Module intrinsics

Raw libdevice math intrinsics.

Note that this file was autogenerated using crude text analysis from the libdevice PDF file, therefore many of the descriptions have broken text, especially for math symbols. The link to the libdevice website is provided for every intrinsic so you can view the non-broken description.

Most of the intrinsics here have “proper” functions, corresponding f32/f64 functions are already codegenned to libdevice intrinsics by the codegen automatically. This module is mostly for exotic intrinsics that have not been added as proper functions yet.

The underlying intrinsic functions have a prefix of __nv_, however this prefix is stripped from the functions in this module just for convenience.

Functions

abs^⚠

Determine the absolute value of the 32-bit signed integer x.

acos^⚠

Calculate the principal value of the arc cosine of the input argument x.

acosf^⚠

Calculate the principal value of the arc cosine of the input argument x.

acosh^⚠

Calculate the nonnegative arc hyperbolic cosine of the input argument x.

acoshf^⚠

Calculate the nonnegative arc hyperbolic cosine of the input argument x.

asin^⚠

Calculate the principal value of the arc sine of the input argument x.

asinf^⚠

Calculate the principal value of the arc sine of the input argument x.

asinh^⚠

Calculate the arc hyperbolic sine of the input argument x.

asinhf^⚠

Calculate the arc hyperbolic sine of the input argument x.

atan^⚠

Calculate the principal value of the arc tangent of the input argument x.

atan2^⚠

Calculate the principal value of the arc tangent of the ratio of first and second input arguments / . The quadrant of the result is determined by the signs of inputs and x. y x y

atan2f^⚠

Calculate the principal value of the arc tangent of the ratio of first and second input arguments / . The quadrant of the result is determined by the signs of inputs and x. y x y

atanf^⚠

Calculate the principal value of the arc tangent of the input argument x.

atanh^⚠

Calculate the arc hyperbolic tangent of the input argument x.

atanhf^⚠

Calculate the arc hyperbolic tangent of the input argument x.

brev^⚠

Reverses the bit order of the 32 bit unsigned integer x.

brevll^⚠

Reverses the bit order of the 64 bit unsigned integer x.

byte_perm^⚠

__nv_byte_perm(x,y,s) returns a 32-bit integer consisting of four bytes from eight input bytes provided in the two input integers and , as specified by a selector, x. y s The input bytes are indexed as follows: input[0] = x<7:0> input[1] = x<15:8> input[2] = x<23:16> input[3] = x<31:24> input[4] = y<7:0> input[5] = y<15:8> input[6] = y<23:16> input[7] = y<31:24>

cbrt^⚠

Calculate the cube root of , x.

cbrtf^⚠

Calculate the cube root of , x.

ceil^⚠

Compute the smallest integer value not less than x.

ceilf^⚠

Compute the smallest integer value not less than x.

clz^⚠

Count the number of consecutive leading zero bits, starting at the most significant bit (bit 31) of x.

clzll^⚠

Count the number of consecutive leading zero bits, starting at the most significant bit (bit 63) of x.

copysign^⚠

Create a floating-point value with the magnitude and the sign of x. y

copysignf^⚠

Create a floating-point value with the magnitude and the sign of x. y

cos^⚠

Calculate the cosine of the input argument (measured in radians)x.

cosf^⚠

Calculate the cosine of the input argument (measured in radians)x.

cosh^⚠

Calculate the hyperbolic cosine of the input argument x.

coshf^⚠

Calculate the hyperbolic cosine of the input argument x.

cospi^⚠

Calculate the cosine of (measured in radians), where is the input argumentx. x

cospif^⚠

Calculate the cosine of (measured in radians), where is the input argumentx. x

dadd_rd^⚠

Adds two floating point values and in round-down (to negative infinity) modex. y

dadd_rn^⚠

Adds two floating point values and in round-to-nearest-even modex. y

dadd_ru^⚠

Adds two floating point values and in round-up (to positive infinity) modex. y

dadd_rz^⚠

Adds two floating point values and in round-towards-zero modex. y

ddiv_rd^⚠

Divides two floating point values by in round-down (to negative infinity) modex. y

ddiv_rn^⚠

Divides two floating point values by in round-to-nearest-even modex. y

ddiv_ru^⚠

Divides two floating point values by in round-up (to positive infinity) modex. y

ddiv_rz^⚠

Divides two floating point values by in round-towards-zero modex. y

dmul_rd^⚠

Multiplies two floating point values and in round-down (to negative infinity) modex. y

dmul_rn^⚠

Multiplies two floating point values and in round-to-nearest-even modex. y

dmul_ru^⚠

Multiplies two floating point values and in round-up (to positive infinity) modex. y

dmul_rz^⚠

Multiplies two floating point values and in round-towards-zero modex. y

double2float_rd^⚠

Convert the double-precision floating point value to a single-precision floating pointx value in round-down (to negative infinity) mode.

