mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Ming Lei <tom.leiming@gmail.com>
To: Jens Axboe <axboe@fb.com>, linux-kernel@vger.kernel.org
Cc: linux-block@vger.kernel.org, linux-fsdevel@vger.kernel.org,
	Christoph Hellwig <hch@infradead.org>,
	"Kirill A . Shutemov" <kirill.shutemov@linux.intel.com>,
	Ming Lei <tom.leiming@gmail.com>, Jens Axboe <axboe@kernel.dk>
Subject: [PATCH 43/60] block: use bio_for_each_segment_mp() to map sg
Date: Sat, 29 Oct 2016 16:08:42 +0800	[thread overview]
Message-ID: <1477728600-12938-44-git-send-email-tom.leiming@gmail.com> (raw)
In-Reply-To: <1477728600-12938-1-git-send-email-tom.leiming@gmail.com>

It is more efficient to use bio_for_each_segment_mp()
for mapping sg, meantime we have to consider splitting
multipage bvec as done in blk_bio_segment_split().

Signed-off-by: Ming Lei <tom.leiming@gmail.com>
---
 block/blk-merge.c | 72 +++++++++++++++++++++++++++++++++++++++----------------
 1 file changed, 52 insertions(+), 20 deletions(-)

diff --git a/block/blk-merge.c b/block/blk-merge.c
index 9142f1fc914b..e3b8cbc8b675 100644
--- a/block/blk-merge.c
+++ b/block/blk-merge.c
@@ -442,6 +442,56 @@ static int blk_phys_contig_segment(struct request_queue *q, struct bio *bio,
 	return 0;
 }
 
+static inline struct scatterlist *blk_next_sg(struct scatterlist **sg,
+		struct scatterlist *sglist)
+{
+	if (!*sg)
+		return sglist;
+	else {
+		/*
+		 * If the driver previously mapped a shorter
+		 * list, we could see a termination bit
+		 * prematurely unless it fully inits the sg
+		 * table on each mapping. We KNOW that there
+		 * must be more entries here or the driver
+		 * would be buggy, so force clear the
+		 * termination bit to avoid doing a full
+		 * sg_init_table() in drivers for each command.
+		 */
+		sg_unmark_end(*sg);
+		return sg_next(*sg);
+	}
+}
+
+static inline unsigned
+blk_bvec_map_sg(struct request_queue *q, struct bio_vec *bvec,
+		struct scatterlist *sglist, struct scatterlist **sg)
+{
+	unsigned nbytes = bvec->bv_len;
+	unsigned nsegs = 0, total = 0;
+
+	while (nbytes > 0) {
+		unsigned seg_size;
+		struct page *pg;
+		unsigned offset, idx;
+
+		*sg = blk_next_sg(sg, sglist);
+
+		seg_size = min(nbytes, queue_max_segment_size(q));
+		offset = (total + bvec->bv_offset) % PAGE_SIZE;
+		idx = (total + bvec->bv_offset) / PAGE_SIZE;
+		pg = nth_page(bvec->bv_page, idx);
+
+		sg_set_page(*sg, pg, seg_size, offset);
+
+		total += seg_size;
+		nbytes -= seg_size;
+		nsegs++;
+	}
+
+	return nsegs;
+}
+
 static inline void
 __blk_segment_map_sg(struct request_queue *q, struct bio_vec *bvec,
 		     struct scatterlist *sglist, struct bio_vec *bvprv,
@@ -475,25 +525,7 @@ __blk_segment_map_sg(struct request_queue *q, struct bio_vec *bvec,
 		(*sg)->length += nbytes;
 	} else {
 new_segment:
-		if (!*sg)
-			*sg = sglist;
-		else {
-			/*
-			 * If the driver previously mapped a shorter
-			 * list, we could see a termination bit
-			 * prematurely unless it fully inits the sg
-			 * table on each mapping. We KNOW that there
-			 * must be more entries here or the driver
-			 * would be buggy, so force clear the
-			 * termination bit to avoid doing a full
-			 * sg_init_table() in drivers for each command.
-			 */
-			sg_unmark_end(*sg);
-			*sg = sg_next(*sg);
-		}
-
-		sg_set_page(*sg, bvec->bv_page, nbytes, bvec->bv_offset);
-		(*nsegs)++;
+		(*nsegs) += blk_bvec_map_sg(q, bvec, sglist, sg);
 
 		/* for making iterator happy */
 		bvec->bv_offset -= advance;
@@ -536,7 +568,7 @@ static int __blk_bios_map_sg(struct request_queue *q, struct bio *bio,
 	}
 
 	for_each_bio(bio)
-		bio_for_each_segment(bvec, bio, iter)
+		bio_for_each_segment_mp(bvec, bio, iter)
 			__blk_segment_map_sg(q, &bvec, sglist, &bvprv, sg,
 					     &nsegs, &cluster);
 
-- 
2.7.4

  parent reply	other threads:[~2016-10-29  8:15 UTC|newest]

