ipsdk::simd::detail::MaxReg< eInstructionSet::eIS_Avx, ipInt32 > Struct Template Reference

MaxReg template specialization for instruction set AVX and type ipInt32. More...

#include <MaxRegDecl.h>

Static Public Member Functions
static IPSDK_FORCEINLINE AvxType< ipInt32 >::Type	act (const AvxType< ipInt32 >::Type &in1, const AvxType< ipInt32 >::Type &in2)
	MaxReg template specialization for instruction set AVX and type ipInt32. More...
static IPSDK_FORCEINLINE void	act (const AvxType< ipInt32 >::Type &in1, const AvxType< ipInt32 >::Type &in2, AvxType< ipInt32 >::Type &out)

Detailed Description

template<>
struct ipsdk::simd::detail::MaxReg< eInstructionSet::eIS_Avx, ipInt32 >

MaxReg template specialization for instruction set AVX and type ipInt32.

Note

no native AVX instruction exist for maximum computation on integers, so we emulated it using SSE2 instructions (if we had not done that, we would not have been able to implement functions like logarithm on ipReal32)

Member Function Documentation

act()

AvxType< ipInt32 >::Type ipsdk::simd::detail::MaxReg< eInstructionSet::eIS_Avx, ipInt32 >::act ( const AvxType< ipInt32 >::Type &  in1, const AvxType< ipInt32 >::Type &  in2  )

static

MaxReg template specialization for instruction set AVX and type ipInt32.

Note

no native AVX instruction exist for maximum computation on integers, so we emulated it using SSE2 instructions (if we had not done that, we would not have been able to implement functions like logarithm on ipReal32)

---

The documentation for this struct was generated from the following files:

• /media/reactivipdev/Nightly/IPSDK_Patch/IPSDK_4_1/IPSDK_Nightly/include/partner/IPSDKUtil/InstructionSet/Arithmetic/detail/AVX/MaxRegDecl.h
• /media/reactivipdev/Nightly/IPSDK_Patch/IPSDK_4_1/IPSDK_Nightly/include/partner/IPSDKUtil/InstructionSet/Arithmetic/detail/AVX/MaxRegImpl.h