2009-04-03 01:34:56

by Alexander Beregalov

Subject: [PATCH] reiserfs: fix printk format warning

Fix following warning message
fs/reiserfs/procfs.c:393: warning: format '%u' expects type 'unsigned int', but argument 22 has type 'long unsigned int'
introduced in 600ed4167
(reiserfs: audit transaction ids to always be unsigned ints)

Signed-off-by: Alexander Beregalov <[email protected]>
---

include/linux/reiserfs_fs_sb.h | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/include/linux/reiserfs_fs_sb.h b/include/linux/reiserfs_fs_sb.h
index 5621d87..1d1594a 100644
--- a/include/linux/reiserfs_fs_sb.h
+++ b/include/linux/reiserfs_fs_sb.h
@@ -193,7 +193,7 @@ struct reiserfs_journal {
atomic_t j_wcount; /* count of writers for current commit */
unsigned long j_bcount; /* batch count. allows turning X transactions into 1 */
unsigned long j_first_unflushed_offset; /* first unflushed transactions offset */
- unsigned long j_last_flush_trans_id; /* last fully flushed journal timestamp */
+ unsigned int j_last_flush_trans_id; /* last fully flushed journal timestamp */
struct buffer_head *j_header_bh;

time_t j_trans_start_time; /* time this transaction started */

2009-04-09 21:00:46

by Andrew Morton

Subject: Re: [PATCH] reiserfs: fix printk format warning

On Fri, 3 Apr 2009 05:34:37 +0400
Alexander Beregalov <[email protected]> wrote:

> Fix following warning message
> fs/reiserfs/procfs.c:393: warning: format '%u' expects type 'unsigned int', but argument 22 has type 'long unsigned int'
> introduced in 600ed4167
> (reiserfs: audit transaction ids to always be unsigned ints)
>
> Signed-off-by: Alexander Beregalov <[email protected]>
> ---
>
> include/linux/reiserfs_fs_sb.h | 2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/include/linux/reiserfs_fs_sb.h b/include/linux/reiserfs_fs_sb.h
> index 5621d87..1d1594a 100644
> --- a/include/linux/reiserfs_fs_sb.h
> +++ b/include/linux/reiserfs_fs_sb.h
> @@ -193,7 +193,7 @@ struct reiserfs_journal {
> atomic_t j_wcount; /* count of writers for current commit */
> unsigned long j_bcount; /* batch count. allows turning X transactions into 1 */
> unsigned long j_first_unflushed_offset; /* first unflushed transactions offset */
> - unsigned long j_last_flush_trans_id; /* last fully flushed journal timestamp */
> + unsigned int j_last_flush_trans_id; /* last fully flushed journal timestamp */
> struct buffer_head *j_header_bh;
>
> time_t j_trans_start_time; /* time this transaction started */

Yes, looking at the code I think it does want to be a u32.