From: Christoph Hellwig <hch@infradead.org>
To: Goldwyn Rodrigues <rgoldwyn@suse.de>
Cc: linux-fsdevel@vger.kernel.org, jack@suse.com, hch@infradead.org,
linux-block@vger.kernel.org, linux-btrfs@vger.kernel.org,
linux-ext4@vger.kernel.org, linux-xfs@vger.kernel.org,
sagi@grimberg.me, avi@scylladb.com, axboe@kernel.dk,
linux-api@vger.kernel.org, willy@infradead.org,
tom.leiming@gmail.com, Goldwyn Rodrigues <rgoldwyn@suse.com>
Subject: Re: [PATCH 5/8] nowait aio: return on congested block device
Date: Tue, 18 Apr 2017 23:45:05 -0700 [thread overview]
Message-ID: <20170419064505.GD20053@infradead.org> (raw)
In-Reply-To: <20170414120257.8932-6-rgoldwyn@suse.de>
On Fri, Apr 14, 2017 at 07:02:54AM -0500, Goldwyn Rodrigues wrote:
> From: Goldwyn Rodrigues <rgoldwyn@suse.com>
>
> A new bio operation flag REQ_NOWAIT is introduced to identify bio's
s/bio/block/
> @@ -1232,6 +1232,11 @@ static struct request *get_request(struct request_queue *q, unsigned int op,
> if (!IS_ERR(rq))
> return rq;
>
> + if (bio && (bio->bi_opf & REQ_NOWAIT)) {
> + blk_put_rl(rl);
> + return ERR_PTR(-EAGAIN);
> + }
Please check the op argument instead of touching bio.
> + if (bio->bi_opf & REQ_NOWAIT) {
> + if (!blk_queue_nowait(q)) {
> + err = -EOPNOTSUPP;
> + goto end_io;
> + }
> + if (!(bio->bi_opf & REQ_SYNC)) {
I don't understand this check at all..
> + if (unlikely(!blk_queue_dying(q) && (bio->bi_opf & REQ_NOWAIT)))
Please break lines after 80 characters.
> @@ -119,6 +119,9 @@ struct request *blk_mq_sched_get_request(struct request_queue *q,
> if (likely(!data->hctx))
> data->hctx = blk_mq_map_queue(q, data->ctx->cpu);
>
> + if (bio && (bio->bi_opf & REQ_NOWAIT))
> + data->flags |= BLK_MQ_REQ_NOWAIT;
Check the op flag again here.
> +++ b/block/blk-mq.c
> @@ -1538,6 +1538,8 @@ static blk_qc_t blk_mq_make_request(struct request_queue *q, struct bio *bio)
> rq = blk_mq_sched_get_request(q, bio, bio->bi_opf, &data);
> if (unlikely(!rq)) {
> __wbt_done(q->rq_wb, wb_acct);
> + if (bio && (bio->bi_opf & REQ_NOWAIT))
> + bio_wouldblock_error(bio);
bio iѕ dereferences unconditionally above, so you can do the same.
> @@ -1662,6 +1664,8 @@ static blk_qc_t blk_sq_make_request(struct request_queue *q, struct bio *bio)
> rq = blk_mq_sched_get_request(q, bio, bio->bi_opf, &data);
> if (unlikely(!rq)) {
> __wbt_done(q->rq_wb, wb_acct);
> + if (bio && (bio->bi_opf & REQ_NOWAIT))
> + bio_wouldblock_error(bio);
Same here. Although blk_sq_make_request is gone anyway in the current
block tree..
> + /* Request queue supports BIO_NOWAIT */
> + queue_flag_set_unlocked(QUEUE_FLAG_NOWAIT, q);
BIO_NOWAIT is gone. And the comment would not be needed if the
flag had a more descriptive name, e.g. QUEUE_FLAG_NOWAIT_SUPPORT.
And I think all request based drivers should set the flag implicitly
as ->queuecommand can't sleep, and ->queue_rq only when it's always
offloaded to a workqueue when the BLK_MQ_F_BLOCKING flag is set.
> --- a/fs/direct-io.c
> +++ b/fs/direct-io.c
> @@ -480,8 +480,12 @@ static int dio_bio_complete(struct dio *dio, struct bio *bio)
> unsigned i;
> int err;
>
> - if (bio->bi_error)
> - dio->io_error = -EIO;
> + if (bio->bi_error) {
> + if (bio->bi_opf & REQ_NOWAIT)
> + dio->io_error = -EAGAIN;
> + else
> + dio->io_error = -EIO;
> + }
Huh? Once REQ_NOWAIT is set all errors are -EAGAIN?
