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

Re: [Xen-devel] Support for VMEbus in Xen

  • To: xen-devel <xen-devel@xxxxxxxxxxxxx>
  • From: Paul Manners <paul.manners@xxxxxxxxx>
  • Date: Wed, 13 Aug 2014 14:41:34 +0200
  • Delivery-date: Wed, 13 Aug 2014 12:42:25 +0000
  • List-id: Xen developer discussion <xen-devel.lists.xen.org>

Hi Stefano,

Thanks for your reply. Â

> the only type of device assignment that is supported out of the box (on
> x86) without having to dig too deep into technicalities is PCI
> passthrough.

Are you saying that device assignment on x86-64 for PCI passthrough does not work out the box? Â

> Xen supports remapping one ore more generic MMIO regions and one or more
> interrupts into a guest, by passing irqs= and iomem= in the VM config
> file, see man xl.cfg for more info.
> Could you use those options to get VMEbus running into a guest?

I'm not sure how to determine whether the existing VME driver in the linux kernel exposes MMIO regions but I did find this:Âhttp://epics.sourceforge.net/devlib2/ which sounded promising. ÂAm I right that going this route would allow full hardware virtualization instead of paravirtualization? ÂWith paravirtualization I'd guess VMEbus would have to be supported in QEMU correct? The only information I can find regarding QEMU and VMEbus isÂhttp://fossies.org/dox/qemu-2.1.0/cmd__universe_8c.html Â


Xen-devel mailing list



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