Re: [PATCH] mm/vmpressure: skip socket pressure for costly order reclaim

[Johannes Weiner]

On Wed, Apr 01, 2026 at 01:37:52PM -0700, JP Kobryn (Meta) wrote:
> @@ -307,7 +308,7 @@ void vmpressure(gfp_t gfp, struct mem_cgroup *memcg, bool tree,
>  
>  		level = vmpressure_calc_level(scanned, reclaimed);
>  
> -		if (level > VMPRESSURE_LOW) {
> +		if (level > VMPRESSURE_LOW && order <= PAGE_ALLOC_COSTLY_ORDER) {

I think a comment would be in, ahem, order.

Once we go above COSTLY_ORDER, reclaim relies heavily on compaction to
make progress. Reclaim efficiency was never a great proxy for pressure
to begin with, but it's outright misleading with these high
orders. Don't throttle sockets because somebody is attempting
something crazy like an order-7 and predictably struggling.