C-7
COMPILER INTRINSICS AND FUNCTIONAL EQUIVALENTS
PMULHUW__m64 _m_pmulhuw(__m64 a, __m64 b)
__m64 _mm_mulhi_pu16(__m64 a, __m64 b)
Multiplies the unsigned words in a and b,
returning the upper 16 bits of the 32-bit
intermediate results.
PMULHW__m64 _m_pmulhw(__m64 m1, __m64 m2)
__m64 _mm_mulhi_pi16(__m64 m1, __m64 m2)
Multiply four signed 16-bit values in m1 by
four signed 16-bit values in m2 and
produce the high 16 bits of the four results.
PMULLW__m64 _m_pmullw(__m64 m1, __m64 m2)
__m64 _mm_mullo_pi16(__m64 m1, __m64 m2)
Multiply four 16-bit values in m1 by four 16-
bit values in m2 and produce the low 16
bits of the four results.
POR
__m64 _m_por(__m64 m1, __m64 m2)
__m64 _mm_or_si64(__m64 m1, __m64 m2)
Perform a bitwise OR of the 64-bit value in
m1 with the 64-bit value in m2.
PREFETCHvoid _mm_prefetch(char *a, int sel)
Loads one cache line of data from address
p to a location "closer" to the processor.
The value i specifies the type of prefetch
operation.
PSHUFW__m64 _m_psadbw(__m64 a, __m64 b)
__m64 _mm_sad_pu8(__m64 a, __m64 b)
Returns a combination of the four words of
a. The selector n must be an immediate.
PSLLW
__m64 _m_pshufw(__m64 a, int n)
__m64 _mm_shuffle_pi16(__m64 a, int n)
Shift four 16-bit values in m left the amount
specified by count while shifting in zeroes.
__m64 _m_psllw(__m64 m, __m64 count)
__m64 _mm_sll_pi16(__m64 m, __m64 count)
Shift four 16-bit values in m left the amount
specified by count while shifting in zeroes.
For the best performance, count should be
a constant.
PSLLD
__m64 _m_psllwi (__m64 m, int count)
__m64 _m_slli_pi16(__m64 m, int count)
Shift two 32-bit values in m left the amount
specified by count while shifting in zeroes.
__m64 _m_pslld (__m64 m, __m64 count)
__m64 _m_sll_pi32(__m64 m, __m64 count)
Shift two 32-bit values in m left the amount
specified by count while shifting in zeroes.
For the best performance, count should be
a constant.
PSLLQ
__m64 _m_psllq (__m64 m, __m64 count)
__m64 _mm_sll_si64(__m64 m, __m64 count)
Shift the 64-bit value in m left the amount
specified by count while shifting in zeroes.
__m64 _m_psllqi (__m64 m, int count)
__m64 _mm_slli_si64(__m64 m, int count)
Shift the 64-bit value in m left the amount
specified by count while shifting in zeroes.
For the best performance, count should be
a constant.
PSRAW
__m64 _m_psraw (__m64 m, __m64 count)
__m64 _mm_sra_pi16(__m64 m, __m64 count)
Shift four 16-bit values in m right the
amount specified by count while shifting in
the sign bit.
__m64 _m_psrawi (__m64 m, int count)
__m64 _mm_srai_pi16(__m64 m, int count)
Shift four 16-bit values in m right the
amount specified by count while shifting in
the sign bit. For the best performance,
count should be a constant.
PSRAD
__m64 _m_psrad (__m64 m, __m64 count)
__m64 _mm_sra_pi32 (__m64 m, __m64 count)
Shift two 32-bit values in m right the
amount specified by count while shifting in
the sign bit.
__m64 _m_psradi (__m64 m, int count)
__m64 _mm_srai_pi32 (__m64 m, int count)
Shift two 32-bit values in m right the
amount specified by count while shifting in
the sign bit. For the best performance,
count should be a constant.
Table C-1. Simple Intrinsics
Mnemonic
Intrinsic
Description