Thread overview: 104+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <1477728600-12938-1-git-send-email-tom.leiming@gmail.com>
2016-10-29  8:08 ` [PATCH 01/60] block: bio: introduce bio_init_with_vec_table() Ming Lei
2016-10-29 15:21   ` Christoph Hellwig
2016-10-29  8:08 ` [PATCH 02/60] block drivers: convert to bio_init_with_vec_table() Ming Lei
2016-10-29  8:08 ` [PATCH 03/60] block: drbd: remove impossible failure handling Ming Lei
2016-10-31 15:25   ` Christoph Hellwig
2016-10-29  8:08 ` [PATCH 04/60] block: floppy: use bio_add_page() Ming Lei
2016-10-31 15:26   ` Christoph Hellwig
2016-10-31 22:54     ` Ming Lei
2016-10-29  8:08 ` [PATCH 05/60] target: avoid to access .bi_vcnt directly Ming Lei
2016-10-31 15:26   ` Christoph Hellwig
2016-10-29  8:08 ` [PATCH 06/60] bcache: debug: avoid to access .bi_io_vec directly Ming Lei
2016-10-29  8:08 ` [PATCH 07/60] dm: crypt: use bio_add_page() Ming Lei
2016-10-29  8:08 ` [PATCH 08/60] dm: use bvec iterator helpers to implement .get_page and .next_page Ming Lei
2016-10-29  8:08 ` [PATCH 09/60] dm: dm.c: replace 'bio->bi_vcnt == 1' with !bio_multiple_segments Ming Lei
2016-10-31 15:29   ` Christoph Hellwig
2016-10-31 22:59     ` Ming Lei
2016-11-02  3:09     ` Kent Overstreet
2016-11-02  7:56       ` Ming Lei
2016-11-02 14:24         ` Mike Snitzer
2016-11-02 23:47           ` Ming Lei
2016-10-29  8:08 ` [PATCH 10/60] fs: logfs: convert to bio_add_page() in sync_request() Ming Lei
2016-10-29  8:08 ` [PATCH 11/60] fs: logfs: use bio_add_page() in __bdev_writeseg() Ming Lei
2016-10-31 15:29   ` Christoph Hellwig
2016-10-29  8:08 ` [PATCH 12/60] fs: logfs: use bio_add_page() in do_erase() Ming Lei
2016-10-31 15:29   ` Christoph Hellwig
2016-10-29  8:08 ` [PATCH 13/60] fs: logfs: remove unnecesary check Ming Lei
2016-10-31 15:29   ` Christoph Hellwig
2016-10-29  8:08 ` [PATCH 14/60] block: drbd: comment on direct access bvec table Ming Lei
2016-10-29  8:08 ` [PATCH 15/60] block: loop: comment on direct access to " Ming Lei
2016-10-31 15:31   ` Christoph Hellwig
2016-10-31 23:08     ` Ming Lei
2016-10-29  8:08 ` [PATCH 16/60] block: pktcdvd: " Ming Lei
2016-10-31 15:33   ` Christoph Hellwig
2016-10-31 23:08     ` Ming Lei
2016-10-29  8:08 ` [PATCH 17/60] kernel/power/swap.c: " Ming Lei
2016-10-29  8:08 ` [PATCH 18/60] mm: page_io.c: " Ming Lei
2016-10-29  8:08 ` [PATCH 19/60] fs/buffer: " Ming Lei
2016-10-31 15:35   ` Christoph Hellwig
2016-10-31 23:12     ` Ming Lei
2016-10-29  8:08 ` [PATCH 20/60] f2fs: f2fs_read_end_io: " Ming Lei
2016-10-29  8:08 ` [PATCH 21/60] bcache: " Ming Lei
2016-10-29  8:08 ` [PATCH 22/60] block: comment on bio_alloc_pages() Ming Lei
2016-10-29  8:08 ` [PATCH 23/60] block: introduce flag QUEUE_FLAG_NO_MP Ming Lei
2016-10-29 15:29   ` Christoph Hellwig
2016-10-29 22:20     ` Ming Lei
2016-10-29  8:08 ` [PATCH 24/60] md: set NO_MP for request queue of md Ming Lei
2016-10-29  8:08 ` [PATCH 25/60] block: pktcdvd: set NO_MP for pktcdvd request queue Ming Lei
2016-10-29  8:08 ` [PATCH 26/60] btrfs: set NO_MP for request queues behind BTRFS Ming Lei
2016-10-31 15:36   ` Christoph Hellwig
2016-10-31 17:58     ` Chris Mason
2016-10-31 18:00       ` Christoph Hellwig
2016-10-29  8:08 ` [PATCH 27/60] block: introduce BIO_SP_MAX_SECTORS Ming Lei
2016-10-29  8:08 ` [PATCH 28/60] block: introduce QUEUE_FLAG_SPLIT_MP Ming Lei
2016-10-31 15:39   ` Christoph Hellwig
2016-10-31 23:56     ` Ming Lei
2016-11-02  3:08     ` Kent Overstreet
