Horizontally adds adjacent signed packed 16-bit integer data elements of two vectors with saturation. The corresponding Intel® AVX2 instruction is VPHADDSW.
extern __m256i _mm256_hadds_epi16(__m256i s1, __m256i s2); |
s1 |
integer source vector used for the operation |
s2 |
integer source vector used for the operation |
Adds two adjacent signed 16-bit integers horizontally from the second and first source vectors, s2 and s1, saturates the signed results, and packs the signed, saturated 16-bit results to the destination vector.
Result of the horizontal addition operation with saturation.
Copyright © 1996-2011, Intel Corporation. All rights reserved.