[vfio-users] IOMMU page table of QEMU vfio device passthrough

[Huaicheng Li]

Hi Alex,

VFIO does not setup
> mappings per DMA buffer, it maps all of guest physical memory through
> the IOMMU so that the device can DMA to any GPA and it's already
> mapped.  We call this transparent assignment, the guest can perform DMA
> without any knowledge of or interaction with the host IOMMU.


Could you briefly explain how all of guest physical memory is mapped
through IOMMU.
I do see a ``vfio_dma_map`` call with `size` almost equal to my guest
memory size, is
this the place where IOMMU is instructed to enable such transparent
assignment you
mentioned? Since the guest driver is issuing IOs with GPA, how does the
device do
DMA to the corresponding physical DMA address? Is the GPA->HPA a static
mapping
so the HPA can be calculated instead of walking the page tables (which
those DMA APIs
do)?


> It's software, anything is possible, but you'll need to subvert
> something to do that for you.  One subversion we have in place is the
> no-iommu mode of vfio, which does not provide any mapping services and
> taints the host kernel.  This mode is explicitly not intended for
> device assignment, you'll be lucky to boot a VM at all and it will
> probably take the host down in the process, thus the tainting.  Please
> don't report bugs against unsafe options that explicitly taint the
> kernel as an indication not to use them.  Thanks,


Seems this is what I need. I will have a try.

Thank you so much!!

Best,
Huaicheng



On Tue, Jun 5, 2018 at 5:25 PM Alex Williamson <alex.williamson at redhat.com>
wrote:

> On Tue, 5 Jun 2018 15:16:55 -0700
> Huaicheng Li <huaicheng at cs.uchicago.edu> wrote:
>
> > Hi,
> >
> > I assigned a PCIe device to my QEMU VM and learnt that in this case,
> IOMMU
> > maintains GPA->HPA translation. What I
> > want to achieve is to pass commands with host physical addresses to the
> > device from inside the VM and see if the DMA
> > will successfully do that for me.
> >
> > (1). How does VFIO setup the GPA->HPA mapping for each DMA buffer? which
> > source file can I refer to for such information?
> > Currently I'm looking at VFIO_IOMMU_DMA_MAP, it seems it only maps the
> base
> > address of QEMU guest memory region to
> > IOVA. Does this mean that later all GPA->HPA translation are based on
> that
> > plus some offsets?
>
> Unless you enable an emulated IOMMU in the guest, VFIO does not setup
> mappings per DMA buffer, it maps all of guest physical memory through
> the IOMMU so that the device can DMA to any GPA and it's already
> mapped.  We call this transparent assignment, the guest can perform DMA
> without any knowledge of or interaction with the host IOMMU.  If you do
> configure an emulated IOMMU, then the guest device driver will
> configure a mapping through the gIOMMU, which will create a mapping
> through the pIOMMU.  In either case, all of this happens through
> MemoryListeners in the vfio code. It's mostly a difference in which
> AddressSpace the device is using.
>
> > (2). Is it possible for me to disable IOMMU page tables as I just want to
> > give the device a host physical address?
>
> It's software, anything is possible, but you'll need to subvert
> something to do that for you.  One subversion we have in place is the
> no-iommu mode of vfio, which does not provide any mapping services and
> taints the host kernel.  This mode is explicitly not intended for
> device assignment, you'll be lucky to boot a VM at all and it will
> probably take the host down in the process, thus the tainting.  Please
> don't report bugs against unsafe options that explicitly taint the
> kernel as an indication not to use them.  Thanks,
>
> Alex
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/vfio-users/attachments/20180605/f8930d83/attachment.htm>