Setting all bits in range [m,n] to 1

Asked Jun 13 '18 at 19:58

Active Jun 13 '18 at 20:08

Viewed 59 times

Im using the code in This Link to generate a uint whose bit are all set to 1 in a desired range of indeces.

eg:

#include <iostream>
#include <string>

int main ()
{
    // range is [32,5] so i want [1 1 1......1 0 0 0 0 0]
    uint mask = ((1 << 32) - 1) ^ ((1 << (5 - 1)) - 1);
    std::cout << mask << std::endl;
}

The output is 4294967280, which contains the desired array of bits according to this site.

But as soon as i do the following:

#include <iostream>
#include <string>

int main () 
{
    uint stop = 32;
    uint start = 5;
    uint mask = ((1 << stop) - 1) ^ ((1 << (start - 1)) - 1);
    std::cout << mask << std::endl; 
}

It outputs 15. Uuuuh what's is going on? are the 32 and 5 in the first example a different type that gets casted when assigning to the uint mask? how do i fix this?

edited Jun 13 '18 at 20:01

asked Jun 13 '18 at 19:58

user2255757

Overflow? You only have 32 bits – JVApen Jun 13 '18 at 20:02
This depends on the size of your `uint` (which isn't a c++ type, I'll assume you meant `unsigned int`). If it's 32 bit, then `1<<32` is an overflow. – François Andrieux Jun 13 '18 at 20:07
why do you subtract 1 in case of 5 but not in case of 32? – phuclv Jun 14 '18 at 00:54
if you dont subtract by 1 the last (from left to right) of the bits you want 1 is actually 0 – user2255757 Jun 14 '18 at 06:51

Setting all bits in range [m,n] to 1

0 Answers0