Re: [RFC, 2.6] a simple FIFO implementation

[Ryan Cumming]

On Friday 17 September 2004 15:35, Andrew Morton wrote:
> Ryan Cumming <ryan@xxxxxxxxxxxxxxxxxxx> wrote:
> > How does this look?
> >
> >  ...
> > +static inline unsigned long __attribute_pure__ roundup_pow_of_two(int x)
> > +{
> > + return (1UL << fls(x));
> > +}
>
> Any reason for making the argument an integer, rather than unsigned long?

I originally had it as unsigned long, but I changed it to match what fls() 
takes. Would it be better as an unsigned long despite that?

-Ryan

Attachment: pgp00000.pgp
Description: PGP signature