double2float_rn^⚠

Convert the double-precision floating point value to a single-precision floating pointx value in round-to-nearest-even mode.

double2float_ru^⚠

Convert the double-precision floating point value to a single-precision floating pointx value in round-up (to positive infinity) mode.

double2float_rz^⚠

Convert the double-precision floating point value to a single-precision floating pointx value in round-towards-zero mode.

double2hiint^⚠

Reinterpret the high 32 bits in the double-precision floating point value as a signedx integer.

double2int_rd^⚠

Convert the double-precision floating point value to a signed integer value in round-x down (to negative infinity) mode.

double2int_rn^⚠

Convert the double-precision floating point value to a signed integer value in round-x to-nearest-even mode.

double2int_ru^⚠

Convert the double-precision floating point value to a signed integer value in round-x up (to positive infinity) mode.

double2int_rz^⚠

Convert the double-precision floating point value to a signed integer value in round-x towards-zero mode.

double2ll_rd^⚠

Convert the double-precision floating point value to a signed 64-bit integer value inx round-down (to negative infinity) mode.

double2ll_rn^⚠

Convert the double-precision floating point value to a signed 64-bit integer value inx round-to-nearest-even mode.

double2ll_ru^⚠

Convert the double-precision floating point value to a signed 64-bit integer value inx round-up (to positive infinity) mode.

double2ll_rz^⚠

Convert the double-precision floating point value to a signed 64-bit integer value inx round-towards-zero mode.

double2loint^⚠

Reinterpret the low 32 bits in the double-precision floating point value as a signedx integer.

double2uint_rd^⚠

Convert the double-precision floating point value to an unsigned integer value inx round-down (to negative infinity) mode.

double2uint_rn^⚠

Convert the double-precision floating point value to an unsigned integer value inx round-to-nearest-even mode.

double2uint_ru^⚠

Convert the double-precision floating point value to an unsigned integer value inx round-up (to positive infinity) mode.

double2uint_rz^⚠

Convert the double-precision floating point value to an unsigned integer value inx round-towards-zero mode.

double2ull_rd^⚠

Convert the double-precision floating point value to an unsigned 64-bit integer valuex in round-down (to negative infinity) mode.

double2ull_rn^⚠

Convert the double-precision floating point value to an unsigned 64-bit integer valuex in round-to-nearest-even mode.

double2ull_ru^⚠

Convert the double-precision floating point value to an unsigned 64-bit integer valuex in round-up (to positive infinity) mode.

double2ull_rz^⚠

Convert the double-precision floating point value to an unsigned 64-bit integer valuex in round-towards-zero mode.

double_as_longlong^⚠

Reinterpret the bits in the double-precision floating point value as a signed 64-bitx integer.

drcp_rd^⚠

Compute the reciprocal of in round-down (to negative infinity) modex.

drcp_rn^⚠

Compute the reciprocal of in round-to-nearest-even modex.

drcp_ru^⚠

Compute the reciprocal of in round-up (to positive infinity) modex.

drcp_rz^⚠

Compute the reciprocal of in round-towards-zero modex.

dsqrt_rd^⚠

Compute the square root of in round-down (to negative infinity) modex.

dsqrt_rn^⚠

Compute the square root of in round-to-nearest-even modex.

dsqrt_ru^⚠

Compute the square root of in round-up (to positive infinity) modex.

dsqrt_rz^⚠

Compute the square root of in round-towards-zero modex.

erf^⚠

Calculate the value of the error function for the input argument , x.

