[PATCH 4/4] drm/i915/panel: sync panel prepared state at register

[Murthy, Arun R]

> -----Original Message-----
> From: Intel-gfx <intel-gfx-bounces at lists.freedesktop.org> On Behalf Of Jani
> Nikula
> Sent: Friday, June 6, 2025 2:35 PM
> To: dri-devel at lists.freedesktop.org
> Cc: intel-gfx at lists.freedesktop.org; intel-xe at lists.freedesktop.org; Heikki
> Krogerus <heikki.krogerus at linux.intel.com>; Wysocki, Rafael J
> <rafael.j.wysocki at intel.com>; Lee, Shawn C <shawn.c.lee at intel.com>; Yang,
> Simon1 <simon1.yang at intel.com>; Doug Anderson <dianders at chromium.org>;
> Maxime Ripard <mripard at kernel.org>; Nikula, Jani <jani.nikula at intel.com>
> Subject: [PATCH 4/4] drm/i915/panel: sync panel prepared state at register
> 
> If the panel is enabled at probe, and we take over the hardware state, the
> drm_panel prepared state will be out of sync. We'll need to notify drm_panel
> framework about the state at probe, so it can in turn notify the panel followers.
> 
> Cc: Lee Shawn C <shawn.c.lee at intel.com>
> Signed-off-by: Jani Nikula <jani.nikula at intel.com>
> ---
Reviewed-by: Arun R Murthy <arun.r.murthy at intel.com>

Thanks and Regards,
Arun R Murthy
-------------------

>  drivers/gpu/drm/i915/display/intel_panel.c | 41 ++++++++++++++++++++++
>  1 file changed, 41 insertions(+)
> 
> diff --git a/drivers/gpu/drm/i915/display/intel_panel.c
> b/drivers/gpu/drm/i915/display/intel_panel.c
> index b1d549e6cf53..f956919dc648 100644
> --- a/drivers/gpu/drm/i915/display/intel_panel.c
> +++ b/drivers/gpu/drm/i915/display/intel_panel.c
> @@ -463,6 +463,45 @@ void intel_panel_fini(struct intel_connector
> *connector)
>  	}
>  }
> 
> +/*
> + * If the panel was already enabled at probe, and we took over the
> +state, the
> + * panel prepared state is out of sync, and the panel followers won't
> +be
> + * notified. We need to call drm_panel_prepare() on enabled panels.
> + *
> + * It would be natural to handle this e.g. in the connector
> +->sync_state hook at
> + * intel_modeset_readout_hw_state(), but that's unfortunately too
> +early. We
> + * don't have drm_connector::kdev at that time. For now, figure out the
> +state at
> + * ->late_register, and sync there.
> + */
> +static void intel_panel_sync_state(struct intel_connector *connector) {
> +	struct intel_display *display = to_intel_display(connector);
> +	struct drm_connector_state *conn_state;
> +	struct intel_crtc *crtc;
> +	int ret;
> +
> +	ret = drm_modeset_lock(&display->drm-
> >mode_config.connection_mutex, NULL);
> +	if (ret)
> +		return;
> +
> +	conn_state = connector->base.state;
> +
> +	crtc = to_intel_crtc(conn_state->crtc);
> +	if (crtc) {
> +		struct intel_crtc_state *crtc_state;
> +
> +		crtc_state = to_intel_crtc_state(crtc->base.state);
> +
> +		if (crtc_state->hw.active) {
> +			drm_dbg_kms(display->drm, "[CONNECTOR:%d:%s]
> Panel prepare\n",
> +				    connector->base.base.id, connector-
> >base.name);
> +			intel_panel_prepare(crtc_state, conn_state);
> +		}
> +	}
> +
> +	drm_modeset_unlock(&display->drm-
> >mode_config.connection_mutex);
> +}
> +
>  const struct drm_panel_funcs dummy_panel_funcs = {  };
> 
> @@ -513,6 +552,8 @@ int intel_panel_register(struct intel_connector
> *connector)
>  		drm_dbg_kms(display->drm, "[CONNECTOR:%d:%s] Registered
> panel device '%s', has fwnode: %s\n",
>  			    connector->base.base.id, connector->base.name,
>  			    dev_name(dev), str_yes_no(dev_fwnode(dev)));
> +
> +		intel_panel_sync_state(connector);
>  	}
> 
>  out:
> --
> 2.39.5