[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20181214113241.GC5343@dhcp22.suse.cz>
Date: Fri, 14 Dec 2018 12:32:41 +0100
From: Michal Hocko <mhocko@...nel.org>
To: Andrea Arcangeli <aarcange@...hat.com>
Cc: David Rientjes <rientjes@...gle.com>,
Linus Torvalds <torvalds@...ux-foundation.org>,
mgorman@...hsingularity.net, Vlastimil Babka <vbabka@...e.cz>,
ying.huang@...el.com, s.priebe@...fihost.ag,
Linux List Kernel Mailing <linux-kernel@...r.kernel.org>,
alex.williamson@...hat.com, lkp@...org, kirill@...temov.name,
Andrew Morton <akpm@...ux-foundation.org>,
zi.yan@...rutgers.edu
Subject: Re: [LKP] [mm] ac5b2c1891: vm-scalability.throughput -61.3%
regression
On Wed 12-12-18 12:00:16, Andrea Arcangeli wrote:
[...]
> Adding MADV_THISNODE/MADV_NODE_RECLAIM, will guarantee his proprietary
> software binary will run at maximum performance without cache
> interference, and he's happy to accept the risk of massive slowdown in
> case the local node is truly OOM. The fallback, despite very
> inefficient, will still happen without OOM killer triggering.
I believe this fits much better into a MPOL_$FOO rather than MADV_$FOO.
But other than that I full agree. There are reasonable usecases for the
node reclaim like behavior. As a bonus you do not get local node only
but all nodes within reclaim distance as well.
--
Michal Hocko
SUSE Labs
Powered by blists - more mailing lists