Hi Gianluca

Thank you very much for your reply!

> On Mon, Feb 3, 2020 at 3:16 PM <branimirp(a)gmail.com&gt; wrote:
> 
> Can you compare the qemu-kvm command line between:

The command lines are (proverbially) huge but I spotted a few (notable) 
differences in '-cpu' option - if we ignore qemu cmd (please see my comment on 
standalone KVM OS):

> - the  nested ESXi 6.7 on top of the standalone KVM hypervisor

/usr/bin/qemu-system-x86_64 -cpu 
Haswell-noTSX-IBRS,vme=on,ss=on,vmx=on,rdrand=on,hypervisor=on,arat=on,tsc_adjust=on,umip=on,md-clear=on,stibp=on,arch-capabilities=on,ssbd=on,xsaveopt=on,pdpe1gb=on,abm=on,fma=off,aes=off,avx=off,f16c=off,bmi1=off,avx2=off,smep=off,bmi2=off

> - the nested ESXi 6.7 on top of oVirt based hypervisor 

Not sure if "host" is a "shortcut" for all CPU flags exposed to the nested VM:

/usr/libexec/qemu-kvm -cpu host,vmx=on
(in oVirt management, "Pass-Through Host CPU" is selected, In Edit Host, 
"Kernel" tab, Kernel command line has "kvm-intel.nested=1" and as mentioned 
vdsm nestedvt hook rpm was installed)

> 
> Also,
> which OS version of the standalone KVM hypervisor?
openSUSE 15.1

> which cpu for the KVM hypervisor
Intel(R) Celeron(R) CPU G1820 @ 2.70GHz

> which oVirt version?
the newest oVirt 4.3.8 with both CentOS 7.7 hypervisors fully updated.

> which cpu for the oVirt hypervisor?
Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz

> did you use e1000 as the network interface type also in oVirt for the ESXi
> 6.7 VM?

Correct. For ESXi, I use e1000 on both standalone KVM and oVirt KVM hypervisor.
> 

If you need any additional info, please let me know.

Thank you!

Regards,

Branimir
_______________________________________________
Users mailing list -- [email protected]
To unsubscribe send an email to [email protected]
Privacy Statement: https://www.ovirt.org/site/privacy-policy/
oVirt Code of Conduct: 
https://www.ovirt.org/community/about/community-guidelines/
List Archives: 
https://lists.ovirt.org/archives/list/[email protected]/message/2MMLANIOJUKM3RCILXGNL2MWAU3F67VI/

Reply via email to