[Intel-gfx] [PATCH] drm/i915: Perform manual conversions for crtc uapi/hw split, v2.

[Maarten Lankhorst]

Op 24-10-2019 om 14:23 schreef Ville Syrjälä:
> On Thu, Oct 24, 2019 at 02:12:46PM +0200, Maarten Lankhorst wrote:
>> Op 22-10-2019 om 20:16 schreef Ville Syrjälä:
>>> On Fri, Oct 18, 2019 at 10:13:23AM +0200, Maarten Lankhorst wrote:
>>>> intel_get_load_detect_pipe() needs to set uapi active,
>>>> uapi enable is set by the call to drm_atomic_set_mode_for_crtc(),
>>>> so we can remove it.
>>>>
>>>> intel_pipe_config_compare() needs to look at hw state, but I didn't
>>>> change spatch to look at it. It's easy enough to do manually.
>>>>
>>>> intel_atomic_check() definitely needs to check for uapi enable,
>>>> otherwise intel_modeset_pipe_config cannot copy uapi state to hw.
>>> We seem to have three totally separate things in this one patch.
>> The patch is about the manual conversions that need to be done because the automated
>>
>> checks get them wrong.
> Does that mean the series is not actually bisectable?

I've checked all places manually and it should be correct.

For any regression we will likely end up at patch 8/14 as
it formalizes the split. You can investigate further by
changing all hw references in a file to uapi.

Realistically you won't hit any issues until bigjoiner is
introduced, and hw and uapi can contain wildly different
values. This is mostly noticeable for plane_state, because
plane_state->uapi.fb/crtc will be null for a slave plane.