immintrin.h

C语言库函数的原型,有用的拿去

 * in the destination
 */
extern __m256d __cdecl _mm256_hsub_pd(__m256d m1, __m256d m2);

/*
 * Subtract Horizontal Single Precision Floating-Point Values
 * **** VHSUBPS ymm1, ymm2, ymm3/m256
 * Subtract pairs of adjacent single-precision floating-point values in
 * the first source operand and second source operand and stores results
 * in the destination.
 */
extern __m256 __cdecl _mm256_hsub_ps(__m256 m1, __m256 m2);

/*
 * Maximum of Packed Double Precision Floating-Point Values
 * **** VMAXPD ymm1, ymm2, ymm3/m256
 * Performs an SIMD compare of the packed double-precision floating-point
 * values in the first source operand and the second source operand and
 * returns the maximum value for each pair of values to the destination
 */
extern __m256d __cdecl _mm256_max_pd(__m256d m1, __m256d m2);

/*
 * Maximum of Packed Single Precision Floating-Point Values
 * **** VMAXPS ymm1, ymm2, ymm3/m256
 * Performs an SIMD compare of the packed single-precision floating-point
 * values in the first source operand and the second source operand and
 * returns the maximum value for each pair of values to the destination
 */
extern __m256 __cdecl _mm256_max_ps(__m256 m1, __m256 m2);

/*
 * Minimum of Packed Double Precision Floating-Point Values
 * **** VMINPD ymm1, ymm2, ymm3/m256
 * Performs an SIMD compare of the packed double-precision floating-point
 * values in the first source operand and the second source operand and
 * returns the minimum value for each pair of values to the destination
 */
extern __m256d __cdecl _mm256_min_pd(__m256d m1, __m256d m2);

/*
 * Minimum of Packed Single Precision Floating-Point Values
 * **** VMINPS ymm1, ymm2, ymm3/m256
 * Performs an SIMD compare of the packed single-precision floating-point
 * values in the first source operand and the second source operand and
 * returns the minimum value for each pair of values to the destination
 */
extern __m256 __cdecl _mm256_min_ps(__m256 m1, __m256 m2);

/*
 * Multiply Packed Double Precision Floating-Point Values
 * **** VMULPD ymm1, ymm2, ymm3/m256
 * Performs a SIMD multiply of the four packed double-precision floating-point
 * values from the first Source operand to the Second Source operand, and
 * stores the packed double-precision floating-point results in the
 * destination
 */
extern __m256d __cdecl _mm256_mul_pd(__m256d m1, __m256d m2);

/*
 * Multiply Packed Single Precision Floating-Point Values
 * **** VMULPS ymm1, ymm2, ymm3/m256
 * Performs an SIMD multiply of the eight packed single-precision
 * floating-point values from the first source operand to the second source
 * operand, and stores the packed double-precision floating-point results in
 * the destination
 */
extern __m256 __cdecl _mm256_mul_ps(__m256 m1, __m256 m2);

/*
 * Bitwise Logical OR of Packed Double Precision Floating-Point Values
 * **** VORPD ymm1, ymm2, ymm3/m256
 * Performs a bitwise logical OR of the four packed double-precision
 * floating-point values from the first source operand and the second
 * source operand, and stores the result in the destination
 */
extern __m256d __cdecl _mm256_or_pd(__m256d m1, __m256d m2);

/*
 * Bitwise Logical OR of Packed Single Precision Floating-Point Values
 * **** VORPS ymm1, ymm2, ymm3/m256
 * Performs a bitwise logical OR of the eight packed single-precision
 * floating-point values from the first source operand and the second
 * source operand, and stores the result in the destination
 */
extern __m256 __cdecl _mm256_or_ps(__m256 m1, __m256 m2);

/*
 * Shuffle Packed Double Precision Floating-Point Values
 * **** VSHUFPD ymm1, ymm2, ymm3/m256, imm8
 * Moves either of the two packed double-precision floating-point values from
 * each double quadword in the first source operand into the low quadword
 * of each double quadword of the destination; moves either of the two packed
 * double-precision floating-point values from the second source operand into
 * the high quadword of each double quadword of the destination operand.
 * The selector operand determines which values are moved to the destination
 */
extern __m256d __cdecl _mm256_shuffle_pd(__m256d m1, __m256d m2, const int select);

