diff options
| author | 2023-04-18 15:16:06 +0100 | |
|---|---|---|
| committer | 2023-04-19 15:31:53 +0000 | |
| commit | 5eb1fd0dae3832ceee2102613bb08c291daca6f3 (patch) | |
| tree | da074eb8ea773b594a455fa9cd7ddde21d8f885b /compiler/optimizing/optimizing_compiler.cc | |
| parent | 2ff7fcccdeb2100cb5a3cc5314411fd23dd5f7c9 (diff) | |
Restrict the use of ConstantFolding's VisitIf
It was taking a lot of time for the improvement it got. We can get
99.99% of the improvement, with only one VisitIf call. This is
roughly 20% of the compile time it used to take.
Bug: 278626992
Fixes: 278626992
Test: art/test/testrunner/testrunner.py --host --64 --optimizing -b
Change-Id: Icc00c9ad6a9eb4f4fd18677bcb65655cbbe9d027
Diffstat (limited to 'compiler/optimizing/optimizing_compiler.cc')
| -rw-r--r-- | compiler/optimizing/optimizing_compiler.cc | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/compiler/optimizing/optimizing_compiler.cc b/compiler/optimizing/optimizing_compiler.cc index 60d5e6f3bd..b97dc881eb 100644 --- a/compiler/optimizing/optimizing_compiler.cc +++ b/compiler/optimizing/optimizing_compiler.cc @@ -656,7 +656,7 @@ void OptimizingCompiler::RunOptimizations(HGraph* graph, OptDef(OptimizationPass::kGlobalValueNumbering), // Simplification (TODO: only if GVN occurred). OptDef(OptimizationPass::kSelectGenerator), - OptDef(OptimizationPass::kConstantFolding, + OptDef(OptimizationPass::kAggressiveConstantFolding, "constant_folding$after_gvn"), OptDef(OptimizationPass::kInstructionSimplifier, "instruction_simplifier$after_gvn"), |