diff options
Diffstat (limited to 'lib/Headers/tmmintrin.h')
| -rw-r--r-- | lib/Headers/tmmintrin.h | 124 |
1 files changed, 64 insertions, 60 deletions
diff --git a/lib/Headers/tmmintrin.h b/lib/Headers/tmmintrin.h index 042bfc7e3b0d1..734cd391be602 100644 --- a/lib/Headers/tmmintrin.h +++ b/lib/Headers/tmmintrin.h @@ -27,9 +27,10 @@ #include <pmmintrin.h> /* Define the default attributes for the functions in this file. */ -#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("ssse3"))) +#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("ssse3"), __min_vector_width__(64))) +#define __DEFAULT_FN_ATTRS_MMX __attribute__((__always_inline__, __nodebug__, __target__("mmx,ssse3"), __min_vector_width__(64))) -/// \brief Computes the absolute value of each of the packed 8-bit signed +/// Computes the absolute value of each of the packed 8-bit signed /// integers in the source operand and stores the 8-bit unsigned integer /// results in the destination. /// @@ -41,13 +42,13 @@ /// A 64-bit vector of [8 x i8]. /// \returns A 64-bit integer vector containing the absolute values of the /// elements in the operand. -static __inline__ __m64 __DEFAULT_FN_ATTRS +static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_abs_pi8(__m64 __a) { return (__m64)__builtin_ia32_pabsb((__v8qi)__a); } -/// \brief Computes the absolute value of each of the packed 8-bit signed +/// Computes the absolute value of each of the packed 8-bit signed /// integers in the source operand and stores the 8-bit unsigned integer /// results in the destination. /// @@ -65,7 +66,7 @@ _mm_abs_epi8(__m128i __a) return (__m128i)__builtin_ia32_pabsb128((__v16qi)__a); } -/// \brief Computes the absolute value of each of the packed 16-bit signed +/// Computes the absolute value of each of the packed 16-bit signed /// integers in the source operand and stores the 16-bit unsigned integer /// results in the destination. /// @@ -77,13 +78,13 @@ _mm_abs_epi8(__m128i __a) /// A 64-bit vector of [4 x i16]. /// \returns A 64-bit integer vector containing the absolute values of the /// elements in the operand. -static __inline__ __m64 __DEFAULT_FN_ATTRS +static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_abs_pi16(__m64 __a) { return (__m64)__builtin_ia32_pabsw((__v4hi)__a); } -/// \brief Computes the absolute value of each of the packed 16-bit signed +/// Computes the absolute value of each of the packed 16-bit signed /// integers in the source operand and stores the 16-bit unsigned integer /// results in the destination. /// @@ -101,7 +102,7 @@ _mm_abs_epi16(__m128i __a) return (__m128i)__builtin_ia32_pabsw128((__v8hi)__a); } -/// \brief Computes the absolute value of each of the packed 32-bit signed +/// Computes the absolute value of each of the packed 32-bit signed /// integers in the source operand and stores the 32-bit unsigned integer /// results in the destination. /// @@ -113,13 +114,13 @@ _mm_abs_epi16(__m128i __a) /// A 64-bit vector of [2 x i32]. /// \returns A 64-bit integer vector containing the absolute values of the /// elements in the operand. -static __inline__ __m64 __DEFAULT_FN_ATTRS +static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_abs_pi32(__m64 __a) { return (__m64)__builtin_ia32_pabsd((__v2si)__a); } -/// \brief Computes the absolute value of each of the packed 32-bit signed +/// Computes the absolute value of each of the packed 32-bit signed /// integers in the source operand and stores the 32-bit unsigned integer /// results in the destination. /// @@ -137,7 +138,7 @@ _mm_abs_epi32(__m128i __a) return (__m128i)__builtin_ia32_pabsd128((__v4si)__a); } -/// \brief Concatenates the two 128-bit integer vector operands, and +/// Concatenates the two 128-bit integer vector operands, and /// right-shifts the result by the number of bytes specified