Re: [patch 1/5] i386 generic cmpxchg

[Nick Piggin]

Zwane Mwaikambo wrote:
> Hi Nick,
>
> On Sun, 30 Oct 2005, Nick Piggin wrote:
>
> +#define cmpxchg(ptr,o,n)						\
> +({									\
> +	__typeof__(*(ptr)) __ret;					\
> +	if (likely(boot_cpu_data.x86 > 3))				\
> +		__ret = __cmpxchg((ptr), (unsigned long)(o),		\
> +					(unsigned long)(n), sizeof(*(ptr))); \
> +	else								\
> +		__ret = cmpxchg_386((ptr), (unsigned long)(o),		\
> +					(unsigned long)(n), sizeof(*(ptr))); \
> +	__ret;								\
> +})
> +#endif
>
> How about something similar to the following to remove the branch on 
> optimised kernels?

Hi Zwane,
This is only in the !CONFIG_X86_CMPXCHG case, though, so the branch would
only be there on a 386 kernel, I think?

--
SUSE Labs, Novell Inc.

Send instant messages to your online friends http://au.messenger.yahoo.com 
-
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/