public inbox for linux-block@vger.kernel.org
 help / color / mirror / Atom feed
From: NeilBrown <neilb@suse.com>
To: Jens Axboe <axboe@fb.com>
Cc: linux-block@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: [PATCH 06/13] rbd: use bio_clone_fast() instead of bio_clone()
Date: Sun, 18 Jun 2017 14:38:58 +1000	[thread overview]
Message-ID: <149776073802.23258.14608712010143858524.stgit@noble> (raw)
In-Reply-To: <149776047907.23258.8058071140236879834.stgit@noble>

bio_clone() makes a copy of the bi_io_vec, but rbd never changes that,
so there is no need for a copy.
bio_clone_fast() can be used instead, which avoids making the copy.

This requires that we provide a bio_set.  bio_clone() uses fs_bio_set,
but it isn't, in general, safe to use the same bio_set at different
levels of the stack, as that can lead to deadlocks.  As filesystems
use fs_bio_set, block devices shouldn't.

As rbd never stacks, it is safe to have a single global bio_set for
all rbd devices to use.  So allocate that when the module is
initialised, and use it with bio_clone_fast().

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: NeilBrown <neilb@suse.com>
---
 drivers/block/rbd.c |   16 +++++++++++++++-
 1 file changed, 15 insertions(+), 1 deletion(-)

diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c
index 5420bc40c544..b008b6a98098 100644
--- a/drivers/block/rbd.c
+++ b/drivers/block/rbd.c
@@ -442,6 +442,8 @@ static DEFINE_SPINLOCK(rbd_client_list_lock);
 static struct kmem_cache	*rbd_img_request_cache;
 static struct kmem_cache	*rbd_obj_request_cache;
 
+static struct bio_set		*rbd_bio_clone;
+
 static int rbd_major;
 static DEFINE_IDA(rbd_dev_id_ida);
 
@@ -1363,7 +1365,7 @@ static struct bio *bio_clone_range(struct bio *bio_src,
 {
 	struct bio *bio;
 
-	bio = bio_clone(bio_src, gfpmask);
+	bio = bio_clone_fast(bio_src, gfpmask, rbd_bio_clone);
 	if (!bio)
 		return NULL;	/* ENOMEM */
 
@@ -6416,8 +6418,16 @@ static int rbd_slab_init(void)
 	if (!rbd_obj_request_cache)
 		goto out_err;
 
+	rbd_assert(!rbd_bio_clone);
+	rbd_bio_clone = bioset_create(BIO_POOL_SIZE, 0, 0);
+	if (!rbd_bio_clone)
+		goto out_err_clone;
+
 	return 0;
 
+out_err_clone:
+	kmem_cache_destroy(rbd_obj_request_cache);
+	rbd_obj_request_cache = NULL;
 out_err:
 	kmem_cache_destroy(rbd_img_request_cache);
 	rbd_img_request_cache = NULL;
@@ -6433,6 +6443,10 @@ static void rbd_slab_exit(void)
 	rbd_assert(rbd_img_request_cache);
 	kmem_cache_destroy(rbd_img_request_cache);
 	rbd_img_request_cache = NULL;
+
+	rbd_assert(rbd_bio_clone);
+	bioset_free(rbd_bio_clone);
+	rbd_bio_clone = NULL;
 }
 
 static int __init rbd_init(void)

  parent reply	other threads:[~2017-06-18  4:40 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-06-18  4:38 [PATCH 00/13] block: assorted cleanup for bio splitting and cloning NeilBrown
2017-06-18  4:38 ` [PATCH 01/13] blk: remove bio_set arg from blk_queue_split() NeilBrown
2017-06-18  4:38 ` [PATCH 04/13] blk: use non-rescuing bioset for q->bio_split NeilBrown
2017-06-18  4:38 ` [PATCH 03/13] blk: make the bioset rescue_workqueue optional NeilBrown
2017-06-18  4:38 ` [PATCH 02/13] blk: replace bioset_create_nobvec() with a flags arg to bioset_create() NeilBrown
2017-06-18  4:38 ` [PATCH 07/13] drbd: use bio_clone_fast() instead of bio_clone() NeilBrown
2017-06-18  4:38 ` [PATCH 05/13] block: Improvements to bounce-buffer handling NeilBrown
2017-06-18  4:38 ` [PATCH 08/13] pktcdvd: use bio_clone_fast() instead of bio_clone() NeilBrown
2017-06-18  4:38 ` [PATCH 09/13] lightnvm/pblk-read: use bio_clone_fast() NeilBrown
2017-06-18  4:38 ` NeilBrown [this message]
2017-06-18  4:38 ` [PATCH 12/13] block: remove bio_clone() and all references NeilBrown
2017-06-18  4:38 ` [PATCH 10/13] xen-blkfront: remove bio splitting NeilBrown
2017-06-18  4:38 ` [PATCH 13/13] block: don't check for BIO_MAX_PAGES in blk_bio_segment_split() NeilBrown
2017-06-18  4:38 ` [PATCH 11/13] bcache: use kmalloc to allocate bio in bch_data_verify() NeilBrown
2017-06-18 18:41 ` [PATCH 00/13] block: assorted cleanup for bio splitting and cloning Jens Axboe
2017-06-18 21:36   ` NeilBrown
2017-11-20 16:43     ` Mike Snitzer
2017-11-21  0:34       ` [dm-devel] " NeilBrown
2017-11-21  1:35         ` Mike Snitzer
2017-11-21 12:10           ` Mike Snitzer
2017-11-21 12:43             ` Mike Snitzer
2017-11-21 19:47               ` new patchset to eliminate DM's use of BIOSET_NEED_RESCUER [was: Re: [PATCH 00/13] block: assorted cleanup for bio splitting and cloning.] Mike Snitzer
2017-11-21 21:23                 ` [dm-devel] " Mikulas Patocka
2017-11-21 22:51                   ` new patchset to eliminate DM's use of BIOSET_NEED_RESCUER Mike Snitzer
2017-11-22  1:21                     ` Mikulas Patocka
2017-11-22  2:32                       ` Mike Snitzer
2017-11-22  4:00                       ` [dm-devel] " NeilBrown
2017-11-22  4:28                         ` Mike Snitzer
2017-11-22 21:18                           ` Mike Snitzer
2017-11-22 18:24                         ` [dm-devel] " Mikulas Patocka
2017-11-22 18:49                           ` Mike Snitzer
2017-11-23  5:12                           ` [dm-devel] " NeilBrown
2017-11-23 22:52                           ` [PATCH] dm: use cloned bio as head, not remainder, in __split_and_process_bio() NeilBrown
2017-11-27 14:23                             ` Mike Snitzer
2017-11-28 22:18                               ` [dm-devel] " NeilBrown
2017-11-21 23:03                   ` [dm-devel] new patchset to eliminate DM's use of BIOSET_NEED_RESCUER [was: Re: [PATCH 00/13] block: assorted cleanup for bio splitting and cloning.] NeilBrown
2017-11-21 19:44             ` [dm-devel] [PATCH 00/13] block: assorted cleanup for bio splitting and cloning NeilBrown
2017-11-21 19:50               ` Mike Snitzer
  -- strict thread matches above, loose matches on Subject: below --
2017-05-02  3:42 NeilBrown
2017-05-02  3:42 ` [PATCH 06/13] rbd: use bio_clone_fast() instead of bio_clone() NeilBrown

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=149776073802.23258.14608712010143858524.stgit@noble \
    --to=neilb@suse.com \
    --cc=axboe@fb.com \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-kernel@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