in the immediate /// operand. /// @@ -157,11 +158,11 @@ _mm_abs_epi32(__m128i __a) /// An immediate operand specifying how many bytes to right-shift the result. /// \returns A 128-bit integer vector containing the concatenated right-shifted /// value. -#define _mm_alignr_epi8(a, b, n) __extension__ ({ \ +#define _mm_alignr_epi8(a, b, n) \ (__m128i)__builtin_ia32_palignr128((__v16qi)(__m128i)(a), \ - (__v16qi)(__m128i)(b), (n)); }) + (__v16qi)(__m128i)(b), (n)) -/// \brief Concatenates the two 64-bit integer vector operands, and right-shifts +/// Concatenates the two 64-bit integer vector operands, and right-shifts /// the result by the number of bytes specified in the immediate operand. /// /// \headerfile <x86intrin.h> @@ -180,10 +181,10 @@ _mm_abs_epi32(__m128i __a) /// An immediate operand specifying how many bytes to right-shift the result. /// \returns A 64-bit integer vector containing the concatenated right-shifted /// value. -#define _mm_alignr_pi8(a, b, n) __extension__ ({ \ - (__m64)__builtin_ia32_palignr((__v8qi)(__m64)(a), (__v8qi)(__m64)(b), (n)); }) +#define _mm_alignr_pi8(a, b, n) \ + (__m64)__builtin_ia32_palignr((__v8qi)(__m64)(a), (__v8qi)(__m64)(b), (n)) -/// \brief Horizontally adds the adjacent pairs of values contained in 2 packed +/// Horizontally adds the adjacent pairs of values contained in 2 packed /// 128-bit vectors of [8 x i16]. /// /// \headerfile <x86intrin.h> @@ -206,7 +207,7 @@ _mm_hadd_epi16(__m128i __a, __m128i __b) return (__m128i)__builtin_ia32_phaddw128((__v8hi)__a, (__v8hi)__b); } -/// \brief Horizontally adds the adjacent pairs of values contained in 2 packed +/// Horizontally adds the adjacent pairs of values contained in 2 packed /// 128-bit vectors of [4 x i32]. /// /// \headerfile <x86intrin.h> @@ -229,7 +230,7 @@ _mm_hadd_epi32(__m128i __a, __m128i __b) return (__m128i)__builtin_ia32_phaddd128((__v4si)__a, (__v4si)__b); } -/// \brief Horizontally adds the adjacent pairs of values contained in 2 packed +/// Horizontally adds the adjacent pairs of values contained in 2 packed /// 64-bit vectors of [4 x i16]. /// /// \headerfile <x86intrin.h> @@ -246,13 +247,13 @@ _mm_hadd_epi32(__m128i __a, __m128i __b) /// destination. /// \returns A 64-bit vector of [4 x i16] containing the horizontal sums of both /// operands. -static __inline__ __m64 __DEFAULT_FN_ATTRS +static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_hadd_pi16(__m64 __a, __m64 __b) { return (__m64)__builtin_ia32_phaddw((__v4hi)__a, (__v4hi)__b); } -/// \brief Horizontally adds the adjacent pairs of values contained in 2 packed +/// Horizontally adds the adjacent pairs of values contained in 2 packed /// 64-bit vectors of [2 x i32]. /// /// \headerfile <x86intrin.h> @@ -269,15 +270,16 @@ _mm_hadd_pi16(__m64 __a, __m64 __b) /// destination. /// \returns A 64-bit vector of [2 x i32] containing the horizontal sums of both /// operands. -static __inline__ __m64 __DEFAULT_FN_ATTRS +static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_hadd_pi32(__m64 __a, __m64 __b) { return (__m64)__builtin_ia32_phaddd((__v2si)__a, (__v2si)__b); } -/// \brief Horizontally adds the adjacent pairs of values contained in 2 packed -/// 128-bit vectors of [8 x i16]. Positive sums greater than 7FFFh are -/// saturated to 7FFFh. Negative sums less than 8000h are saturated to 8000h. +/// Horizontally adds the adjacent pairs of values contained in 2 packed +/// 128-bit vectors of [8 x i16]. Positive sums greater than 0x7FFF are +/// saturated to 0x7FFF. Negative sums less than 0x8000 are saturated to +/// 0x8000. /// /// \headerfile <x86intrin.h> /// @@ -299,9 +301,10 @@ _mm_hadds_epi16(__m128i __a, __m128i __b) return (__m128i)__builtin_ia32_phaddsw128((__v8hi)__a, (__v8hi)__b); } -/// \brief Horizontally adds the adjacent pairs of values contained