如何使用6位存储值？

Question

My data unit (a network packet header) i am currently working on has 2 flags in its definition, stored in a byte field and accessed via bitwise operators. Unfortunately, i need only 2 bits and thinking what i can do with other 6 bits? Can i use them to store number?

Can i use them to store some internal state code (value range smaller than char ?) and do not just waste them.

Is there any data types smaller than byte and how can i use them in C++? If no, should i waste those bits and left them without meaning?

Answer 1

You could use a bit field, as described here .

Adapted from that page:

#include <iostream>
struct S {
 // 6-bit unsigned field,
 // allowed values are 0...63
 unsigned int b : 6;
};
int main()
{
    S s = {7};
    ++s.b;
    std::cout << s.b << '\n'; // output: 8
}

Answer 2

In C++, there is no datatype smaller than a char , which is - by definition - one byte. However, you do not need a dedicated datatype to access the bits of a value. Bitwise logic and Bitwise Shift operators are sufficient.

If you cannot live with sacrificing 6 bits (this is assuming 8-bit bytes) you might want to consider the std::vector<bool> specialization. Note, though, that there are a number of restrictions and differences to a regular std::vector .

Another option to make individual (consecutive) bits of a datatype accessible by name is to use bit fields :

struct S {
    unsigned int flags : 2;
    unsigned int state : 6;
};
static_assert( sizeof( S ) == 1, "Packing is implementation-defined." );

This declares a structure that can hold two pieces of information: flags and state , which occupy 2 and 6 bits, respectively. Adjacent bit fields are usually packed together (although this behavior is implementation-defined).