Hi, Christian,

On Wed, 2026-01-21 at 10:20 +0100, Christian König wrote:
> On 1/20/26 15:07, Leon Romanovsky wrote:
> > From: Leon Romanovsky <[email protected]>
> > 
> > dma-buf invalidation is performed asynchronously by hardware, so
> > VFIO must
> > wait until all affected objects have been fully invalidated.
> > 
> > Fixes: 5d74781ebc86 ("vfio/pci: Add dma-buf export support for MMIO
> > regions")
> > Signed-off-by: Leon Romanovsky <[email protected]>
> 
> Reviewed-by: Christian König <[email protected]>
> 
> Please also keep in mind that the while this wait for all fences for
> correctness you also need to keep the mapping valid until
> dma_buf_unmap_attachment() was called.

I'm wondering shouldn't we require DMA_RESV_USAGE_BOOKKEEP here, as
*any* unsignaled fence could indicate access through the map?

/Thomas

> 
> In other words you can only redirect the DMA-addresses previously
> given out into nirvana (or a dummy memory or similar), but you still
> need to avoid re-using them for something else.
> 
> Regards,
> Christian.
> 
> > ---
> >  drivers/vfio/pci/vfio_pci_dmabuf.c | 5 +++++
> >  1 file changed, 5 insertions(+)
> > 
> > diff --git a/drivers/vfio/pci/vfio_pci_dmabuf.c
> > b/drivers/vfio/pci/vfio_pci_dmabuf.c
> > index d4d0f7d08c53..33bc6a1909dd 100644
> > --- a/drivers/vfio/pci/vfio_pci_dmabuf.c
> > +++ b/drivers/vfio/pci/vfio_pci_dmabuf.c
> > @@ -321,6 +321,9 @@ void vfio_pci_dma_buf_move(struct
> > vfio_pci_core_device *vdev, bool revoked)
> >                     dma_resv_lock(priv->dmabuf->resv, NULL);
> >                     priv->revoked = revoked;
> >                     dma_buf_move_notify(priv->dmabuf);
> > +                   dma_resv_wait_timeout(priv->dmabuf->resv,
> > +                                        
> > DMA_RESV_USAGE_KERNEL, false,
> > +                                        
> > MAX_SCHEDULE_TIMEOUT);
> >                     dma_resv_unlock(priv->dmabuf->resv);
> >             }
> >             fput(priv->dmabuf->file);
> > @@ -342,6 +345,8 @@ void vfio_pci_dma_buf_cleanup(struct
> > vfio_pci_core_device *vdev)
> >             priv->vdev = NULL;
> >             priv->revoked = true;
> >             dma_buf_move_notify(priv->dmabuf);
> > +           dma_resv_wait_timeout(priv->dmabuf->resv,
> > DMA_RESV_USAGE_KERNEL,
> > +                                 false,
> > MAX_SCHEDULE_TIMEOUT);
> >             dma_resv_unlock(priv->dmabuf->resv);
> >             vfio_device_put_registration(&vdev->vdev);
> >             fput(priv->dmabuf->file);
> > 

Reply via email to