Problem - Using WSL2 in a Windows virtual machine

[fth0]

FWIW, contrary to some statements in the Ubuntu forums' thread, VirtualBox doesn't need contiguous memory on the host, at least not for the VM's RAM (System > Motherboard > Base Memory).

[mdalacu]

Everything is working fine now after installing the latest BIOS!!!
Victory!

[fth0]

Thanks for reporting back!

FYI @mpack, @scottgus1:
It looks like we've had a few users in the last months with AMD Ryzen 7 7840HS and 7940HS CPUs where a BIOS update was needed, independent of NEM (Hyper-V) or AMD-V being used. In all cases, the Linux guest OS panicked during the CPU initialization.

[scottgus1]

Interesting, fth0! I wonder if new processors having trouble running Virtualbox could include a host BIOS update as a possible diagnostic step.

[mpack]

could include a host BIOS update as a possible diagnostic step.

I would hate to start recommending that. I'm of the "don't fix it if it ain't broke" school of thought, and updating the BIOS comes with possibly low but still significant risk of a bricked PC.

If specific manufacturers are recommending it for their users then fine, but I wouldn't want to promote it for them even then. Let users know about the discussion but otherwise waive all responsibility is the most I'd do.

[fth0]

I also wouldn't recommend a BIOS update as a general measure (although BIOS updates could make sense from a security perspective, depending on your security needs). Personally, I'd use the following strategy:

If a user's VBox.log file indicates one of the known problematic CPUs (or a quite similar CPU) trying to run a Linux VM that hangs in the early boot phase ("initrd"), I'd tell the user that other users were able to solve the issue with a BIOS update on the host.

That's even less than my usual "I'd suggest to ..." and leaves the responsibility even more at the user's side.

[scottgus1]

Thanks, Mpack and fth0! No general recommendation, then, but a "hey, you have this CPU, others have done this" might not be a problem.