From: Ekaterina Romanova Date: Thu, 3 Mar 2016 00:20:11 +0000 (+0000) Subject: This patch adds doxygen comments for all the intrinsincs in the header file tmmintrin.h. X-Git-Url: https://granicus.if.org/sourcecode?a=commitdiff_plain;h=d795e064f74f0da38e442187efce434ffd8e3c3a;p=clang This patch adds doxygen comments for all the intrinsincs in the header file tmmintrin.h. The doxygen comments are automatically generated based on Sony's intrinsics document. I got an OK from Eric Christopher to commit doxygen comments without prior code review upstream. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@262565 91177308-0d34-0410-b5e6-96231b3b80d8 --- diff --git a/lib/Headers/tmmintrin.h b/lib/Headers/tmmintrin.h index 0002890c13..a72796ba4a 100644 --- a/lib/Headers/tmmintrin.h +++ b/lib/Headers/tmmintrin.h @@ -29,187 +29,739 @@ /* Define the default attributes for the functions in this file. */ #define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("ssse3"))) +/// \brief 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. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c PABSB instruction. +/// +/// \param __a +/// 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 _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 +/// integers in the source operand and stores the 8-bit unsigned integer +/// results in the destination. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c VPABSB instruction. +/// +/// \param __a +/// A 128-bit vector of [16 x i8]. +/// \returns A 128-bit integer vector containing the absolute values of the +/// elements in the operand. static __inline__ __m128i __DEFAULT_FN_ATTRS _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 +/// integers in the source operand and stores the 16-bit unsigned integer +/// results in the destination. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c PABSW instruction. +/// +/// \param __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 _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 +/// integers in the source operand and stores the 16-bit unsigned integer +/// results in the destination. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c VPABSW instruction. +/// +/// \param __a +/// A 128-bit vector of [8 x i16]. +/// \returns A 128-bit integer vector containing the absolute values of the +/// elements in the operand. static __inline__ __m128i __DEFAULT_FN_ATTRS _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 +/// integers in the source operand and stores the 32-bit unsigned integer +/// results in the destination. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c PABSD instruction. +/// +/// \param __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 _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 +/// integers in the source operand and stores the 32-bit unsigned integer +/// results in the destination. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c VPABSD instruction. +/// +/// \param __a +/// A 128-bit vector of [4 x i32]. +/// \returns A 128-bit integer vector containing the absolute values of the +/// elements in the operand. static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_abs_epi32(__m128i __a) { return (__m128i)__builtin_ia32_pabsd128((__v4si)__a); } +/// \brief Concatenates the two 128-bit integer vector operands, and +/// right-shifts the result by the number of bytes specified in the immediate +/// operand. +/// +/// \headerfile +/// +/// \code +/// __m128i _mm_alignr_epi8(__m128i a, __m128i b, const int n); +/// \endcode +/// +/// This intrinsic corresponds to the \c PALIGNR instruction. +/// +/// \param a +/// A 128-bit vector of [16 x i8] containing one of the source operands. +/// \param b +/// A 128-bit vector of [16 x i8] containing one of the source operands. +/// \param n +/// 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__ ({ \ (__m128i)__builtin_ia32_palignr128((__v16qi)(__m128i)(a), \ (__v16qi)(__m128i)(b), (n)); }) +/// \brief Concatenates the two 64-bit integer vector operands, and right-shifts +/// the result by the number of bytes specified in the immediate operand. +/// +/// \headerfile +/// +/// \code +/// __m64 _mm_alignr_pi8(__m64 a, __m64 b, const int n); +/// \endcode +/// +/// This intrinsic corresponds to the \c PALIGNR instruction. +/// +/// \param a +/// A 64-bit vector of [8 x i8] containing one of the source operands. +/// \param b +/// A 64-bit vector of [8 x i8] containing one of the source operands. +/// \param n +/// 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)); }) +/// \brief Horizontally adds the adjacent pairs of values contained in 2 packed +/// 128-bit vectors of [8 x i16]. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c VPHADDW instruction. +/// +/// \param __a +/// A 128-bit vector of [8 x i16] containing one of the source operands. The +/// horizontal sums of the values are stored in the lower bits of the +/// destination. +/// \param __b +/// A 128-bit vector of [8 x i16] containing one of the source operands. The +/// horizontal sums of the values are stored in the upper bits of the +/// destination. +/// \returns A 128-bit vector of [8 x i16] containing the horizontal sums of +/// both operands. static __inline__ __m128i __DEFAULT_FN_ATTRS _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 +/// 128-bit vectors of [4 x i32]. