Re: [PATCH 2.6.18-rc4-mm2] Generic boolean

[Richard Knutsson <ricknu-0@student.ltu.se>]

Jan Engelhardt wrote:

> Hi,
>
>
>  
>
> > There has been concern about adding other values then 0 and 1. There has been
> > ideas to do something like:
> > bool b = i & 1 : 0;
> >    
>
> I think you miseed a '?'
>  
Yes, thanks...

> bool b = (i & 1) ? : 0;
>  
...but I meant: bool b = i ? 1 : 0;

>  
>
> > /*or*/
> > bool b = !!i;
> >
> > but all that is needed is just a casting:
> >
> > bool b = (bool) i;
> >    
>
> No casting needed (in fact, casting is more evil than !!). If bool is a
>  
Please inform me why casting is evil (other then risking losing data, by 
cutting the value). But also, _Bool-casting is after all a bit special 
since it does not seem (at least by me) to be able to get a wrong value 
(giving it a value other then 0 but reciving 0).

> bool, then the compiler will (hopefully) ensure that b will only get
> values valid for bools.
>  
No, not really. Because _Bool is a byte and not a bit, you can put in a 
value other then 0 or 1 if you cast the pointer (if you insert 256 it 
becomes 0). But after all, there should not be any:
if (b == true)
in the code anyway, so it should be ok.

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