Re: [PATCH] Remove pointless <0 comparison for unsigned variable infs/fcntl.c

[Jesper Juhl]

On Tue, 23 Nov 2004, Timur Tabi wrote:

> Linus Torvalds wrote:
> 
> > which might warn on an architecture where "pid_t" is just sixteen bits wide.
> > Does that make the code wrong? Hell no.
> 
> Wouldn't something like "sizeof(pid_t) > 2" be a better test?  It certainly
> would be a lot easier to understand than comparing with 0xffff.
> 
That was not the point of the example Linus gave.
The example Linus gave was a function taking a pid_t argument and then 
comparing the value of the argument passed against 0xffff - the /value/ of 
the pid_t argument passed, not the size of the datatype.

        int fn(pid_t a)
        {
                if (a > 0xffff)
                        ...
        }

if pid_t is 16 bit, then the value can never be greater than 0xffff but, 
if pid_t is greater than 16 bit, say 32 bit, then the argument "a" could 
very well contain a value greater than 0xffff and then the comparison 
makes perfect sense. So, while you'd get a warning on architectures where 
pid_t is 16bit or less you won't get a warning when pid_t is greater than 
16 bit. "fixing" that warning would clearly be wrong, no argument about 
that.


--
Jesper Juhl

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/