Hi Dmitry,

On Sun, Feb 25, 2024 at 04:50:00PM +0200, Dmitry Baryshkov wrote:

On Thu, 22 Feb 2024 at 20:14, Maxime Ripard [off-list ref] wrote:

quoted

Here's a series that creates some extra infrastructure specifically
targeted at HDMI controllers.

The idea behind this series came from a recent discussion on IRC during
which we discussed infoframes generation of i915 vs everything else.

Infoframes generation code still requires some decent boilerplate, with
each driver doing some variation of it.

In parallel, while working on vc4, we ended up converting a lot of i915
logic (mostly around format / bpc selection, and scrambler setup) to
apply on top of a driver that relies only on helpers.

While currently sitting in the vc4 driver, none of that logic actually
relies on any driver or hardware-specific behaviour.

The only missing piece to make it shareable are a bunch of extra
variables stored in a state (current bpc, format, RGB range selection,
etc.).

The initial implementation was relying on some generic subclass of
drm_connector to address HDMI connectors, with a bunch of helpers that
will take care of all the "HDMI Spec" related code. Scrambler setup is
missing at the moment but can easily be plugged in.

The feedback was that creating a connector subclass like was done for
writeback would prevent the adoption of those helpers since it couldn't
be used in all situations (like when the connector driver can implement
multiple output) and required more churn to cast between the
drm_connector and its subclass. The decision was thus to provide a set
of helper and to store the required variables in drm_connector and
drm_connector_state. This what has been implemented now.

Hans Verkuil also expressed interest in implementing a mechanism in v4l2
to retrieve infoframes from HDMI receiver and implementing a tool to
decode (and eventually check) infoframes. His current work on
edid-decode to enable that based on that series can be found here:
https://git.linuxtv.org/hverkuil/edid-decode.git/log/?h=hverkuil

And some more context here:
https://lore.kernel.org/dri-devel/50db7366-cd3d-4675-aaad-b857202234de@xs4all.nl/ (local)

This series thus leverages the infoframe generation code to expose it
through debugfs.

[...]

quoted

Let me know what you think,
Maxime

The overall idea looks great. I've started checking how I can use that
for our msm devices family, which makes use of bridges and
drm_bridge_connector.

Yeah, I had that in mind for a while too. I think it would be valuable,
but it's difficult as it is already so I didn't tackle it at first :)

The format negociation especially seems to divert quite a lot from what
i915 and vc4 have been doing, and that's user facing to some extent.

My current idea is to extend the drm_bridge_funcs with the new
callback to be called once the drm_connector has been instantiated.
This way all the bridges can influence new connector.
Another possibility is to follow drm_bridge_connector design closely
and let it call into drm_connector_hdmi code if it detects that the
last bridge is the HDMI one.
WDYT?

I had the latter in mind, but I haven't really tried to reconcile the
connector state output_format with the
drm_atomic_helper_bridge_propagate_bus_fmt, and plug in the output_bpc
count too.

We would have to create the max_bpc properties from the HDMI connector,
and then propagate that upstream along the bridges I guess (or we can
just ignore it for now).

So, yeah, I had the latter in mind but it might turn out that the former
is actually easier. Both make sense to me at least.

Maxime