> From: Jason Gunthorpe <jgg@xxxxxxxxxx>
> Sent: Saturday, October 2, 2021 7:22 AM
>
[...]
> +static void vfio_group_release(struct device *dev)
> {
> - struct vfio_group *group, *existing_group;
> - struct device *dev;
> - int ret, minor;
> + struct vfio_group *group = container_of(dev, struct vfio_group, dev);
> + struct vfio_unbound_dev *unbound, *tmp;
> +
> + list_for_each_entry_safe(unbound, tmp,
> + &group->unbound_list, unbound_next) {
> + list_del(&unbound->unbound_next);
> + kfree(unbound);
> + }
move to vfio_group_put()? this is not paired with vfio_group_alloc()...
[...]
> static int vfio_group_fops_open(struct inode *inode, struct file *filep)
> {
> - struct vfio_group *group;
> + struct vfio_group *group =
> + container_of(inode->i_cdev, struct vfio_group, cdev);
> int opened;
A curiosity question. According to cdev_device_del() any cdev already
open will remain with their fops callable. What prevents vfio_group
from being released after cdev_device_del() has been called? Is it
because cdev open will hold a reference to device thus put_device()
will not hit zero in vfio_group_put()?
Thanks
Kevin
