「KVM」- CPU model and topology（未完） @20210222

basic introduction

The following elements can be used to specify the CPU "model", "feature", and "topology".

<cpu match='exact'>
  <model fallback='allow'>core2duo</model>
  <vendor>Intel</vendor>
  <topology sockets='1' cores='2' threads='1'/>
  <cache level='3' mode='emulate'/>
  <feature policy='disable' name='lahf_lm'/>
</cpu>

<cpu mode='host-model'>
  <model fallback='forbid'/>
  <topology sockets='1' cores='2' threads='1'/>
</cpu>

<cpu mode='host-passthrough'>
  <cache mode='passthrough'/>
  <feature policy='disable' name='lahf_lm'/>
...

If there is no need to restrict the CPU "model" and its "features", you can simply use a cpu element:

<cpu>
  <topology sockets='1' cores='2' threads='1'/>
</cpu>

Configuration structure

A complete structure is as follows:

<cpu match="minimum/exact/strict" check="none/partial/full" mode="custom/host-model/host-passthrough">
	<model />
	<vendor />
	<topology sockets='1' cores='2' threads='1'/>
	<feature />
	<cache />
	<numa>
		<cell id="" cpus="" memory="" unit="" memAccess=""  discard="">
		<distances>
        	<sibling id='' value=''/>
      	</distances>
		</cell>
	</numa>
</cpu>

references

KVM cpu configuration mode classification