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

[Xen-API] Xenserver 6.2 pci passtrhough problem: Internal error: xenopsd internal error: Xenctrl.Error("14: Bad address")

Hi all,

I'm not sure if this would be the right place to ask this since it's about xenserver, still I want to exploit any help I can get ;)

The Problem:
As it turns out I can run 2 VM's with passedthrough pci devices without any problems, but when I try to start a third one I get the "Internal error: xenopsd internal error: Xenctrl.Error("14: Bad address")" error which I can't seem to get rid off... I tried any possible combination of booting the VM's but I can get only two working at the same time.

When looking at the xl dmesg log this looks most interesting:
(XEN) AMD-Vi: Setup I/O page table: device id = 0x0200, root table = 0x284120000, domain = 3, paging mode = 2
(XEN) AMD-Vi: Re-assign 02:00.0 from domain 0 to domain 3
(XEN) AMD-Vi: Cannot allocate I/O page table
(XEN) AMD-Vi: Invalid IO pagetable entry gfn = 13ec00
(XEN) domain_crash called from iommu_map.c:585

But I've no idea what might cause this and how to fix it...

System summary:
  • mainboard: Gigabyte GA-990FXA-UD3 rev. 1.2 (modified bios by gigabyte to get pci passthrough working in the first place)
  • Cpu: AMD FX8350
  • RAM: 16GB
  • GPU's: 2x AMD HD5870 + nVidia GTS450

Other possible relevant info is in the attached files...

Thanks in advance for any help!


Attachment: xl_dmesg_error.txt
Description: Text document

Attachment: dmesg.txt
Description: Text document

Attachment: lspci.txt
Description: Text document

Attachment: extlinux.conf
Description: Text document

Attachment: VM1.txt
Description: Text document

Attachment: VM2.txt
Description: Text document

Attachment: VM3.txt
Description: Text document

Xen-api mailing list



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