[igt-dev] [PATCH v2] tests/kms_atomic_transition: Enable Primary Plane on Enabling CRTC.

[Kazlauskas, Nicholas]

On 2021-04-22 2:13 a.m., Petri Latvala wrote:
> On Fri, Apr 16, 2021 at 01:27:24PM -0400, Mark Yacoub wrote:
>> From: Mark Yacoub <markyacoub at google.com>
>>
>> Setup all plane params before committing anything.
>> Do not disable Primary Plane whenever a CRTC is getting enabled.
>>
>> AMD Driver requires the primary plane to be enabled whenever the CRTC is, otherwise
>> drm_mode_cursor_universal may end up trying to enable the cursor plane while all other
>> planes are disabled, which is not supported by the hardware.
>>
>> === Changes from v1 ===
>> Remove product names from commit message
>>
>> Cc: Nicholas.Kazlauskas at amd.com
>> ---
>>   tests/kms_atomic_transition.c | 7 +++----
>>   1 file changed, 3 insertions(+), 4 deletions(-)
> 
> 
> I've asked Intel display folks to chime in whether i915 needs any of
> this flow needs to be unchanged with is_i915_device(), no response
> yet. We also need a word from someone to confirm that this patch does
> what it says on the tin for amdgpu. Nicholas, are you able to get
> someone to look at this?
> 
> 

Shifting the initial commit after setup_parms seems correct, but I'm not 
sure if avoiding the disable has any impact to the test itself. I would 
think that it does from initial inspection.

The test comment talks about the RMFB CRTC disable case but I don't 
actually see anything testing this specifically in the test? I didn't 
think out fencing was related.

Regards,
Nicholas Kazlauskas