* Re: [PATCH v4 1/3] vfio: add dma-buf get_tph callback and DMA_BUF_TPH feature
2026-05-21 22:04 ` [PATCH v4 1/3] vfio: add dma-buf get_tph callback and DMA_BUF_TPH feature Alex Williamson
@ 2026-05-21 22:24 ` Alex Williamson
0 siblings, 0 replies; 2+ messages in thread
From: Alex Williamson @ 2026-05-21 22:24 UTC (permalink / raw)
To: Zhiping Zhang
Cc: Jason Gunthorpe, Leon Romanovsky, Bjorn Helgaas, kvm, linux-rdma,
linux-pci, netdev, dri-devel, Keith Busch, Yochai Cohen,
Yishai Hadas, alex
On Thu, 21 May 2026 16:04:12 -0600
Alex Williamson <alex@shazbot.org> wrote:
> On Tue, 19 May 2026 13:13:49 -0700
> Zhiping Zhang <zhipingz@meta.com> wrote:
>
> > Add a dma-buf get_tph callback for exporters to return TPH
> > (TLP Processing Hints) metadata, and add VFIO_DEVICE_FEATURE_DMA_BUF_TPH
> > so userspace can attach that metadata to a VFIO-exported dma-buf.
>
> This should be two patches, the first extending the dma-buf framework
> for the get_tph callback for explicit approval from dma-buf maintainers
> (who are not even copied here). The second the vfio-pci implementation
> of get_tph.
>
> > 8-bit ST and 16-bit Extended ST are distinct PCIe TPH namespaces; the
> > uAPI carries both with explicit validity flags so importers get the
> > value matching their requested width. SET is write-once per dma-buf;
> > the existing VFIO_DEVICE_FEATURE_DMA_BUF uAPI is unchanged.
>
> I didn't see what motivated this write-once change, I thought we
> understood that it was a userspace problem that the tph values need to
> be set before providing the dma-buf fd to the importer and that races
> relative to that are a userspace ordering problem. Write-once seems
> unnecessarily restrictive and there's no justification provided here.
>
> > Signed-off-by: Zhiping Zhang <zhipingz@meta.com>
> > ---
> > drivers/vfio/pci/vfio_pci_core.c | 3 +
> > drivers/vfio/pci/vfio_pci_dmabuf.c | 134 +++++++++++++++++++++++++++--
> > drivers/vfio/pci/vfio_pci_priv.h | 12 +++
> > include/linux/dma-buf.h | 21 +++++
> > include/uapi/linux/vfio.h | 35 ++++++++
> > 5 files changed, 198 insertions(+), 7 deletions(-)
> >
> > diff --git a/drivers/vfio/pci/vfio_pci_core.c b/drivers/vfio/pci/vfio_pci_core.c
> > index 3f8d093aacf8..94aa6dd95701 100644
> > --- a/drivers/vfio/pci/vfio_pci_core.c
> > +++ b/drivers/vfio/pci/vfio_pci_core.c
> > @@ -1534,6 +1534,9 @@ int vfio_pci_core_ioctl_feature(struct vfio_device *device, u32 flags,
> > return vfio_pci_core_feature_token(vdev, flags, arg, argsz);
> > case VFIO_DEVICE_FEATURE_DMA_BUF:
> > return vfio_pci_core_feature_dma_buf(vdev, flags, arg, argsz);
> > + case VFIO_DEVICE_FEATURE_DMA_BUF_TPH:
> > + return vfio_pci_core_feature_dma_buf_tph(vdev, flags, arg,
> > + argsz);
> > default:
> > return -ENOTTY;
> > }
> > diff --git a/drivers/vfio/pci/vfio_pci_dmabuf.c b/drivers/vfio/pci/vfio_pci_dmabuf.c
> > index f87fd32e4a01..be1c65385670 100644
> > --- a/drivers/vfio/pci/vfio_pci_dmabuf.c
> > +++ b/drivers/vfio/pci/vfio_pci_dmabuf.c
> > @@ -19,7 +19,24 @@ struct vfio_pci_dma_buf {
> > u32 nr_ranges;
> > struct kref kref;
> > struct completion comp;
> > - u8 revoked : 1;
> > + /*
> > + * TPH metadata published by VFIO_DEVICE_FEATURE_DMA_BUF_TPH and
> > + * consumed by the @get_tph dma-buf callback.
