virt-manager PCI passthrough gives error "No IOMMU Found" even w/ Intel IOMMU enabled

Question

I am trying to set up PCI passthrough in virt-manager to grant my Debian Wheezy guest VM access to my laptop's Intel Centrino N wireless card.

The Debian default kernel has Intel IOMMU enabled, and I have passed intel_iommu=on to the kernel at boot (by editing /etc/default/grub and setting GRUB_CMDLINE_LINUX="intel_iommu=on", and then running grub-update) ...

Looking at dmesg to verify that IOMMU is indeed enabled, I see: [ 0.000000] Intel-IOMMU: enabled ...

I tried following the instructions here, and used virsh to detach the PCI device before adding it to the VM (i.e. virsh nodedev-detach pci_0000_01_00_0) just to be sure.

However, I cannot boot into the guest VM at all with PCI passthrough enabled for this wireless card, and I am getting the following error (from /var/log/libvirt/libvirtd.log):

2014-09-22 04:45:19.325+0000: 2990: error : qemuMonitorOpenUnix:305 : failed to connect to monitor socket: No such process
2014-09-22 04:45:19.325+0000: 2990: error : qemuProcessWaitForMonitor:1593 : internal error process exited while connecting to monitor: char device redirected to /dev/pts/1
**No IOMMU found.  Unable to assign device "hostdev0"
kvm: -device pci-assign,host=01:00.0,id=hostdev0,configfd=22,bus=pci.0,addr=0x7: Device 'pci-assign' could not be initialized**

Here is the full output from /var/log/libvirt/qemu/:

2014-09-22 04:10:47.002+0000: starting up
LC_ALL=C PATH=/usr/local/sbin:/usr/local/bin:/sbin:/bin:/usr/sbin:/usr/bin HOME=/ QEMU_AUDIO_DRV=none /usr/bin/kvm -S -M pc-1.1 -cpu Nehalem,+rdtscp,+avx,+osxsave,+xsave,+tsc-deadline,+x2apic,+pdcm,+xtpr,+tm2,+est,+vmx,+ds_cpl,+monitor,+dtes64,+pclmuldq,+pbe,+tm,+ht,+ss,+acpi,+ds,+vme -enable-kvm -m 2500 -smp 4,sockets=4,cores=1,threads=1 -name DebianStableSandbox -uuid c18e2808-c98a-2b4d-54f1-00977d3ec1d2 -no-user-config -nodefaults -chardev socket,id=charmonitor,path=/var/lib/libvirt/qemu/DebianStableSandbox.monitor,server,nowait -mon chardev=charmonitor,id=monitor,mode=control -rtc base=utc -no-shutdown -device piix3-usb-uhci,id=usb,bus=pci.0,addr=0x1.0x2 -drive file=/var/lib/libvirt/images/DebianNET.img,if=none,id=drive-virtio-disk0,format=raw -device virtio-blk-pci,scsi=off,bus=pci.0,addr=0x5,drive=drive-virtio-disk0,id=virtio-disk0,bootindex=1 -drive if=none,id=drive-ide0-1-0,readonly=on,format=raw -device ide-cd,bus=ide.1,unit=0,drive=drive-ide0-1-0,id=ide0-1-0 -netdev tap,fd=21,id=hostnet0 -device virtio-net-pci,netdev=hostnet0,id=net0,mac=52:54:00:c5:4a:a9,bus=pci.0,addr=0x3 -chardev pty,id=charserial0 -device isa-serial,chardev=charserial0,id=serial0 -device usb-tablet,id=input0 -vnc 127.0.0.1:0 -vga cirrus -device intel-hda,id=sound0,bus=pci.0,addr=0x4 -device hda-duplex,id=sound0-codec0,bus=sound0.0,cad=0 -device pci-assign,host=01:00.0,id=hostdev0,configfd=22,bus=pci.0,addr=0x7 -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x6
char device redirected to /dev/pts/1
No IOMMU found.  Unable to assign device "hostdev0"
kvm: -device pci-assign,host=01:00.0,id=hostdev0,configfd=22,bus=pci.0,addr=0x7: Device 'pci-assign' could not be initialized
2014-09-22 04:10:47.610+0000: shutting down

