche
2017-12-01 3:00 ` Jens Axboe
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=20171201000848.2656-4-bart.vanassche@wdc.com \
--to=bart.vanassche@wdc.com \
--cc=axboe@kernel.dk \
--cc=hare@suse.de \
--cc=hch@lst.de \
--cc=jthumshirn@suse.de \
--cc=linux-block@vger.kernel.org \
--cc=osandov@fb.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.