next prev parent reply other threads:[~2017-04-19 6:45 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-04-14 12:02 [PATCH 0/8 v6] No wait AIO Goldwyn Rodrigues
[not found] ` <20170414120257.8932-1-rgoldwyn-l3A5Bk7waGM@public.gmane.org>
2017-04-14 12:02 ` [PATCH 1/8] Use RWF_* flags for AIO operations Goldwyn Rodrigues
[not found] ` <20170414120257.8932-2-rgoldwyn-l3A5Bk7waGM@public.gmane.org>
2017-04-19 6:37 ` Christoph Hellwig
2017-04-14 12:02 ` [PATCH 2/8] nowait aio: Introduce RWF_NOWAIT Goldwyn Rodrigues
2017-04-19 6:39 ` Christoph Hellwig
[not found] ` <20170419063930.GB20053-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
2017-04-19 10:30 ` Goldwyn Rodrigues
[not found] ` <2d816e3e-3625-7ebd-8550-68a7cc9c1a2e-l3A5Bk7waGM@public.gmane.org>
2017-04-19 11:03 ` Jan Kara
2017-04-14 12:02 ` [PATCH 3/8] nowait aio: return if direct write will trigger writeback Goldwyn Rodrigues
2017-04-14 12:02 ` [PATCH 4/8] nowait-aio: Introduce IOMAP_NOWAIT Goldwyn Rodrigues
2017-04-19 6:39 ` Christoph Hellwig
[not found] ` <20170414120257.8932-5-rgoldwyn-l3A5Bk7waGM@public.gmane.org>
2017-04-19 7:12 ` Jan Kara
2017-04-14 12:02 ` [PATCH 5/8] nowait aio: return on congested block device Goldwyn Rodrigues
2017-04-19 6:45 ` Christoph Hellwig [this message]
2017-04-19 15:21 ` Goldwyn Rodrigues
2017-04-20 13:43 ` Jan Kara
2017-04-24 21:10 ` Goldwyn Rodrigues
2017-04-25 2:28 ` Jens Axboe
2017-04-14 12:02 ` [PATCH 6/8] nowait aio: ext4 Goldwyn Rodrigues
2017-04-14 12:02 ` [PATCH 7/8] nowait aio: xfs Goldwyn Rodrigues
2017-04-19 6:45 ` Christoph Hellwig
2017-04-14 12:02 ` [PATCH 8/8] nowait aio: btrfs Goldwyn Rodrigues
[not found] ` <20170414120257.8932-9-rgoldwyn-l3A5Bk7waGM@public.gmane.org>
2017-04-18 11:39 ` David Sterba
-- strict thread matches above, loose matches on Subject: below --
2017-05-09 12:22 [PATCH 0/8 v7] No wait AIO Goldwyn Rodrigues
2017-05-09 12:22 ` [PATCH 5/8] nowait aio: return on congested block device Goldwyn Rodrigues
2017-05-11 7:44 ` Christoph Hellwig
[not found] ` <20170511074451.GD15626-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
2017-05-11 18:16 ` Goldwyn Rodrigues
2017-04-03 18:52 [PATCH 0/8 v4] No wait AIO Goldwyn Rodrigues
2017-04-03 18:53 ` [PATCH 5/8] nowait aio: return on congested block device Goldwyn Rodrigues
2017-04-04 6:49 ` Christoph Hellwig
2017-03-15 21:50 [PATCH 0/8 v3] No wait AIO Goldwyn Rodrigues
2017-03-15 21:51 ` [PATCH 5/8] nowait aio: return on congested block device Goldwyn Rodrigues
[not found] ` <20170315215107.5628-6-rgoldwyn-l3A5Bk7waGM@public.gmane.org>
2017-03-16 14:33 ` Jens Axboe
2017-03-17 2:03 ` Ming Lei
2017-03-17 12:23 ` Goldwyn Rodrigues
[not found] ` <eee4683d-9f44-434f-b97f-b0b24c7b3dab-tSWWG44O7X1aa/9Udqfwiw@public.gmane.org>
2017-03-24 11:32 ` Goldwyn Rodrigues
2017-03-24 14:39 ` Jens Axboe
2017-03-16 21:31 ` Dave Chinner
2017-03-17 12:23 ` Goldwyn Rodrigues
2017-03-20 17:33 ` Jan Kara
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=20170419064505.GD20053@infradead.org \
--to=hch@infradead.org \
--cc=avi@scylladb.com \
--cc=axboe@kernel.dk \
--cc=jack@suse.com \
--cc=linux-api@vger.kernel.org \
--cc=linux-block@vger.kernel.org \
--cc=linux-btrfs@vger.kernel.org \
--cc=linux-ext4@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-xfs@vger.kernel.org \
--cc=rgoldwyn@suse.com \
--cc=rgoldwyn@suse.de \
--cc=sagi@grimberg.me \
--cc=tom.leiming@gmail.com \
--cc=willy@infradead.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).