[Intel-gfx] [RFC] dpms handling on atomic drivers

[Daniel Vetter]

On Fri, Nov 7, 2014 at 12:33 AM, Daniel Vetter <daniel at ffwll.ch> wrote:
> All this only matters for cloned configs anyway only, which is about
> i915 for gen2, roughly. Well Ville implemented cloning for some more
> recent stuff too.

Ok, I've grepped: Beside i915 there's only radeon which has a
possible_clones setting which isn't totally bogus. Everyone else
either sets nonsense, 0 (which is also nonsense, the own encoder bit
should be in there), nothing (which means 0 because kzalloc) or a
complete lie like exynos claiming to be able to clone encoders between
fimd and mixer.

So for everyone else: I simply propose to clamp dpms to on/off and
move it to the crtc as a new boolean called "active". All the other
talk is only interesting if you have cloned configs in hw (i.e. 1
crtc, 2 outputs from the same crtc).

Another (rather quick chase, haven't followed call-chains fully
really) audit shows that besides i915 no one has a disdinction between
crtc enable and active tracking. At least I haven't found how radeon
tracks shared dplls even, that one should have some checks though
since they have 5+ output cards with just 2-3 dplls. So at least right
now not a lot of drivers which bother with even handling this.
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch