[igt-dev] [i-g-t,v2] tests/kms_color: Change in commit style

[Maarten Lankhorst]

Op 28-02-2019 om 15:07 schreef Sharma, Swati2:
> Hi Maarten,
>
> Thank you for the review :) When I made changes as you said, tests are getting skipped because of test reqt data.display.is_atomic not getting met.
>
> ./build/tests/kms_color --run-subtest pipe-A-ctm-red-to-blue
> IGT-Version: 1.23-g3f2aaed (x86_64) (Linux: 5.0.0-rc8+ x86_64)
> Test requirement not met in function __real_main1161, file ../tests/kms_color.c:1174:
> Test requirement: data.display.is_atomic
> Subtest pipe-A-ctm-red-to-blue: SKIP

What system?


If it's pre gen5, probably no big issue.

> Thanks and Regards,
> Swati
>
> -----Original Message-----
> From: Maarten Lankhorst <maarten.lankhorst at linux.intel.com> 
> Sent: Wednesday, February 13, 2019 8:45 PM
> To: Sharma, Swati2 <swati2.sharma at intel.com>; igt-dev at lists.freedesktop.org
> Subject: Re: [igt-dev] [i-g-t,v2] tests/kms_color: Change in commit style
>
> Hey,
>
> Op 29-01-2019 om 15:08 schreef swati2.sharma at intel.com:
>> From: Swati Sharma <swati2.sharma at intel.com>
>>
>> Existing kms_color i-g-t, commit style by default is legacy for all 
>> the ctm/gamma/degamma subtests.
>>
>> In this patch, provision for both legacy and atomic commit is 
>> provided. Patch is floated as RFC to gather feedback on the idea and 
>> infrastructure proposed.
>>
>> Other approach could be getting the value of is_atomic as done in 
>> kms_available_modes_crc like data.commit = data.display.is_atomic ? 
>> COMMIT_ATOMIC : COMMIT_LEGACY However with this approach we won't be 
>> able to test both the paths.
>>
>> v1: As per Daniel's comments switching over to atomic.
>> v2: Rebase
>>
>> Signed-off-by: Swati Sharma <swati2.sharma at intel.com>
>
> Just do COMMIT_ATOMIC, and add igt_require(display.is_atomic), drivers should support it for this test. :)
>
> ~Maarten
>