On Mon, Mar 10, 2025 at 03:51:53PM +0100, Maxime Ripard wrote:
> On Sun, Mar 09, 2025 at 10:13:58AM +0200, Dmitry Baryshkov wrote:
> > From: Dmitry Baryshkov <dmitry.baryshkov@xxxxxxxxxx>
> >
> > Use drm_hdmi_acr_get_n_cts() helper instead of calculating N and CTS
> > values in the VC4 driver.
> >
> > Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@xxxxxxxxxx>
> > ---
> > drivers/gpu/drm/vc4/vc4_hdmi.c | 10 +++-------
> > drivers/gpu/drm/vc4/vc4_hdmi.h | 7 +++++++
> > 2 files changed, 10 insertions(+), 7 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.h b/drivers/gpu/drm/vc4/vc4_hdmi.h
> > index e3d989ca302b72533c374dfa3fd0d5bd7fe64a82..0a775dbfe99d45521f3d0a2016555aefa81d7934 100644
> > --- a/drivers/gpu/drm/vc4/vc4_hdmi.h
> > +++ b/drivers/gpu/drm/vc4/vc4_hdmi.h
> > @@ -211,6 +211,13 @@ struct vc4_hdmi {
> > * KMS hooks. Protected by @mutex.
> > */
> > enum hdmi_colorspace output_format;
> > +
> > + /**
> > + * @tmds_char_rate: Copy of
> > + * @drm_connector_state.hdmi.tmds_char_rate for use outside of
> > + * KMS hooks. Protected by @mutex.
> > + */
> > + unsigned long long tmds_char_rate;
> > };
>
> This should be in drm_connector_hdmi if it's useful
That would mean bringing the state to a non-state structure on the
framework level. Is it fine from your POV? Is it also fine to use
drm_connector.mutex for protecting this? Or should we be using something
like drm_connector_hdmi.infoframes.mutex (maybe after moving it from
.infoframes to the top level)?
--
With best wishes
Dmitry
