Christoph Hellwig wrote:
> On Thu, Jan 28, 2010 at 05:30:25PM -0600, Eric Sandeen wrote:
>> 053 and 077 can be generic w/ a little tweaking.
>> Also change 077's filler to something more likely
>> to be on a random system under test, and add it
>> to the enospc group.
>> Signed-off-by: Eric Sandeen <sandeen@xxxxxxxxxxx>
>> diff --git a/053 b/053
>> index 98644a7..5542885 100755
>> --- a/053
>> +++ b/053
>> @@ -38,18 +38,19 @@ trap "rm -f $tmp.*; exit \$status" 0 1 2 3 15
>> . ./common.attr
>> # real QA test starts here
>> -_supported_fs xfs
>> +_supported_fs generic
>> _supported_os Linux
>> [ ! -x /bin/chacl -a ! -x /usr/bin/chacl ] && _notrun "chacl command not
> _acl_requirements includes the chacl check, so it can be removed.
oh good point :)
> Btw, any reason _acl_requirements isn't called _require_acl?
It already existed, I just used it ... renaming is probably a good idea.
>> diff --git a/077 b/077
>> index cdee8da..d23305f 100755
>> --- a/077
>> +++ b/077
>> @@ -31,7 +31,8 @@ here=`pwd`
>> +filler=/lib/modules/`uname -r`/
> Oooops. Maybe just use /lib/modules/ to get as much as possible.
> And no need to let the other bits comment out there. Maybe we need
> to generate the data instead to really be safe - at least for my
> tests under qemu I don't have any modules installed. But that should
> be left for another patch.
hm, on the other hand some of my machines have so many kernels installed,
picking up all of /lib/modules might slow it down a lot.
So generating might be better, yeah.
> Anyway, patch looks good enough to go in for now, so
thanks, I'll pull the chacl test & the commented lines before that.
> Reviewed-by: Christoph Hellwig <hch@xxxxxx>