Sdílet prostřednictvím


Fma Třída

Definice

Důležité

Toto rozhraní API neodpovídá specifikaci CLS.

Tato třída poskytuje přístup k hardwarovým instrukcím Intel FMA prostřednictvím vnitřních objektů.

public ref class Fma abstract : System::Runtime::Intrinsics::X86::Avx
[System.CLSCompliant(false)]
public abstract class Fma : System.Runtime.Intrinsics.X86.Avx
[<System.CLSCompliant(false)>]
type Fma = class
    inherit Avx
Public MustInherit Class Fma
Inherits Avx
Dědičnost
Dědičnost
Atributy

Vlastnosti

IsSupported

Tato třída poskytuje přístup k hardwarovým instrukcím Intel FMA prostřednictvím vnitřních objektů.

Metody

Equals(Object)

Určí, zda se zadaný objekt rovná aktuálnímu objektu.

(Zděděno od Object)
GetHashCode()

Slouží jako výchozí hashovací funkce.

(Zděděno od Object)
GetType()

Type Získá z aktuální instance.

(Zděděno od Object)
MemberwiseClone()

Vytvoří mělkou kopii aktuálního Objectsouboru .

(Zděděno od Object)
MultiplyAdd(Vector128<Double>, Vector128<Double>, Vector128<Double>)

__m128d _mm_fmadd_pd (__m128d a, __m128d b, __m128d c)

VFMADDPD xmm, xmm, xmm/m128

MultiplyAdd(Vector128<Single>, Vector128<Single>, Vector128<Single>)

__m128 _mm_fmadd_ps (__m128 a, __m128 b, __m128 c)

VFMADDPS xmm, xmm, xmm/m128

MultiplyAdd(Vector256<Double>, Vector256<Double>, Vector256<Double>)

__m256d _mm256_fmadd_pd (__m256d a, __m256d b, __m256d c)

VFMADDPS ymm, ymm, ymm/m256

MultiplyAdd(Vector256<Single>, Vector256<Single>, Vector256<Single>)

__m256 _mm256_fmadd_ps (__m256 a, __m256 b, __m256 c)

VFMADDPS ymm, ymm, ymm/m256

MultiplyAddNegated(Vector128<Double>, Vector128<Double>, Vector128<Double>)

__m128d _mm_fnmadd_pd (__m128d a, __m128d b, __m128d c)

VFNMADDPD xmm, xmm, xmm/m128

MultiplyAddNegated(Vector128<Single>, Vector128<Single>, Vector128<Single>)

__m128 _mm_fnmadd_ps (__m128 a, __m128 b, __m128 c)

VFNMADDPS xmm, xmm, xmm/m128

MultiplyAddNegated(Vector256<Double>, Vector256<Double>, Vector256<Double>)

__m256d _mm256_fnmadd_pd (__m256d a, __m256d b, __m256d c)

VFNMADDPD ymm, ymm, ymm/m256

MultiplyAddNegated(Vector256<Single>, Vector256<Single>, Vector256<Single>)

__m256 _mm256_fnmadd_ps (__m256 a, __m256 b, __m256 c)

VFNMADDPS ymm, ymm, ymm/m256

MultiplyAddNegatedScalar(Vector128<Double>, Vector128<Double>, Vector128<Double>)

__m128d _mm_fnmadd_sd (__m128d a, __m128d b, __m128d c)

VFNMADDSD xmm, xmm, xmm/m64

MultiplyAddNegatedScalar(Vector128<Single>, Vector128<Single>, Vector128<Single>)

__m128 _mm_fnmadd_ss (__m128 a, __m128 b, __m128 c)

VFNMADDSS xmm, xmm, xmm/m32

MultiplyAddScalar(Vector128<Double>, Vector128<Double>, Vector128<Double>)

__m128d _mm_fmadd_sd (__m128d a, __m128d b, __m128d c)

VFMADDSS xmm, xmm, xmm/m64

MultiplyAddScalar(Vector128<Single>, Vector128<Single>, Vector128<Single>)

__m128 _mm_fmadd_ss (__m128 a, __m128 b, __m128 c)

VFMADDSS xmm, xmm, xmm/m32

MultiplyAddSubtract(Vector128<Double>, Vector128<Double>, Vector128<Double>)

__m128d _mm_fmaddsub_pd (__m128d a, __m128d b, __m128d c)

VFMADDSUBPD xmm, xmm, xmm/m128

MultiplyAddSubtract(Vector128<Single>, Vector128<Single>, Vector128<Single>)

__m128 _mm_fmaddsub_ps (__m128 a, __m128 b, __m128 c)

VFMADDSUBPS xmm, xmm, xmm/m128

MultiplyAddSubtract(Vector256<Double>, Vector256<Double>, Vector256<Double>)

