Multiplication Intrinsics

These Supplemental Streaming SIMD Extensions 3 (SSSE3) intrinsics are used for multiplication. The prototypes for these intrinsics are in tmmintrin.h.

To use these intrinsics, include the immintrin.h file as follows:

#include <immintrin.h>

_mm_maddubs_epi16

extern __m128i _mm_maddubs_epi16(__m128i a, __m128i b);

Multiplies signed and unsigned bytes, adds horizontal pair of signed words, and packs saturated signed words.

Interpreting a as array of unsigned 8-bit integers, b as arrays of signed 8-bit integers, and r as arrays of 16-bit signed integers:

for (i = 0; i < 8; i++) {
r[i] = signed_saturate_to_word(a[2*i+1] * b[2*i+1] + a[2*i]*b[2*i]);
}

_mm_maddubs_pi16

extern __m64 _mm_maddubs_pi16(__m64 a, __m64 b);

Multiplies signed and unsigned bytes, adds horizontal pair of signed words, and packs saturated signed words.

Interpreting a as array of unsigned 8-bit integers, b as arrays of signed 8-bit integers, and r as arrays of 16-bit signed integers:

for (i = 0; i < 4; i++) {
r[i] = signed_saturate_to_word(a[2*i+1] * b[2*i+1] + a[2*i]*b[2*i]);
}

_mm_mulhrs_epi16

extern __m128i _mm_mulhrs_epi16(__m128i a, __m128i b);

Multiplies signed words, scales and rounds signed doublewords, and packs high 16-bits.

Interpreting a, b, and r as arrays of signed 16-bit integers:

for (i = 0; i < 8; i++) {
r[i] =  (( (int32)((a[i] * b[i]) >> 14) + 1) >> 1) & 0xFFFF;
}

_mm_mulhrs_pi16

extern __m64 _mm_mulhrs_pi16(__m64 a, __m64 b);

Multiplies signed words, scales and rounds signed doublewords, and packs high 16-bits.

Interpreting a, b, and r as arrays of signed 16-bit integers:

for (i = 0; i < 4; i++) {
r[i] =  (( (int32)((a[i] * b[i]) >> 14) + 1) >> 1) & 0xFFFF;
}
Para obtener información más completa sobre las optimizaciones del compilador, consulte nuestro Aviso de optimización.