diff --git a/sdk/include/crt/mmintrin.h b/sdk/include/crt/mmintrin.h index 3b1b313a643..61a1427e6dc 100644 --- a/sdk/include/crt/mmintrin.h +++ b/sdk/include/crt/mmintrin.h @@ -1,7 +1,21 @@ -/** - * This file has no copyright assigned and is placed in the Public Domain. - * This file is part of the w64 mingw-runtime package. - * No warranty is given; refer to the file DISCLAIMER within this package. +/* + * mmintrin.h + * + * This file is part of the ReactOS CRT package. + * + * Contributors: + * Timo Kreuzer (timo.kreuzer@reactos.org) + * + * THIS SOFTWARE IS NOT COPYRIGHTED + * + * This source code is offered for use in the public domain. You may + * use, modify or distribute it freely. + * + * This code is distributed in the hope that it will be useful but + * WITHOUT ANY WARRANTY. ALL WARRANTIES, EXPRESS OR IMPLIED ARE HEREBY + * DISCLAIMED. This includes but is not limited to warranties of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * */ #pragma once @@ -10,6 +24,639 @@ #include +#ifdef __cplusplus +extern "C" { +#endif +#ifdef _MSC_VER +#define DECLSPEC_INTRINTYPE __declspec(intrin_type) +#else +#define DECLSPEC_INTRINTYPE +#endif + +#if defined(_MSC_VER) && !defined(__clang__) + + typedef union DECLSPEC_INTRINTYPE _CRT_ALIGN(8) __m64 + { + unsigned __int64 m64_u64; + float m64_f32[2]; + __int8 m64_i8[8]; + __int16 m64_i16[4]; + __int32 m64_i32[2]; + __int64 m64_i64; + unsigned __int8 m64_u8[8]; + unsigned __int16 m64_u16[4]; + unsigned __int32 m64_u32[2]; + } __m64; + +#else /* _MSC_VER */ + + typedef long long __v1di __attribute__((__vector_size__(8))); + typedef int __v2si __attribute__((__vector_size__(8))); + typedef short __v4hi __attribute__((__vector_size__(8))); + typedef char __v8qi __attribute__((__vector_size__(8))); + + typedef float __m64 __attribute__((__vector_size__(8), __aligned__(16))); + +#ifdef __clang__ +#define __INTRIN_INLINE_MMX __INTRIN_INLINE __attribute__((__target__("mmx"),__min_vector_width__(64))) +#else +#define __INTRIN_INLINE_MMX __INTRIN_INLINE __attribute__((__target__("mmx"))) +#endif + +#endif /* _MSC_VER */ + +#ifdef _M_IX86 + +void _m_empty(void); +__m64 _m_from_int(int i); +int _m_to_int(__m64 m); +__m64 _m_packsswb(__m64 a, __m64 b); +__m64 _m_packssdw(__m64 a, __m64 b); +__m64 _m_packuswb(__m64 a, __m64 b); +__m64 _m_punpckhbw(__m64 a, __m64 b); +__m64 _m_punpckhwd(__m64 a, __m64 b); +__m64 _m_punpckhdq(__m64 a, __m64 b); +__m64 _m_punpcklbw(__m64 a, __m64 b); +__m64 _m_punpcklwd(__m64 a, __m64 b); +__m64 _m_punpckldq(__m64 a, __m64 b); +__m64 _m_paddb(__m64 a, __m64 b); +__m64 _m_paddw(__m64 a, __m64 b); +__m64 _m_paddd(__m64 a, __m64 b); +__m64 _m_paddsb(__m64 a, __m64 b); +__m64 _m_paddsw(__m64 a, __m64 b); +__m64 _m_paddusb(__m64 a, __m64 b); +__m64 _m_paddusw(__m64 a, __m64 b); +__m64 _m_psubb(__m64 a, __m64 b); +__m64 _m_psubw(__m64 a, __m64 b); +__m64 _m_psubd(__m64 a, __m64 b); +__m64 _m_psubsb(__m64 a, __m64 b); +__m64 _m_psubsw(__m64 a, __m64 b); +__m64 _m_psubusb(__m64 a, __m64 