Re: [PATCH blktests] loop/011: skip if running on kernel older than v6.10

[Nilay Shroff <nilay@linux.ibm.com>]

On 8/1/24 11:57, Shinichiro Kawasaki wrote:
> On Jul 31, 2024 / 19:29, Nilay Shroff wrote:
>> Hi Shinichiro,
>>
>> On 7/31/24 18:16, Cyril Hrubis wrote:
>>> Hi!
>>>>> According to www.kernel.org, the 6.9 stable branch is already EOL. Is it planned
>>>>> to backport the kernel fix to other longterm branches?
>>>> I just checked this commit 5f75e081ab5c ("loop: Disable fallocate() zero and discard
>>>> if not supported") hasn't been backported to any of the longterm stable kernel yet.
>>>> However I don't know if there's any plan to backport it on longterm stable kernel.
>>>
>>> The patch will not apply into older branches since the in kernel API did
>>> change, so I suppose that nobody will invest into rewriting the patch
>>> since it's mostly cosmetic.
>>>
>>
>> This commit 5f75e081ab5c has been backported to kernel v6.9.11 and per the above 
>> comment from Cyril, this commit shall not be backported further to any other longterm 
>> kernel.
>>
>> So is it reasonable to assume that this test would fail on kernel older than v6.9.11?
> 
> I have the same guess.
> 
>> And if this is true then how about rewriting the patch as below ?
>>
>> diff --git a/tests/loop/011 b/tests/loop/011
>> index 35eb39b..a454848 100755
>> --- a/tests/loop/011
>> +++ b/tests/loop/011
>> @@ -9,6 +9,7 @@
>>  DESCRIPTION="Make sure unsupported backing file fallocate does not fill dmesg with errors"
>>  
>>  requires() {
>> +       _have_kver 6 9 11
>>         _have_program mkfs.ext2
>>  }
> 
> I think this change is reasonable. Would you repost the patch?
Sure, I will send another patch.

Thanks,
--Nilay