ARM: Avoid branches to branches

Generally speaking, this optimization applies to all code
generation visitors ending with a call to Bind(), which
includes intrinsics with kNoCall CallKind. However, no
changes are done for slow paths (which frequently end with
a branch to an exit label that is bound at the end of a
visitor).

Test: m test-art-target
Change-Id: Ie1a0c8c54ef76b01e7f0b23962c56c29ca8984a9
diff --git a/compiler/optimizing/code_generator_arm.h b/compiler/optimizing/code_generator_arm.h
index 5b15902..59a7f7c 100644
--- a/compiler/optimizing/code_generator_arm.h
+++ b/compiler/optimizing/code_generator_arm.h
@@ -237,7 +237,7 @@
   void HandleBitwiseOperation(HBinaryOperation* operation);
   void HandleCondition(HCondition* condition);
   void HandleIntegerRotate(LocationSummary* locations);
-  void HandleLongRotate(LocationSummary* locations);
+  void HandleLongRotate(HRor* ror);
   void HandleShift(HBinaryOperation* operation);
 
   void GenerateWideAtomicStore(Register addr, uint32_t offset,