Fix a typo in VectorizeIfCondition.
This patch fixes an issue where the type of lhs operand was checked
twice and rhs type was ignored when processing conditions during
predicated vectorization.
The issue doesn't affect the actual vectorization behaviour.
Test: ./art/test.py --host --optimizing --jit
Test: ./art/test.py --target --optimizing --jit
Test: 661-checker-simd-cf-loops.
Test: target tests on arm64 with SVE (for details see
art/test/README.arm_fvp).
Change-Id: I486f29579a307548c321024a41c7fd30bdb97258
diff --git a/compiler/optimizing/loop_optimization.cc b/compiler/optimizing/loop_optimization.cc
index f62a355..0f8c95f 100644
--- a/compiler/optimizing/loop_optimization.cc
+++ b/compiler/optimizing/loop_optimization.cc
@@ -2739,7 +2739,7 @@
HInstruction* opa_promoted = opa;
HInstruction* opb_promoted = opb;
bool is_int_case = DataType::Type::kInt32 == opa->GetType() &&
- DataType::Type::kInt32 == opa->GetType();
+ DataType::Type::kInt32 == opb->GetType();
// Condition arguments should be either both int32 or consistently extended signed/unsigned
// narrower operands.