diff options
Diffstat (limited to 'contrib/gcc/config/i386/mmintrin.h')
-rw-r--r-- | contrib/gcc/config/i386/mmintrin.h | 336 |
1 files changed, 333 insertions, 3 deletions
diff --git a/contrib/gcc/config/i386/mmintrin.h b/contrib/gcc/config/i386/mmintrin.h index 7b4aa01..00e77e4 100644 --- a/contrib/gcc/config/i386/mmintrin.h +++ b/contrib/gcc/config/i386/mmintrin.h @@ -1,4 +1,4 @@ -/* Copyright (C) 2002 Free Software Foundation, Inc. +/* Copyright (C) 2002, 2003 Free Software Foundation, Inc. This file is part of GNU CC. @@ -25,7 +25,7 @@ Public License. */ /* Implemented from the specification included in the Intel C++ Compiler - User Guide and Reference, version 5.0. */ + User Guide and Reference, version 8.0. */ #ifndef _MMINTRIN_H_INCLUDED #define _MMINTRIN_H_INCLUDED @@ -48,6 +48,12 @@ _mm_empty (void) __builtin_ia32_emms (); } +static __inline void +_m_empty (void) +{ + _mm_empty (); +} + /* Convert I to a __m64 object. The integer is zero-extended to 64-bits. */ static __inline __m64 _mm_cvtsi32_si64 (int __i) @@ -56,6 +62,12 @@ _mm_cvtsi32_si64 (int __i) return (__m64) __tmp; } +static __inline __m64 +_m_from_int (int __i) +{ + return _mm_cvtsi32_si64 (__i); +} + #ifdef __x86_64__ /* Convert I to a __m64 object. */ static __inline __m64 @@ -80,6 +92,12 @@ _mm_cvtsi64_si32 (__m64 __i) return __tmp; } +static __inline int +_m_to_int (__m64 __i) +{ + return _mm_cvtsi64_si32 (__i); +} + #ifdef __x86_64__ /* Convert the lower 32 bits of the __m64 object into an integer. */ static __inline long long @@ -98,6 +116,12 @@ _mm_packs_pi16 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_packsswb ((__v4hi)__m1, (__v4hi)__m2); } +static __inline __m64 +_m_packsswb (__m64 __m1, __m64 __m2) +{ + return _mm_packs_pi16 (__m1, __m2); +} + /* Pack the two 32-bit values from M1 in to the lower two 16-bit values of the result, and the two 32-bit values from M2 into the upper two 16-bit values of the result, all with signed saturation. */ @@ -107,6 +131,12 @@ _mm_packs_pi32 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_packssdw ((__v2si)__m1, (__v2si)__m2); } +static __inline __m64 +_m_packssdw (__m64 __m1, __m64 __m2) +{ + return _mm_packs_pi32 (__m1, __m2); +} + /* Pack the four 16-bit values from M1 into the lower four 8-bit values of the result, and the four 16-bit values from M2 into the upper four 8-bit values of the result, all with unsigned saturation. */ @@ -116,6 +146,12 @@ _mm_packs_pu16 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_packuswb ((__v4hi)__m1, (__v4hi)__m2); } +static __inline __m64 +_m_packuswb (__m64 __m1, __m64 __m2) +{ + return _mm_packs_pu16 (__m1, __m2); +} + /* Interleave the four 8-bit values from the high half of M1 with the four 8-bit values from the high half of M2. */ static __inline __m64 @@ -124,6 +160,12 @@ _mm_unpackhi_pi8 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_punpckhbw ((__v8qi)__m1, (__v8qi)__m2); } +static __inline __m64 +_m_punpckhbw (__m64 __m1, __m64 __m2) +{ + return _mm_unpackhi_pi8 (__m1, __m2); +} + /* Interleave the two 16-bit values from the high half of M1 with the two 16-bit values from the high half of M2. */ static __inline __m64 @@ -132,6 +174,12 @@ _mm_unpackhi_pi16 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_punpckhwd ((__v4hi)__m1, (__v4hi)__m2); } +static __inline __m64 +_m_punpckhwd (__m64 __m1, __m64 __m2) +{ + return _mm_unpackhi_pi16 (__m1, __m2); +} + /* Interleave the 32-bit value from the high half of M1 with the 32-bit value from the high half of M2. */ static __inline __m64 @@ -140,6 +188,12 @@ _mm_unpackhi_pi32 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_punpckhdq ((__v2si)__m1, (__v2si)__m2); } +static __inline __m64 +_m_punpckhdq (__m64 __m1, __m64 __m2) +{ + return _mm_unpackhi_pi32 (__m1, __m2); +} + /* Interleave the four 8-bit values from the low half of M1 with the four 8-bit values from the low half of M2. */ static __inline __m64 @@ -148,6 +202,12 @@ _mm_unpacklo_pi8 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_punpcklbw ((__v8qi)__m1, (__v8qi)__m2); } +static __inline __m64 +_m_punpcklbw (__m64 __m1, __m64 __m2) +{ + return _mm_unpacklo_pi8 (__m1, __m2); +} + /* Interleave the two 16-bit values from the low half of M1 with the two 16-bit values from the low half of M2. */ static __inline __m64 @@ -156,6 +216,12 @@ _mm_unpacklo_pi16 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_punpcklwd ((__v4hi)__m1, (__v4hi)__m2); } +static __inline __m64 +_m_punpcklwd (__m64 __m1, __m64 __m2) +{ + return _mm_unpacklo_pi16 (__m1, __m2); +} + /* Interleave the 32-bit value from the low half of M1 with the 32-bit value from the low half of M2. */ static __inline __m64 @@ -164,6 +230,12 @@ _mm_unpacklo_pi32 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_punpckldq ((__v2si)__m1, (__v2si)__m2); } +static __inline __m64 +_m_punpckldq (__m64 __m1, __m64 __m2) +{ + return _mm_unpacklo_pi32 (__m1, __m2); +} + /* Add the 8-bit values in M1 to the 8-bit values in M2. */ static __inline __m64 _mm_add_pi8 (__m64 __m1, __m64 __m2) @@ -171,6 +243,12 @@ _mm_add_pi8 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_paddb ((__v8qi)__m1, (__v8qi)__m2); } +static __inline __m64 +_m_paddb (__m64 __m1, __m64 __m2) +{ + return _mm_add_pi8 (__m1, __m2); +} + /* Add the 16-bit values in M1 to the 16-bit values in M2. */ static __inline __m64 _mm_add_pi16 (__m64 __m1, __m64 __m2) @@ -178,6 +256,12 @@ _mm_add_pi16 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_paddw ((__v4hi)__m1, (__v4hi)__m2); } +static __inline __m64 +_m_paddw (__m64 __m1, __m64 __m2) +{ + return _mm_add_pi16 (__m1, __m2); +} + /* Add the 32-bit values in M1 to the 32-bit values in M2. */ static __inline __m64 _mm_add_pi32 (__m64 __m1, __m64 __m2) @@ -185,6 +269,12 @@ _mm_add_pi32 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_paddd ((__v2si)__m1, (__v2si)__m2); } +static __inline __m64 +_m_paddd (__m64 __m1, __m64 __m2) +{ + return _mm_add_pi32 (__m1, __m2); +} + /* Add the 64-bit values in M1 to the 64-bit values in M2. */ static __inline __m64 _mm_add_si64 (__m64 __m1, __m64 __m2) @@ -200,6 +290,12 @@ _mm_adds_pi8 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_paddsb ((__v8qi)__m1, (__v8qi)__m2); } +static __inline __m64 +_m_paddsb (__m64 __m1, __m64 __m2) +{ + return _mm_adds_pi8 (__m1, __m2); +} + /* Add the 16-bit values in M1 to the 16-bit values in M2 using signed saturated arithmetic. */ static __inline __m64 @@ -208,6 +304,12 @@ _mm_adds_pi16 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_paddsw ((__v4hi)__m1, (__v4hi)__m2); } +static __inline __m64 +_m_paddsw (__m64 __m1, __m64 __m2) +{ + return _mm_adds_pi16 (__m1, __m2); +} + /* Add the 8-bit values in M1 to the 8-bit values in M2 using unsigned saturated arithmetic. */ static __inline __m64 @@ -216,6 +318,12 @@ _mm_adds_pu8 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_paddusb ((__v8qi)__m1, (__v8qi)__m2); } +static __inline __m64 +_m_paddusb (__m64 __m1, __m64 __m2) +{ + return _mm_adds_pu8 (__m1, __m2); +} + /* Add the 16-bit values in M1 to the 16-bit values in M2 using unsigned saturated arithmetic. */ static __inline __m64 @@ -224,6 +332,12 @@ _mm_adds_pu16 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_paddusw ((__v4hi)__m1, (__v4hi)__m2); } +static __inline __m64 +_m_paddusw (__m64 __m1, __m64 __m2) +{ + return _mm_adds_pu16 (__m1, __m2); +} + /* Subtract the 8-bit values in M2 from the 8-bit values in M1. */ static __inline __m64 _mm_sub_pi8 (__m64 __m1, __m64 __m2) @@ -231,6 +345,12 @@ _mm_sub_pi8 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_psubb ((__v8qi)__m1, (__v8qi)__m2); } +static __inline __m64 +_m_psubb (__m64 __m1, __m64 __m2) +{ + return _mm_sub_pi8 (__m1, __m2); +} + /* Subtract the 16-bit values in M2 from the 16-bit values in M1. */ static __inline __m64 _mm_sub_pi16 (__m64 __m1, __m64 __m2) @@ -238,6 +358,12 @@ _mm_sub_pi16 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_psubw ((__v4hi)__m1, (__v4hi)__m2); } +static __inline __m64 +_m_psubw (__m64 __m1, __m64 __m2) +{ + return _mm_sub_pi16 (__m1, __m2); +} + /* Subtract the 32-bit values in M2 from the 32-bit values in M1. */ static __inline __m64 _mm_sub_pi32 (__m64 __m1, __m64 __m2) @@ -245,6 +371,12 @@ _mm_sub_pi32 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_psubd ((__v2si)__m1, (__v2si)__m2); } +static __inline __m64 +_m_psubd (__m64 __m1, __m64 __m2) +{ + return _mm_sub_pi32 (__m1, __m2); +} + /* Add the 64-bit values in M1 to the 64-bit values in M2. */ static __inline __m64 _mm_sub_si64 (__m64 __m1, __m64 __m2) @@ -260,6 +392,12 @@ _mm_subs_pi8 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_psubsb ((__v8qi)__m1, (__v8qi)__m2); } +static __inline __m64 +_m_psubsb (__m64 __m1, __m64 __m2) +{ + return _mm_subs_pi8 (__m1, __m2); +} + /* Subtract the 16-bit values in M2 from the 16-bit values in M1 using signed saturating arithmetic. */ static __inline __m64 @@ -268,6 +406,12 @@ _mm_subs_pi16 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_psubsw ((__v4hi)__m1, (__v4hi)__m2); } +static __inline __m64 +_m_psubsw (__m64 __m1, __m64 __m2) +{ + return _mm_subs_pi16 (__m1, __m2); +} + /* Subtract the 8-bit values in M2 from the 8-bit values in M1 using unsigned saturating arithmetic. */ static __inline __m64 @@ -276,6 +420,12 @@ _mm_subs_pu8 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_psubusb ((__v8qi)__m1, (__v8qi)__m2); } +static __inline __m64 +_m_psubusb (__m64 __m1, __m64 __m2) +{ + return _mm_subs_pu8 (__m1, __m2); +} + /* Subtract the 16-bit values in M2 from the 16-bit values in M1 using unsigned saturating arithmetic. */ static __inline __m64 @@ -284,6 +434,12 @@ _mm_subs_pu16 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_psubusw ((__v4hi)__m1, (__v4hi)__m2); } +static __inline __m64 +_m_psubusw (__m64 __m1, __m64 __m2) +{ + return _mm_subs_pu16 (__m1, __m2); +} + /* Multiply four 16-bit values in M1 by four 16-bit values in M2 producing four 32-bit intermediate results, which are then summed by pairs to produce two 32-bit results. */ @@ -293,6 +449,12 @@ _mm_madd_pi16 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_pmaddwd ((__v4hi)__m1, (__v4hi)__m2); } +static __inline __m64 +_m_pmaddwd (__m64 __m1, __m64 __m2) +{ + return _mm_madd_pi16 (__m1, __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 32-bit results. */ static __inline __m64 @@ -301,6 +463,12 @@ _mm_mulhi_pi16 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_pmulhw ((__v4hi)__m1, (__v4hi)__m2); } +static __inline __m64 +_m_pmulhw (__m64 __m1, __m64 __m2) +{ + return _mm_mulhi_pi16 (__m1, __m2); +} + /* Multiply four 16-bit values in M1 by four 16-bit values in M2 and produce the low 16 bits of the results. */ static __inline __m64 @@ -309,6 +477,12 @@ _mm_mullo_pi16 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_pmullw ((__v4hi)__m1, (__v4hi)__m2); } +static __inline __m64 +_m_pmullw (__m64 __m1, __m64 __m2) +{ + return _mm_mullo_pi16 (__m1, __m2); +} + /* Shift four 16-bit values in M left by COUNT. */ static __inline __m64 _mm_sll_pi16 (__m64 __m, __m64 __count) @@ -317,11 +491,23 @@ _mm_sll_pi16 (__m64 __m, __m64 __count) } static __inline __m64 +_m_psllw (__m64 __m, __m64 __count) +{ + return _mm_sll_pi16 (__m, __count); +} + +static __inline __m64 _mm_slli_pi16 (__m64 __m, int __count) { return (__m64) __builtin_ia32_psllw ((__v4hi)__m, __count); } +static __inline __m64 +_m_psllwi (__m64 __m, int __count) +{ + return _mm_slli_pi16 (__m, __count); +} + /* Shift two 32-bit values in M left by COUNT. */ static __inline __m64 _mm_sll_pi32 (__m64 __m, __m64 __count) @@ -330,11 +516,23 @@ _mm_sll_pi32 (__m64 __m, __m64 __count) } static __inline __m64 +_m_pslld (__m64 __m, __m64 __count) +{ + return _mm_sll_pi32 (__m, __count); +} + +static __inline __m64 _mm_slli_pi32 (__m64 __m, int __count) { return (__m64) __builtin_ia32_pslld ((__v2si)__m, __count); } +static __inline __m64 +_m_pslldi (__m64 __m, int __count) +{ + return _mm_slli_pi32 (__m, __count); +} + /* Shift the 64-bit value in M left by COUNT. */ static __inline __m64 _mm_sll_si64 (__m64 __m, __m64 __count) @@ -343,11 +541,23 @@ _mm_sll_si64 (__m64 __m, __m64 __count) } static __inline __m64 +_m_psllq (__m64 __m, __m64 __count) +{ + return _mm_sll_si64 (__m, __count); +} + +static __inline __m64 _mm_slli_si64 (__m64 __m, int __count) { return (__m64) __builtin_ia32_psllq ((long