linux-kernel - Re: [PATCH 3/4] sched/numa: Apply imbalance limitations consistently

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

Date:   Wed, 18 May 2022 16:39:47 +0100
From:   Mel Gorman <mgorman@...hsingularity.net>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     Ingo Molnar <mingo@...nel.org>,
        Vincent Guittot <vincent.guittot@...aro.org>,
        Valentin Schneider <valentin.schneider@....com>,
        Aubrey Li <aubrey.li@...ux.intel.com>,
        LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 3/4] sched/numa: Apply imbalance limitations consistently

On Wed, May 18, 2022 at 03:59:34PM +0200, Peter Zijlstra wrote:
> On Wed, May 18, 2022 at 11:46:52AM +0100, Mel Gorman wrote:
> 
> > > (Although I do wonder about that 25% figure in the comment; that doesn't
> > > seem to relate to any actual code anymore)
> > > 
> > 
> > You're right, by the end of the series it's completely inaccurate and
> > currently it's not accurate if there are multiple LLCs per node. I
> > adjusted the wording to "Allow a NUMA imbalance if busy CPUs is less
> > than the maximum threshold. Above this threshold, individual tasks may
> > be contending for both memory bandwidth and any shared HT resources."
> > 
> 
> Looks good. Meanwhile I saw a 0-day complaint that this regresses
> something something unixbench by a bit. Do we care enough? I suppose
> this is one of those trade-off patches again, win some, loose some.

I think it's a trade-off. I made a more complete response to the 0-day
people at https://lore.kernel.org/all/20220518152258.GR3441@techsingularity.net/

-- 
Mel Gorman
SUSE Labs