From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-yk0-f173.google.com (mail-yk0-f173.google.com [209.85.160.173]) by kanga.kvack.org (Postfix) with ESMTP id 056DE6B0253 for ; Wed, 29 Jul 2015 11:34:54 -0400 (EDT) Received: by ykdu72 with SMTP id u72so10791509ykd.2 for ; Wed, 29 Jul 2015 08:34:53 -0700 (PDT) Received: from mail-yk0-x22c.google.com (mail-yk0-x22c.google.com. [2607:f8b0:4002:c07::22c]) by mx.google.com with ESMTPS id z188si18867815ywa.99.2015.07.29.08.34.52 for (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Wed, 29 Jul 2015 08:34:53 -0700 (PDT) Received: by ykay190 with SMTP id y190so10774462yka.3 for ; Wed, 29 Jul 2015 08:34:52 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: <20150729153135.GW8100@esperanza> References: <20150729123629.GI15801@dhcp22.suse.cz> <20150729135907.GT8100@esperanza> <20150729144539.GU8100@esperanza> <20150729153135.GW8100@esperanza> Date: Wed, 29 Jul 2015 08:34:52 -0700 Message-ID: Subject: Re: [PATCH -mm v9 0/8] idle memory tracking From: Michel Lespinasse Content-Type: multipart/alternative; boundary=94eb2c0335d0f2164a051c055333 Sender: owner-linux-mm@kvack.org List-ID: To: Vladimir Davydov Cc: Michal Hocko , Andrew Morton , Andres Lagar-Cavilla , Minchan Kim , Raghavendra K T , Johannes Weiner , Greg Thelen , David Rientjes , Pavel Emelyanov , Cyrill Gorcunov , Jonathan Corbet , linux-api@vger.kernel.org, linux-doc@vger.kernel.org, linux-mm@kvack.org, cgroups@vger.kernel.org, linux-kernel@vger.kernel.org --94eb2c0335d0f2164a051c055333 Content-Type: text/plain; charset=UTF-8 On Wed, Jul 29, 2015 at 8:31 AM, Vladimir Davydov wrote: > On Wed, Jul 29, 2015 at 08:08:22AM -0700, Michel Lespinasse wrote: > > On Wed, Jul 29, 2015 at 7:45 AM, Vladimir Davydov < > vdavydov@parallels.com> > > wrote: > > > Page table scan approach has the inherent problem - it ignores unmapped > > > page cache. If a workload does a lot of read/write or map-access-unmap > > > operations, we won't be able to even roughly estimate its wss. > > > > You can catch that in mark_page_accessed on those paths, though. > > Actually, the problem here is how to find an unmapped page cache page > *to mark it idle*, not to mark it accessed. > Ah, yes. When I tried that I was still scanning memory by address at the end just to compute such totals - but I did not have to do rmap at that point anymore. It did look incredibly lame, though. -- Michel "Walken" Lespinasse A program is never fully debugged until the last user dies. --94eb2c0335d0f2164a051c055333 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
On Wed, Jul 29, 2015 at 8:31 AM, Vladimir Davydo= v <vdavydov@parallels.com> wrote:
On Wed, Jul 29, 2015 at 08:08:22AM -0700, Michel Lespinasse w= rote:
> On Wed, Jul 29, 2015 at 7:45 AM, Vladimir Davydov <vdavydov@parallels.com>
> wrote:
> > Page table scan approach has the inherent problem - it ignores un= mapped
> > page cache. If a workload does a lot of read/write or map-access-= unmap
> > operations, we won't be able to even roughly estimate its wss= .
>
> You can catch that in mark_page_accessed on those paths, though.

Actually, the problem here is how to find an unmapped page cache pag= e
*to mark it idle*, not to mark it accessed.

Ah, ye= s.

When I tried that I was still scanning memory by address at the e= nd just to compute such totals - but I did not have to do rmap at that poin= t anymore.

It did look incredibly lame, though.

--
Michel "Walken"= ; Lespinasse
A program is never fully debugged until the last user dies.=
--94eb2c0335d0f2164a051c055333-- -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org