b); +__m64 _m_psubusw(__m64 a, __m64 b); +__m64 _m_pmaddwd(__m64 a, __m64 b); +__m64 _m_pmulhw(__m64 a, __m64 b); +__m64 _m_pmullw(__m64 a, __m64 b); +__m64 _m_psllw(__m64 a, __m64 count); +__m64 _m_psllwi(__m64 a, int imm8); +__m64 _m_pslld(__m64 a, __m64 count); +__m64 _m_pslldi(__m64 a, int imm8); +__m64 _m_psllq(__m64 a, __m64 count); +__m64 _m_psllqi(__m64 a, int imm8); +__m64 _m_psraw(__m64 a, __m64 count); +__m64 _m_psrawi(__m64 a, int imm8); +__m64 _m_psrad(__m64 a, __m64 count); +__m64 _m_psradi(__m64 a, int imm8); +__m64 _m_psrlw(__m64 a, __m64 count); +__m64 _m_psrlwi(__m64 a, int imm8); +__m64 _m_psrld(__m64 a, __m64 count); +__m64 _m_psrldi(__m64 a, int imm8); +__m64 _m_psrlq(__m64 a, __m64 count); +__m64 _m_psrlqi(__m64 a, int imm8); +__m64 _m_pand(__m64 a, __m64 b); +__m64 _m_pandn(__m64 a, __m64 b); +__m64 _m_por(__m64 a, __m64 b); +__m64 _m_pxor(__m64 a, __m64 b); +__m64 _m_pcmpeqb(__m64 a, __m64 b); +__m64 _m_pcmpgtb(__m64 a, __m64 b); +__m64 _m_pcmpeqw(__m64 a, __m64 b); +__m64 _m_pcmpgtw(__m64 a, __m64 b); +__m64 _m_pcmpeqd(__m64 a, __m64 b); +__m64 _m_pcmpgtd(__m64 a, __m64 b); +__m64 _mm_setzero_si64(void); +__m64 _mm_set_pi32(int i1, int i0); +__m64 _mm_set_pi16(short s3, short s2, short s1, short s0); +__m64 _mm_set_pi8(char b7, char b6, char b5, char b4, + char b3, char b2, char b1, char b0); +__m64 _mm_setr_pi32(int i1, int i0); +__m64 _mm_setr_pi16(short s3, short s2, short s1, short s0); +__m64 _mm_setr_pi8(char b7, char b6, char b5, char b4, + char b3, char b2, char b1, char b0); +__m64 _mm_set1_pi32(int i); +__m64 _mm_set1_pi16(short s); +__m64 _mm_set1_pi8(char b); + +/* Alternate names */ +#define _mm_empty _m_empty +#define _mm_cvtsi32_si64 _m_from_int +#define _mm_cvtsi64_si32 _m_to_int +#define _mm_packs_pi16 _m_packsswb +#define _mm_packs_pi32 _m_packssdw +#define _mm_packs_pu16 _m_packuswb +#define _mm_unpackhi_pi8 _m_punpckhbw +#define _mm_unpackhi_pi16 _m_punpckhwd +#define _mm_unpackhi_pi32 _m_punpckhdq +#define _mm_unpacklo_pi8 _m_punpcklbw +#define _mm_unpacklo_pi16 _m_punpcklwd +#define _mm_unpacklo_pi32 _m_punpckldq +#define _mm_add_pi8 _m_paddb +#define _mm_add_pi16 _m_paddw +#define _mm_add_pi32 _m_paddd +#define _mm_adds_pi8 _m_paddsb +#define _mm_adds_pi16 _m_paddsw +#define _mm_adds_pu8 _m_paddusb +#define _mm_adds_pu16 _m_paddusw +#define _mm_sub_pi8 _m_psubb +#define _mm_sub_pi16 _m_psubw +#define _mm_sub_pi32 _m_psubd +#define _mm_subs_pi8 _m_psubsb +#define _mm_subs_pi16 _m_psubsw +#define _mm_subs_pu8 _m_psubusb +#define _mm_subs_pu16 _m_psubusw +#define _mm_madd_pi16 _m_pmaddwd +#define _mm_mulhi_pi16 _m_pmulhw +#define _mm_mullo_pi16 _m_pmullw +#define _mm_sll_pi16 _m_psllw +#define _mm_slli_pi16 _m_psllwi +#define _mm_sll_pi32 _m_pslld +#define _mm_slli_pi32 _m_pslldi +#define _mm_sll_si64 _m_psllq +#define _mm_slli_si64 _m_psllqi +#define _mm_sra_pi16 _m_psraw +#define _mm_srai_pi16 _m_psrawi +#define _mm_sra_pi32 _m_psrad +#define _mm_srai_pi32 _m_psradi +#define _mm_srl_pi16 _m_psrlw +#define _mm_srli_pi16 _m_psrlwi +#define _mm_srl_pi32 _m_psrld +#define _mm_srli_pi32 _m_psrldi +#define _mm_srl_si64 _m_psrlq +#define _mm_srli_si64 _m_psrlqi +#define _mm_and_si64 _m_pand +#define _mm_andnot_si64 _m_pandn +#define _mm_or_si64 _m_por +#define _mm_xor_si64 _m_pxor +#define _mm_cmpeq_pi8 _m_pcmpeqb +#define _mm_cmpgt_pi8 _m_pcmpgtb +#define _mm_cmpeq_pi16 _m_pcmpeqw +#define _mm_cmpgt_pi16 _m_pcmpgtw +#define _mm_cmpeq_pi32 _m_pcmpeqd +#define _mm_cmpgt_pi32 _m_pcmpgtd + +/* Use intrinsics on MSVC */ +#if defined(_MSC_VER) && !defined(__clang__) +#pragma intrinsic(_m_empty) +#pragma intrinsic(_m_from_int) +#pragma intrinsic(_m_to_int) +#pragma intrinsic(_m_packsswb) +#pragma intrinsic(_m_packssdw) +#pragma intrinsic(_m_packuswb) +#pragma intrinsic(_m_punpckhbw) +#pragma intrinsic(_m_punpckhwd) +#pragma intrinsic(_m_punpckhdq) +#pragma intrinsic(_m_punpcklbw) +#pragma intrinsic(_m_punpcklwd) +#pragma intrinsic(_m_punpckldq) +#pragma intrinsic(_m_paddb) +#pragma intrinsic(_m_paddw) +#pragma intrinsic(_m_paddd) +#pragma intrinsic(_m_paddsb) +#pragma intrinsic(_m_paddsw) +#pragma intrinsic(_m_paddusb) +#pragma intrinsic(_m_paddusw) +#pragma intrinsic(_m_psubb) +#pragma intrinsic(_m_psubw) +#pragma intrinsic(_m_psubd) +#pragma intrinsic(_m_psubsb) +#pragma intrinsic(_m_psubsw) +#pragma intrinsic(_m_psubusb) +#pragma intrinsic(_m_psubusw) +#pragma intrinsic(_m_pmaddwd) +#pragma intrinsic(_m_pmulhw) +#pragma intrinsic(_m_pmullw) +#pragma intrinsic(_m_psllw) +#pragma intrinsic(_m_psllwi) +#pragma intrinsic(_m_pslld) +#pragma intrinsic(_m_pslldi) +#pragma intrinsic(_m_psllq) +#pragma intrinsic(_m_psllqi) +#pragma intrinsic(_m_psraw) +#pragma intrinsic(_m_psrawi) +#pragma intrinsic(_m_psrad) +#pragma intrinsic(_m_psradi) +#pragma intrinsic(_m_psrlw) +#pragma intrinsic(_m_psrlwi) +#pragma intrinsic(_m_psrld) +#pragma intrinsic(_m_psrldi) +#pragma intrinsic(_m_psrlq) +#pragma intrinsic(_m_psrlqi) +#pragma intrinsic(_m_pand) +#pragma intrinsic(_m_pandn) +#pragma intrinsic(_m_por) +#pragma intrinsic(_m_pxor) +#pragma intrinsic(_m_pcmpeqb) +#pragma intrinsic(_m_pcmpgtb) +#pragma intrinsic(_m_pcmpeqw) +#pragma intrinsic(_m_pcmpgtw) +#pragma intrinsic(_m_pcmpeqd) +#pragma intrinsic(_m_pcmpgtd) +#pragma intrinsic(_mm_setzero_si64) +#pragma intrinsic(_mm_set_pi32) +#pragma intrinsic(_mm_set_pi16) +#pragma intrinsic(_mm_set_pi8) +#pragma intrinsic(_mm_setr_pi32) +#pragma intrinsic(_mm_setr_pi16) +#pragma intrinsic(_mm_setr_pi8) +#pragma intrinsic(_mm_set1_pi32) +#pragma intrinsic(_mm_set1_pi16) +#pragma intrinsic(_mm_set1_pi8) + +/* Use inline functions on GCC/Clang */ +#else // GCC / Clang Clang-CL + +/* +- GCC: https://github.com/gcc-mirror/gcc/blob/master/gcc/config/i386/mmintrin.h +- Clang: https://github.com/llvm/llvm-project/blob/main/clang/lib/Headers/mmintrin.h +*/ + +// _m_empty +__INTRIN_INLINE_MMX void _mm_empty(void) +{ + __builtin_ia32_emms(); +} + +// _m_from_int +__INTRIN_INLINE_MMX __m64 _mm_cvtsi32_si64(int i) +{ + return (__m64)__builtin_ia32_vec_init_v2si(i, 