Single & Double Math library reference (x86)

Table of contents

• Function naming convention
• Data types
• Trigonometric functions
• Power, exponential, and logarithmic functions
• Inverse trigonometric functions
• Hyperbolic functions and inverse hyperbolic functions
• Error and gamma functions
• Nearest integer functions
• Other functions
• Supplementary functions

Function naming convention

The naming convention for the vectorized math functions is shown in Fig. 3.1 and 3.2. The function name is a concatenation of the following items, in this order.

• String “Sleef_”
• Name of the corresponding double precision function in math.h
• Data type specifier of a vector element, “d” and “f” for double and single precision functions, respectively
• The number of elements in a vector
• Accuracy specifier, which is a concatenation of “_u” and 10 times the maximum error for typical input domain in ULP (two digits). If the function is expected to always return the correctly rounded value, its name has no accuracy specifier.
• Vector extension specifier
  • (Nothing) : Dispatcher automatically chooses the fastest available vector extension
  • sse2 : SSE2
  • sse4 : SSE4.1
  • avx2128 : AVX2+FMA3 instructions utilized for 128 bit computation
  • avx : AVX
  • avx2 : AVX2+FMA3
  • avx512f : AVX512F
• Attributes
  • c : The function gives bit-wise consistent results across all platforms
  • f : The function requires FMA support and gives bit-wise consistent results across all platforms
  • i : Infinity is correctly handled
  • n : NaN is correctly handled
  • z : Negative zero is correctly handled

Fig. 3.1: Naming convention of vectorized functions

Fig. 3.2: Naming convention of vectorized functions with attributes

Data types for x86 architecture

Sleef___m128_2

Sleef___m128_2 is a data type for storing two __m128 values, which is defined in sleef.h as follows:

typedef struct {
  __m128 x, y;
} Sleef___m128_2;

Sleef___m128d_2

Sleef___m128d_2 is a data type for storing two __m128d values, which is defined in sleef.h as follows:

typedef struct {
  __m128d x, y;
} Sleef___m128d_2;

Sleef___m256_2

Sleef___m256_2 is a data type for storing two __m256 values, which is defined in sleef.h as follows:

typedef struct {
  __m256 x, y;
} Sleef___m256_2;

Sleef___m256d_2

Sleef___m256d_2 is a data type for storing two __m256d values, which is defined in sleef.h as follows:

typedef struct {
  __m256d x, y;
} Sleef___m256d_2;

Sleef___m512_2

Sleef___m512_2 is a data type for storing two __m512 values, which is defined in sleef.h as follows:

typedef struct {
  __m512 x, y;
} Sleef___m512_2;

Sleef___m512d_2

Sleef___m512d_2 is a data type for storing two __m512d values, which is defined in sleef.h as follows:

typedef struct {
  __m512d x, y;
} Sleef___m512d_2;

Trigonometric Functions

Vectorized double precision sine functions with 1.0 ULP error bound

#include <sleef.h>

__m128d Sleef_sind2_u10(__m128d a);
__m128d Sleef_sind2_u10sse2(__m128d a);
__m128d Sleef_sind2_u10sse4(__m128d a);
__m128d Sleef_sind2_u10avx2128(__m128d a);
__m128d Sleef_cinz_sind2_u10sse2(__m128d a);
__m128d Sleef_cinz_sind2_u10sse4(__m128d a);
__m128d Sleef_finz_sind2_u10avx2128(__m128d a);

__m256d Sleef_sind4_u10(__m256d a);
__m256d Sleef_sind4_u10avx(__m256d a);
__m256d Sleef_sind4_u10avx2(__m256d a);
__m256d Sleef_sind4_u10fma4(__m256d a);
__m256d Sleef_cinz_sind4_u10avx(__m256d a);
__m256d Sleef_finz_sind4_u10avx2(__m256d a);
__m256d Sleef_finz_sind4_u10fma4(__m256d a);

__m512d Sleef_sind8_u10(__m512d a);
__m512d Sleef_sind8_u10avx512f(__m512d a);
__m512d Sleef_cinz_sind8_u10avx512fnofma(__m512d a);
__m512d Sleef_finz_sind8_u10avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_sin_u10 with the same accuracy specification.

Vectorized single precision sine functions with 1.0 ULP error bound

#include <sleef.h>

__m128 Sleef_sinf4_u10(__m128 a);
__m128 Sleef_sinf4_u10sse2(__m128 a);
__m128 Sleef_sinf4_u10sse4(__m128 a);
__m128 Sleef_sinf4_u10avx2128(__m128 a);
__m128 Sleef_cinz_sinf4_u10sse2(__m128 a);
__m128 Sleef_cinz_sinf4_u10sse4(__m128 a);
__m128 Sleef_finz_sinf4_u10avx2128(__m128 a);

__m256 Sleef_sinf8_u10(__m256 a);
__m256 Sleef_sinf8_u10avx(__m256 a);
__m256 Sleef_sinf8_u10avx2(__m256 a);
__m256 Sleef_sinf8_u10fma4(__m256 a);
__m256 Sleef_cinz_sinf8_u10avx(__m256 a);
__m256 Sleef_finz_sinf8_u10avx2(__m256 a);
__m256 Sleef_finz_sinf8_u10fma4(__m256 a);

__m512 Sleef_sinf16_u10(__m512 a);
__m512 Sleef_sinf16_u10avx512f(__m512 a);
__m512 Sleef_cinz_sinf16_u10avx512fnofma(__m512 a);
__m512 Sleef_finz_sinf16_u10avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_sinf_u10 with the same accuracy specification.

Vectorized double precision sine functions with 3.5 ULP error bound

#include <sleef.h>

__m128d Sleef_sind2_u35(__m128d a);
__m128d Sleef_sind2_u35sse2(__m128d a);
__m128d Sleef_sind2_u35sse4(__m128d a);
__m128d Sleef_sind2_u35avx2128(__m128d a);
__m128d Sleef_cinz_sind2_u35sse2(__m128d a);
__m128d Sleef_cinz_sind2_u35sse4(__m128d a);
__m128d Sleef_finz_sind2_u35avx2128(__m128d a);

__m256d Sleef_sind4_u35(__m256d a);
__m256d Sleef_sind4_u35avx(__m256d a);
__m256d Sleef_sind4_u35avx2(__m256d a);
__m256d Sleef_sind4_u35fma4(__m256d a);
__m256d Sleef_cinz_sind4_u35avx(__m256d a);
__m256d Sleef_finz_sind4_u35avx2(__m256d a);
__m256d Sleef_finz_sind4_u35fma4(__m256d a);

__m512d Sleef_sind8_u35(__m512d a);
__m512d Sleef_sind8_u35avx512f(__m512d a);
__m512d Sleef_cinz_sind8_u35avx512fnofma(__m512d a);
__m512d Sleef_finz_sind8_u35avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_sin_u35 with the same accuracy specification.

Vectorized single precision sine functions with 3.5 ULP error bound

#include <sleef.h>

__m128 Sleef_sinf4_u35(__m128 a);
__m128 Sleef_sinf4_u35sse2(__m128 a);
__m128 Sleef_sinf4_u35sse4(__m128 a);
__m128 Sleef_sinf4_u35avx2128(__m128 a);
__m128 Sleef_cinz_sinf4_u35sse2(__m128 a);
__m128 Sleef_cinz_sinf4_u35sse4(__m128 a);
__m128 Sleef_finz_sinf4_u35avx2128(__m128 a);

__m256 Sleef_sinf8_u35(__m256 a);
__m256 Sleef_sinf8_u35avx(__m256 a);
__m256 Sleef_sinf8_u35avx2(__m256 a);
__m256 Sleef_sinf8_u35fma4(__m256 a);
__m256 Sleef_cinz_sinf8_u35avx(__m256 a);
__m256 Sleef_finz_sinf8_u35avx2(__m256 a);
__m256 Sleef_finz_sinf8_u35fma4(__m256 a);

__m512 Sleef_sinf16_u35(__m512 a);
__m512 Sleef_sinf16_u35avx512f(__m512 a);
__m512 Sleef_cinz_sinf16_u35avx512fnofma(__m512 a);
__m512 Sleef_finz_sinf16_u35avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_sinf_u35 with the same accuracy specification.

Vectorized double precision cosine functions with 1.0 ULP error bound

#include <sleef.h>

__m128d Sleef_cosd2_u10(__m128d a);
__m128d Sleef_cosd2_u10sse2(__m128d a);
__m128d Sleef_cosd2_u10sse4(__m128d a);
__m128d Sleef_cosd2_u10avx2128(__m128d a);
__m128d Sleef_cinz_cosd2_u10sse2(__m128d a);
__m128d Sleef_cinz_cosd2_u10sse4(__m128d a);
__m128d Sleef_finz_cosd2_u10avx2128(__m128d a);

__m256d Sleef_cosd4_u10(__m256d a);
__m256d Sleef_cosd4_u10avx(__m256d a);
__m256d Sleef_cosd4_u10avx2(__m256d a);
__m256d Sleef_cosd4_u10fma4(__m256d a);
__m256d Sleef_cinz_cosd4_u10avx(__m256d a);
__m256d Sleef_finz_cosd4_u10avx2(__m256d a);
__m256d Sleef_finz_cosd4_u10fma4(__m256d a);

__m512d Sleef_cosd8_u10(__m512d a);
__m512d Sleef_cosd8_u10avx512f(__m512d a);
__m512d Sleef_cinz_cosd8_u10avx512fnofma(__m512d a);
__m512d Sleef_finz_cosd8_u10avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_cos_u10 with the same accuracy specification.

Vectorized single precision cosine functions with 1.0 ULP error bound

#include <sleef.h>

__m128 Sleef_cosf4_u10(__m128 a);
__m128 Sleef_cosf4_u10sse2(__m128 a);
__m128 Sleef_cosf4_u10sse4(__m128 a);
__m128 Sleef_cosf4_u10avx2128(__m128 a);
__m128 Sleef_cinz_cosf4_u10sse2(__m128 a);
__m128 Sleef_cinz_cosf4_u10sse4(__m128 a);
__m128 Sleef_finz_cosf4_u10avx2128(__m128 a);

__m256 Sleef_cosf8_u10(__m256 a);
__m256 Sleef_cosf8_u10avx(__m256 a);
__m256 Sleef_cosf8_u10avx2(__m256 a);
__m256 Sleef_cosf8_u10fma4(__m256 a);
__m256 Sleef_cinz_cosf8_u10avx(__m256 a);
__m256 Sleef_finz_cosf8_u10avx2(__m256 a);
__m256 Sleef_finz_cosf8_u10fma4(__m256 a);

__m512 Sleef_cosf16_u10(__m512 a);
__m512 Sleef_cosf16_u10avx512f(__m512 a);
__m512 Sleef_cinz_cosf16_u10avx512fnofma(__m512 a);
__m512 Sleef_finz_cosf16_u10avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_cosf_u10 with the same accuracy specification.

Vectorized double precision cosine functions with 3.5 ULP error bound

#include <sleef.h>

__m128d Sleef_cosd2_u35(__m128d a);
__m128d Sleef_cosd2_u35sse2(__m128d a);
__m128d Sleef_cosd2_u35sse4(__m128d a);
__m128d Sleef_cosd2_u35avx2128(__m128d a);
__m128d Sleef_cinz_cosd2_u35sse2(__m128d a);
__m128d Sleef_cinz_cosd2_u35sse4(__m128d a);
__m128d Sleef_finz_cosd2_u35avx2128(__m128d a);

__m256d Sleef_cosd4_u35(__m256d a);
__m256d Sleef_cosd4_u35avx(__m256d a);
__m256d Sleef_cosd4_u35avx2(__m256d a);
__m256d Sleef_cosd4_u35fma4(__m256d a);
__m256d Sleef_cinz_cosd4_u35avx(__m256d a);
__m256d Sleef_finz_cosd4_u35avx2(__m256d a);
__m256d Sleef_finz_cosd4_u35fma4(__m256d a);

__m512d Sleef_cosd8_u35(__m512d a);
__m512d Sleef_cosd8_u35avx512f(__m512d a);
__m512d Sleef_cinz_cosd8_u35avx512fnofma(__m512d a);
__m512d Sleef_finz_cosd8_u35avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_cos_u35 with the same accuracy specification.

Vectorized single precision cosine functions with 3.5 ULP error bound

#include <sleef.h>

__m128 Sleef_cosf4_u35(__m128 a);
__m128 Sleef_cosf4_u35sse2(__m128 a);
__m128 Sleef_cosf4_u35sse4(__m128 a);
__m128 Sleef_cosf4_u35avx2128(__m128 a);
__m128 Sleef_cinz_cosf4_u35sse2(__m128 a);
__m128 Sleef_cinz_cosf4_u35sse4(__m128 a);
__m128 Sleef_finz_cosf4_u35avx2128(__m128 a);

__m256 Sleef_cosf8_u35(__m256 a);
__m256 Sleef_cosf8_u35avx(__m256 a);
__m256 Sleef_cosf8_u35avx2(__m256 a);
__m256 Sleef_cosf8_u35fma4(__m256 a);
__m256 Sleef_cinz_cosf8_u35avx(__m256 a);
__m256 Sleef_finz_cosf8_u35avx2(__m256 a);
__m256 Sleef_finz_cosf8_u35fma4(__m256 a);

__m512 Sleef_cosf16_u35(__m512 a);
__m512 Sleef_cosf16_u35avx512f(__m512 a);
__m512 Sleef_cinz_cosf16_u35avx512fnofma(__m512 a);
__m512 Sleef_finz_cosf16_u35avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_cosf_u35 with the same accuracy specification.

Vectorized double precision combined sine and cosine functions with 1.0 ULP error bound

#include <sleef.h>

Sleef___m128d_2 Sleef_sincosd2_u10(__m128d a);
Sleef___m128d_2 Sleef_sincosd2_u10sse2(__m128d a);
Sleef___m128d_2 Sleef_sincosd2_u10sse4(__m128d a);
Sleef___m128d_2 Sleef_sincosd2_u10avx2128(__m128d a);
Sleef___m128d_2 Sleef_cinz_sincosd2_u10sse2(__m128d a);
Sleef___m128d_2 Sleef_cinz_sincosd2_u10sse4(__m128d a);
Sleef___m128d_2 Sleef_finz_sincosd2_u10avx2128(__m128d a);

Sleef___m256d_2 Sleef_sincosd4_u10(__m256d a);
Sleef___m256d_2 Sleef_sincosd4_u10avx(__m256d a);
Sleef___m256d_2 Sleef_sincosd4_u10avx2(__m256d a);
Sleef___m256d_2 Sleef_sincosd4_u10fma4(__m256d a);
Sleef___m256d_2 Sleef_cinz_sincosd4_u10avx(__m256d a);
Sleef___m256d_2 Sleef_finz_sincosd4_u10avx2(__m256d a);
Sleef___m256d_2 Sleef_finz_sincosd4_u10fma4(__m256d a);

Sleef___m512d_2 Sleef_sincosd8_u10(__m512d a);
Sleef___m512d_2 Sleef_sincosd8_u10avx512f(__m512d a);
Sleef___m512d_2 Sleef_cinz_sincosd8_u10avx512fnofma(__m512d a);
Sleef___m512d_2 Sleef_finz_sincosd8_u10avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_sincos_u10 with the same accuracy specification.

Vectorized single precision combined sine and cosine functions with 1.0 ULP error bound

#include <sleef.h>

Sleef___m128_2 Sleef_sincosf4_u10(__m128 a);
Sleef___m128_2 Sleef_sincosf4_u10sse2(__m128 a);
Sleef___m128_2 Sleef_sincosf4_u10sse4(__m128 a);
Sleef___m128_2 Sleef_sincosf4_u10avx2128(__m128 a);
Sleef___m128_2 Sleef_cinz_sincosf4_u10sse2(__m128 a);
Sleef___m128_2 Sleef_cinz_sincosf4_u10sse4(__m128 a);
Sleef___m128_2 Sleef_finz_sincosf4_u10avx2128(__m128 a);

Sleef___m256_2 Sleef_sincosf8_u10(__m256 a);
Sleef___m256_2 Sleef_sincosf8_u10avx(__m256 a);
Sleef___m256_2 Sleef_sincosf8_u10avx2(__m256 a);
Sleef___m256_2 Sleef_sincosf8_u10fma4(__m256 a);
Sleef___m256_2 Sleef_cinz_sincosf8_u10avx(__m256 a);
Sleef___m256_2 Sleef_finz_sincosf8_u10avx2(__m256 a);
Sleef___m256_2 Sleef_finz_sincosf8_u10fma4(__m256 a);

Sleef___m512_2 Sleef_sincosf16_u10(__m512 a);
Sleef___m512_2 Sleef_sincosf16_u10avx512f(__m512 a);
Sleef___m512_2 Sleef_cinz_sincosf16_u10avx512fnofma(__m512 a);
Sleef___m512_2 Sleef_finz_sincosf16_u10avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_sincosf_u10 with the same accuracy specification.

Vectorized double precision combined sine and cosine functions with 3.5 ULP error bound

#include <sleef.h>

Sleef___m128d_2 Sleef_sincosd2_u35(__m128d a);
Sleef___m128d_2 Sleef_sincosd2_u35sse2(__m128d a);
Sleef___m128d_2 Sleef_sincosd2_u35sse4(__m128d a);
Sleef___m128d_2 Sleef_sincosd2_u35avx2128(__m128d a);
Sleef___m128d_2 Sleef_cinz_sincosd2_u35sse2(__m128d a);
Sleef___m128d_2 Sleef_cinz_sincosd2_u35sse4(__m128d a);
Sleef___m128d_2 Sleef_finz_sincosd2_u35avx2128(__m128d a);

Sleef___m256d_2 Sleef_sincosd4_u35(__m256d a);
Sleef___m256d_2 Sleef_sincosd4_u35avx(__m256d a);
Sleef___m256d_2 Sleef_sincosd4_u35avx2(__m256d a);
Sleef___m256d_2 Sleef_sincosd4_u35fma4(__m256d a);
Sleef___m256d_2 Sleef_cinz_sincosd4_u35avx(__m256d a);
Sleef___m256d_2 Sleef_finz_sincosd4_u35avx2(__m256d a);
Sleef___m256d_2 Sleef_finz_sincosd4_u35fma4(__m256d a);

Sleef___m512d_2 Sleef_sincosd8_u35(__m512d a);
Sleef___m512d_2 Sleef_sincosd8_u35avx512f(__m512d a);
Sleef___m512d_2 Sleef_cinz_sincosd8_u35avx512fnofma(__m512d a);
Sleef___m512d_2 Sleef_finz_sincosd8_u35avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_sincos_u35 with the same accuracy specification.

Vectorized single precision combined sine and cosine functions with 3.5 ULP error bound

#include <sleef.h>

Sleef___m128_2 Sleef_sincosf4_u35(__m128 a);
Sleef___m128_2 Sleef_sincosf4_u35sse2(__m128 a);
Sleef___m128_2 Sleef_sincosf4_u35sse4(__m128 a);
Sleef___m128_2 Sleef_sincosf4_u35avx2128(__m128 a);
Sleef___m128_2 Sleef_cinz_sincosf4_u35sse2(__m128 a);
Sleef___m128_2 Sleef_cinz_sincosf4_u35sse4(__m128 a);
Sleef___m128_2 Sleef_finz_sincosf4_u35avx2128(__m128 a);

Sleef___m256_2 Sleef_sincosf8_u35(__m256 a);
Sleef___m256_2 Sleef_sincosf8_u35avx(__m256 a);
Sleef___m256_2 Sleef_sincosf8_u35avx2(__m256 a);
Sleef___m256_2 Sleef_sincosf8_u35fma4(__m256 a);
Sleef___m256_2 Sleef_cinz_sincosf8_u35avx(__m256 a);
Sleef___m256_2 Sleef_finz_sincosf8_u35avx2(__m256 a);
Sleef___m256_2 Sleef_finz_sincosf8_u35fma4(__m256 a);

Sleef___m512_2 Sleef_sincosf16_u35(__m512 a);
Sleef___m512_2 Sleef_sincosf16_u35avx512f(__m512 a);
Sleef___m512_2 Sleef_cinz_sincosf16_u35avx512fnofma(__m512 a);
Sleef___m512_2 Sleef_finz_sincosf16_u35avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_sincosf_u35 with the same accuracy specification.

Vectorized double precision sine functions with 0.506 ULP error bound

#include <sleef.h>

__m128d Sleef_sinpid2_u05(__m128d a);
__m128d Sleef_sinpid2_u05sse2(__m128d a);
__m128d Sleef_sinpid2_u05sse4(__m128d a);
__m128d Sleef_sinpid2_u05avx2128(__m128d a);
__m128d Sleef_cinz_sinpid2_u05sse2(__m128d a);
__m128d Sleef_cinz_sinpid2_u05sse4(__m128d a);
__m128d Sleef_finz_sinpid2_u05avx2128(__m128d a);

__m256d Sleef_sinpid4_u05(__m256d a);
__m256d Sleef_sinpid4_u05avx(__m256d a);
__m256d Sleef_sinpid4_u05avx2(__m256d a);
__m256d Sleef_sinpid4_u05fma4(__m256d a);
__m256d Sleef_cinz_sinpid4_u05avx(__m256d a);
__m256d Sleef_finz_sinpid4_u05avx2(__m256d a);
__m256d Sleef_finz_sinpid4_u05fma4(__m256d a);

__m512d Sleef_sinpid8_u05(__m512d a);
__m512d Sleef_sinpid8_u05avx512f(__m512d a);
__m512d Sleef_cinz_sinpid8_u05avx512fnofma(__m512d a);
__m512d Sleef_finz_sinpid8_u05avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_sinpi_u05 with the same accuracy specification.

