linux-nilfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jan Kara <jack@suse.cz>
To: linux-mm@kvack.org
Cc: cluster-devel@redhat.com, linux-nilfs@vger.kernel.org,
	tytso@mit.edu, linux-xfs@vger.kernel.org, "Yan,
	Zheng" <zyan@redhat.com>, Jan Kara <jack@suse.cz>,
	"Darrick J . Wong" <darrick.wong@oracle.com>,
	Hugh Dickins <hughd@google.com>,
	linux-f2fs-devel@lists.sourceforge.net,
	David Howells <dhowells@redhat.com>,
	David Sterba <dsterba@suse.com>,
	ceph-devel@vger.kernel.org,
	Nadia Yvette Chambers <nyc@holomorphy.com>,
	Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>,
	Jaegeuk Kim <jaegeuk@kernel.org>,
	Ilya Dryomov <idryomov@gmail.com>,
	linux-ext4@vger.kernel.org, linux-afs@lists.infradead.org,
	linux-btrfs@vger.kernel.org
Subject: [PATCH 08/35] fs: Fix performance regression in clean_bdev_aliases()
Date: Thu,  1 Jun 2017 11:32:18 +0200	[thread overview]
Message-ID: <20170601093245.29238-9-jack@suse.cz> (raw)
In-Reply-To: <20170601093245.29238-1-jack@suse.cz>

Commit e64855c6cfaa "fs: Add helper to clean bdev aliases under a bh and
use it" added a wrapper for clean_bdev_aliases() that invalidates bdev
aliases underlying a single buffer head. However this has caused a
performance regression for bonnie++ benchmark on ext4 filesystem when
delayed allocation is turned off (ext3 mode) - average of 3 runs:

Hmean SeqOut Char  164787.55 (  0.00%) 107189.06 (-34.95%)
Hmean SeqOut Block 219883.89 (  0.00%) 168870.32 (-23.20%)

The reason for this regression is that clean_bdev_aliases() is slower
when called for a single block because pagevec_lookup() it uses will end
up iterating through the radix tree until it finds a page (which may
take a while) but we are only interested whether there's a page at a
particular index.

Fix the problem by using pagevec_lookup_range() instead which avoids the
needless iteration.

Fixes: e64855c6cfaa0a80c1b71c5f647cb792dc436668
Signed-off-by: Jan Kara <jack@suse.cz>
---
 fs/buffer.c | 14 ++++++++------
 1 file changed, 8 insertions(+), 6 deletions(-)

diff --git a/fs/buffer.c b/fs/buffer.c
index fe0ee01c5a44..d63b22e50f38 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -1632,19 +1632,18 @@ void clean_bdev_aliases(struct block_device *bdev, sector_t block, sector_t len)
 	struct pagevec pvec;
 	pgoff_t index = block >> (PAGE_SHIFT - bd_inode->i_blkbits);
 	pgoff_t end;
-	int i;
+	int i, count;
 	struct buffer_head *bh;
 	struct buffer_head *head;
 
 	end = (block + len - 1) >> (PAGE_SHIFT - bd_inode->i_blkbits);
 	pagevec_init(&pvec, 0);
-	while (index <= end && pagevec_lookup(&pvec, bd_mapping, &index,
-			min(end - index, (pgoff_t)PAGEVEC_SIZE - 1) + 1)) {
-		for (i = 0; i < pagevec_count(&pvec); i++) {
+	while (pagevec_lookup_range(&pvec, bd_mapping, &index, end,
+				    PAGEVEC_SIZE)) {
+		count = pagevec_count(&pvec);
+		for (i = 0; i < count; i++) {
 			struct page *page = pvec.pages[i];
 
-			if (page->index > end)
-				break;
 			if (!page_has_buffers(page))
 				continue;
 			/*
@@ -1674,6 +1673,9 @@ void clean_bdev_aliases(struct block_device *bdev, sector_t block, sector_t len)
 		}
 		pagevec_release(&pvec);
 		cond_resched();
+		/* End of range already reached? */
+		if (index > end || !index)
+			break;
 	}
 }
 EXPORT_SYMBOL(clean_bdev_aliases);
-- 
2.12.3


  parent reply	other threads:[~2017-06-01  9:32 UTC|newest]

