[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20190130071759.GR18811@dhcp22.suse.cz>
Date: Wed, 30 Jan 2019 08:17:59 +0100
From: Michal Hocko <mhocko@...nel.org>
To: Andrea Arcangeli <aarcange@...hat.com>
Cc: lsf-pc@...ts.linux-foundation.org, linux-mm@...ck.org,
linux-kernel@...r.kernel.org, Peter Xu <peterx@...hat.com>,
Blake Caldwell <blake.caldwell@...orado.edu>,
Mike Rapoport <rppt@...ux.vnet.ibm.com>,
Mike Kravetz <mike.kravetz@...cle.com>,
Mel Gorman <mgorman@...e.de>, Vlastimil Babka <vbabka@...e.cz>,
David Rientjes <rientjes@...gle.com>
Subject: Re: [LSF/MM TOPIC] NUMA remote THP vs NUMA local non-THP under
MADV_HUGEPAGE
On Tue 29-01-19 18:40:58, Andrea Arcangeli wrote:
> Hello,
>
> I'd like to attend the LSF/MM Summit 2019. I'm interested in most MM
> topics and it's enlightening to listen to the common non-MM topics
> too.
>
> One current topic that could be of interest is the THP / NUMA tradeoff
> in subject.
>
> One issue about a change in MADV_HUGEPAGE behavior made ~3 years ago
> kept floating around for the last 6 months (~12 months since it was
> initially reported as regression through an enterprise-like workload)
> and it was hot-fixed in commit
> ac5b2c18911ffe95c08d69273917f90212cf5659, but it got quickly reverted
> for various reasons.
>
> I posted some benchmark results showing that for tasks without strong
> NUMA locality the __GFP_THISNODE logic is not guaranteed to be optimal
> (and here of course I mean even if we ignore the large slowdown with
> swap storms at allocation time that might be caused by
> __GFP_THISNODE). The results also show NUMA remote THPs help
> intrasocket as well as intersocket.
>
> https://lkml.kernel.org/r/20181210044916.GC24097@redhat.com
> https://lkml.kernel.org/r/20181212104418.GE1130@redhat.com
>
> The following seems the interim conclusion which I happen to be in
> agreement with Michal and Mel:
>
> https://lkml.kernel.org/r/20181212095051.GO1286@dhcp22.suse.cz
> https://lkml.kernel.org/r/20181212170016.GG1130@redhat.com
I am definitely interested in discussing this topic and actually wanted
to propose it myself. I would add that part of the discussion was
proposing a neww memory policy that would effectively enable per-vma
node-reclaim like behavior.
--
Michal Hocko
SUSE Labs
Powered by blists - more mailing lists