How to support various hardware blocks in drm driver

[Daniel Vetter]

On Tue, Mar 18, 2014 at 1:12 PM, Rob Clark <robdclark at gmail.com> wrote:
>> What's the difference here compared to an encoder_slave? I don't really
>> see the point of adding yet another such thing to the drm core ...
>
> so I think at one point the rough idea was to add additional fxn ptrs
> to bridge as the need arose.  While I still need to give a closer read
> to the original msg, but seems like if you s/drm_hw_block/drm_bridge/g
> it is basically the same thing.  (Not saying "bridge" is the best
> name.. it was just one of those things where no one had a better
> suggestion at the time ;-))

Erhm yeah, drm_bridge is what I've meant ;-) Somehow our efforts in
i915-land to port our dsi driver over to it have stalled a bit, so I
didn't have any examples at hand. Like Rob said, the plan should be to
add necessary functionality to drm_bridge or provide convenience
helper functions to e.g. wire this up with the new componentized
device driver helpers. But in the end you can subclass drm_bridge
within exynos and add fancy stuff yourself if needed. We even talked
about how we could still allow back-end drivers to be shared between
drivers with different drm_bridge subclassing.
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch