Developer Guide and Reference

Contents

_mm256_sad_epu8

Computes sum of absolute differences between extended packed unsigned values of two vectors. The corresponding Intel® AVX2 instruction is
VPSADBW
.

Syntax

extern __m256i _mm256_sad_epu8(__m256i s1, __m256i s2);
Arguments
s1
integer source vector used for the operation
s2
integer source vector used for the operation
Description
Computes the absolute value of the difference of packed groups of eight unsigned byte integers from the source vectors
s1
and
s2
. Four blocks of eight differences are stored at specific locations in the destination vector. Remaining bits in the destination vector are set to zero.
Returns
Result of the single sum-absolute-difference operation.