Re: NUMA node swapping V3

[Mika Penttilä]

> 	if (pg == orig) {
> 		z->pageset[cpu].numa_hit++;
> +		/*
> +		 * If zone allocation has left less than
> +		 * (sysctl_node_swap / 10) %  of the zone free invoke kswapd.
> +		 * (the page limit is obtained through (pages*limit)/1024 to
> +		 * make the calculation more efficient)
> +		 */
> +		if (z->free_pages < (z->present_pages * sysctl_node_swap) << 10)
> +			wakeup_kswapd(z);
> 	} else {
> 		p->numa_miss++;
> 		zonelist->zones[0]->pageset[cpu].numa_foreign++;
> Index: linux-2.6.9/kernel/sysctl.c
> ===================================================================
>  

I think you mean >> 10 though.

--Mika


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/