Re: [PATCH] kernel.h: Convert rounding macros to statementexpressions, add ADD_MOD

[Randy Dunlap]

On Mon, 29 Mar 2010 14:09:49 -0700 Joe Perches wrote:

> Do you prefer something like this?

Yes, thanks.

> Convert rounding macros to statement expressions
> so arguments are only evaluated once.
> Add kernel-doc to rounding macros
> Add ADD_MOD statement expression for "(x + y) % y"
> 
> Signed-off-by: Joe Perches <joe@xxxxxxxxxxx>
> ---
>  include/linux/kernel.h |   65 +++++++++++++++++++++++++++++++++++++++++------
>  1 files changed, 56 insertions(+), 9 deletions(-)
> 
> diff --git a/include/linux/kernel.h b/include/linux/kernel.h
> index c96b1ac..6b92b20 100644
> --- a/include/linux/kernel.h
> +++ b/include/linux/kernel.h
> @@ -55,15 +55,62 @@ extern const char linux_proc_banner[];
>  #define round_down(x, y) ((x) & ~__round_mask(x, y))
>  
>  #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));	\
> -}							\
> -)
> -#define ADD_MOD(x, y) (((x) + (y)) % (y))
> +
> +/**
> + * roundup() - Returns x rounded up to the next whole multiple of y
> + * @x:	dividend
> + * @y:	divisor
> + *
> + * ie: roundup(4, 2) is 4, roundup(5, 2) is 6
> + * c type conversion rules apply when x and y types differ
> + */
> +#define roundup(x, y)				\
> +({						\
> +	typeof(y) _y = y;			\
> +	((x) + (_y - 1) / _y) * _y;		\

Above needs an extra set of parens:

	(((x) + (_y - 1)) / _y) * _y;		\


> +})
> +
> +/**
> + * DIV_ROUND_UP() - Returns x/y rounded up to the next whole number
> + * @x:	dividend
> + * @y:	divisor
> + *
> + * ie: DIV_ROUND_UP(4, 2) is 2, DIV_ROUND_UP(5, 2) is 3
> + * c type conversion rules apply when x and y types differ
> + */
> +#define DIV_ROUND_UP(x, y)			\
> +({						\
> +	typeof(y) _y = y;			\
> +	((x) + _y - 1) / _y;			\
> +})
> +
> +/**
> + * DIV_ROUND_CLOSEST() - Returns x/y rounded to the nearest whole number
> + * @x:	dividend
> + * @y:	divisor
> + *
> + * ie: DIV_ROUND_CLOSEST(4, 3) is 1, DIV_ROUND_CLOSEST(5, 3) is 2
> + * c type conversion rules apply when x and y types differ
> + */
> +#define DIV_ROUND_CLOSEST(x, y)			\
> +({						\
> +	typeof(y) _y = y;			\
> +	((x) + (_y / 2)) / _y;			\
> +})
> +
> +/**
> + * ADD_MOD() - Returns (x + y) % y
> + * @x:	initial value
> + * @y:	value added to x then used as modulo
> + *
> + * ie: ADD_MOD(4, 2) is 0, ADD_MOD(5, 2) is 1
> + * c type conversion rules apply when x and y types differ
> + */
> +#define ADD_MOD(x, y)				\
> +({						\
> +	typeof(y) _y = y;			\
> +	((x) + _y) % _y;			\
> +})
>  
>  #define _RET_IP_		(unsigned long)__builtin_return_address(0)
>  #define _THIS_IP_  ({ __label__ __here; __here: (unsigned long)&&__here; })


Reviewed-by: Randy Dunlap <randy.dunlap@xxxxxxxxxx>

---
~Randy
--
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/