__m256d _mm256_fmaddsub_pd (__m256d a, __m256d b, __m256d c)

VFMADDSUBPD ymm, ymm, ymm/m256

MultiplyAddSubtract(Vector256<Single>, Vector256<Single>, Vector256<Single>)

__m256 _mm256_fmaddsub_ps (__m256 a, __m256 b, __m256 c)

VFMADDSUBPS ymm, ymm, ymm/m256

MultiplySubtract(Vector128<Double>, Vector128<Double>, Vector128<Double>)

__m128d _mm_fmsub_pd (__m128d a, __m128d b, __m128d c)

VFMSUBPS xmm, xmm, xmm/m128

MultiplySubtract(Vector128<Single>, Vector128<Single>, Vector128<Single>)

__m128 _mm_fmsub_ps (__m128 a, __m128 b, __m128 c)

VFMSUBPS xmm, xmm, xmm/m128

MultiplySubtract(Vector256<Double>, Vector256<Double>, Vector256<Double>)

__m256d _mm256_fmsub_pd (__m256d a, __m256d b, __m256d c)

VFMSUBPD ymm, ymm, ymm/m256

MultiplySubtract(Vector256<Single>, Vector256<Single>, Vector256<Single>)

__m256 _mm256_fmsub_ps (__m256 a, __m256 b, __m256 c)

VFMSUBPS ymm, ymm, ymm/m256

MultiplySubtractAdd(Vector128<Double>, Vector128<Double>, Vector128<Double>)

__m128d _mm_fmsubadd_pd (__m128d a, __m128d b, __m128d c)

VFMSUBADDPD xmm, xmm, xmm/m128

MultiplySubtractAdd(Vector128<Single>, Vector128<Single>, Vector128<Single>)

__m128 _mm_fmsubadd_ps (__m128 a, __m128 b, __m128 c)

VFMSUBADDPS xmm, xmm, xmm/m128

MultiplySubtractAdd(Vector256<Double>, Vector256<Double>, Vector256<Double>)

__m256d _mm256_fmsubadd_pd (__m256d a, __m256d b, __m256d c)

VFMSUBADDPD ymm, ymm, ymm/m256

MultiplySubtractAdd(Vector256<Single>, Vector256<Single>, Vector256<Single>)

__m256 _mm256_fmsubadd_ps (__m256 a, __m256 b, __m256 c)

VFMSUBADDPS ymm, ymm, ymm/m256

MultiplySubtractNegated(Vector128<Double>, Vector128<Double>, Vector128<Double>)

__m128d _mm_fnmsub_pd (__m128d a, __m128d b, __m128d c)

VFNMSUBPD xmm, xmm, xmm/m128

MultiplySubtractNegated(Vector128<Single>, Vector128<Single>, Vector128<Single>)

__m128 _mm_fnmsub_ps (__m128 a, __m128 b, __m128 c)

VFNMSUBPS xmm, xmm, xmm/m128

MultiplySubtractNegated(Vector256<Double>, Vector256<Double>, Vector256<Double>)

__m256d _mm256_fnmsub_pd (__m256d a, __m256d b, __m256d c)

VFNMSUBPD ymm, ymm, ymm/m256

MultiplySubtractNegated(Vector256<Single>, Vector256<Single>, Vector256<Single>)

__m256 _mm256_fnmsub_ps (__m256 a, __m256 b, __m256 c)

VFNMSUBPS ymm, ymm, ymm/m256

MultiplySubtractNegatedScalar(Vector128<Double>, Vector128<Double>, Vector128<Double>)

__m128d _mm_fnmsub_sd (__m128d a, __m128d b, __m128d c)

VFNMSUBSD xmm, xmm, xmm/m64

MultiplySubtractNegatedScalar(Vector128<Single>, Vector128<Single>, Vector128<Single>)

__m128 _mm_fnmsub_ss (__m128 a, __m128 b, __m128 c)

VFNMSUBSS xmm, xmm, xmm/m32

MultiplySubtractScalar(Vector128<Double>, Vector128<Double>, Vector128<Double>)

__m128d _mm_fmsub_sd (__m128d a, __m128d b, __m128d c)

VFMSUBSD xmm, xmm, xmm/m64

MultiplySubtractScalar(Vector128<Single>, Vector128<Single>, Vector128<Single>)

__m128 _mm_fmsub_ss (__m128 a, __m128 b, __m128 c)

VFMSUBSS xmm, xmm, xmm/m32

ToString()

Vrátí řetězec, který představuje aktuální objekt.

(Zděděno od Object)

Platí pro