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

Re: [Xen-users] pciback vs. qemu-dm vs. ?


Sorry, am not familiar with the current state of PCI passthrough, so I'm going 
to tackle your second query for now:

> Secondly, thinking that giving a PCI card to the guest is not a good
> idea, got searching for a way to emulate the device, to preserve Dom0
> stability. Ran into the QEMU-DM discussion. So, will this capability be
> usable on para-virtualized non-VT enabled processors? How do we give
> users the capability to generically attach devices to Xen guests that do
> not run the newer processors?

Qemu-dm isn't (for the moment, anyhow) plumbed through to PV guests.  But in 
any case it only emulates platform hardware and a set of "typical" IO 
devices, so it wouldn't be able to pass through random PCI cards...

It does offer the potential to pass through individual USB devices to HVM 
domainsheard good things about how well that works - although maybe things 
have improved.  Right now we're missing a way to pass through USB devices to 
PV domains entirely - there was one in tree in the Xen 2.x timeframe, but it 
suffered from bitrot and was removed.

If you have a dedicated PCI card and really want to use Xen then I suspect 
that the most likely way to do things will be to fiddle with pciback a bit 
more.  Bear in mind that you're making the domain with the PCI card trusted 
(potentially as much as dom0 itself) by giving it a PCI device to play with.


Dave: Just a question. What use is a unicyle with no seat?  And no pedals!
Mark: To answer a question with a question: What use is a skateboard?
Dave: Skateboards have wheels.
Mark: My wheel has a wheel!

Xen-users mailing list



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