On Wed, Aug 02, 2006 at 05:02:11AM +0200, Andi Kleen wrote:
> > --- a/arch/x86_64/kernel/smp.c
> > +++ b/arch/x86_64/kernel/smp.c
> > @@ -203,7 +203,7 @@ int __cpuinit init_smp_flush(void)
> > {
> > int i;
> > for_each_cpu_mask(i, cpu_possible_map) {
> > - spin_lock_init(&per_cpu(flush_state.tlbstate_lock, i));
> > + spin_lock_init(&per_cpu(flush_state, i).tlbstate_lock);
>
> What advantage does this have over the earlier form?
I've grepped tree after seeing "[PATCH] fix vmstat per cpu usage"¹.
Rationale mentioned in that thread are
1) invalid asm on s390
2) it only works because per-cpu macros are very simple
> In general this should be split up into three patches.
Yep, I see Andrew splitted them.
¹ http://marc.theaimsgroup.com/?l=linux-kernel&m=115445399826223&w=2
-
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
