Use ScopedArenaAllocator for register allocation.
Memory needed to compile the two most expensive methods for
aosp_angler-userdebug boot image:
BatteryStats.dumpCheckinLocked() : 25.1MiB -> 21.1MiB
BatteryStats.dumpLocked(): 49.6MiB -> 42.0MiB
This is because all the memory previously used by Scheduler
is reused by the register allocator; the register allocator
has a higher peak usage of the ArenaStack.
And continue the "arena"->"allocator" renaming.
Test: m test-art-host-gtest
Test: testrunner.py --host
Bug: 64312607
Change-Id: Idfd79a9901552b5147ec0bf591cb38120de86b01
diff --git a/compiler/optimizing/ssa_liveness_analysis_test.cc b/compiler/optimizing/ssa_liveness_analysis_test.cc
index 82ee441..9b78e0e 100644
--- a/compiler/optimizing/ssa_liveness_analysis_test.cc
+++ b/compiler/optimizing/ssa_liveness_analysis_test.cc
@@ -75,7 +75,7 @@
block->AddInstruction(new (GetAllocator()) HExit());
graph_->BuildDominatorTree();
- SsaLivenessAnalysis ssa_analysis(graph_, codegen_.get());
+ SsaLivenessAnalysis ssa_analysis(graph_, codegen_.get(), GetScopedAllocator());
ssa_analysis.Analyze();
std::ostringstream arg_dump;
@@ -127,7 +127,7 @@
block->AddInstruction(array_set);
graph_->BuildDominatorTree();
- SsaLivenessAnalysis ssa_analysis(graph_, codegen_.get());
+ SsaLivenessAnalysis ssa_analysis(graph_, codegen_.get(), GetScopedAllocator());
ssa_analysis.Analyze();
EXPECT_FALSE(graph_->IsDebuggable());
@@ -201,7 +201,7 @@
block->AddInstruction(array_set);
graph_->BuildDominatorTree();
- SsaLivenessAnalysis ssa_analysis(graph_, codegen_.get());
+ SsaLivenessAnalysis ssa_analysis(graph_, codegen_.get(), GetScopedAllocator());
ssa_analysis.Analyze();
EXPECT_FALSE(graph_->IsDebuggable());