https://jenkins-fedora-atomic-process.apps.ci.centos.org/job/str_pr_test/141/console

W0611: 11,0: : Unused import multiprocessing
Running: pycodestyle --max-line-length=120 /home/jenkins/workspace/str_pr_test/inventory/standard-inventory-qcow2
/home/jenkins/workspace/str_pr_test/inventory/standard-inventory-qcow2:116:1: E302 expected 2 blank lines, found 1
/home/jenkins/workspace/str_pr_test/inventory/standard-inventory-qcow2:132:1: E302 expected 2 blank lines, found 1

OK, updated :arrow_up:

lscpu is part of util-linux RPM. OK
nproc is part of coreutils. OK

About next code:

This PR has: '{},sockets={},cores=1,threads=1'.format(sockets, sockets)

From qemu documentation:

-smp [cpus=]n[,cores=cores][,threads=threads][,sockets=sockets][,maxcpus=maxcpus]
... For the PC target, the number of cores per socket, the number of threads per cores and the total number of sockets can be specified. Missing values will be computed.

Then from the email: https://www.redhat.com/archives/libvirt-users/2017-June/msg00025.html

By defaulting to giving the guest **only sockets**, you get fixed predictable
behaviour from the guest OS, as the host OS moves vCPUs threads around.

I am not sure why we need to set cores=1,threads=1.

Are there performance benefits from cores=1,threads=1 ?

I am not sure why we need to set cores=1,threads=1.

It's probably implicit, but the idea here is just to match what libvirt is doing, makes things easier to compare.

@walters thank you!

Commit e442d09e fixes this pull-request