in 2 packed -/// 64-bit vectors of [4 x i16]. Positive sums greater than 7FFFh are -/// saturated to 7FFFh. Negative sums less than 8000h are saturated to 8000h. +/// Horizontally adds the adjacent pairs of values contained in 2 packed +/// 64-bit vectors of [4 x i16]. Positive sums greater than 0x7FFF are +/// saturated to 0x7FFF. Negative sums less than 0x8000 are saturated to +/// 0x8000. /// /// \headerfile <x86intrin.h> /// @@ -317,13 +320,13 @@ _mm_hadds_epi16(__m128i __a, __m128i __b) /// destination. /// \returns A 64-bit vector of [4 x i16] containing the horizontal saturated /// sums of both operands. -static __inline__ __m64 __DEFAULT_FN_ATTRS +static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_hadds_pi16(__m64 __a, __m64 __b) { return (__m64)__builtin_ia32_phaddsw((__v4hi)__a, (__v4hi)__b); } -/// \brief Horizontally subtracts the adjacent pairs of values contained in 2 +/// Horizontally subtracts the adjacent pairs of values contained in 2 /// packed 128-bit vectors of [8 x i16]. /// /// \headerfile <x86intrin.h> @@ -346,7 +349,7 @@ _mm_hsub_epi16(__m128i __a, __m128i __b) return (__m128i)__builtin_ia32_phsubw128((__v8hi)__a, (__v8hi)__b); } -/// \brief Horizontally subtracts the adjacent pairs of values contained in 2 +/// Horizontally subtracts the adjacent pairs of values contained in 2 /// packed 128-bit vectors of [4 x i32]. /// /// \headerfile <x86intrin.h> @@ -369,7 +372,7 @@ _mm_hsub_epi32(__m128i __a, __m128i __b) return (__m128i)__builtin_ia32_phsubd128((__v4si)__a, (__v4si)__b); } -/// \brief Horizontally subtracts the adjacent pairs of values contained in 2 +/// Horizontally subtracts the adjacent pairs of values contained in 2 /// packed 64-bit vectors of [4 x i16]. /// /// \headerfile <x86intrin.h> @@ -386,13 +389,13 @@ _mm_hsub_epi32(__m128i __a, __m128i __b) /// the destination. /// \returns A 64-bit vector of [4 x i16] containing the horizontal differences /// of both operands. -static __inline__ __m64 __DEFAULT_FN_ATTRS +static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_hsub_pi16(__m64 __a, __m64 __b) { return (__m64)__builtin_ia32_phsubw((__v4hi)__a, (__v4hi)__b); } -/// \brief Horizontally subtracts the adjacent pairs of values contained in 2 +/// Horizontally subtracts the adjacent pairs of values contained in 2 /// packed 64-bit vectors of [2 x i32]. /// /// \headerfile <x86intrin.h> @@ -409,16 +412,16 @@ _mm_hsub_pi16(__m64 __a, __m64 __b) /// the destination. /// \returns A 64-bit vector of [2 x i32] containing the horizontal differences /// of both operands. -static __inline__ __m64 __DEFAULT_FN_ATTRS +static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_hsub_pi32(__m64 __a, __m64 __b) { return (__m64)__builtin_ia32_phsubd((__v2si)__a, (__v2si)__b); } -/// \brief Horizontally subtracts the adjacent pairs of values contained in 2 +/// Horizontally subtracts the adjacent pairs of values contained in 2 /// packed 128-bit vectors of [8 x i16]. Positive differences greater than -/// 7FFFh are saturated to 7FFFh. Negative differences less than 8000h are -/// saturated to 8000h. +/// 0x7FFF are saturated to 0x7FFF. Negative differences less than 0x8000 are +/// saturated to 0x8000. /// /// \headerfile <x86intrin.h> /// @@ -440,10 +443,10 @@ _mm_hsubs_epi16(__m128i __a, __m128i __b) return (__m128i)__builtin_ia32_phsubsw128((__v8hi)__a, (__v8hi)__b); } -/// \brief Horizontally subtracts the adjacent pairs of values contained in 2 +/// Horizontally subtracts the adjacent pairs of values contained in 2 /// packed 64-bit vectors of [4 x i16]. Positive differences greater than -/// 7FFFh are saturated to 7FFFh. Negative differences less than 8000h are -/// saturated to 8000h. +/// 0x7FFF are saturated to 0x7FFF. Negative differences less than 0x8000 are +/// saturated to 