RE: [PATCH v6 00/11] vfio/pci: Allow MMIO regions to be exported through dma-buf

[Shameer Kolothum]

Hi Cédric,

> -----Original Message-----
> From: Cédric Le Goater <clg@xxxxxxxxxx>
> Sent: 17 December 2025 17:02
> To: Nicolin Chen <nicolinc@xxxxxxxxxx>; Leon Romanovsky <leon@xxxxxxxxxx>
> Cc: Bjorn Helgaas <bhelgaas@xxxxxxxxxx>; Logan Gunthorpe
> <logang@xxxxxxxxxxxx>; Jens Axboe <axboe@xxxxxxxxx>; Robin Murphy
> <robin.murphy@xxxxxxx>; Joerg Roedel <joro@xxxxxxxxxx>; Will Deacon
> <will@xxxxxxxxxx>; Marek Szyprowski <m.szyprowski@xxxxxxxxxxx>; Jason
> Gunthorpe <jgg@xxxxxxxx>; Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>;
> Jonathan Corbet <corbet@xxxxxxx>; Sumit Semwal
> <sumit.semwal@xxxxxxxxxx>; Christian König <christian.koenig@xxxxxxx>;
> Alex Williamson <alex.williamson@xxxxxxxxxx>; Kees Cook
> <kees@xxxxxxxxxx>; Gustavo A. R. Silva <gustavoars@xxxxxxxxxx>; Ankit
> Agrawal <ankita@xxxxxxxxxx>; Yishai Hadas <yishaih@xxxxxxxxxx>; Shameer
> Kolothum <skolothumtho@xxxxxxxxxx>; Kevin Tian <kevin.tian@xxxxxxxxx>;
> Krishnakant Jaju <kjaju@xxxxxxxxxx>; Matt Ochs <mochs@xxxxxxxxxx>;
> linux-pci@xxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx; linux-
> block@xxxxxxxxxxxxxxx; iommu@xxxxxxxxxxxxxxx; linux-mm@xxxxxxxxx; linux-
> doc@xxxxxxxxxxxxxxx; linux-media@xxxxxxxxxxxxxxx; dri-
> devel@xxxxxxxxxxxxxxxxxxxxx; linaro-mm-sig@xxxxxxxxxxxxxxxx;
> kvm@xxxxxxxxxxxxxxx; linux-hardening@xxxxxxxxxxxxxxx; Vivek Kasireddy
> <vivek.kasireddy@xxxxxxxxx>
> Subject: Re: [PATCH v6 00/11] vfio/pci: Allow MMIO regions to be exported
> through dma-buf
> 
> External email: Use caution opening links or attachments
> 
> 
> Hello Nicolin,
> 
> On 11/4/25 20:19, Nicolin Chen wrote:
> > On Sun, Nov 02, 2025 at 10:00:48AM +0200, Leon Romanovsky wrote:
> >> Changelog:
> >> v6:
> >>   * Fixed wrong error check from pcim_p2pdma_init().
> >>   * Documented pcim_p2pdma_provider() function.
> >>   * Improved commit messages.
> >>   * Added VFIO DMA-BUF selftest.
> >>   * Added __counted_by(nr_ranges) annotation to struct
> vfio_device_feature_dma_buf.
> >>   * Fixed error unwind when dma_buf_fd() fails.
> >>   * Document latest changes to p2pmem.
> >>   * Removed EXPORT_SYMBOL_GPL from pci_p2pdma_map_type.
> >>   * Moved DMA mapping logic to DMA-BUF.
> >>   * Removed types patch to avoid dependencies between subsystems.
> >>   * Moved vfio_pci_dma_buf_move() in err_undo block.
> >>   * Added nvgrace patch.
> >
> > I have verified this v6 using Jason's iommufd dmabuf branch:
> > https://github.com/jgunthorpe/linux/commits/iommufd_dmabuf/
> >
> > by drafting a QEMU patch on top of Shameer's vSMMU v5 series:
> > https://github.com/nicolinc/qemu/commits/wip/iommufd_dmabuf/
> >
> > with that, I see GPU BAR memory be correctly fetched in the QEMU:
> > vfio_region_dmabuf Device 0009:01:00.0, region "0009:01:00.0 BAR 0",
> > offset: 0x0, size: 0x1000000 vfio_region_dmabuf Device 0009:01:00.0,
> > region "0009:01:00.0 BAR 2", offset: 0x0, size: 0x44f00000
> > vfio_region_dmabuf Device 0009:01:00.0, region "0009:01:00.0 BAR 4",
> > offset: 0x0, size: 0x17a0000000
> >
> > Tested-by: Nicolin Chen <nicolinc@xxxxxxxxxx>
> 
> Do you plan to provide P2P support with IOMMUFD for QEMU ?

Yes. We do plan to send that support patch soon.

Thanks,
Shameer