[PATCH v2] xfs: use null daddr for unset first bad log block

[PATCH v2] xfs: use null daddr for unset first bad log block

[Yousef Alhouseen]

xlog_do_recovery_pass() may return before setting first_bad.  The caller
must distinguish that case from an error at a valid log block, including
block zero after the log wraps.

Initialize first_bad to XFS_BUF_DADDR_NULL and test it explicitly before
treating the error as a torn write.

Fixes: 7088c4136fa1 ("xfs: detect and trim torn writes during log recovery")
Suggested-by: Darrick J. Wong <djwong@kernel.org>
Reported-by: syzbot+b7dfbed0c6c2b5e9fd34@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=b7dfbed0c6c2b5e9fd34
Cc: stable@vger.kernel.org
Signed-off-by: Yousef Alhouseen <alhouseenyousef@gmail.com>
---
Changes in v2:
- Use XFS_BUF_DADDR_NULL instead of zero as the unset sentinel.
- Test the sentinel explicitly before handling a torn write.

 fs/xfs/xfs_log_recover.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
index 09e6678ca487..5f984bf5698a 100644
--- a/fs/xfs/xfs_log_recover.c
+++ b/fs/xfs/xfs_log_recover.c
@@ -1028,7 +1028,7 @@ xlog_verify_head(
 {
 	struct xlog_rec_header	*tmp_rhead;
 	char			*tmp_buffer;
-	xfs_daddr_t		first_bad;
+	xfs_daddr_t		first_bad = XFS_BUF_DADDR_NULL;
 	xfs_daddr_t		tmp_rhead_blk;
 	int			found;
 	int			error;
@@ -1057,7 +1057,8 @@ xlog_verify_head(
 	 */
 	error = xlog_do_recovery_pass(log, *head_blk, tmp_rhead_blk,
 				      XLOG_RECOVER_CRCPASS, &first_bad);
-	if ((error == -EFSBADCRC || error == -EFSCORRUPTED) && first_bad) {
+	if ((error == -EFSBADCRC || error == -EFSCORRUPTED) &&
+	    first_bad != XFS_BUF_DADDR_NULL) {
 		/*
 		 * We've hit a potential torn write. Reset the error and warn
 		 * about it.
@@ -3575,4 +3576,3 @@ xlog_recover_cancel(
 	if (xlog_recovery_needed(log))
 		xlog_recover_cancel_intents(log);
 }
-
-- 
2.54.0

Re: [PATCH v2] xfs: use null daddr for unset first bad log block

[Darrick J. Wong]

On Tue, Jun 30, 2026 at 12:06:07PM +0200, Yousef Alhouseen wrote:
> xlog_do_recovery_pass() may return before setting first_bad.  The caller
> must distinguish that case from an error at a valid log block, including
> block zero after the log wraps.
> 
> Initialize first_bad to XFS_BUF_DADDR_NULL and test it explicitly before
> treating the error as a torn write.
> 
> Fixes: 7088c4136fa1 ("xfs: detect and trim torn writes during log recovery")
> Suggested-by: Darrick J. Wong <djwong@kernel.org>
> Reported-by: syzbot+b7dfbed0c6c2b5e9fd34@syzkaller.appspotmail.com
> Closes: https://syzkaller.appspot.com/bug?extid=b7dfbed0c6c2b5e9fd34
> Cc: stable@vger.kernel.org

Cc: <stable@vger.kernel.org> # v4.5

> Signed-off-by: Yousef Alhouseen <alhouseenyousef@gmail.com>

Much better now, thanks.
Reviewed-by: "Darrick J. Wong" <djwong@kernel.org>

--D

> ---
> Changes in v2:
> - Use XFS_BUF_DADDR_NULL instead of zero as the unset sentinel.
> - Test the sentinel explicitly before handling a torn write.
> 
>  fs/xfs/xfs_log_recover.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
> index 09e6678ca487..5f984bf5698a 100644
> --- a/fs/xfs/xfs_log_recover.c
> +++ b/fs/xfs/xfs_log_recover.c
> @@ -1028,7 +1028,7 @@ xlog_verify_head(
>  {
>  	struct xlog_rec_header	*tmp_rhead;
>  	char			*tmp_buffer;
> -	xfs_daddr_t		first_bad;
> +	xfs_daddr_t		first_bad = XFS_BUF_DADDR_NULL;
>  	xfs_daddr_t		tmp_rhead_blk;
>  	int			found;
>  	int			error;
> @@ -1057,7 +1057,8 @@ xlog_verify_head(
>  	 */
>  	error = xlog_do_recovery_pass(log, *head_blk, tmp_rhead_blk,
>  				      XLOG_RECOVER_CRCPASS, &first_bad);
> -	if ((error == -EFSBADCRC || error == -EFSCORRUPTED) && first_bad) {
> +	if ((error == -EFSBADCRC || error == -EFSCORRUPTED) &&
> +	    first_bad != XFS_BUF_DADDR_NULL) {
>  		/*
>  		 * We've hit a potential torn write. Reset the error and warn
>  		 * about it.
> @@ -3575,4 +3576,3 @@ xlog_recover_cancel(
>  	if (xlog_recovery_needed(log))
>  		xlog_recover_cancel_intents(log);
>  }
> -
> -- 
> 2.54.0
> 

Re: [PATCH v2] xfs: use null daddr for unset first bad log block

[Christoph Hellwig]

Looks good:

Reviewed-by: Christoph Hellwig <hch@lst.de>

Re: [PATCH v2] xfs: use null daddr for unset first bad log block

[Carlos Maiolino]

On Tue, 30 Jun 2026 12:06:07 +0200, Yousef Alhouseen wrote:
> xlog_do_recovery_pass() may return before setting first_bad.  The caller
> must distinguish that case from an error at a valid log block, including
> block zero after the log wraps.
> 
> Initialize first_bad to XFS_BUF_DADDR_NULL and test it explicitly before
> treating the error as a torn write.
> 
> [...]

Applied to for-next, thanks!

[1/1] xfs: use null daddr for unset first bad log block
      commit: cc9af5e461ea5f6e37738f3f1e41c45a9b7f45d6

Best regards,
-- 
Carlos Maiolino <cem@kernel.org>