On Tue, Apr 14, 2009 at 03:09:29PM -0700, Andrew Morton wrote: > We need a comment here explaining why we can't use the much preferable > lock_page(). > > Why can't we use the much preferable lock_page()? We might but then it'd risk to waste time waiting. It's not worth waiting, we want kksmd to be allowed to keep one (in future more than one as we scale it smp/numa) CPU busy at all times running memcmp and not schedule (other than for need_resched()) to try to free memory at the fastest peace possible. -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