2016-11-03 10:38       ` Ming Lei
2016-11-03 11:20         ` Kent Overstreet
2016-11-03 11:26           ` Ming Lei
2016-11-03 11:30             ` Kent Overstreet
2016-10-29  8:08 ` [PATCH 29/60] dm: limit the max bio size as BIO_SP_MAX_SECTORS << SECTOR_SHIFT Ming Lei
2016-10-29  8:08 ` [PATCH 30/60] bcache: set flag of QUEUE_FLAG_SPLIT_MP Ming Lei
2016-10-29  8:08 ` [PATCH 31/60] block: introduce multipage/single page bvec helpers Ming Lei
2016-10-29  8:08 ` [PATCH 32/60] block: implement sp version of bvec iterator helpers Ming Lei
2016-10-29 11:06   ` kbuild test robot
2016-12-17 11:38     ` Ming Lei
2016-10-29  8:08 ` [PATCH 33/60] block: introduce bio_for_each_segment_mp() Ming Lei
2016-10-29  8:08 ` [PATCH 34/60] block: introduce bio_clone_sp() Ming Lei
2016-10-29  8:08 ` [PATCH 35/60] bvec_iter: introduce BVEC_ITER_ALL_INIT Ming Lei
2016-10-29  8:08 ` [PATCH 36/60] block: bounce: avoid direct access to bvec from bio->bi_io_vec Ming Lei
2016-10-29  8:08 ` [PATCH 37/60] block: bounce: don't access bio->bi_io_vec in copy_to_high_bio_irq Ming Lei
2016-10-29  8:08 ` [PATCH 38/60] block: bounce: convert multipage bvecs into singlepage Ming Lei
2016-10-29  8:08 ` [PATCH 39/60] bcache: debug: switch to bio_clone_sp() Ming Lei
2016-10-29  8:08 ` [PATCH 40/60] blk-merge: compute bio->bi_seg_front_size efficiently Ming Lei
2016-10-29  8:08 ` [PATCH 41/60] block: blk-merge: try to make front segments in full size Ming Lei
2016-10-29  8:08 ` [PATCH 42/60] block: use bio_for_each_segment_mp() to compute segments count Ming Lei
2016-10-29  8:08 ` Ming Lei [this message]
2016-10-29  8:08 ` [PATCH 44/60] block: introduce bvec_for_each_sp_bvec() Ming Lei
2016-10-29  8:08 ` [PATCH 45/60] block: bio: introduce bio_for_each_segment_all_rd() and its write pair Ming Lei
2016-10-31 13:59   ` Theodore Ts'o
2016-10-31 15:11     ` Christoph Hellwig
2016-10-31 22:50       ` Ming Lei
2016-11-02  3:01       ` Kent Overstreet
2016-10-31 22:46     ` Ming Lei
2016-10-31 23:51       ` Ming Lei
2016-11-01 14:17         ` Theodore Ts'o
2016-11-02  1:58           ` Ming Lei
2016-10-29  8:08 ` [PATCH 46/60] block: deal with dirtying pages for multipage bvec Ming Lei
2016-10-31 15:40   ` Christoph Hellwig
2016-11-01  0:19     ` Ming Lei
2016-10-29  8:08 ` [PATCH 47/60] block: convert to bio_for_each_segment_all_rd() Ming Lei
2016-10-29  8:08 ` [PATCH 48/60] fs/mpage: " Ming Lei
2016-10-29  8:08 ` [PATCH 49/60] fs/direct-io: " Ming Lei
2016-10-29  8:08 ` [PATCH 50/60] ext4: " Ming Lei
2016-10-29  8:08 ` [PATCH 51/60] xfs: " Ming Lei
2016-10-29  8:08 ` [PATCH 52/60] logfs: " Ming Lei
2016-10-29  8:08 ` [PATCH 53/60] gfs2: " Ming Lei
2016-10-29  8:08 ` [PATCH 54/60] f2fs: " Ming Lei
2016-10-29  8:08 ` [PATCH 55/60] exofs: " Ming Lei
2016-10-29  8:08 ` [PATCH 56/60] fs: crypto: " Ming Lei
2016-10-29  8:08 ` [PATCH 57/60] bcache: " Ming Lei
2016-10-29  8:08 ` [PATCH 58/60] dm-crypt: " Ming Lei
2016-10-29  8:08 ` [PATCH 59/60] fs/buffer.c: use bvec iterator to truncate the bio Ming Lei
2016-10-29  8:08 ` [PATCH 60/60] block: enable multipage bvecs Ming Lei

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=1477728600-12938-44-git-send-email-tom.leiming@gmail.com \
    --to=tom.leiming@gmail.com \
    --cc=axboe@fb.com \
    --cc=axboe@kernel.dk \
    --cc=hch@infradead.org \
    --cc=kirill.shutemov@linux.intel.com \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-fsdevel@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

Powered by JetHome