[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <1341404740.2979.91.camel@sauron.fi.intel.com>
Date: Wed, 04 Jul 2012 15:25:40 +0300
From: Artem Bityutskiy <dedekind1@...il.com>
To: Jan Kara <jack@...e.cz>
Cc: Theodore Tso <tytso@....edu>,
Linux Kernel Maling List <linux-kernel@...r.kernel.org>,
Ext4 Mailing List <linux-ext4@...r.kernel.org>,
Linux FS Maling List <linux-fsdevel@...r.kernel.org>
Subject: Re: [PATCHv3 3/4] ext4: remove unnecessary superblock dirtying
On Tue, 2012-07-03 at 14:48 +0200, Jan Kara wrote:
> > 1. When creating the first large file on a file
> > system without EXT4_FEATURE_RO_COMPAT_LARGE_FILE feature.
> > 2. When re-sizing the file-system
> > 3. When creating an xattr on a file-system without the
> > EXT4_FEATURE_COMPAT_EXT_ATTR feature.
> Since this was written, we seem to have grown a use of
> ext4_handle_dirty_super() in fs/ext4/namei.c when adding inode to orphan
> list.
Re-worked the commit message a bit in v4.
> > + err = ext4_commit_super(sb, 0);
> When you look at the whole __ext4_handle_dirty_super() function now, it
> doesn't make much sense like this. When now == 1, we mark superblock buffer
> as dirty, when now == 0, we set wtime, kbytes_written, free blocks and
> inodes in the superblock buffer and then mark it dirty. Looking into all
> the places calling this, I'd just drop the 'now' argument and make
> everything behave as in now == 1 case.
Done in v4 in patch N5. Thanks a lot for suggestions and the help!
--
Best Regards,
Artem Bityutskiy
Download attachment "signature.asc" of type "application/pgp-signature" (837 bytes)
Powered by blists - more mailing lists