From: Ville Syrjälä <[email protected]> To avoid accidentally breaking things in the future add a comment explaining why we misconfigure the pipe_mask.
Also toss in a TODO for investigating a single encoder approach as opposed to the encoder-per-pipe approach. Signed-off-by: Ville Syrjälä <[email protected]> --- drivers/gpu/drm/i915/display/intel_dp_mst.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c b/drivers/gpu/drm/i915/display/intel_dp_mst.c index 7be82cf926ca..cb3047fe2d02 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_mst.c +++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c @@ -615,6 +615,18 @@ intel_dp_create_fake_mst_encoder(struct intel_digital_port *intel_dig_port, enum intel_encoder->power_domain = intel_dig_port->base.power_domain; intel_encoder->port = intel_dig_port->base.port; intel_encoder->cloneable = 0; + /* + * This is wrong, but broken userspace uses the intersection + * of possible_crtcs of all the encoders of a given connector + * to figure out which crtcs can drive said connector. What + * should be used instead is the union of possible_crtcs. + * To keep such userspace functioning we must misconfigure + * this to make sure the intersection is not empty :( + * + * TODO: figure out if we could eliminate the per-pipe + * encoders here and just have a single encoder for each + * MST connector. That would sidestep the userspace bug. + */ intel_encoder->pipe_mask = ~0; intel_encoder->compute_config = intel_dp_mst_compute_config; -- 2.21.0 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