0); +} + +// _m_to_int +__INTRIN_INLINE_MMX int _mm_cvtsi64_si32(__m64 m) +{ + return __builtin_ia32_vec_ext_v2si((__v2si)m, 0); +} + +// _m_packsswb +__INTRIN_INLINE_MMX __m64 _mm_packs_pi16(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_packsswb((__v4hi)a, (__v4hi)b); +} + +// _m_packssdw +__INTRIN_INLINE_MMX __m64 _mm_packs_pi32(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_packssdw((__v2si)a, (__v2si)b); +} + +// _m_packuswb +__INTRIN_INLINE_MMX __m64 _mm_packs_pu16(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_packuswb((__v4hi)a, (__v4hi)b); +} + +// _m_punpckhbw +__INTRIN_INLINE_MMX __m64 _mm_unpackhi_pi8(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_punpckhbw((__v8qi)a, (__v8qi)b); +} + +// _m_punpckhwd +__INTRIN_INLINE_MMX __m64 _mm_unpackhi_pi16(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_punpckhwd((__v4hi)a, (__v4hi)b); +} + +// _m_punpckhdq +__INTRIN_INLINE_MMX __m64 _mm_unpackhi_pi32(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_punpckhdq((__v2si)a, (__v2si)b); +} + +// _m_punpcklbw +__INTRIN_INLINE_MMX __m64 _mm_unpacklo_pi8(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_punpcklbw((__v8qi)a, (__v8qi)b); +} + +// _m_punpcklwd +__INTRIN_INLINE_MMX __m64 _mm_unpacklo_pi16(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_punpcklwd((__v4hi)a, (__v4hi)b); +} + +// _m_punpckldq +__INTRIN_INLINE_MMX __m64 _mm_unpacklo_pi32(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_punpckldq((__v2si)a, (__v2si)b); +} + +// _m_paddb +__INTRIN_INLINE_MMX __m64 _mm_add_pi8(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_paddb((__v8qi)a, (__v8qi)b); +} + +// _m_paddw +__INTRIN_INLINE_MMX __m64 _mm_add_pi16(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_paddw((__v4hi)a, (__v4hi)b); +} + +// _m_paddd +__INTRIN_INLINE_MMX __m64 _mm_add_pi32(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_paddd((__v2si)a, (__v2si)b); +} + +// _m_paddsb +__INTRIN_INLINE_MMX __m64 _mm_adds_pi8(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_paddsb((__v8qi)a, (__v8qi)b); +} + +// _m_paddsw +__INTRIN_INLINE_MMX __m64 _mm_adds_pi16(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_paddsw((__v4hi)a, (__v4hi)b); +} + +// _m_paddusb +__INTRIN_INLINE_MMX __m64 _mm_adds_pu8(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_paddusb((__v8qi)a, (__v8qi)b); +} + +// _m_paddusw +__INTRIN_INLINE_MMX __m64 _mm_adds_pu16(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_paddusw((__v4hi)a, (__v4hi)b); +} + +// _m_psubb +__INTRIN_INLINE_MMX __m64 _mm_sub_pi8(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_psubb((__v8qi)a, (__v8qi)b); +} + +// _m_psubw +__INTRIN_INLINE_MMX __m64 _mm_sub_pi16(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_psubw((__v4hi)a, (__v4hi)b); +} + +// _m_psubd +__INTRIN_INLINE_MMX __m64 _mm_sub_pi32(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_psubd((__v2si)a, (__v2si)b); +} + +// _m_psubsb +__INTRIN_INLINE_MMX __m64 _mm_subs_pi8(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_psubsb((__v8qi)a, (__v8qi)b); +} + +// _m_psubsw +__INTRIN_INLINE_MMX __m64 _mm_subs_pi16(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_psubsw((__v4hi)a, (__v4hi)b); +} + +// _m_psubusb +__INTRIN_INLINE_MMX __m64 _mm_subs_pu8(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_psubusb((__v8qi)a, (__v8qi)b); +} + +// _m_psubusw +__INTRIN_INLINE_MMX __m64 _mm_subs_pu16(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_psubusw((__v4hi)a, (__v4hi)b); +} + +// _m_pmaddwd +__INTRIN_INLINE_MMX __m64 _mm_madd_pi16(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_pmaddwd((__v4hi)a, (__v4hi)b); +} + +// _m_pmulhw +__INTRIN_INLINE_MMX __m64 _mm_mulhi_pi16(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_pmulhw((__v4hi)a, (__v4hi)b); +} + +// _m_pmullw +__INTRIN_INLINE_MMX __m64 _mm_mullo_pi16(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_pmullw((__v4hi)a, (__v4hi)b); +} + +// _m_psllw +__INTRIN_INLINE_MMX __m64 _mm_sll_pi16(__m64 a, __m64 count) +{ + return (__m64)__builtin_ia32_psllw((__v4hi)a, (__v4hi)count); +} + +// _m_psllwi +__INTRIN_INLINE_MMX __m64 _mm_slli_pi16(__m64 a, int imm8) +{ + return (__m64)__builtin_ia32_psllwi((__v4hi)a, imm8); +} + +// _m_pslld +__INTRIN_INLINE_MMX __m64 _mm_sll_pi32(__m64 a, __m64 count) +{ + return (__m64)__builtin_ia32_pslld((__v2si)a, (__v2si)count); +} + +// _m_pslldi +__INTRIN_INLINE_MMX __m64 _mm_slli_pi32(__m64 a, int imm8) +{ + return (__m64)__builtin_ia32_pslldi((__v2si)a, imm8); +} + +// _m_psllq +__INTRIN_INLINE_MMX __m64 _mm_sll_si64(__m64 a, __m64 count) +{ + return (__m64)__builtin_ia32_psllq((__v1di)a, (__v1di)count); +} + +// _m_psllqi +__INTRIN_INLINE_MMX __m64 _mm_slli_si64(__m64 a, int imm8) +{ + return (__m64)__builtin_ia32_psllqi((__v1di)a, imm8); +} + +// _m_psraw +__INTRIN_INLINE_MMX __m64 _mm_sra_pi16(__m64 a, __m64 count) +{ + return (__m64)__builtin_ia32_psraw((__v4hi)a, (__v4hi)count); +} + +// _m_psrawi +__INTRIN_INLINE_MMX __m64 _mm_srai_pi16(__m64 a, int imm8) +{ + return (__m64)__builtin_ia32_psrawi((__v4hi)a, imm8); +} + +// _m_psrad +__INTRIN_INLINE_MMX __m64 _mm_sra_pi32(__m64 a, __m64 count) +{ + return (__m64)__builtin_ia32_psrad((__v2si)a, (__v2si)count); +} + +// _m_psradi +__INTRIN_INLINE_MMX __m64 _mm_srai_pi32(__m64 a, int imm8) +{ + return (__m64)__builtin_ia32_psradi((__v2si)a, imm8); +} + +// _m_psrlw +__INTRIN_INLINE_MMX __m64 _mm_srl_pi16(__m64 a, __m64 count) +{ + return (__m64)__builtin_ia32_psrlw((__v4hi)a, (__v4hi)count); +} + +// _m_psrlwi +__INTRIN_INLINE_MMX __m64 _mm_srli_pi16(__m64 a, int imm8) +{ + return (__m64)__builtin_ia32_psrlwi((__v4hi)a, imm8); +} + +// _m_psrld +__INTRIN_INLINE_MMX __m64 _mm_srl_pi32(__m64 a, __m64 count) +{ + return (__m64)__builtin_ia32_psrld((__v2si)a, (__v2si)count); +} + +// _m_psrldi +__INTRIN_INLINE_MMX __m64 _mm_srli_pi32(__m64 a, int imm8) +{ + return (__m64)__builtin_ia32_psrldi((__v2si)a, imm8); +} + +// _m_psrlq +__INTRIN_INLINE_MMX __m64 _mm_srl_si64(__m64 