ART: ARM64: Support DotProd SIMD idiom.

Implement support for vectorization idiom which performs dot
product of two vectors and adds the result to wider precision
components in the accumulator.

viz. DOT_PRODUCT([ a1, .. , am], [ x1, .. , xn ], [ y1, .. , yn ]) =
                 [ a1 + sum(xi * yi), .. , am + sum(xj * yj) ],
     for m <= n, non-overlapping sums,
     for either both signed or both unsigned operands x, y.

The patch shows up to 7x performance improvement on a micro
benchmark on Cortex-A57.

Test: 684-checker-simd-dotprod.
Test: test-art-host, test-art-target.

Change-Id: Ibab0d51f537fdecd1d84033197be3ebf5ec4e455

1 files changed, 8 insertions, 0 deletions

diff --git a/compiler/optimizing/graph_visualizer.cc b/compiler/optimizing/graph_visualizer.cc
index 31db8c205f..21f22af3c3 100644
--- a/compiler/optimizing/graph_visualizer.cc
+++ b/compiler/optimizing/graph_visualizer.cc
@@ -564,6 +564,14 @@ class HGraphVisualizerPrinter : public HGraphDelegateVisitor {
     StartAttributeStream("kind") << instruction->GetOpKind();
   }
 
+  void VisitVecDotProd(HVecDotProd* instruction) override {
+    VisitVecOperation(instruction);
+    DataType::Type arg_type = instruction->InputAt(1)->AsVecOperation()->GetPackedType();
+    StartAttributeStream("type") << (instruction->IsZeroExtending() ?
+                                    DataType::ToUnsigned(arg_type) :
+                                    DataType::ToSigned(arg_type));
+  }
+
 #if defined(ART_ENABLE_CODEGEN_arm) || defined(ART_ENABLE_CODEGEN_arm64)
   void VisitMultiplyAccumulate(HMultiplyAccumulate* instruction) override {
     StartAttributeStream("kind") << instruction->GetOpKind();