On 2016/6/15 23:40, Steve Novakov wrote:
Hello Paolo, thank you for joining in,
Yeah absolutely, it freezes no matter what I do. I get all sorts of
interesting error messages during boot. Dmesg goes to like [64.4XXXX] or
higher and spits out random kernel functions, addresses, etc. then
freezes. I've tried
intel_iommu=on
intel_iommu=on vfio_iommu_type1.allow_unsafe_interrupts=1
all sorts of combinations. I've verified that all of the relevant
modules load (vfio* kvm kvm_intel pci_stub) through dmesg. That that
"DMAR-IR" message is suspicious though, no? That there is some setting
in the kernel that disables remapping by default for my BIOS. I've seen
posts with working vt-d even with an i7-950. I wonder if they have the
C3 stepping of the northbridge. Does that sound like it might be an issue?
I can't believe that there is not some way to make this work, as it
works flawlessly with ESXi (which sits on top of a linux kernel, though
it's vmware's proprietary fork I think). If my system boots in "Plug and
play" mode, the BIOS configuration for non critical devices shouldn't
matter, right?
It seems there are something wrong with vfio. Have you tried with legacy
device pass through? It should be disabled on new distribution(like
ubuntu 16.04) but RHEL6 may still support it.
Steve Novakov
B.A.Sc. Engineering Physics
PhD Student - Physics
University of Michigan
On 6/15/2016 2:59 AM, Paolo Bonzini wrote:
On 15/06/2016 07:04, Steve Novakov wrote:
So I swapped over to ubuntu 16.04 server and I just saw this line in
dmesg:
[0.100541] DMAR-IR: This system BIOS has enabled interrupt remapping on
a chipset that contains an erratum making that feature unstable. To
maintain system stability interrupt remapping is being disabled.please
contact your BIOS vendor for an update.
This line was actually present while using arch as well I just never
took the time to read past the first line (which just says enabled
interrupt remapping). This seems to be an issue with northbridge
stepping B3, and was apparently fixed in stepping C2 ( I have B3 : ( ).
I wonder if it's even possible to get this to work with my motherboard
now....**though ESXi can manage just fine...**
Maybe there's some way to trick linux into just trying to do it. I mean,
it seems to pass devices to pci-stub just fine.
Passing devices to pci-stub doesn't require the IOMMU, it's just
swapping the driver that binds to the device. I think this is why with
VT-d disabled you cannot assign the device to vfio-pci.
Does it still freeze with Ubuntu 16.04 (VT-d enabled,
vfio_iommu_type1.allow_unsafe_interrupts=1)?
Thanks,
Paolo
--
best regards
yang
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