> > + *
> > + * @tph_flags is the publish/consume gate: writers populate
> > + * @steering_tag, @steering_tag_ext and @ph first, then store
> > + * @tph_flags with smp_store_release(); readers do
> > + * smp_load_acquire(&tph_flags) before accessing the value fields.
> > + * @tph_flags == 0 means "TPH not set". Writers publish a non-zero
> > + * value only once per dma-buf and serialize via vdev->memory_lock;
> > + * readers stay lockless to avoid AB-BA against the dma_resv_lock held
> > + * by importers.
> > + */
>
> Can you outline the ABBA hazard, I'm not seeing it. You're acquiring
> memory_lock in the feature SET and dma_resv_lock doesn't appear to be
> held when calling .get_tph(). There's a lot of lockless complication
> here balanced on this claim of avoiding a hazard that doesn't appear
> present.
>
> > + u32 tph_flags;
> > + u16 steering_tag_ext;
> > + u8 steering_tag;
> > + u8 ph;
> > + bool revoked;
>
> If we still used memory_lock for tph, these could be:
>
> u8 tph_st_valid:1; /* memory_lock */
> u8 tph_st_ext_valid:1; /* memory_lock */
> u8 tph_ph:2; /* memory_lock */
> u8 tph_st;
> u16 tph_st_ext;
> u8 revoked:1; /* dma_resv_lock */
>
> The existing change of @revoked from bitfield to bool has no rationale
> noted for it in the commit log.
On second thought, what dependency does anything here have on
memory_lock? I think we're jumping through hoops to avoid a lock we
don't even need. If we just want to serialize SET vs get_tph we could
have a mutex on the dma-buf structure, or use RCU if we want to manage
it locklessly and make sure get_tph always sees a fully consistent set
of values. Thanks,
Alex
> > };
> >
> > static int vfio_pci_dma_buf_attach(struct dma_buf *dmabuf,
> > @@ -69,6 +86,36 @@ vfio_pci_dma_buf_map(struct dma_buf_attachment *attachment,
> > return ret;
> > }
> >
> > +static int vfio_pci_dma_buf_get_tph(struct dma_buf *dmabuf, u16 *steering_tag,
> > + u8 *ph, u8 st_width)
> > +{
> > + struct vfio_pci_dma_buf *priv = dmabuf->priv;
> > + u32 flags;
> > +
> > + /* Pair with the smp_store_release() in VFIO_DEVICE_FEATURE_DMA_BUF_TPH. */
> > + flags = smp_load_acquire(&priv->tph_flags);
> > + if (!flags)
> > + return -EOPNOTSUPP;
> > +
> > + switch (st_width) {
> > + case 8:
> > + if (!(flags & VFIO_DMA_BUF_TPH_ST))
> > + return -EOPNOTSUPP;
> > + *steering_tag = priv->steering_tag;
> > + break;
> > + case 16:
> > + if (!(flags & VFIO_DMA_BUF_TPH_ST_EXT))
> > + return -EOPNOTSUPP;
> > + *steering_tag = priv->steering_tag_ext;
> > + break;
> > + default:
> > + return -EINVAL;
> > + }
> > +
> > + *ph = priv->ph;
> > + return 0;
> > +}
> > +
> > static void vfio_pci_dma_buf_unmap(struct dma_buf_attachment *attachment,
> > struct sg_table *sgt,
> > enum dma_data_direction dir)
> > @@ -84,16 +131,17 @@ static void vfio_pci_dma_buf_unmap(struct dma_buf_attachment *attachment,
> > static void vfio_pci_dma_buf_release(struct dma_buf *dmabuf)
> > {
> > struct vfio_pci_dma_buf *priv = dmabuf->priv;
> > + struct vfio_pci_core_device *vdev = READ_ONCE(priv->vdev);
> >
> > /*
> > * Either this or vfio_pci_dma_buf_cleanup() will remove from the list.
