linux-kernel - Re: [PATCH v2] sched/core: Don't mix isolcpus and housekeeping 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] [day] [month] [year] [list]

Date:   Fri, 26 Oct 2018 15:00:55 +0530
From:   Srikar Dronamraju <srikar@...ux.vnet.ibm.com>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     Ingo Molnar <mingo@...nel.org>,
        LKML <linux-kernel@...r.kernel.org>,
        Mel Gorman <mgorman@...hsingularity.net>,
        Rik van Riel <riel@...riel.com>,
        Yi Wang <wang.yi59@....com.cn>, zhong.weidong@....com.cn,
        Yi Liu <liu.yi24@....com.cn>,
        Frederic Weisbecker <frederic@...nel.org>,
        Thomas Gleixner <tglx@...utronix.de>
Subject: Re: [PATCH v2] sched/core: Don't mix isolcpus and housekeeping CPUs

* Peter Zijlstra <peterz@...radead.org> [2018-10-26 10:41:05]:

> > #
> > # ./hist.sh
> > 10:35:21  IST   UID      TGID       TID    %usr %system  %guest    %CPU   CPU  Command
> > 10:35:21  IST     0      2645         -    8.70    0.01    0.00    8.71     1  ebizzy
> > 10:35:21  IST     0         -      2645    0.01    0.00    0.00    0.01     1  |__ebizzy
> > 10:35:21  IST     0         -      2647    0.14    0.00    0.00    0.14     1  |__ebizzy
> > 10:35:21  IST     0         -      2648    0.13    0.00    0.00    0.13     1  |__ebizzy
> > 10:35:21  IST     0         -      2649    0.13    0.00    0.00    0.13     1  |__ebizzy
> > 10:35:21  IST     0         -      2650    0.13    0.00    0.00    0.13     1  |__ebizzy
> > 10:35:21  IST     0         -      2651    0.13    0.00    0.00    0.13     1  |__ebizzy
> > 10:35:21  IST     0         -      2652    0.13    0.00    0.00    0.13     1  |__ebizzy
> > 10:35:21  IST     0         -      2653    0.13    0.00    0.00    0.13     1  |__ebizzy
> > 10:35:23  IST   UID      TGID       TID    %usr %system  %guest    %CPU   CPU  Command
> > 10:35:23  IST     0      2642         -    0.00    0.00    0.00    0.00     1  hist.sh
> > 10:35:23  IST     0         -      2642    0.00    0.00    0.00    0.00     1  |__hist.sh
> > #
> 
> That's correct and specified behaviour.
> 
> cpu1 has no sched domain (or the singleton domaon, which is the same)
> and thus its tasks will not be migrated.
> 

As I mentioned, If every run was the same, then it would have been good.
However if we increase the cpuset from 1-8 to 1-31 then we seen spreading
more often but not always.

./hist.sh
02:59:10  IST   UID      TGID       TID    %usr %system  %guest    %CPU   CPU  Command
02:59:10  IST     0      2244         -  100.00    0.13    0.00  100.00    30  ebizzy
02:59:10  IST     0         -      2244    0.00    0.02    0.00    0.02    30  |__ebizzy
02:59:10  IST     0         -      2246    4.37    0.00    0.00    4.37    21  |__ebizzy
02:59:10  IST     0         -      2247    2.14    0.00    0.00    2.14     8  |__ebizzy
02:59:10  IST     0         -      2248    3.13    0.00    0.00    3.13    28  |__ebizzy
02:59:10  IST     0         -      2249    5.57    0.00    0.00    5.57    18  |__ebizzy
02:59:10  IST     0         -      2250    4.29    0.00    0.00    4.29    11  |__ebizzy
02:59:10  IST     0         -      2251    5.32    0.00    0.00    5.32    19  |__ebizzy
02:59:10  IST     0         -      2252    5.62    0.00    0.00    5.62    17  |__ebizzy
02:59:11  IST   UID      TGID       TID    %usr %system  %guest    %CPU   CPU  Command
02:59:11  IST     0      2241         -    0.00    0.00    0.00    0.00    29  hist.sh
02:59:11  IST     0         -      2241    0.00    0.00    0.00    0.00    29  |__hist.sh


> If there is a problem, it is with cpuset and its interaction with
> isolcpus. I still have to look at the last version of cpuset-v2, but
> iirc that was better in this regard.

Okay.

-- 
Thanks and Regards
Srikar Dronamraju