Re: trying to figure out why VT-d isn't working on my kernel --solved

[Chris Wright]

* Chris Friesen (chris.friesen@xxxxxxxxxxx) wrote:
> On 03/06/2012 10:42 PM, Chris Friesen wrote:
> >I've got a current (pulled today) kernel running on a Xeon E5-2648L CPU
> >on an Intel motherboard, and I'm trying to get VT-d working without much success.
> >
> >lspci -vv (version 3.1.4) doesn't show anything related to IOV.
> >(Full output below.)  Since I know the I350 devices are capable of virtual
> >functions, am I correct in assuming it's something similar to the situation
> >described at https://bugzilla.redhat.com/show_bug.cgi?id=652210 and there is
> >a problem with my BIOS?  Based on comment 15 there, the fact that all
> >the DevCtl2 lines show "ARIFwd-" seems interesting.
> >
> >Among other things, I've enabled the following in the kernel config:
> >
> >CONFIG_DMAR_TABLE=y
> >CONFIG_INTEL_IOMMU=y
> >CONFIG_INTEL_IOMMU_DEFAULT_ON=y
> >CONFIG_PCI_MSI=y
> >CONFIG_PCI_STUB=y
> >CONFIG_PCI_IOV=y
> 
> Just in case anyone else runs into this, it turns out that enabling
> CONFIG_PCI_MMCONFIG allowed the virtual functions to be detected and
> configured.

Yeah, you need to be able to access PCI Extended Configuration Space to
enable SR-IOV.

thanks,
-chris
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/