aco: fix spills_entry heuristic for branch blocks in init_live_in_vars()

fossil-db (Navi):
Totals from 222 (0.16% of 135946) affected shaders:
SpillSGPRs: 9121 -> 9117 (-0.04%)
SpillVGPRs: 2820 -> 1821 (-35.43%)
CodeSize: 5134264 -> 5053336 (-1.58%); split: -1.63%, +0.05%
Instrs: 953435 -> 938761 (-1.54%); split: -1.59%, +0.05%
Cycles: 100567688 -> 97252432 (-3.30%); split: -3.34%, +0.04%
VMEM: 40752 -> 39219 (-3.76%); split: +0.04%, -3.80%
SMEM: 15416 -> 15509 (+0.60%); split: +0.64%, -0.03%
VClause: 20120 -> 19091 (-5.11%)
SClause: 23540 -> 23544 (+0.02%); split: -0.11%, +0.12%
Copies: 125912 -> 122017 (-3.09%); split: -3.36%, +0.26%
Branches: 31131 -> 30009 (-3.60%)

Mostly affects parallel-rdp ubershaders.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Cc: 20.2 <mesa-stable>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6212>

diff --git a/src/amd/compiler/aco_spill.cpp b/src/amd/compiler/aco_spill.cpp
index 45c67469d94f97c6636183f009c56d92151c4a0b..94364bfa54ec1de365ce7fe404985b62844d1591 100644 (file)
--- a/src/amd/compiler/aco_spill.cpp
+++ b/src/amd/compiler/aco_spill.cpp
@@ -502,7 +502,7 @@ RegisterDemand init_live_in_vars(spill_ctx& ctx, Block* block, unsigned block_id
       for (std::pair<Temp, uint32_t> pair : ctx.spills_exit[pred_idx]) {
          if (pair.first.type() == RegType::sgpr &&
              ctx.next_use_distances_start[block_idx].find(pair.first) != ctx.next_use_distances_start[block_idx].end() &&
       for (std::pair<Temp, uint32_t> pair : ctx.spills_exit[pred_idx]) {
          if (pair.first.type() == RegType::sgpr &&
              ctx.next_use_distances_start[block_idx].find(pair.first) != ctx.next_use_distances_start[block_idx].end() &&

-             ctx.next_use_distances_start[block_idx][pair.first].second > block_idx) {
+             ctx.next_use_distances_start[block_idx][pair.first].first != block_idx) {

             ctx.spills_entry[block_idx].insert(pair);
             spilled_registers.sgpr += pair.first.size();
          }
             ctx.spills_entry[block_idx].insert(pair);
             spilled_registers.sgpr += pair.first.size();
          }


@@ -512,7 +512,7 @@ RegisterDemand init_live_in_vars(spill_ctx& ctx, Block* block, unsigned block_id
          for (std::pair<Temp, uint32_t> pair : ctx.spills_exit[pred_idx]) {
             if (pair.first.type() == RegType::vgpr &&
                 ctx.next_use_distances_start[block_idx].find(pair.first) != ctx.next_use_distances_start[block_idx].end() &&
          for (std::pair<Temp, uint32_t> pair : ctx.spills_exit[pred_idx]) {
             if (pair.first.type() == RegType::vgpr &&
                 ctx.next_use_distances_start[block_idx].find(pair.first) != ctx.next_use_distances_start[block_idx].end() &&

-                ctx.next_use_distances_start[block_idx][pair.first].second > block_idx) {
+                ctx.next_use_distances_start[block_idx][pair.first].first != block_idx) {

                ctx.spills_entry[block_idx].insert(pair);
                spilled_registers.vgpr += pair.first.size();
             }
                ctx.spills_entry[block_idx].insert(pair);
                spilled_registers.vgpr += pair.first.size();
             }