PCI Passthrough error

Issues related to hardware problems
Post Reply
rdeterre
Posts: 6
Joined: 2014/07/15 13:45:44

PCI Passthrough error

Post by rdeterre » 2014/10/22 18:42:45

I'd like to use PCI passthrough to make a Ubuntu 14.04 guest see a pci device plugged in a CentOS 7 host. I get an error when I try to detach my PCI device:

Code: Select all

[rdeterre@localhost ~]$ virsh nodedev-detach pci_0000_03_00_0
error: Failed to detach device pci_0000_03_00_0
error: Operation not supported: neither VFIO nor KVM device assignment is currently supported on this system   
VT-d is enabled both in the BIOS and in linux with the 'intel_iommu' command line argument. The vmx, ept, cpid, tpr_shadow and vnmi flags are set in /proc/cpuinfo. lsmod contains kvm and kvm_intel. Also, dmesg does not show any message complaining about an issue with hardware support or bios for KVM. Instead, "dmesg | grep -e DMAR -e IOMMU " returns "[ 0.000000] Intel-IOMMU: enabled".

Yet, there is nothing in "/sys/kernel/iommu_groups" and from looking at the libvirt source (http://libvirt.org/git/?p=libvirt.git;a ... b=HEAD#l96), it looks like it could cause my issue.

Am I missing something?

Thanks,

Romain

effbiai
Posts: 2
Joined: 2015/03/06 21:02:19

Re: PCI Passthrough error

Post by effbiai » 2015/03/06 21:07:42

Hi,

Did you figure this one out? I have the exact same problem.

Code: Select all

# dmesg|grep IOMMU
[    0.000000] Intel-IOMMU: enabled

]# cat /proc/cpuinfo |grep flags
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc aperfmperf eagerfpu pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr pdcm pcid sse4_1 sse4_2 popcnt tsc_deadline_timer aes xsave avx lahf_lm ida arat epb pln pts dtherm tpr_shadow vnmi flexpriority ept vpid xsaveopt

# virsh nodedev-dettach pci_0000_05_01_0
error: Failed to detach device pci_0000_05_01_0
error: Operation not supported: neither VFIO nor KVM device assignment is currently supported on this system
KVM is however working, the virtual server is running if I don't add a physical PCI device to it;

Code: Select all

# virsh list --all
 Id    Name                           State
----------------------------------------------------
 2     ossim                          running

Post Reply