Developer Guide and Reference

Contents

_mm256_hadd_epi16/32

Horizontally adds adjacent signed packed 16/32-bit integer data elements of two vectors. The corresponding Intel® AVX2 instruction is
VPHADDW
or
VPHADDD
.

Syntax

extern __m256i _mm256_hadd_epi16(__m256i s1, __m256i s2);
extern __m256i _mm256_hadd_epi32(__m256i s1, __m256i s2);
Arguments
s1
integer source vector used for the operation
s2
integer source vector used for the operation
Description
Adds two adjacent 16- or 32-bit signed integers horizontally from source vectors,
s1
and
s2
and packs the 16 or 32-bit signed results to the destination vector.
Horizontal addition of two adjacent data elements of the low 16- or 32-bytes of the first and second source vectors are packed into the low 16- or 32-bytes of the destination vector. Horizontal addition of two adjacent data elements of the high 16- or 32-bytes of the first and second source vectors are packed into the high 16- or 32-bytes of the destination vector.
Returns
Result of the horizontal addition operation.