[Intel-gfx] [PATCH 00/19] [RFC] introduce struct intel_pipe_config

[Jesse Barnes]

On Fri, 15 Feb 2013 11:05:26 +0100
Daniel Vetter <daniel at ffwll.ch> wrote:

> On Thu, Feb 14, 2013 at 02:28:19PM -0800, Jesse Barnes wrote:
> > Well, I think we'll have to punt in quite a few configs due to hw
> > limitations about what can be changed without a full pipe shutdown, but
> > we can definitely do better than we do today, and may be able to take
> > some shortcuts (provided we get lots of testing on the given hw).
> 
> Yeah, as discussed on irc my idea (doesn't exists as code yet) is to add a
> few special modes to the pipe_config compare function:
> - strict: Used after a modeset to check whether the new hw state matches
>   up with what we wanted to set.
> - modeset: More permissive for fastboot to decide whether the new
>   configuration is essentially the same as the old one. Things like fuzzy
>   clock matching and things like that would imo fit here.
> - plane-update-only: Even more permissive to figure out whether we only
>   have different plane/pfit/whatever settings. For a new fastboot plane
>   update mode that everyone seems to dream about.
> - Feel free to go nuts ;-)

On a related note, getting the clock for gen5+ plus is really easy if
you want to get the fitted mode.  You can just read the linkm reg.  The
problem is that it doesn't tell us what the clock would be for the
native mode which we read from the htotal/vtotal etc regs.  However
forcing it to the right value works, even though the pipe m/n values
are wrong for the native mode.

> > > Comments, flames and ideas for the patches themselves, but also for the above
> > > issues in general highly welcome.
> > 
> > Overall I like this approach, it should make fastboot state tracking a
> > lot easier.
> > 
> > We'll also want to track gamma enable; unfortunately that can only be
> > changed with the plane off (according to docs), but we should check for
> > it regardless.
> 
> Hey, that's a new one! I've we're lucky we could subsume this with the
> "update plane only" mode to get rid of the pfit and similar stuff we don't
> want ...
> 
> A similar issue is all the various bits&pieces to support hdmi/audio. I
> don't know yet how we should exactly track this. Might be that fastboot
> for external screens is a long way off still.

Well, first step is to track that stuff.  We can still fastboot in many
cases but then punt to a full mode set if we find any state we don't
deal with.

-- 
Jesse Barnes, Intel Open Source Technology Center