0x8000. /// /// \headerfile <x86intrin.h> /// @@ -459,13 +462,13 @@ _mm_hsubs_epi16(__m128i __a, __m128i __b) /// the destination. /// \returns A 64-bit vector of [4 x i16] containing the horizontal saturated /// differences of both operands. -static __inline__ __m64 __DEFAULT_FN_ATTRS +static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_hsubs_pi16(__m64 __a, __m64 __b) { return (__m64)__builtin_ia32_phsubsw((__v4hi)__a, (__v4hi)__b); } -/// \brief Multiplies corresponding pairs of packed 8-bit unsigned integer +/// Multiplies corresponding pairs of packed 8-bit unsigned integer /// values contained in the first source operand and packed 8-bit signed /// integer values contained in the second source operand, adds pairs of /// contiguous products with signed saturation, and writes the 16-bit sums to @@ -499,7 +502,7 @@ _mm_maddubs_epi16(__m128i __a, __m128i __b) return (__m128i)__builtin_ia32_pmaddubsw128((__v16qi)__a, (__v16qi)__b); } -/// \brief Multiplies corresponding pairs of packed 8-bit unsigned integer +/// Multiplies corresponding pairs of packed 8-bit unsigned integer /// values contained in the first source operand and packed 8-bit signed /// integer values contained in the second source operand, adds pairs of /// contiguous products with signed saturation, and writes the 16-bit sums to @@ -523,13 +526,13 @@ _mm_maddubs_epi16(__m128i __a, __m128i __b) /// \a R1 := (\a __a2 * \a __b2) + (\a __a3 * \a __b3) \n /// \a R2 := (\a __a4 * \a __b4) + (\a __a5 * \a __b5) \n /// \a R3 := (\a __a6 * \a __b6) + (\a __a7 * \a __b7) -static __inline__ __m64 __DEFAULT_FN_ATTRS +static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_maddubs_pi16(__m64 __a, __m64 __b) { return (__m64)__builtin_ia32_pmaddubsw((__v8qi)__a, (__v8qi)__b); } -/// \brief Multiplies packed 16-bit signed integer values, truncates the 32-bit +/// Multiplies packed 16-bit signed integer values, truncates the 32-bit /// products to the 18 most significant bits by right-shifting, rounds the /// truncated value by adding 1, and writes bits [16:1] to the destination. /// @@ -549,7 +552,7 @@ _mm_mulhrs_epi16(__m128i __a, __m128i __b) return (__m128i)__builtin_ia32_pmulhrsw128((__v8hi)__a, (__v8hi)__b); } -/// \brief Multiplies packed 16-bit signed integer values, truncates the 32-bit +/// Multiplies packed 16-bit signed integer values, truncates the 32-bit /// products to the 18 most significant bits by right-shifting, rounds the /// truncated value by adding 1, and writes bits [16:1] to the destination. /// @@ -563,13 +566,13 @@ _mm_mulhrs_epi16(__m128i __a, __m128i __b) /// A 64-bit vector of [4 x i16] containing one of the source operands. /// \returns A 64-bit vector of [4 x i16] containing the rounded and scaled /// products of both operands. -static __inline__ __m64 __DEFAULT_FN_ATTRS +static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_mulhrs_pi16(__m64 __a, __m64 __b) { return (__m64)__builtin_ia32_pmulhrsw((__v4hi)__a, (__v4hi)__b); } -/// \brief Copies the 8-bit integers from a 128-bit integer vector to the +/// Copies the 8-bit integers from a 128-bit integer vector to the /// destination or clears 8-bit values in the destination, as specified by /// the second source operand. /// @@ -595,7 +598,7 @@ _mm_shuffle_epi8(__m128i __a, __m128i __b) return (__m128i)__builtin_ia32_pshufb128((__v16qi)__a, (__v16qi)__b); } -/// \brief Copies the 8-bit integers from a 64-bit integer vector to the +/// Copies the 8-bit integers from a 64-bit integer vector to the /// destination or clears 8-bit values in the destination, as specified by /// the second source operand. /// @@ -614,13 +617,13 @@ _mm_shuffle_epi8(__m128i __a, __m128i __b) /// destination. \n /// Bits [3:0] select the source byte to be copied. /// \returns A 64-bit integer vector containing the copied or