On Mon, 2 Dec 2024 at 16:07, Shameerali Kolothum Thodi <shameerali.kolothum.th...@huawei.com> wrote: > > > > > -----Original Message----- > > From: Zhangfei Gao <zhangfei....@linaro.org> > > Sent: Monday, December 2, 2024 6:05 AM > > To: Jason Gunthorpe <j...@nvidia.com> > > Cc: Nicolin Chen <nicol...@nvidia.com>; Shameerali Kolothum Thodi > > <shameerali.kolothum.th...@huawei.com>; Eric Auger > > <eric.au...@redhat.com>; Mostafa Saleh <smost...@google.com>; qemu- > > a...@nongnu.org; qemu-devel@nongnu.org; Peter Maydell > > <peter.mayd...@linaro.org>; Jean-Philippe Brucker <jean- > > phili...@linaro.org>; Moritz Fischer <m...@kernel.org>; Michael Shavit > > <msha...@google.com>; Andrea Bolognani <abolo...@redhat.com>; > > Michael S. Tsirkin <m...@redhat.com>; Peter Xu <pet...@redhat.com>; > > nath...@nvidia.com; ari...@nvidia.com; i...@nvidia.com; > > j...@nvidia.com; mo...@nvidia.com > > Subject: Re: nested-smmuv3 topic for QEMU/libvirt, Nov 2024 > > > > Hi, Nico > > > > On Fri, 1 Nov 2024 at 19:55, Jason Gunthorpe <j...@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.william...@redhat.com/
Thanks Shameer
