devtools/eigen
1
0
Fork 0
mirror of https://gitlab.com/libeigen/eigen.git synced 2026-04-10 11:34:33 +08:00
Code Issues Packages Projects Releases Wiki Activity

Extend support for Packet16b:

Browse Source 
* Add ptranspose<*,4> to support matmul and add unit test for Matrix<bool> * Matrix<bool>
* work around a bug in slicing of Tensor<bool>.
* Add tensor tests

This speeds up matmul for boolean matrices by about 10x

name                            old time/op             new time/op             delta
BM_MatMul<bool>/8                267ns ± 0%              479ns ± 0%  +79.25%          (p=0.008 n=5+5)
BM_MatMul<bool>/32              6.42µs ± 0%             0.87µs ± 0%  -86.50%          (p=0.008 n=5+5)
BM_MatMul<bool>/64              43.3µs ± 0%              5.9µs ± 0%  -86.42%          (p=0.008 n=5+5)
BM_MatMul<bool>/128              315µs ± 0%               44µs ± 0%  -85.98%          (p=0.008 n=5+5)
BM_MatMul<bool>/256             2.41ms ± 0%             0.34ms ± 0%  -85.68%          (p=0.008 n=5+5)
BM_MatMul<bool>/512             18.8ms ± 0%              2.7ms ± 0%  -85.53%          (p=0.008 n=5+5)
BM_MatMul<bool>/1k               149ms ± 0%               22ms ± 0%  -85.40%          (p=0.008 n=5+5)
This commit is contained in:
Rasmus Munk Larsen
2020-04-24 17:29:25 -07:00
parent b47c777993
commit ab773c7e91
10 changed files with 267 additions and 162 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
unsupported/Eigen/CXX11/src/Tensor/TensorMorphing.h
View File

@@ -456,7 +456,9 @@ struct TensorEvaluator<const TensorSlicingOp<StartIndices, Sizes, ArgType>, Devi
// slice offsets and sizes.
IsAligned = false,
PacketAccess = TensorEvaluator<ArgType, Device>::PacketAccess,
BlockAccess = TensorEvaluator<ArgType, Device>::BlockAccess,
BlockAccess = TensorEvaluator<ArgType, Device>::BlockAccess &&
// FIXME: Temporary workaround for bug in slicing of bool tensors.
!internal::is_same<typename internal::remove_const<Scalar>::type, bool>::value,
PreferBlockAccess = true,
Layout = TensorEvaluator<ArgType, Device>::Layout,
CoordAccess = false,
@@ -525,7 +527,6 @@ struct TensorEvaluator<const TensorSlicingOp<StartIndices, Sizes, ArgType>, Devi
EIGEN_DEVICE_FUNC EIGEN_STRONG_INLINE const Dimensions& dimensions() const { return m_dimensions; }
EIGEN_DEVICE_FUNC EIGEN_STRONG_INLINE bool evalSubExprsIfNeeded(EvaluatorPointerType data) {
m_impl.evalSubExprsIfNeeded(NULL);
if (!NumTraits<typename internal::remove_const<Scalar>::type>::RequireInitialization
@@ -559,14 +560,14 @@ struct TensorEvaluator<const TensorSlicingOp<StartIndices, Sizes, ArgType>, Devi
}
return true;
}
#ifdef EIGEN_USE_THREADS
template <typename EvalSubExprsCallback>
EIGEN_DEVICE_FUNC EIGEN_STRONG_INLINE void evalSubExprsIfNeededAsync(
EvaluatorPointerType data, EvalSubExprsCallback done) {
m_impl.evalSubExprsIfNeededAsync(nullptr, [done](bool) { done(true); });
}
#endif // EIGEN_USE_THREADS
#endif // EIGEN_USE_THREADS
EIGEN_DEVICE_FUNC EIGEN_STRONG_INLINE void cleanup() {
m_impl.cleanup();