Vectorized single precision sine functions with 0.506 ULP error bound

#include <sleef.h>

__m128 Sleef_sinpif4_u05(__m128 a);
__m128 Sleef_sinpif4_u05sse2(__m128 a);
__m128 Sleef_sinpif4_u05sse4(__m128 a);
__m128 Sleef_sinpif4_u05avx2128(__m128 a);
__m128 Sleef_cinz_sinpif4_u05sse2(__m128 a);
__m128 Sleef_cinz_sinpif4_u05sse4(__m128 a);
__m128 Sleef_finz_sinpif4_u05avx2128(__m128 a);

__m256 Sleef_sinpif8_u05(__m256 a);
__m256 Sleef_sinpif8_u05avx(__m256 a);
__m256 Sleef_sinpif8_u05avx2(__m256 a);
__m256 Sleef_sinpif8_u05fma4(__m256 a);
__m256 Sleef_cinz_sinpif8_u05avx(__m256 a);
__m256 Sleef_finz_sinpif8_u05avx2(__m256 a);
__m256 Sleef_finz_sinpif8_u05fma4(__m256 a);

__m512 Sleef_sinpif16_u05(__m512 a);
__m512 Sleef_sinpif16_u05avx512f(__m512 a);
__m512 Sleef_cinz_sinpif16_u05avx512fnofma(__m512 a);
__m512 Sleef_finz_sinpif16_u05avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_sinpif_u05 with the same accuracy specification.

Vectorized double precision cosine functions with 0.506 ULP error bound

#include <sleef.h>

__m128d Sleef_cospid2_u05(__m128d a);
__m128d Sleef_cospid2_u05sse2(__m128d a);
__m128d Sleef_cospid2_u05sse4(__m128d a);
__m128d Sleef_cospid2_u05avx2128(__m128d a);
__m128d Sleef_cinz_cospid2_u05sse2(__m128d a);
__m128d Sleef_cinz_cospid2_u05sse4(__m128d a);
__m128d Sleef_finz_cospid2_u05avx2128(__m128d a);

__m256d Sleef_cospid4_u05(__m256d a);
__m256d Sleef_cospid4_u05avx(__m256d a);
__m256d Sleef_cospid4_u05avx2(__m256d a);
__m256d Sleef_cospid4_u05fma4(__m256d a);
__m256d Sleef_cinz_cospid4_u05avx(__m256d a);
__m256d Sleef_finz_cospid4_u05avx2(__m256d a);
__m256d Sleef_finz_cospid4_u05fma4(__m256d a);

__m512d Sleef_cospid8_u05(__m512d a);
__m512d Sleef_cospid8_u05avx512f(__m512d a);
__m512d Sleef_cinz_cospid8_u05avx512fnofma(__m512d a);
__m512d Sleef_finz_cospid8_u05avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_cospi_u05 with the same accuracy specification.

Vectorized single precision cosine functions with 0.506 ULP error bound

#include <sleef.h>

__m128 Sleef_cospif4_u05(__m128 a);
__m128 Sleef_cospif4_u05sse2(__m128 a);
__m128 Sleef_cospif4_u05sse4(__m128 a);
__m128 Sleef_cospif4_u05avx2128(__m128 a);
__m128 Sleef_cinz_cospif4_u05sse2(__m128 a);
__m128 Sleef_cinz_cospif4_u05sse4(__m128 a);
__m128 Sleef_finz_cospif4_u05avx2128(__m128 a);

__m256 Sleef_cospif8_u05(__m256 a);
__m256 Sleef_cospif8_u05avx(__m256 a);
__m256 Sleef_cospif8_u05avx2(__m256 a);
__m256 Sleef_cospif8_u05fma4(__m256 a);
__m256 Sleef_cinz_cospif8_u05avx(__m256 a);
__m256 Sleef_finz_cospif8_u05avx2(__m256 a);
__m256 Sleef_finz_cospif8_u05fma4(__m256 a);

__m512 Sleef_cospif16_u05(__m512 a);
__m512 Sleef_cospif16_u05avx512f(__m512 a);
__m512 Sleef_cinz_cospif16_u05avx512fnofma(__m512 a);
__m512 Sleef_finz_cospif16_u05avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_cospif_u05 with the same accuracy specification.

Vectorized double precision combined sine and cosine functions with 0.506 ULP error bound

#include <sleef.h>

Sleef___m128d_2 Sleef_sincospid2_u05(__m128d a);
Sleef___m128d_2 Sleef_sincospid2_u05sse2(__m128d a);
Sleef___m128d_2 Sleef_sincospid2_u05sse4(__m128d a);
Sleef___m128d_2 Sleef_sincospid2_u05avx2128(__m128d a);
Sleef___m128d_2 Sleef_cinz_sincospid2_u05sse2(__m128d a);
Sleef___m128d_2 Sleef_cinz_sincospid2_u05sse4(__m128d a);
Sleef___m128d_2 Sleef_finz_sincospid2_u05avx2128(__m128d a);

Sleef___m256d_2 Sleef_sincospid4_u05(__m256d a);
Sleef___m256d_2 Sleef_sincospid4_u05avx(__m256d a);
Sleef___m256d_2 Sleef_sincospid4_u05avx2(__m256d a);
Sleef___m256d_2 Sleef_sincospid4_u05fma4(__m256d a);
Sleef___m256d_2 Sleef_cinz_sincospid4_u05avx(__m256d a);
Sleef___m256d_2 Sleef_finz_sincospid4_u05avx2(__m256d a);
Sleef___m256d_2 Sleef_finz_sincospid4_u05fma4(__m256d a);

Sleef___m512d_2 Sleef_sincospid8_u05(__m512d a);
Sleef___m512d_2 Sleef_sincospid8_u05avx512f(__m512d a);
Sleef___m512d_2 Sleef_cinz_sincospid8_u05avx512fnofma(__m512d a);
Sleef___m512d_2 Sleef_finz_sincospid8_u05avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_sincospi_u05 with the same accuracy specification.

Vectorized single precision combined sine and cosine functions with 0.506 ULP error bound

#include <sleef.h>

Sleef___m128_2 Sleef_sincospif4_u05(__m128 a);
Sleef___m128_2 Sleef_sincospif4_u05sse2(__m128 a);
Sleef___m128_2 Sleef_sincospif4_u05sse4(__m128 a);
Sleef___m128_2 Sleef_sincospif4_u05avx2128(__m128 a);
Sleef___m128_2 Sleef_cinz_sincospif4_u05sse2(__m128 a);
Sleef___m128_2 Sleef_cinz_sincospif4_u05sse4(__m128 a);
Sleef___m128_2 Sleef_finz_sincospif4_u05avx2128(__m128 a);

Sleef___m256_2 Sleef_sincospif8_u05(__m256 a);
Sleef___m256_2 Sleef_sincospif8_u05avx(__m256 a);
Sleef___m256_2 Sleef_sincospif8_u05avx2(__m256 a);
Sleef___m256_2 Sleef_sincospif8_u05fma4(__m256 a);
Sleef___m256_2 Sleef_cinz_sincospif8_u05avx(__m256 a);
Sleef___m256_2 Sleef_finz_sincospif8_u05avx2(__m256 a);
Sleef___m256_2 Sleef_finz_sincospif8_u05fma4(__m256 a);

Sleef___m512_2 Sleef_sincospif16_u05(__m512 a);
Sleef___m512_2 Sleef_sincospif16_u05avx512f(__m512 a);
Sleef___m512_2 Sleef_cinz_sincospif16_u05avx512fnofma(__m512 a);
Sleef___m512_2 Sleef_finz_sincospif16_u05avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_sincospif_u05 with the same accuracy specification.

Vectorized double precision combined sine and cosine functions with 3.5 ULP error bound

#include <sleef.h>

Sleef___m128d_2 Sleef_sincospid2_u35(__m128d a);
Sleef___m128d_2 Sleef_sincospid2_u35sse2(__m128d a);
Sleef___m128d_2 Sleef_sincospid2_u35sse4(__m128d a);
Sleef___m128d_2 Sleef_sincospid2_u35avx2128(__m128d a);
Sleef___m128d_2 Sleef_cinz_sincospid2_u35sse2(__m128d a);
Sleef___m128d_2 Sleef_cinz_sincospid2_u35sse4(__m128d a);
Sleef___m128d_2 Sleef_finz_sincospid2_u35avx2128(__m128d a);

Sleef___m256d_2 Sleef_sincospid4_u35(__m256d a);
Sleef___m256d_2 Sleef_sincospid4_u35avx(__m256d a);
Sleef___m256d_2 Sleef_sincospid4_u35avx2(__m256d a);
Sleef___m256d_2 Sleef_sincospid4_u35fma4(__m256d a);
Sleef___m256d_2 Sleef_cinz_sincospid4_u35avx(__m256d a);
Sleef___m256d_2 Sleef_finz_sincospid4_u35avx2(__m256d a);
Sleef___m256d_2 Sleef_finz_sincospid4_u35fma4(__m256d a);

Sleef___m512d_2 Sleef_sincospid8_u35(__m512d a);
Sleef___m512d_2 Sleef_sincospid8_u35avx512f(__m512d a);
Sleef___m512d_2 Sleef_cinz_sincospid8_u35avx512fnofma(__m512d a);
Sleef___m512d_2 Sleef_finz_sincospid8_u35avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_sincospi_u35 with the same accuracy specification.

Vectorized single precision combined sine and cosine functions with 3.5 ULP error bound

#include <sleef.h>

Sleef___m128_2 Sleef_sincospif4_u35(__m128 a);
Sleef___m128_2 Sleef_sincospif4_u35sse2(__m128 a);
Sleef___m128_2 Sleef_sincospif4_u35sse4(__m128 a);
Sleef___m128_2 Sleef_sincospif4_u35avx2128(__m128 a);
Sleef___m128_2 Sleef_cinz_sincospif4_u35sse2(__m128 a);
Sleef___m128_2 Sleef_cinz_sincospif4_u35sse4(__m128 a);
Sleef___m128_2 Sleef_finz_sincospif4_u35avx2128(__m128 a);

Sleef___m256_2 Sleef_sincospif8_u35(__m256 a);
Sleef___m256_2 Sleef_sincospif8_u35avx(__m256 a);
Sleef___m256_2 Sleef_sincospif8_u35avx2(__m256 a);
Sleef___m256_2 Sleef_sincospif8_u35fma4(__m256 a);
Sleef___m256_2 Sleef_cinz_sincospif8_u35avx(__m256 a);
Sleef___m256_2 Sleef_finz_sincospif8_u35avx2(__m256 a);
Sleef___m256_2 Sleef_finz_sincospif8_u35fma4(__m256 a);

Sleef___m512_2 Sleef_sincospif16_u35(__m512 a);
Sleef___m512_2 Sleef_sincospif16_u35avx512f(__m512 a);
Sleef___m512_2 Sleef_cinz_sincospif16_u35avx512fnofma(__m512 a);
Sleef___m512_2 Sleef_finz_sincospif16_u35avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_sincospif_u35 with the same accuracy specification.

Vectorized double precision tangent functions with 1.0 ULP error bound

#include <sleef.h>

__m128d Sleef_tand2_u10(__m128d a);
__m128d Sleef_tand2_u10sse2(__m128d a);
__m128d Sleef_tand2_u10sse4(__m128d a);
__m128d Sleef_tand2_u10avx2128(__m128d a);
__m128d Sleef_cinz_tand2_u10sse2(__m128d a);
__m128d Sleef_cinz_tand2_u10sse4(__m128d a);
__m128d Sleef_finz_tand2_u10avx2128(__m128d a);

__m256d Sleef_tand4_u10(__m256d a);
__m256d Sleef_tand4_u10avx(__m256d a);
__m256d Sleef_tand4_u10avx2(__m256d a);
__m256d Sleef_tand4_u10fma4(__m256d a);
__m256d Sleef_cinz_tand4_u10avx(__m256d a);
__m256d Sleef_finz_tand4_u10avx2(__m256d a);
__m256d Sleef_finz_tand4_u10fma4(__m256d a);

__m512d Sleef_tand8_u10(__m512d a);
__m512d Sleef_tand8_u10avx512f(__m512d a);
__m512d Sleef_cinz_tand8_u10avx512fnofma(__m512d a);
__m512d Sleef_finz_tand8_u10avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_tan_u10 with the same accuracy specification.

Vectorized single precision tangent functions with 1.0 ULP error bound

#include <sleef.h>

__m128 Sleef_tanf4_u10(__m128 a);
__m128 Sleef_tanf4_u10sse2(__m128 a);
__m128 Sleef_tanf4_u10sse4(__m128 a);
__m128 Sleef_tanf4_u10avx2128(__m128 a);
__m128 Sleef_cinz_tanf4_u10sse2(__m128 a);
__m128 Sleef_cinz_tanf4_u10sse4(__m128 a);
__m128 Sleef_finz_tanf4_u10avx2128(__m128 a);

__m256 Sleef_tanf8_u10(__m256 a);
__m256 Sleef_tanf8_u10avx(__m256 a);
__m256 Sleef_tanf8_u10avx2(__m256 a);
__m256 Sleef_tanf8_u10fma4(__m256 a);
__m256 Sleef_cinz_tanf8_u10avx(__m256 a);
__m256 Sleef_finz_tanf8_u10avx2(__m256 a);
__m256 Sleef_finz_tanf8_u10fma4(__m256 a);

__m512 Sleef_tanf16_u10(__m512 a);
__m512 Sleef_tanf16_u10avx512f(__m512 a);
__m512 Sleef_cinz_tanf16_u10avx512fnofma(__m512 a);
__m512 Sleef_finz_tanf16_u10avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_tanf_u10 with the same accuracy specification.

Vectorized double precision tangent functions with 3.5 ULP error bound

#include <sleef.h>

__m128d Sleef_tand2_u35(__m128d a);
__m128d Sleef_tand2_u35sse2(__m128d a);
__m128d Sleef_tand2_u35sse4(__m128d a);
__m128d Sleef_tand2_u35avx2128(__m128d a);
__m128d Sleef_cinz_tand2_u35sse2(__m128d a);
__m128d Sleef_cinz_tand2_u35sse4(__m128d a);
__m128d Sleef_finz_tand2_u35avx2128(__m128d a);

__m256d Sleef_tand4_u35(__m256d a);
__m256d Sleef_tand4_u35avx(__m256d a);
__m256d Sleef_tand4_u35avx2(__m256d a);
__m256d Sleef_tand4_u35fma4(__m256d a);
__m256d Sleef_cinz_tand4_u35avx(__m256d a);
__m256d Sleef_finz_tand4_u35avx2(__m256d a);
__m256d Sleef_finz_tand4_u35fma4(__m256d a);

__m512d Sleef_tand8_u35(__m512d a);
__m512d Sleef_tand8_u35avx512f(__m512d a);
__m512d Sleef_cinz_tand8_u35avx512fnofma(__m512d a);
__m512d Sleef_finz_tand8_u35avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_tan_u35 with the same accuracy specification.

Vectorized single precision tangent functions with 3.5 ULP error bound

#include <sleef.h>

__m128 Sleef_tanf4_u35(__m128 a);
__m128 Sleef_tanf4_u35sse2(__m128 a);
__m128 Sleef_tanf4_u35sse4(__m128 a);
__m128 Sleef_tanf4_u35avx2128(__m128 a);
__m128 Sleef_cinz_tanf4_u35sse2(__m128 a);
__m128 Sleef_cinz_tanf4_u35sse4(__m128 a);
__m128 Sleef_finz_tanf4_u35avx2128(__m128 a);

__m256 Sleef_tanf8_u35(__m256 a);
__m256 Sleef_tanf8_u35avx(__m256 a);
__m256 Sleef_tanf8_u35avx2(__m256 a);
__m256 Sleef_tanf8_u35fma4(__m256 a);
__m256 Sleef_cinz_tanf8_u35avx(__m256 a);
__m256 Sleef_finz_tanf8_u35avx2(__m256 a);
__m256 Sleef_finz_tanf8_u35fma4(__m256 a);

__m512 Sleef_tanf16_u35(__m512 a);
__m512 Sleef_tanf16_u35avx512f(__m512 a);
__m512 Sleef_cinz_tanf16_u35avx512fnofma(__m512 a);
__m512 Sleef_finz_tanf16_u35avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_tanf_u35 with the same accuracy specification.

Power, exponential, and logarithmic functions

Vectorized double precision power functions with 1.0 ULP error bound

#include <sleef.h>

__m128d Sleef_powd2_u10(__m128d a, __m128d b);
__m128d Sleef_powd2_u10sse2(__m128d a, __m128d b);
__m128d Sleef_powd2_u10sse4(__m128d a, __m128d b);
__m128d Sleef_powd2_u10avx2128(__m128d a, __m128d b);
__m128d Sleef_cinz_powd2_u10sse2(__m128d a, __m128d b);
__m128d Sleef_cinz_powd2_u10sse4(__m128d a, __m128d b);
__m128d Sleef_finz_powd2_u10avx2128(__m128d a, __m128d b);

__m256d Sleef_powd4_u10(__m256d a, __m256d b);
__m256d Sleef_powd4_u10avx(__m256d a, __m256d b);
__m256d Sleef_powd4_u10avx2(__m256d a, __m256d b);
__m256d Sleef_powd4_u10fma4(__m256d a, __m256d b);
__m256d Sleef_cinz_powd4_u10avx(__m256d a, __m256d b);
__m256d Sleef_finz_powd4_u10avx2(__m256d a, __m256d b);
__m256d Sleef_finz_powd4_u10fma4(__m256d a, __m256d b);

__m512d Sleef_powd8_u10(__m512d a, __m512d b);
__m512d Sleef_powd8_u10avx512f(__m512d a, __m512d b);
__m512d Sleef_cinz_powd8_u10avx512fnofma(__m512d a, __m512d b);
__m512d Sleef_finz_powd8_u10avx512f(__m512d a, __m512d b);

Link with -lsleef.

These are the vectorized functions of Sleef_pow_u10 with the same accuracy specification.

Vectorized single precision power functions with 1.0 ULP error bound

#include <sleef.h>

__m128 Sleef_powf4_u10(__m128 a, __m128 b);
__m128 Sleef_powf4_u10sse2(__m128 a, __m128 b);
__m128 Sleef_powf4_u10sse4(__m128 a, __m128 b);
__m128 Sleef_powf4_u10avx2128(__m128 a, __m128 b);
__m128 Sleef_cinz_powf4_u10sse2(__m128 a, __m128 b);
__m128 Sleef_cinz_powf4_u10sse4(__m128 a, __m128 b);
__m128 Sleef_finz_powf4_u10avx2128(__m128 a, __m128 b);

__m256 Sleef_powf8_u10(__m256 a, __m256 b);
__m256 Sleef_powf8_u10avx(__m256 a, __m256 b);
__m256 Sleef_powf8_u10avx2(__m256 a, __m256 b);
__m256 Sleef_powf8_u10fma4(__m256 a, __m256 b);
__m256 Sleef_cinz_powf8_u10avx(__m256 a, __m256 b);
__m256 Sleef_finz_powf8_u10avx2(__m256 a, __m256 b);
__m256 Sleef_finz_powf8_u10fma4(__m256 a, __m256 b);

__m512 Sleef_powf16_u10(__m512 a, __m512 b);
__m512 Sleef_powf16_u10avx512f(__m512 a, __m512 b);
__m512 Sleef_cinz_powf16_u10avx512fnofma(__m512 a, __m512 b);
__m512 Sleef_finz_powf16_u10avx512f(__m512 a, __m512 b);

Link with -lsleef.

These are the vectorized functions of Sleef_powf_u10 with the same accuracy specification.

Vectorized double precision natural logarithmic functions with 1.0 ULP error bound

#include <sleef.h>

__m128d Sleef_logd2_u10(__m128d a);
__m128d Sleef_logd2_u10sse2(__m128d a);
__m128d Sleef_logd2_u10sse4(__m128d a);
__m128d Sleef_logd2_u10avx2128(__m128d a);
__m128d Sleef_cinz_logd2_u10sse2(__m128d a);
__m128d Sleef_cinz_logd2_u10sse4(__m128d a);
__m128d Sleef_finz_logd2_u10avx2128(__m128d a);

__m256d Sleef_logd4_u10(__m256d a);
__m256d Sleef_logd4_u10avx(__m256d a);
__m256d Sleef_logd4_u10avx2(__m256d a);
__m256d Sleef_logd4_u10fma4(__m256d a);
__m256d Sleef_cinz_logd4_u10avx(__m256d a);
__m256d Sleef_finz_logd4_u10avx2(__m256d a);
__m256d Sleef_finz_logd4_u10fma4(__m256d a);

__m512d Sleef_logd8_u10(__m512d a);
__m512d Sleef_logd8_u10avx512f(__m512d a);
__m512d Sleef_cinz_logd8_u10avx512fnofma(__m512d a);
__m512d Sleef_finz_logd8_u10avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_log_u10 with the same accuracy specification.

Vectorized single precision natural logarithmic functions with 1.0 ULP error bound

#include <sleef.h>

__m128 Sleef_logf4_u10(__m128 a);
__m128 Sleef_logf4_u10sse2(__m128 a);
__m128 Sleef_logf4_u10sse4(__m128 a);
__m128 Sleef_logf4_u10avx2128(__m128 a);
__m128 Sleef_cinz_logf4_u10sse2(__m128 a);
__m128 Sleef_cinz_logf4_u10sse4(__m128 a);
__m128 Sleef_finz_logf4_u10avx2128(__m128 a);

__m256 Sleef_logf8_u10(__m256 a);
__m256 Sleef_logf8_u10avx(__m256 a);
__m256 Sleef_logf8_u10avx2(__m256 a);
__m256 Sleef_logf8_u10fma4(__m256 a);
__m256 Sleef_cinz_logf8_u10avx(__m256 a);
__m256 Sleef_finz_logf8_u10avx2(__m256 a);
__m256 Sleef_finz_logf8_u10fma4(__m256 a);

__m512 Sleef_logf16_u10(__m512 a);
__m512 Sleef_logf16_u10avx512f(__m512 a);
__m512 Sleef_cinz_logf16_u10avx512fnofma(__m512 a);
__m512 Sleef_finz_logf16_u10avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_logf_u10 with the same accuracy specification.