erfc^⚠

Calculate the complementary error function of the input argument , 1 - erf( )x. x

erfcf^⚠

Calculate the complementary error function of the input argument , 1 - erf( )x. x

erfcinv^⚠

Calculate the inverse complementary error function of the input argument , for in they y interval [0, 2]. The inverse complementary error function find the value that satisfiesx the equation = erfc( ), for , and .y x

erfcinvf^⚠

Calculate the inverse complementary error function of the input argument , for in they y interval [0, 2]. The inverse complementary error function find the value that satisfiesx the equation = erfc( ), for , and .y x

erfcx^⚠

Calculate the scaled complementary error function of the input argument , x.

erfcxf^⚠

Calculate the scaled complementary error function of the input argument , x.

erff^⚠

Calculate the value of the error function for the input argument , x.

erfinv^⚠

Calculate the inverse error function of the input argument , for in the interval [-1,y y 1]. The inverse error function finds the value that satisfies the equation = erf( ), forx y x , and .

erfinvf^⚠

Calculate the inverse error function of the input argument , for in the interval [-1,y y 1]. The inverse error function finds the value that satisfies the equation = erf( ), forx y x , and .

exp^⚠

Calculate the base exponential of the input argument x.

exp2^⚠

Calculate the base 2 exponential of the input argument x.

exp2f^⚠

Calculate the base 2 exponential of the input argument x.

exp10^⚠

Calculate the base 10 exponential of the input argument x.

exp10f^⚠

Calculate the base 10 exponential of the input argument x.

expf^⚠

Calculate the base exponential of the input argument x.

expm1^⚠

Calculate the base exponential of the input argument , minus 1x.

expm1f^⚠

Calculate the base exponential of the input argument , minus 1x.

fabs^⚠

Calculate the absolute value of the input argument x.

fabsf^⚠

Calculate the absolute value of the input argument x.

fadd_rd^⚠

Compute the sum of and in round-down (to negative infinity) modex. y

fadd_rn^⚠

Compute the sum of and in round-to-nearest-even rounding modex. y

fadd_ru^⚠

Compute the sum of and in round-up (to positive infinity) modex. y

fadd_rz^⚠

Compute the sum of and in round-towards-zero modex. y

fast_cosf^⚠

Calculate the fast approximate cosine of the input argument , measured in radiansx.

fast_exp10f^⚠

Calculate the fast approximate base 10 exponential of the input argument , x.

fast_expf^⚠

Calculate the fast approximate base exponential of the input argument , x.

fast_fdividef^⚠

Calculate the fast approximate division of by x. y

fast_log2f^⚠

Calculate the fast approximate base 2 logarithm of the input argument x.

fast_log10f^⚠

Calculate the fast approximate base 10 logarithm of the input argument x.

fast_logf^⚠

Calculate the fast approximate base logarithm of the input argument x.

fast_powf^⚠

Calculate the fast approximate of , the first input argument, raised to the power of ,x y the second input argument, .

fast_sincosf^⚠

Calculate the fast approximate of sine and cosine of the first input argument x (measured in radians). The results for sine and cosine are written into the second argument, , and, respectively, third argument, . sptr zptr

fast_sinf^⚠

Calculate the fast approximate sine of the input argument , measured in radiansx.

fast_tanf^⚠

Calculate the fast approximate tangent of the input argument , measured in radiansx.

fdim^⚠

Compute the positive difference between and . The positive difference is - when x y x y x

fdimf^⚠

Compute the positive difference between and . The positive difference is - when x y x y x

fdiv_rd^⚠

Divide two floating point values by in round-down (to negative infinity) modex. y

fdiv_rn^⚠

Divide two floating point values by in round-to-nearest-even modex. y

fdiv_ru^⚠

Divide two floating point values by in round-up (to positive infinity) modex. y

fdiv_rz^⚠

Divide two floating point values by in round-towards-zero modex. y

ffs^⚠

Find the position of the first (least significant) bit set to 1 in , where the least significantx bit position is 1.

ffsll^⚠

Find the position of the first (least significant) bit set to 1 in , where the least significantx bit position is 1.

finitef^⚠

Determine whether the floating-point value is a finite valuex.

float2half_rn^⚠

Convert the single-precision float value to a half-precision floating point valuex represented in format, in round-to-nearest-even mode. unsigned short

float2int_rd^⚠

Convert the single-precision floating point value to a signed integer in round-down (tox negative infinity) mode.

float2int_rn^⚠

Convert the single-precision floating point value to a signed integer in round-to-x nearest-even mode.

float2int_ru^⚠

Convert the single-precision floating point value to a signed integer in round-up (tox positive infinity) mode.

