linux-ext4.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jan Kara <jack@suse.cz>
To: Theodore Ts'o <tytso@mit.edu>
Cc: Ext4 Developers List <linux-ext4@vger.kernel.org>
Subject: Re: [PATCH 07/12] ext4: start handle at the last possible moment in ext4_unlink()
Date: Mon, 11 Feb 2013 17:21:33 +0100	[thread overview]
Message-ID: <20130211162133.GJ5318@quack.suse.cz> (raw)
In-Reply-To: <1360446832-12724-8-git-send-email-tytso@mit.edu>

On Sat 09-02-13 16:53:47, Ted Tso wrote:
> Don't start the jbd2 transaction handle until after the directory
> entry has been found, to minimize the amount of time that a handle is
> held active.
  Looks good. You can add:
Reviewed-by: Jan Kara <jack@suse.cz>

								Honza
> 
> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
> ---
>  fs/ext4/namei.c | 24 ++++++++++++++----------
>  1 file changed, 14 insertions(+), 10 deletions(-)
> 
> diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c
> index 5184103..4a27069 100644
> --- a/fs/ext4/namei.c
> +++ b/fs/ext4/namei.c
> @@ -2787,7 +2787,7 @@ static int ext4_unlink(struct inode *dir, struct dentry *dentry)
>  	struct inode *inode;
>  	struct buffer_head *bh;
>  	struct ext4_dir_entry_2 *de;
> -	handle_t *handle;
> +	handle_t *handle = NULL;
>  
>  	trace_ext4_unlink_enter(dir, dentry);
>  	/* Initialize quotas before so that eventual writes go
> @@ -2795,14 +2795,6 @@ static int ext4_unlink(struct inode *dir, struct dentry *dentry)
>  	dquot_initialize(dir);
>  	dquot_initialize(dentry->d_inode);
>  
> -	handle = ext4_journal_start(dir, EXT4_HT_DIR,
> -				    EXT4_DELETE_TRANS_BLOCKS(dir->i_sb));
> -	if (IS_ERR(handle))
> -		return PTR_ERR(handle);
> -
> -	if (IS_DIRSYNC(dir))
> -		ext4_handle_sync(handle);
> -
>  	retval = -ENOENT;
>  	bh = ext4_find_entry(dir, &dentry->d_name, &de, NULL);
>  	if (!bh)
> @@ -2814,6 +2806,17 @@ static int ext4_unlink(struct inode *dir, struct dentry *dentry)
>  	if (le32_to_cpu(de->inode) != inode->i_ino)
>  		goto end_unlink;
>  
> +	handle = ext4_journal_start(dir, EXT4_HT_DIR,
> +				    EXT4_DELETE_TRANS_BLOCKS(dir->i_sb));
> +	if (IS_ERR(handle)) {
> +		retval = PTR_ERR(handle);
> +		handle = NULL;
> +		goto end_unlink;
> +	}
> +
> +	if (IS_DIRSYNC(dir))
> +		ext4_handle_sync(handle);
> +
>  	if (!inode->i_nlink) {
>  		ext4_warning(inode->i_sb,
>  			     "Deleting nonexistent file (%lu), %d",
> @@ -2834,8 +2837,9 @@ static int ext4_unlink(struct inode *dir, struct dentry *dentry)
>  	retval = 0;
>  
>  end_unlink:
> -	ext4_journal_stop(handle);
>  	brelse(bh);
> +	if (handle)
> +		ext4_journal_stop(handle);
>  	trace_ext4_unlink_exit(dentry, retval);
>  	return retval;
>  }
> -- 
> 1.7.12.rc0.22.gcdd159b
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
-- 
Jan Kara <jack@suse.cz>
SUSE Labs, CR

  reply	other threads:[~2013-02-11 16:21 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-02-09 21:53 [PATCH 00/12] jbd2 optimization and bug fixes Theodore Ts'o
2013-02-09 21:53 ` [PATCH 01/12] jbd2: track request delay statistics Theodore Ts'o
2013-02-11 15:57   ` Jan Kara
2013-02-09 21:53 ` [PATCH 02/12] jbd2: revert "jbd2: add COW fields to struct jbd2_journal_handle" Theodore Ts'o
2013-02-11 15:58   ` Jan Kara
2013-02-09 21:53 ` [PATCH 03/12] jbd2: add tracepoints which provide per-handle statistics Theodore Ts'o
2013-02-09 21:53 ` [PATCH 04/12] ext4: move the jbd2 wrapper functions out of super.c Theodore Ts'o
2013-02-09 21:53 ` [PATCH 05/12] ext4: pass context information to jbd2__journal_start() Theodore Ts'o
2013-02-11 16:16   ` Jan Kara
2013-02-11 18:13     ` Theodore Ts'o
2013-02-11 19:58       ` Jan Kara
2013-02-11 20:14         ` Theodore Ts'o
2013-02-09 21:53 ` [PATCH 06/12] ext4: grab page before starting transaction handle in write_begin() Theodore Ts'o
2013-02-11 16:35   ` Jan Kara
2013-02-09 21:53 ` [PATCH 07/12] ext4: start handle at the last possible moment in ext4_unlink() Theodore Ts'o
2013-02-11 16:21   ` Jan Kara [this message]
2013-02-09 21:53 ` [PATCH 08/12] ext4: start handle at the last possible moment in ext4_rmdir() Theodore Ts'o
2013-02-11 16:22   ` Jan Kara
2013-02-09 21:53 ` [PATCH 09/12] ext4: fix the number of credits needed for ext4_ext_migrate() Theodore Ts'o
2013-02-11 16:26   ` Jan Kara
2013-02-09 21:53 ` [PATCH 10/12] ext4: fix the number of credits needed for ext4_unlink() and ext4_rmdir() Theodore Ts'o
2013-02-11 16:28   ` Jan Kara
2013-02-11 18:30     ` Theodore Ts'o
2013-02-11 19:30       ` Jan Kara
2013-02-09 21:53 ` [PATCH 11/12] ext4: fix the number of credits needed for acl ops with inline data Theodore Ts'o
2013-02-10 13:42   ` Tao Ma
2013-02-10 18:15     ` Shentino
2013-02-10 19:43       ` Theodore Ts'o
2013-02-11 16:16         ` Andreas Dilger
2013-02-11 16:30   ` Jan Kara
2013-02-09 21:53 ` [PATCH 12/12] ext4: start handle at the last possible moment when creating inodes Theodore Ts'o
2013-02-11  1:47   ` Theodore Ts'o
2013-02-11 16:00   ` Andreas Dilger
2013-02-11 15:52 ` [PATCH 00/12] jbd2 optimization and bug fixes 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=20130211162133.GJ5318@quack.suse.cz \
    --to=jack@suse.cz \
    --cc=linux-ext4@vger.kernel.org \
    --cc=tytso@mit.edu \
    /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).