Implement irreducible loop support in optimizing.

So we don't fallback to the interpreter in the presence of
irreducible loops.

Implications:
- A loop pre-header does not necessarily dominate a loop header.
- Non-constant redundant phis will be kept in loop headers, to
  satisfy our linear scan register allocation algorithm.
- while-graph optimizations, such as gvn, licm, lse, and dce
  need to know when they are dealing with irreducible loops.

Change-Id: I2cea8934ce0b40162d215353497c7f77d6c9137e

1 files changed, 1 insertions, 1 deletions

diff --git a/compiler/optimizing/optimizing_unit_test.h b/compiler/optimizing/optimizing_unit_test.h
index af3a005304..5a910433b4 100644
--- a/compiler/optimizing/optimizing_unit_test.h
+++ b/compiler/optimizing/optimizing_unit_test.h
@@ -117,7 +117,7 @@ inline bool IsRemoved(HInstruction* instruction) {
 inline void TransformToSsa(HGraph* graph) {
   ScopedObjectAccess soa(Thread::Current());
   StackHandleScopeCollection handles(soa.Self());
-  EXPECT_EQ(graph->TryBuildingSsa(&handles), kBuildSsaSuccess);
+  EXPECT_EQ(graph->TryBuildingSsa(&handles), kAnalysisSuccess);
 }
 
 }  // namespace art