float2int_rz^⚠

Convert the single-precision floating point value to a signed integer in round-towards-x zero mode.

float2ll_rd^⚠

Convert the single-precision floating point value to a signed 64-bit integer in round-x down (to negative infinity) mode.

float2ll_rn^⚠

Convert the single-precision floating point value to a signed 64-bit integer in round-to-x nearest-even mode.

float2ll_ru^⚠

Convert the single-precision floating point value to a signed 64-bit integer in round-upx (to positive infinity) mode.

float2ll_rz^⚠

Convert the single-precision floating point value to a signed 64-bit integer in round-x towards-zero mode.

float2uint_rd^⚠

Convert the single-precision floating point value to an unsigned integer in round-x down (to negative infinity) mode.

float2uint_rn^⚠

Convert the single-precision floating point value to an unsigned integer in round-to-x nearest-even mode.

float2uint_ru^⚠

Convert the single-precision floating point value to an unsigned integer in round-upx (to positive infinity) mode.

float2uint_rz^⚠

Convert the single-precision floating point value to an unsigned integer in round-x towards-zero mode.

float2ull_rd^⚠

Convert the single-precision floating point value to an unsigned 64-bit integer inx round-down (to negative infinity) mode.

float2ull_rn^⚠

Convert the single-precision floating point value to an unsigned 64-bit integer inx round-to-nearest-even mode.

float2ull_ru^⚠

Convert the single-precision floating point value to an unsigned 64-bit integer inx round-up (to positive infinity) mode.

float2ull_rz^⚠

Convert the single-precision floating point value to an unsigned 64-bit integer inx round-towards_zero mode.

float_as_int^⚠

Reinterpret the bits in the single-precision floating point value as a signed integerx.

floor^⚠

Calculates the largest integer value which is less than or equal to x.

floorf^⚠

Calculates the largest integer value which is less than or equal to x.

fma^⚠

Compute the value of as a single ternary operation. After computing the value to infinite precision, the value is rounded once.

fma_rd^⚠

Computes the value of as a single ternary operation, rounding the result once in round-down (to negative infinity) mode.

fma_rn^⚠

Computes the value of as a single ternary operation, rounding the result once in round-to-nearest-even mode.

fma_ru^⚠

Computes the value of as a single ternary operation, rounding the result once in round-up (to positive infinity) mode.

fma_rz^⚠

Computes the value of as a single ternary operation, rounding the result once in round-towards-zero mode.

fmaf^⚠

Compute the value of as a single ternary operation. After computing the value to infinite precision, the value is rounded once.

fmaf_rd^⚠

Computes the value of as a single ternary operation, rounding the result once in round-down (to negative infinity) mode.

fmaf_rn^⚠

Computes the value of as a single ternary operation, rounding the result once in round-to-nearest-even mode.

fmaf_ru^⚠

Computes the value of as a single ternary operation, rounding the result once in round-up (to positive infinity) mode.

fmaf_rz^⚠

Computes the value of as a single ternary operation, rounding the result once in round-towards-zero mode.

fmax^⚠

Determines the maximum numeric value of the arguments and . Treats NaNx y arguments as missing data. If one argument is a NaN and the other is legitimate numeric value, the numeric value is chosen.

fmaxf^⚠

Determines the maximum numeric value of the arguments and . Treats NaNx y arguments as missing data. If one argument is a NaN and the other is legitimate numeric value, the numeric value is chosen.

fmin^⚠

Determines the minimum numeric value of the arguments and . Treats NaNx y arguments as missing data. If one argument is a NaN and the other is legitimate numeric value, the numeric value is chosen.

fminf^⚠

Determines the minimum numeric value of the arguments and . Treats NaNx y arguments as missing data. If one argument is a NaN and the other is legitimate numeric value, the numeric value is chosen.

fmod^⚠

Calculate the floating-point remainder of / . The absolute value of the computedx y value is always less than absolute value and will have the same sign as .y’s x

fmodf^⚠

Calculate the floating-point remainder of / . The absolute value of the computedx y value is always less than absolute value and will have the same sign as .y’s x

fmul_rd^⚠

Compute the product of and in round-down (to negative infinity) modex. y

fmul_rn^⚠

Compute the product of and in round-to-nearest-even modex. y

fmul_ru^⚠

Compute the product of and in round-up (to positive infinity) modex. y

fmul_rz^⚠

Compute the product of and in round-towards-zero modex. y

frcp_rd^⚠

Compute the reciprocal of in round-down (to negative infinity) modex.

