Am Di., 1. Apr. 2025 um 02:29 Uhr schrieb Alex Hung <alex.h...@amd.com>: > > > > On 3/31/25 12:53, Xaver Hugl wrote: > >> Cursor plane has no color pipeline and thus it has no colorop either. It > >> inherits color processing from its parent plane. > > > > Just to be sure: That means amdgpu will reject atomic commits that try > > to set a color pipeline on the primary plane while showing the cursor > > plane on top of it? Just like with scaling? > > > In theory that should be the case, and I will investigate and confirm > it. Is this a beavhiour expected by a compositor?
It's not just an expectation, the API requires it. If you were to apply a color pipeline to a plane that doesn't have any, or not apply one that is set, then that would be incredibly broken. Commits where that would happen must never ever pass atomic tests.
