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

Change inline hint for general_matrix_vector_product<>::run() to gain performance

Browse Source 
libeigen/eigen!2092
This commit is contained in:
Yu You
2026-01-09 11:46:37 -08:00
committed by Antonio Sánchez
parent 7aea350ba1
commit 0315fb319a
1 changed files with 6 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
Eigen/src/Core/products/GeneralMatrixVector.h
View File

@@ -97,14 +97,13 @@ struct general_matrix_vector_product<Index, LhsScalar, LhsMapper, ColMajor, Conj
typedef typename QuarterTraits::RhsPacket RhsPacketQuarter;
typedef typename QuarterTraits::ResPacket ResPacketQuarter;
EIGEN_DEVICE_FUNC EIGEN_DONT_INLINE static void run(Index rows, Index cols, const LhsMapper& lhs,
const RhsMapper& rhs, ResScalar* res, Index resIncr,
RhsScalar alpha);
EIGEN_DEVICE_FUNC inline static void run(Index rows, Index cols, const LhsMapper& lhs, const RhsMapper& rhs,
ResScalar* res, Index resIncr, RhsScalar alpha);
};
template <typename Index, typename LhsScalar, typename LhsMapper, bool ConjugateLhs, typename RhsScalar,
typename RhsMapper, bool ConjugateRhs, int Version>
EIGEN_DEVICE_FUNC EIGEN_DONT_INLINE void
EIGEN_DEVICE_FUNC inline void
general_matrix_vector_product<Index, LhsScalar, LhsMapper, ColMajor, ConjugateLhs, RhsScalar, RhsMapper, ConjugateRhs,
Version>::run(Index rows, Index cols, const LhsMapper& alhs, const RhsMapper& rhs,
ResScalar* res, Index resIncr, RhsScalar alpha) {
@@ -290,14 +289,13 @@ struct general_matrix_vector_product<Index, LhsScalar, LhsMapper, RowMajor, Conj
typedef typename QuarterTraits::RhsPacket RhsPacketQuarter;
typedef typename QuarterTraits::ResPacket ResPacketQuarter;
EIGEN_DEVICE_FUNC EIGEN_DONT_INLINE static void run(Index rows, Index cols, const LhsMapper& lhs,
const RhsMapper& rhs, ResScalar* res, Index resIncr,
ResScalar alpha);
EIGEN_DEVICE_FUNC static inline void run(Index rows, Index cols, const LhsMapper& lhs, const RhsMapper& rhs,
ResScalar* res, Index resIncr, ResScalar alpha);
};
template <typename Index, typename LhsScalar, typename LhsMapper, bool ConjugateLhs, typename RhsScalar,
typename RhsMapper, bool ConjugateRhs, int Version>
EIGEN_DEVICE_FUNC EIGEN_DONT_INLINE void
EIGEN_DEVICE_FUNC inline void
general_matrix_vector_product<Index, LhsScalar, LhsMapper, RowMajor, ConjugateLhs, RhsScalar, RhsMapper, ConjugateRhs,
Version>::run(Index rows, Index cols, const LhsMapper& alhs, const RhsMapper& rhs,
ResScalar* res, Index resIncr, ResScalar alpha) {