Javier Martinez Canillas <javierm@xxxxxxxxxx> writes: > Thomas Zimmermann <tzimmermann@xxxxxxx> writes: > > Hello Thomas, > [...] >> >> You have a atomic_disable in that plane, so you're taking the branch at >> [2] for disabling the plane. No atomic_update then. If the plane has >> been enabled, you should take the branch at [3]. Without being able to >> move/scale the primary plane, I don't see how plane_state->visible could >> be false here. Right? >> >> AFAIKT there should not be a NULL-deref here. Can you do a test? >> > > You are correct, there's no NULL-deref and in fact you are fixing the > plane not disable bug, so your fix is correct and should be applied. > > I still prefer as mentioned keeping the drm_plane_helper_atomic_check() > call instead of open coding it, but regardless of what is decided: > > Reviewed-by: Javier Martinez Canillas <javierm@xxxxxxxxxx> > And also for patches #1, #2, #3, #6 and #7, feel free to add: Tested-by: Javier Martinez Canillas <javierm@xxxxxxxxxx> -- Best regards, Javier Martinez Canillas Core Platforms Red Hat