/*
 * Shuffle Packed Single Precision Floating-Point Values
 * **** VSHUFPS ymm1, ymm2, ymm3/m256, imm8
 * Moves two of the four packed single-precision floating-point values
 * from each double qword of the first source operand into the low
 * quadword of each double qword of the destination; moves two of the four
 * packed single-precision floating-point values from each double qword of
 * the second source operand into to the high quadword of each double qword
 * of the destination. The selector operand determines which values are moved
 * to the destination.
 */
extern __m256 __cdecl _mm256_shuffle_ps(__m256 m1, __m256 m2, const int select);

/*
 * Subtract Packed Double Precision Floating-Point Values
 * **** VSUBPD ymm1, ymm2, ymm3/m256
 * Performs an SIMD subtract of the four packed double-precision floating-point
 * values of the second Source operand from the first Source operand, and
 * stores the packed double-precision floating-point results in the destination
 */
extern __m256d __cdecl _mm256_sub_pd(__m256d m1, __m256d m2);

/*
 * Subtract Packed Single Precision Floating-Point Values
 * **** VSUBPS ymm1, ymm2, ymm3/m256
 * Performs an SIMD subtract of the eight packed single-precision
 * floating-point values in the second Source operand from the First Source
 * operand, and stores the packed single-precision floating-point results in
 * the destination
 */
extern __m256 __cdecl _mm256_sub_ps(__m256 m1, __m256 m2);

/*
 * Bitwise Logical XOR of Packed Double Precision Floating-Point Values
 * **** VXORPD ymm1, ymm2, ymm3/m256
 * Performs a bitwise logical XOR of the four packed double-precision
 * floating-point values from the first source operand and the second
 * source operand, and stores the result in the destination
 */
extern __m256d __cdecl _mm256_xor_pd(__m256d m1, __m256d m2);

/*
 * Bitwise Logical XOR of Packed Single Precision Floating-Point Values
 * **** VXORPS ymm1, ymm2, ymm3/m256
 * Performs a bitwise logical XOR of the eight packed single-precision
 * floating-point values from the first source operand and the second
 * source operand, and stores the result in the destination
 */
extern __m256 __cdecl _mm256_xor_ps(__m256 m1, __m256 m2);

/*
 * Compare Packed Double-Precision Floating-Point Values
 * **** VCMPPD xmm1, xmm2, xmm3/m128, imm8
 * **** VCMPPD ymm1, ymm2, ymm3/m256, imm8
 * Performs an SIMD compare of the four packed double-precision floating-point
 * values in the second source operand (third operand) and the first source
 * operand (second operand) and returns the results of the comparison to the
 * destination operand (first operand). The comparison predicate operand
 * (immediate) specifies the type of comparison performed on each of the pairs
 * of packed values.
 * For 128-bit intrinsic with compare predicate values in range 0-7 compiler
 * may generate SSE2 instructions if it is warranted for performance reasons.
 */
extern __m128d __cdecl _mm_cmp_pd(__m128d m1, __m128d m2, const int predicate);
extern __m256d __cdecl _mm256_cmp_pd(__m256d m1, __m256d m2, const int predicate);

/*
 * Compare Packed Single-Precision Floating-Point Values
 * **** VCMPPS xmm1, xmm2, xmm3/m256, imm8
 * **** VCMPPS ymm1, ymm2, ymm3/m256, imm8
 * Performs a SIMD compare of the packed single-precision floating-point values
 * in the second source operand (third operand) and the first source operand
 * (second operand) and returns the results of the comparison to the destination
 * operand (first operand). The comparison predicate operand (immediate)
 * specifies the type of comparison performed on each of the pairs of packed
 * values.
 * For 128-bit intrinsic with compare predicate values in range 0-7 compiler
 * may generate SSE2 instructions if it is warranted for performance reasons.
 */
extern __m128 __cdecl _mm_cmp_ps(__m128 m1, __m128 m2, const int predicate);
extern __m256 __cdecl _mm256_cmp_ps(__m256 m1, __m256 m2, const int predicate);