> > * The refcount prevents both.
> > */
> > - if (priv->vdev) {
> > - down_write(&priv->vdev->memory_lock);
> > + if (vdev) {
> > + down_write(&vdev->memory_lock);
> > list_del_init(&priv->dmabufs_elm);
> > - up_write(&priv->vdev->memory_lock);
> > - vfio_device_put_registration(&priv->vdev->vdev);
> > + up_write(&vdev->memory_lock);
> > + vfio_device_put_registration(&vdev->vdev);
> > }
> > kfree(priv->phys_vec);
> > kfree(priv);
>
>
> This seems unnecessary. I think this is just because priv->vdev is now
> (unnecessarily) set via WRITE_ONCE, right? These are very well ordered
> paths, prior to exposing the dma-buf, while the device is opened, during
> release, after release. They don't seem to need the READ/WRITE_ONCE
> treatment. This looks like noise from trying to make it lockless.
>
>
> > @@ -101,6 +149,7 @@ static void vfio_pci_dma_buf_release(struct dma_buf *dmabuf)
> >
> > static const struct dma_buf_ops vfio_pci_dmabuf_ops = {
> > .attach = vfio_pci_dma_buf_attach,
> > + .get_tph = vfio_pci_dma_buf_get_tph,
> > .map_dma_buf = vfio_pci_dma_buf_map,
> > .unmap_dma_buf = vfio_pci_dma_buf_unmap,
> > .release = vfio_pci_dma_buf_release,
> > @@ -269,7 +318,7 @@ int vfio_pci_core_feature_dma_buf(struct vfio_pci_core_device *vdev, u32 flags,
> > goto err_free_priv;
> > }
> >
> > - priv->vdev = vdev;
> > + WRITE_ONCE(priv->vdev, vdev);
> > priv->nr_ranges = get_dma_buf.nr_ranges;
> > priv->size = length;
> > ret = vdev->pci_ops->get_dmabuf_phys(vdev, &priv->provider,
> > @@ -331,6 +380,77 @@ int vfio_pci_core_feature_dma_buf(struct vfio_pci_core_device *vdev, u32 flags,
> > return ret;
> > }
> >
> > +int vfio_pci_core_feature_dma_buf_tph(struct vfio_pci_core_device *vdev,
> > + u32 flags,
> > + struct vfio_device_feature_dma_buf_tph __user *arg,
> > + size_t argsz)
> > +{
> > + struct vfio_device_feature_dma_buf_tph set_tph;
> > + struct vfio_pci_dma_buf *priv;
> > + struct dma_buf *dmabuf;
> > + int ret;
> > +
> > + ret = vfio_check_feature(flags, argsz, VFIO_DEVICE_FEATURE_SET,
> > + sizeof(set_tph));
> > + if (ret != 1)
> > + return ret;
> > +
> > + if (copy_from_user(&set_tph, arg, sizeof(set_tph)))
> > + return -EFAULT;
> > +
> > + if (set_tph.flags & ~(VFIO_DMA_BUF_TPH_ST | VFIO_DMA_BUF_TPH_ST_EXT))
> > + return -EINVAL;
> > +
> > + if (!set_tph.flags)
> > + return -EINVAL;
> > +
> > + /* PCIe TLP Processing Hint is a 2-bit field. */
> > + if (set_tph.ph & ~0x3)
> > + return -EINVAL;
> > +
> > + dmabuf = dma_buf_get(set_tph.dmabuf_fd);
> > + if (IS_ERR(dmabuf))
> > + return PTR_ERR(dmabuf);
> > +
> > + if (dmabuf->ops != &vfio_pci_dmabuf_ops) {
> > + ret = -EINVAL;
> > + goto out_put;
> > + }
> > +
> > + priv = dmabuf->priv;
> > + down_write(&vdev->memory_lock);
> > + if (READ_ONCE(priv->vdev) != vdev) {
> > + ret = -EINVAL;
> > + goto out_unlock;
> > + }
> > +
> > + /*
> > + * TPH metadata is write-once per dma-buf so that lockless readers only
> > + * have to observe a single release-published transition from 0 -> flags.
