All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ming Lei <ming.lei@redhat.com>
To: Jens Axboe <axboe@kernel.dk>,
	linux-block@vger.kernel.org,
	Christoph Hellwig <hch@infradead.org>,
	Mike Snitzer <snitzer@redhat.com>
Cc: dm-devel@redhat.com, Ming Lei <ming.lei@redhat.com>,
	Laurence Oberman <loberman@redhat.com>,
	Bart Van Assche <bart.vanassche@sandisk.com>
Subject: [PATCH 4/5] blk-mq: introduce blk_get_request_notify
Date: Mon, 22 Jan 2018 11:35:49 +0800	[thread overview]
Message-ID: <20180122033550.27855-5-ming.lei@redhat.com> (raw)
In-Reply-To: <20180122033550.27855-1-ming.lei@redhat.com>

DM-MPATH need to allocate request from underlying queue, but when the
allocation fails, there is no way to make underlying queue's RESTART
to restart DM's queue.

This patch introduces blk_get_request_notify() for this purpose, and
caller need to pass 'wait_queue_entry_t' to this function, and make
sure it is initialized well, so after the current allocation fails,
DM will get notified when there is request available from underlying
queue.

This approach is suggested by Jens, and has been used in blk-mq dispatch
patch for a while, see blk_mq_mark_tag_wait().

Suggested-by: Jens Axboe <axboe@kernel.dk>
Cc: Mike Snitzer <snitzer@redhat.com>
Cc: Laurence Oberman <loberman@redhat.com>
Cc: Bart Van Assche <bart.vanassche@sandisk.com>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
---
 block/blk-mq-tag.c     | 27 ++++++++++++++++++++++++++-
 block/blk-mq.c         | 24 +++++++++++++++++++++---
 block/blk-mq.h         |  1 +
 include/linux/blk-mq.h |  5 +++++
 4 files changed, 53 insertions(+), 4 deletions(-)

diff --git a/block/blk-mq-tag.c b/block/blk-mq-tag.c
index 336dde07b230..911fc9bd1bab 100644
--- a/block/blk-mq-tag.c
+++ b/block/blk-mq-tag.c
@@ -128,10 +128,35 @@ unsigned int blk_mq_get_tag(struct blk_mq_alloc_data *data)
 	if (tag != -1)
 		goto found_tag;
 
-	if (data->flags & BLK_MQ_REQ_NOWAIT)
+	if ((data->flags & BLK_MQ_REQ_NOWAIT) && !(data->notifier &&
+			(data->flags & BLK_MQ_REQ_ALLOC_NOTIFY)))
 		return BLK_MQ_TAG_FAIL;
 
 	ws = bt_wait_ptr(bt, data->hctx);
+
+	/*
+	 * If caller requires notification when tag is available, add
+	 * wait entry of 'data->notifier' to the wait queue.
+	 */
+	if (data->flags & BLK_MQ_REQ_NOWAIT) {
+		bool added = false;
+
+		spin_lock_irq(&ws->wait.lock);
+		if (list_empty(&data->notifier->entry))
+			__add_wait_queue(&ws->wait, data->notifier);
+		else
+			added = true;
+		spin_unlock_irq(&ws->wait.lock);
+
+		if (added)
+			return BLK_MQ_TAG_FAIL;
+
+		tag = __blk_mq_get_tag(data, bt);
+		if (tag != -1)
+			goto found_tag;
+		return BLK_MQ_TAG_FAIL;
+	}
+
 	drop_ctx = data->ctx == NULL;
 	do {
 		/*
diff --git a/block/blk-mq.c b/block/blk-mq.c
index e91d688792a9..a71e44f17e7f 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -384,10 +384,14 @@ static struct request *blk_mq_get_request(struct request_queue *q,
 	return rq;
 }
 
-struct request *blk_mq_alloc_request(struct request_queue *q, unsigned int op,
-		blk_mq_req_flags_t flags)
+static struct request *__blk_mq_alloc_request(struct request_queue *q,
+		unsigned int op, blk_mq_req_flags_t flags,
+		wait_queue_entry_t *notifier)
 {
-	struct blk_mq_alloc_data alloc_data = { .flags = flags };
+	struct blk_mq_alloc_data alloc_data = {
+		.flags = flags,
+		.notifier = notifier,
+	};
 	struct request *rq;
 	int ret;
 
@@ -408,8 +412,22 @@ struct request *blk_mq_alloc_request(struct request_queue *q, unsigned int op,
 	rq->bio = rq->biotail = NULL;
 	return rq;
 }
+
+struct request *blk_mq_alloc_request(struct request_queue *q, unsigned int op,
+		blk_mq_req_flags_t flags)
+{
+	return __blk_mq_alloc_request(q, op, flags, NULL);
+}
 EXPORT_SYMBOL(blk_mq_alloc_request);
 