/*
 * Compare Scalar Double-Precision Floating-Point Values
 * **** VCMPSD xmm1, xmm2, xmm3/m64, imm8
 * Compares the low double-precision floating-point values in the second source
 * operand (third operand) and the first source operand (second operand) and
 * returns the results in of the comparison to the destination operand (first
 * operand). The comparison predicate operand (immediate operand) specifies the
 * type of comparison performed.
 * For compare predicate values in range 0-7 compiler may generate SSE2
 * instructions if it is warranted for performance reasons.
 */
extern __m128d __cdecl _mm_cmp_sd(__m128d m1, __m128d m2, const int predicate);

/*
 * Compare Scalar Single-Precision Floating-Point Values
 * **** VCMPSS xmm1, xmm2, xmm3/m64, imm8
 * Compares the low single-precision floating-point values in the second source
 * operand (third operand) and the first source operand (second operand) and
 * returns the results of the comparison to the destination operand (first
 * operand). The comparison predicate operand (immediate operand) specifies
 * the type of comparison performed.
 * For compare predicate values in range 0-7 compiler may generate SSE2
 * instructions if it is warranted for performance reasons.
 */
extern __m128 __cdecl _mm_cmp_ss(__m128 m1, __m128 m2, const int predicate);

/*
 * Convert Packed Doubleword Integers to
 * Packed Double-Precision Floating-Point Values
 * **** VCVTDQ2PD ymm1, xmm2/m128
 * Converts four packed signed doubleword integers in the source operand to
 * four packed double-precision floating-point values in the destination
 */
extern __m256d __cdecl _mm256_cvtepi32_pd(__m128i m1);

/*
 * Convert Packed Doubleword Integers to
 * Packed Single-Precision Floating-Point Values
 * **** VCVTDQ2PS ymm1, ymm2/m256
 * Converts eight packed signed doubleword integers in the source operand to
 * eight packed double-precision floating-point values in the destination
 */
extern __m256  __cdecl _mm256_cvtepi32_ps(__m256i m1);

/*
 * Convert Packed Double-Precision Floating-point values to
 * Packed Single-Precision Floating-Point Values
 * **** VCVTPD2PS xmm1, ymm2/m256
 * Converts four packed double-precision floating-point values in the source
 * operand to four packed single-precision floating-point values in the
 * destination
 */
extern __m128  __cdecl _mm256_cvtpd_ps(__m256d m1);

/*
 * Convert Packed Single Precision Floating-Point Values to
 * Packed Singed Doubleword Integer Values
 * **** VCVTPS2DQ ymm1, ymm2/m256
 * Converts eight packed single-precision floating-point values in the source
 * operand to eight signed doubleword integers in the destination
 */
extern __m256i __cdecl _mm256_cvtps_epi32(__m256 m1);

/*
 * Convert Packed Single Precision Floating-point values to
 * Packed Double Precision Floating-Point Values
 * **** VCVTPS2PD ymm1, xmm2/m128
 * Converts four packed single-precision floating-point values in the source
 * operand to four packed double-precision floating-point values in the
 * destination
 */
extern __m256d __cdecl _mm256_cvtps_pd(__m128 m1);

/*
 * Convert with Truncation Packed Double-Precision Floating-Point values to
 * Packed Doubleword Integers
 * **** VCVTTPD2DQ xmm1, ymm2/m256
 * Converts four packed double-precision floating-point values in the source
 * operand to four packed signed doubleword integers in the destination.
 * When a conversion is inexact, a truncated (round toward zero) value is
 * returned. If a converted result is larger than the maximum signed doubleword
 * integer, the floating-point invalid exception is raised, and if this
 * exception is masked, the indefinite integer value (80000000H) is returned
*/
extern __m128i __cdecl _mm256_cvttpd_epi32(__m256d m1);

/*
 * Convert Packed Double-Precision Floating-point values to
 * Packed Doubleword Integers