[PATCH] xfstests 042: point xfs_fsr right at the fragmented file

[PATCH] xfstests 042: point xfs_fsr right at the fragmented file

[Eric Sandeen]

Somehow 042 as it stands isn't actually defragging the fragmented
file we created (see 042.full post-run)

Probably need to look into that as well, but for now this should
make the call to fsr actually _do_ something.

This has let 2 fsr bugs slip through qa in the past few months.  :(

Signed-off-by: Eric Sandeen <sandeen@sandeen.net>
---

diff --git a/042 b/042
index 6628040..1ad81e4 100755
--- a/042
+++ b/042
@@ -114,7 +114,7 @@ _do "sum $SCRATCH_MNT/fragmented >$tmp.sum1"
 _do "Remove other files" "rm -rf $SCRATCH_MNT/{pad,hole*}"
 
 # defragment
-_do "Run xfs_fsr on filesystem" "$XFS_FSR_PROG -v $SCRATCH_DEV"
+_do "Run xfs_fsr on filesystem" "$XFS_FSR_PROG -v $SCRATCH_MNT/fragmented"
 _do "xfs_bmap -v $SCRATCH_MNT/fragmented"
 _do "Check 4k files" "src/fill2fs_check $tmp.manifest"
 


_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

Re: [PATCH] xfstests 042: point xfs_fsr right at the fragmented file

[Christoph Hellwig]

On Thu, May 07, 2009 at 08:12:29PM -0500, Eric Sandeen wrote:
> Somehow 042 as it stands isn't actually defragging the fragmented
> file we created (see 042.full post-run)
> 
> Probably need to look into that as well, but for now this should
> make the call to fsr actually _do_ something.
> 
> This has let 2 fsr bugs slip through qa in the past few months.  :(

Looks good.  Can you open a bugzilla for the problem of fsr not actually
finding the fragmented file so that we don't forget about that one?

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

Re: [PATCH] xfstests 042: point xfs_fsr right at the fragmented file

[Eric Sandeen]

Christoph Hellwig wrote:
> On Thu, May 07, 2009 at 08:12:29PM -0500, Eric Sandeen wrote:
>> Somehow 042 as it stands isn't actually defragging the fragmented
>> file we created (see 042.full post-run)
>>
>> Probably need to look into that as well, but for now this should
>> make the call to fsr actually _do_ something.
>>
>> This has let 2 fsr bugs slip through qa in the past few months.  :(
> 
> Looks good.  Can you open a bugzilla for the problem of fsr not actually
> finding the fragmented file so that we don't forget about that one?
> 

Good idea and done: http://oss.sgi.com/bugzilla/show_bug.cgi?id=821

-Eric

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs