linux-kernel - Re: [PATCH 2/3] sched/fair: trigger asym

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:   Thu, 20 Dec 2018 16:12:18 +0000
From:   Valentin Schneider <valentin.schneider@....com>
To:     Vincent Guittot <vincent.guittot@...aro.org>
Cc:     Peter Zijlstra <peterz@...radead.org>,
        Ingo Molnar <mingo@...nel.org>,
        linux-kernel <linux-kernel@...r.kernel.org>,
        Morten Rasmussen <Morten.Rasmussen@....com>
Subject: Re: [PATCH 2/3] sched/fair: trigger asym_packing during idle load
 balance

On 20/12/2018 14:33, Vincent Guittot wrote:
[...]
>> As in the previous thread, I'll still argue that if you want to *reliably*
>> exploit newidle balances to do asym packing active balancing, you should
>> add some logic to raise rq->rd->overload when we notice some asym packing
>> could be done, so that it can be leveraged by a higher-priority CPU doing
>> a newidle balance.
> 
> The source cpu with the task is never overloaded.
> We need to start the load balance to know that it's worth migrating the task.
> 

Yep, that's my point exactly:  if you ever want to active balance a task
on a rq with nr_running == 1 when a higher priority CPU goes newidle, you'd
need an asym-packing version of:

  757ffdd705ee ("sched/fair: Set rq->rd->overload when misfit")

It's somewhat orthogonal to this patch, but the reason I'm bringing this up
is that the commit log says

  "newly idle load balance is not always triggered when a cpu becomes idle"

And not having root_domain->overload raised is a reason for that issue.

>>
>> Otherwise the few newidle asym-packing active balances you'll get will be
>> due to somewhat random luck because we happened to set that overload flag
>> at some point.