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
diff --git a/compiler/optimizing/loop_optimization.h b/compiler/optimizing/loop_optimization.h
index 2b202fd..1a842c4 100644
--- a/compiler/optimizing/loop_optimization.h
+++ b/compiler/optimizing/loop_optimization.h
@@ -82,6 +82,7 @@
kNoReduction = 1 << 9, // no reduction
kNoSAD = 1 << 10, // no sum of absolute differences (SAD)
kNoWideSAD = 1 << 11, // no sum of absolute differences (SAD) with operand widening
+ kNoDotProd = 1 << 12, // no dot product
};
/*
@@ -217,6 +218,11 @@
bool generate_code,
DataType::Type type,
uint64_t restrictions);
+ bool VectorizeDotProdIdiom(LoopNode* node,
+ HInstruction* instruction,
+ bool generate_code,
+ DataType::Type type,
+ uint64_t restrictions);
// Vectorization heuristics.
Alignment ComputeAlignment(HInstruction* offset,