Re: Node affine NUMA scheduler

[Pete Zaitcev]

> diff -urN 2.4.17-ia64-kdbv2.1-K3+/kernel/fork.c 2.4.17-ia64-kdbv2.1-k3z-nod9/kernel/fork.c
> --- 2.4.17-ia64-kdbv2.1-K3+/kernel/fork.c	Mon Mar  4 11:39:18 2002
> +++ 2.4.17-ia64-kdbv2.1-k3z-nod9/kernel/fork.c	Thu Mar  7 13:50:42 2002
> @@ -640,10 +640,6 @@
>  	{
>  		int i;
>  
> -		if (likely(p->cpus_allowed & (1UL<<smp_processor_id())))
> -			p->cpu = smp_processor_id();
> -		else
> -			p->cpu = __ffs(p->cpus_allowed);
>  		/* ?? should we just memset this ?? */
>  		for(i = 0; i < smp_num_cpus; i++)
>  			p->per_cpu_utime[cpu_logical_map(i)] =

OK, I am glad that we kinda converge on something common here
with ia-64 people.

-- Pete