[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20250930090048.GG324804@unreal>
Date: Tue, 30 Sep 2025 12:00:48 +0300
From: Leon Romanovsky <leon@...nel.org>
To: Alex Williamson <alex.williamson@...hat.com>
Cc: Jason Gunthorpe <jgg@...dia.com>,
Andrew Morton <akpm@...ux-foundation.org>,
Bjorn Helgaas <bhelgaas@...gle.com>,
Christian König <christian.koenig@....com>,
dri-devel@...ts.freedesktop.org, iommu@...ts.linux.dev,
Jens Axboe <axboe@...nel.dk>, Joerg Roedel <joro@...tes.org>,
kvm@...r.kernel.org, linaro-mm-sig@...ts.linaro.org,
linux-block@...r.kernel.org, linux-kernel@...r.kernel.org,
linux-media@...r.kernel.org, linux-mm@...ck.org,
linux-pci@...r.kernel.org, Logan Gunthorpe <logang@...tatee.com>,
Marek Szyprowski <m.szyprowski@...sung.com>,
Robin Murphy <robin.murphy@....com>,
Sumit Semwal <sumit.semwal@...aro.org>,
Vivek Kasireddy <vivek.kasireddy@...el.com>,
Will Deacon <will@...nel.org>
Subject: Re: [PATCH v4 10/10] vfio/pci: Add dma-buf export support for MMIO
regions
On Mon, Sep 29, 2025 at 03:17:49PM -0600, Alex Williamson wrote:
> On Sun, 28 Sep 2025 17:50:20 +0300
> Leon Romanovsky <leon@...nel.org> wrote:
> > +static int validate_dmabuf_input(struct vfio_pci_core_device *vdev,
> > + struct vfio_device_feature_dma_buf *dma_buf,
> > + struct vfio_region_dma_range *dma_ranges,
> > + struct p2pdma_provider **provider)
> > +{
> > + struct pci_dev *pdev = vdev->pdev;
> > + u32 bar = dma_buf->region_index;
> > + resource_size_t bar_size;
> > + u64 sum;
> > + int i;
> > +
> > + if (dma_buf->flags)
> > + return -EINVAL;
> > + /*
> > + * For PCI the region_index is the BAR number like everything else.
> > + */
> > + if (bar >= VFIO_PCI_ROM_REGION_INDEX)
> > + return -ENODEV;
> > +
> > + *provider = pcim_p2pdma_provider(pdev, bar);
> > + if (!provider)
>
> This needs to be IS_ERR_OR_NULL() or the function needs to settle on a
> consistent error return value regardless of CONFIG_PCI_P2PDMA.
pcim_p2pdma_provider() doesn't return errors after split to _init() and _get().
The more accurate check needs to be if (!*provider) and not what is written.
>
> > + return -EINVAL;
> > +
> > + bar_size = pci_resource_len(pdev, bar);
>
> We get to this feature via vfio_pci_core_ioctl_feature(), which is used
> by several variant drivers, some of which mangle the BAR size exposed
> to the user, ex. hisi_acc. I'm afraid this might actually be giving
> dmabuf access to a portion of the BAR that isn't exposed otherwise.
Doe you mean that part?
1185 static int hisi_acc_vf_qm_init(struct hisi_acc_vf_core_device *hisi_acc_vdev)
1186 {
...
1204 * Also the HiSilicon ACC VF devices supported by this driver on
1205 * HiSilicon hardware platforms are integrated end point devices
1206 * and the platform lacks the capability to perform any PCIe P2P
1207 * between these devices.
1208 */
1209
1210 vf_qm->io_base =
1211 ioremap(pci_resource_start(vf_dev, VFIO_PCI_BAR2_REGION_INDEX),
1212 pci_resource_len(vf_dev, VFIO_PCI_BAR2_REGION_INDEX));
1213 if (!vf_qm->io_base)
1214 return -EIO;
1215
According to the comment, it doesn't support p2p and in any case we will
fail that platform in vfio_pci_dma_buf_attach() by taking "default" case:
34 switch (pci_p2pdma_map_type(priv->provider, attachment->dev)) {
35 case PCI_P2PDMA_MAP_THRU_HOST_BRIDGE:
36 break;
37 case PCI_P2PDMA_MAP_BUS_ADDR:
38 /*
39 * There is no need in IOVA at all for this flow.
40 * We rely on attachment->priv == NULL as a marker
41 * for this mode.
42 */
43 return 0;
44 default:
45 return -EINVAL;
46 }
47
>
> > + for (i = 0; i < dma_buf->nr_ranges; i++) {
> > + u64 offset = dma_ranges[i].offset;
> > + u64 len = dma_ranges[i].length;
> > +
> > + if (!PAGE_ALIGNED(offset) || !PAGE_ALIGNED(len))
> > + return -EINVAL;
> > +
> > + if (check_add_overflow(offset, len, &sum) || sum > bar_size)
> > + return -EINVAL;
> > + }
> > +
> > + return 0;
> > +}
> > +
> > +int vfio_pci_core_feature_dma_buf(struct vfio_pci_core_device *vdev, u32 flags,
> > + struct vfio_device_feature_dma_buf __user *arg,
> > + size_t argsz)
> > +{
> > + struct vfio_device_feature_dma_buf get_dma_buf = {};
> > + struct vfio_region_dma_range *dma_ranges;
> > + DEFINE_DMA_BUF_EXPORT_INFO(exp_info);
> > + struct p2pdma_provider *provider;
> > + struct vfio_pci_dma_buf *priv;
> > + int ret;
> > +
> > + ret = vfio_check_feature(flags, argsz, VFIO_DEVICE_FEATURE_GET,
> > + sizeof(get_dma_buf));
> > + if (ret != 1)
> > + return ret;
> > +
> > + if (copy_from_user(&get_dma_buf, arg, sizeof(get_dma_buf)))
> > + return -EFAULT;
> > +
> > + if (!get_dma_buf.nr_ranges)
> > + return -EINVAL;
> > +
> > + dma_ranges = memdup_array_user(&arg->dma_ranges, get_dma_buf.nr_ranges,
> > + sizeof(*dma_ranges));
> > + if (IS_ERR(dma_ranges))
> > + return PTR_ERR(dma_ranges);
> > +
> > + ret = validate_dmabuf_input(vdev, &get_dma_buf, dma_ranges, &provider);
> > + if (ret)
> > + return ret;
>
> goto err_free_ranges;
Thanks
>
> Thanks,
> Alex
>
>
Powered by blists - more mailing lists