Re: [PATCH 1/5]: DCCP Fix use of invalid loss intervals

[Eddie Kohler <kohler@cs.ucla.edu>]

But the two bit patterns are the same, and the comparison will promote 
them both to unsigned.

Try running the following program:

int main (int c, char **v)
{
     if (~0 != ~0U)
	printf("not equal as unknown");
     if ((int) ~0 != (int) ~0U)
	printf("not equal as ints");
     if ((unsigned) ~0 != (unsigned) ~0U)
	printf("not equal as unsigneds");
     if ((int) ~0 != (unsigned) ~0U)
	printf("not equal as int/unsigned");
     if ((unsigned) ~0 != (int) ~0U)
	printf("not equal as unsigned/int");
}

It prints nothing.

If one of the two numbers is 64-bit, your analysis works.  Maybe I'm 
missing something but I don't think so...
Eddie



Ian McDonald wrote:
> On 1/5/07, Eddie Kohler <kohler@cs.ucla.edu> wrote:
>> Ian (catching up slowly slowly), here is a nit as nitty as they come.
>>
>> This diff seems strange to me, since ~ actually does the same thing on
>> integers and unsigned integers.  (This code:
>>
>>      printf("%u %u\n", ~0, ~0U);
>>
>> will print the same thing twice.)
>>
>> Perhaps dccplih_interval is a 64-bit number?  In which case you want to
>> say something like ~0ULL?
>>
>> Eddie
> 
> Printing gives them the same result as you are using a %u mask. If you
> do it with a %d mask you will get a different result.
> 
> And that is the issue dccp_lih_interval is unsigned 32 bit and ~0 is a
> signed number and is large negative and they therefore can't be equal.
> 
> Ian