Comment 26 for bug 1953671

Hi there, VMware here...
We're able to repro this and have the following temporary workaround while we figure out who's code needs adjusting:

We are able to boot by adding:
acpi=force

to: GRUB_CMDLINE_LINUX_DEFAULT
in: /etc/default/grub
Such that the full line reads:
GRUB_CMDLINE_LINUX_DEFAULT="quiet splash acpi=force"

Remember to also disable Wayland (which is not presently supported in this Fusion tech preview) in /etc/gdm3/custom.conf by uncommenting WaylandEnable=false.

When it's in the unresponsive state it still boots so ssh works... you can check /proc/iomem and it seems like the PCI memory is already reserved for BOOTFB..

Not working state:
50000000-7fffffff : pcie@0x40000000
  78000000-7fffffff : 0000:00:0f.0
    78000000-782fffff : BOOTFB

Working state: (i.e. with acpi=force)

50000000-7fffffff : PCI Bus 0000:00
  50000000-501fffff : PCI Bus 0000:20
  50200000-503fffff : PCI Bus 0000:21
  50400000-505fffff : PCI Bus 0000:21
  78000000-782fffff : BOOTFB

We have an internal bug filed so I'll update if we have more we can share.

Thanks,
-Michael