From: Jan Kara Subject: [PATCH] ext3: Speedup WB_SYNC_ALL pass Date: Fri, 10 Oct 2014 16:23:12 +0200 Message-ID: <1412951028-4085-8-git-send-email-jack@suse.cz> References: <1412951028-4085-1-git-send-email-jack@suse.cz> Cc: linux-ext4@vger.kernel.org, Dave Chinner , xfs@oss.sgi.com, cluster-devel@redhat.com, Steven Whitehouse , Mark Fasheh , Joel Becker , ocfs2-devel@oss.oracle.com, reiserfs-devel@vger.kernel.org, Jeff Mahoney , Dave Kleikamp , jfs-discussion@lists.sourceforge.net, tytso@mit.edu, viro@zeniv.linux.org.uk, Jan Kara To: linux-fsdevel@vger.kernel.org Return-path: In-Reply-To: <1412951028-4085-1-git-send-email-jack@suse.cz> Sender: linux-fsdevel-owner@vger.kernel.org List-Id: linux-ext4.vger.kernel.org When doing filesystem wide sync, there's no need to force transaction commit separately for each inode because ext3_sync_fs() takes care of forcing commit at the end. Most of the time this slowness doesn't manifest because previous WB_SYNC_NONE writeback doesn't leave much to write but when there are processes aggressively creating new files and several filesystems to sync, the sync slowness can be noticeable. In the following test script sync(1) takes around 6 minutes when there are two ext3 filesystems mounted on a standard SATA drive. After this patch sync takes a couple of seconds so we have about two orders of magnitude improvement. function run_writers { for (( i = 0; i < 10; i++ )); do mkdir $1/dir$i for (( j = 0; j < 40000; j++ )); do dd if=/dev/zero of=$1/dir$i/$j bs=4k count=4 &>/dev/null done & done } for dir in "$@"; do run_writers $dir done sleep 40 time sync Signed-off-by: Jan Kara --- fs/ext3/inode.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index 384b6ebb655f..14c2faf3c312 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -3253,7 +3253,12 @@ int ext3_write_inode(struct inode *inode, struct writeback_control *wbc) return -EIO; } - if (wbc->sync_mode != WB_SYNC_ALL) + /* + * No need to force transaction in WB_SYNC_NONE mode. Also + * ext3_sync_fs() will force the commit after everything is + * written. + */ + if (wbc->sync_mode != WB_SYNC_ALL || wbc->for_sync) return 0; return ext3_force_commit(inode->i_sb); -- 1.8.1.4