Vectorized double precision natural logarithmic functions with 3.5 ULP error bound

#include <sleef.h>

__m128d Sleef_logd2_u35(__m128d a);
__m128d Sleef_logd2_u35sse2(__m128d a);
__m128d Sleef_logd2_u35sse4(__m128d a);
__m128d Sleef_logd2_u35avx2128(__m128d a);
__m128d Sleef_cinz_logd2_u35sse2(__m128d a);
__m128d Sleef_cinz_logd2_u35sse4(__m128d a);
__m128d Sleef_finz_logd2_u35avx2128(__m128d a);

__m256d Sleef_logd4_u35(__m256d a);
__m256d Sleef_logd4_u35avx(__m256d a);
__m256d Sleef_logd4_u35avx2(__m256d a);
__m256d Sleef_logd4_u35fma4(__m256d a);
__m256d Sleef_cinz_logd4_u35avx(__m256d a);
__m256d Sleef_finz_logd4_u35avx2(__m256d a);
__m256d Sleef_finz_logd4_u35fma4(__m256d a);

__m512d Sleef_logd8_u35(__m512d a);
__m512d Sleef_logd8_u35avx512f(__m512d a);
__m512d Sleef_cinz_logd8_u35avx512fnofma(__m512d a);
__m512d Sleef_finz_logd8_u35avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_log_u35 with the same accuracy specification.

Vectorized single precision natural logarithmic functions with 3.5 ULP error bound

#include <sleef.h>

__m128 Sleef_logf4_u35(__m128 a);
__m128 Sleef_logf4_u35sse2(__m128 a);
__m128 Sleef_logf4_u35sse4(__m128 a);
__m128 Sleef_logf4_u35avx2128(__m128 a);
__m128 Sleef_cinz_logf4_u35sse2(__m128 a);
__m128 Sleef_cinz_logf4_u35sse4(__m128 a);
__m128 Sleef_finz_logf4_u35avx2128(__m128 a);

__m256 Sleef_logf8_u35(__m256 a);
__m256 Sleef_logf8_u35avx(__m256 a);
__m256 Sleef_logf8_u35avx2(__m256 a);
__m256 Sleef_logf8_u35fma4(__m256 a);
__m256 Sleef_cinz_logf8_u35avx(__m256 a);
__m256 Sleef_finz_logf8_u35avx2(__m256 a);
__m256 Sleef_finz_logf8_u35fma4(__m256 a);

__m512 Sleef_logf16_u35(__m512 a);
__m512 Sleef_logf16_u35avx512f(__m512 a);
__m512 Sleef_cinz_logf16_u35avx512fnofma(__m512 a);
__m512 Sleef_finz_logf16_u35avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_logf_u35 with the same accuracy specification.

Vectorized double precision base-10 logarithmic functions with 1.0 ULP error bound

#include <sleef.h>

__m128d Sleef_log10d2_u10(__m128d a);
__m128d Sleef_log10d2_u10sse2(__m128d a);
__m128d Sleef_log10d2_u10sse4(__m128d a);
__m128d Sleef_log10d2_u10avx2128(__m128d a);
__m128d Sleef_cinz_log10d2_u10sse2(__m128d a);
__m128d Sleef_cinz_log10d2_u10sse4(__m128d a);
__m128d Sleef_finz_log10d2_u10avx2128(__m128d a);

__m256d Sleef_log10d4_u10(__m256d a);
__m256d Sleef_log10d4_u10avx(__m256d a);
__m256d Sleef_log10d4_u10avx2(__m256d a);
__m256d Sleef_log10d4_u10fma4(__m256d a);
__m256d Sleef_cinz_log10d4_u10avx(__m256d a);
__m256d Sleef_finz_log10d4_u10avx2(__m256d a);
__m256d Sleef_finz_log10d4_u10fma4(__m256d a);

__m512d Sleef_log10d8_u10(__m512d a);
__m512d Sleef_log10d8_u10avx512f(__m512d a);
__m512d Sleef_cinz_log10d8_u10avx512fnofma(__m512d a);
__m512d Sleef_finz_log10d8_u10avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_log10_u10 with the same accuracy specification.

Vectorized single precision base-10 logarithmic functions with 1.0 ULP error bound

#include <sleef.h>

__m128 Sleef_log10f4_u10(__m128 a);
__m128 Sleef_log10f4_u10sse2(__m128 a);
__m128 Sleef_log10f4_u10sse4(__m128 a);
__m128 Sleef_log10f4_u10avx2128(__m128 a);
__m128 Sleef_cinz_log10f4_u10sse2(__m128 a);
__m128 Sleef_cinz_log10f4_u10sse4(__m128 a);
__m128 Sleef_finz_log10f4_u10avx2128(__m128 a);

__m256 Sleef_log10f8_u10(__m256 a);
__m256 Sleef_log10f8_u10avx(__m256 a);
__m256 Sleef_log10f8_u10avx2(__m256 a);
__m256 Sleef_log10f8_u10fma4(__m256 a);
__m256 Sleef_cinz_log10f8_u10avx(__m256 a);
__m256 Sleef_finz_log10f8_u10avx2(__m256 a);
__m256 Sleef_finz_log10f8_u10fma4(__m256 a);

__m512 Sleef_log10f16_u10(__m512 a);
__m512 Sleef_log10f16_u10avx512f(__m512 a);
__m512 Sleef_cinz_log10f16_u10avx512fnofma(__m512 a);
__m512 Sleef_finz_log10f16_u10avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_log10f_u10 with the same accuracy specification.

Vectorized double precision base-2 logarithmic functions with 1.0 ULP error bound

#include <sleef.h>

__m128d Sleef_log2d2_u10(__m128d a);
__m128d Sleef_log2d2_u10sse2(__m128d a);
__m128d Sleef_log2d2_u10sse4(__m128d a);
__m128d Sleef_log2d2_u10avx2128(__m128d a);
__m128d Sleef_cinz_log2d2_u10sse2(__m128d a);
__m128d Sleef_cinz_log2d2_u10sse4(__m128d a);
__m128d Sleef_finz_log2d2_u10avx2128(__m128d a);

__m256d Sleef_log2d4_u10(__m256d a);
__m256d Sleef_log2d4_u10avx(__m256d a);
__m256d Sleef_log2d4_u10avx2(__m256d a);
__m256d Sleef_log2d4_u10fma4(__m256d a);
__m256d Sleef_cinz_log2d4_u10avx(__m256d a);
__m256d Sleef_finz_log2d4_u10avx2(__m256d a);
__m256d Sleef_finz_log2d4_u10fma4(__m256d a);

__m512d Sleef_log2d8_u10(__m512d a);
__m512d Sleef_log2d8_u10avx512f(__m512d a);
__m512d Sleef_cinz_log2d8_u10avx512fnofma(__m512d a);
__m512d Sleef_finz_log2d8_u10avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_log2_u10 with the same accuracy specification.

Vectorized single precision base-2 logarithmic functions with 1.0 ULP error bound

#include <sleef.h>

__m128 Sleef_log2f4_u10(__m128 a);
__m128 Sleef_log2f4_u10sse2(__m128 a);
__m128 Sleef_log2f4_u10sse4(__m128 a);
__m128 Sleef_log2f4_u10avx2128(__m128 a);
__m128 Sleef_cinz_log2f4_u10sse2(__m128 a);
__m128 Sleef_cinz_log2f4_u10sse4(__m128 a);
__m128 Sleef_finz_log2f4_u10avx2128(__m128 a);

__m256 Sleef_log2f8_u10(__m256 a);
__m256 Sleef_log2f8_u10avx(__m256 a);
__m256 Sleef_log2f8_u10avx2(__m256 a);
__m256 Sleef_log2f8_u10fma4(__m256 a);
__m256 Sleef_cinz_log2f8_u10avx(__m256 a);
__m256 Sleef_finz_log2f8_u10avx2(__m256 a);
__m256 Sleef_finz_log2f8_u10fma4(__m256 a);

__m512 Sleef_log2f16_u10(__m512 a);
__m512 Sleef_log2f16_u10avx512f(__m512 a);
__m512 Sleef_cinz_log2f16_u10avx512fnofma(__m512 a);
__m512 Sleef_finz_log2f16_u10avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_log2f_u10 with the same accuracy specification.

Vectorized double precision logarithm of one plus argument with 1.0 ULP error bound

#include <sleef.h>

__m128d Sleef_log1pd2_u10(__m128d a);
__m128d Sleef_log1pd2_u10sse2(__m128d a);
__m128d Sleef_log1pd2_u10sse4(__m128d a);
__m128d Sleef_log1pd2_u10avx2128(__m128d a);
__m128d Sleef_cinz_log1pd2_u10sse2(__m128d a);
__m128d Sleef_cinz_log1pd2_u10sse4(__m128d a);
__m128d Sleef_finz_log1pd2_u10avx2128(__m128d a);

__m256d Sleef_log1pd4_u10(__m256d a);
__m256d Sleef_log1pd4_u10avx(__m256d a);
__m256d Sleef_log1pd4_u10avx2(__m256d a);
__m256d Sleef_log1pd4_u10fma4(__m256d a);
__m256d Sleef_cinz_log1pd4_u10avx(__m256d a);
__m256d Sleef_finz_log1pd4_u10avx2(__m256d a);
__m256d Sleef_finz_log1pd4_u10fma4(__m256d a);

__m512d Sleef_log1pd8_u10(__m512d a);
__m512d Sleef_log1pd8_u10avx512f(__m512d a);
__m512d Sleef_cinz_log1pd8_u10avx512fnofma(__m512d a);
__m512d Sleef_finz_log1pd8_u10avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_log1p_u10 with the same accuracy specification.

Vectorized single precision logarithm of one plus argument with 1.0 ULP error bound

#include <sleef.h>

__m128 Sleef_log1pf4_u10(__m128 a);
__m128 Sleef_log1pf4_u10sse2(__m128 a);
__m128 Sleef_log1pf4_u10sse4(__m128 a);
__m128 Sleef_log1pf4_u10avx2128(__m128 a);
__m128 Sleef_cinz_log1pf4_u10sse2(__m128 a);
__m128 Sleef_cinz_log1pf4_u10sse4(__m128 a);
__m128 Sleef_finz_log1pf4_u10avx2128(__m128 a);

__m256 Sleef_log1pf8_u10(__m256 a);
__m256 Sleef_log1pf8_u10avx(__m256 a);
__m256 Sleef_log1pf8_u10avx2(__m256 a);
__m256 Sleef_log1pf8_u10fma4(__m256 a);
__m256 Sleef_cinz_log1pf8_u10avx(__m256 a);
__m256 Sleef_finz_log1pf8_u10avx2(__m256 a);
__m256 Sleef_finz_log1pf8_u10fma4(__m256 a);

__m512 Sleef_log1pf16_u10(__m512 a);
__m512 Sleef_log1pf16_u10avx512f(__m512 a);
__m512 Sleef_cinz_log1pf16_u10avx512fnofma(__m512 a);
__m512 Sleef_finz_log1pf16_u10avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_log1pf_u10 with the same accuracy specification.

Vectorized double precision base-e exponential functions functions with 1.0 ULP error bound

#include <sleef.h>

__m128d Sleef_expd2_u10(__m128d a);
__m128d Sleef_expd2_u10sse2(__m128d a);
__m128d Sleef_expd2_u10sse4(__m128d a);
__m128d Sleef_expd2_u10avx2128(__m128d a);
__m128d Sleef_cinz_expd2_u10sse2(__m128d a);
__m128d Sleef_cinz_expd2_u10sse4(__m128d a);
__m128d Sleef_finz_expd2_u10avx2128(__m128d a);

__m256d Sleef_expd4_u10(__m256d a);
__m256d Sleef_expd4_u10avx(__m256d a);
__m256d Sleef_expd4_u10avx2(__m256d a);
__m256d Sleef_expd4_u10fma4(__m256d a);
__m256d Sleef_cinz_expd4_u10avx(__m256d a);
__m256d Sleef_finz_expd4_u10avx2(__m256d a);
__m256d Sleef_finz_expd4_u10fma4(__m256d a);

__m512d Sleef_expd8_u10(__m512d a);
__m512d Sleef_expd8_u10avx512f(__m512d a);
__m512d Sleef_cinz_expd8_u10avx512fnofma(__m512d a);
__m512d Sleef_finz_expd8_u10avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_exp_u10 with the same accuracy specification.

Vectorized single precision base-e exponential functions functions with 1.0 ULP error bound

#include <sleef.h>

__m128 Sleef_expf4_u10(__m128 a);
__m128 Sleef_expf4_u10sse2(__m128 a);
__m128 Sleef_expf4_u10sse4(__m128 a);
__m128 Sleef_expf4_u10avx2128(__m128 a);
__m128 Sleef_cinz_expf4_u10sse2(__m128 a);
__m128 Sleef_cinz_expf4_u10sse4(__m128 a);
__m128 Sleef_finz_expf4_u10avx2128(__m128 a);

__m256 Sleef_expf8_u10(__m256 a);
__m256 Sleef_expf8_u10avx(__m256 a);
__m256 Sleef_expf8_u10avx2(__m256 a);
__m256 Sleef_expf8_u10fma4(__m256 a);
__m256 Sleef_cinz_expf8_u10avx(__m256 a);
__m256 Sleef_finz_expf8_u10avx2(__m256 a);
__m256 Sleef_finz_expf8_u10fma4(__m256 a);

__m512 Sleef_expf16_u10(__m512 a);
__m512 Sleef_expf16_u10avx512f(__m512 a);
__m512 Sleef_cinz_expf16_u10avx512fnofma(__m512 a);
__m512 Sleef_finz_expf16_u10avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_expf_u10 with the same accuracy specification.

Vectorized double precision base-2 exponential functions functions with 1.0 ULP error bound

#include <sleef.h>

__m128d Sleef_exp2d2_u10(__m128d a);
__m128d Sleef_exp2d2_u10sse2(__m128d a);
__m128d Sleef_exp2d2_u10sse4(__m128d a);
__m128d Sleef_exp2d2_u10avx2128(__m128d a);
__m128d Sleef_cinz_exp2d2_u10sse2(__m128d a);
__m128d Sleef_cinz_exp2d2_u10sse4(__m128d a);
__m128d Sleef_finz_exp2d2_u10avx2128(__m128d a);

__m256d Sleef_exp2d4_u10(__m256d a);
__m256d Sleef_exp2d4_u10avx(__m256d a);
__m256d Sleef_exp2d4_u10avx2(__m256d a);
__m256d Sleef_exp2d4_u10fma4(__m256d a);
__m256d Sleef_cinz_exp2d4_u10avx(__m256d a);
__m256d Sleef_finz_exp2d4_u10avx2(__m256d a);
__m256d Sleef_finz_exp2d4_u10fma4(__m256d a);

__m512d Sleef_exp2d8_u10(__m512d a);
__m512d Sleef_exp2d8_u10avx512f(__m512d a);
__m512d Sleef_cinz_exp2d8_u10avx512fnofma(__m512d a);
__m512d Sleef_finz_exp2d8_u10avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_exp2_u10 with the same accuracy specification.

Vectorized single precision base-2 exponential functions functions with 1.0 ULP error bound

#include <sleef.h>

__m128 Sleef_exp2f4_u10(__m128 a);
__m128 Sleef_exp2f4_u10sse2(__m128 a);
__m128 Sleef_exp2f4_u10sse4(__m128 a);
__m128 Sleef_exp2f4_u10avx2128(__m128 a);
__m128 Sleef_cinz_exp2f4_u10sse2(__m128 a);
__m128 Sleef_cinz_exp2f4_u10sse4(__m128 a);
__m128 Sleef_finz_exp2f4_u10avx2128(__m128 a);

__m256 Sleef_exp2f8_u10(__m256 a);
__m256 Sleef_exp2f8_u10avx(__m256 a);
__m256 Sleef_exp2f8_u10avx2(__m256 a);
__m256 Sleef_exp2f8_u10fma4(__m256 a);
__m256 Sleef_cinz_exp2f8_u10avx(__m256 a);
__m256 Sleef_finz_exp2f8_u10avx2(__m256 a);
__m256 Sleef_finz_exp2f8_u10fma4(__m256 a);

__m512 Sleef_exp2f16_u10(__m512 a);
__m512 Sleef_exp2f16_u10avx512f(__m512 a);
__m512 Sleef_cinz_exp2f16_u10avx512fnofma(__m512 a);
__m512 Sleef_finz_exp2f16_u10avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_exp2f_u10 with the same accuracy specification.

Vectorized double precision base-10 exponential functions functions with 1.09 ULP error bound

#include <sleef.h>

__m128d Sleef_exp10d2_u10(__m128d a);
__m128d Sleef_exp10d2_u10sse2(__m128d a);
__m128d Sleef_exp10d2_u10sse4(__m128d a);
__m128d Sleef_exp10d2_u10avx2128(__m128d a);
__m128d Sleef_cinz_exp10d2_u10sse2(__m128d a);
__m128d Sleef_cinz_exp10d2_u10sse4(__m128d a);
__m128d Sleef_finz_exp10d2_u10avx2128(__m128d a);

__m256d Sleef_exp10d4_u10(__m256d a);
__m256d Sleef_exp10d4_u10avx(__m256d a);
__m256d Sleef_exp10d4_u10avx2(__m256d a);
__m256d Sleef_exp10d4_u10fma4(__m256d a);
__m256d Sleef_cinz_exp10d4_u10avx(__m256d a);
__m256d Sleef_finz_exp10d4_u10avx2(__m256d a);
__m256d Sleef_finz_exp10d4_u10fma4(__m256d a);

__m512d Sleef_exp10d8_u10(__m512d a);
__m512d Sleef_exp10d8_u10avx512f(__m512d a);
__m512d Sleef_cinz_exp10d8_u10avx512fnofma(__m512d a);
__m512d Sleef_finz_exp10d8_u10avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_exp10_u10 with the same accuracy specification.

Vectorized single precision base-10 exponential functions functions with 1.0 ULP error bound

#include <sleef.h>

__m128 Sleef_exp10f4_u10(__m128 a);
__m128 Sleef_exp10f4_u10sse2(__m128 a);
__m128 Sleef_exp10f4_u10sse4(__m128 a);
__m128 Sleef_exp10f4_u10avx2128(__m128 a);
__m128 Sleef_cinz_exp10f4_u10sse2(__m128 a);
__m128 Sleef_cinz_exp10f4_u10sse4(__m128 a);
__m128 Sleef_finz_exp10f4_u10avx2128(__m128 a);

__m256 Sleef_exp10f8_u10(__m256 a);
__m256 Sleef_exp10f8_u10avx(__m256 a);
__m256 Sleef_exp10f8_u10avx2(__m256 a);
__m256 Sleef_exp10f8_u10fma4(__m256 a);
__m256 Sleef_cinz_exp10f8_u10avx(__m256 a);
__m256 Sleef_finz_exp10f8_u10avx2(__m256 a);
__m256 Sleef_finz_exp10f8_u10fma4(__m256 a);

__m512 Sleef_exp10f16_u10(__m512 a);
__m512 Sleef_exp10f16_u10avx512f(__m512 a);
__m512 Sleef_cinz_exp10f16_u10avx512fnofma(__m512 a);
__m512 Sleef_finz_exp10f16_u10avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_exp10f_u10 with the same accuracy specification.

Vectorized double precision base-e exponential functions minus 1 with 1.0 ULP error bound

#include <sleef.h>

__m128d Sleef_expm1d2_u10(__m128d a);
__m128d Sleef_expm1d2_u10sse2(__m128d a);
__m128d Sleef_expm1d2_u10sse4(__m128d a);
__m128d Sleef_expm1d2_u10avx2128(__m128d a);
__m128d Sleef_cinz_expm1d2_u10sse2(__m128d a);
__m128d Sleef_cinz_expm1d2_u10sse4(__m128d a);
__m128d Sleef_finz_expm1d2_u10avx2128(__m128d a);

__m256d Sleef_expm1d4_u10(__m256d a);
__m256d Sleef_expm1d4_u10avx(__m256d a);
__m256d Sleef_expm1d4_u10avx2(__m256d a);
__m256d Sleef_expm1d4_u10fma4(__m256d a);
__m256d Sleef_cinz_expm1d4_u10avx(__m256d a);
__m256d Sleef_finz_expm1d4_u10avx2(__m256d a);
__m256d Sleef_finz_expm1d4_u10fma4(__m256d a);

__m512d Sleef_expm1d8_u10(__m512d a);
__m512d Sleef_expm1d8_u10avx512f(__m512d a);
__m512d Sleef_cinz_expm1d8_u10avx512fnofma(__m512d a);
__m512d Sleef_finz_expm1d8_u10avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_expm1_u10 with the same accuracy specification.

