Linux bcache driver list
 help / color / mirror / Atom feed
From: Yu Kuai <yukuai@kernel.org>
To: Jens Axboe <axboe@kernel.dk>, Tejun Heo <tj@kernel.org>
Cc: Christoph Hellwig <hch@lst.de>, Keith Busch <kbusch@kernel.org>,
	Sagi Grimberg <sagi@grimberg.me>,
	Alasdair Kergon <agk@redhat.com>,
	Benjamin Marzinski <bmarzins@redhat.com>,
	Mike Snitzer <snitzer@kernel.org>,
	Mikulas Patocka <mpatocka@redhat.com>,
	Dongsheng Yang <dongsheng.yang@linux.dev>,
	Zheng Gu <cengku@gmail.com>, Coly Li <colyli@fygo.io>,
	Kent Overstreet <kent.overstreet@linux.dev>,
	Josef Bacik <josef@toxicpanda.com>, Yu Kuai <yukuai@fygo.io>,
	Nilay Shroff <nilay@linux.ibm.com>,
	linux-block@vger.kernel.org, cgroups@vger.kernel.org,
	linux-nvme@lists.infradead.org, dm-devel@lists.linux.dev,
	linux-bcache@vger.kernel.org
Subject: [RFC PATCH v1 07/17] block: support non-blocking bio allocation with a bdev
Date: Sun,  5 Jul 2026 03:51:14 +0800	[thread overview]
Message-ID: <20260704195124.1375075-8-yukuai@kernel.org> (raw)
In-Reply-To: <20260704195124.1375075-1-yukuai@kernel.org>

From: Yu Kuai <yukuai@fygo.io>

bio_alloc_clone(), bio_init_clone(), and bio_alloc_bioset() can be called
with non-blocking GFP masks.  Passing a bdev into bio initialization may
need to associate blkcg state and, after missing blkg creation is serialized
by q->blkcg_mutex, that association can sleep.

Keep the generic block layer simple by letting bio_alloc_bioset() handle this
case directly.  Non-blocking allocations initialize the bio without a bdev,
set the bdev fields, and associate the blkg with nowait=true.  If the needed
blkg is missing and would have to be created, allocation fails normally so the
caller can retry from a blocking context.

Blocking callers keep the existing allocation-time association behavior.

Signed-off-by: Yu Kuai <yukuai@fygo.io>
---
 block/bio.c | 46 ++++++++++++++++++++++++++++++++++++++++------
 1 file changed, 40 insertions(+), 6 deletions(-)

diff --git a/block/bio.c b/block/bio.c
index b74e9961c8ee..863ae73a4222 100644
--- a/block/bio.c
+++ b/block/bio.c
@@ -259,6 +259,20 @@ void bio_init(struct bio *bio, struct block_device *bdev, struct bio_vec *table,
 }
 EXPORT_SYMBOL(bio_init);
 
+static bool bio_init_nowait(struct bio *bio, struct block_device *bdev,
+		struct bio_vec *table, unsigned short max_vecs, blk_opf_t opf)
+{
+	bio_init(bio, NULL, table, max_vecs, opf);
+	if (bdev) {
+		bio_set_dev_no_blkg(bio, bdev);
+		if (bio_associate_blkg(bio, true))
+			return true;
+		bio_uninit(bio);
+		return false;
+	}
+	return true;
+}
+
 /**
  * bio_reset - reinitialize a bio
  * @bio:	bio to reset
@@ -599,12 +613,25 @@ struct bio *bio_alloc_bioset(struct block_device *bdev, unsigned short nr_vecs,
 		}
 	}
 
-	if (nr_vecs && nr_vecs <= BIO_INLINE_VECS)
-		bio_init_inline(bio, bdev, nr_vecs, opf);
-	else
-		bio_init(bio, bdev, bvecs, nr_vecs, opf);
+	if (nr_vecs && nr_vecs <= BIO_INLINE_VECS) {
+		bvecs = bio_inline_vecs(bio);
+		if (gfpflags_allow_blocking(saved_gfp))
+			bio_init(bio, bdev, bvecs, nr_vecs, opf);
+		else if (!bio_init_nowait(bio, bdev, bvecs, nr_vecs, opf))
+			goto fail_free_bio;
+	} else {
+		if (gfpflags_allow_blocking(saved_gfp))
+			bio_init(bio, bdev, bvecs, nr_vecs, opf);
+		else if (!bio_init_nowait(bio, bdev, bvecs, nr_vecs, opf))
+			goto fail_free_bio;
+	}
 	bio->bi_pool = bs;
 	return bio;
+
+fail_free_bio:
+	bio->bi_pool = bs;
+	bio_put(bio);
+	return NULL;
 }
 EXPORT_SYMBOL(bio_alloc_bioset);
 
