There are only 3. (without XEN which is not linux but rather xenux, without bosch etc). The third one is bought by corporation worth than Oracle or similar and closed source. The second one can be used simultaneously with the first one when it use hypervisor from the first one (KVM) = take the best from 2 worlds.

This is exactly what I`ve wrote already.

P.S. As always, everybody is free to use whatever s/he likes and everybody is free to give an advise based on h.. own experience.

1. The best ssd accelerator for Linux:
https://github.com/firelzrd/adios

2.

If it's not AES then it's too slow.

Yo do not understand what you talk about the openrc IS in Debian,

I understand. And more, I understand that openrc and all other than systemd inits are half-banned in Debian. Why would Debian like to develop them? There is no reason.

OK. But Debian  is systemd while Devuan is not. Devuan follows Debian. It's OK,  the idea is clear. But why does Devuan follows obsolet versions of inits that are not inits in Debian? Here is the difference between Debian and Devuan, why not to improve init? It's done already in other linuxes so it should not be very hard.

It's funny, Oracle decided to move towards KVM https://github.com/VirtualBox/virtualbo … 3848644498

They could not do it before because of the lack of resources. Oracle .... But other people could

Vulnerability exploited over a network https://www.oracle.com/security-alerts/ … pendixOVIR
KVM is better.

why there is no support for user services in openrc? Pipewire starts by services at least in alpine, artix and gentoo. Who knows

https://blog.programster.org/kvm-missin … lt-network

just start the network permanently with

virsh net-autostart default 

What does this

sudo usermod -a -G operator <user>

do?

check your /etc/libvirt/libvirtd.conf/etc/libvirt/libvirtd.conf

It’s worth to have there

unix_sock_group = "libvirt" 
unix_sock_rw_perms = "0770"

in

/etc/libvirt/qemu.conf

set

user = "user" 
group = "kvm"

Do not forget to add user to aall this groups: kvm,  libvirt, *qemu*

then rc-update add libvirtd
and
rc-service libvirtd start (if you are using openrc, if sysvinit or runit then find appropriate commands to enable service)

After reboot you'll have started networ "default", running libvirt and user wil be granted access to use virt-manager and virt-viewer (I find virt-viewer more comfortable to run VMs, then virt-manager which is more suitable for creating them).

Also note that uefi can be broken in virt-manager. last time I have to copy uefi bios from somwhere and I don't know if thhe situation have been changed since.

If  you are looking for alternatives you can try  https://github.com/quickemu-project/quickemu  and GUI for it https://github.com/quickemu-project/quickgui

The gui will download and install guest for you. You can also copy uefi bios from this install in the cse you need it for virt-manager and you can run all these guests in parallel (virt-manager/viewer and quickemu).

More of that you can use https://github.com/ivan-hc/VirtualBox-appimage - this is an appimage of virtualbox that use KVM as hypervisor, not Oracle's core component!  It can be run in parallel with qemu, aqemu, quickemu, virt-manager etc.

Have fun.

Why not use windows though?

Or, ... at least Zorin?

I run it manually

ntpdate -s pool.ntp.org

when my system clock is more than ~ 20 secconds away. I remember also adding it to rc.local to let time to be synchronized once at boot time.

did you believe in BitLocker or in Windows?

sddm boots almost everythhing and in most cases does not require manual intervention.

ryzen_smu for RyzenAdj
ths is due to  the fact that amd pstate on my laptop is wrongly configured using very high frequences while idle.

And again, I've never heared that someting does not work on pure Intel under Linux.