i965/cnl: Don't write to Cache Mode Register 1 on gen10+
authorAnuj Phogat <anuj.phogat@gmail.com>
Tue, 13 Jun 2017 21:22:06 +0000 (14:22 -0700)
committerAnuj Phogat <anuj.phogat@gmail.com>
Fri, 23 Jun 2017 18:16:00 +0000 (11:16 -0700)
With below optimizations gone in gen10+ we have nothing left out to
write to CACHE_MODE_1:
Float Blend Optimization Enable: This bit have been removed in gen10+
Partial Resolve Disable in VC: Recommendation is to always set this
field to 0 in gen10+ and that's the default value of the bit.

Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
src/mesa/drivers/dri/i965/brw_state_upload.c

index 926597b6258e9e3ebb3d106848bb334c55cb96e1..5e82c1b4ce80a40f6640800faf68c3dfa4f6a95a 100644 (file)
@@ -60,8 +60,10 @@ brw_upload_initial_gpu_state(struct brw_context *brw)
 
    brw_upload_invariant_state(brw);
 
-   /* Recommended optimization for Victim Cache eviction in pixel backend. */
-   if (brw->gen >= 9) {
+   if (brw->gen == 9) {
+      /* Recommended optimizations for Victim Cache eviction and floating
+       * point blending.
+       */
       BEGIN_BATCH(3);
       OUT_BATCH(MI_LOAD_REGISTER_IMM | (3 - 2));
       OUT_BATCH(GEN7_CACHE_MODE_1);