frcp_rn^⚠

Compute the reciprocal of in round-to-nearest-even modex.

frcp_ru^⚠

Compute the reciprocal of in round-up (to positive infinity) modex.

frcp_rz^⚠

Compute the reciprocal of in round-towards-zero modex.

frexp^⚠

Decompose the floating-point value into a component for the normalized fractionx m element and another term for the exponent. The absolute value of will be greater n m than or equal to 0.5 and less than 1.0 or it will be equal to 0; . The integer exponent will be stored in the location to which points. n nptr

frexpf^⚠

Decompose the floating-point value into a component for the normalized fractionx m element and another term for the exponent. The absolute value of will be greater n m than or equal to 0.5 and less than 1.0 or it will be equal to 0; . The integer exponent will be stored in the location to which points. n nptr

frsqrt_rn^⚠

Compute the reciprocal square root of in round-to-nearest-even modex.

fsqrt_rd^⚠

Compute the square root of in round-down (to negative infinity) modex.

fsqrt_rn^⚠

Compute the square root of in round-to-nearest-even modex.

fsqrt_ru^⚠

Compute the square root of in round-up (to positive infinity) modex.

fsqrt_rz^⚠

Compute the square root of in round-towards-zero modex.

fsub_rd^⚠

Compute the difference of and in round-down (to negative infinity) modex. y

fsub_rn^⚠

Compute the difference of and in round-to-nearest-even rounding modex. y

fsub_ru^⚠

Compute the difference of and in round-up (to positive infinity) modex. y

fsub_rz^⚠

Compute the difference of and in round-towards-zero modex. y

hadd^⚠

Compute average of signed input arguments and as ( + ) >> 1, avoiding overflowx y x y in the intermediate sum.

half2float^⚠

Convert the half-precision floating point value represented in x unsigned short format to a single-precision floating point value.

hiloint2double^⚠

Reinterpret the integer value of as the high 32 bits of a double-precision floating hi point value and the integer value of as the low 32 bits of the same double-precision lo floating point value.

hypot^⚠

Calculate the length of the hypotenuse of a right triangle whose two sides have lengths x and without undue overflow or underflow.y

hypotf^⚠

Calculate the length of the hypotenuse of a right triangle whose two sides have lengths x and without undue overflow or underflow.y

ilogb^⚠

Calculates the unbiased integer exponent of the input argument x.

ilogbf^⚠

Calculates the unbiased integer exponent of the input argument x.

int2double_rn^⚠

Convert the signed integer value to a double-precision floating point valuex.

int2float_rd^⚠

Convert the signed integer value to a single-precision floating point value in round-x down (to negative infinity) mode.

int2float_rn^⚠

Convert the signed integer value to a single-precision floating point value in round-to-x nearest-even mode.

int2float_ru^⚠

Convert the signed integer value to a single-precision floating point value in round-upx (to positive infinity) mode.

int2float_rz^⚠

Convert the signed integer value to a single-precision floating point value in round-x towards-zero mode.

int_as_float^⚠

Reinterpret the bits in the signed integer value as a single-precision floating pointx value.

isfinited^⚠

Determine whether the floating-point value is a finite value (zero, subnormal, orx normal and not infinity or NaN).

isinfd^⚠

Determine whether the floating-point value is an infinite value (positive or negative)x.

isinff^⚠

Determine whether the floating-point value is an infinite value (positive or negative)x.

isnand^⚠

Determine whether the floating-point value is a NaNx.

isnanf^⚠

Determine whether the floating-point value is a NaNx.

j0^⚠

Calculate the value of the Bessel function of the first kind of order 0 for the input argument , x.

j0f^⚠

Calculate the value of the Bessel function of the first kind of order 0 for the input argument , x.

j1^⚠

Calculate the value of the Bessel function of the first kind of order 1 for the input argument , x.

j1f^⚠

Calculate the value of the Bessel function of the first kind of order 1 for the input argument , x.

jn^⚠

Calculate the value of the Bessel function of the first kind of order for the input n argument , x.

jnf^⚠

Calculate the value of the Bessel function of the first kind of order for the input n argument , x.

