Optimizing: Run gtests without creating the Runtime.

The only Optimizing test that actually needs a Runtime is
the ReferenceTypePropagationTest, so we make it subclass
CommonCompilerTest explicitly and change OptimizingUnitTest
to subclass CommonArtTest for the other tests.

On host, each test that initializes the Runtime takes ~220ms
more than without initializing the Runtime. For example, the
ConstantFoldingTest that has 10 individual tests previously
took over 2.2s to run but without the Runtime initialization
it takes around 3-5ms. On target, running 32-bit gtests on
taimen with run-gtests.sh (single-threaded) goes from
~28m47s to ~26m13s, a reduction of ~9%.

Test: m test-art-host-gtest
Test: run-gtests.sh
Change-Id: I43e50ed58e52cc0ad04cdb4d39801bfbae840a3d

1 files changed, 4 insertions, 2 deletions

diff --git a/compiler/optimizing/liveness_test.cc b/compiler/optimizing/liveness_test.cc
index f11f7a9779..ba3787e9be 100644
--- a/compiler/optimizing/liveness_test.cc
+++ b/compiler/optimizing/liveness_test.cc
@@ -47,8 +47,10 @@ static void DumpBitVector(BitVector* vector,
 void LivenessTest::TestCode(const std::vector<uint16_t>& data, const char* expected) {
   HGraph* graph = CreateCFG(data);
   // `Inline` conditions into ifs.
-  PrepareForRegisterAllocation(graph, *compiler_options_).Run();
-  std::unique_ptr<CodeGenerator> codegen = CodeGenerator::Create(graph, *compiler_options_);
+  std::unique_ptr<CompilerOptions> compiler_options =
+      CommonCompilerTest::CreateCompilerOptions(kRuntimeISA, "default");
+  PrepareForRegisterAllocation(graph, *compiler_options).Run();
+  std::unique_ptr<CodeGenerator> codegen = CodeGenerator::Create(graph, *compiler_options);
   SsaLivenessAnalysis liveness(graph, codegen.get(), GetScopedAllocator());
   liveness.Analyze();