Host Reboot when using PCIe Passthrough

[cjones]

Hello,

I have a VM that I'm passing a GPU through to. The passthrough itself works great. The issue is that whenever this VM is powered on, the host will reboot without any interaction from me. The reboot happens anywhere from 3 - 15 minutes after the VM has been powered on. I have many other VMs that don't cause this. The only difference between them and this one is the passthrough GPU. Attached are some potentially helpful outputs. The syslogs have been truncated from when the VM had been powered on to the last entry right before the host rebooted.

Thanks,
Craig

 

[oguz]

hi,

it seems like the reboot doesn't happen gracefully (at least according to the syslog you provided, but maybe you cut these part out so i can't tell). i also see the line

usb 3-12: reset full-speed USB device number 5 using xhci_hcd

popping up quite often.. so what is this device exactly?

another thing you can do is to check the `dmesg` output around the times of reboot which could help uncover more (maybe a kernel crash is happening or something of similar nature)

 

[cjones]

i also see the line

usb 3-12: reset full-speed USB device number 5 using xhci_hcd

popping up quite often.. so what is this device exactly?

That device is a USB controller that I have passed-through to the VM also. Here is the latest dmesg entries showing that device:

[    1.920067] usb 3-12: new full-speed USB device number 5 using xhci_hcd
[    2.069714] usb 3-12: New USB device found, idVendor=0e8f, idProduct=0003
[    2.069716] usb 3-12: New USB device strings: Mfr=1, Product=2, SerialNumber=0
[    2.069717] usb 3-12: Product: PLAYSTATION(R)3 Controller
[    2.069718] usb 3-12: Manufacturer: Sony

I will start the VM without that device attached and see if I still experience the reboot. I will also double check for any kernel crash events.

 

[oguz]

hi,

any update?

 

[cjones]

hi,

any update?

Apologies for the delay,

I removed the passthrough USB device also and was still getting host reboots.

I attempted configuring things to get a kernel crash dump, but did not have any luck. Have not had a chance to try something else. Is this the recommended method?

Thanks,
Craig

 

[cjones]

No more reboots after swapping the GPU with another one. I guess there's something funky with the original GPU I was using. Thanks for the suggestions!