ldexp^⚠

Calculate the value of of the input arguments and x. exp

ldexpf^⚠

Calculate the value of of the input arguments and x. exp

lgamma^⚠

Calculate the natural logarithm of the absolute value of the gamma function of the input argument , namely the value of x

lgammaf^⚠

Calculate the natural logarithm of the absolute value of the gamma function of the input argument , namely the value of x

ll2double_rd^⚠

Convert the signed 64-bit integer value to a double-precision floating point value inx round-down (to negative infinity) mode.

ll2double_rn^⚠

Convert the signed 64-bit integer value to a double-precision floating point value inx round-to-nearest-even mode.

ll2double_ru^⚠

Convert the signed 64-bit integer value to a double-precision floating point value inx round-up (to positive infinity) mode.

ll2double_rz^⚠

Convert the signed 64-bit integer value to a double-precision floating point value inx round-towards-zero mode.

ll2float_rd^⚠

Convert the signed integer value to a single-precision floating point value in round-x down (to negative infinity) mode.

ll2float_rn^⚠

Convert the signed 64-bit integer value to a single-precision floating point value inx round-to-nearest-even mode.

ll2float_ru^⚠

Convert the signed integer value to a single-precision floating point value in round-upx (to positive infinity) mode.

ll2float_rz^⚠

Convert the signed integer value to a single-precision floating point value in round-x towards-zero mode.

llabs^⚠

Determine the absolute value of the 64-bit signed integer x.

llmax^⚠

Determine the maximum value of the two 64-bit signed integers and x. y

llmin^⚠

Determine the minimum value of the two 64-bit signed integers and x. y

llrint^⚠

Round to the nearest integer value, with halfway cases rounded towards zero. If thex result is outside the range of the return type, the result is undefined.

llrintf^⚠

Round to the nearest integer value, with halfway cases rounded towards zero. If thex result is outside the range of the return type, the result is undefined.

llround^⚠

Round to the nearest integer value, with halfway cases rounded away from zero. If thex result is outside the range of the return type, the result is undefined.

llroundf^⚠

Round to the nearest integer value, with halfway cases rounded away from zero. If thex result is outside the range of the return type, the result is undefined.

log^⚠

Calculate the base logarithm of the input argument x.

log2^⚠

Calculate the base 2 logarithm of the input argument x.

log1p^⚠

Calculate the value of of the input argument x.

log1pf^⚠

Calculate the value of of the input argument x.

log2f^⚠

Calculate the base 2 logarithm of the input argument x.

log10^⚠

Calculate the base 10 logarithm of the input argument x.

log10f^⚠

Calculate the base 10 logarithm of the input argument x.

logb^⚠

Calculate the floating point representation of the exponent of the input argument x.

logbf^⚠

Calculate the floating point representation of the exponent of the input argument x.

logf^⚠

Calculate the base logarithm of the input argument x.

longlong_as_double^⚠

Reinterpret the bits in the 64-bit signed integer value as a double-precision floatingx point value.

max^⚠

Determine the maximum value of the two 32-bit signed integers and x. y

min^⚠

Determine the minimum value of the two 32-bit signed integers and x. y

modf^⚠

Break down the argument into fractional and integral parts. The integral part is storedx in the argument . Fractional and integral parts are given the same sign as the iptr argument x.

modff^⚠

Break down the argument into fractional and integral parts. The integral part is storedx in the argument . Fractional and integral parts are given the same sign as the iptr argument x.

mul24^⚠

Calculate the least significant 32 bits of the product of the least significant 24 bits of x and . The high order 8 bits of and are ignored.y x y

mul64hi^⚠

Calculate the most significant 64 bits of the 128-bit product * , where and are 64-x y x y bit integers.

mulhi^⚠

Calculate the most significant 32 bits of the 64-bit product * , where and are 32-bitx y x y integers.

nan^⚠

Return a representation of a quiet NaN. Argument selects one of the possible tagp representations.

nanf^⚠

Return a representation of a quiet NaN. Argument selects one of the possible tagp representations.

nearbyint^⚠

Round argument to an integer value in double precision floating-point formatx.

nearbyintf^⚠

Round argument to an integer value in double precision floating-point formatx.

