On Wed, Dec 9, 2020 at 12:17 PM Tomi Valkeinen <tomi.valkei...@ti.com> wrote:
>
> Hi Daniel,
>
> On 09/12/2020 02:51, Daniel Vetter wrote:
>
> >>> diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h
> >>> index ba839e5e357d..4d9e217e5040 100644
> >>> --- a/include/drm/drm_crtc.h
> >>> +++ b/include/drm/drm_crtc.h
> >>> @@ -1084,6 +1084,9 @@ struct drm_crtc {
> >>> */
> >>> uint16_t *gamma_store;
> >>>
> >>> + bool has_gamma_prop : 1;
> >>> + bool has_degamma_prop : 1;
> >
> > I'm a bit behind on patches, but in case this got missed please remove
> > this and replace with the (obj, prop) lookup function thing or something
> > like that. Makes sure everything stays in sync, plus like I said atomic
> > uses this a ton. So not a problem here.
>
> The drm_mode_obj_find_prop_id() is in core drm.ko, and not exported, but I
> need it also from
> drm_kms_helper.ko.
>
> drm_mode_obj_find_prop_id() is declared in drm_crtc_internal.h. Are those
> functions supposed to be
> not exported from drm.ko?
>
> So, is it fine to just export drm_mode_obj_find_prop_id? If I export, should
> I move it to
> drm_mode_object.h?
It's not exported to drivers so they don't play games in their own
ioctls or do something else than the atomic set/get_property hooks. I
guess we could export, but I think here the better solution is to lift
the legacy gamma compat function to core code status, since I think
that's the direction we want to go anyway. Bit more churn since the
name changes.
-Daniel
--
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