a, __m64 count) +{ + return (__m64)__builtin_ia32_psrlq((__v1di)a, (__v1di)count); +} + +// _m_psrlqi +__INTRIN_INLINE_MMX __m64 _mm_srli_si64(__m64 a, int imm8) +{ + return (__m64)__builtin_ia32_psrlqi((__v1di)a, imm8); +} + +// _m_pand +__INTRIN_INLINE_MMX __m64 _mm_and_si64(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_pand((__v2si)a, (__v2si)b); +} + +// _m_pandn +__INTRIN_INLINE_MMX __m64 _mm_andnot_si64(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_pandn((__v2si)a, (__v2si)b); +} + +// _m_por +__INTRIN_INLINE_MMX __m64 _mm_or_si64(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_por((__v2si)a, (__v2si)b); +} + +// _m_pxor +__INTRIN_INLINE_MMX __m64 _mm_xor_si64(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_pxor((__v2si)a, (__v2si)b); +} + +// _m_pcmpeqb +__INTRIN_INLINE_MMX __m64 _mm_cmpeq_pi8(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_pcmpeqb((__v8qi)a, (__v8qi)b); +} + +// _m_pcmpgtb +__INTRIN_INLINE_MMX __m64 _mm_cmpgt_pi8(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_pcmpgtb((__v8qi)a, (__v8qi)b); +} + +// _m_pcmpeqw +__INTRIN_INLINE_MMX __m64 _mm_cmpeq_pi16(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_pcmpeqw((__v4hi)a, (__v4hi)b); +} + +// _m_pcmpgtw +__INTRIN_INLINE_MMX __m64 _mm_cmpgt_pi16(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_pcmpgtw((__v4hi)a, (__v4hi)b); +} + +// _m_pcmpeqd +__INTRIN_INLINE_MMX __m64 _mm_cmpeq_pi32(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_pcmpeqd((__v2si)a, (__v2si)b); +} + +// _m_pcmpgtd +__INTRIN_INLINE_MMX __m64 _mm_cmpgt_pi32(__m64 a, __m64 b) +{ + return (__m64)__builtin_ia32_pcmpgtd((__v2si)a, (__v2si)b); +} + +__INTRIN_INLINE_MMX __m64 _mm_setzero_si64(void) +{ + return (__m64) { 0 }; +} + +__INTRIN_INLINE_MMX __m64 _mm_set_pi32(int i1, int i0) +{ + return (__m64)__builtin_ia32_vec_init_v2si(i0, i1); +} + +__INTRIN_INLINE_MMX __m64 _mm_set_pi16(short s3, short s2, short s1, short s0) +{ + return (__m64)__builtin_ia32_vec_init_v4hi(s0, s1, s2, s3); +} + +__INTRIN_INLINE_MMX __m64 _mm_set_pi8(char b7, char b6, char b5, char b4, + char b3, char b2, char b1, char b0) +{ + return (__m64)__builtin_ia32_vec_init_v8qi(b0, b1, b2, b3, b4, b5, b6, b7); +} + +__INTRIN_INLINE_MMX __m64 _mm_setr_pi32(int i1, int i0) +{ + return _mm_set_pi32(i0, i1); +} + +__INTRIN_INLINE_MMX __m64 _mm_setr_pi16(short s3, short s2, short s1, short s0) +{ + return _mm_set_pi16(s0, s1, s2, s3); +} + +__INTRIN_INLINE_MMX __m64 _mm_setr_pi8(char b7, char b6, char b5, char b4, + char b3, char b2, char b1, char b0) +{ + return _mm_set_pi8(b7, b6, b5, b4, b3, b2, b1, b0); +} + +__INTRIN_INLINE_MMX __m64 _mm_set1_pi32(int i) +{ + return _mm_set_pi32(i, i); +} + +__INTRIN_INLINE_MMX __m64 _mm_set1_pi16(short s) +{ + return _mm_set_pi16(s, s, s, s); +} + +__INTRIN_INLINE_MMX __m64 _mm_set1_pi8(char b) +{ + return _mm_set_pi8(b, b, b, b, b, b, b, b); +} + +#endif /* __GNUC__ */ + +#endif /* _M_IX86 */ + +#ifdef __cplusplus +} +#endif #endif /* _MMINTRIN_H_INCLUDED */