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A newer version of this document is available.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

_mm256_permute4x64_pd

Permutes quadword double-precision floating-point values of the source vector into the destination vector. The corresponding Intel® AVX2 instruction is VPERMPD.

Syntax

extern __m256i _mm256_permute4x64_epi64(__m256d val, const int control);

Arguments

val

the vector of 64-bit qword double-precision floating-point elements to be permuted

control

an integer specified as an 8-bit immediate

Description

Use two-bit index values in the immediate byte to select a qword double-precision floating-point element from the source vector val. The result element is copied to the corresponding element of destination vector. The intrinsic allows to copy the same element of the source vector to more than one element of the destination vector.

Below is the pseudo-code for the intrinsic:

RESULT[63:0] <- (VAL[255:0] >> (CONTROL[1:0] * 64))[63:0];
RESULT[127:64] <- (VAL[255:0] >> (CONTROL[3:2] * 64))[63:0];
RESULT[191:128] <- (VAL[255:0] >> (CONTROL[5:4] * 64))[63:0];
RESULT[255:192] <- (VAL[255:0] >> (CONTROL[7:6] * 64))[63:0];

Returns

Result of the permute operation.