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A newer version of this document is available.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 Test 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_conflict_epi32, _mm512_mask_conflict_epi32, _mm512_maskz_conflict_epi32

Test int32 elements for equality.

VPCONFLICTD

_mm512_conflict_epi64, _mm512_mask_conflict_epi64, _mm512_maskz_conflict_epi64

Test int64 elements for equality.

VPCONFLICTQ

_mm512_test_epi32_mask, _mm512_mask_test_epi32_mask

Performs a bitwise logical AND operation and stores the logical comparison result.

VPTESTMD

_mm512_testn_epi32_mask, _mm512_mask_testn_epi32_mask

Performs a bitwise logical AND NOT operation and stores the logical comparison result.

VPTESTNMD

_mm512_test_epi64_mask, _mm512_mask_test_epi64_mask

Performs a bitwise logical AND operation and stores the logical comparison result.

VPTESTMQ

_mm512_testn_epi64_mask, _mm512_mask_testn_epi64_mask

Performs a bitwise logical AND NOT operation and stores the logical comparison result.

VPTESTNMQ


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


_mm512_conflict_epi32

extern __m512i __cdecl _mm512_conflict_epi32(__m512i a);

Tests each 32-bit element of a for equality with all other elements in a closer to the least significant bit.

Each element's comparison forms a zero extended bit vector in the destination.



_mm512_mask_conflict_epi32

extern __m512i __cdecl _mm512_mask_conflict_epi32(__m512i src, __mmask16 k, __m512i a);

Tests each 32-bit element of a for equality with all other elements in a closer to the least significant bit using writemask k (elements are copied from src when the corresponding mask bit is not set).

Each element's comparison forms a zero extended bit vector in the destination.



_mm512_maskz_conflict_epi32

extern __m512i __cdecl _mm512_maskz_conflict_epi32(__mmask16 k, __m512i a);

Tests each 32-bit element of a for equality with other elements in a closer to the least significant bit using zeromask k (elements are zeroed out when the corresponding mask bit is not set).

Each element's comparison forms a zero extended bit vector in the destination.



_mm512_conflict_epi64

extern __m512i __cdecl _mm512_conflict_epi64(__m512i a);

Tests each 64-bit element of a for equality with other elements in a closer to the least significant bit.

Each element's comparison forms a zero extended bit vector in the destination.



_mm512_mask_conflict_epi64

extern __m512i __cdecl _mm512_mask_conflict_epi64(__m512i src, __mmask8 k, __m512i a);

Tests each 64-bit element of a for equality with other elements in a closer to the least significant bit using writemask k (elements are copied from src when the corresponding mask bit is not set).

Each element's comparison forms a zero extended bit vector in the destination.



_mm512_maskz_conflict_epi64

extern __m512i __cdecl _mm512_maskz_conflict_epi64(__mmask8 k, __m512i a);

Tests each 64-bit element of a for equality with other elements in a closer to the least significant bit using zeromask k (elements are zeroed out when the corresponding mask bit is not set).

Each element's comparison forms a zero extended bit vector the destination.



_mm512_test_epi32_mask

extern __mmask16 __cdecl _mm512_test_epi32_mask(__m512i a, __m512i b);

Computes the bitwise AND of packed 32-bit integers in a and b, producing intermediate 32-bit values, and sets the corresponding bit in result mask k if the intermediate value is non-zero.



_mm512_mask_test_epi32_mask

extern __mmask16 __cdecl _mm512_mask_test_epi32_mask(__mmask16 k, __m512i a, __m512i b);

Computes the bitwise AND of packed 32-bit integers in a and b, producing intermediate 32-bit values, and sets the corresponding bit in result mask k (subject to writemask k) if the intermediate value is non-zero.



_mm512_test_epi64_mask

extern __mmask8 __cdecl _mm512_test_epi64_mask(__m512i a, __m512i b);

Computes the bitwise AND of packed 64-bit integers in a and b, producing intermediate 64-bit values, and sets the corresponding bit in result mask k if the intermediate value is non-zero.



_mm512_mask_test_epi64_mask

extern __mmask8 __cdecl _mm512_mask_test_epi64_mask(__mmask8 k, __m512i a, __m512i b);

Computes the bitwise AND of packed 64-bit integers in a and b, producing intermediate 64-bit values, and sets the corresponding bit in result mask k (subject to writemask k) if the intermediate value is non-zero.



_mm512_testn_epi32_mask

extern __mmask16 __cdecl _mm512_testn_epi32_mask(__m512i a, __m512i b);

Computes the bitwise AND NOT of packed 32-bit integers in a and b, producing intermediate 32-bit values, and sets the corresponding bit in result mask k if the intermediate value is zero.



_mm512_mask_testn_epi32_mask

extern __mmask16 __cdecl _mm512_mask_testn_epi32_mask(__mmask16 k, __m512i a, __m512i b);

Computes the bitwise AND NOT of packed 32-bit integers in a and b, producing intermediate 32-bit values, and sets the corresponding bit in result mask k (subject to writemask k) if the intermediate value is zero.



_mm512_testn_epi64_mask

extern __mmask8 __cdecl _mm512_testn_epi64_mask(__m512i a, __m512i b);

Computes the bitwise AND NOT of packed 64-bit integers in a and b, producing intermediate 64-bit values, and sets the corresponding bit in result mask k if the intermediate value is zero.



_mm512_mask_testn_epi64_mask

extern __mmask8 __cdecl _mm512_mask_testn_epi64_mask(__mmask8 k, __m512i a, __m512i b);

Computes the bitwise AND NOT of packed 64-bit integers in a and b, producing intermediate 64-bit values, and sets the corresponding bit in result mask k (subject to writemask k) if the intermediate value is zero.