+struct request *blk_mq_alloc_request_notify(struct request_queue *q,
+		unsigned int op, blk_mq_req_flags_t flags,
+		wait_queue_entry_t *notifier)
+{
+	return __blk_mq_alloc_request(q, op, flags, notifier);
+}
+EXPORT_SYMBOL(blk_mq_alloc_request_notify);
+
 struct request *blk_mq_alloc_request_hctx(struct request_queue *q,
 	unsigned int op, blk_mq_req_flags_t flags, unsigned int hctx_idx)
 {
diff --git a/block/blk-mq.h b/block/blk-mq.h
index 88c558f71819..bec2f675f8f1 100644
--- a/block/blk-mq.h
+++ b/block/blk-mq.h
@@ -160,6 +160,7 @@ struct blk_mq_alloc_data {
 	struct request_queue *q;
 	blk_mq_req_flags_t flags;
 	unsigned int shallow_depth;
+	wait_queue_entry_t *notifier;
 
 	/* input & output parameter */
 	struct blk_mq_ctx *ctx;
diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h
index 8efcf49796a3..335c7dace5a7 100644
--- a/include/linux/blk-mq.h
+++ b/include/linux/blk-mq.h
@@ -219,10 +219,15 @@ enum {
 	BLK_MQ_REQ_INTERNAL	= (__force blk_mq_req_flags_t)(1 << 2),
 	/* set RQF_PREEMPT */
 	BLK_MQ_REQ_PREEMPT	= (__force blk_mq_req_flags_t)(1 << 3),
+	/* notify when new rq is available */
+	BLK_MQ_REQ_ALLOC_NOTIFY	= (__force blk_mq_req_flags_t)(1 << 4),
 };
 
 struct request *blk_mq_alloc_request(struct request_queue *q, unsigned int op,
 		blk_mq_req_flags_t flags);
+struct request *blk_mq_alloc_request_notify(struct request_queue *q,
+		unsigned int op, blk_mq_req_flags_t flags,
+		wait_queue_entry_t *notifier);
 struct request *blk_mq_alloc_request_hctx(struct request_queue *q,
 		unsigned int op, blk_mq_req_flags_t flags,
 		unsigned int hctx_idx);
-- 
2.9.5

  parent reply	other threads:[~2018-01-22  3:35 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-01-22  3:35 [PATCH 0/5] blk-mq & dm: fix IO hang and deal with one performance issue Ming Lei
2018-01-22  3:35 ` [PATCH 1/5] blk-mq: introduce BLK_STS_DEV_RESOURCE Ming Lei
2018-01-22 16:32   ` Christoph Hellwig
2018-01-22 16:49   ` Bart Van Assche
2018-01-22 16:49     ` Bart Van Assche
2018-01-23  0:57     ` Ming Lei
2018-01-23 16:17       ` Bart Van Assche
2018-01-23 16:26         ` Ming Lei
2018-01-23 16:37           ` Bart Van Assche
2018-01-23 16:41             ` Ming Lei
2018-01-23 16:47               ` Bart Van Assche
2018-01-23 16:47                 ` Bart Van Assche
2018-01-23 16:49                 ` Ming Lei
2018-01-23 16:54                   ` Bart Van Assche
2018-01-23 16:54                     ` Bart Van Assche
2018-01-23 16:59                     ` Ming Lei
2018-01-23 16:59                       ` Ming Lei
2018-01-23 22:01                       ` Bart Van Assche
2018-01-23 22:01                         ` Bart Van Assche
2018-01-24  2:31                         ` Ming Lei
2018-01-22  3:35 ` [PATCH 2/5] dm-rq: handle dispatch exception in dm_dispatch_clone_request() Ming Lei
2018-01-22  3:35   ` Ming Lei
2018-01-22  3:35 ` [PATCH 3/5] dm-rq: return BLK_STS_* from map_request() Ming Lei
2018-01-22  3:35   ` Ming Lei
2018-01-22  5:35   ` Ming Lei
2018-01-22  3:35 ` Ming Lei [this message]
2018-01-22 10:19   ` [PATCH 4/5] blk-mq: introduce blk_get_request_notify Ming Lei
2018-01-22 17:13   ` Bart Van Assche
2018-01-22 17:13     ` Bart Van Assche
2018-01-23  1:29     ` Ming Lei
2018-01-22  3:35 ` [PATCH 5/5] dm-mpath: use blk_mq_alloc_request_notify for allocating blk-mq req Ming Lei

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=20180122033550.27855-5-ming.lei@redhat.com \
    --to=ming.lei@redhat.com \
    --cc=axboe@kernel.dk \
    --cc=bart.vanassche@sandisk.com \
    --cc=dm-devel@redhat.com \
    --cc=hch@infradead.org \
    --cc=linux-block@vger.kernel.org \
    --cc=loberman@redhat.com \
    --cc=snitzer@redhat.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.