Re: [PATCH v3] vfio/pci: Fix vf_token mechanism when device-specific VF drivers are used

[Jason Gunthorpe <jgg@nvidia.com>]

On Wed, Apr 13, 2022 at 11:45:25AM -0600, Alex Williamson wrote:
> On Wed, 13 Apr 2022 10:10:36 -0300
> Jason Gunthorpe <jgg@nvidia.com> wrote:
> > @@ -1732,8 +1705,30 @@ static int vfio_pci_bus_notifier(struct notifier_block *nb,
> >  static int vfio_pci_vf_init(struct vfio_pci_core_device *vdev)
> >  {
> >  	struct pci_dev *pdev = vdev->pdev;
> > +	struct vfio_pci_core_device *cur;
> > +	struct pci_dev *physfn;
> >  	int ret;
> >  
> > +	if (pdev->is_virtfn) {
> > +		/*
> > +		 * If this VF was created by our vfio_pci_core_sriov_configure()
> > +		 * then we can find the PF vfio_pci_core_device now, and due to
> > +		 * the locking in pci_disable_sriov() it cannot change until
> > +		 * this VF device driver is removed.
> > +		 */
> > +		physfn = pci_physfn(vdev->pdev);
> > +		mutex_lock(&vfio_pci_sriov_pfs_mutex);
> > +		list_for_each_entry (cur, &vfio_pci_sriov_pfs, sriov_pfs_item) {
>                                    ^
>                                    |
> checkpatch noted the space here ----

Yeah, I usually ignore that.. we don't write "for()" after all..

> Fixed on commit, looks good otherwise.  Applied to vfio for-linus
> branch for v5.18.  Thanks,

Thanks!

Jason