Re: [PATCH v2] ext4/306: require 64bit feature to run ext4/306

[Xiao Yang <yangx.jy@cn.fujitsu.com>]

On 2017/03/02 12:22, Eryu Guan   wrote:
> On Thu, Mar 02, 2017 at 10:24:26AM +0800, Xiao Yang wrote:
>> The 64bit feature isn't supported by mkfs.ext4 on RHEL6, so
>> we add _require_ext4_mkfs_64bit to check it.  If 64bit feature
>> isn't supported, we could skip this case.  This feature has
>> been produced by 'commit 02d6f47e9647d ("mke2fs: Fix up mke2fs
>> to be able to make 64-bit file systems")'.
>>
>> Signed-off-by: Xiao Yang<yangx.jy@cn.fujitsu.com>
>> ---
>>   common/rc      | 11 +++++++++++
>>   tests/ext4/306 |  1 +
>>   2 files changed, 12 insertions(+)
>>
>> diff --git a/common/rc b/common/rc
>> index 8f23334..8c4f1c3 100644
>> --- a/common/rc
>> +++ b/common/rc
>> @@ -1669,6 +1669,17 @@ _require_scratch_ext4_crc()
>>   	_scratch_unmount
>>   }
>>
>> +# this test requires the 64bit feature to be available in mkfs.ext4
>> +_require_ext4_mkfs_64bit()
>> +{
>> +	local testfile=$TEST_DIR/$$.64bit
>> +
>> +	touch $testfile
>> +	$MKFS_EXT4_PROG -F -O 64bit -n $testfile 512m>/dev/null 2>&1 \
>> +	   || _notrun "mkfs.ext4 doesn't have 64bit feature"
>> +	rm -f $testfile
>> +}
> Sorry, I didn't realize earlier that _require_mkfs_mkfs_bigalloc() runs
> mkfs test against $SCRATCH_DEV too. Seems we should fix it too.
Hi Eryu

I will fix it too. :-)
> Can you please factor out a new helper, e.g. _mkfs_ext4_supported(), to
> accept mkfs options and return mkfs status? So that _require_ext4_mkfs_*
> could use this new helper, just pass different feature bits it wants to
> test. e.g.
>
> _require_ext4_mkfs_64bit()
> {
> 	_mkfs_ext4_supported -O 64bit || \
> 		_notrun "mkfs.ext4 doesn't have 64bit feature"
> }
>
That's good idea, and I am trying to write this new helper as you 
sugguested.

Thanks,
Xiao Yang
> Thanks!
>
> Eryu
>
>
> .
>