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

Added initial experimental support for explicit vectorization.

Browse Source 
Currently only the following platform/operations are supported:
 - SSE2 compatible architecture
 - compiler compatible with intel's SSE2 intrinsics
 - float, double and int data types
 - fixed size matrices with a storage major dimension multiple of 4 (or 2 for double)
 - scalar-matrix product, component wise: +,-,*,min,max
 - matrix-matrix product only if the left matrix is vectorizable and column major
   or the right matrix is vectorizable and row major, e.g.:
   a.transpose() * b is not vectorized with the default column major storage.
To use it you must define EIGEN_VECTORIZE and EIGEN_INTEL_PLATFORM.
This commit is contained in:
Gael Guennebaud
2008-04-09 12:31:55 +00:00
parent 4920f2011e
commit 1985fb0551
25 changed files with 436 additions and 93 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
Eigen/src/Core/Redux.h
View File

@@ -94,9 +94,9 @@ struct ei_traits<PartialRedux<Direction, BinaryOp, MatrixType> >
ColsAtCompileTime = Direction==Horizontal ? 1 : MatrixType::ColsAtCompileTime,
MaxRowsAtCompileTime = Direction==Vertical ? 1 : MatrixType::MaxRowsAtCompileTime,
MaxColsAtCompileTime = Direction==Horizontal ? 1 : MatrixType::MaxColsAtCompileTime,
Flags = (RowsAtCompileTime == Dynamic || ColsAtCompileTime == Dynamic)
Flags = ((RowsAtCompileTime == Dynamic || ColsAtCompileTime == Dynamic)
? (unsigned int)_MatrixTypeXprCopy::Flags
: (unsigned int)_MatrixTypeXprCopy::Flags & ~LargeBit,
: (unsigned int)_MatrixTypeXprCopy::Flags & ~LargeBit) & ~VectorizableBit,
TraversalSize = Direction==Vertical ? RowsAtCompileTime : ColsAtCompileTime,
CoeffReadCost = TraversalSize * _MatrixTypeXprCopy::CoeffReadCost
+ (TraversalSize - 1) * ei_functor_traits<BinaryOp>::Cost