Linux-Raid Archives on lore.kernel.org
 help / color / Atom feed
From: Christoph Hellwig <hch@lst.de>
To: Jens Axboe <axboe@kernel.dk>
Cc: Song Liu <song@kernel.org>,
	"Martin K. Petersen" <martin.petersen@oracle.com>,
	linux-block@vger.kernel.org, linux-raid@vger.kernel.org
Subject: [PATCH 04/11] block: streamline bvec_alloc
Date: Tue,  2 Feb 2021 18:19:22 +0100
Message-ID: <20210202171929.1504939-5-hch@lst.de> (raw)
In-Reply-To: <20210202171929.1504939-1-hch@lst.de>

Avoid the pointless goto by trying the slab allocation first and falling
through to the mempool.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 block/bio.c | 26 ++++++++++----------------
 1 file changed, 10 insertions(+), 16 deletions(-)

diff --git a/block/bio.c b/block/bio.c
index c2152c4bf8a317..321b3479a154da 100644
--- a/block/bio.c
+++ b/block/bio.c
@@ -172,8 +172,6 @@ static inline gfp_t bvec_alloc_gfp(gfp_t gfp)
 struct bio_vec *bvec_alloc(gfp_t gfp_mask, int nr, unsigned long *idx,
 			   mempool_t *pool)
 {
-	struct bio_vec *bvl;
-
 	/*
 	 * see comment near bvec_array define!
 	 */
@@ -201,28 +199,24 @@ struct bio_vec *bvec_alloc(gfp_t gfp_mask, int nr, unsigned long *idx,
 	}
 
 	/*
-	 * idx now points to the pool we want to allocate from. only the
-	 * 1-vec entry pool is mempool backed.
+	 * Try a slab allocation first for all smaller allocations.  If that
+	 * fails and __GFP_DIRECT_RECLAIM is set retry with the mempool.
+	 * The mempool is sized to handle up to BIO_MAX_PAGES entries.
 	 */
-	if (*idx == BVEC_POOL_MAX) {
-fallback:
-		bvl = mempool_alloc(pool, gfp_mask);
-	} else {
+	if (*idx < BVEC_POOL_MAX) {
 		struct biovec_slab *bvs = bvec_slabs + *idx;
+		struct bio_vec *bvl;
 
-		/*
-		 * Try a slab allocation. If this fails and __GFP_DIRECT_RECLAIM
-		 * is set, retry with the 1-entry mempool
-		 */
 		bvl = kmem_cache_alloc(bvs->slab, bvec_alloc_gfp(gfp_mask));
-		if (unlikely(!bvl && (gfp_mask & __GFP_DIRECT_RECLAIM))) {
-			*idx = BVEC_POOL_MAX;
-			goto fallback;
+		if (likely(bvl) || !(gfp_mask & __GFP_DIRECT_RECLAIM)) {
+			(*idx)++;
+			return bvl;
 		}
+		*idx = BVEC_POOL_MAX;
 	}
 
 	(*idx)++;
-	return bvl;
+	return mempool_alloc(pool, gfp_mask);
 }
 
 void bio_uninit(struct bio *bio)
-- 
2.29.2


  parent reply index

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-02-02 17:19 cleanup bvec allocation Christoph Hellwig
2021-02-02 17:19 ` [PATCH 01/11] block: reuse BIO_INLINE_VECS for integrity bvecs Christoph Hellwig
2021-02-02 17:19 ` [PATCH 02/11] block: move struct biovec_slab to bio.c Christoph Hellwig
2021-02-02 17:19 ` [PATCH 03/11] block: factor out a bvec_alloc_gfp helper Christoph Hellwig
2021-02-02 17:19 ` Christoph Hellwig [this message]
2021-02-02 17:19 ` [PATCH 05/11] block: remove the 1 and 4 vec bvec_slabs entries Christoph Hellwig
2021-02-02 17:19 ` [PATCH 06/11] block: turn the nr_iovecs argument to bio_alloc* into an unsigned short Christoph Hellwig
2021-02-02 17:19 ` [PATCH 07/11] block: remove a layer of indentation in bio_iov_iter_get_pages Christoph Hellwig
2021-02-02 17:19 ` [PATCH 08/11] block: set BIO_NO_PAGE_REF in bio_iov_bvec_set Christoph Hellwig
2021-02-02 17:19 ` [PATCH 09/11] block: mark the bio as cloned " Christoph Hellwig
2021-02-02 17:19 ` [PATCH 10/11] md/raid10: remove dead code in reshape_request Christoph Hellwig
2021-02-04 17:57   ` Song Liu
2021-02-02 17:19 ` [PATCH 11/11] block: use bi_max_vecs to find the bvec pool Christoph Hellwig
2021-02-07 16:21 ` cleanup bvec allocation Christoph Hellwig
2021-02-08 15:33 ` Jens Axboe

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=20210202171929.1504939-5-hch@lst.de \
    --to=hch@lst.de \
    --cc=axboe@kernel.dk \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-raid@vger.kernel.org \
    --cc=martin.petersen@oracle.com \
    --cc=song@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

Linux-Raid Archives on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-raid/0 linux-raid/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-raid linux-raid/ https://lore.kernel.org/linux-raid \
		linux-raid@vger.kernel.org
	public-inbox-index linux-raid

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-raid


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git