On Wed, Dec 03, 2025 at 10:54:45AM +0100, Linus Walleij wrote: > On Wed, Dec 3, 2025 at 4:10 AM Chaoyi Chen <[email protected]> wrote: > > > It seems that multiple drivers currently depend on the CRTC being > > enabled and they implement the same atomic_commit_tail(). > > > > Why not implement this in drm_atomic_helper_commit_tail_rpm() instead? > > Or why not use another common helper function for this? > > So my v2 version of the patch series added a new special case > helper tail function to do that: > https://lore.kernel.org/dri-devel/[email protected]/ > > It was politely NACKed for complicating the helpers library (short story).
The longer story is that it wasn't NACKed at all, but if we want to do that: - We need to figure out the bridge ordering mess in the first place - Improve the commit_tail helpers doc to make it clearer what each are doing - Probably change their name too, since the rpm suffix is a gross simplification of what is happening - Create a few kunit tests to make sure whatever order they guarantee is properly implemented - And maybe create a few helpers and convert drivers to that. None of that have any place for a regression fix, so it wasn't a reasonable ask. Maxime
signature.asc
Description: PGP signature
