fabs, fabsf, fabsl, fabsd32, fabsd64, fabsd128

Defined in header <math.h>
float       fabsf( float arg );
(1) (since C99)
double      fabs( double arg );
long double fabsl( long double arg );
(3) (since C99)
_Decimal32  fabsd32( _Decimal32 arg );
(4) (since C23)
_Decimal64  fabsd64( _Decimal64 arg );
(5) (since C23)
_Decimal128 fabsd128( _Decimal128 arg );
(6) (since C23)
Defined in header <tgmath.h>
#define fabs( arith )
(7) (since C99)
1-6) Computes the absolute value of a floating point value arg.

The functions with decimal floating point parameters are declared if and only the implementation predefines __STDC_IEC_60559_DFP__ (i.e. the implementation supports decimal floating point numbers).

(since C23)
7) Type-generic macro: If the argument has type _Decimal128, _Decimal64, _Decimal32, (since C23)long double, double, or float, fabsd128, fabsd64, fabsd32, (since C23)fabsl, fabs, or fabsf is called, respectively. Otherwise, if the argument has integer type, fabs is called. Otherwise, if the argument is complex, then the macro invokes the corresponding complex function (cabsf, cabs, cabsl). Otherwise, the behavior is undefined.


arg - floating point value
arith - floating point or integer value

Return value

If successful, returns the absolute value of arg (\(\small |arg| \)|arg|). The value returned is exact and does not depend on any rounding modes.

Error handling

This function is not subject to any of the error conditions specified in math_errhandling.

If the implementation supports IEEE floating-point arithmetic (IEC 60559),

  • If the argument is ±0, +0 is returned
  • If the argument is ±∞, +∞ is returned
  • If the argument is NaN, NaN is returned


#include <stdio.h>
#include <math.h>
#define PI 3.14159
/* This numerical integration assumes all area is positive. */
double integrate(double a, double b, /* assume a < b */
                 double f(double),
                 unsigned steps) { /* assume steps > 0 */
    const double dx = (b-a)/steps;
    double sum = 0.0;
    for (double x = a; x < b; x += dx)
        sum += fabs(f(x));
    return dx*sum;
int main(void)
    printf("fabs(+3) = %f\n", fabs(+3.0));
    printf("fabs(-3) = %f\n", fabs(-3.0));
    // special values
    printf("fabs(-0) = %f\n", fabs(-0.0));
    printf("fabs(-Inf) = %f\n", fabs(-INFINITY));
    printf("%f\n", integrate(-PI,+PI,sin,10*1000));


fabs(+3) = 3.000000
fabs(-3) = 3.000000
fabs(-0) = 0.000000
fabs(-Inf) = inf


  • C17 standard (ISO/IEC 9899:2018):
    • The fabs functions (p: 181)
    • 7.25 Type-generic math <tgmath.h> (p: 272-273)
    • F.10.4.2 The fabs functions (p: 382)
  • C11 standard (ISO/IEC 9899:2011):
    • The fabs functions (p: 248)
    • 7.25 Type-generic math <tgmath.h> (p: 373-375)
    • F.10.4.2 The fabs functions (p: 524)
  • C99 standard (ISO/IEC 9899:1999):
    • The fabs functions (p: 228-229)
    • 7.22 Type-generic math <tgmath.h> (p: 335-337)
    • F.9.4.2 The fabs functions (p: 460)
  • C89/C90 standard (ISO/IEC 9899:1990):
    • The fabs function

See also

computes absolute value of an integral value (\(\small{|x|}\)|x|)
produces a value with the magnitude of a given value and the sign of another given value
checks if the given number is negative
(function macro)
computes the magnitude of a complex number
C++ documentation for fabs

© cppreference.com
Licensed under the Creative Commons Attribution-ShareAlike Unported License v3.0.