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c VPHADDD instruction. +/// +/// \param __a +/// A 128-bit vector of [4 x i32] containing one of the source operands. The +/// horizontal sums of the values are stored in the lower bits of the +/// destination. +/// \param __b +/// A 128-bit vector of [4 x i32] containing one of the source operands. The +/// horizontal sums of the values are stored in the upper bits of the +/// destination. +/// \returns A 128-bit vector of [4 x i32] containing the horizontal sums of +/// both operands. static __inline__ __m128i __DEFAULT_FN_ATTRS _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 +/// 64-bit vectors of [4 x i16]. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c PHADDW instruction. +/// +/// \param __a +/// A 64-bit vector of [4 x i16] containing one of the source operands. The +/// horizontal sums of the values are stored in the lower bits of the +/// destination. +/// \param __b +/// A 64-bit vector of [4 x i16] containing one of the source operands. The +/// horizontal sums of the values are stored in the upper bits of the +/// destination. +/// \returns A 64-bit vector of [4 x i16] containing the horizontal sums of both +/// operands. static __inline__ __m64 __DEFAULT_FN_ATTRS _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 +/// 64-bit vectors of [2 x i32]. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c PHADDD instruction. +/// +/// \param __a +/// A 64-bit vector of [2 x i32] containing one of the source operands. The +/// horizontal sums of the values are stored in the lower bits of the +/// destination. +/// \param __b +/// A 64-bit vector of [2 x i32] containing one of the source operands. The +/// horizontal sums of the values are stored in the upper bits of the +/// destination. +/// \returns A 64-bit vector of [2 x i32] containing the horizontal sums of both +/// operands. static __inline__ __m64 __DEFAULT_FN_ATTRS _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. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c VPHADDSW instruction. +/// +/// \param __a +/// A 128-bit vector of [8 x i16] containing one of the source operands. The +/// horizontal sums of the values are stored in the lower bits of the +/// destination. +/// \param __b +/// A 128-bit vector of [8 x i16] containing one of the source operands. The +/// horizontal sums of the values are stored in the upper bits of the +/// destination. +/// \returns A 128-bit vector of [8 x i16] containing the horizontal saturated +/// sums of both operands. static __inline__ __m128i __DEFAULT_FN_ATTRS _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. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c PHADDSW instruction. +/// +/// \param __a +/// A 64-bit vector of [4 x i16] containing one of the source operands. The +/// horizontal sums of the values are stored in the lower bits of the +/// destination. +/// \param __b +/// A 64-bit vector of [4 x i16] containing one of the source operands. The +/// horizontal sums of the values are stored in the upper bits of the +/// destination. +/// \returns A 64-bit vector of [4 x i16] containing the horizontal saturated +/// sums of both operands. static __inline__ __m64 __DEFAULT_FN_ATTRS _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 +/// packed 128-bit vectors of [8 x i16]. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c VPHSUBW instruction. +/// +/// \param __a +/// A 128-bit vector of [8 x i16] containing one of the source operands. The +/// horizontal differences between the values are stored in the lower bits of +/// the destination. +/// \param __b +/// A 128-bit vector of [8 x i16] containing one of the source operands. The +/// horizontal differences between the values are stored in the upper bits of +/// the destination. +/// \returns A 128-bit vector of [8 x i16] containing the horizontal differences +/// of both operands. static __inline__ __m128i __DEFAULT_FN_ATTRS _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 +/// packed 128-bit vectors of [4 x i32]. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c VPHSUBD instruction. +/// +/// \param __a +/// A 128-bit vector of [4 x i32] containing one of the source operands. The +/// horizontal differences between the values are stored in the lower bits of +/// the destination. +/// \param __b +/// A 128-bit vector of [4 x i32] containing one of the source operands. The +/// horizontal differences between the values are stored in the upper bits of +/// the destination. +/// \returns A 128-bit vector of [4 x i32] containing the horizontal differences +/// of both operands. static __inline__ __m128i __DEFAULT_FN_ATTRS _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 +/// packed 64-bit vectors of [4 x i16]. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c PHSUBW instruction. +/// +/// \param __a +/// A 64-bit vector of [4 x i16] containing one of the source operands. The +/// horizontal differences between the values are stored in the lower bits of +/// the destination. +/// \param __b +/// A 64-bit vector of [4 x i16] containing one of the source operands. The +/// horizontal differences between the values are stored in the upper bits of +/// the destination. +/// \returns A 64-bit vector of [4 x i16] containing the horizontal differences +/// of both operands. static __inline__ __m64 __DEFAULT_FN_ATTRS _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 +/// packed 64-bit vectors of [2 x i32]. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c PHSUBD instruction. +/// +/// \param __a +/// A 64-bit vector of [2 x i32] containing one of the source operands. The +/// horizontal differences between the values are stored in the lower bits of +/// the destination. +/// \param __b +/// A 64-bit vector of [2 x i32] containing one of the source operands. The +/// horizontal differences between the values are stored in the upper bits of +/// the destination. +/// \returns A 64-bit vector of [2 x i32] containing the horizontal differences +/// of both operands. static __inline__ __m64 __DEFAULT_FN_ATTRS _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 +/// 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. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c VPHSUBSW instruction. +/// +/// \param __a +/// A 128-bit vector of [8 x i16] containing one of the source operands. The +/// horizontal differences between the values are stored in the lower bits of +/// the destination. +/// \param __b +/// A 128-bit vector of [8 x i16] containing one of the source operands. The +/// horizontal differences between the values are stored in the upper bits of +/// the destination. +/// \returns A 128-bit vector of [8 x i16] containing the horizontal saturated +/// differences of both operands. static __inline__ __m128i __DEFAULT_FN_ATTRS _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 +/// 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. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c PHSUBSW instruction. +/// +/// \param __a +/// A 64-bit vector of [4 x i16] containing one of the source operands. The +/// horizontal differences between the values are stored in the lower bits of +/// the destination. +/// \param __b +/// A 64-bit vector of [4 x i16] containing one of the source operands. The +/// horizontal differences between the values are stored in the upper bits of +/// 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 _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 +/// 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 +/// the corresponding bits in the destination. For example, bits [7:0] of +/// both operands are multiplied, bits [15:8] of both operands are +/// multiplied, and the sum of both results is written to bits [15:0] of the +/// destination. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c VPMADDUBSW instruction. +/// +/// \param __a +/// A 128-bit integer vector containing the first source operand. +/// \param __b +/// A 128-bit integer vector containing the second source operand. +/// \returns A 128-bit integer vector containing the sums of products of both +/// operands: +/// R0 := (__a0 * __b0) + (__a1 * __b1) +/// R1 := (__a2 * __b2) + (__a3 * __b3) +/// R2 := (__a4 * __b4) + (__a5 * __b5) +/// R3 := (__a6 * __b6) + (__a7 * __b7) +/// R4 := (__a8 * __b8) + (__a9 * __b9) +/// R5 := (__a10 * __b10) + (__a11 * __b11) +/// R6 := (__a12 * __b12) + (__a13 * __b13) +/// R7 := (__a14 * __b14) + (__a15 * __b15) static __inline__ __m128i __DEFAULT_FN_ATTRS _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 +/// 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 +/// the corresponding bits in the destination. For example, bits [7:0] of +/// both operands are multiplied, bits [15:8] of both operands are +/// multiplied, and the sum of both results is written to bits [15:0] of the +/// destination. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c PMADDUBSW instruction. +/// +/// \param __a +/// A 64-bit integer vector containing the first source operand. +/// \param __b +/// A 64-bit integer vector containing the second source operand. +/// \returns A 64-bit integer vector containing the sums of products of both +/// operands: +/// R0 := (__a0 * __b0) + (__a1 * __b1) +/// R1 := (__a2 * __b2) + (__a3 * __b3) +/// R2 := (__a4 * __b4) + (__a5 * __b5) +/// R3 := (__a6 * __b6) + (__a7 * __b7) static __inline__ __m64 __DEFAULT_FN_ATTRS _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 +/// 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. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c VPMULHRSW instruction. +/// +/// \param __a +/// A 128-bit vector of [8 x i16] containing one of the source operands. +/// \param __b +/// A 128-bit vector of [8 x i16] containing one of the source operands. +/// \returns A 128-bit vector of [8 x i16] containing the rounded and scaled +/// products of both operands. static __inline__ __m128i __DEFAULT_FN_ATTRS _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 +/// 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. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c PMULHRSW instruction. +/// +/// \param __a +/// A 64-bit vector of [4 x i16] containing one of the source operands. +/// \param __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 _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 +/// destination or clears 8-bit values in the destination, as specified by +/// the second source operand. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c VPSHUFB instruction. +/// +/// \param __a +/// A 128-bit integer vector containing the values to be copied. +/// \param __b +/// A 128-bit integer vector containing control bytes corresponding to +/// positions in the destination: +/// Bit 7: +/// 1: Clear the corresponding byte in the destination. +/// 0: Copy the selected source byte to the corresponding byte in the +/// destination. +/// Bits [6:4] Reserved. +/// Bits [3:0] select the source byte to be copied. +/// \returns A 128-bit integer vector containing the copied or cleared values. static __inline__ __m128i __DEFAULT_FN_ATTRS _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 +/// destination or clears 8-bit values in the destination, as specified by +/// the second source operand. