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A newer version of this document is available.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

_mm_cmp_pd, _mm256_cmp_pd

Compares packed 128-bit and 256-bit float64 vector elements. The corresponding Intel® AVX instruction is VCMPPD.

Syntax

extern __m128d _mm_cmp_pd(__m128d m1, __m128d m2, const int predicate);

extern __m256d _mm256_cmp_pd(__m256d m1, __m256d m2, const int predicate);

Arguments

m1

float64 vector used for the operation

m2

float64 vector also used for the operation

predicate

an immediate operand that specifies the type of comparison to be performed of the packed values; see immintrin.h file for the values to specify the type of comparison

Description

Performs a SIMD compare of the four packed double-precision floating-point (float64) values in the first source operand, m1, and the second source operand, m2, and returns the results of the comparison.

The _mm_cmp_pd intrinsic is used for comparing 128-bit float64 values while the _m256_cmp_pd intrinsic is used for comparing 256-bit float64 values.

The comparison predicate parameter (immediate) specifies the type of comparison performed on each of the pairs of packed values.

Returns

Result of the compare operation.