@@ -857,7 +884,9 @@ static int __bio_clone(struct bio *bio, struct bio *bio_src, gfp_t gfp)
 		if (bio->bi_bdev == bio_src->bi_bdev &&
 		    bio_flagged(bio_src, BIO_REMAPPED))
 			bio_set_flag(bio, BIO_REMAPPED);
-		bio_clone_blkg_association(bio, bio_src, false);
+		if (!bio_clone_blkg_association(bio, bio_src,
+					!gfpflags_allow_blocking(gfp)))
+			return -ENOMEM;
 	}
 
 	if (bio_crypt_clone(bio, bio_src, gfp) < 0)
@@ -913,9 +942,14 @@ EXPORT_SYMBOL(bio_alloc_clone);
 int bio_init_clone(struct block_device *bdev, struct bio *bio,
 		struct bio *bio_src, gfp_t gfp)
 {
+	bool blocking = gfpflags_allow_blocking(gfp);
 	int ret;
 
-	bio_init(bio, bdev, bio_src->bi_io_vec, 0, bio_src->bi_opf);
+	if (blocking)
+		bio_init(bio, bdev, bio_src->bi_io_vec, 0, bio_src->bi_opf);
+	else if (!bio_init_nowait(bio, bdev, bio_src->bi_io_vec, 0,
+				bio_src->bi_opf))
+		return -ENOMEM;
 	ret = __bio_clone(bio, bio_src, gfp);
 	if (ret)
 		bio_uninit(bio);
-- 
2.51.0


  parent reply	other threads:[~2026-07-04 19:53 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-07-04 19:51 [RFC PATCH v1 00/17] blk-cgroup: protect blkgs with blkcg_mutex Yu Kuai
2026-07-04 19:51 ` [RFC PATCH v1 01/17] nvme-multipath: retarget failedover bios from requeue work Yu Kuai
2026-07-04 19:51 ` [RFC PATCH v1 02/17] dm thin: avoid bio_set_dev under pool lock Yu Kuai
2026-07-04 19:51 ` [RFC PATCH v1 03/17] dm snapshot: avoid bio_set_dev in locked map paths Yu Kuai
2026-07-04 19:51 ` [RFC PATCH v1 04/17] blk-throttle: protect throttle state with td lock Yu Kuai
2026-07-04 19:51 ` [RFC PATCH v1 05/17] block: add bio_alloc_atomic() for atomic bio users Yu Kuai
2026-07-04 19:51 ` [RFC PATCH v1 06/17] blk-cgroup: support non-blocking bio association Yu Kuai
2026-07-04 19:51 ` Yu Kuai [this message]
2026-07-04 19:51 ` [RFC PATCH v1 08/17] bcache: avoid sleeping blkg association from locked paths Yu Kuai
2026-07-04 19:51 ` [RFC PATCH v1 09/17] dm bufio: avoid blkg association from GFP_NOWAIT bio init Yu Kuai
2026-07-04 19:51 ` [RFC PATCH v1 10/17] dm pcache: handle non-blocking bio clone init failure Yu Kuai
2026-07-04 19:51 ` [RFC PATCH v1 11/17] block: avoid scheduling from non-blocking helper allocations Yu Kuai
2026-07-04 19:51 ` [RFC PATCH v1 12/17] dm: avoid sleeping blkg association from NOWAIT remaps Yu Kuai
2026-07-04 19:51 ` [RFC PATCH v1 13/17] bfq: avoid blkg lookup from locked cgroup update Yu Kuai
2026-07-04 19:51 ` [RFC PATCH v1 14/17] blk-cgroup: protect blkgs with blkcg_mutex Yu Kuai
2026-07-04 19:51 ` [RFC PATCH v1 15/17] blk-cgroup: remove blkg radix tree preloading Yu Kuai
2026-07-04 19:51 ` [RFC PATCH v1 16/17] blk-cgroup: allocate blkgs in blkg_create Yu Kuai
2026-07-04 19:51 ` [RFC PATCH v1 17/17] blk-cgroup: share blkg creation between lookup and config prep Yu Kuai

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=20260704195124.1375075-8-yukuai@kernel.org \
    --to=yukuai@kernel.org \
    --cc=agk@redhat.com \
    --cc=axboe@kernel.dk \
    --cc=bmarzins@redhat.com \
    --cc=cengku@gmail.com \
    --cc=cgroups@vger.kernel.org \
    --cc=colyli@fygo.io \
    --cc=dm-devel@lists.linux.dev \
    --cc=dongsheng.yang@linux.dev \
    --cc=hch@lst.de \
    --cc=josef@toxicpanda.com \
    --cc=kbusch@kernel.org \
    --cc=kent.overstreet@linux.dev \
    --cc=linux-bcache@vger.kernel.org \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-nvme@lists.infradead.org \
    --cc=mpatocka@redhat.com \
    --cc=nilay@linux.ibm.com \
    --cc=sagi@grimberg.me \
    --cc=snitzer@kernel.org \
    --cc=tj@kernel.org \
    --cc=yukuai@fygo.io \
    /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