mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Kent Overstreet <koverstreet@google.com>
To: linux-bcache@vger.kernel.org, linux-kernel@vger.kernel.org,
	dm-devel@redhat.com
Cc: Kent Overstreet <koverstreet@google.com>,
	tj@kernel.org, vgoyal@redhat.com, mpatocka@redhat.com,
	bharrosh@panasas.com, Jens Axboe <axboe@kernel.dk>,
	Alasdair Kergon <agk@redhat.com>, Sage Weil <sage@inktank.com>
Subject: [PATCH v6 13/13] block: Only clone bio vecs that are in use
Date: Wed, 22 Aug 2012 10:04:10 -0700	[thread overview]
Message-ID: <1345655050-28199-14-git-send-email-koverstreet@google.com> (raw)
In-Reply-To: <1345655050-28199-1-git-send-email-koverstreet@google.com>

bcache creates large bios internally, and then splits them according to
the device requirements before it sends them down. If a lower level
device tries to clone the bio, and the original bio had more than
BIO_MAX_PAGES, the clone will fail unecessarily.

We can fix this by only cloning the bio vecs that are actually in use.

Signed-off-by: Kent Overstreet <koverstreet@google.com>
CC: Jens Axboe <axboe@kernel.dk>
CC: Alasdair Kergon <agk@redhat.com>
CC: Sage Weil <sage@inktank.com>
---
 drivers/block/rbd.c |  2 +-
 drivers/md/dm.c     |  5 ++---
 fs/bio.c            | 15 ++++++++++-----
 3 files changed, 13 insertions(+), 9 deletions(-)

diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c
index 63e5852..5c3457f 100644
--- a/drivers/block/rbd.c
+++ b/drivers/block/rbd.c
@@ -734,7 +734,7 @@ static struct bio *bio_chain_clone(struct bio **old, struct bio **next,
 	}
 
 	while (old_chain && (total < len)) {
-		tmp = bio_kmalloc(gfpmask, old_chain->bi_max_vecs);
+		tmp = bio_kmalloc(gfpmask, bio_segments(old_chain));
 		if (!tmp)
 			goto err_out;
 
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index a3c38b9..3aeb108 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -1080,11 +1080,10 @@ static struct bio *clone_bio(struct bio *bio, sector_t sector,
 {
 	struct bio *clone;
 
-	clone = bio_alloc_bioset(GFP_NOIO, bio->bi_max_vecs, bs);
+	clone = bio_alloc_bioset(GFP_NOIO, bv_count, bs);
 	__bio_clone(clone, bio);
 	clone->bi_sector = sector;
-	clone->bi_idx = idx;
-	clone->bi_vcnt = idx + bv_count;
+	clone->bi_vcnt = bv_count;
 	clone->bi_size = to_bytes(len);
 	clone->bi_flags &= ~(1 << BIO_SEG_VALID);
 
diff --git a/fs/bio.c b/fs/bio.c
index a58c3c6..3f43e50 100644
--- a/fs/bio.c
+++ b/fs/bio.c
@@ -450,11 +450,16 @@ EXPORT_SYMBOL(bio_phys_segments);
  *	Clone a &bio. Caller will own the returned bio, but not
  *	the actual data it points to. Reference count of returned
  * 	bio will be one.
+ *
+ *	We don't clone the entire bvec, just the part from bi_idx to b_vcnt
+ *	(i.e. what the bio currently points to, so the new bio is still
+ *	equivalent to the old bio).
  */
 void __bio_clone(struct bio *bio, struct bio *bio_src)
 {
-	memcpy(bio->bi_io_vec, bio_src->bi_io_vec,
-		bio_src->bi_max_vecs * sizeof(struct bio_vec));
+	memcpy(bio->bi_io_vec,
+	       bio_iovec(bio_src),
+	       bio_segments(bio_src) * sizeof(struct bio_vec));
 
 	/*
 	 * most users will be overriding ->bi_bdev with a new target,
@@ -463,10 +468,10 @@ void __bio_clone(struct bio *bio, struct bio *bio_src)
 	bio->bi_sector = bio_src->bi_sector;
 	bio->bi_bdev = bio_src->bi_bdev;
 	bio->bi_flags |= 1 << BIO_CLONED;
+	bio->bi_flags &= ~(1 << BIO_SEG_VALID);
 	bio->bi_rw = bio_src->bi_rw;
-	bio->bi_vcnt = bio_src->bi_vcnt;
+	bio->bi_vcnt = bio_segments(bio_src);
 	bio->bi_size = bio_src->bi_size;
-	bio->bi_idx = bio_src->bi_idx;
 }
 EXPORT_SYMBOL(__bio_clone);
 
@@ -483,7 +488,7 @@ struct bio *bio_clone_bioset(struct bio *bio, gfp_t gfp_mask,
 {
 	struct bio *b;
 
