Re: [PATCH v7 13/13] vmstat: add pcp remote node draining via cpu_vm_stats_fold

[Marcelo Tosatti]

On Mon, Mar 20, 2023 at 01:43:09PM -0700, Tim Chen wrote:
> On Mon, 2023-03-20 at 15:03 -0300, Marcelo Tosatti wrote:
> > 
> > +
> > +		if (do_pagesets) {
> > +			cond_resched();
> > +			/*
> > +			 * Deal with draining the remote pageset of a
> > +			 * processor
> > +			 *
> > +			 * Check if there are pages remaining in this pageset
> > +			 * if not then there is nothing to expire.
> > +			 */
> > +			if (!pcp->expire || !pcp->count)
> > +				continue;refresh_cpu_vm_stats
> > +
> > +			/*
> > +			 * We never drain zones local to this processor.
> > +			 */
> > +			if (zone_to_nid(zone) == cpu_to_node(cpu)) {
> > +				pcp->expire = 0;
> > +				continue;
> > +			}
> > +
> > +			WARN_ON(pcp->expire < 0);
> > +			/*
> > +			 * pcp->expire is only accessed from vmstat_shepherd context,
> > +			 * therefore no locking is required.
> > +			 */
> > +			if (--pcp->expire)
> > +				continue;
> > +
> > +			if (pcp->count)
> > +				drain_zone_pages(zone, pcp);
> > +		}
> 
> This logic is the same to that for the do_pagesets portion of code in refresh_cpu_vm_stats().
> Is it possible to consolidate to avoid replicating the logic across two functions?
> 
> Tim 

Tim,

One of the versions is for remote access, the other version is for local
access (thats why merging the two is not so simple).

I suppose the next logical step would be ensure that the current 
!CONFIG_HAVE_CMPXCHG_LOCAL also uses the cmpxchg remotely (then 
refresh_cpu_vm_stats can be removed).