[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <672a45a5-722f-af57-dad4-a285acb0e53a@huawei.com>
Date: Sat, 14 Sep 2024 10:15:47 +0800
From: zhengzucheng <zhengzucheng@...wei.com>
To: Waiman Long <longman@...hat.com>, <peterz@...radead.org>,
<juri.lelli@...hat.com>, <vincent.guittot@...aro.org>,
<dietmar.eggemann@....com>, <rostedt@...dmis.org>, <bsegall@...gle.com>,
<mgorman@...e.de>, <vschneid@...hat.com>, <oleg@...hat.com>, Frederic
Weisbecker <frederic@...nel.org>, <mingo@...nel.org>, <peterx@...hat.com>,
<tj@...nel.org>, <tjcao980311@...il.com>
CC: <linux-kernel@...r.kernel.org>
Subject: Re: [Question] sched:the load is unbalanced in the VM overcommitment scenario
在 2024/9/14 1:17, Waiman Long 写道:
> you don't actually need to use 2 different cpusets if they all get the
> same set of CPUs and memory nodes
Yes, you're right. The purpose of setting two different cpusets is to
simulate the scenario of two VMs.
each cpuset is a VM.
For example, the VM configuration is as follows:
<domain type='kvm' id='12676'>
<name>master</name>
<vcpu placement='static' cpuset='0-3,80-83'>8</vcpu>
<iothreads>1</iothreads>
<iothreadids>
<iothread id='1'/>
</iothreadids>
<cputune>
<vcpupin vcpu='0' cpuset='0-3,80-83'/>
<vcpupin vcpu='1' cpuset='0-3,80-83'/>
<vcpupin vcpu='2' cpuset='0-3,80-83'/>
<vcpupin vcpu='3' cpuset='0-3,80-83'/>
<vcpupin vcpu='4' cpuset='0-3,80-83'/>
<vcpupin vcpu='5' cpuset='0-3,80-83'/>
<vcpupin vcpu='6' cpuset='0-3,80-83'/>
<vcpupin vcpu='7' cpuset='0-3,80-83'/>
<emulatorpin cpuset='0-79'/>
</cputune>
<numatune>
<memory mode='strict' nodeset='0'/>
<memnode cellid='0' mode='strict' nodeset='0'/>
</numatune>
Powered by blists - more mailing lists