Vectorized single precision base-e exponential functions minus 1 with 1.0 ULP error bound

#include <sleef.h>

__m128 Sleef_expm1f4_u10(__m128 a);
__m128 Sleef_expm1f4_u10sse2(__m128 a);
__m128 Sleef_expm1f4_u10sse4(__m128 a);
__m128 Sleef_expm1f4_u10avx2128(__m128 a);
__m128 Sleef_cinz_expm1f4_u10sse2(__m128 a);
__m128 Sleef_cinz_expm1f4_u10sse4(__m128 a);
__m128 Sleef_finz_expm1f4_u10avx2128(__m128 a);

__m256 Sleef_expm1f8_u10(__m256 a);
__m256 Sleef_expm1f8_u10avx(__m256 a);
__m256 Sleef_expm1f8_u10avx2(__m256 a);
__m256 Sleef_expm1f8_u10fma4(__m256 a);
__m256 Sleef_cinz_expm1f8_u10avx(__m256 a);
__m256 Sleef_finz_expm1f8_u10avx2(__m256 a);
__m256 Sleef_finz_expm1f8_u10fma4(__m256 a);

__m512 Sleef_expm1f16_u10(__m512 a);
__m512 Sleef_expm1f16_u10avx512f(__m512 a);
__m512 Sleef_cinz_expm1f16_u10avx512fnofma(__m512 a);
__m512 Sleef_finz_expm1f16_u10avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_expm1f_u10 with the same accuracy specification.

Vectorized double precision square root functions with 0.5001 ULP error bound

#include <sleef.h>

__m128d Sleef_sqrtd2_u05(__m128d a);
__m128d Sleef_sqrtd2_u05sse2(__m128d a);
__m128d Sleef_sqrtd2_u05sse4(__m128d a);
__m128d Sleef_sqrtd2_u05avx2128(__m128d a);
__m128d Sleef_cinz_sqrtd2_u05sse2(__m128d a);
__m128d Sleef_cinz_sqrtd2_u05sse4(__m128d a);
__m128d Sleef_finz_sqrtd2_u05avx2128(__m128d a);

__m256d Sleef_sqrtd4_u05(__m256d a);
__m256d Sleef_sqrtd4_u05avx(__m256d a);
__m256d Sleef_sqrtd4_u05avx2(__m256d a);
__m256d Sleef_sqrtd4_u05fma4(__m256d a);
__m256d Sleef_cinz_sqrtd4_u05avx(__m256d a);
__m256d Sleef_finz_sqrtd4_u05avx2(__m256d a);
__m256d Sleef_finz_sqrtd4_u05fma4(__m256d a);

__m512d Sleef_sqrtd8_u05(__m512d a);
__m512d Sleef_sqrtd8_u05avx512f(__m512d a);
__m512d Sleef_cinz_sqrtd8_u05avx512fnofma(__m512d a);
__m512d Sleef_finz_sqrtd8_u05avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_sqrt_u05 with the same accuracy specification.

Vectorized single precision square root functions with 0.5001 ULP error bound

#include <sleef.h>

__m128 Sleef_sqrtf4_u05(__m128 a);
__m128 Sleef_sqrtf4_u05sse2(__m128 a);
__m128 Sleef_sqrtf4_u05sse4(__m128 a);
__m128 Sleef_sqrtf4_u05avx2128(__m128 a);
__m128 Sleef_cinz_sqrtf4_u05sse2(__m128 a);
__m128 Sleef_cinz_sqrtf4_u05sse4(__m128 a);
__m128 Sleef_finz_sqrtf4_u05avx2128(__m128 a);

__m256 Sleef_sqrtf8_u05(__m256 a);
__m256 Sleef_sqrtf8_u05avx(__m256 a);
__m256 Sleef_sqrtf8_u05avx2(__m256 a);
__m256 Sleef_sqrtf8_u05fma4(__m256 a);
__m256 Sleef_cinz_sqrtf8_u05avx(__m256 a);
__m256 Sleef_finz_sqrtf8_u05avx2(__m256 a);
__m256 Sleef_finz_sqrtf8_u05fma4(__m256 a);

__m512 Sleef_sqrtf16_u05(__m512 a);
__m512 Sleef_sqrtf16_u05avx512f(__m512 a);
__m512 Sleef_cinz_sqrtf16_u05avx512fnofma(__m512 a);
__m512 Sleef_finz_sqrtf16_u05avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_sqrtf_u05 with the same accuracy specification.

Vectorized double precision square root functions with 3.5 ULP error bound

#include <sleef.h>

__m128d Sleef_sqrtd2_u35(__m128d a);
__m128d Sleef_sqrtd2_u35sse2(__m128d a);
__m128d Sleef_sqrtd2_u35sse4(__m128d a);
__m128d Sleef_sqrtd2_u35avx2128(__m128d a);
__m128d Sleef_cinz_sqrtd2_u35sse2(__m128d a);
__m128d Sleef_cinz_sqrtd2_u35sse4(__m128d a);
__m128d Sleef_finz_sqrtd2_u35avx2128(__m128d a);

__m256d Sleef_sqrtd4_u35(__m256d a);
__m256d Sleef_sqrtd4_u35avx(__m256d a);
__m256d Sleef_sqrtd4_u35avx2(__m256d a);
__m256d Sleef_sqrtd4_u35fma4(__m256d a);
__m256d Sleef_cinz_sqrtd4_u35avx(__m256d a);
__m256d Sleef_finz_sqrtd4_u35avx2(__m256d a);
__m256d Sleef_finz_sqrtd4_u35fma4(__m256d a);

__m512d Sleef_sqrtd8_u35(__m512d a);
__m512d Sleef_sqrtd8_u35avx512f(__m512d a);
__m512d Sleef_cinz_sqrtd8_u35avx512fnofma(__m512d a);
__m512d Sleef_finz_sqrtd8_u35avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_sqrt_u35 with the same accuracy specification.

Vectorized single precision square root functions with 3.5 ULP error bound

#include <sleef.h>

__m128 Sleef_sqrtf4_u35(__m128 a);
__m128 Sleef_sqrtf4_u35sse2(__m128 a);
__m128 Sleef_sqrtf4_u35sse4(__m128 a);
__m128 Sleef_sqrtf4_u35avx2128(__m128 a);
__m128 Sleef_cinz_sqrtf4_u35sse2(__m128 a);
__m128 Sleef_cinz_sqrtf4_u35sse4(__m128 a);
__m128 Sleef_finz_sqrtf4_u35avx2128(__m128 a);

__m256 Sleef_sqrtf8_u35(__m256 a);
__m256 Sleef_sqrtf8_u35avx(__m256 a);
__m256 Sleef_sqrtf8_u35avx2(__m256 a);
__m256 Sleef_sqrtf8_u35fma4(__m256 a);
__m256 Sleef_cinz_sqrtf8_u35avx(__m256 a);
__m256 Sleef_finz_sqrtf8_u35avx2(__m256 a);
__m256 Sleef_finz_sqrtf8_u35fma4(__m256 a);

__m512 Sleef_sqrtf16_u35(__m512 a);
__m512 Sleef_sqrtf16_u35avx512f(__m512 a);
__m512 Sleef_cinz_sqrtf16_u35avx512fnofma(__m512 a);
__m512 Sleef_finz_sqrtf16_u35avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_sqrtf_u35 with the same accuracy specification.

Vectorized double precision cubic root functions with 1.0 ULP error bound

#include <sleef.h>

__m128d Sleef_cbrtd2_u10(__m128d a);
__m128d Sleef_cbrtd2_u10sse2(__m128d a);
__m128d Sleef_cbrtd2_u10sse4(__m128d a);
__m128d Sleef_cbrtd2_u10avx2128(__m128d a);
__m128d Sleef_cinz_cbrtd2_u10sse2(__m128d a);
__m128d Sleef_cinz_cbrtd2_u10sse4(__m128d a);
__m128d Sleef_finz_cbrtd2_u10avx2128(__m128d a);

__m256d Sleef_cbrtd4_u10(__m256d a);
__m256d Sleef_cbrtd4_u10avx(__m256d a);
__m256d Sleef_cbrtd4_u10avx2(__m256d a);
__m256d Sleef_cbrtd4_u10fma4(__m256d a);
__m256d Sleef_cinz_cbrtd4_u10avx(__m256d a);
__m256d Sleef_finz_cbrtd4_u10avx2(__m256d a);
__m256d Sleef_finz_cbrtd4_u10fma4(__m256d a);

__m512d Sleef_cbrtd8_u10(__m512d a);
__m512d Sleef_cbrtd8_u10avx512f(__m512d a);
__m512d Sleef_cinz_cbrtd8_u10avx512fnofma(__m512d a);
__m512d Sleef_finz_cbrtd8_u10avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_cbrt_u10 with the same accuracy specification.

Vectorized single precision cubic root functions with 1.0 ULP error bound

#include <sleef.h>

__m128 Sleef_cbrtf4_u10(__m128 a);
__m128 Sleef_cbrtf4_u10sse2(__m128 a);
__m128 Sleef_cbrtf4_u10sse4(__m128 a);
__m128 Sleef_cbrtf4_u10avx2128(__m128 a);
__m128 Sleef_cinz_cbrtf4_u10sse2(__m128 a);
__m128 Sleef_cinz_cbrtf4_u10sse4(__m128 a);
__m128 Sleef_finz_cbrtf4_u10avx2128(__m128 a);

__m256 Sleef_cbrtf8_u10(__m256 a);
__m256 Sleef_cbrtf8_u10avx(__m256 a);
__m256 Sleef_cbrtf8_u10avx2(__m256 a);
__m256 Sleef_cbrtf8_u10fma4(__m256 a);
__m256 Sleef_cinz_cbrtf8_u10avx(__m256 a);
__m256 Sleef_finz_cbrtf8_u10avx2(__m256 a);
__m256 Sleef_finz_cbrtf8_u10fma4(__m256 a);

__m512 Sleef_cbrtf16_u10(__m512 a);
__m512 Sleef_cbrtf16_u10avx512f(__m512 a);
__m512 Sleef_cinz_cbrtf16_u10avx512fnofma(__m512 a);
__m512 Sleef_finz_cbrtf16_u10avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_cbrtf_u10 with the same accuracy specification.

Vectorized double precision cubic root functions with 3.5 ULP error bound

#include <sleef.h>

__m128d Sleef_cbrtd2_u35(__m128d a);
__m128d Sleef_cbrtd2_u35sse2(__m128d a);
__m128d Sleef_cbrtd2_u35sse4(__m128d a);
__m128d Sleef_cbrtd2_u35avx2128(__m128d a);
__m128d Sleef_cinz_cbrtd2_u35sse2(__m128d a);
__m128d Sleef_cinz_cbrtd2_u35sse4(__m128d a);
__m128d Sleef_finz_cbrtd2_u35avx2128(__m128d a);

__m256d Sleef_cbrtd4_u35(__m256d a);
__m256d Sleef_cbrtd4_u35avx(__m256d a);
__m256d Sleef_cbrtd4_u35avx2(__m256d a);
__m256d Sleef_cbrtd4_u35fma4(__m256d a);
__m256d Sleef_cinz_cbrtd4_u35avx(__m256d a);
__m256d Sleef_finz_cbrtd4_u35avx2(__m256d a);
__m256d Sleef_finz_cbrtd4_u35fma4(__m256d a);

__m512d Sleef_cbrtd8_u35(__m512d a);
__m512d Sleef_cbrtd8_u35avx512f(__m512d a);
__m512d Sleef_cinz_cbrtd8_u35avx512fnofma(__m512d a);
__m512d Sleef_finz_cbrtd8_u35avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_cbrt_u35 with the same accuracy specification.

Vectorized single precision cubic root functions with 3.5 ULP error bound

#include <sleef.h>

__m128 Sleef_cbrtf4_u35(__m128 a);
__m128 Sleef_cbrtf4_u35sse2(__m128 a);
__m128 Sleef_cbrtf4_u35sse4(__m128 a);
__m128 Sleef_cbrtf4_u35avx2128(__m128 a);
__m128 Sleef_cinz_cbrtf4_u35sse2(__m128 a);
__m128 Sleef_cinz_cbrtf4_u35sse4(__m128 a);
__m128 Sleef_finz_cbrtf4_u35avx2128(__m128 a);

__m256 Sleef_cbrtf8_u35(__m256 a);
__m256 Sleef_cbrtf8_u35avx(__m256 a);
__m256 Sleef_cbrtf8_u35avx2(__m256 a);
__m256 Sleef_cbrtf8_u35fma4(__m256 a);
__m256 Sleef_cinz_cbrtf8_u35avx(__m256 a);
__m256 Sleef_finz_cbrtf8_u35avx2(__m256 a);
__m256 Sleef_finz_cbrtf8_u35fma4(__m256 a);

__m512 Sleef_cbrtf16_u35(__m512 a);
__m512 Sleef_cbrtf16_u35avx512f(__m512 a);
__m512 Sleef_cinz_cbrtf16_u35avx512fnofma(__m512 a);
__m512 Sleef_finz_cbrtf16_u35avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_cbrtf_u35 with the same accuracy specification.

Vectorized double precision 2D Euclidian distance functions with 0.5 ULP error bound

#include <sleef.h>

__m128d Sleef_hypotd2_u05(__m128d a, __m128d b);
__m128d Sleef_hypotd2_u05sse2(__m128d a, __m128d b);
__m128d Sleef_hypotd2_u05sse4(__m128d a, __m128d b);
__m128d Sleef_hypotd2_u05avx2128(__m128d a, __m128d b);
__m128d Sleef_cinz_hypotd2_u05sse2(__m128d a, __m128d b);
__m128d Sleef_cinz_hypotd2_u05sse4(__m128d a, __m128d b);
__m128d Sleef_finz_hypotd2_u05avx2128(__m128d a, __m128d b);

__m256d Sleef_hypotd4_u05(__m256d a, __m256d b);
__m256d Sleef_hypotd4_u05avx(__m256d a, __m256d b);
__m256d Sleef_hypotd4_u05avx2(__m256d a, __m256d b);
__m256d Sleef_hypotd4_u05fma4(__m256d a, __m256d b);
__m256d Sleef_cinz_hypotd4_u05avx(__m256d a, __m256d b);
__m256d Sleef_finz_hypotd4_u05avx2(__m256d a, __m256d b);
__m256d Sleef_finz_hypotd4_u05fma4(__m256d a, __m256d b);

__m512d Sleef_hypotd8_u05(__m512d a, __m512d b);
__m512d Sleef_hypotd8_u05avx512f(__m512d a, __m512d b);
__m512d Sleef_cinz_hypotd8_u05avx512fnofma(__m512d a, __m512d b);
__m512d Sleef_finz_hypotd8_u05avx512f(__m512d a, __m512d b);

Link with -lsleef.

These are the vectorized functions of Sleef_hypot_u05 with the same accuracy specification.

Vectorized single precision 2D Euclidian distance functions with 0.5 ULP error bound

#include <sleef.h>

__m128 Sleef_hypotf4_u05(__m128 a, __m128 b);
__m128 Sleef_hypotf4_u05sse2(__m128 a, __m128 b);
__m128 Sleef_hypotf4_u05sse4(__m128 a, __m128 b);
__m128 Sleef_hypotf4_u05avx2128(__m128 a, __m128 b);
__m128 Sleef_cinz_hypotf4_u05sse2(__m128 a, __m128 b);
__m128 Sleef_cinz_hypotf4_u05sse4(__m128 a, __m128 b);
__m128 Sleef_finz_hypotf4_u05avx2128(__m128 a, __m128 b);

__m256 Sleef_hypotf8_u05(__m256 a, __m256 b);
__m256 Sleef_hypotf8_u05avx(__m256 a, __m256 b);
__m256 Sleef_hypotf8_u05avx2(__m256 a, __m256 b);
__m256 Sleef_hypotf8_u05fma4(__m256 a, __m256 b);
__m256 Sleef_cinz_hypotf8_u05avx(__m256 a, __m256 b);
__m256 Sleef_finz_hypotf8_u05avx2(__m256 a, __m256 b);
__m256 Sleef_finz_hypotf8_u05fma4(__m256 a, __m256 b);

__m512 Sleef_hypotf16_u05(__m512 a, __m512 b);
__m512 Sleef_hypotf16_u05avx512f(__m512 a, __m512 b);
__m512 Sleef_cinz_hypotf16_u05avx512fnofma(__m512 a, __m512 b);
__m512 Sleef_finz_hypotf16_u05avx512f(__m512 a, __m512 b);

Link with -lsleef.

These are the vectorized functions of Sleef_hypotf_u05 with the same accuracy specification.

Vectorized double precision 2D Euclidian distance functions with 3.5 ULP error bound

#include <sleef.h>

__m128d Sleef_hypotd2_u35(__m128d a, __m128d b);
__m128d Sleef_hypotd2_u35sse2(__m128d a, __m128d b);
__m128d Sleef_hypotd2_u35sse4(__m128d a, __m128d b);
__m128d Sleef_hypotd2_u35avx2128(__m128d a, __m128d b);
__m128d Sleef_cinz_hypotd2_u35sse2(__m128d a, __m128d b);
__m128d Sleef_cinz_hypotd2_u35sse4(__m128d a, __m128d b);
__m128d Sleef_finz_hypotd2_u35avx2128(__m128d a, __m128d b);

__m256d Sleef_hypotd4_u35(__m256d a, __m256d b);
__m256d Sleef_hypotd4_u35avx(__m256d a, __m256d b);
__m256d Sleef_hypotd4_u35avx2(__m256d a, __m256d b);
__m256d Sleef_hypotd4_u35fma4(__m256d a, __m256d b);
__m256d Sleef_cinz_hypotd4_u35avx(__m256d a, __m256d b);
__m256d Sleef_finz_hypotd4_u35avx2(__m256d a, __m256d b);
__m256d Sleef_finz_hypotd4_u35fma4(__m256d a, __m256d b);

__m512d Sleef_hypotd8_u35(__m512d a, __m512d b);
__m512d Sleef_hypotd8_u35avx512f(__m512d a, __m512d b);
__m512d Sleef_cinz_hypotd8_u35avx512fnofma(__m512d a, __m512d b);
__m512d Sleef_finz_hypotd8_u35avx512f(__m512d a, __m512d b);

Link with -lsleef.

These are the vectorized functions of Sleef_hypot_u35 with the same accuracy specification.

Vectorized single precision 2D Euclidian distance functions with 3.5 ULP error bound

#include <sleef.h>

__m128 Sleef_hypotf4_u35(__m128 a, __m128 b);
__m128 Sleef_hypotf4_u35sse2(__m128 a, __m128 b);
__m128 Sleef_hypotf4_u35sse4(__m128 a, __m128 b);
__m128 Sleef_hypotf4_u35avx2128(__m128 a, __m128 b);
__m128 Sleef_cinz_hypotf4_u35sse2(__m128 a, __m128 b);
__m128 Sleef_cinz_hypotf4_u35sse4(__m128 a, __m128 b);
__m128 Sleef_finz_hypotf4_u35avx2128(__m128 a, __m128 b);

__m256 Sleef_hypotf8_u35(__m256 a, __m256 b);
__m256 Sleef_hypotf8_u35avx(__m256 a, __m256 b);
__m256 Sleef_hypotf8_u35avx2(__m256 a, __m256 b);
__m256 Sleef_hypotf8_u35fma4(__m256 a, __m256 b);
__m256 Sleef_cinz_hypotf8_u35avx(__m256 a, __m256 b);
__m256 Sleef_finz_hypotf8_u35avx2(__m256 a, __m256 b);
__m256 Sleef_finz_hypotf8_u35fma4(__m256 a, __m256 b);

__m512 Sleef_hypotf16_u35(__m512 a, __m512 b);
__m512 Sleef_hypotf16_u35avx512f(__m512 a, __m512 b);
__m512 Sleef_cinz_hypotf16_u35avx512fnofma(__m512 a, __m512 b);
__m512 Sleef_finz_hypotf16_u35avx512f(__m512 a, __m512 b);

Link with -lsleef.

These are the vectorized functions of Sleef_hypotf_u35 with the same accuracy specification.

Inverse Trigonometric Functions

Vectorized double precision arc sine functions with 1.0 ULP error bound

#include <sleef.h>

__m128d Sleef_asind2_u10(__m128d a);
__m128d Sleef_asind2_u10sse2(__m128d a);
__m128d Sleef_asind2_u10sse4(__m128d a);
__m128d Sleef_asind2_u10avx2128(__m128d a);
__m128d Sleef_cinz_asind2_u10sse2(__m128d a);
__m128d Sleef_cinz_asind2_u10sse4(__m128d a);
__m128d Sleef_finz_asind2_u10avx2128(__m128d a);

__m256d Sleef_asind4_u10(__m256d a);
__m256d Sleef_asind4_u10avx(__m256d a);
__m256d Sleef_asind4_u10avx2(__m256d a);
__m256d Sleef_asind4_u10fma4(__m256d a);
__m256d Sleef_cinz_asind4_u10avx(__m256d a);
__m256d Sleef_finz_asind4_u10avx2(__m256d a);
__m256d Sleef_finz_asind4_u10fma4(__m256d a);

__m512d Sleef_asind8_u10(__m512d a);
__m512d Sleef_asind8_u10avx512f(__m512d a);
__m512d Sleef_cinz_asind8_u10avx512fnofma(__m512d a);
__m512d Sleef_finz_asind8_u10avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_asin_u10 with the same accuracy specification.

Vectorized single precision arc sine functions with 3.5 ULP error bound

#include <sleef.h>

__m128 Sleef_asinf4_u10(__m128 a);
__m128 Sleef_asinf4_u10sse2(__m128 a);
__m128 Sleef_asinf4_u10sse4(__m128 a);
__m128 Sleef_asinf4_u10avx2128(__m128 a);
__m128 Sleef_cinz_asinf4_u10sse2(__m128 a);
__m128 Sleef_cinz_asinf4_u10sse4(__m128 a);
__m128 Sleef_finz_asinf4_u10avx2128(__m128 a);

__m256 Sleef_asinf8_u10(__m256 a);
__m256 Sleef_asinf8_u10avx(__m256 a);
__m256 Sleef_asinf8_u10avx2(__m256 a);
__m256 Sleef_asinf8_u10fma4(__m256 a);
__m256 Sleef_cinz_asinf8_u10avx(__m256 a);
__m256 Sleef_finz_asinf8_u10avx2(__m256 a);
__m256 Sleef_finz_asinf8_u10fma4(__m256 a);

__m512 Sleef_asinf16_u10(__m512 a);
__m512 Sleef_asinf16_u10avx512f(__m512 a);
__m512 Sleef_cinz_asinf16_u10avx512fnofma(__m512 a);
__m512 Sleef_finz_asinf16_u10avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_asinf_u10 with the same accuracy specification.

Vectorized double precision arc sine functions with 3.5 ULP error bound

#include <sleef.h>

__m128d Sleef_asind2_u35(__m128d a);
__m128d Sleef_asind2_u35sse2(__m128d a);
__m128d Sleef_asind2_u35sse4(__m128d a);
__m128d Sleef_asind2_u35avx2128(__m128d a);
__m128d Sleef_cinz_asind2_u35sse2(__m128d a);
__m128d Sleef_cinz_asind2_u35sse4(__m128d a);
__m128d Sleef_finz_asind2_u35avx2128(__m128d a);

__m256d Sleef_asind4_u35(__m256d a);
__m256d Sleef_asind4_u35avx(__m256d a);
__m256d Sleef_asind4_u35avx2(__m256d a);
__m256d Sleef_asind4_u35fma4(__m256d a);
__m256d Sleef_cinz_asind4_u35avx(__m256d a);
__m256d Sleef_finz_asind4_u35avx2(__m256d a);
__m256d Sleef_finz_asind4_u35fma4(__m256d a);

__m512d Sleef_asind8_u35(__m512d a);
__m512d Sleef_asind8_u35avx512f(__m512d a);
__m512d Sleef_cinz_asind8_u35avx512fnofma(__m512d a);
__m512d Sleef_finz_asind8_u35avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_asin_u35 with the same accuracy specification.

Vectorized single precision arc sine functions with 3.5 ULP error bound

#include <sleef.h>

__m128 Sleef_asinf4_u35(__m128 a);
__m128 Sleef_asinf4_u35sse2(__m128 a);
__m128 Sleef_asinf4_u35sse4(__m128 a);
__m128 Sleef_asinf4_u35avx2128(__m128 a);
__m128 Sleef_cinz_asinf4_u35sse2(__m128 a);
__m128 Sleef_cinz_asinf4_u35sse4(__m128 a);
__m128 Sleef_finz_asinf4_u35avx2128(__m128 a);

__m256 Sleef_asinf8_u35(__m256 a);
__m256 Sleef_asinf8_u35avx(__m256 a);
__m256 Sleef_asinf8_u35avx2(__m256 a);
__m256 Sleef_asinf8_u35fma4(__m256 a);
__m256 Sleef_cinz_asinf8_u35avx(__m256 a);
__m256 Sleef_finz_asinf8_u35avx2(__m256 a);
__m256 Sleef_finz_asinf8_u35fma4(__m256 a);

__m512 Sleef_asinf16_u35(__m512 a);
__m512 Sleef_asinf16_u35avx512f(__m512 a);
__m512 Sleef_cinz_asinf16_u35avx512fnofma(__m512 a);
__m512 Sleef_finz_asinf16_u35avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_asinf_u35 with the same accuracy specification.

Vectorized double precision arc cosine functions with 1.0 ULP error bound

#include <sleef.h>

__m128d Sleef_acosd2_u10(__m128d a);
__m128d Sleef_acosd2_u10sse2(__m128d a);
__m128d Sleef_acosd2_u10sse4(__m128d a);
__m128d Sleef_acosd2_u10avx2128(__m128d a);
__m128d Sleef_cinz_acosd2_u10sse2(__m128d a);
__m128d Sleef_cinz_acosd2_u10sse4(__m128d a);
__m128d Sleef_finz_acosd2_u10avx2128(__m128d a);

__m256d Sleef_acosd4_u10(__m256d a);
__m256d Sleef_acosd4_u10avx(__m256d a);
__m256d Sleef_acosd4_u10avx2(__m256d a);
__m256d Sleef_acosd4_u10fma4(__m256d a);
__m256d Sleef_cinz_acosd4_u10avx(__m256d a);
__m256d Sleef_finz_acosd4_u10avx2(__m256d a);
__m256d Sleef_finz_acosd4_u10fma4(__m256d a);

__m512d Sleef_acosd8_u10(__m512d a);
__m512d Sleef_acosd8_u10avx512f(__m512d a);
__m512d Sleef_cinz_acosd8_u10avx512fnofma(__m512d a);
__m512d Sleef_finz_acosd8_u10avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_acos_u10 with the same accuracy specification.

Vectorized single precision arc cosine functions with 1.0 ULP error bound

#include <sleef.h>

__m128 Sleef_acosf4_u10(__m128 a);
__m128 Sleef_acosf4_u10sse2(__m128 a);
__m128 Sleef_acosf4_u10sse4(__m128 a);
__m128 Sleef_acosf4_u10avx2128(__m128 a);
__m128 Sleef_cinz_acosf4_u10sse2(__m128 a);
__m128 Sleef_cinz_acosf4_u10sse4(__m128 a);
__m128 Sleef_finz_acosf4_u10avx2128(__m128 a);

__m256 Sleef_acosf8_u10(__m256 a);
__m256 Sleef_acosf8_u10avx(__m256 a);
__m256 Sleef_acosf8_u10avx2(__m256 a);
__m256 Sleef_acosf8_u10fma4(__m256 a);
__m256 Sleef_cinz_acosf8_u10avx(__m256 a);
__m256 Sleef_finz_acosf8_u10avx2(__m256 a);
__m256 Sleef_finz_acosf8_u10fma4(__m256 a);

__m512 Sleef_acosf16_u10(__m512 a);
__m512 Sleef_acosf16_u10avx512f(__m512 a);
__m512 Sleef_cinz_acosf16_u10avx512fnofma(__m512 a);
__m512 Sleef_finz_acosf16_u10avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_acosf_u10 with the same accuracy specification.

Vectorized double precision arc cosine functions with 3.5 ULP error bound

#include <sleef.h>

__m128d Sleef_acosd2_u35(__m128d a);
__m128d Sleef_acosd2_u35sse2(__m128d a);
__m128d Sleef_acosd2_u35sse4(__m128d a);
__m128d Sleef_acosd2_u35avx2128(__m128d a);
__m128d Sleef_cinz_acosd2_u35sse2(__m128d a);
__m128d Sleef_cinz_acosd2_u35sse4(__m128d a);
__m128d Sleef_finz_acosd2_u35avx2128(__m128d a);

__m256d Sleef_acosd4_u35(__m256d a);
__m256d Sleef_acosd4_u35avx(__m256d a);
__m256d Sleef_acosd4_u35avx2(__m256d a);
__m256d Sleef_acosd4_u35fma4(__m256d a);
__m256d Sleef_cinz_acosd4_u35avx(__m256d a);
__m256d Sleef_finz_acosd4_u35avx2(__m256d a);
__m256d Sleef_finz_acosd4_u35fma4(__m256d a);

__m512d Sleef_acosd8_u35(__m512d a);
__m512d Sleef_acosd8_u35avx512f(__m512d a);
__m512d Sleef_cinz_acosd8_u35avx512fnofma(__m512d a);
__m512d Sleef_finz_acosd8_u35avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_acos_u35 with the same accuracy specification.

Vectorized single precision arc cosine functions with 3.5 ULP error bound

#include <sleef.h>

__m128 Sleef_acosf4_u35(__m128 a);
__m128 Sleef_acosf4_u35sse2(__m128 a);
__m128 Sleef_acosf4_u35sse4(__m128 a);
__m128 Sleef_acosf4_u35avx2128(__m128 a);
__m128 Sleef_cinz_acosf4_u35sse2(__m128 a);
__m128 Sleef_cinz_acosf4_u35sse4(__m128 a);
__m128 Sleef_finz_acosf4_u35avx2128(__m128 a);

__m256 Sleef_acosf8_u35(__m256 a);
__m256 Sleef_acosf8_u35avx(__m256 a);
__m256 Sleef_acosf8_u35avx2(__m256 a);
__m256 Sleef_acosf8_u35fma4(__m256 a);
__m256 Sleef_cinz_acosf8_u35avx(__m256 a);
__m256 Sleef_finz_acosf8_u35avx2(__m256 a);
__m256 Sleef_finz_acosf8_u35fma4(__m256 a);

__m512 Sleef_acosf16_u35(__m512 a);
__m512 Sleef_acosf16_u35avx512f(__m512 a);
__m512 Sleef_cinz_acosf16_u35avx512fnofma(__m512 a);
__m512 Sleef_finz_acosf16_u35avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_acosf_u35 with the same accuracy specification.

Vectorized double precision arc tangent functions with 1.0 ULP error bound

#include <sleef.h>

__m128d Sleef_atand2_u10(__m128d a);
__m128d Sleef_atand2_u10sse2(__m128d a);
__m128d Sleef_atand2_u10sse4(__m128d a);
__m128d Sleef_atand2_u10avx2128(__m128d a);
__m128d Sleef_cinz_atand2_u10sse2(__m128d a);
__m128d Sleef_cinz_atand2_u10sse4(__m128d a);
__m128d Sleef_finz_atand2_u10avx2128(__m128d a);

__m256d Sleef_atand4_u10(__m256d a);
__m256d Sleef_atand4_u10avx(__m256d a);
__m256d Sleef_atand4_u10avx2(__m256d a);
__m256d Sleef_atand4_u10fma4(__m256d a);
__m256d Sleef_cinz_atand4_u10avx(__m256d a);
__m256d Sleef_finz_atand4_u10avx2(__m256d a);
__m256d Sleef_finz_atand4_u10fma4(__m256d a);

__m512d Sleef_atand8_u10(__m512d a);
__m512d Sleef_atand8_u10avx512f(__m512d a);
__m512d Sleef_cinz_atand8_u10avx512fnofma(__m512d a);
__m512d Sleef_finz_atand8_u10avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_atan_u10 with the same accuracy specification.

Vectorized single precision arc tangent functions with 1.0 ULP error bound

#include <sleef.h>

__m128 Sleef_atanf4_u10(__m128 a);
__m128 Sleef_atanf4_u10sse2(__m128 a);
__m128 Sleef_atanf4_u10sse4(__m128 a);
__m128 Sleef_atanf4_u10avx2128(__m128 a);
__m128 Sleef_cinz_atanf4_u10sse2(__m128 a);
__m128 Sleef_cinz_atanf4_u10sse4(__m128 a);
__m128 Sleef_finz_atanf4_u10avx2128(__m128 a);

__m256 Sleef_atanf8_u10(__m256 a);
__m256 Sleef_atanf8_u10avx(__m256 a);
__m256 Sleef_atanf8_u10avx2(__m256 a);
__m256 Sleef_atanf8_u10fma4(__m256 a);
__m256 Sleef_cinz_atanf8_u10avx(__m256 a);
__m256 Sleef_finz_atanf8_u10avx2(__m256 a);
__m256 Sleef_finz_atanf8_u10fma4(__m256 a);

__m512 Sleef_atanf16_u10(__m512 a);
__m512 Sleef_atanf16_u10avx512f(__m512 a);
__m512 Sleef_cinz_atanf16_u10avx512fnofma(__m512 a);
__m512 Sleef_finz_atanf16_u10avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_atanf_u10 with the same accuracy specification.

Vectorized double precision arc tangent functions with 3.5 ULP error bound

#include <sleef.h>

__m128d Sleef_atand2_u35(__m128d a);
__m128d Sleef_atand2_u35sse2(__m128d a);
__m128d Sleef_atand2_u35sse4(__m128d a);
__m128d Sleef_atand2_u35avx2128(__m128d a);
__m128d Sleef_cinz_atand2_u35sse2(__m128d a);
__m128d Sleef_cinz_atand2_u35sse4(__m128d a);
__m128d Sleef_finz_atand2_u35avx2128(__m128d a);

__m256d Sleef_atand4_u35(__m256d a);
__m256d Sleef_atand4_u35avx(__m256d a);
__m256d Sleef_atand4_u35avx2(__m256d a);
__m256d Sleef_atand4_u35fma4(__m256d a);
__m256d Sleef_cinz_atand4_u35avx(__m256d a);
__m256d Sleef_finz_atand4_u35avx2(__m256d a);
__m256d Sleef_finz_atand4_u35fma4(__m256d a);

__m512d Sleef_atand8_u35(__m512d a);
__m512d Sleef_atand8_u35avx512f(__m512d a);
__m512d Sleef_cinz_atand8_u35avx512fnofma(__m512d a);
__m512d Sleef_finz_atand8_u35avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_atan_u35 with the same accuracy specification.

Vectorized single precision arc tangent functions with 3.5 ULP error bound

#include <sleef.h>

__m128 Sleef_atanf4_u35(__m128 a);
__m128 Sleef_atanf4_u35sse2(__m128 a);
__m128 Sleef_atanf4_u35sse4(__m128 a);
__m128 Sleef_atanf4_u35avx2128(__m128 a);
__m128 Sleef_cinz_atanf4_u35sse2(__m128 a);
__m128 Sleef_cinz_atanf4_u35sse4(__m128 a);
__m128 Sleef_finz_atanf4_u35avx2128(__m128 a);

__m256 Sleef_atanf8_u35(__m256 a);
__m256 Sleef_atanf8_u35avx(__m256 a);
__m256 Sleef_atanf8_u35avx2(__m256 a);
__m256 Sleef_atanf8_u35fma4(__m256 a);
__m256 Sleef_cinz_atanf8_u35avx(__m256 a);
__m256 Sleef_finz_atanf8_u35avx2(__m256 a);
__m256 Sleef_finz_atanf8_u35fma4(__m256 a);

__m512 Sleef_atanf16_u35(__m512 a);
__m512 Sleef_atanf16_u35avx512f(__m512 a);
__m512 Sleef_cinz_atanf16_u35avx512fnofma(__m512 a);
__m512 Sleef_finz_atanf16_u35avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_atanf_u35 with the same accuracy specification.

Vectorized double precision arc tangent functions of two variables with 1.0 ULP error bound

#include <sleef.h>

__m128d Sleef_atan2d2_u10(__m128d a, __m128d b);
__m128d Sleef_atan2d2_u10sse2(__m128d a, __m128d b);
__m128d Sleef_atan2d2_u10sse4(__m128d a, __m128d b);
__m128d Sleef_atan2d2_u10avx2128(__m128d a, __m128d b);
__m128d Sleef_cinz_atan2d2_u10sse2(__m128d a, __m128d b);
__m128d Sleef_cinz_atan2d2_u10sse4(__m128d a, __m128d b);
__m128d Sleef_finz_atan2d2_u10avx2128(__m128d a, __m128d b);

__m256d Sleef_atan2d4_u10(__m256d a, __m256d b);
__m256d Sleef_atan2d4_u10avx(__m256d a, __m256d b);
__m256d Sleef_atan2d4_u10avx2(__m256d a, __m256d b);
__m256d Sleef_atan2d4_u10fma4(__m256d a, __m256d b);
__m256d Sleef_cinz_atan2d4_u10avx(__m256d a, __m256d b);
__m256d Sleef_finz_atan2d4_u10avx2(__m256d a, __m256d b);
__m256d Sleef_finz_atan2d4_u10fma4(__m256d a, __m256d b);

__m512d Sleef_atan2d8_u10(__m512d a, __m512d b);
__m512d Sleef_atan2d8_u10avx512f(__m512d a, __m512d b);
__m512d Sleef_cinz_atan2d8_u10avx512fnofma(__m512d a, __m512d b);
__m512d Sleef_finz_atan2d8_u10avx512f(__m512d a, __m512d b);

Link with -lsleef.

These are the vectorized functions of Sleef_atan2_u10 with the same accuracy specification.

Vectorized single precision arc tangent functions of two variables with 1.0 ULP error bound

#include <sleef.h>

__m128 Sleef_atan2f4_u10(__m128 a, __m128 b);
__m128 Sleef_atan2f4_u10sse2(__m128 a, __m128 b);
__m128 Sleef_atan2f4_u10sse4(__m128 a, __m128 b);
__m128 Sleef_atan2f4_u10avx2128(__m128 a, __m128 b);
__m128 Sleef_cinz_atan2f4_u10sse2(__m128 a, __m128 b);
__m128 Sleef_cinz_atan2f4_u10sse4(__m128 a, __m128 b);
__m128 Sleef_finz_atan2f4_u10avx2128(__m128 a, __m128 b);

__m256 Sleef_atan2f8_u10(__m256 a, __m256 b);
__m256 Sleef_atan2f8_u10avx(__m256 a, __m256 b);
__m256 Sleef_atan2f8_u10avx2(__m256 a, __m256 b);
__m256 Sleef_atan2f8_u10fma4(__m256 a, __m256 b);
__m256 Sleef_cinz_atan2f8_u10avx(__m256 a, __m256 b);
__m256 Sleef_finz_atan2f8_u10avx2(__m256 a, __m256 b);
__m256 Sleef_finz_atan2f8_u10fma4(__m256 a, __m256 b);

__m512 Sleef_atan2f16_u10(__m512 a, __m512 b);
__m512 Sleef_atan2f16_u10avx512f(__m512 a, __m512 b);
__m512 Sleef_cinz_atan2f16_u10avx512fnofma(__m512 a, __m512 b);
__m512 Sleef_finz_atan2f16_u10avx512f(__m512 a, __m512 b);

Link with -lsleef.

These are the vectorized functions of Sleef_atan2f_u10 with the same accuracy specification.

Vectorized double precision arc tangent functions of two variables with 3.5 ULP error bound

#include <sleef.h>

__m128d Sleef_atan2d2_u35(__m128d a, __m128d b);
__m128d Sleef_atan2d2_u35sse2(__m128d a, __m128d b);
__m128d Sleef_atan2d2_u35sse4(__m128d a, __m128d b);
__m128d Sleef_atan2d2_u35avx2128(__m128d a, __m128d b);
__m128d Sleef_cinz_atan2d2_u35sse2(__m128d a, __m128d b);
__m128d Sleef_cinz_atan2d2_u35sse4(__m128d a, __m128d b);
__m128d Sleef_finz_atan2d2_u35avx2128(__m128d a, __m128d b);

__m256d Sleef_atan2d4_u35(__m256d a, __m256d b);
__m256d Sleef_atan2d4_u35avx(__m256d a, __m256d b);
__m256d Sleef_atan2d4_u35avx2(__m256d a, __m256d b);
__m256d Sleef_atan2d4_u35fma4(__m256d a, __m256d b);
__m256d Sleef_cinz_atan2d4_u35avx(__m256d a, __m256d b);
__m256d Sleef_finz_atan2d4_u35avx2(__m256d a, __m256d b);
__m256d Sleef_finz_atan2d4_u35fma4(__m256d a, __m256d b);

__m512d Sleef_atan2d8_u35(__m512d a, __m512d b);
__m512d Sleef_atan2d8_u35avx512f(__m512d a, __m512d b);
__m512d Sleef_cinz_atan2d8_u35avx512fnofma(__m512d a, __m512d b);
__m512d Sleef_finz_atan2d8_u35avx512f(__m512d a, __m512d b);

Link with -lsleef.

These are the vectorized functions of Sleef_atan2_u35 with the same accuracy specification.

Vectorized single precision arc tangent functions of two variables with 3.5 ULP error bound

#include <sleef.h>

__m128 Sleef_atan2f4_u35(__m128 a, __m128 b);
__m128 Sleef_atan2f4_u35sse2(__m128 a, __m128 b);
__m128 Sleef_atan2f4_u35sse4(__m128 a, __m128 b);
__m128 Sleef_atan2f4_u35avx2128(__m128 a, __m128 b);
__m128 Sleef_cinz_atan2f4_u35sse2(__m128 a, __m128 b);
__m128 Sleef_cinz_atan2f4_u35sse4(__m128 a, __m128 b);
__m128 Sleef_finz_atan2f4_u35avx2128(__m128 a, __m128 b);

__m256 Sleef_atan2f8_u35(__m256 a, __m256 b);
__m256 Sleef_atan2f8_u35avx(__m256 a, __m256 b);
__m256 Sleef_atan2f8_u35avx2(__m256 a, __m256 b);
__m256 Sleef_atan2f8_u35fma4(__m256 a, __m256 b);
__m256 Sleef_cinz_atan2f8_u35avx(__m256 a, __m256 b);
__m256 Sleef_finz_atan2f8_u35avx2(__m256 a, __m256 b);
__m256 Sleef_finz_atan2f8_u35fma4(__m256 a, __m256 b);

__m512 Sleef_atan2f16_u35(__m512 a, __m512 b);
__m512 Sleef_atan2f16_u35avx512f(__m512 a, __m512 b);
__m512 Sleef_cinz_atan2f16_u35avx512fnofma(__m512 a, __m512 b);
__m512 Sleef_finz_atan2f16_u35avx512f(__m512 a, __m512 b);

Link with -lsleef.

These are the vectorized functions of Sleef_atan2f_u35 with the same accuracy specification.

