Guru meditation, then lockup, then crash, then lockup

[workerbee]

Sequence:

1. Running the following:
   • Ubuntu guest, Windows 10 Host
   • Multiple VDI files representing multiple disk drives
   • VDI files are on an external USB3 SSD
2. Logged in to the Ubuntu guest, running a Xilinx PetaLinux build
3. VBOX halted with a guru meditation
4. Waited a long time for VBox to finish shutting down
5. After a long time, tried to use Task Manager to kill the process
6. Took a while, including multiple attempts, to kill the process
   • Didn't succeed until I closed my open VirtualBox Manager window; maybe coincidence?
7. Started the VM up again
8. Deleted the PetaLinux project and temporary files
9. Started a new Xilinx PetaLinux build
10. Second crash (but apparently not a "guru meditation" this time?)
11. While writing all this have been waiting for "Stopping the virtual machine" to complete, still seems hung up 10 minutes later
12. Just closed both my open Manager windows, "Stopping the virtual machine" still hung
13. Trying first "End Task" from Task Manager
14. Been a couple minutes, trying again...still not responding
15. Task Manager reports 0% CPU and 198.2MB for VirtualBox Virtual Machine

[workerbee]

logs of first crash (guru meditation, crash1 folder) and second crash (crash2 folder...non-guru crash)

[workerbee]

3rd crash, another "guru meditation", while synthesizing a Xilinx Vivado fpga project.

[workerbee]

PNG at time of crash4.zip

[workerbee]

4th crash. Still while using Xilinx tools; the other times were while Xilinx tools were potentially doing lots of memory intensive things and intensive disk I/O.

This time it was just getting ready to download an FPGA image over TCP/IP.

I believe for the earlier crashes I was using the AHCI storage driver? I believe before this crash I switched to the Lsi SAS driver, in the hope that that might change something, but it would seem not.

Vbox is still hanging up at the "stopping the virtual machine" stage, as well.