# std::bit_floor

> C++ Atlas: `cppatlas.com/w/cpp/numeric/bit_floor` | Source: https://en.cppreference.com/w/cpp/numeric/bit_floor | Domain: Numerics | Standards: C++20

If x is not zero, calculates the largest integral power of two that is not greater than x. If x is zero, returns zero.

## Official Specification Summary (cppreference)

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

template< class T >
constexpr T bit_floor( T x ) noexcept; // (since C++20)
```

If `x` is not zero, calculates the largest integral power of two that is not greater than `x`. If `x` is zero, returns zero.

This overload participates in overload resolution only if `T` is an unsigned integer type (that is, `unsigned char`, `unsigned short`, `unsigned int`, `unsigned long`, `unsigned long long`, or an extended unsigned integer type).

### Parameters

- `x` - unsigned integer value

### Return value

Zero if `x` is zero; otherwise, the largest integral power of two that is not greater than `x`.

### Notes

Prior to P1956R1, the proposed name for this function template was `floor2`.

| Feature-test macro | Value | Std | Feature |
| --- | --- | --- | --- |
| `__cpp_lib_int_pow2` | `202002L` | (C++20) | Integral power-of-2 operations |

### Possible implementation

```cpp
template<std::unsigned_integral T>
    requires !std::same_as<T, bool> && !std::same_as<T, char> &&
             !std::same_as<T, char8_t> && !std::same_as<T, char16_t> &&
             !std::same_as<T, char32_t> && !std::same_as<T, wchar_t>
constexpr T bit_floor(T x) noexcept
{
    if (x != 0)
        return T{1} << (std::bit_width(x) - 1);
    return 0;
}
```

### Example

```cpp
#include <bit>
#include <bitset>
#include <iostream>

int main()
{
    using bin = std::bitset<8>;
    for (unsigned x{}; x != 012; ++x)
        std::cout << "bit_floor( " << bin(x) << " ) = "
                  << bin(std::bit_floor(x)) << '\n';
}
```

Output:

```text
bit_floor( 00000000 ) = 00000000
bit_floor( 00000001 ) = 00000001
bit_floor( 00000010 ) = 00000010
bit_floor( 00000011 ) = 00000010
bit_floor( 00000100 ) = 00000100
bit_floor( 00000101 ) = 00000100
bit_floor( 00000110 ) = 00000100
bit_floor( 00000111 ) = 00000100
bit_floor( 00001000 ) = 00001000
bit_floor( 00001001 ) = 00001000
```

### See also

- [`bit_ceil`](https://en.cppreference.com/w/cpp/numeric/bit_ceil) (C++20): finds the smallest integral power of 2 not less than the given value (function template)
- [`rotr`](https://en.cppreference.com/w/cpp/numeric/rotr) (C++20): computes the result of bitwise right-rotation (function template)
- [`bit_width`](https://en.cppreference.com/w/cpp/numeric/bit_width) (C++20): finds the smallest number of bits needed to represent the given value (function template)
- [`has_single_bit`](https://en.cppreference.com/w/cpp/numeric/has_single_bit) (C++20): checks if a number is an integral power of 2 (function template)

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