On Tue, Dec 19, 2023 at 09:54:45AM +0900, Dominique Martinet wrote:
Naresh Kamboju wrote on Mon, Dec 18, 2023 at 08:54:13PM +0530:
commit that is causing build failure, drm/atomic: Pass the full state to CRTC atomic begin and flush [ Upstream commit f6ebe9f9c9233a6114eb922aba9a0c9ccc2d2e14 ]
I also had to fix up a few nxp-provided drivers because of this commits, it seems a bit heavy-handed for stable trees when users can have a couple of out of tree modules.
It's marked as a stable dep of fe4c5f662097 ("drm/mediatek: Add spinlock for setting vblank event in atomic_begin") but that looks like it's only because of the context and it should be easy to backport that commit without the crtc atomic begin/flush rework -- what do you think?
Good catch, I've dropped those two larger dependent patches now and fixed up the spinlock patch by hand to work properly.
I'll push out a -rc2 now, thanks!
greg k-h