Re: [KJ][RFC][PATCH] BIT macro cleanup

[Richard Knutsson]

Milind Choudhary wrote:
> Hi all
>     working towards the cleanup of BIT macro,
> I've added one to <linux/bitops.h> & cleaned some obvious users.
>
> include/linux/input.h also has a BIT macro
> which does a wrap
> so currently i've done something like
>
> +#undef BIT
> #define BIT(nr)        (1UL << ((nr) % BITS_PER_LONG))
>
> Is it advisible to move this macro to bitops.h with some other name
>
> +#define BITWRAP(nr)    (1UL << ((nr) % BITS_PER_LONG))
>
> & make the whole input subsystem use it
> The change is huge, more than 125 files using input.h
> & almost all use the BIT macro.
It is as a big of change, but have you dismissed the "BIT(nr % 
BITS_PER_LONG)" approach?
I really think this has to be broken down into a patch-set.

<snip>
> diff --git a/fs/select.c b/fs/select.c
> index fe0893a..4bbe8ed 100644
> --- a/fs/select.c
> +++ b/fs/select.c
> @@ -180,7 +180,6 @@ get_max:
>        return max;
> }
>
> -#define BIT(i)         (1UL << ((i)&(__NFDBITS-1)))
Are you sure you can just delete this one?

<snip>
> diff --git a/include/linux/input.h b/include/linux/input.h
> index bde65c8..e4203d1 100644
> --- a/include/linux/input.h
> +++ b/include/linux/input.h
> @@ -908,9 +908,11 @@ struct ff_effect {
> #include <linux/fs.h>
> #include <linux/timer.h>
> #include <linux/mod_devicetable.h>
> +//#include <linux/bitops.h>
You added and commented it out?
> #define NBITS(x) (((x)/BITS_PER_LONG)+1)
> -#define BIT(x) (1UL<<((x)%BITS_PER_LONG))
> +#undef BIT
> +#define BIT(nr)        (1UL << ((nr) % BITS_PER_LONG))
Why did you change x to nr? The other defines seems to use x.
> #define LONG(x) ((x)/BITS_PER_LONG)
>
> #define INPUT_KEYCODE(dev, scancode) ((dev->keycodesize == 1) ?
> ((u8*)dev->keycode)[scancode] : \
> diff --git a/include/video/sstfb.h b/include/video/sstfb.h
> index baa163f..b52f073 100644
> --- a/include/video/sstfb.h
> +++ b/include/video/sstfb.h
> @@ -68,7 +68,6 @@
> #  define print_var(X,Y...)
> #endif
>
> -#define BIT(x)         (1ul<<(x))
> #define POW2(x)                (1ul<<(x))
Maybe you can clean up POW2 as well (or define it as "#define POW2(x) 
BIT(x)")

Also, it seems your mail-client swapped the tabs to spaces (aka not able 
to apply).

Other then what I have commented on, it looks good.

Richard Knutsson

-
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/