Hyperbolic functions and inverse hyperbolic functions

Vectorized double precision hyperbolic sine functions

#include <sleef.h>

__m128d Sleef_sinhd2_u10(__m128d a);
__m128d Sleef_sinhd2_u10sse2(__m128d a);
__m128d Sleef_sinhd2_u10sse4(__m128d a);
__m128d Sleef_sinhd2_u10avx2128(__m128d a);
__m128d Sleef_cinz_sinhd2_u10sse2(__m128d a);
__m128d Sleef_cinz_sinhd2_u10sse4(__m128d a);
__m128d Sleef_finz_sinhd2_u10avx2128(__m128d a);

__m256d Sleef_sinhd4_u10(__m256d a);
__m256d Sleef_sinhd4_u10avx(__m256d a);
__m256d Sleef_sinhd4_u10avx2(__m256d a);
__m256d Sleef_sinhd4_u10fma4(__m256d a);
__m256d Sleef_cinz_sinhd4_u10avx(__m256d a);
__m256d Sleef_finz_sinhd4_u10avx2(__m256d a);
__m256d Sleef_finz_sinhd4_u10fma4(__m256d a);

__m512d Sleef_sinhd8_u10(__m512d a);
__m512d Sleef_sinhd8_u10avx512f(__m512d a);
__m512d Sleef_cinz_sinhd8_u10avx512fnofma(__m512d a);
__m512d Sleef_finz_sinhd8_u10avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_sinh_u10 with the same accuracy specification.

Vectorized single precision hyperbolic sine functions

#include <sleef.h>

__m128 Sleef_sinhf4_u10(__m128 a);
__m128 Sleef_sinhf4_u10sse2(__m128 a);
__m128 Sleef_sinhf4_u10sse4(__m128 a);
__m128 Sleef_sinhf4_u10avx2128(__m128 a);
__m128 Sleef_cinz_sinhf4_u10sse2(__m128 a);
__m128 Sleef_cinz_sinhf4_u10sse4(__m128 a);
__m128 Sleef_finz_sinhf4_u10avx2128(__m128 a);

__m256 Sleef_sinhf8_u10(__m256 a);
__m256 Sleef_sinhf8_u10avx(__m256 a);
__m256 Sleef_sinhf8_u10avx2(__m256 a);
__m256 Sleef_sinhf8_u10fma4(__m256 a);
__m256 Sleef_cinz_sinhf8_u10avx(__m256 a);
__m256 Sleef_finz_sinhf8_u10avx2(__m256 a);
__m256 Sleef_finz_sinhf8_u10fma4(__m256 a);

__m512 Sleef_sinhf16_u10(__m512 a);
__m512 Sleef_sinhf16_u10avx512f(__m512 a);
__m512 Sleef_cinz_sinhf16_u10avx512fnofma(__m512 a);
__m512 Sleef_finz_sinhf16_u10avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_sinhf_u10 with the same accuracy specification.

Vectorized double precision hyperbolic sine functions

#include <sleef.h>

__m128d Sleef_sinhd2_u35(__m128d a);
__m128d Sleef_sinhd2_u35sse2(__m128d a);
__m128d Sleef_sinhd2_u35sse4(__m128d a);
__m128d Sleef_sinhd2_u35avx2128(__m128d a);
__m128d Sleef_cinz_sinhd2_u35sse2(__m128d a);
__m128d Sleef_cinz_sinhd2_u35sse4(__m128d a);
__m128d Sleef_finz_sinhd2_u35avx2128(__m128d a);

__m256d Sleef_sinhd4_u35(__m256d a);
__m256d Sleef_sinhd4_u35avx(__m256d a);
__m256d Sleef_sinhd4_u35avx2(__m256d a);
__m256d Sleef_sinhd4_u35fma4(__m256d a);
__m256d Sleef_cinz_sinhd4_u35avx(__m256d a);
__m256d Sleef_finz_sinhd4_u35avx2(__m256d a);
__m256d Sleef_finz_sinhd4_u35fma4(__m256d a);

__m512d Sleef_sinhd8_u35(__m512d a);
__m512d Sleef_sinhd8_u35avx512f(__m512d a);
__m512d Sleef_cinz_sinhd8_u35avx512fnofma(__m512d a);
__m512d Sleef_finz_sinhd8_u35avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_sinh_u35 with the same accuracy specification.

Vectorized single precision hyperbolic sine functions

#include <sleef.h>

__m128 Sleef_sinhf4_u35(__m128 a);
__m128 Sleef_sinhf4_u35sse2(__m128 a);
__m128 Sleef_sinhf4_u35sse4(__m128 a);
__m128 Sleef_sinhf4_u35avx2128(__m128 a);
__m128 Sleef_cinz_sinhf4_u35sse2(__m128 a);
__m128 Sleef_cinz_sinhf4_u35sse4(__m128 a);
__m128 Sleef_finz_sinhf4_u35avx2128(__m128 a);

__m256 Sleef_sinhf8_u35(__m256 a);
__m256 Sleef_sinhf8_u35avx(__m256 a);
__m256 Sleef_sinhf8_u35avx2(__m256 a);
__m256 Sleef_sinhf8_u35fma4(__m256 a);
__m256 Sleef_cinz_sinhf8_u35avx(__m256 a);
__m256 Sleef_finz_sinhf8_u35avx2(__m256 a);
__m256 Sleef_finz_sinhf8_u35fma4(__m256 a);

__m512 Sleef_sinhf16_u35(__m512 a);
__m512 Sleef_sinhf16_u35avx512f(__m512 a);
__m512 Sleef_cinz_sinhf16_u35avx512fnofma(__m512 a);
__m512 Sleef_finz_sinhf16_u35avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_sinhf_u35 with the same accuracy specification.

Vectorized double precision hyperbolic cosine functions

#include <sleef.h>

__m128d Sleef_coshd2_u10(__m128d a);
__m128d Sleef_coshd2_u10sse2(__m128d a);
__m128d Sleef_coshd2_u10sse4(__m128d a);
__m128d Sleef_coshd2_u10avx2128(__m128d a);
__m128d Sleef_cinz_coshd2_u10sse2(__m128d a);
__m128d Sleef_cinz_coshd2_u10sse4(__m128d a);
__m128d Sleef_finz_coshd2_u10avx2128(__m128d a);

__m256d Sleef_coshd4_u10(__m256d a);
__m256d Sleef_coshd4_u10avx(__m256d a);
__m256d Sleef_coshd4_u10avx2(__m256d a);
__m256d Sleef_coshd4_u10fma4(__m256d a);
__m256d Sleef_cinz_coshd4_u10avx(__m256d a);
__m256d Sleef_finz_coshd4_u10avx2(__m256d a);
__m256d Sleef_finz_coshd4_u10fma4(__m256d a);

__m512d Sleef_coshd8_u10(__m512d a);
__m512d Sleef_coshd8_u10avx512f(__m512d a);
__m512d Sleef_cinz_coshd8_u10avx512fnofma(__m512d a);
__m512d Sleef_finz_coshd8_u10avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_cosh_u10 with the same accuracy specification.

Vectorized single precision hyperbolic cosine functions

#include <sleef.h>

__m128 Sleef_coshf4_u10(__m128 a);
__m128 Sleef_coshf4_u10sse2(__m128 a);
__m128 Sleef_coshf4_u10sse4(__m128 a);
__m128 Sleef_coshf4_u10avx2128(__m128 a);
__m128 Sleef_cinz_coshf4_u10sse2(__m128 a);
__m128 Sleef_cinz_coshf4_u10sse4(__m128 a);
__m128 Sleef_finz_coshf4_u10avx2128(__m128 a);

__m256 Sleef_coshf8_u10(__m256 a);
__m256 Sleef_coshf8_u10avx(__m256 a);
__m256 Sleef_coshf8_u10avx2(__m256 a);
__m256 Sleef_coshf8_u10fma4(__m256 a);
__m256 Sleef_cinz_coshf8_u10avx(__m256 a);
__m256 Sleef_finz_coshf8_u10avx2(__m256 a);
__m256 Sleef_finz_coshf8_u10fma4(__m256 a);

__m512 Sleef_coshf16_u10(__m512 a);
__m512 Sleef_coshf16_u10avx512f(__m512 a);
__m512 Sleef_cinz_coshf16_u10avx512fnofma(__m512 a);
__m512 Sleef_finz_coshf16_u10avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_coshf_u10 with the same accuracy specification.

Vectorized double precision hyperbolic cosine functions

#include <sleef.h>

__m128d Sleef_coshd2_u35(__m128d a);
__m128d Sleef_coshd2_u35sse2(__m128d a);
__m128d Sleef_coshd2_u35sse4(__m128d a);
__m128d Sleef_coshd2_u35avx2128(__m128d a);
__m128d Sleef_cinz_coshd2_u35sse2(__m128d a);
__m128d Sleef_cinz_coshd2_u35sse4(__m128d a);
__m128d Sleef_finz_coshd2_u35avx2128(__m128d a);

__m256d Sleef_coshd4_u35(__m256d a);
__m256d Sleef_coshd4_u35avx(__m256d a);
__m256d Sleef_coshd4_u35avx2(__m256d a);
__m256d Sleef_coshd4_u35fma4(__m256d a);
__m256d Sleef_cinz_coshd4_u35avx(__m256d a);
__m256d Sleef_finz_coshd4_u35avx2(__m256d a);
__m256d Sleef_finz_coshd4_u35fma4(__m256d a);

__m512d Sleef_coshd8_u35(__m512d a);
__m512d Sleef_coshd8_u35avx512f(__m512d a);
__m512d Sleef_cinz_coshd8_u35avx512fnofma(__m512d a);
__m512d Sleef_finz_coshd8_u35avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_cosh_u35 with the same accuracy specification.

Vectorized single precision hyperbolic cosine functions

#include <sleef.h>

__m128 Sleef_coshf4_u35(__m128 a);
__m128 Sleef_coshf4_u35sse2(__m128 a);
__m128 Sleef_coshf4_u35sse4(__m128 a);
__m128 Sleef_coshf4_u35avx2128(__m128 a);
__m128 Sleef_cinz_coshf4_u35sse2(__m128 a);
__m128 Sleef_cinz_coshf4_u35sse4(__m128 a);
__m128 Sleef_finz_coshf4_u35avx2128(__m128 a);

__m256 Sleef_coshf8_u35(__m256 a);
__m256 Sleef_coshf8_u35avx(__m256 a);
__m256 Sleef_coshf8_u35avx2(__m256 a);
__m256 Sleef_coshf8_u35fma4(__m256 a);
__m256 Sleef_cinz_coshf8_u35avx(__m256 a);
__m256 Sleef_finz_coshf8_u35avx2(__m256 a);
__m256 Sleef_finz_coshf8_u35fma4(__m256 a);

__m512 Sleef_coshf16_u35(__m512 a);
__m512 Sleef_coshf16_u35avx512f(__m512 a);
__m512 Sleef_cinz_coshf16_u35avx512fnofma(__m512 a);
__m512 Sleef_finz_coshf16_u35avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_coshf_u35 with the same accuracy specification.

Vectorized double precision hyperbolic tangent functions

#include <sleef.h>

__m128d Sleef_tanhd2_u10(__m128d a);
__m128d Sleef_tanhd2_u10sse2(__m128d a);
__m128d Sleef_tanhd2_u10sse4(__m128d a);
__m128d Sleef_tanhd2_u10avx2128(__m128d a);
__m128d Sleef_cinz_tanhd2_u10sse2(__m128d a);
__m128d Sleef_cinz_tanhd2_u10sse4(__m128d a);
__m128d Sleef_finz_tanhd2_u10avx2128(__m128d a);

__m256d Sleef_tanhd4_u10(__m256d a);
__m256d Sleef_tanhd4_u10avx(__m256d a);
__m256d Sleef_tanhd4_u10avx2(__m256d a);
__m256d Sleef_tanhd4_u10fma4(__m256d a);
__m256d Sleef_cinz_tanhd4_u10avx(__m256d a);
__m256d Sleef_finz_tanhd4_u10avx2(__m256d a);
__m256d Sleef_finz_tanhd4_u10fma4(__m256d a);

__m512d Sleef_tanhd8_u10(__m512d a);
__m512d Sleef_tanhd8_u10avx512f(__m512d a);
__m512d Sleef_cinz_tanhd8_u10avx512fnofma(__m512d a);
__m512d Sleef_finz_tanhd8_u10avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_tanh_u10 with the same accuracy specification.

Vectorized single precision hyperbolic tangent functions

#include <sleef.h>

__m128 Sleef_tanhf4_u10(__m128 a);
__m128 Sleef_tanhf4_u10sse2(__m128 a);
__m128 Sleef_tanhf4_u10sse4(__m128 a);
__m128 Sleef_tanhf4_u10avx2128(__m128 a);
__m128 Sleef_cinz_tanhf4_u10sse2(__m128 a);
__m128 Sleef_cinz_tanhf4_u10sse4(__m128 a);
__m128 Sleef_finz_tanhf4_u10avx2128(__m128 a);

__m256 Sleef_tanhf8_u10(__m256 a);
__m256 Sleef_tanhf8_u10avx(__m256 a);
__m256 Sleef_tanhf8_u10avx2(__m256 a);
__m256 Sleef_tanhf8_u10fma4(__m256 a);
__m256 Sleef_cinz_tanhf8_u10avx(__m256 a);
__m256 Sleef_finz_tanhf8_u10avx2(__m256 a);
__m256 Sleef_finz_tanhf8_u10fma4(__m256 a);

__m512 Sleef_tanhf16_u10(__m512 a);
__m512 Sleef_tanhf16_u10avx512f(__m512 a);
__m512 Sleef_cinz_tanhf16_u10avx512fnofma(__m512 a);
__m512 Sleef_finz_tanhf16_u10avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_tanhf_u10 with the same accuracy specification.

Vectorized double precision hyperbolic tangent functions

#include <sleef.h>

__m128d Sleef_tanhd2_u35(__m128d a);
__m128d Sleef_tanhd2_u35sse2(__m128d a);
__m128d Sleef_tanhd2_u35sse4(__m128d a);
__m128d Sleef_tanhd2_u35avx2128(__m128d a);
__m128d Sleef_cinz_tanhd2_u35sse2(__m128d a);
__m128d Sleef_cinz_tanhd2_u35sse4(__m128d a);
__m128d Sleef_finz_tanhd2_u35avx2128(__m128d a);

__m256d Sleef_tanhd4_u35(__m256d a);
__m256d Sleef_tanhd4_u35avx(__m256d a);
__m256d Sleef_tanhd4_u35avx2(__m256d a);
__m256d Sleef_tanhd4_u35fma4(__m256d a);
__m256d Sleef_cinz_tanhd4_u35avx(__m256d a);
__m256d Sleef_finz_tanhd4_u35avx2(__m256d a);
__m256d Sleef_finz_tanhd4_u35fma4(__m256d a);

__m512d Sleef_tanhd8_u35(__m512d a);
__m512d Sleef_tanhd8_u35avx512f(__m512d a);
__m512d Sleef_cinz_tanhd8_u35avx512fnofma(__m512d a);
__m512d Sleef_finz_tanhd8_u35avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_tanh_u35 with the same accuracy specification.

Vectorized single precision hyperbolic tangent functions

#include <sleef.h>

__m128 Sleef_tanhf4_u35(__m128 a);
__m128 Sleef_tanhf4_u35sse2(__m128 a);
__m128 Sleef_tanhf4_u35sse4(__m128 a);
__m128 Sleef_tanhf4_u35avx2128(__m128 a);
__m128 Sleef_cinz_tanhf4_u35sse2(__m128 a);
__m128 Sleef_cinz_tanhf4_u35sse4(__m128 a);
__m128 Sleef_finz_tanhf4_u35avx2128(__m128 a);

__m256 Sleef_tanhf8_u35(__m256 a);
__m256 Sleef_tanhf8_u35avx(__m256 a);
__m256 Sleef_tanhf8_u35avx2(__m256 a);
__m256 Sleef_tanhf8_u35fma4(__m256 a);
__m256 Sleef_cinz_tanhf8_u35avx(__m256 a);
__m256 Sleef_finz_tanhf8_u35avx2(__m256 a);
__m256 Sleef_finz_tanhf8_u35fma4(__m256 a);

__m512 Sleef_tanhf16_u35(__m512 a);
__m512 Sleef_tanhf16_u35avx512f(__m512 a);
__m512 Sleef_cinz_tanhf16_u35avx512fnofma(__m512 a);
__m512 Sleef_finz_tanhf16_u35avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_tanhf_u35 with the same accuracy specification.

Vectorized double precision inverse hyperbolic sine functions

#include <sleef.h>

__m128d Sleef_asinhd2_u10(__m128d a);
__m128d Sleef_asinhd2_u10sse2(__m128d a);
__m128d Sleef_asinhd2_u10sse4(__m128d a);
__m128d Sleef_asinhd2_u10avx2128(__m128d a);
__m128d Sleef_cinz_asinhd2_u10sse2(__m128d a);
__m128d Sleef_cinz_asinhd2_u10sse4(__m128d a);
__m128d Sleef_finz_asinhd2_u10avx2128(__m128d a);

__m256d Sleef_asinhd4_u10(__m256d a);
__m256d Sleef_asinhd4_u10avx(__m256d a);
__m256d Sleef_asinhd4_u10avx2(__m256d a);
__m256d Sleef_asinhd4_u10fma4(__m256d a);
__m256d Sleef_cinz_asinhd4_u10avx(__m256d a);
__m256d Sleef_finz_asinhd4_u10avx2(__m256d a);
__m256d Sleef_finz_asinhd4_u10fma4(__m256d a);

__m512d Sleef_asinhd8_u10(__m512d a);
__m512d Sleef_asinhd8_u10avx512f(__m512d a);
__m512d Sleef_cinz_asinhd8_u10avx512fnofma(__m512d a);
__m512d Sleef_finz_asinhd8_u10avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_asinh_u10 with the same accuracy specification.

Vectorized single precision inverse hyperbolic sine functions

#include <sleef.h>

__m128 Sleef_asinhf4_u10(__m128 a);
__m128 Sleef_asinhf4_u10sse2(__m128 a);
__m128 Sleef_asinhf4_u10sse4(__m128 a);
__m128 Sleef_asinhf4_u10avx2128(__m128 a);
__m128 Sleef_cinz_asinhf4_u10sse2(__m128 a);
__m128 Sleef_cinz_asinhf4_u10sse4(__m128 a);
__m128 Sleef_finz_asinhf4_u10avx2128(__m128 a);

__m256 Sleef_asinhf8_u10(__m256 a);
__m256 Sleef_asinhf8_u10avx(__m256 a);
__m256 Sleef_asinhf8_u10avx2(__m256 a);
__m256 Sleef_asinhf8_u10fma4(__m256 a);
__m256 Sleef_cinz_asinhf8_u10avx(__m256 a);
__m256 Sleef_finz_asinhf8_u10avx2(__m256 a);
__m256 Sleef_finz_asinhf8_u10fma4(__m256 a);

__m512 Sleef_asinhf16_u10(__m512 a);
__m512 Sleef_asinhf16_u10avx512f(__m512 a);
__m512 Sleef_cinz_asinhf16_u10avx512fnofma(__m512 a);
__m512 Sleef_finz_asinhf16_u10avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_asinhf_u10 with the same accuracy specification.

Vectorized double precision inverse hyperbolic cosine functions

#include <sleef.h>

__m128d Sleef_acoshd2_u10(__m128d a);
__m128d Sleef_acoshd2_u10sse2(__m128d a);
__m128d Sleef_acoshd2_u10sse4(__m128d a);
__m128d Sleef_acoshd2_u10avx2128(__m128d a);
__m128d Sleef_cinz_acoshd2_u10sse2(__m128d a);
__m128d Sleef_cinz_acoshd2_u10sse4(__m128d a);
__m128d Sleef_finz_acoshd2_u10avx2128(__m128d a);

__m256d Sleef_acoshd4_u10(__m256d a);
__m256d Sleef_acoshd4_u10avx(__m256d a);
__m256d Sleef_acoshd4_u10avx2(__m256d a);
__m256d Sleef_acoshd4_u10fma4(__m256d a);
__m256d Sleef_cinz_acoshd4_u10avx(__m256d a);
__m256d Sleef_finz_acoshd4_u10avx2(__m256d a);
__m256d Sleef_finz_acoshd4_u10fma4(__m256d a);

__m512d Sleef_acoshd8_u10(__m512d a);
__m512d Sleef_acoshd8_u10avx512f(__m512d a);
__m512d Sleef_cinz_acoshd8_u10avx512fnofma(__m512d a);
__m512d Sleef_finz_acoshd8_u10avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_acosh_u10 with the same accuracy specification.

Vectorized single precision inverse hyperbolic cosine functions

#include <sleef.h>

__m128 Sleef_acoshf4_u10(__m128 a);
__m128 Sleef_acoshf4_u10sse2(__m128 a);
__m128 Sleef_acoshf4_u10sse4(__m128 a);
__m128 Sleef_acoshf4_u10avx2128(__m128 a);
__m128 Sleef_cinz_acoshf4_u10sse2(__m128 a);
__m128 Sleef_cinz_acoshf4_u10sse4(__m128 a);
__m128 Sleef_finz_acoshf4_u10avx2128(__m128 a);

__m256 Sleef_acoshf8_u10(__m256 a);
__m256 Sleef_acoshf8_u10avx(__m256 a);
__m256 Sleef_acoshf8_u10avx2(__m256 a);
__m256 Sleef_acoshf8_u10fma4(__m256 a);
__m256 Sleef_cinz_acoshf8_u10avx(__m256 a);
__m256 Sleef_finz_acoshf8_u10avx2(__m256 a);
__m256 Sleef_finz_acoshf8_u10fma4(__m256 a);

__m512 Sleef_acoshf16_u10(__m512 a);
__m512 Sleef_acoshf16_u10avx512f(__m512 a);
__m512 Sleef_cinz_acoshf16_u10avx512fnofma(__m512 a);
__m512 Sleef_finz_acoshf16_u10avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_acoshf_u10 with the same accuracy specification.

