At 00:34 06/07/03, Hans Reiser wrote:
>Chris Mason wrote:
>
>>
>>Thanks, this one looks good.
>>
>>
>Hisashi, please submit it to akpm.
Hi, Andrew.
When write() extends a file(i_size is increased) and fsync() is called,
change of inode must be written to journaling area through fsync().
But,currently the i_trans_id is not correctly updated when i_size
is increased. So fsync() does not kick the journal writer.
Reiserfs_file_write() already updates the transaction when blocks are allocated,
but the case when i_size increases and new blocks are not added is not
correctly treated.
Following patch fix this bug.
Signed-off-by :Hisashi Hifumi <hifumi.hisashi@xxxxxxxxxxxxx>
diff -Nru linux-2.6.17/fs/reiserfs/file.c linux-2.6.17_fix/fs/reiserfs/file.c
--- linux-2.6.17/fs/reiserfs/file.c 2006-06-18 10:49:35.000000000 +0900
+++ linux-2.6.17_fix/fs/reiserfs/file.c 2006-07-03 09:52:51.000000000 +0900
@@ -860,8 +860,12 @@
// this sets the proper flags for O_SYNC to trigger a commit
mark_inode_dirty(inode);
reiserfs_write_unlock(inode->i_sb);
- } else
+ } else {
+ reiserfs_write_lock(inode->i_sb);
+ reiserfs_update_inode_transaction(inode);
mark_inode_dirty(inode);
+ reiserfs_write_unlock(inode->i_sb);
+ }
sd_update = 1;
}
Thanks.
-
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
