[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-users] HVM DomU+Dom0 crash with PCI USB passthrough

On Wed, 31 Jul 2013 17:54:26 +0200, Gustav Sorenson <gu.sorenson@xxxxxxxxx> wrote:

What I eventually noticed was that this only seemed to happen
when passing through USB devices that happened to share an
IRQ with something else (in my case it wafs various ICH10
devices sharing interrupts, USB and SMBus).

is there a way I can see which devices share IRQs? Is it possible to
change the IRQ mappings if they are unconvenient?

Point being - make sure that whatever you are passing
through to domU isn't on a shared interrupt. Also
bear in mind that the binding between the physical port
and the PCI USB controller device can change depending
on what you plug into the port.

I'm not sure I understood. Are you saying that when I plugin in device
X in the USB port A, it could get handled by a different PCI device
than if I had plugged in device Y into A instead?

In a word - yes. On my system (ICH10) some USB controllers don't
correspond to any of the ports on the machine (mapped it out).
At the same time, multiple ports can correspond to the same PCI
device that is the USB controller. But - if I plug in a USB hub,
suddenly the PCI device handling that port changes to one of
the ones that didn't seem to handle any of the physical ports.

Weird, I know, but that's what it seems to be doing.

I just installed a USB2 PCI controller card, manufactured by LogiLink
and advertising a NEC Chip. It appears in lspci as
01:06.0 USB controller: NEC Corporation OHCI USB Controller (rev 41)
01:06.1 USB controller: NEC Corporation OHCI USB Controller (rev 41)
01:06.2 USB controller: NEC Corporation uPD72010x USB 2.0 Controller (rev 02)
I assigned all three of these to pciback, rebooted, assigned them to
the DomU and launched it. However, every time I try to start the DomU
with the new USB controller attached, both Dom0 and DomU hang after a
few seconds (this surely is not enough time for DomU to finish
booting, but maybe it started doing that.)

It sounds more like it locks up during initializing the USB driver.
My USB3 ports are handled by a NEC controller and it seems to
work fine for me.

What may be the reason for this? My mainboard has three PCI slots.
Could it make a difference which slot I put the card into?

Plausibly, yes. Can you check lspci -vvv and see if any of those
USB devices are sharing an IRQ with anything else on the system?

It might also be handy to check the output of lspci -t and see
if there is an odd bridging topology in play.


Xen-users mailing list



Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.