RE: nested-smmuv3 topic for QEMU/libvirt, Nov 2024

[Shameerali Kolothum Thodi]

> -----Original Message-----
> From: Zhangfei Gao <zhangfei.gao@linaro.org>
> Sent: Monday, December 2, 2024 6:05 AM
> To: Jason Gunthorpe <jgg@nvidia.com>
> Cc: Nicolin Chen <nicolinc@nvidia.com>; Shameerali Kolothum Thodi
> <shameerali.kolothum.thodi@huawei.com>; Eric Auger
> <eric.auger@redhat.com>; Mostafa Saleh <smostafa@google.com>; qemu-
> arm@nongnu.org; qemu-devel@nongnu.org; Peter Maydell
> <peter.maydell@linaro.org>; Jean-Philippe Brucker <jean-
> philippe@linaro.org>; Moritz Fischer <mdf@kernel.org>; Michael Shavit
> <mshavit@google.com>; Andrea Bolognani <abologna@redhat.com>;
> Michael S. Tsirkin <mst@redhat.com>; Peter Xu <peterx@redhat.com>;
> nathanc@nvidia.com; arighi@nvidia.com; ianm@nvidia.com;
> jan@nvidia.com; mochs@nvidia.com
> Subject: Re: nested-smmuv3 topic for QEMU/libvirt, Nov 2024
> 
> Hi, Nico
> 
> On Fri, 1 Nov 2024 at 19:55, Jason Gunthorpe <jgg@nvidia.com> wrote:
> >
> > On Thu, Oct 31, 2024 at 09:09:20PM -0700, Nicolin Chen wrote:
> >
> > > FWIW, Robin requested a different solution for MSI mapping [1], v.s.
> > > the RMR one that we have been using since Eric's work. I drafted a
> > > few VFIO/IOMMUFD patches for that,
> >
> > I also talked to MarcZ about this at LPC and he seems willing to
> > consider it. It took a bit to explain everything though. So I think we
> > should try in Nov/Dec
> 
> When boot qemu, reports this
> 
> qemu-system-aarch64: warning: IOMMU_IOAS_MAP failed: Bad address, PCI
> BAR?
> qemu-system-aarch64: vfio_container_dma_map(0xaaaadd30f110,
> 0x8000200000, 0x10000, 0xffffb8031000) = -14 (Bad address)
> qemu-system-aarch64: warning: IOMMU_IOAS_MAP failed: Bad address, PCI
> BAR?
> qemu-system-aarch64: vfio_container_dma_map(0xaaaadd2bc310,
> 0x8000200000, 0x10000, 0xffffb8031000) = -14 (Bad address)
> qemu-system-aarch64: warning: IOMMU_IOAS_MAP failed: Bad address, PCI
> BAR?
> qemu-system-aarch64: vfio_container_dma_map(0xaaaadcf90000,
> 0x8000200000, 0x10000, 0xffffb8031000) = -14 (Bad address)
> 
> Will this also be solved in the new MSI mapping patchset?

Nope. These are not related to MSIs. These mappings are required for 
P2P DMA between devices and is not supported by IOMMUFD at the moment.

See the discussion here,
https://lore.kernel.org/all/20220426103507.5693a0ca.alex.williamson@redhat.com/

Thanks,
Shameer