long)__m, (long long)__count); } +static __inline __m64 +_m_psllqi (__m64 __m, int __count) +{ + return _mm_slli_si64 (__m, __count); +} + /* Shift four 16-bit values in M right by COUNT; shift in the sign bit. */ static __inline __m64 _mm_sra_pi16 (__m64 __m, __m64 __count) @@ -356,11 +566,23 @@ _mm_sra_pi16 (__m64 __m, __m64 __count) } static __inline __m64 +_m_psraw (__m64 __m, __m64 __count) +{ + return _mm_sra_pi16 (__m, __count); +} + +static __inline __m64 _mm_srai_pi16 (__m64 __m, int __count) { return (__m64) __builtin_ia32_psraw ((__v4hi)__m, __count); } +static __inline __m64 +_m_psrawi (__m64 __m, int __count) +{ + return _mm_srai_pi16 (__m, __count); +} + /* Shift two 32-bit values in M right by COUNT; shift in the sign bit. */ static __inline __m64 _mm_sra_pi32 (__m64 __m, __m64 __count) @@ -369,11 +591,23 @@ _mm_sra_pi32 (__m64 __m, __m64 __count) } static __inline __m64 +_m_psrad (__m64 __m, __m64 __count) +{ + return _mm_sra_pi32 (__m, __count); +} + +static __inline __m64 _mm_srai_pi32 (__m64 __m, int __count) { return (__m64) __builtin_ia32_psrad ((__v2si)__m, __count); } +static __inline __m64 +_m_psradi (__m64 __m, int __count) +{ + return _mm_srai_pi32 (__m, __count); +} + /* Shift four 16-bit values in M right by COUNT; shift in zeros. */ static __inline __m64 _mm_srl_pi16 (__m64 __m, __m64 __count) @@ -382,11 +616,23 @@ _mm_srl_pi16 (__m64 __m, __m64 __count) } static __inline __m64 +_m_psrlw (__m64 __m, __m64 __count) +{ + return _mm_srl_pi16 (__m, __count); +} + +static __inline __m64 _mm_srli_pi16 (__m64 __m, int __count) { return (__m64) __builtin_ia32_psrlw ((__v4hi)__m, __count); } +static __inline __m64 +_m_psrlwi (__m64 __m, int __count) +{ + return _mm_srli_pi16 (__m, __count); +} + /* Shift two 32-bit values in M right by COUNT; shift in zeros. */ static __inline __m64 _mm_srl_pi32 (__m64 __m, __m64 __count) @@ -395,11 +641,23 @@ _mm_srl_pi32 (__m64 __m, __m64 __count) } static __inline __m64 +_m_psrld (__m64 __m, __m64 __count) +{ + return _mm_srl_pi32 (__m, __count); +} + +static __inline __m64 _mm_srli_pi32 (__m64 __m, int __count) { return (__m64) __builtin_ia32_psrld ((__v2si)__m, __count); } +static __inline __m64 +_m_psrldi (__m64 __m, int __count) +{ + return _mm_srli_pi32 (__m, __count); +} + /* Shift the 64-bit value in M left by COUNT; shift in zeros. */ static __inline __m64 _mm_srl_si64 (__m64 __m, __m64 __count) @@ -408,11 +666,23 @@ _mm_srl_si64 (__m64 __m, __m64 __count) } static __inline __m64 +_m_psrlq (__m64 __m, __m64 __count) +{ + return _mm_srl_si64 (__m, __count); +} + +static __inline __m64 _mm_srli_si64 (__m64 __m, int __count) { return (__m64) __builtin_ia32_psrlq ((long long)__m, (long long)__count); } +static __inline __m64 +_m_psrlqi (__m64 __m, int __count) +{ + return _mm_srli_si64 (__m, __count); +} + /* Bit-wise AND the 64-bit values in M1 and M2. */ static __inline __m64 _mm_and_si64 (__m64 __m1, __m64 __m2) @@ -420,6 +690,12 @@ _mm_and_si64 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_pand ((long long)__m1, (long long)__m2); } +static __inline __m64 +_m_pand (__m64 __m1, __m64 __m2) +{ + return _mm_and_si64 (__m1, __m2); +} + /* Bit-wise complement the 64-bit value in M1 and bit-wise AND it with the 64-bit value in M2. */ static __inline __m64 @@ -428,6 +704,12 @@ _mm_andnot_si64 (__m64 __m1, __m64 __m2) return (__m64) __builtin_ia32_pandn ((long long)__m1, (long long)__m2); } +static __inline __m64 +_m_pandn (__m64 __m1, __m64 __m2) +{ + return _mm_andnot_si64 (__m1, __m2); +} + /* Bit-wise inclusive OR the 64-bit values in M1 and M2. */ static __inline __m64 _mm_or_si64 (__m64 __m1, __m64 __m2) @@ -435,6 +717,12 @@ _mm_or_si64 (__m64 __m1, __m64 __m2) return (__m64)__builtin_ia32_por ((long long)__m1, (long long)__m2); } +static __inline __m64 +_m_por (__m64 __m1, __m64 __m2) +{ + return _mm_or_si64 (__m1, __m2); +} + /* Bit-wise exclusive OR the 64-bit values in M1 and M2. */ static __inline __m64 _mm_xor_si64 (__m64 __m1, __m64 __m2) @@ -442,6 +730,12 @@ _mm_xor_si64 (__m64 __m1, __m64 __m2) return (__m64)__builtin_ia32_pxor ((long long)__m1, (long long)__m2); } +static __inline __m64 +_m_pxor (__m64 __m1, __m64 __m2) +{ + return _mm_xor_si64 (__m1, __m2); +} + /* Compare eight 8-bit values. The result of the comparison is 0xFF if the test is true and zero if false. */ static __inline __m64 @@ -451,11 +745,23 @@ _mm_cmpeq_pi8 (__m64 __m1, __m64 __m2) } static __inline __m64 +_m_pcmpeqb (__m64 __m1, __m64 __m2) +{ + return _mm_cmpeq_pi8 (__m1, __m2); +} + +static __inline __m64 _mm_cmpgt_pi8 (__m64 __m1, __m64 __m2) { return (__m64) __builtin_ia32_pcmpgtb ((__v8qi)__m1, (__v8qi)__m2); } +static __inline __m64 +_m_pcmpgtb (__m64 __m1, __m64 __m2) +{ + return _mm_cmpgt_pi8 (__m1, __m2); +} + /* Compare four 16-bit values. The result of the comparison is 0xFFFF if the test is true and zero if false. */ static __inline __m64 @@ -465,11 +771,23 @@ _mm_cmpeq_pi16 (__m64 __m1, __m64 __m2) } static __inline __m64 +_m_pcmpeqw (__m64 __m1, __m64 __m2) +{ + return _mm_cmpeq_pi16 (__m1, __m2); +} + +static __inline __m64 _mm_cmpgt_pi16 (__m64 __m1, __m64 __m2) { return (__m64) __builtin_ia32_pcmpgtw ((__v4hi)__m1, (__v4hi)__m2); } +static __inline __m64 +_m_pcmpgtw (__m64 __m1, __m64 __m2) +{ + return _mm_cmpgt_pi16 (__m1, __m2); +} + /* Compare two 32-bit values. The result of the comparison is 0xFFFFFFFF if the test is true and zero if false. */ static __inline __m64 @@ -479,11 +797,23 @@ _mm_cmpeq_pi32 (__m64 __m1, __m64 __m2) } static __inline __m64 +_m_pcmpeqd (__m64 __m1, __m64 __m2) +{ + return _mm_cmpeq_pi32 (__m1, __m2); +} + +static __inline __m64 _mm_cmpgt_pi32 (__m64 __m1, __m64 __m2) { return (__m64) __builtin_ia32_pcmpgtd ((__v2si)__m1, (__v2si)__m2); } +static __inline __m64 +_m_pcmpgtd (__m64 __m1, __m64 __m2) +{ + return _mm_cmpgt_pi32 (__m1, __m2); +} + /* Creates a 64-bit zero. */ static __inline __m64 _mm_setzero_si64 (void) @@ -574,7 +904,7 @@ _mm_set1_pi16 (short __w) return _mm_set1_pi32 (__i); } -/* Creates a vector of four 16-bit values, all elements containing B. */ +/* Creates a vector of eight 8-bit values, all elements containing B. */ static __inline __m64 _mm_set1_pi8 (char __b) { |