On Tue, 2008-11-11 at 09:07 -0800, Andrew Morton wrote:
> On Tue, 11 Nov 2008 11:04:54 +0100 Jean Delvare <khali at linux-fr.org> wrote:
> > On Mon, 10 Nov 2008 17:01:32 -0800, Darrick J. Wong wrote:
> > > diff --git a/include/linux/kernel.h b/include/linux/kernel.h
> > > index fba141d..fb02266 100644
> > > --- a/include/linux/kernel.h
> > > +++ b/include/linux/kernel.h
> > > @@ -48,6 +48,12 @@ extern const char linux_proc_banner[];
> > > #define FIELD_SIZEOF(t, f) (sizeof(((t*)0)->f))
> > > #define DIV_ROUND_UP(n,d) (((n) + (d) - 1) / (d))
> > > #define roundup(x, y) ((((x) + ((y) - 1)) / (y)) * (y))
> > > +#define DIV_ROUND_CLOSEST(x, divisor)( \
> > > +{ \
> > > + typeof(divisor) __divisor = divisor; \
> > > + (((x) + ((__divisor) / 2)) / (__divisor)); \
> > > +} \
> > > +)
> > I don't get why you implement this as a macro rather than an inline
> > function? A function would look much better.
> The idea is that DIV_ROUND_CLOSEST() can be used with arguments of any
> size (char, short, ... long long) and will do all the suitable
> promotion and will return a type of the appropriate width and
> signedness.
Perhaps the macro should be placed directly after
DIV_ROUND_UP and should use the same argument naming.
Perhaps HALF_UP is more descriptive and fairly common.
http://java.sun.com/j2se/1.5.0/docs/api/java/math/RoundingMode.html