> > + */
> > + if (READ_ONCE(priv->tph_flags)) {
> > + ret = -EBUSY;
> > + goto out_unlock;
> > + }
> > +
> > + priv->steering_tag = set_tph.steering_tag;
> > + priv->steering_tag_ext = set_tph.steering_tag_ext;
> > + priv->ph = set_tph.ph;
> > + /*
> > + * Publish the TPH values before the gate flag, so that lockless
> > + * readers in vfio_pci_dma_buf_get_tph() see fully-initialized
> > + * fields once they observe a non-zero tph_flags.
> > + */
> > + smp_store_release(&priv->tph_flags, set_tph.flags);
> > + ret = 0;
> > +
> > +out_unlock:
> > + up_write(&vdev->memory_lock);
> > +out_put:
> > + dma_buf_put(dmabuf);
> > + return ret;
> > +}
> > +
> > void vfio_pci_dma_buf_move(struct vfio_pci_core_device *vdev, bool revoked)
> > {
> > struct vfio_pci_dma_buf *priv;
> > @@ -388,7 +508,7 @@ void vfio_pci_dma_buf_cleanup(struct vfio_pci_core_device *vdev)
> >
> > dma_resv_lock(priv->dmabuf->resv, NULL);
> > list_del_init(&priv->dmabufs_elm);
> > - priv->vdev = NULL;
> > + WRITE_ONCE(priv->vdev, NULL);
> > priv->revoked = true;
> > dma_buf_invalidate_mappings(priv->dmabuf);
> > dma_resv_wait_timeout(priv->dmabuf->resv,
> > diff --git a/drivers/vfio/pci/vfio_pci_priv.h b/drivers/vfio/pci/vfio_pci_priv.h
> > index fca9d0dfac90..c58f369be4b3 100644
> > --- a/drivers/vfio/pci/vfio_pci_priv.h
> > +++ b/drivers/vfio/pci/vfio_pci_priv.h
> > @@ -118,6 +118,10 @@ static inline bool vfio_pci_is_vga(struct pci_dev *pdev)
> > 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);
> > +int vfio_pci_core_feature_dma_buf_tph(struct vfio_pci_core_device *vdev,
> > + u32 flags,
> > + struct vfio_device_feature_dma_buf_tph __user *arg,
> > + size_t argsz);
> > void vfio_pci_dma_buf_cleanup(struct vfio_pci_core_device *vdev);
> > void vfio_pci_dma_buf_move(struct vfio_pci_core_device *vdev, bool revoked);
> > #else
> > @@ -128,6 +132,14 @@ vfio_pci_core_feature_dma_buf(struct vfio_pci_core_device *vdev, u32 flags,
> > {
> > return -ENOTTY;
> > }
> > +
> > +static inline int
> > +vfio_pci_core_feature_dma_buf_tph(struct vfio_pci_core_device *vdev, u32 flags,
> > + struct vfio_device_feature_dma_buf_tph __user *arg,
> > + size_t argsz)
> > +{
> > + return -ENOTTY;
> > +}
> > static inline void vfio_pci_dma_buf_cleanup(struct vfio_pci_core_device *vdev)
> > {
> > }
> > diff --git a/include/linux/dma-buf.h b/include/linux/dma-buf.h
> > index d1203da56fc5..49eb6ad644a2 100644
> > --- a/include/linux/dma-buf.h
> > +++ b/include/linux/dma-buf.h
> > @@ -113,6 +113,27 @@ struct dma_buf_ops {
> > */
> > void (*unpin)(struct dma_buf_attachment *attach);
> >
> > + /**
> > + * @get_tph:
> > + * @dmabuf: DMA buffer for which to retrieve TPH metadata
> > + * @steering_tag: Returns the raw TPH steering tag for @st_width
> > + * @ph: Returns the TPH processing hint (2-bit value)
> > + * @st_width: Consumer's supported steering tag width in bits (8 or 16)
> > + *
> > + * Return the TPH (TLP Processing Hints) metadata associated with this
> > + * DMA buffer for the requested steering-tag width. 8-bit ST and 16-bit
> > + * Extended ST are distinct namespaces in the PCIe TPH ST table and may
> > + * both be present with different values, so the exporter must select the
> > + * value that matches @st_width and must not substitute one for the other.
