linux-kernel - Re: [PATCH] sched: Fix numabalancing to work with 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]

Date:   Wed, 5 Apr 2017 07:20:06 +0530
From:   Srikar Dronamraju <srikar@...ux.vnet.ibm.com>
To:     Mel Gorman <mgorman@...hsingularity.net>
Cc:     Ingo Molnar <mingo@...nel.org>,
        Peter Zijlstra <peterz@...radead.org>,
        LKML <linux-kernel@...r.kernel.org>,
        Rik van Riel <riel@...hat.com>
Subject: Re: [PATCH] sched: Fix numabalancing to work with isolated cpus

> > 
> > To avoid this, please check for isolated cpus before choosing a target
> > cpu.
> > 
> 
> Hmm, would this also prevent a task running inside a cgroup that is
> allowed accessed to isolated CPUs from balancing? I severely doubt it

Scheduler doesn't do any kind of load balancing for isolated cpus.

# grep -o "isolcpus=.*" /proc/cmdline 
isolcpus=56,64,72,80,88
# taskset -c "56,64,72,80" ebizzy -S 100 -t 44

(on another terminal)
# pgrep ebizzy
10437
# cat /proc/10437/status |grep Cpus_allowed_list
Cpus_allowed_list:      56,64,72,80
#

But the all the tasks would only run on cpu 56, even if its fully
overloaded.  Cpus 64,72,80 would be completely idle.  So on isolated
cpus, its the user who is in full control of scheduling the tasks on the
cpu.

> matters because if a process is isolated from interference then it
> follows that automatic NUMA balancing should not be involved. If

Yes, as an extension of the above, numa balancing should not be
involved.

> anything the protection should be absolute but either way;
> 
> Acked-by: Mel Gorman <mgorman@...hsingularity.net>
> 

Thanks.