Thread overview: 43+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-06-01  9:32 [PATCH 00/35 v1] pagevec API cleanups Jan Kara
2017-06-01  9:32 ` [PATCH 01/35] fscache: Remove unused ->now_uncached callback Jan Kara
2017-06-01  9:32 ` [PATCH 02/35] ext4: Fix SEEK_HOLE Jan Kara
     [not found] ` <20170601093245.29238-1-jack-AlSwsSmVLrQ@public.gmane.org>
2017-06-01  9:32   ` [PATCH 03/35] ext4: Fix off-by-in in loop termination in ext4_find_unwritten_pgoff() Jan Kara
2017-06-01  9:32 ` [PATCH 04/35] dax: Fix inefficiency in dax_writeback_mapping_range() Jan Kara
2017-06-01  9:32 ` [PATCH 05/35] mm: Fix THP handling in invalidate_mapping_pages() Jan Kara
2017-06-01  9:32 ` [PATCH 06/35] mm: Make pagevec_lookup() update index Jan Kara
2017-06-01  9:32 ` [PATCH 07/35] mm: Implement find_get_pages_range() Jan Kara
2017-06-01  9:32 ` Jan Kara [this message]
2017-06-01  9:32 ` [PATCH 09/35] ext4: Use pagevec_lookup_range() in ext4_find_unwritten_pgoff() Jan Kara
2017-06-01  9:32 ` [PATCH 10/35] ext4: Use pagevec_lookup_range() in writeback code Jan Kara
2017-06-01  9:32 ` [PATCH 11/35] hugetlbfs: Use pagevec_lookup_range() in remove_inode_hugepages() Jan Kara
2017-06-01  9:32 ` [PATCH 12/35] xfs: Use pagevec_lookup_range() in xfs_find_get_desired_pgoff() Jan Kara
2017-06-01  9:32 ` [PATCH 13/35] mm: Remove nr_pages argument from pagevec_lookup{, _range}() Jan Kara
2017-06-01  9:32 ` [PATCH 14/35] mm: Implement find_get_pages_range_tag() Jan Kara
2017-06-01  9:32 ` [PATCH 15/35] btrfs: Use pagevec_lookup_range_tag() Jan Kara
2017-06-01  9:32 ` [PATCH 16/35] ceph: " Jan Kara
2017-06-01  9:32 ` [PATCH 17/35] ext4: " Jan Kara
2017-06-01  9:32 ` [PATCH 18/35] f2fs: " Jan Kara
2017-06-01  9:32 ` [PATCH 19/35] f2fs: Simplify page iteration loops Jan Kara
2017-06-01 13:00   ` kbuild test robot
2017-06-01  9:32 ` [PATCH 20/35] f2fs: Use find_get_pages_tag() for looking up single page Jan Kara
2017-06-01  9:32 ` [PATCH 21/35] gfs2: Use pagevec_lookup_range_tag() Jan Kara
2017-06-01  9:32 ` [PATCH 22/35] nilfs2: " Jan Kara
2017-06-01  9:32 ` [PATCH 23/35] mm: Use pagevec_lookup_range_tag() in __filemap_fdatawait_range() Jan Kara
2017-06-01  9:32 ` [PATCH 24/35] mm: Use pagevec_lookup_range_tag() in write_cache_pages() Jan Kara
2017-06-01  9:32 ` [PATCH 25/35] mm: Remove nr_pages argument from pagevec_lookup_{, range}_tag() Jan Kara
2017-06-01  9:32 ` [PATCH 26/35] afs: Use find_get_pages_range_tag() Jan Kara
2017-06-01  9:32 ` [PATCH 27/35] shmem: Use pagevec_lookup() in shmem_unlock_mapping() Jan Kara
2017-06-01  9:32 ` [PATCH 28/35] shmem: Use pagevec_lookup_entries() Jan Kara
2017-06-01  9:32 ` [PATCH 29/35] mm: Make pagevec_lookup_entries() update index Jan Kara
2017-06-01  9:32 ` [PATCH 30/35] mm: Implement find_get_entries_range() Jan Kara
2017-06-01  9:32 ` [PATCH 31/35] shmem: Convert to pagevec_lookup_entries_range() Jan Kara
2017-06-01 13:55   ` kbuild test robot
2017-06-01  9:32 ` [PATCH 32/35] mm: Convert truncate code " Jan Kara
2017-06-01  9:32 ` [PATCH 33/35] mm: Remove nr_entries argument from pagevec_lookup_entries{, _range}() Jan Kara
2017-06-01  9:32 ` [PATCH 34/35] mm: Make find_get_entries_tag() update index Jan Kara
2017-06-01  9:32 ` [PATCH 35/35] mm: Implement find_get_entries_range_tag() Jan Kara
2017-06-01 10:26 ` [PATCH 01/35] fscache: Remove unused ->now_uncached callback David Howells
2017-06-01 11:34   ` Jan Kara
2017-06-19 13:12     ` Jan Kara
2017-06-01 11:36 ` [Cluster-devel] [PATCH 00/35 v1] pagevec API cleanups Christoph Hellwig
2017-06-01 12:05   ` Jan Kara

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=20170601093245.29238-9-jack@suse.cz \
    --to=jack@suse.cz \
    --cc=ceph-devel@vger.kernel.org \
    --cc=cluster-devel@redhat.com \
    --cc=darrick.wong@oracle.com \
    --cc=dhowells@redhat.com \
    --cc=dsterba@suse.com \
    --cc=hughd@google.com \
    --cc=idryomov@gmail.com \
    --cc=jaegeuk@kernel.org \
    --cc=konishi.ryusuke@lab.ntt.co.jp \
    --cc=linux-afs@lists.infradead.org \
    --cc=linux-btrfs@vger.kernel.org \
    --cc=linux-ext4@vger.kernel.org \
    --cc=linux-f2fs-devel@lists.sourceforge.net \
    --cc=linux-mm@kvack.org \
    --cc=linux-nilfs@vger.kernel.org \
    --cc=linux-xfs@vger.kernel.org \
    --cc=nyc@holomorphy.com \
    --cc=tytso@mit.edu \
    --cc=zyan@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;
as well as URLs for NNTP newsgroup(s).