Comment 12 for bug 1845562

I had this error today.
It was on a 20.04 development test server, upon which I was trying to start a freshly installed 20.04 server guest.
The error occurred because I had this on my grub command line:

cgroup_no_v1=all

Once I removed it, the VM started fine.
Note that I do not use virt-manager, I was using "virsh start serv-dev".

Why did I originally have "cgroup_no_v1=all"?
Well, I forgot to get rid of it, after debugging problems with the schedutil governor for either the intel_cpufreq (intel_pstate in passive mode) or acpi-cpufreq CPU frequency scaling drivers for mainline kernel 5.4, where it gets busted due to "UCLAMP_TASK_GROUP=Y"in the Ubuntu kernel configuration.

I was using kernel 5.3.0-18-generic today.