Developer Guide and Reference

Contents

_mm256_packs_epi16/32

Pack signed word/doubleword integers to signed byte/words integers and saturates. The corresponding Intel® AVX2 instruction is
VPACKSSWB
or
VPACKSSDW
.

Syntax

extern __m256i _mm256_packs_epi16(__m256i a, __m256i b);
extern __m256i _mm256_packs_epi32(__m256i a, __m256i b);
Arguments
a
integer source vector used for the operation
b
integer source vector used for the operation
Description
The
_mm256_packs_epi16
intrinsic converts 16 packed signed word integers from the first and the second source operands into 32 packed signed byte integers. The
_mm256_packs_epi32
intrinsic converts eight packed signed doubleword integers from the first and the second source operands into 16 packed signed word integers.
Returns
Result of the pack operation.