nextafter^⚠

Calculate the next representable double-precision floating-point value following inx the direction of . For example, if is greater than , nextafter() returns the smallesty y x representable number greater than x

nextafterf^⚠

Calculate the next representable double-precision floating-point value following inx the direction of . For example, if is greater than , nextafter() returns the smallesty y x representable number greater than x

normcdf^⚠

Calculate the cumulative distribution function of the standard normal distribution for input argument , .y

normcdff^⚠

Calculate the cumulative distribution function of the standard normal distribution for input argument , .y

normcdfinv^⚠

Calculate the inverse of the standard normal cumulative distribution function for input argument , . The function is defined for input values in the interval .y

normcdfinvf^⚠

Calculate the inverse of the standard normal cumulative distribution function for input argument , . The function is defined for input values in the interval .y

popc^⚠

Count the number of bits that are set to 1 in x.

popcll^⚠

Count the number of bits that are set to 1 in x.

pow^⚠

Calculate the value of to the power of x y

powf^⚠

Calculate the value of to the power of x y

powi^⚠

Calculate the value of to the power of x y

powif^⚠

Calculate the value of to the power of x. y

rcbrt^⚠

Calculate reciprocal cube root function of x

rcbrtf^⚠

Calculate reciprocal cube root function of x

remainder^⚠

Compute double-precision floating-point remainder of dividing by for nonzero . r x y y Thus . The value is the integer value nearest . In the case when , n the even value is chosen. n

remainderf^⚠

Compute double-precision floating-point remainder of dividing by for nonzero . r x y y Thus . The value is the integer value nearest . In the case when , n the even value is chosen. n

remquo^⚠

Compute a double-precision floating-point remainder in the same way as the remainder() function. Argument returns part of quotient upon division of by . quo x y Value has the same sign as and may not be the exact quotient but agrees with the quo exact quotient in the low order 3 bits.

remquof^⚠

Compute a double-precision floating-point remainder in the same way as the remainder() function. Argument returns part of quotient upon division of by . quo x y Value has the same sign as and may not be the exact quotient but agrees with the quo exact quotient in the low order 3 bits.

rhadd^⚠

Compute average of signed input arguments and as ( + + 1 ) >> 1, avoidingx y x y overflow in the intermediate sum.

rint^⚠

Round to the nearest integer value in floating-point format, with halfway casesx rounded to the nearest even integer value.

rintf^⚠

Round to the nearest integer value in floating-point format, with halfway casesx rounded to the nearest even integer value.

round^⚠

Round to the nearest integer value in floating-point format, with halfway casesx rounded away from zero.

roundf^⚠

Round to the nearest integer value in floating-point format, with halfway casesx rounded away from zero.

rsqrt^⚠

Calculate the reciprocal of the nonnegative square root of , x.

rsqrtf^⚠

Calculate the reciprocal of the nonnegative square root of , x.

sad^⚠

Calculate , the 32-bit sum of the third argument plus and the absolute value z of the difference between the first argument, , and second argument, x. y Inputs and are signed 32-bit integers, input is a 32-bit unsigned integerx. y z

saturatef^⚠

Clamp the input argument to be within the interval [+0.0, 1.0]x.

scalbn^⚠

Scale by by efficient manipulation of the floating-point exponentx.

scalbnf^⚠

Scale by by efficient manipulation of the floating-point exponentx.

signbitd^⚠

Determine whether the floating-point value is negativex.

signbitf^⚠

Determine whether the floating-point value is negativex.

sin^⚠

Calculate the sine of the input argument (measured in radians)x.

sincos^⚠

Calculate the sine and cosine of the first input argument (measured in radians). Thex results for sine and cosine are written into the second argument, , and, respectively, sptr third argument, . zptr

sincosf^⚠

Calculate the sine and cosine of the first input argument (measured in radians). Thex results for sine and cosine are written into the second argument, , and, respectively, sptr third argument, . zptr

sincospi^⚠

Calculate the sine and cosine of the first input argument, (measured in radians),x . The results for sine and cosine are written into the second argument, , and, sptr respectively, third argument, . zptr

sincospif^⚠

Calculate the sine and cosine of the first input argument, (measured in radians),x . The results for sine and cosine are written into the second argument, , and, sptr respectively, third argument, . zptr

sinf^⚠

Calculate the sine of the input argument (measured in radians)x.

sinh^⚠

Calculate the hyperbolic sine of the input argument x.

sinhf^⚠

Calculate the hyperbolic sine of the input argument x.

sinpi^⚠

Calculate the sine of (measured in radians), where is the input argumentx. x

sinpif^⚠

Calculate the sine of (measured in radians), where is the input argumentx. x

sqrt^⚠

Calculate the nonnegative square root of , x.

sqrtf^⚠

Calculate the nonnegative square root of , x.

tan^⚠

Calculate the tangent of the input argument (measured in radians)x.

tanf^⚠

Calculate the tangent of the input argument (measured in radians)x.

tanh^⚠

Calculate the hyperbolic tangent of the input argument x.

tanhf^⚠

Calculate the hyperbolic tangent of the input argument x.

tgamma^⚠

Calculate the gamma function of the input argument , namely the value of x.

tgammaf^⚠

Calculate the gamma function of the input argument , namely the value of x.

trunc^⚠

Round to the nearest integer value that does not exceed in magnitudex. x

truncf^⚠

Round to the nearest integer value that does not exceed in magnitudex. x

uhadd^⚠

Compute average of unsigned input arguments and as ( + ) >> 1, avoidingx y x y overflow in the intermediate sum.

uint2double_rn^⚠

Convert the unsigned integer value to a double-precision floating point valuex.

uint2float_rd^⚠

Convert the unsigned integer value to a single-precision floating point value in round-x down (to negative infinity) mode.

uint2float_rn^⚠

Convert the unsigned integer value to a single-precision floating point value in round-x to-nearest-even mode.

uint2float_ru^⚠

Convert the unsigned integer value to a single-precision floating point value in round-x up (to positive infinity) mode.

uint2float_rz^⚠

Convert the unsigned integer value to a single-precision floating point value in round-x towards-zero mode.

ull2double_rd^⚠

Convert the unsigned 64-bit integer value to a double-precision floating point value inx round-down (to negative infinity) mode.

ull2double_rn^⚠

Convert the unsigned 64-bit integer value to a double-precision floating point value inx round-to-nearest-even mode.

ull2double_ru^⚠

Convert the unsigned 64-bit integer value to a double-precision floating point value inx round-up (to positive infinity) mode.

ull2double_rz^⚠

Convert the unsigned 64-bit integer value to a double-precision floating point value inx round-towards-zero mode.

ull2float_rd^⚠

Convert the unsigned integer value to a single-precision floating point value in round-x down (to negative infinity) mode.

ull2float_rn^⚠

Convert the unsigned integer value to a single-precision floating point value in round-x to-nearest-even mode.

ull2float_ru^⚠

Convert the unsigned integer value to a single-precision floating point value in round-x up (to positive infinity) mode.

ull2float_rz^⚠

Convert the unsigned integer value to a single-precision floating point value in round-x towards-zero mode.

ullmax^⚠

Determine the maximum value of the two 64-bit unsigned integers and x. y

ullmin^⚠

Determine the minimum value of the two 64-bit unsigned integers and x. y

umax^⚠

Determine the maximum value of the two 32-bit unsigned integers and x. y

umin^⚠

Determine the minimum value of the two 32-bit unsigned integers and x. y

umul24^⚠

Calculate the least significant 32 bits of the product of the least significant 24 bits of x and . The high order 8 bits of and are ignored.y x y

umul64hi^⚠

Calculate the most significant 64 bits of the 128-bit product * , where and are 64-x y x y bit unsigned integers.

umulhi^⚠

Calculate the most significant 32 bits of the 64-bit product * , where and are 32-bitx y x y unsigned integers.

urhadd^⚠

Compute average of unsigned input arguments and as ( + + 1 ) >> 1, avoidingx y x y overflow in the intermediate sum.

usad^⚠

Calculate , the 32-bit sum of the third argument plus and the absolute value z of the difference between the first argument, , and second argument, x. y Inputs , , and are unsigned 32-bit integersx. y z

y0^⚠

Calculate the value of the Bessel function of the second kind of order 0 for the input argument , x.

y0f^⚠

Calculate the value of the Bessel function of the second kind of order 0 for the input argument , x.

y1^⚠

Calculate the value of the Bessel function of the second kind of order 1 for the input argument , x.

y1f^⚠

Calculate the value of the Bessel function of the second kind of order 1 for the input argument , x.

yn^⚠

Calculate the value of the Bessel function of the second kind of order for the input n argument , x.

ynf^⚠

Calculate the value of the Bessel function of the second kind of order for the input n argument , x.