Re: [PATCH 2/5] bitops: compile time optimization for hweight_long(CONSTANT)

[Andrew Morton <akpm@linux-foundation.org>]

On Sat, 30 Jan 2010 17:45:17 +0800
Wu Fengguang <fengguang.wu@intel.com> wrote:

> This allows use of hweight_long() in BUILD_BUG_ON().
> Suggested by Jamie.
> 
> CC: Jamie Lokier <jamie@shareable.org>
> CC: Roland Dreier <rdreier@cisco.com>
> Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
> ---
>  include/linux/bitops.h |   12 ++++++++----
>  1 file changed, 8 insertions(+), 4 deletions(-)
> 
> --- linux-mm.orig/include/linux/bitops.h	2009-07-20 20:10:19.000000000 +0800
> +++ linux-mm/include/linux/bitops.h	2010-01-30 17:41:15.000000000 +0800
> @@ -40,10 +40,14 @@ static __inline__ int get_count_order(un
>  	return order;
>  }
>  
> -static inline unsigned long hweight_long(unsigned long w)
> -{
> -	return sizeof(w) == 4 ? hweight32(w) : hweight64(w);
> -}
> +#define hweight_long(x)				\
> +(						\
> +	__builtin_constant_p(x) ?		\
> +		__builtin_popcountl(x) :	\
> +		(sizeof(x) <= 4 ?		\
> +			 hweight32(x) :		\
> +			 hweight64(x))		\
> +)
>  
>  /**
>   * rol32 - rotate a 32-bit value left

Peter's been mucking with a compile-time HWEIGHT().  An outdated
version of that is presently in linux-next.

(I wonder if it could have used __builtin_popcount)

(I wonder which gcc versions support __builtin_popcount)

Anyway, I suspect you should be using that rather than tweaking
hweight_long().