Re: [PATCH 2/9] ksm: let shared pages be swappable

[Chris Wright]

* KAMEZAWA Hiroyuki (kamezawa.hiroyu@xxxxxxxxxxxxxx) wrote:
> Hmm. Can KSM coalesce 10000+ of pages to a page ?

Yes.  The zero page is a prime example of this.

> In such case, lru
> need to scan 10000+ ptes with 10000+ anon_vma->lock and 10000+ pte locks
> for reclaiming a page.

Would likely be a poor choice too.  With so many references it's likely
to be touched soon and swapped right back in.

thanks,
-chris
--
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/