aco: enable nir_opt_sink

SGPRS: 880272 -> 838936 (-4.70 %)
VGPRS: 705316 -> 680988 (-3.45 %)
Spilled SGPRs: 1032 -> 832 (-19.38 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Private memory VGPRs: 0 -> 0 (0.00 %)
Scratch size: 252 -> 252 (0.00 %) dwords per thread
Code Size: 55150788 -> 55172436 (0.04 %) bytes
LDS: 451 -> 451 (0.00 %) blocks
Max Waves: 66178 -> 68706 (3.82 %)

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>

diff --git a/src/amd/compiler/aco_instruction_selection_setup.cpp b/src/amd/compiler/aco_instruction_selection_setup.cpp
index 2d06ac87b9d990a08837f6117a2751e1de970415..7505707d4bb904e1358a94dfa5a7daab2c4f034a 100644 (file)
--- a/src/amd/compiler/aco_instruction_selection_setup.cpp
+++ b/src/amd/compiler/aco_instruction_selection_setup.cpp
@@ -1343,7 +1343,7 @@ setup_isel_context(Program* program,
       nir_opt_shrink_load(nir);
       nir_move_options move_opts = (nir_move_options)(
          nir_move_const_undef | nir_move_load_ubo | nir_move_load_input | nir_move_comparisons);
-      //nir_opt_sink(nir, move_opts); // TODO: enable this once !1664 is merged
+      nir_opt_sink(nir, move_opts);
       nir_opt_move(nir, move_opts);
       nir_convert_to_lcssa(nir, true, false);
       nir_lower_phis_to_scalar(nir);