Intel® C++ Compiler Classic Developer Guide and Reference

ID 767249
Date 3/31/2023
Public

A newer version of this document is available. Customers should click here to go to the newest version.

Document Table of Contents

Intrinsics for FP Shuffle Operations

The prototypes for Intel® Advanced Vector Extensions 512 (Intel® AVX-512) intrinsics are located in the zmmintrin.h header file.

To use these intrinsics, include the immintrin.h file as follows:

#include <immintrin.h>


Intrinsic Name

Operation

Corresponding
Intel® AVX-512 Instruction

_mm512_shuffle_pd, _mm512_mask_shuffle_pd, _mm512_maskz_shuffle_pd

Shuffle float64 values.

VSHUFPD

_mm512_shuffle_ps, _mm512_mask_shuffle_ps, _mm512_maskz_shuffle_ps

Shuffle float32 values.

VSHUFPS

_mm512_shuffle_f64x2, _mm512_mask_shuffle_f64x2, _mm512_maskz_shuffle_f64x2

Shuffle float64 values and store using mask.

VSHUFF64X2

_mm512_shuffle_f32x4, _mm512_mask_shuffle_f32x4, _mm512_maskz_shuffle_f32x4

Shuffle float32 values and store using mask.

VSHUFF32X4


variable definition
k

writemask used as a selector

a

first source vector element

b

second source vector element

src

source element to use based on writemask result

imm

vector element selector


_mm512_shuffle_f32x4

extern __m512 __cdecl _mm512_shuffle_f32x4(__m512 a, __m512 b, const int imm);

Shuffles four float32 elements from a and b, selected by imm, and stores the result.


_mm512_mask_shuffle_f32x4

extern __m512 __cdecl _mm512_mask_shuffle_f32x4(__m512 src, __mmask16 k, __m512 a, __m512 b, const int imm);

Shuffles four float32 elements from a and b, selected by imm, and stores the result using writemask k (elements are copied from src when the corresponding mask bit is not set).


_mm512_maskz_shuffle_f32x4

extern __m512 __cdecl _mm512_maskz_shuffle_f32x4(__mmask16 k, __m512 a, __m512 b, const int imm);

Shuffles four float32 elements from a and b, selected by imm, and stores the result using zeromask k (elements are zeroed out when the corresponding mask bit is not set).



_mm512_shuffle_f64x2

extern __m512d __cdecl _mm512_shuffle_f64x2(__m512d a, __m512d b, const int imm);

Shuffles 128-bits (composed of two float64 elements from a and b, selected by imm, and stores the result.


_mm512_mask_shuffle_f64x2

extern __m512d __cdecl _mm512_mask_shuffle_f64x2(__m512d src, __mmask8 k, __m512d a, __m512d b, const int imm);

Shuffles 128-bits (composed of two float64 elements from a and b, selected by imm, and stores the result using writemask k (elements are copied from src when the corresponding mask bit is not set).


_mm512_maskz_shuffle_f64x2

extern __m512d __cdecl _mm512_maskz_shuffle_f64x2(__mmask8 k, __m512d a, __m512d b, const int imm);

Shuffles 128-bits (composed of two float64 elements from a and b, selected by imm, and stores the result using zeromask k (elements are zeroed out when the corresponding mask bit is not set).



_mm512_shuffle_pd

extern __m512d __cdecl _mm512_shuffle_pd(__m512d a, __m512d b, const int imm);

Shuffles float64 elements from vectors a and b within 128-bit lanes using the control in imm, and stores the result.


_mm512_mask_shuffle_pd

extern __m512d __cdecl _mm512_mask_shuffle_pd(__m512d src, __mmask8 k, __m512d a, __m512d b, const int imm);

Shuffle float64 elements from vectors a and b within 128-bit lanes using the control in imm, and stores the result using writemask k (elements are copied from src when the corresponding mask bit is not set).


_mm512_maskz_shuffle_pd

extern __m512d __cdecl _mm512_maskz_shuffle_pd(__mmask8 k, __m512d a, __m512d b, const int imm);

Shuffle float64 elements from vectors a and b within 128-bit lanes using the control in imm, and stores the result using zeromask k (elements are zeroed out when the corresponding mask bit is not set).



_mm512_shuffle_ps

extern __m512 __cdecl _mm512_shuffle_ps(__m512 a, __m512 b, const int imm);

Shuffles float32 elements from vectors a and b within 128-bit lanes using the control in imm, and stores the result.


_mm512_mask_shuffle_ps

extern __m512 __cdecl _mm512_mask_shuffle_ps(__m512 src, __mmask16 k, __m512 a, __m512 b, const int imm);

Shuffle float32 elements from vectors a and b within 128-bit lanes using the control in imm, and stores the result using writemask k.

Elements are copied from src when the corresponding mask bit is not set.


_mm512_maskz_shuffle_ps

extern __m512 __cdecl _mm512_maskz_shuffle_ps(__mmask16 k, __m512 a, __m512 b, const int imm);

Shuffle float32 elements from vectors a and b within 128-bit lanes using the control in imm, and stores the result using zeromask k.

Elements are zeroed out when the corresponding mask bit is not set.