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

44
Eigen/src/Core/Functors.h
View File

@@ -34,12 +34,15 @@
*/
template<typename Scalar> struct ei_scalar_sum_op EIGEN_EMPTY_STRUCT {
const Scalar operator() (const Scalar& a, const Scalar& b) const { return a + b; }
template<typename PacketScalar>
PacketScalar packetOp(const PacketScalar& a, const PacketScalar& b) const
{ return ei_padd(a,b); }
};
template<typename Scalar>
struct ei_functor_traits<ei_scalar_sum_op<Scalar> > {
enum {
Cost = NumTraits<Scalar>::AddCost,
IsVectorizable = NumTraits<Scalar>::PacketSize>0
IsVectorizable = ei_packet_traits<Scalar>::size>1
};
};
@@ -50,12 +53,15 @@ struct ei_functor_traits<ei_scalar_sum_op<Scalar> > {
*/
template<typename Scalar> struct ei_scalar_product_op EIGEN_EMPTY_STRUCT {
const Scalar operator() (const Scalar& a, const Scalar& b) const { return a * b; }
template<typename PacketScalar>
PacketScalar packetOp(const PacketScalar& a, const PacketScalar& b) const
{ return ei_pmul(a,b); }
};
template<typename Scalar>
struct ei_functor_traits<ei_scalar_product_op<Scalar> > {
enum {
Cost = NumTraits<Scalar>::MulCost,
IsVectorizable = NumTraits<Scalar>::PacketSize>0
IsVectorizable = ei_packet_traits<Scalar>::size>1
};
};
@@ -66,12 +72,15 @@ struct ei_functor_traits<ei_scalar_product_op<Scalar> > {
*/
template<typename Scalar> struct ei_scalar_min_op EIGEN_EMPTY_STRUCT {
const Scalar operator() (const Scalar& a, const Scalar& b) const { return std::min(a, b); }
template<typename PacketScalar>
PacketScalar packetOp(const PacketScalar& a, const PacketScalar& b) const
{ return ei_pmin(a,b); }
};
template<typename Scalar>
struct ei_functor_traits<ei_scalar_min_op<Scalar> > {
enum {
Cost = NumTraits<Scalar>::AddCost,
IsVectorizable = NumTraits<Scalar>::PacketSize>0
IsVectorizable = ei_packet_traits<Scalar>::size>1
};
};
@@ -82,12 +91,15 @@ struct ei_functor_traits<ei_scalar_min_op<Scalar> > {
*/
template<typename Scalar> struct ei_scalar_max_op EIGEN_EMPTY_STRUCT {
const Scalar operator() (const Scalar& a, const Scalar& b) const { return std::max(a, b); }
template<typename PacketScalar>
PacketScalar packetOp(const PacketScalar& a, const PacketScalar& b) const
{ return ei_pmax(a,b); }
};
template<typename Scalar>
struct ei_functor_traits<ei_scalar_max_op<Scalar> > {
enum {
Cost = NumTraits<Scalar>::AddCost,
IsVectorizable = NumTraits<Scalar>::PacketSize>0
IsVectorizable = ei_packet_traits<Scalar>::size>1
};
};
@@ -100,13 +112,16 @@ struct ei_functor_traits<ei_scalar_max_op<Scalar> > {
* \sa class CwiseBinaryOp, MatrixBase::operator-
*/
template<typename Scalar> struct ei_scalar_difference_op EIGEN_EMPTY_STRUCT {
const Scalar operator() (const Scalar& a, const Scalar& b) const { return a - b; }
const Scalar operator() (const Scalar& a, const Scalar& b) const { return a - b; }
template<typename PacketScalar>
PacketScalar packetOp(const PacketScalar& a, const PacketScalar& b) const
{ return ei_psub(a,b); }
};
template<typename Scalar>
struct ei_functor_traits<ei_scalar_difference_op<Scalar> > {
enum {
Cost = NumTraits<Scalar>::AddCost,
IsVectorizable = NumTraits<Scalar>::PacketSize>0
IsVectorizable = ei_packet_traits<Scalar>::size>1
};
};
@@ -194,15 +209,26 @@ struct ei_functor_traits<ei_scalar_cast_op<Scalar,NewType> >
*
* \sa class CwiseUnaryOp, MatrixBase::operator*, MatrixBase::operator/
*/
template<typename Scalar, bool IsVectorizable = (int(ei_packet_traits<Scalar>::size)>1?true:false) > struct ei_scalar_multiple_op;
template<typename Scalar>
struct ei_scalar_multiple_op {
ei_scalar_multiple_op(const Scalar& other) : m_other(other) {}
struct ei_scalar_multiple_op<Scalar,true> {
typedef typename ei_packet_traits<Scalar>::type PacketScalar;
ei_scalar_multiple_op(const Scalar& other) : m_other(ei_pset1(other)) { }
Scalar operator() (const Scalar& a) const { return a * ei_pfirst(m_other); }
PacketScalar packetOp(const PacketScalar& a) const
{ return ei_pmul(a, m_other); }
const PacketScalar m_other;
};
template<typename Scalar>
struct ei_scalar_multiple_op<Scalar,false> {
ei_scalar_multiple_op(const Scalar& other) : m_other(other) { }
Scalar operator() (const Scalar& a) const { return a * m_other; }
const Scalar m_other;
};
template<typename Scalar>
struct ei_functor_traits<ei_scalar_multiple_op<Scalar> >
{ enum { Cost = NumTraits<Scalar>::MulCost, IsVectorizable = false }; };
{ enum { Cost = NumTraits<Scalar>::MulCost, IsVectorizable = ei_packet_traits<Scalar>::size>1 }; };
template<typename Scalar, bool HasFloatingPoint>
struct ei_scalar_quotient1_impl {