-	b = bio_alloc_bioset(gfp_mask, bio->bi_max_vecs, bs);
+	b = bio_alloc_bioset(gfp_mask, bio_segments(bio), bs);
 	if (!b)
 		return NULL;
 
-- 
1.7.12


  parent reply	other threads:[~2012-08-22 17:05 UTC|newest]

Thread overview: 53+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-08-22 17:03 [PATCH v6 00/13] Block cleanups Kent Overstreet
2012-08-22 17:03 ` [PATCH v6 01/13] block: Generalized bio pool freeing Kent Overstreet
2012-08-22 21:27   ` Nicholas A. Bellinger
2012-08-22 17:03 ` [PATCH v6 02/13] dm: Use bioset's front_pad for dm_rq_clone_bio_info Kent Overstreet
2012-08-22 18:32   ` Tejun Heo
2012-08-22 21:30   ` Vivek Goyal
2012-08-24  7:14     ` Kent Overstreet
2012-08-24 18:40       ` Vivek Goyal
2012-08-22 17:04 ` [PATCH v6 03/13] block: Add bio_reset() Kent Overstreet
2012-08-22 18:34   ` Tejun Heo
2012-08-22 19:51     ` Tejun Heo
2012-08-22 17:04 ` [PATCH v6 04/13] pktcdvd: Switch to bio_kmalloc() Kent Overstreet
2012-08-22 19:55   ` Tejun Heo
2012-08-28 23:19     ` Jiri Kosina
2012-08-29  4:35       ` Peter Osterlund
2012-09-03 16:15     ` Jiri Kosina
2012-08-22 17:04 ` [PATCH v6 05/13] block: Kill bi_destructor Kent Overstreet
2012-08-22 20:00   ` Tejun Heo
2012-08-24  5:09     ` Kent Overstreet
2012-08-22 17:04 ` [PATCH v6 06/13] block: Consolidate bio_alloc_bioset(), bio_kmalloc() Kent Overstreet
2012-08-22 20:17   ` Tejun Heo
2012-08-24  5:04     ` Kent Overstreet
2012-08-24 20:08       ` Tejun Heo
2012-08-22 17:04 ` [PATCH v6 07/13] block: Avoid deadlocks with bio allocation by stacking drivers Kent Overstreet
2012-08-22 20:30   ` Tejun Heo
2012-08-24  5:55     ` Kent Overstreet
2012-08-24 20:28       ` Tejun Heo
2012-08-22 17:04 ` [PATCH v6 08/13] block: Add an explicit bio flag for bios that own their bvec Kent Overstreet
2012-08-22 17:43   ` Adrian Bunk
2012-08-22 19:22     ` Kent Overstreet
2012-08-22 20:00       ` Adrian Bunk
2012-08-28 17:23         ` Kent Overstreet
2012-08-22 17:04 ` [PATCH v6 09/13] block: Rename bio_split() -> bio_pair_split() Kent Overstreet
2012-08-22 17:04 ` [PATCH v6 10/13] block: Introduce new bio_split() Kent Overstreet
2012-08-22 20:46   ` Tejun Heo
2012-08-22 17:04 ` [PATCH v6 11/13] block: Rework bio_pair_split() Kent Overstreet
2012-08-22 21:04   ` Tejun Heo
2012-08-24  2:25     ` Martin K. Petersen
2012-08-24 10:37       ` Kent Overstreet
2012-08-24 20:58       ` Tejun Heo
2012-08-24 10:30     ` Kent Overstreet
2012-08-24 20:53       ` Tejun Heo
2012-08-22 17:04 ` [PATCH v6 12/13] block: Add bio_clone_bioset(), bio_clone_kmalloc() Kent Overstreet
2012-08-22 17:13   ` Jeff Garzik
2012-08-22 21:07   ` Tejun Heo
2012-08-24  6:24     ` Kent Overstreet
2012-08-24 20:36       ` Tejun Heo
2012-08-22 17:04 ` Kent Overstreet [this message]
2012-08-22 21:10   ` [PATCH v6 13/13] block: Only clone bio vecs that are in use Tejun Heo
2012-08-24  7:05     ` Kent Overstreet
2012-08-24 20:42       ` Tejun Heo
2012-08-23 18:00 ` [PATCH v6 00/13] Block cleanups Vivek Goyal
2012-08-24 12:46   ` Kent Overstreet

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=1345655050-28199-14-git-send-email-koverstreet@google.com \
    --to=koverstreet@google.com \
    --cc=agk@redhat.com \
    --cc=axboe@kernel.dk \
    --cc=bharrosh@panasas.com \
    --cc=dm-devel@redhat.com \
    --cc=linux-bcache@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mpatocka@redhat.com \
    --cc=sage@inktank.com \
    --cc=tj@kernel.org \
    --cc=vgoyal@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

Powered by JetHome