linux-kernel - Re: [PATCH v3] mm/khugepaged: sched to numa node when collapse huge page

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:   Thu, 12 May 2022 19:40:13 -0700
From:   Yang Shi <shy828301@...il.com>
To:     Andrew Morton <akpm@...ux-foundation.org>
Cc:     maobibo <maobibo@...ngson.cn>, Peter Xu <peterx@...hat.com>,
        David Hildenbrand <david@...hat.com>,
        Linux MM <linux-mm@...ck.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v3] mm/khugepaged: sched to numa node when collapse huge page

On Thu, May 12, 2022 at 6:49 PM Andrew Morton <akpm@...ux-foundation.org> wrote:
>
> On Fri, 13 May 2022 09:29:07 +0800 maobibo <maobibo@...ngson.cn> wrote:
>
> >
> >
> > >> and/or changelogging.
> > > Sorry for the late response, the mail is filtered and I did not notice that. The result is not so obvious after bandwidth is improved between nodes, it is only about 1% improvement for specint2006 for 32 core on my box.
> > >
> > > Now I do not see negative effective about this patch unless user wants to keep some cores separated from khugepaged daemon process.
> >
> > Can we provide an extra parameter to let khugepaged daemon scheduling binded to node or freely? If can, I will provide updated patch.
>
> It has always surprised me that we have a single khugepaged thread.  If
> we had a thread per node, you'd be all fixed up, yes?

Actually I was thinking about this before, but I didn't see too much
benefit with this approach TBH. The khugepaged scans vmas and the
mapped pages may spread on all nodes. It is not like kswapd which
could scan the LRUs for the specific node.

>
> Ditto ksmd.
>