[PATCH] [PATCH] xfstests: use -F when making scratch ext4 fs

[PATCH] [PATCH] xfstests: use -F when making scratch ext4 fs

[Josef Bacik]

mkfs.ext4 will ask you if you are sure if you try to mkfs an entire disk instead
of just a partition, so any xfstest that uses a scratch device and you specify a
full disk will appear to hang because it's waiting for you to say yes.  This
fixes the problem by just forcing it to do its thing.  Thanks,

Signed-off-by: Josef Bacik <jbacik@fusionio.com>
---
 common/rc |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/common/rc b/common/rc
index ff18cc2..9d1c3fc 100644
--- a/common/rc
+++ b/common/rc
@@ -485,7 +485,7 @@ _scratch_mkfs_ext4()
 {
 	local tmp_dir=/tmp/
 
-	/sbin/mkfs -t $FSTYP -- $MKFS_OPTIONS $* $SCRATCH_DEV \
+	/sbin/mkfs -t $FSTYP -- -F $MKFS_OPTIONS $* $SCRATCH_DEV \
 			2>$tmp_dir.mkfserr 1>$tmp_dir.mkfsstd
 	local mkfs_status=$?
 
-- 
1.7.7.6

Re: [PATCH] [PATCH] xfstests: use -F when making scratch ext4 fs

[Eric Sandeen]

On 4/26/13 2:16 PM, Josef Bacik wrote:
> mkfs.ext4 will ask you if you are sure if you try to mkfs an entire disk instead
> of just a partition, so any xfstest that uses a scratch device and you specify a
> full disk will appear to hang because it's waiting for you to say yes.  This
> fixes the problem by just forcing it to do its thing.  Thanks,
> 
> Signed-off-by: Josef Bacik <jbacik@fusionio.com>

Reviewed-by: Eric Sandeen <sandeen@redhat.com>

The whole-disk test is pretty derpy anyway. Too bad -F overrides
much more than just that, but this is probably a decent
thing to do.

Thanks,
-Eric

> ---
>  common/rc |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/common/rc b/common/rc
> index ff18cc2..9d1c3fc 100644
> --- a/common/rc
> +++ b/common/rc
> @@ -485,7 +485,7 @@ _scratch_mkfs_ext4()
>  {
>  	local tmp_dir=/tmp/
>  
> -	/sbin/mkfs -t $FSTYP -- $MKFS_OPTIONS $* $SCRATCH_DEV \
> +	/sbin/mkfs -t $FSTYP -- -F $MKFS_OPTIONS $* $SCRATCH_DEV \
>  			2>$tmp_dir.mkfserr 1>$tmp_dir.mkfsstd
>  	local mkfs_status=$?
>  
> 

Re: [PATCH] [PATCH] xfstests: use -F when making scratch ext4 fs

[Andreas Dilger]

On 2013-04-26, at 1:18 PM, Eric Sandeen wrote:
> On 4/26/13 2:16 PM, Josef Bacik wrote:
>> mkfs.ext4 will ask you if you are sure if you try to mkfs an entire disk instead of just a partition, so any xfstest that uses a scratch device and you specify a full disk will appear to hang because it's waiting for you to say yes.  This fixes the problem by just forcing it to do its thing.  Thanks,
>> 
>> Signed-off-by: Josef Bacik <jbacik@fusionio.com>
> 
> Reviewed-by: Eric Sandeen <sandeen@redhat.com>
> 
> The whole-disk test is pretty derpy anyway. Too bad -F overrides
> much more than just that, but this is probably a decent
> thing to do.

Yeah, we've had to work around the whole-device check for years,
because the 512-byte sector offset for the DOS partition table
kills performance on RAID systems.

Given that the kernel should hold the whole block device busy these
days for O_EXCL opens if there are partitions on the device, it
should be safe to remove this check from mke2fs, though the above
workaround would still be needed for a while.

Cheers, Andreas

Re: [PATCH] [PATCH] xfstests: use -F when making scratch ext4 fs

[Rich Johnston]

Hi Josef,

Thanks for this patch, it has been committed,

Regards,
--Rich
commit c150d937f506912da8c48419d6939731d5063967
Author: Josef Bacik <jbacik@fusionio.com>
Date:   Fri Apr 26 19:16:53 2013 +0000

     xfstests: use -F when making scratch ext4 fs