GitHub-Proxy/eigen
1
0
Fork 0
mirror of https://gitlab.com/libeigen/eigen.git synced 2025-08-12 03:39:01 +08:00
Code Issues Packages Projects Releases Wiki Activity

packet math functions:

Browse Source 
- take const Packet& args like the other packet funcs
 - SSE specializations: make them be actual template specializations
This commit is contained in:
Benoit Jacob 2010-06-15 08:29:21 -04:00
parent 7958797648
commit 134ca4acb3
2 changed files with 19 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
Eigen/src/Core/GenericPacketMath.h
View File

@ -210,19 +210,19 @@ template<typename Packet> inline Packet ei_preverse(const Packet& a)
***************************/
/** \internal \returns the sin of \a a (coeff-wise) */
template<typename Packet> inline static Packet ei_psin(Packet a) { return ei_sin(a); }
template<typename Packet> inline static Packet ei_psin(const Packet& a) { return ei_sin(a); }
/** \internal \returns the cos of \a a (coeff-wise) */
template<typename Packet> inline static Packet ei_pcos(Packet a) { return ei_cos(a); }
template<typename Packet> inline static Packet ei_pcos(const Packet& a) { return ei_cos(a); }
/** \internal \returns the exp of \a a (coeff-wise) */
template<typename Packet> inline static Packet ei_pexp(Packet a) { return ei_exp(a); }
template<typename Packet> inline static Packet ei_pexp(const Packet& a) { return ei_exp(a); }
/** \internal \returns the log of \a a (coeff-wise) */
template<typename Packet> inline static Packet ei_plog(Packet a) { return ei_log(a); }
template<typename Packet> inline static Packet ei_plog(const Packet& a) { return ei_log(a); }
/** \internal \returns the square-root of \a a (coeff-wise) */
template<typename Packet> inline static Packet ei_psqrt(Packet a) { return ei_sqrt(a); }
template<typename Packet> inline static Packet ei_psqrt(const Packet& a) { return ei_sqrt(a); }
/***************************************************************************
* The following functions might not have to be overwritten for vectorized types

19
Eigen/src/Core/arch/SSE/MathFunctions.h
View File

@ -30,8 +30,10 @@
#ifndef EIGEN_MATH_FUNCTIONS_SSE_H
#define EIGEN_MATH_FUNCTIONS_SSE_H
static EIGEN_DONT_INLINE EIGEN_UNUSED Packet4f ei_plog(Packet4f x)
template<> EIGEN_DONT_INLINE EIGEN_UNUSED
Packet4f ei_plog<Packet4f>(const Packet4f& _x)
{
Packet4f x = _x;
_EIGEN_DECLARE_CONST_Packet4f(1 , 1.0f);
_EIGEN_DECLARE_CONST_Packet4f(half, 0.5f);
_EIGEN_DECLARE_CONST_Packet4i(0x7f, 0x7f);
@ -108,8 +110,10 @@ static EIGEN_DONT_INLINE EIGEN_UNUSED Packet4f ei_plog(Packet4f x)
return _mm_or_ps(x, invalid_mask); // negative arg will be NAN
}
static EIGEN_DONT_INLINE EIGEN_UNUSED Packet4f ei_pexp(Packet4f x)
template<> EIGEN_DONT_INLINE EIGEN_UNUSED
Packet4f ei_pexp<Packet4f>(const Packet4f& _x)
{
Packet4f x = _x;
_EIGEN_DECLARE_CONST_Packet4f(1 , 1.0f);
_EIGEN_DECLARE_CONST_Packet4f(half, 0.5f);
_EIGEN_DECLARE_CONST_Packet4i(0x7f, 0x7f);
@ -181,8 +185,10 @@ static EIGEN_DONT_INLINE EIGEN_UNUSED Packet4f ei_pexp(Packet4f x)
surprising but correct result.
*/
static EIGEN_DONT_INLINE EIGEN_UNUSED Packet4f ei_psin(Packet4f x)
template<> EIGEN_DONT_INLINE EIGEN_UNUSED
Packet4f ei_psin<Packet4f>(const Packet4f& _x)
{
Packet4f x = _x;
_EIGEN_DECLARE_CONST_Packet4f(1 , 1.0f);
_EIGEN_DECLARE_CONST_Packet4f(half, 0.5f);
@ -280,8 +286,10 @@ static EIGEN_DONT_INLINE EIGEN_UNUSED Packet4f ei_psin(Packet4f x)
}
/* almost the same as ei_psin */
static EIGEN_DONT_INLINE EIGEN_UNUSED Packet4f ei_pcos(Packet4f x)
template<> EIGEN_DONT_INLINE EIGEN_UNUSED
Packet4f ei_pcos<Packet4f>(const Packet4f& _x)
{
Packet4f x = _x;
_EIGEN_DECLARE_CONST_Packet4f(1 , 1.0f);
_EIGEN_DECLARE_CONST_Packet4f(half, 0.5f);
@ -367,7 +375,8 @@ static EIGEN_DONT_INLINE EIGEN_UNUSED Packet4f ei_pcos(Packet4f x)
// This is Quake3's fast inverse square root.
// For detail see here: http://www.beyond3d.com/content/articles/8/
static EIGEN_UNUSED Packet4f ei_psqrt(Packet4f _x)
template<> EIGEN_UNUSED
Packet4f ei_psqrt<Packet4f>(const Packet4f& _x)
{
Packet4f half = ei_pmul(_x, ei_pset1(.5f));