cleared values. -static __inline__ __m64 __DEFAULT_FN_ATTRS +static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_shuffle_pi8(__m64 __a, __m64 __b) { return (__m64)__builtin_ia32_pshufb((__v8qi)__a, (__v8qi)__b); } -/// \brief For each 8-bit integer in the first source operand, perform one of +/// For each 8-bit integer in the first source operand, perform one of /// the following actions as specified by the second source operand. /// /// If the byte in the second source is negative, calculate the two's @@ -646,7 +649,7 @@ _mm_sign_epi8(__m128i __a, __m128i __b) return (__m128i)__builtin_ia32_psignb128((__v16qi)__a, (__v16qi)__b); } -/// \brief For each 16-bit integer in the first source operand, perform one of +/// For each 16-bit integer in the first source operand, perform one of /// the following actions as specified by the second source operand. /// /// If the word in the second source is negative, calculate the two's @@ -672,7 +675,7 @@ _mm_sign_epi16(__m128i __a, __m128i __b) return (__m128i)__builtin_ia32_psignw128((__v8hi)__a, (__v8hi)__b); } -/// \brief For each 32-bit integer in the first source operand, perform one of +/// For each 32-bit integer in the first source operand, perform one of /// the following actions as specified by the second source operand. /// /// If the doubleword in the second source is negative, calculate the two's @@ -698,7 +701,7 @@ _mm_sign_epi32(__m128i __a, __m128i __b) return (__m128i)__builtin_ia32_psignd128((__v4si)__a, (__v4si)__b); } -/// \brief For each 8-bit integer in the first source operand, perform one of +/// For each 8-bit integer in the first source operand, perform one of /// the following actions as specified by the second source operand. /// /// If the byte in the second source is negative, calculate the two's @@ -718,13 +721,13 @@ _mm_sign_epi32(__m128i __a, __m128i __b) /// A 64-bit integer vector containing control bytes corresponding to /// positions in the destination. /// \returns A 64-bit integer vector containing the resultant values. -static __inline__ __m64 __DEFAULT_FN_ATTRS +static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_sign_pi8(__m64 __a, __m64 __b) { return (__m64)__builtin_ia32_psignb((__v8qi)__a, (__v8qi)__b); } -/// \brief For each 16-bit integer in the first source operand, perform one of +/// For each 16-bit integer in the first source operand, perform one of /// the following actions as specified by the second source operand. /// /// If the word in the second source is negative, calculate the two's @@ -744,13 +747,13 @@ _mm_sign_pi8(__m64 __a, __m64 __b) /// A 64-bit integer vector containing control words corresponding to /// positions in the destination. /// \returns A 64-bit integer vector containing the resultant values. -static __inline__ __m64 __DEFAULT_FN_ATTRS +static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_sign_pi16(__m64 __a, __m64 __b) { return (__m64)__builtin_ia32_psignw((__v4hi)__a, (__v4hi)__b); } -/// \brief For each 32-bit integer in the first source operand, perform one of +/// For each 32-bit integer in the first source operand, perform one of /// the following actions as specified by the second source operand. /// /// If the doubleword in the second source is negative, calculate the two's @@ -770,12 +773,13 @@ _mm_sign_pi16(__m64 __a, __m64 __b) /// A 64-bit integer vector containing two control doublewords corresponding /// to positions in the destination. /// \returns A 64-bit integer vector containing the resultant values. -static __inline__ __m64 __DEFAULT_FN_ATTRS +static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_sign_pi32(__m64 __a, __m64 __b) { return (__m64)__builtin_ia32_psignd((__v2si)__a, (__v2si)__b); } #undef __DEFAULT_FN_ATTRS +#undef __DEFAULT_FN_ATTRS_MMX #endif /* __TMMINTRIN_H */ |