[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-Id: <20160510121504.257c2600f7bc4571c8cbcfed@linux-foundation.org>
Date: Tue, 10 May 2016 12:15:04 -0700
From: Andrew Morton <akpm@...ux-foundation.org>
To: Salah Triki <salah.triki@....org>
Cc: linux-kernel@...r.kernel.org
Subject: Re: [PATCH] fs: befs: use flags field to validate fs state
On Tue, 10 May 2016 19:59:42 +0100 Salah Triki <salah.triki@....org> wrote:
> flags field records the state of the superblock, so it is
> more appropriate to use this field for validating the fs state than
> using the fields log_start and log_end.
>
> Signed-off-by: Salah Triki <salah.triki@....org>
> ---
> fs/befs/super.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/fs/befs/super.c b/fs/befs/super.c
> index aeafc4d..079fc00 100644
> --- a/fs/befs/super.c
> +++ b/fs/befs/super.c
> @@ -101,7 +101,7 @@ befs_check_sb(struct super_block *sb)
> return BEFS_ERR;
> }
>
> - if (befs_sb->log_start != befs_sb->log_end) {
> + if (befs_sb->flags == BFS_DIRTY) {
> befs_error(sb, "Filesystem not clean! There are blocks in the "
> "journal. You must boot into BeOS and mount this volume "
> "to make it clean.");
Has this been tested?
Powered by blists - more mailing lists