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c PSHUFB instruction. +/// +/// \param __a +/// A 64-bit integer vector containing the values to be copied. +/// \param __b +/// A 64-bit integer vector containing control bytes corresponding to +/// positions in the destination: +/// Bit 7: +/// 1: Clear the corresponding byte in the destination. +/// 0: Copy the selected source byte to the corresponding byte in the +/// destination. +/// 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 _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 +/// the following actions as specified by the second source operand: If the +/// byte in the second source is negative, calculate the two's complement of +/// the corresponding byte in the first source, and write that value to the +/// destination. If the byte in the second source is positive, copy the +/// corresponding byte from the first source to the destination. If the byte +/// in the second source is zero, clear the corresponding byte in the +/// destination. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c VPSIGNB instruction. +/// +/// \param __a +/// A 128-bit integer vector containing the values to be copied. +/// \param __b +/// A 128-bit integer vector containing control bytes corresponding to +/// positions in the destination. +/// \returns A 128-bit integer vector containing the resultant values. static __inline__ __m128i __DEFAULT_FN_ATTRS _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 +/// the following actions as specified by the second source operand: If the +/// word in the second source is negative, calculate the two's complement of +/// the corresponding word in the first source, and write that value to the +/// destination. If the word in the second source is positive, copy the +/// corresponding word from the first source to the destination. If the word +/// in the second source is zero, clear the corresponding word in the +/// destination. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c VPSIGNW instruction. +/// +/// \param __a +/// A 128-bit integer vector containing the values to be copied. +/// \param __b +/// A 128-bit integer vector containing control words corresponding to +/// positions in the destination. +/// \returns A 128-bit integer vector containing the resultant values. static __inline__ __m128i __DEFAULT_FN_ATTRS _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 +/// the following actions as specified by the second source operand: If the +/// doubleword in the second source is negative, calculate the two's +/// complement of the corresponding word in the first source, and write that +/// value to the destination. If the doubleword in the second source is +/// positive, copy the corresponding word from the first source to the +/// destination. If the doubleword in the second source is zero, clear the +/// corresponding word in the destination. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c VPSIGND instruction. +/// +/// \param __a +/// A 128-bit integer vector containing the values to be copied. +/// \param __b +/// A 128-bit integer vector containing control doublewords corresponding to +/// positions in the destination. +/// \returns A 128-bit integer vector containing the resultant values. static __inline__ __m128i __DEFAULT_FN_ATTRS _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 +/// the following actions as specified by the second source operand: If the +/// byte in the second source is negative, calculate the two's complement of +/// the corresponding byte in the first source, and write that value to the +/// destination. If the byte in the second source is positive, copy the +/// corresponding byte from the first source to the destination. If the byte +/// in the second source is zero, clear the corresponding byte in the +/// destination. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c PSIGNB instruction. +/// +/// \param __a +/// A 64-bit integer vector containing the values to be copied. +/// \param __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 _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 +/// the following actions as specified by the second source operand: If the +/// word in the second source is negative, calculate the two's complement of +/// the corresponding word in the first source, and write that value to the +/// destination. If the word in the second source is positive, copy the +/// corresponding word from the first source to the destination. If the word +/// in the second source is zero, clear the corresponding word in the +/// destination. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c PSIGNW instruction. +/// +/// \param __a +/// A 64-bit integer vector containing the values to be copied. +/// \param __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 _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 +/// the following actions as specified by the second source operand: If the +/// doubleword in the second source is negative, calculate the two's +/// complement of the corresponding doubleword in the first source, and +/// write that value to the destination. If the doubleword in the second +/// source is positive, copy the corresponding doubleword from the first +/// source to the destination. If the doubleword in the second source is +/// zero, clear the corresponding doubleword in the destination. +/// +/// \headerfile +/// +/// This intrinsic corresponds to the \c PSIGND instruction. +/// +/// \param __a +/// A 64-bit integer vector containing the values to be copied. +/// \param __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 _mm_sign_pi32(__m64 __a, __m64 __b) {