drm/scdc: Document hotplug gotchas

There's some interactions between the SCDC setup and the disconnection /
reconnection of displays. Let's document it and a solution.

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
Link: https://lore.kernel.org/r/20220829134731.213478-9-maxime@cerno.tech

diff --git a/drivers/gpu/drm/display/drm_scdc_helper.c b/drivers/gpu/drm/display/drm_scdc_helper.c
index 81881e8..c3ad4ab 100644 (file)
--- a/drivers/gpu/drm/display/drm_scdc_helper.c
+++ b/drivers/gpu/drm/display/drm_scdc_helper.c
@@ -35,6 +35,19 @@
  * HDMI 2.0 specification. It is a point-to-point protocol that allows the
  * HDMI source and HDMI sink to exchange data. The same I2C interface that
  * is used to access EDID serves as the transport mechanism for SCDC.
+ *
+ * Note: The SCDC status is going to be lost when the display is
+ * disconnected. This can happen physically when the user disconnects
+ * the cable, but also when a display is switched on (such as waking up
+ * a TV).
+ *
+ * This is further complicated by the fact that, upon a disconnection /
+ * reconnection, KMS won't change the mode on its own. This means that
+ * one can't just rely on setting the SCDC status on enable, but also
+ * has to track the connector status changes using interrupts and
+ * restore the SCDC status. The typical solution for this is to trigger an
+ * empty modeset in drm_connector_helper_funcs.detect_ctx(), like what vc4 does
+ * in vc4_hdmi_reset_link().
  */
 
 #define SCDC_I2C_SLAVE_ADDRESS 0x54