Hi Alex,
> On 01/23/2016 04:23 AM, Alex Williamson wrote:
> > Using iommu_present() to determine whether an IOMMU group is real or
> > fake has some problems. First, apparently Power systems don't
> > register an IOMMU on the device bus, so the groups and containers get
> > marked as noiommu and then won't bind to their actual IOMMU driver.
> > Second, I expect we'll run into the same issue as we try to support
> > vGPUs through vfio, since they're likely to emulate this behavior of
> > creating an IOMMU group on a virtual device and then providing a vfio
> > IOMMU backend tailored to the sort of isolation they provide, which
> > won't necessarily be fully compatible with the IOMMU API.
> >
> > The solution here is to use the existing iommudata interface to IOMMU
> > groups, which allows us to easily identify the fake groups we've
> > created for noiommu purposes. The iommudata we set is purely
> > arbitrary since we're only comparing the address, so we use the
> > address of the noiommu switch itself.
> >
> > Reported-by: Alexey Kardashevskiy <aik at ozlabs.ru>
> > Fixes: 03a76b60f8ba ("vfio: Include No-IOMMU mode")
> > Signed-off-by: Alex Williamson <alex.williamson at redhat.com>
>
>
>
> Reviewed-by: Alexey Kardashevskiy <aik at ozlabs.ru>
> Tested-by: Alexey Kardashevskiy <aik at ozlabs.ru>
Tested bringing the NIC's up, encountered no issues. Curious if it also works
for Santosh (CC'd) as he's one of the intended users of the No-IOMMU
functionality, but otherwise seems to work.
Thanks,
Anatoly
