2008-09-22 10:58:11

by Takashi Sato

[permalink] [raw]
Subject: [PATCH 4/10] xfs: Fix error handling in write_super_lockfs/unlockfs

I've changed write_super_lockfs/unlockfs so that they always return
0 (success) to keep a current behavior.

Signed-off-by: Takashi Sato <[email protected]>
Signed-off-by: Masayuki Hamaguchi <[email protected]>
---
xfs_super.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)

diff -uprN -X linux-2.6.27-rc7-lockfs/Documentation/dontdiff linux-2.6.27-rc7-lockfs/fs/xfs/linux-2.6/xfs_super.c linux-
2.6.27-rc7-lockfs-xfs/fs/xfs/linux-2.6/xfs_super.c
--- linux-2.6.27-rc7-lockfs/fs/xfs/linux-2.6/xfs_super.c 2008-09-22 07:29:55.000000000 +0900
+++ linux-2.6.27-rc7-lockfs-xfs/fs/xfs/linux-2.6/xfs_super.c 2008-09-22 10:13:39.000000000 +0900
@@ -1351,7 +1351,7 @@ xfs_fs_remount(
* need to take care of themetadata. Once that's done write a dummy
* record to dirty the log in case of a crash while frozen.
*/
-STATIC void
+STATIC int
xfs_fs_lockfs(
struct super_block *sb)
{
@@ -1359,6 +1359,7 @@ xfs_fs_lockfs(

xfs_attr_quiesce(mp);
xfs_fs_log_dummy(mp);
+ return 0;
}

STATIC int


2008-09-22 23:41:20

by Dave Chinner

[permalink] [raw]
Subject: Re: [PATCH 4/10] xfs: Fix error handling in write_super_lockfs/unlockfs

On Mon, Sep 22, 2008 at 07:56:37PM +0900, Takashi Sato wrote:
> I've changed write_super_lockfs/unlockfs so that they always return
> 0 (success) to keep a current behavior.
>
> Signed-off-by: Takashi Sato <[email protected]>
> Signed-off-by: Masayuki Hamaguchi <[email protected]>
> ---
> xfs_super.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff -uprN -X linux-2.6.27-rc7-lockfs/Documentation/dontdiff linux-2.6.27-rc7-lockfs/fs/xfs/linux-2.6/xfs_super.c linux-
> 2.6.27-rc7-lockfs-xfs/fs/xfs/linux-2.6/xfs_super.c
> --- linux-2.6.27-rc7-lockfs/fs/xfs/linux-2.6/xfs_super.c 2008-09-22 07:29:55.000000000 +0900
> +++ linux-2.6.27-rc7-lockfs-xfs/fs/xfs/linux-2.6/xfs_super.c 2008-09-22 10:13:39.000000000 +0900
> @@ -1351,7 +1351,7 @@ xfs_fs_remount(
> * need to take care of themetadata. Once that's done write a dummy
> * record to dirty the log in case of a crash while frozen.
> */
> -STATIC void
> +STATIC int
> xfs_fs_lockfs(
> struct super_block *sb)
> {
> @@ -1359,6 +1359,7 @@ xfs_fs_lockfs(
>
> xfs_attr_quiesce(mp);
> xfs_fs_log_dummy(mp);
> + return 0;
> }

xfs_fs_log_dummy() currently is void because it had nowhere to
return errors to. It silently throws away errors because of this.
Hence the correct thing to do here is to fix xfs_fs_log_dummy()
to return the errors it currently ignores. i.e. catch the errors
from xfs_trans_reserve() and xfs_trans_commit() calls in
xfs_fs_log_dummy().

Cheers,

Dave.
--
Dave Chinner
[email protected]

2008-09-25 21:22:37

by Eric Sandeen

[permalink] [raw]
Subject: Re: [PATCH 4/10] xfs: Fix error handling in write_super_lockfs/unlockfs

Dave Chinner wrote:

> xfs_fs_log_dummy() currently is void because it had nowhere to
> return errors to. It silently throws away errors because of this.
> Hence the correct thing to do here is to fix xfs_fs_log_dummy()
> to return the errors it currently ignores. i.e. catch the errors
> from xfs_trans_reserve() and xfs_trans_commit() calls in
> xfs_fs_log_dummy().
>
> Cheers,
>
> Dave.
>
Here's that part in case it's helpful :) Feel free to roll it into
your patch.

-Eric

Make xfs_fs_log_dummy() return errors which can then be returned by
xfs_fs_lockfs(), once it is capable of doing so.

Signed-off-by: Eric Sandeen <[email protected]>

---

Index: linux-2.6/fs/xfs/xfs_fsops.c
===================================================================
--- linux-2.6.orig/fs/xfs/xfs_fsops.c 2008-08-04 15:30:31.000000000 -0500
+++ linux-2.6/fs/xfs/xfs_fsops.c 2008-09-25 16:16:46.916001110 -0500
@@ -589,17 +589,19 @@ out:
return 0;
}

-void
+int
xfs_fs_log_dummy(
xfs_mount_t *mp)
{
xfs_trans_t *tp;
xfs_inode_t *ip;
+ int error = 0;

tp = _xfs_trans_alloc(mp, XFS_TRANS_DUMMY1);
- if (xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp), 0, 0, 0)) {
+ error = xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp), 0, 0, 0);
+ if (error) {
xfs_trans_cancel(tp, 0);
- return;
+ return error;
}

ip = mp->m_rootip;
@@ -609,9 +611,10 @@ xfs_fs_log_dummy(
xfs_trans_ihold(tp, ip);
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
xfs_trans_set_sync(tp);
- xfs_trans_commit(tp, 0);
+ error = xfs_trans_commit(tp, 0);

xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ return error;
}

int
Index: linux-2.6/fs/xfs/xfs_fsops.h
===================================================================
--- linux-2.6.orig/fs/xfs/xfs_fsops.h 2008-06-05 13:44:24.000000000 -0500
+++ linux-2.6/fs/xfs/xfs_fsops.h 2008-09-25 16:16:58.303000603 -0500
@@ -25,6 +25,6 @@ extern int xfs_fs_counts(xfs_mount_t *mp
extern int xfs_reserve_blocks(xfs_mount_t *mp, __uint64_t *inval,
xfs_fsop_resblks_t *outval);
extern int xfs_fs_goingdown(xfs_mount_t *mp, __uint32_t inflags);
-extern void xfs_fs_log_dummy(xfs_mount_t *mp);
+extern int xfs_fs_log_dummy(xfs_mount_t *mp);

#endif /* __XFS_FSOPS_H__ */