linux-block.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jens Axboe <axboe@kernel.dk>
To: linux-block@vger.kernel.org
Cc: Jens Axboe <axboe@kernel.dk>
Subject: [PATCH 05/11] block: avoid ordered task state change for polled IO
Date: Tue, 13 Nov 2018 08:42:27 -0700	[thread overview]
Message-ID: <20181113154233.15256-6-axboe@kernel.dk> (raw)
In-Reply-To: <20181113154233.15256-1-axboe@kernel.dk>

Ensure that writes to the dio/bio waiter field are ordered
correctly. With the smp_rmb() before the READ_ONCE() check,
we should be able to use a more relaxed ordering for the
task state setting.

Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 fs/block_dev.c | 11 +++++++++--
 fs/iomap.c     |  6 +++++-
 2 files changed, 14 insertions(+), 3 deletions(-)

diff --git a/fs/block_dev.c b/fs/block_dev.c
index c039abfb2052..2f920c03996e 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -181,6 +181,7 @@ static void blkdev_bio_end_io_simple(struct bio *bio)
 	struct task_struct *waiter = bio->bi_private;
 
 	WRITE_ONCE(bio->bi_private, NULL);
+	smp_wmb();
 	wake_up_process(waiter);
 }
 
@@ -237,9 +238,12 @@ __blkdev_direct_IO_simple(struct kiocb *iocb, struct iov_iter *iter,
 
 	qc = submit_bio(&bio);
 	for (;;) {
-		set_current_state(TASK_UNINTERRUPTIBLE);
+		__set_current_state(TASK_UNINTERRUPTIBLE);
+
+		smp_rmb();
 		if (!READ_ONCE(bio.bi_private))
 			break;
+
 		if (!(iocb->ki_flags & IOCB_HIPRI) ||
 		    !blk_poll(bdev_get_queue(bdev), qc))
 			io_schedule();
@@ -305,6 +309,7 @@ static void blkdev_bio_end_io(struct bio *bio)
 			struct task_struct *waiter = dio->waiter;
 
 			WRITE_ONCE(dio->waiter, NULL);
+			smp_wmb();
 			wake_up_process(waiter);
 		}
 	}
@@ -403,7 +408,9 @@ __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages)
 		return -EIOCBQUEUED;
 
 	for (;;) {
-		set_current_state(TASK_UNINTERRUPTIBLE);
+		__set_current_state(TASK_UNINTERRUPTIBLE);
+
+		smp_rmb();
 		if (!READ_ONCE(dio->waiter))
 			break;
 
diff --git a/fs/iomap.c b/fs/iomap.c
index f61d13dfdf09..7898927e758e 100644
--- a/fs/iomap.c
+++ b/fs/iomap.c
@@ -1524,7 +1524,9 @@ static void iomap_dio_bio_end_io(struct bio *bio)
 	if (atomic_dec_and_test(&dio->ref)) {
 		if (dio->wait_for_completion) {
 			struct task_struct *waiter = dio->submit.waiter;
+
 			WRITE_ONCE(dio->submit.waiter, NULL);
+			smp_wmb();
 			wake_up_process(waiter);
 		} else if (dio->flags & IOMAP_DIO_WRITE) {
 			struct inode *inode = file_inode(dio->iocb->ki_filp);
@@ -1888,7 +1890,9 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
 			return -EIOCBQUEUED;
 
 		for (;;) {
-			set_current_state(TASK_UNINTERRUPTIBLE);
+			__set_current_state(TASK_UNINTERRUPTIBLE);
+
+			smp_rmb();
 			if (!READ_ONCE(dio->submit.waiter))
 				break;
 
-- 
2.17.1


  parent reply	other threads:[~2018-11-13 15:42 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-11-13 15:42 [PATCHSET v2 0/11] Various block optimizations Jens Axboe
2018-11-13 15:42 ` [PATCH 01/11] nvme: don't disable local ints for polled queue Jens Axboe
2018-11-13 15:42 ` [PATCH 02/11] block: add queue_is_mq() helper Jens Axboe
2018-11-14 15:23   ` Christoph Hellwig
2018-11-13 15:42 ` [PATCH 03/11] blk-mq: embed blk_mq_ops directly in the request queue Jens Axboe
2018-11-13 15:42 ` [PATCH 04/11] blk-rq-qos: inline check for q->rq_qos functions Jens Axboe
2018-11-13 15:42 ` Jens Axboe [this message]
2018-11-14  2:29   ` [PATCH 05/11] block: avoid ordered task state change for polled IO jianchao.wang
2018-11-14  2:35     ` Jens Axboe
2018-11-13 15:42 ` [PATCH 06/11] block: add polled wakeup task helper Jens Axboe
2018-11-13 15:52   ` Keith Busch
2018-11-13 15:58     ` Jens Axboe
2018-11-13 16:59       ` Jens Axboe
2018-11-13 15:42 ` [PATCH 07/11] block: have ->poll_fn() return number of entries polled Jens Axboe
2018-11-13 15:42 ` [PATCH 08/11] blk-mq: when polling for IO, look for any completion Jens Axboe
2018-11-13 15:42 ` [PATCH 09/11] block: make blk_poll() take a parameter on whether to spin or not Jens Axboe
2018-11-13 15:42 ` [PATCH 10/11] block: for async O_DIRECT, mark us as polling if asked to Jens Axboe
2018-11-13 15:42 ` [PATCH 11/11] block: don't plug for aio/O_DIRECT HIPRI IO 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=20181113154233.15256-6-axboe@kernel.dk \
    --to=axboe@kernel.dk \
    --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).