linux-kernel - Re: [PATCH] sched/numa: do not balance tasks onto isolated cpus

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]

Message-ID: <201807240911372861670@zte.com.cn>
Date:   Tue, 24 Jul 2018 09:11:37 +0800 (CST)
From:   <jiang.biao2@....com.cn>
To:     <peterz@...radead.org>
Cc:     <chen.lin130@....com.cn>, <mingo@...hat.com>,
        <linux-kernel@...r.kernel.org>, <zhong.weidong@....com.cn>,
        <tan.hu@....com.cn.cn>, <cheng.lin130@....com.cn>,
        <tan.hu@....com.cn>
Subject: Re: [PATCH] sched/numa: do not balance tasks onto isolated cpus

>
> On Mon, Jul 23, 2018 at 01:39:30PM +0800, Chen Lin wrote:
>> From: Chen Lin <cheng.lin130@....com.cn>
>>
>> NUMA balancing has not taken *isolcpus(isolated cpus)* into
>> consideration. It may migrate tasks onto isolated cpus and the
>> migrated tasks will never escape from the isolated cpus, which will
>> break the isolation provided by *isolcpus* boot parameter and
>> intrduce various problems.
>>
>> This patch ensure NUMA balancing not to balance tasks onto iaolated
>> cpus.
>
> I'm not sure what kernel you're patching, but cpu_isolated_map doesn't
> exist anymore. Also, if it steps on isolated CPUs, this is the wrong fix
> anyway. Load-balancing should be constrained to the current root domain.
Indeed, we used 4.14 version and made wrong assumption, sorry for that.
We will retest on latest mainline version, rework the patch and send another
version.

>> Signed-off-by: Cheng Lin <cheng.lin130@....com.cn>
>> Signed-off-by: Tan Hu <tan.hu@....com.cn>
>> Signed-off-by: Jiang Biao <jiang.biao2@....com.cn>
> 
>This SoB chain is invalid.
Mm, we don't quite understand what the *Signed-off-by* precisely means,
Does it only mean DCO(developer certificate of origin)?
As we understood, multiple SoBs could be used to indicate co-authors.
If SoB only means DCO, how can the patches reflect co-authors?

Thanks a lot.
Jiang,
Regards