[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20110217172523.GJ26780@thunk.org>
Date: Thu, 17 Feb 2011 12:25:23 -0500
From: Ted Ts'o <tytso@....edu>
To: Linus Torvalds <torvalds@...ux-foundation.org>
Cc: Steven Liu <lingjiujianke@...il.com>, viro@...iv.linux.org.uk,
randy.dunlap@...cle.com, linux-fsdevel@...r.kernel.org,
linux-kernel@...r.kernel.org, dchinner@...hat.com,
liuqi <liuqi@...nderst.com>
Subject: Re: [PATCH] code cleanup on fs/super.c
On Thu, Feb 17, 2011 at 08:20:31AM -0800, Linus Torvalds wrote:
> On Thu, Feb 17, 2011 at 12:59 AM, Steven Liu <lingjiujianke@...il.com> wrote:
> >
> > Clean up the unsed code on fs/super.c, all filesystem using mount_bdev
> > and mount_nodev to replace get_sb_bdev and get_sb_nodev.
>
> There might easily be external modules that still use this.
>
> Also, if you do this, then the same patch would also need to remove
> the declarations and the documentation entry, so that there is no sign
> at all of those functions in the whole kernel tree.
The documentation update should just state that previous users of
get_sb_bdev() and get_sb_bdev() can and should just use mount_bdev()
and mount_nodev() instead. The lines of code in the get_sb_*()
functions that do this:
mnt->mnt_root = root;
mnt->mnt_sb = root->d_sb;
Actually aren't necessary, since these functions are used in a
filesystem's file_system_type->mount() function, and vfs_kern_mount(),
which calls the type->mount() function, already does the above call.
So telling external modules that if you use get_sb_bdev(), you should
use mount_bdev() instead, should be just fine.
- Ted
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists