# std::fdim, std::fdimf, std::fdiml

> C++ Atlas: `cppatlas.com/w/cpp/numeric/math/fdim` | Source: https://en.cppreference.com/w/cpp/numeric/math/fdim | Domain: Numerics | Standards: C++11, C++23, C++26

If successful, returns the positive difference between x and y.

## Official Specification Summary (cppreference)

```cpp
// Defined in header <cmath>

float       fdim ( float x, float y );
double      fdim ( double x, double y );
long double fdim ( long double x, long double y ); // (1) (until C++23)

constexpr /*floating-point-type*/
            fdim ( /*floating-point-type*/ x,
                   /*floating-point-type*/ y ); // (1) (since C++23)

float       fdimf( float x, float y ); // (2) (since C++11) (constexpr since C++23)

long double fdiml( long double x, long double y ); // (3) (since C++11) (constexpr since C++23)

// SIMD overload (since C++26)

// Defined in header <simd>

template< class V0, class V1 >
constexpr /*math-common-simd-t*/<V0, V1>
            fdim ( const V0& v_x, const V1& v_y ); // (S) (since C++26)

// Additional overloads (since C++11)

// Defined in header <cmath>

template< class Integer >
double      fdim ( Integer x, Integer y ); // (A) (constexpr since C++23)
```

1-3) Returns the positive difference between `x` and `y`, that is, if `x > y`, returns `x - y`, otherwise (i.e. if `x <= y`) returns `+0`. The library provides overloads of `std::fdim` for all cv-unqualified floating-point types as the type of the parameters. (since C++23)

(since C++26):

S) The SIMD overload performs an element-wise `std::fdim` on `v_x`and `v_y`.

  (See `math-common-simd-t` for its definition.)

A) Additional overloads are provided for all integer types, which are treated as `double`. (since C++11)

### Parameters

- `x, y` - floating-point or integer values

### Return value

If successful, returns the positive difference between `x` and `y`.

If a range error due to overflow occurs, `+HUGE_VAL`, `+HUGE_VALF`, or `+HUGE_VALL` is returned.

If a range error due to underflow occurs, the correct value (after rounding) is returned.

### Error handling

Errors are reported as specified in `math_errhandling`.

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

- If either argument is NaN, NaN is returned.

### Notes

Equivalent to `std::fmax(x - y, 0)`, except for the NaN handling requirements.

The additional overloads are not required to be provided exactly as (A). They only need to be sufficient to ensure that for their first argument `num1` and second argument `num2`:

(until C++23):

- If `num1` or `num2` has type `long double`, then `std::fdim(num1, num2)` has the same effect as `std::fdim(static_cast<long double>(num1), static_cast<long double>(num2))`.
- Otherwise, if `num1` and/or `num2` has type `double` or an integer type, then `std::fdim(num1, num2)` has the same effect as `std::fdim(static_cast<double>(num1), static_cast<double>(num2))`.
- Otherwise, if `num1` or `num2` has type `float`, then `std::fdim(num1, num2)` has the same effect as `std::fdim(static_cast<float>(num1), static_cast<float>(num2))`.

(since C++23):

If `num1` and `num2` have arithmetic types, then `std::fdim(num1, num2)` has the same effect as `std::fdim(static_cast</*common-floating-point-type*/>(num1), static_cast</*common-floating-point-type*/>(num2))`, where `/*common-floating-point-type*/` is the floating-point type with the greatest floating-point conversion rank and greatest floating-point conversion subrank between the types of `num1` and `num2`, arguments of integer type are considered to have the same floating-point conversion rank as `double`.

If no such floating-point type with the greatest rank and subrank exists, then overload resolution does not result in a usable candidate from the overloads provided.

### Example

```cpp
#include <cerrno>
#include <cfenv>
#include <cmath>
#include <cstring>
#include <iostream>

#ifndef __GNUC__
#pragma STDC FENV_ACCESS ON
#endif

int main()
{
    std::cout << "fdim(4, 1) = " << std::fdim(4, 1) << '\n'
              << "fdim(1, 4) = " << std::fdim(1, 4) << '\n'
              << "fdim(4,-1) = " << std::fdim(4, -1) << '\n'
              << "fdim(1,-4) = " << std::fdim(1, -4) << '\n';

    // error handling
    errno = 0;
    std::feclearexcept(FE_ALL_EXCEPT);

    std::cout << "fdim(1e308, -1e308) = " << std::fdim(1e308, -1e308) << '\n';

    if (errno == ERANGE)
        std::cout << "    errno == ERANGE: " << std::strerror(errno) << '\n';
    if (std::fetestexcept(FE_OVERFLOW))
        std::cout << "    FE_OVERFLOW raised\n";
}
```

Output:

```text
fdim(4, 1) = 3
fdim(1, 4) = 0
fdim(4,-1) = 5
fdim(1,-4) = 5
fdim(1e308, -1e308) = inf
    errno == ERANGE: Numerical result out of range
    FE_OVERFLOW raised
```

### See also

- [`abs(int)`](https://en.cppreference.com/w/cpp/numeric/math/abs), [`labs`](https://en.cppreference.com/w/cpp/numeric/math/abs), [`llabs`](https://en.cppreference.com/w/cpp/numeric/math/abs) (C++11): computes absolute value of an integral value (|x|) (function)
- [`fmax`](https://en.cppreference.com/w/cpp/numeric/math/fmax) (C++11), [`fmaxf`](https://en.cppreference.com/w/cpp/numeric/math/fmax) (C++11), [`fmaxl`](https://en.cppreference.com/w/cpp/numeric/math/fmax) (C++11): larger of two floating-point values (function)
- [C documentation for `fdim`](https://en.cppreference.com/w/c/numeric/math/fdim)

---
Attribution: content from https://en.cppreference.com/w/cpp/numeric/math/fdim (cppreference.com), licensed under CC-BY-SA 3.0 and GFDL. Modified: converted to markdown, study notes added. Offline snapshot of 9 February 2025.