Am 15.08.2017 05:26, schrieb taii...@gmx.com:
On 08/14/2017 11:16 PM, Alex Williamson wrote:
On Mon, Aug 14, 2017 at 9:07 PM, taii...@gmx.com <taii...@gmx.com>
wrote:
I am not sure but I think there are no pci-e audio cards that support
IOMMU AFAIK.
False, especially for device assignment, the drivers in the guest and
the
guest itself is completely unaware of the IOMMU.
I heard an Asus Xonar (cmedia chipset) PCI-e (all of them are based
off that chip AFAIK) wouldn't work and resulted in IO_PAGE_FAULT dmesg
spam with iommu enabled.
https://bbs.archlinux.org/viewtopic.php?id=168555
The statement is definitely false. I'm currently passing the built-in
PCIe
Intel HDA of my X99 motherboard to the VM. The HDA is nicely isolated in
its
own IOMMU group and the VM controls it flawlessly. Furthermore, libvirt
does
automatic unbind/rebind with the host driver upon booting up and
shutting
down the VM. PulseAudio plays nicely along and automatically reroutes
all
active streams to the other sound device on the host, which happens to
be...
an Asus Xonar STX II. When it comes to Xonars, those use a PCI chip with
an
on-card PCIe-to-PCI bridge:
07:00.0 PCI bridge [0604]: ASMedia Technology Inc. ASM1083/1085 PCIe to
PCI Bridge [1b21:1080] (rev 04) (prog-if 00 [Normal decode])
Flags: bus master, fast devsel, latency 0, IRQ 18, NUMA node 0
Bus: primary=07, secondary=08, subordinate=08, sec-latency=32
I/O behind bridge: 0000c000-0000cfff [size=4K]
Memory behind bridge: None
Prefetchable memory behind bridge: None
Capabilities: [50] MSI: Enable- Count=1/1 Maskable- 64bit+
Capabilities: [78] Power Management version 3
Capabilities: [80] Express PCI-Express to PCI/PCI-X Bridge, MSI
00
Capabilities: [c0] Subsystem: Device [0000:0000]
Capabilities: [100] Virtual Channel
Kernel modules: shpchp
08:04.0 Multimedia audio controller [0401]: C-Media Electronics Inc
CMI8788 [Oxygen HD Audio] [13f6:8788]
Subsystem: ASUSTeK Computer Inc. Virtuoso 100 (Xonar Essence STX
II) [1043:85f4]
Flags: bus master, medium devsel, latency 32, IRQ 18, NUMA node
0
I/O ports at c000 [size=256]
Capabilities: [c0] Power Management version 2
Kernel driver in use: snd_virtuoso
Kernel modules: snd_virtuoso
Works with iommu=pt, though I believe it also worked without it (I'm not
really sure why I added the iommu parameter). I still haven't tried
passing
the Xonar to the VM. The two devices appear to be grouped together with
the
PCH PCI bridge. I'll try in the future.
I had a similar error - one of my GFX cards broke so I slotted in an
old PCI-e Nvidia 7800gt and when booting dom0 and trying to do
modesetting with IOMMU enabled it would spew io page faults and fail
to modeset - I had to replace it with a newer card...I do not know why
that occurred.
_______________________________________________
vfio-users mailing list
vfio-users@redhat.com
https://www.redhat.com/mailman/listinfo/vfio-users
_______________________________________________
vfio-users mailing list
vfio-users@redhat.com
https://www.redhat.com/mailman/listinfo/vfio-users