Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755252AbbHCU4P (ORCPT ); Mon, 3 Aug 2015 16:56:15 -0400 Received: from gum.cmpxchg.org ([85.214.110.215]:48755 "EHLO gum.cmpxchg.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754533AbbHCU4N (ORCPT ); Mon, 3 Aug 2015 16:56:13 -0400 Date: Mon, 3 Aug 2015 16:55:32 -0400 From: Johannes Weiner To: Vladimir Davydov Cc: Andrew Morton , Michal Hocko , Minchan Kim , Rik van Riel , Mel Gorman , linux-mm@kvack.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH 2/3] mm: make workingset detection logic memcg aware Message-ID: <20150803205532.GA19478@cmpxchg.org> References: <9662034e14549b9e1445684f674063ce8b092cb0.1438599199.git.vdavydov@parallels.com> <20150803132358.GA18399@cmpxchg.org> <20150803135229.GA11971@esperanza> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20150803135229.GA11971@esperanza> User-Agent: Mutt/1.5.23+102 (2ca89bed6448) (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2514 Lines: 50 On Mon, Aug 03, 2015 at 04:52:29PM +0300, Vladimir Davydov wrote: > On Mon, Aug 03, 2015 at 09:23:58AM -0400, Johannes Weiner wrote: > > On Mon, Aug 03, 2015 at 03:04:22PM +0300, Vladimir Davydov wrote: > > > @@ -179,8 +180,9 @@ static void unpack_shadow(void *shadow, > > > eviction = entry; > > > > > > *zone = NODE_DATA(nid)->node_zones + zid; > > > + *lruvec = mem_cgroup_page_lruvec(page, *zone); > > > > > > - refault = atomic_long_read(&(*zone)->inactive_age); > > > + refault = atomic_long_read(&(*lruvec)->inactive_age); > > > mask = ~0UL >> (NODES_SHIFT + ZONES_SHIFT + > > > RADIX_TREE_EXCEPTIONAL_SHIFT); > > > /* > > > > You can not compare an eviction shadow entry from one lruvec with the > > inactive age of another lruvec. The inactive ages are not related and > > might differ significantly: memcgs are created ad hoc, memory hotplug, > > page allocator fairness drift. In those cases the result will be pure > > noise. > > That's true. If a page is evicted in one cgroup and then refaulted in > another, the activation will be random. However, is it a frequent event > when a page used by and evicted from one cgroup is refaulted in another? > If there is no active file sharing (is it common?), this should only > happen to code pages, but those will most likely end up in the cgroup > that has the greatest limit, so they shouldn't be evicted and refaulted > frequently. So the question is can we tolerate some noise here? It's not just the memcg, it's also the difference between zones themselves. > > As much as I would like to see a simpler way, I am pessimistic that > > there is a way around storing memcg ids in the shadow entries. > > On 32 bit there is too little space for storing memcg id. We can shift > the distance so that it would fit and still contain something meaningful > though, but that would take much more code, so I'm trying to try the > simplest way first. It should be easy to trim quite a few bits from the timestamp, both in terms of available memory as well as in terms of distance granularity. We probably don't care if the refault distance is only accurate to say 2MB, and how many pages do we have to represent on 32-bit in the first place? Once we trim that, we should be able to fit a CSS ID. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/