diff options
| author | Artem Serov <artem.serov@linaro.org> | 2018-08-07 00:52:22 +0100 |
|---|---|---|
| committer | Artem Serov <artem.serov@linaro.org> | 2018-09-25 14:47:48 +0100 |
| commit | aaac0e3cbfe72217cad204d0122f2b73a602d2dd (patch) | |
| tree | d148274452b3a409c9d6b8ef749c34185375d2ea /compiler/optimizing/loop_optimization.h | |
| parent | 7dca45b9677c16a54347cdc0d08bfa2bdd94b464 (diff) | |
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
Diffstat (limited to 'compiler/optimizing/loop_optimization.h')
| -rw-r--r-- | compiler/optimizing/loop_optimization.h | 6 |
1 files changed, 6 insertions, 0 deletions
diff --git a/compiler/optimizing/loop_optimization.h b/compiler/optimizing/loop_optimization.h index 2b202fda75..1a842c4bf3 100644 --- a/compiler/optimizing/loop_optimization.h +++ b/compiler/optimizing/loop_optimization.h @@ -82,6 +82,7 @@ class HLoopOptimization : public HOptimization { 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 @@ class HLoopOptimization : public HOptimization { 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, |