Re: [PATCH 2/6] futex: Implement FUTEX2_NUMA

[Peter Zijlstra <peterz@infradead.org>]

On Fri, Oct 25, 2024 at 11:30:26AM -0700, Davidlohr Bueso wrote:
> On Fri, 25 Oct 2024, Peter Zijlstra wrote:\n
> 
> > static int __init futex_init(void)
> > {
> > -	unsigned int futex_shift;
> > -	unsigned long i;
> > +	unsigned int order, n;
> > +	unsigned long size, i;
> > 
> > #ifdef CONFIG_BASE_SMALL
> > 	futex_hashsize = 16;
> > #else
> > -	futex_hashsize = roundup_pow_of_two(256 * num_possible_cpus());
> > +	futex_hashsize = 256 * num_possible_cpus();
> > +	futex_hashsize /= num_possible_nodes();
> > +	futex_hashsize = roundup_pow_of_two(futex_hashsize);
> > #endif
> > +	futex_hashshift = ilog2(futex_hashsize);
> > +	size = sizeof(struct futex_hash_bucket) * futex_hashsize;
> > +	order = get_order(size);
> > +
> > +	for_each_node(n) {
> 
> Probably want to skip nodes that don't have CPUs, those will never
> have the remote for .node value.

What if the CPU-less node is placed equidistant between two (or more)
regular nodes and it is the best location for a futex that is spanning
those nodes?

That is to say, just because it doesn't have CPUs, doesn't mean it is
never the right node.

Hmm?