Re: [tip:sched/core] x86/numa: Check for nonsensical topologies on real hw as well

[Peter Zijlstra <peterz@infradead.org>]

On Thu, 2012-05-10 at 13:49 +0200, Peter Zijlstra wrote:
> +static bool __cpuinit match_mc(struct cpuinfo_x86 *c, struct cpuinfo_x86 *o)
> +{
> +       if (!topology_sane(c->cpu_index, o->cpu_index, "mc"))
> +               return false;
> +
> +       if (c->phys_proc_id == cpu_data(i).phys_proc_id)
> +               return true;
> +
> +       return false;
> +} 

Argh, that has the very same problem.. it (and the same for the other
two match funcions).

Something like:

  if (c->phys_proc_id == cpu_data(i).phys_proc_id && topology_sane())

might do, since then we only verify the topology after we would
otherwise have already accepted it.