Comment 16 for bug 1749420

Attached is the difference between kernel 4.7 and 4.8 in the Intel graphics driver. It's still a big diff, but at least eliminates my above theories about the cause of the regression. Now the main suspect, I think, would be:

@@ -12095,21 +12181,11 @@ connected_sink_compute_bpp(struct intel_connector *connector,
                pipe_config->pipe_bpp = connector->base.display_info.bpc*3;
        }

- /* Clamp bpp to default limit on screens without EDID 1.4 */
- if (connector->base.display_info.bpc == 0) {
- int type = connector->base.connector_type;
- int clamp_bpp = 24;
-
- /* Fall back to 18 bpp when DP sink capability is unknown. */
- if (type == DRM_MODE_CONNECTOR_DisplayPort ||
- type == DRM_MODE_CONNECTOR_eDP)
- clamp_bpp = 18;
-
- if (bpp > clamp_bpp) {
- DRM_DEBUG_KMS("clamping display bpp (was %d) to default limit of %d\n",
- bpp, clamp_bpp);
- pipe_config->pipe_bpp = clamp_bpp;
- }
+ /* Clamp bpp to 8 on screens without EDID 1.4 */
+ if (connector->base.display_info.bpc == 0 && bpp > 24) {
+ DRM_DEBUG_KMS("clamping display bpp (was %d) to default limit of 24\n",
+ bpp);
+ pipe_config->pipe_bpp = 24;
        }
 }