I think that the "failed to connect to monitor" message is somehow related. If I use virt-manager to add the wireless card as a PCI device, without first detaching it with virsh nodedev-detach (as mentioned above), then I get an error message from virt-manager saying:

Error starting domain: Unable to read from monitor: Connection reset by peer

Traceback (most recent call last):
  File "/usr/share/virt-manager/virtManager/asyncjob.py", line 45, in cb_wrapper
    callback(asyncjob, *args, **kwargs)
  File "/usr/share/virt-manager/virtManager/asyncjob.py", line 66, in tmpcb
    callback(*args, **kwargs)
  File "/usr/share/virt-manager/virtManager/domain.py", line 1114, in startup
    self._backend.create()
  File "/usr/lib/python2.7/dist-packages/libvirt.py", line 620, in create
    if ret == -1: raise libvirtError ('virDomainCreate() failed', dom=self)
libvirtError: Unable to read from monitor: Connection reset by peer

... and all that shows up in /var/log/libvirt/libvirtd.log in this case is: 2014-09-22 04:48:26.387+0000: 2985: error : qemuMonitorIORead:523 : Unable to read from monitor: Connection reset by peer

In case it is helpful, here is some of the relevant output from lspci -v:

00:00.0 Host bridge: Intel Corporation 2nd Generation Core Processor Family DRAM Controller (rev 09)
    Subsystem: Dell Device 04d8
    Flags: bus master, fast devsel, latency 0
    Capabilities: [e0] Vendor Specific Information: Len=0c <?>
    Kernel driver in use: agpgart-intel

00:1c.0 PCI bridge: Intel Corporation 6 Series/C200 Series Chipset Family PCI Express Root Port 1 (rev b5) (prog-if 00 [Normal decode])
    Flags: bus master, fast devsel, latency 0
    Bus: primary=00, secondary=01, subordinate=01, sec-latency=0
    Memory behind bridge: d1600000-d16fffff
    Capabilities: [40] Express Root Port (Slot+), MSI 00
    Capabilities: [80] MSI: Enable- Count=1/1 Maskable- 64bit-
    Capabilities: [90] Subsystem: Dell Device 04d8
    Capabilities: [a0] Power Management version 2
    Kernel driver in use: pcieport

01:00.0 Network controller: Intel Corporation Centrino Wireless-N 1030 [Rainbow Peak] (rev 34)
    Subsystem: Intel Corporation Centrino Wireless-N 1030 BGN
    Flags: fast devsel, IRQ 16
    Memory at d1600000 (64-bit, non-prefetchable) [size=8K]
    Capabilities: [c8] Power Management version 3
    Capabilities: [d0] MSI: Enable- Count=1/1 Maskable- 64bit+
    Capabilities: [e0] Express Endpoint, MSI 00
    Capabilities: [100] Advanced Error Reporting
    Capabilities: [140] Device Serial Number bc-77-37-ff-ff-0d-3f-90
    Kernel driver in use: pci-stub

Also, just to be clear, I was able to boot into this VM fine before adding the wireless card as a PCI passthrough device, and once I remove it via virt-manager, everything goes back to working as usual.

Does anyone have any idea how to fix this? Thanks!

Solution

Actually, I just found out that this processor supports VT-x but not VT-d, so I can't do IOMMU.

Comments

[StrongBad]

Are you sure your BIOS really supports IOMMU. The Xen wiki suggests that "there are many buggy BIOSes causing Xen to disable IO virtualization because of errors in the BIOS DMAR/ACPI tables."

[J. Taylor]

I am using a Core-i5 2410M processor, and my BIOS is Phoenix SecureCore Tiano ... I can't find any info one way or the other whether this BIOS supports IOMMU or not.

[J. Taylor]

Actually, I just found out that this processor supports VT-x but not VT-d, so I can't do IOMMU.

[William Levi]

I want to know if the problem was solved by the author, as if the latest Linux version of the kernel suggested abolishing the KVM_ASSIGN mechanism and supporting only vfio

[Amitav Pajni]

@WilliamLevi The problem was solved by the author; he even posted an answer about it. If you read his answer, you will see his solution. Though it may or may not apply to you because your situation is different.