> > + *
> > + * Return 0 on success, -EOPNOTSUPP if no metadata is available for the
> > + * requested width, or -EINVAL if @st_width is not 8 or 16.
> > + *
> > + * This callback is optional.
> > + */
> > + int (*get_tph)(struct dma_buf *dmabuf, u16 *steering_tag, u8 *ph,
> > + u8 st_width);
> > +
> > /**
> > * @map_dma_buf:
> > *
> > diff --git a/include/uapi/linux/vfio.h b/include/uapi/linux/vfio.h
> > index 5de618a3a5ee..a9cb6cbc6ade 100644
> > --- a/include/uapi/linux/vfio.h
> > +++ b/include/uapi/linux/vfio.h
> > @@ -1534,6 +1534,41 @@ struct vfio_device_feature_dma_buf {
> > */
> > #define VFIO_DEVICE_FEATURE_MIG_PRECOPY_INFOv2 12
> >
> > +/**
> > + * Upon VFIO_DEVICE_FEATURE_SET associate TPH (TLP Processing Hints) metadata
> > + * with a vfio-exported dma-buf. The dma-buf must have been created by
> > + * VFIO_DEVICE_FEATURE_DMA_BUF on this device.
> > + *
> > + * dmabuf_fd is the file descriptor returned by VFIO_DEVICE_FEATURE_DMA_BUF.
> > + *
> > + * 8-bit ST (steering_tag) and 16-bit Extended ST (steering_tag_ext) are
> > + * distinct namespaces in the PCIe TPH ST table and may both be present with
> > + * different values. Userspace should populate the value(s) it has from the
> > + * firmware ST table for this device and set the matching VFIO_DMA_BUF_TPH_ST /
> > + * VFIO_DMA_BUF_TPH_ST_EXT bit in @flags. An importer requests a specific
> > + * width and receives the matching value; if the requested width is not
> > + * present, the importer is told TPH is unavailable for this dma-buf.
> > + *
> > + * ph is the 2-bit TLP Processing Hint and must be in the range [0, 3].
> > + *
> > + * The user must set TPH on the dma-buf before the importer consumes it.
> > + * TPH metadata is write-once per dma-buf; a second SET returns -EBUSY.
> > + *
> > + * Return: 0 on success, -errno on failure.
> > + */
> > +#define VFIO_DEVICE_FEATURE_DMA_BUF_TPH 13
> > +
> > +#define VFIO_DMA_BUF_TPH_ST (1 << 0) /* steering_tag valid */
> > +#define VFIO_DMA_BUF_TPH_ST_EXT (1 << 1) /* steering_tag_ext valid */
> > +
> > +struct vfio_device_feature_dma_buf_tph {
> > + __s32 dmabuf_fd;
> > + __u32 flags;
> > + __u8 steering_tag;
> > + __u8 ph;
> > + __u16 steering_tag_ext;
> > +};
>
> Sure is tempting to make the ph field the first 2-bits of u8 flags.
> Thanks,
>
> Alex
>
> > +
> > /* -------- API for Type1 VFIO IOMMU -------- */
> >
> > /**
>
^ permalink raw reply [flat|nested] 2+ messages in thread