Vectorized double precision inverse hyperbolic tangent functions

#include <sleef.h>

__m128d Sleef_atanhd2_u10(__m128d a);
__m128d Sleef_atanhd2_u10sse2(__m128d a);
__m128d Sleef_atanhd2_u10sse4(__m128d a);
__m128d Sleef_atanhd2_u10avx2128(__m128d a);
__m128d Sleef_cinz_atanhd2_u10sse2(__m128d a);
__m128d Sleef_cinz_atanhd2_u10sse4(__m128d a);
__m128d Sleef_finz_atanhd2_u10avx2128(__m128d a);

__m256d Sleef_atanhd4_u10(__m256d a);
__m256d Sleef_atanhd4_u10avx(__m256d a);
__m256d Sleef_atanhd4_u10avx2(__m256d a);
__m256d Sleef_atanhd4_u10fma4(__m256d a);
__m256d Sleef_cinz_atanhd4_u10avx(__m256d a);
__m256d Sleef_finz_atanhd4_u10avx2(__m256d a);
__m256d Sleef_finz_atanhd4_u10fma4(__m256d a);

__m512d Sleef_atanhd8_u10(__m512d a);
__m512d Sleef_atanhd8_u10avx512f(__m512d a);
__m512d Sleef_cinz_atanhd8_u10avx512fnofma(__m512d a);
__m512d Sleef_finz_atanhd8_u10avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_atanh_u10 with the same accuracy specification.

Vectorized single precision inverse hyperbolic tangent functions

#include <sleef.h>

__m128 Sleef_atanhf4_u10(__m128 a);
__m128 Sleef_atanhf4_u10sse2(__m128 a);
__m128 Sleef_atanhf4_u10sse4(__m128 a);
__m128 Sleef_atanhf4_u10avx2128(__m128 a);
__m128 Sleef_cinz_atanhf4_u10sse2(__m128 a);
__m128 Sleef_cinz_atanhf4_u10sse4(__m128 a);
__m128 Sleef_finz_atanhf4_u10avx2128(__m128 a);

__m256 Sleef_atanhf8_u10(__m256 a);
__m256 Sleef_atanhf8_u10avx(__m256 a);
__m256 Sleef_atanhf8_u10avx2(__m256 a);
__m256 Sleef_atanhf8_u10fma4(__m256 a);
__m256 Sleef_cinz_atanhf8_u10avx(__m256 a);
__m256 Sleef_finz_atanhf8_u10avx2(__m256 a);
__m256 Sleef_finz_atanhf8_u10fma4(__m256 a);

__m512 Sleef_atanhf16_u10(__m512 a);
__m512 Sleef_atanhf16_u10avx512f(__m512 a);
__m512 Sleef_cinz_atanhf16_u10avx512fnofma(__m512 a);
__m512 Sleef_finz_atanhf16_u10avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_atanhf_u10 with the same accuracy specification.

Error and gamma functions

Vectorized double precision error functions

#include <sleef.h>

__m128 Sleef_erfd2_u10(__m128 a);
__m128 Sleef_erfd2_u10sse2(__m128 a);
__m128 Sleef_erfd2_u10sse4(__m128 a);
__m128 Sleef_erfd2_u10avx2128(__m128 a);
__m128 Sleef_cinz_erfd2_u10sse2(__m128 a);
__m128 Sleef_cinz_erfd2_u10sse4(__m128 a);
__m128 Sleef_finz_erfd2_u10avx2128(__m128 a);

__m128 Sleef_erfd4_u10(__m128 a);
__m128 Sleef_erfd4_u10avx(__m128 a);
__m128 Sleef_erfd4_u10avx2(__m128 a);
__m128 Sleef_erfd4_u10fma4(__m128 a);
__m128 Sleef_cinz_erfd4_u10avx(__m128 a);
__m128 Sleef_finz_erfd4_u10avx2(__m128 a);
__m128 Sleef_finz_erfd4_u10fma4(__m128 a);

__m256 Sleef_erfd8_u10(__m256 a);
__m256 Sleef_erfd8_u10avx512f(__m256 a);
__m256 Sleef_cinz_erfd8_u10avx512fnofma(__m256 a);
__m256 Sleef_finz_erfd8_u10avx512f(__m256 a);

Link with -lsleef.

These are the vectorized functions of Sleef_erf_u10 with the same accuracy specification.

Vectorized single precision error functions

#include <sleef.h>

__m128 Sleef_erff4_u10(__m128 a);
__m128 Sleef_erff4_u10sse2(__m128 a);
__m128 Sleef_erff4_u10sse4(__m128 a);
__m128 Sleef_erff4_u10avx2128(__m128 a);
__m128 Sleef_cinz_erff4_u10sse2(__m128 a);
__m128 Sleef_cinz_erff4_u10sse4(__m128 a);
__m128 Sleef_finz_erff4_u10avx2128(__m128 a);

__m256 Sleef_erff8_u10(__m256 a);
__m256 Sleef_erff8_u10avx(__m256 a);
__m256 Sleef_erff8_u10avx2(__m256 a);
__m256 Sleef_erff8_u10fma4(__m256 a);
__m256 Sleef_cinz_erff8_u10avx(__m256 a);
__m256 Sleef_finz_erff8_u10avx2(__m256 a);
__m256 Sleef_finz_erff8_u10fma4(__m256 a);

__m512 Sleef_erff16_u10(__m512 a);
__m512 Sleef_erff16_u10avx512f(__m512 a);
__m512 Sleef_cinz_erff16_u10avx512fnofma(__m512 a);
__m512 Sleef_finz_erff16_u10avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_erff_u10 with the same accuracy specification.

Vectorized double precision complementary error functions

#include <sleef.h>

__m128d Sleef_erfcd2_u15(__m128d a);
__m128d Sleef_erfcd2_u15sse2(__m128d a);
__m128d Sleef_erfcd2_u15sse4(__m128d a);
__m128d Sleef_erfcd2_u15avx2128(__m128d a);
__m128d Sleef_cinz_erfcd2_u15sse2(__m128d a);
__m128d Sleef_cinz_erfcd2_u15sse4(__m128d a);
__m128d Sleef_finz_erfcd2_u15avx2128(__m128d a);

__m256d Sleef_erfcd4_u15(__m256d a);
__m256d Sleef_erfcd4_u15avx(__m256d a);
__m256d Sleef_erfcd4_u15avx2(__m256d a);
__m256d Sleef_erfcd4_u15fma4(__m256d a);
__m256d Sleef_cinz_erfcd4_u15avx(__m256d a);
__m256d Sleef_finz_erfcd4_u15avx2(__m256d a);
__m256d Sleef_finz_erfcd4_u15fma4(__m256d a);

__m512d Sleef_erfcd8_u15(__m512d a);
__m512d Sleef_erfcd8_u15avx512f(__m512d a);
__m512d Sleef_cinz_erfcd8_u15avx512fnofma(__m512d a);
__m512d Sleef_finz_erfcd8_u15avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_erfc_u15 with the same accuracy specification.

Vectorized single precision complementary error functions

#include <sleef.h>

__m128 Sleef_erfcf4_u15(__m128 a);
__m128 Sleef_erfcf4_u15sse2(__m128 a);
__m128 Sleef_erfcf4_u15sse4(__m128 a);
__m128 Sleef_erfcf4_u15avx2128(__m128 a);
__m128 Sleef_cinz_erfcf4_u15sse2(__m128 a);
__m128 Sleef_cinz_erfcf4_u15sse4(__m128 a);
__m128 Sleef_finz_erfcf4_u15avx2128(__m128 a);

__m256 Sleef_erfcf8_u15(__m256 a);
__m256 Sleef_erfcf8_u15avx(__m256 a);
__m256 Sleef_erfcf8_u15avx2(__m256 a);
__m256 Sleef_erfcf8_u15fma4(__m256 a);
__m256 Sleef_cinz_erfcf8_u15avx(__m256 a);
__m256 Sleef_finz_erfcf8_u15avx2(__m256 a);
__m256 Sleef_finz_erfcf8_u15fma4(__m256 a);

__m512 Sleef_erfcf16_u15(__m512 a);
__m512 Sleef_erfcf16_u15avx512f(__m512 a);
__m512 Sleef_cinz_erfcf16_u15avx512fnofma(__m512 a);
__m512 Sleef_finz_erfcf16_u15avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_erfcf_u15 with the same accuracy specification.

Vectorized double precision gamma functions

#include <sleef.h>

__m128d Sleef_tgammad2_u10(__m128d a);
__m128d Sleef_tgammad2_u10sse2(__m128d a);
__m128d Sleef_tgammad2_u10sse4(__m128d a);
__m128d Sleef_tgammad2_u10avx2128(__m128d a);
__m128d Sleef_cinz_tgammad2_u10sse2(__m128d a);
__m128d Sleef_cinz_tgammad2_u10sse4(__m128d a);
__m128d Sleef_finz_tgammad2_u10avx2128(__m128d a);

__m256d Sleef_tgammad4_u10(__m256d a);
__m256d Sleef_tgammad4_u10avx(__m256d a);
__m256d Sleef_tgammad4_u10avx2(__m256d a);
__m256d Sleef_tgammad4_u10fma4(__m256d a);
__m256d Sleef_cinz_tgammad4_u10avx(__m256d a);
__m256d Sleef_finz_tgammad4_u10avx2(__m256d a);
__m256d Sleef_finz_tgammad4_u10fma4(__m256d a);

__m512d Sleef_tgammad8_u10(__m512d a);
__m512d Sleef_tgammad8_u10avx512f(__m512d a);
__m512d Sleef_cinz_tgammad8_u10avx512fnofma(__m512d a);
__m512d Sleef_finz_tgammad8_u10avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_tgamma_u10 with the same accuracy specification.

Vectorized single precision gamma functions

#include <sleef.h>

__m128 Sleef_tgammaf4_u10(__m128 a);
__m128 Sleef_tgammaf4_u10sse2(__m128 a);
__m128 Sleef_tgammaf4_u10sse4(__m128 a);
__m128 Sleef_tgammaf4_u10avx2128(__m128 a);
__m128 Sleef_cinz_tgammaf4_u10sse2(__m128 a);
__m128 Sleef_cinz_tgammaf4_u10sse4(__m128 a);
__m128 Sleef_finz_tgammaf4_u10avx2128(__m128 a);

__m256 Sleef_tgammaf8_u10(__m256 a);
__m256 Sleef_tgammaf8_u10avx(__m256 a);
__m256 Sleef_tgammaf8_u10avx2(__m256 a);
__m256 Sleef_tgammaf8_u10fma4(__m256 a);
__m256 Sleef_cinz_tgammaf8_u10avx(__m256 a);
__m256 Sleef_finz_tgammaf8_u10avx2(__m256 a);
__m256 Sleef_finz_tgammaf8_u10fma4(__m256 a);

__m512 Sleef_tgammaf16_u10(__m512 a);
__m512 Sleef_tgammaf16_u10avx512f(__m512 a);
__m512 Sleef_cinz_tgammaf16_u10avx512fnofma(__m512 a);
__m512 Sleef_finz_tgammaf16_u10avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_tgammaf_u10 with the same accuracy specification.

Vectorized double precision log gamma functions

#include <sleef.h>

__m128d Sleef_lgammad2_u10(__m128d a);
__m128d Sleef_lgammad2_u10sse2(__m128d a);
__m128d Sleef_lgammad2_u10sse4(__m128d a);
__m128d Sleef_lgammad2_u10avx2128(__m128d a);
__m128d Sleef_cinz_lgammad2_u10sse2(__m128d a);
__m128d Sleef_cinz_lgammad2_u10sse4(__m128d a);
__m128d Sleef_finz_lgammad2_u10avx2128(__m128d a);

__m256d Sleef_lgammad4_u10(__m256d a);
__m256d Sleef_lgammad4_u10avx(__m256d a);
__m256d Sleef_lgammad4_u10avx2(__m256d a);
__m256d Sleef_lgammad4_u10fma4(__m256d a);
__m256d Sleef_cinz_lgammad4_u10avx(__m256d a);
__m256d Sleef_finz_lgammad4_u10avx2(__m256d a);
__m256d Sleef_finz_lgammad4_u10fma4(__m256d a);

__m512d Sleef_lgammad8_u10(__m512d a);
__m512d Sleef_lgammad8_u10avx512f(__m512d a);
__m512d Sleef_cinz_lgammad8_u10avx512fnofma(__m512d a);
__m512d Sleef_finz_lgammad8_u10avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_lgamma_u10 with the same accuracy specification.

Vectorized single precision log gamma functions

#include <sleef.h>

__m128 Sleef_lgammaf4_u10(__m128 a);
__m128 Sleef_lgammaf4_u10sse2(__m128 a);
__m128 Sleef_lgammaf4_u10sse4(__m128 a);
__m128 Sleef_lgammaf4_u10avx2128(__m128 a);
__m128 Sleef_cinz_lgammaf4_u10sse2(__m128 a);
__m128 Sleef_cinz_lgammaf4_u10sse4(__m128 a);
__m128 Sleef_finz_lgammaf4_u10avx2128(__m128 a);

__m256 Sleef_lgammaf8_u10(__m256 a);
__m256 Sleef_lgammaf8_u10avx(__m256 a);
__m256 Sleef_lgammaf8_u10avx2(__m256 a);
__m256 Sleef_lgammaf8_u10fma4(__m256 a);
__m256 Sleef_cinz_lgammaf8_u10avx(__m256 a);
__m256 Sleef_finz_lgammaf8_u10avx2(__m256 a);
__m256 Sleef_finz_lgammaf8_u10fma4(__m256 a);

__m512 Sleef_lgammaf16_u10(__m512 a);
__m512 Sleef_lgammaf16_u10avx512f(__m512 a);
__m512 Sleef_cinz_lgammaf16_u10avx512fnofma(__m512 a);
__m512 Sleef_finz_lgammaf16_u10avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_lgammaf_u10 with the same accuracy specification.

Nearest integer functions

Vectorized double precision functions for rounding to integer towards zero

#include <sleef.h>

__m128d Sleef_truncd2(__m128d a);
__m128d Sleef_truncd2_sse2(__m128d a);
__m128d Sleef_truncd2_sse4(__m128d a);
__m128d Sleef_truncd2_avx2128(__m128d a);

__m256d Sleef_truncd4(__m256d a);
__m256d Sleef_truncd4_avx(__m256d a);
__m256d Sleef_truncd4_fma4(__m256d a);
__m256d Sleef_truncd4_avx2(__m256d a);

__m512d Sleef_truncd8(__m512d a);
__m512d Sleef_truncd8_avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_trunc with the same accuracy specification.

Vectorized single precision functions for rounding to integer towards zero

#include <sleef.h>

__m128 Sleef_truncf4(__m128 a);
__m128 Sleef_truncf4_sse2(__m128 a);
__m128 Sleef_truncf4_sse4(__m128 a);
__m128 Sleef_truncf4_avx2128(__m128 a);

__m256 Sleef_truncf8(__m256 a);
__m256 Sleef_truncf8_avx(__m256 a);
__m256 Sleef_truncf8_fma4(__m256 a);
__m256 Sleef_truncf8_avx2(__m256 a);

__m512 Sleef_truncf16(__m512 a);
__m512 Sleef_truncf16_avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_truncf with the same accuracy specification.

Vectorized double precision functions for rounding to integer towards negative infinity

#include <sleef.h>

__m128d Sleef_floord2(__m128d a);
__m128d Sleef_floord2_sse2(__m128d a);
__m128d Sleef_floord2_sse4(__m128d a);
__m128d Sleef_floord2_avx2128(__m128d a);

__m256d Sleef_floord4(__m256d a);
__m256d Sleef_floord4_avx(__m256d a);
__m256d Sleef_floord4_fma4(__m256d a);
__m256d Sleef_floord4_avx2(__m256d a);

__m512d Sleef_floord8(__m512d a);
__m512d Sleef_floord8_avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_floor with the same accuracy specification.

Vectorized single precision functions for rounding to integer towards negative infinity

#include <sleef.h>

__m128 Sleef_floorf4(__m128 a);
__m128 Sleef_floorf4_sse2(__m128 a);
__m128 Sleef_floorf4_sse4(__m128 a);
__m128 Sleef_floorf4_avx2128(__m128 a);

__m256 Sleef_floorf8(__m256 a);
__m256 Sleef_floorf8_avx(__m256 a);
__m256 Sleef_floorf8_fma4(__m256 a);
__m256 Sleef_floorf8_avx2(__m256 a);

__m512 Sleef_floorf16(__m512 a);
__m512 Sleef_floorf16_avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_floorf with the same accuracy specification.

Vectorized double precision functions for rounding to integer towards positive infinity

#include <sleef.h>

__m128d Sleef_ceild2(__m128d a);
__m128d Sleef_ceild2_sse2(__m128d a);
__m128d Sleef_ceild2_sse4(__m128d a);
__m128d Sleef_ceild2_avx2128(__m128d a);

__m256d Sleef_ceild4(__m256d a);
__m256d Sleef_ceild4_avx(__m256d a);
__m256d Sleef_ceild4_fma4(__m256d a);
__m256d Sleef_ceild4_avx2(__m256d a);

__m512d Sleef_ceild8(__m512d a);
__m512d Sleef_ceild8_avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_ceil with the same accuracy specification.

Vectorized single precision functions for rounding to integer towards positive infinity

#include <sleef.h>

__m128 Sleef_ceilf4(__m128 a);
__m128 Sleef_ceilf4_sse2(__m128 a);
__m128 Sleef_ceilf4_sse4(__m128 a);
__m128 Sleef_ceilf4_avx2128(__m128 a);

__m256 Sleef_ceilf8(__m256 a);
__m256 Sleef_ceilf8_avx(__m256 a);
__m256 Sleef_ceilf8_fma4(__m256 a);
__m256 Sleef_ceilf8_avx2(__m256 a);

__m512 Sleef_ceilf16(__m512 a);
__m512 Sleef_ceilf16_avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_ceilf with the same accuracy specification.

Vectorized double precision functions for rounding to nearest integer

#include <sleef.h>

__m128d Sleef_roundd2(__m128d a);
__m128d Sleef_roundd2_sse2(__m128d a);
__m128d Sleef_roundd2_sse4(__m128d a);
__m128d Sleef_roundd2_avx2128(__m128d a);

__m256d Sleef_roundd4(__m256d a);
__m256d Sleef_roundd4_avx(__m256d a);
__m256d Sleef_roundd4_fma4(__m256d a);
__m256d Sleef_roundd4_avx2(__m256d a);

__m512d Sleef_roundd8(__m512d a);
__m512d Sleef_roundd8_avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_round with the same accuracy specification.

Vectorized single precision functions for rounding to nearest integer

#include <sleef.h>

__m128 Sleef_roundf4(__m128 a);
__m128 Sleef_roundf4_sse2(__m128 a);
__m128 Sleef_roundf4_sse4(__m128 a);
__m128 Sleef_roundf4_avx2128(__m128 a);

__m256 Sleef_roundf8(__m256 a);
__m256 Sleef_roundf8_avx(__m256 a);
__m256 Sleef_roundf8_fma4(__m256 a);
__m256 Sleef_roundf8_avx2(__m256 a);

__m512 Sleef_roundf16(__m512 a);
__m512 Sleef_roundf16_avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_roundf with the same accuracy specification.

Vectorized double precision functions for rounding to nearest integer

#include <sleef.h>

__m128d Sleef_rintd2(__m128d a);
__m128d Sleef_rintd2_sse2(__m128d a);
__m128d Sleef_rintd2_sse4(__m128d a);
__m128d Sleef_rintd2_avx2128(__m128d a);

__m256d Sleef_rintd4(__m256d a);
__m256d Sleef_rintd4_avx(__m256d a);
__m256d Sleef_rintd4_fma4(__m256d a);
__m256d Sleef_rintd4_avx2(__m256d a);

__m512d Sleef_rintd8(__m512d a);
__m512d Sleef_rintd8_avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_rint with the same accuracy specification.

Vectorized single precision functions for rounding to nearest integer

#include <sleef.h>

__m128 Sleef_rintf4(__m128 a);
__m128 Sleef_rintf4_sse2(__m128 a);
__m128 Sleef_rintf4_sse4(__m128 a);
__m128 Sleef_rintf4_avx2128(__m128 a);

__m256 Sleef_rintf8(__m256 a);
__m256 Sleef_rintf8_avx(__m256 a);
__m256 Sleef_rintf8_fma4(__m256 a);
__m256 Sleef_rintf8_avx2(__m256 a);

__m512 Sleef_rintf16(__m512 a);
__m512 Sleef_rintf16_avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_rintf with the same accuracy specification.

Other functions

Vectorized double precision functions for fused multiply-accumulation

#include <sleef.h>

__m128d Sleef_fmad2(__m128d a, __m128d b, __m128d c);
__m128d Sleef_fmad2_sse2(__m128d a, __m128d b, __m128d c);
__m128d Sleef_fmad2_sse4(__m128d a, __m128d b, __m128d c);
__m128d Sleef_fmad2_avx2128(__m128d a, __m128d b, __m128d c);

__m256d Sleef_fmad4(__m256d a, __m256d b, __m256d c);
__m256d Sleef_fmad4_avx(__m256d a, __m256d b, __m256d c);
__m256d Sleef_fmad4_fma4(__m256d a, __m256d b, __m256d c);
__m256d Sleef_fmad4_avx2(__m256d a, __m256d b, __m256d c);

__m512d Sleef_fmad8(__m512d a, __m512d b, __m512d c);
__m512d Sleef_fmad8_avx512f(__m512d a, __m512d b, __m512d c);

Link with -lsleef.

