Skip to content

radv: fix R_02881C_PA_CL_VS_OUT_CNTL with mixed cull/clip distances

Rhys Perry requested to merge pendingchaos/mesa:radv_zink_clip_cull_dist_fix into main

Matches radeonsi.

Seems Vulkan CTS doesn't really test cull distances. Removing VARYING_SLOT_CULL_DIST0/VARYING_SLOT_CULL_DIST1 variables doesn't break any of dEQP-VK.clipping.*, except for tests which read the variables in the fragment shader.

Merge request reports