These are the vectorized functions of Sleef_fma with the same accuracy specification.

Vectorized single precision functions for fused multiply-accumulation

#include <sleef.h>

__m128 Sleef_fmaf4(__m128 a, __m128 b, __m128 c);
__m128 Sleef_fmaf4_sse2(__m128 a, __m128 b, __m128 c);
__m128 Sleef_fmaf4_sse4(__m128 a, __m128 b, __m128 c);
__m128 Sleef_fmaf4_avx2128(__m128 a, __m128 b, __m128 c);

__m256 Sleef_fmaf8(__m256 a, __m256 b, __m256 c);
__m256 Sleef_fmaf8_avx(__m256 a, __m256 b, __m256 c);
__m256 Sleef_fmaf8_fma4(__m256 a, __m256 b, __m256 c);
__m256 Sleef_fmaf8_avx2(__m256 a, __m256 b, __m256 c);

__m512 Sleef_fmaf16(__m512 a, __m512 b, __m512 c);
__m512 Sleef_fmaf16_avx512f(__m512 a, __m512 b, __m512 c);

Link with -lsleef.

These are the vectorized functions of Sleef_fmaf with the same accuracy specification.

Vectorized double precision FP remainder

#include <sleef.h>

__m128d Sleef_fmodd2(__m128d a, __m128d b);
__m128d Sleef_fmodd2_sse2(__m128d a, __m128d b);
__m128d Sleef_fmodd2_sse4(__m128d a, __m128d b);
__m128d Sleef_fmodd2_avx2128(__m128d a, __m128d b);

__m256d Sleef_fmodd4(__m256d a, __m256d b);
__m256d Sleef_fmodd4_avx(__m256d a, __m256d b);
__m256d Sleef_fmodd4_fma4(__m256d a, __m256d b);
__m256d Sleef_fmodd4_avx2(__m256d a, __m256d b);

__m512d Sleef_fmodd8(__m512d a, __m512d b);
__m512d Sleef_fmodd8_avx512f(__m512d a, __m512d b);

Link with -lsleef.

These are the vectorized functions of Sleef_fmod with the same accuracy specification.

Vectorized single precision FP remainder

#include <sleef.h>

__m128 Sleef_fmodf4(__m128 a, __m128 b);
__m128 Sleef_fmodf4_sse2(__m128 a, __m128 b);
__m128 Sleef_fmodf4_sse4(__m128 a, __m128 b);
__m128 Sleef_fmodf4_avx2128(__m128 a, __m128 b);

__m256 Sleef_fmodf8(__m256 a, __m256 b);
__m256 Sleef_fmodf8_avx(__m256 a, __m256 b);
__m256 Sleef_fmodf8_fma4(__m256 a, __m256 b);
__m256 Sleef_fmodf8_avx2(__m256 a, __m256 b);

__m512 Sleef_fmodf16(__m512 a, __m512 b);
__m512 Sleef_fmodf16_avx512f(__m512 a, __m512 b);

Link with -lsleef.

These are the vectorized functions of Sleef_fmodf with the same accuracy specification.

Vectorized double precision FP remainder

#include <sleef.h>

__m128d Sleef_remainderd2(__m128d a, __m128d b);
__m128d Sleef_remainderd2_sse2(__m128d a, __m128d b);
__m128d Sleef_remainderd2_sse4(__m128d a, __m128d b);
__m128d Sleef_remainderd2_avx2128(__m128d a, __m128d b);

__m256d Sleef_remainderd4(__m256d a, __m256d b);
__m256d Sleef_remainderd4_avx(__m256d a, __m256d b);
__m256d Sleef_remainderd4_fma4(__m256d a, __m256d b);
__m256d Sleef_remainderd4_avx2(__m256d a, __m256d b);

__m512d Sleef_remainderd8(__m512d a, __m512d b);
__m512d Sleef_remainderd8_avx512f(__m512d a, __m512d b);

Link with -lsleef.

These are the vectorized functions of Sleef_remainder with the same accuracy specification.

Vectorized single precision FP remainder

#include <sleef.h>

__m128 Sleef_remainderf4(__m128 a, __m128 b);
__m128 Sleef_remainderf4_sse2(__m128 a, __m128 b);
__m128 Sleef_remainderf4_sse4(__m128 a, __m128 b);
__m128 Sleef_remainderf4_avx2128(__m128 a, __m128 b);

__m256 Sleef_remainderf8(__m256 a, __m256 b);
__m256 Sleef_remainderf8_avx(__m256 a, __m256 b);
__m256 Sleef_remainderf8_fma4(__m256 a, __m256 b);
__m256 Sleef_remainderf8_avx2(__m256 a, __m256 b);

__m512 Sleef_remainderf16(__m512 a, __m512 b);
__m512 Sleef_remainderf16_avx512f(__m512 a, __m512 b);

Link with -lsleef.

These are the vectorized functions of Sleef_remainderf with the same accuracy specification.

Vectorized double precision functions for multiplying by integral power of 2

#include <sleef.h>

__m128d Sleef_ldexpd2(__m128d a, __m128i b);
__m128d Sleef_ldexpd2_sse2(__m128d a, __m128i b);
__m128d Sleef_ldexpd2_sse4(__m128d a, __m128i b);
__m128d Sleef_ldexpd2_avx2128(__m128d a, __m128i b);

__m256d Sleef_ldexpd4(__m256d a, __m128i b);
__m256d Sleef_ldexpd4_avx(__m256d a, __m128i b);
__m256d Sleef_ldexpd4_fma4(__m256d a, __m128i b);
__m256d Sleef_ldexpd4_avx2(__m256d a, __m128i b);

__m512d Sleef_ldexpd8(__m512d a, __m256i b);
__m512d Sleef_ldexpd8_avx512f(__m512d a, __m256i b);

Link with -lsleef.

These are the vectorized functions of Sleef_ldexp with the same accuracy specification.

Vectorized double precision functions for obtaining fractional component of an FP number

#include <sleef.h>

__m128d Sleef_frfrexpd2(__m128d a);
__m128d Sleef_frfrexpd2_sse2(__m128d a);
__m128d Sleef_frfrexpd2_sse4(__m128d a);
__m128d Sleef_frfrexpd2_avx2128(__m128d a);

__m256d Sleef_frfrexpd4(__m256d a);
__m256d Sleef_frfrexpd4_avx(__m256d a);
__m256d Sleef_frfrexpd4_fma4(__m256d a);
__m256d Sleef_frfrexpd4_avx2(__m256d a);

__m512d Sleef_frfrexpd8(__m512d a);
__m512d Sleef_frfrexpd8_avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_frfrexp with the same accuracy specification.

Vectorized single precision functions for obtaining fractional component of an FP number

#include <sleef.h>

__m128 Sleef_frfrexpf4(__m128 a);
__m128 Sleef_frfrexpf4_sse2(__m128 a);
__m128 Sleef_frfrexpf4_sse4(__m128 a);
__m128 Sleef_frfrexpf4_avx2128(__m128 a);

__m256 Sleef_frfrexpf8(__m256 a);
__m256 Sleef_frfrexpf8_avx(__m256 a);
__m256 Sleef_frfrexpf8_fma4(__m256 a);
__m256 Sleef_frfrexpf8_avx2(__m256 a);

__m512 Sleef_frfrexpf16(__m512 a);
__m512 Sleef_frfrexpf16_avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_frfrexpf with the same accuracy specification.

Vectorized double precision function for obtaining integral component of an FP number

#include <sleef.h>

__m128i Sleef_expfrexpd2(__m128d a);
__m128i Sleef_expfrexpd2_sse2(__m128d a);
__m128i Sleef_expfrexpd2_sse4(__m128d a);
__m128i Sleef_expfrexpd2_avx2128(__m128d a);

__m128i Sleef_expfrexpd4(__m256d a);
__m128i Sleef_expfrexpd4_avx(__m256d a);
__m128i Sleef_expfrexpd4_fma4(__m256d a);
__m128i Sleef_expfrexpd4_avx2(__m256d a);

__m256i Sleef_expfrexpd8(__m512d a);
__m256i Sleef_expfrexpd8_avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_expfrexp with the same accuracy specification.

Vectorized double precision functions for getting integer exponent

#include <sleef.h>

__m128i Sleef_ilogbd2(__m128d a);
__m128i Sleef_ilogbd2_sse2(__m128d a);
__m128i Sleef_ilogbd2_sse4(__m128d a);
__m128i Sleef_ilogbd2_avx2128(__m128d a);

__m128i Sleef_ilogbd4(__m256d a);
__m128i Sleef_ilogbd4_avx(__m256d a);
__m128i Sleef_ilogbd4_fma4(__m256d a);
__m128i Sleef_ilogbd4_avx2(__m256d a);

__m256i Sleef_ilogbd8(__m512d a);
__m256i Sleef_ilogbd8_avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_ilogb with the same accuracy specification.

Vectorized double precision signed integral and fractional values

#include <sleef.h>

Sleef___m128d_2 Sleef_modfd2(__m128d a);
Sleef___m128d_2 Sleef_modfd2_sse2(__m128d a);
Sleef___m128d_2 Sleef_modfd2_sse4(__m128d a);
Sleef___m128d_2 Sleef_modfd2_avx2128(__m128d a);

Sleef___m256d_2 Sleef_modfd4(__m256d a);
Sleef___m256d_2 Sleef_modfd4_avx(__m256d a);
Sleef___m256d_2 Sleef_modfd4_fma4(__m256d a);
Sleef___m256d_2 Sleef_modfd4_avx2(__m256d a);

Sleef___m512d_2 Sleef_modfd8(__m512d a);
Sleef___m512d_2 Sleef_modfd8_avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_modf with the same accuracy specification.

Vectorized single precision signed integral and fractional values

#include <sleef.h>

Sleef___m128_2 Sleef_modff4(__m128 a);
Sleef___m128_2 Sleef_modff4_sse2(__m128 a);
Sleef___m128_2 Sleef_modff4_sse4(__m128 a);
Sleef___m128_2 Sleef_modff4_avx2128(__m128 a);

Sleef___m256_2 Sleef_modff8(__m256 a);
Sleef___m256_2 Sleef_modff8_avx(__m256 a);
Sleef___m256_2 Sleef_modff8_fma4(__m256 a);
Sleef___m256_2 Sleef_modff8_avx2(__m256 a);

Sleef___m512_2 Sleef_modff16(__m512 a);
Sleef___m512_2 Sleef_modff16_avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_modff with the same accuracy specification.

Vectorized double precision functions for calculating the absolute value

#include <sleef.h>

__m128d Sleef_fabsd2(__m128d a);
__m128d Sleef_fabsd2_sse2(__m128d a);
__m128d Sleef_fabsd2_sse4(__m128d a);
__m128d Sleef_fabsd2_avx2128(__m128d a);

__m256d Sleef_fabsd4(__m256d a);
__m256d Sleef_fabsd4_avx(__m256d a);
__m256d Sleef_fabsd4_fma4(__m256d a);
__m256d Sleef_fabsd4_avx2(__m256d a);

__m512d Sleef_fabsd8(__m512d a);
__m512d Sleef_fabsd8_avx512f(__m512d a);

Link with -lsleef.

These are the vectorized functions of Sleef_fabs with the same accuracy specification.

Vectorized single precision functions for calculating the absolute value

#include <sleef.h>

__m128 Sleef_fabsf4(__m128 a);
__m128 Sleef_fabsf4_sse2(__m128 a);
__m128 Sleef_fabsf4_sse4(__m128 a);
__m128 Sleef_fabsf4_avx2128(__m128 a);

__m256 Sleef_fabsf8(__m256 a);
__m256 Sleef_fabsf8_avx(__m256 a);
__m256 Sleef_fabsf8_fma4(__m256 a);
__m256 Sleef_fabsf8_avx2(__m256 a);

__m512 Sleef_fabsf16(__m512 a);
__m512 Sleef_fabsf16_avx512f(__m512 a);

Link with -lsleef.

These are the vectorized functions of Sleef_fabsf with the same accuracy specification.

Vectorized double precision functions for copying signs

#include <sleef.h>

__m128d Sleef_copysignd2(__m128d a, __m128d b);
__m128d Sleef_copysignd2_sse2(__m128d a, __m128d b);
__m128d Sleef_copysignd2_sse4(__m128d a, __m128d b);
__m128d Sleef_copysignd2_avx2128(__m128d a, __m128d b);

__m256d Sleef_copysignd4(__m256d a, __m256d b);
__m256d Sleef_copysignd4_avx(__m256d a, __m256d b);
__m256d Sleef_copysignd4_fma4(__m256d a, __m256d b);
__m256d Sleef_copysignd4_avx2(__m256d a, __m256d b);

__m512d Sleef_copysignd8(__m512d a, __m512d b);
__m512d Sleef_copysignd8_avx512f(__m512d a, __m512d b);

Link with -lsleef.

These are the vectorized functions of Sleef_copysign with the same accuracy specification.

Vectorized single precision functions for copying signs

#include <sleef.h>

__m128 Sleef_copysignf4(__m128 a, __m128 b);
__m128 Sleef_copysignf4_sse2(__m128 a, __m128 b);
__m128 Sleef_copysignf4_sse4(__m128 a, __m128 b);
__m128 Sleef_copysignf4_avx2128(__m128 a, __m128 b);

__m256 Sleef_copysignf8(__m256 a, __m256 b);
__m256 Sleef_copysignf8_avx(__m256 a, __m256 b);
__m256 Sleef_copysignf8_fma4(__m256 a, __m256 b);
__m256 Sleef_copysignf8_avx2(__m256 a, __m256 b);

__m512 Sleef_copysignf16(__m512 a, __m512 b);
__m512 Sleef_copysignf16_avx512f(__m512 a, __m512 b);

Link with -lsleef.

These are the vectorized functions of Sleef_copysignf with the same accuracy specification.

Vectorized double precision functions for determining maximum of two values

#include <sleef.h>

__m128d Sleef_fmaxd2(__m128d a, __m128d b);
__m128d Sleef_fmaxd2_sse2(__m128d a, __m128d b);
__m128d Sleef_fmaxd2_sse4(__m128d a, __m128d b);
__m128d Sleef_fmaxd2_avx2128(__m128d a, __m128d b);

__m256d Sleef_fmaxd4(__m256d a, __m256d b);
__m256d Sleef_fmaxd4_avx(__m256d a, __m256d b);
__m256d Sleef_fmaxd4_fma4(__m256d a, __m256d b);
__m256d Sleef_fmaxd4_avx2(__m256d a, __m256d b);

__m512d Sleef_fmaxd8(__m512d a, __m512d b);
__m512d Sleef_fmaxd8_avx512f(__m512d a, __m512d b);

Link with -lsleef.

These are the vectorized functions of Sleef_fmax with the same accuracy specification.

Vectorized single precision functions for determining maximum of two values

#include <sleef.h>

__m128 Sleef_fmaxf4(__m128 a, __m128 b);
__m128 Sleef_fmaxf4_sse2(__m128 a, __m128 b);
__m128 Sleef_fmaxf4_sse4(__m128 a, __m128 b);
__m128 Sleef_fmaxf4_avx2128(__m128 a, __m128 b);

__m256 Sleef_fmaxf8(__m256 a, __m256 b);
__m256 Sleef_fmaxf8_avx(__m256 a, __m256 b);
__m256 Sleef_fmaxf8_fma4(__m256 a, __m256 b);
__m256 Sleef_fmaxf8_avx2(__m256 a, __m256 b);

__m512 Sleef_fmaxf16(__m512 a, __m512 b);
__m512 Sleef_fmaxf16_avx512f(__m512 a, __m512 b);

Link with -lsleef.

These are the vectorized functions of Sleef_fmaxf with the same accuracy specification.

Vectorized double precision functions for determining minimum of two values

#include <sleef.h>

__m128d Sleef_fmind2(__m128d a, __m128d b);
__m128d Sleef_fmind2_sse2(__m128d a, __m128d b);
__m128d Sleef_fmind2_sse4(__m128d a, __m128d b);
__m128d Sleef_fmind2_avx2128(__m128d a, __m128d b);

__m256d Sleef_fmind4(__m256d a, __m256d b);
__m256d Sleef_fmind4_avx(__m256d a, __m256d b);
__m256d Sleef_fmind4_fma4(__m256d a, __m256d b);
__m256d Sleef_fmind4_avx2(__m256d a, __m256d b);

__m512d Sleef_fmind8(__m512d a, __m512d b);
__m512d Sleef_fmind8_avx512f(__m512d a, __m512d b);

Link with -lsleef.

These are the vectorized functions of Sleef_fmin with the same accuracy specification.

Vectorized single precision functions for determining minimum of two values

#include <sleef.h>

__m128 Sleef_fminf4(__m128 a, __m128 b);
__m128 Sleef_fminf4_sse2(__m128 a, __m128 b);
__m128 Sleef_fminf4_sse4(__m128 a, __m128 b);
__m128 Sleef_fminf4_avx2128(__m128 a, __m128 b);

__m256 Sleef_fminf8(__m256 a, __m256 b);
__m256 Sleef_fminf8_avx(__m256 a, __m256 b);
__m256 Sleef_fminf8_fma4(__m256 a, __m256 b);
__m256 Sleef_fminf8_avx2(__m256 a, __m256 b);

__m512 Sleef_fminf16(__m512 a, __m512 b);
__m512 Sleef_fminf16_avx512f(__m512 a, __m512 b);

Link with -lsleef.

These are the vectorized functions of Sleef_fminf with the same accuracy specification.

Vectorized double precision functions to calculate positive difference of two values

#include <sleef.h>

__m128d Sleef_fdimd2(__m128d a, __m128d b);
__m128d Sleef_fdimd2_sse2(__m128d a, __m128d b);
__m128d Sleef_fdimd2_sse4(__m128d a, __m128d b);
__m128d Sleef_fdimd2_avx2128(__m128d a, __m128d b);

__m256d Sleef_fdimd4(__m256d a, __m256d b);
__m256d Sleef_fdimd4_avx(__m256d a, __m256d b);
__m256d Sleef_fdimd4_fma4(__m256d a, __m256d b);
__m256d Sleef_fdimd4_avx2(__m256d a, __m256d b);

__m512d Sleef_fdimd8(__m512d a, __m512d b);
__m512d Sleef_fdimd8_avx512f(__m512d a, __m512d b);

Link with -lsleef.

These are the vectorized functions of Sleef_fdim with the same accuracy specification.

Vectorized single precision functions to calculate positive difference of two values

#include <sleef.h>

__m128 Sleef_fdimf4(__m128 a, __m128 b);
__m128 Sleef_fdimf4_sse2(__m128 a, __m128 b);
__m128 Sleef_fdimf4_sse4(__m128 a, __m128 b);
__m128 Sleef_fdimf4_avx2128(__m128 a, __m128 b);

__m256 Sleef_fdimf8(__m256 a, __m256 b);
__m256 Sleef_fdimf8_avx(__m256 a, __m256 b);
__m256 Sleef_fdimf8_fma4(__m256 a, __m256 b);
__m256 Sleef_fdimf8_avx2(__m256 a, __m256 b);

__m512 Sleef_fdimf16(__m512 a, __m512 b);
__m512 Sleef_fdimf16_avx512f(__m512 a, __m512 b);

Link with -lsleef.

These are the vectorized functions of Sleef_fdimf with the same accuracy specification.

Vectorized double precision functions for obtaining the next representable FP value

#include <sleef.h>

__m128d Sleef_nextafterd2(__m128d a, __m128d b);
__m128d Sleef_nextafterd2_sse2(__m128d a, __m128d b);
__m128d Sleef_nextafterd2_sse4(__m128d a, __m128d b);
__m128d Sleef_nextafterd2_avx2128(__m128d a, __m128d b);

__m256d Sleef_nextafterd4(__m256d a, __m256d b);
__m256d Sleef_nextafterd4_avx(__m256d a, __m256d b);
__m256d Sleef_nextafterd4_fma4(__m256d a, __m256d b);
__m256d Sleef_nextafterd4_avx2(__m256d a, __m256d b);

__m512d Sleef_nextafterd8(__m512d a, __m512d b);
__m512d Sleef_nextafterd8_avx512f(__m512d a, __m512d b);

Link with -lsleef.

These are the vectorized functions of Sleef_nextafter with the same accuracy specification.

Vectorized single precision functions for obtaining the next representable FP value

#include <sleef.h>

__m128 Sleef_nextafterf4(__m128 a, __m128 b);
__m128 Sleef_nextafterf4_sse2(__m128 a, __m128 b);
__m128 Sleef_nextafterf4_sse4(__m128 a, __m128 b);
__m128 Sleef_nextafterf4_avx2128(__m128 a, __m128 b);

__m256 Sleef_nextafterf8(__m256 a, __m256 b);
__m256 Sleef_nextafterf8_avx(__m256 a, __m256 b);
__m256 Sleef_nextafterf8_fma4(__m256 a, __m256 b);
__m256 Sleef_nextafterf8_avx2(__m256 a, __m256 b);

__m512 Sleef_nextafterf16(__m512 a, __m512 b);
__m512 Sleef_nextafterf16_avx512f(__m512 a, __m512 b);

Link with -lsleef.

These are the vectorized functions of Sleef_nextafterf with the same accuracy specification.

Supplementary Functions

CPU feature identification function

#include <sleef.h>

void Sleef_x86CpuID(int32_t out[4], uint32_t eax, uint32_t ecx);

Link with -lsleef.

This is a function for obtaining CPU features by the cpuid instruction available on x86 architecture. This function executes cpuid instruction by passing the values eax and ecx to the corresponding registers. The information returned in EAX, EBX, ECX, and EDX registers will be stored in array out.