void *iommu_data;
};
+struct vfio_unbound_dev {
+ struct device *dev;
+ struct list_head unbound_next;
+};
+
struct vfio_group {
struct kref kref;
int minor;
struct notifier_block nb;
struct list_head vfio_next;
struct list_head container_next;
+ struct list_head unbound_list;
+ struct mutex unbound_lock;
atomic_t opened;
};
kref_init(&group->kref);
INIT_LIST_HEAD(&group->device_list);
mutex_init(&group->device_lock);
+ INIT_LIST_HEAD(&group->unbound_list);
+ mutex_init(&group->unbound_lock);
atomic_set(&group->container_users, 0);
atomic_set(&group->opened, 0);
group->iommu_group = iommu_group;
static void vfio_group_release(struct kref *kref)
{
struct vfio_group *group = container_of(kref, struct vfio_group, kref);
+ struct vfio_unbound_dev *unbound, *tmp;
+ struct iommu_group *iommu_group = group->iommu_group;
WARN_ON(!list_empty(&group->device_list));
+ list_for_each_entry_safe(unbound, tmp,
+ &group->unbound_list, unbound_next) {
+ list_del(&unbound->unbound_next);
+ kfree(unbound);
+ }
+
device_destroy(vfio.class, MKDEV(MAJOR(vfio.group_devt), group->minor));
list_del(&group->vfio_next);
vfio_free_group_minor(group->minor);
vfio_group_unlock_and_free(group);
+ iommu_group_put(iommu_group);
}
static void vfio_group_put(struct vfio_group *group)
}
/*
- * A vfio group is viable for use by userspace if all devices are either
- * driver-less or bound to a vfio or whitelisted driver. We test the
- * latter by the existence of a struct vfio_device matching the dev.
+ * A vfio group is viable for use by userspace if all devices are in
+ * one of the following states:
+ * - driver-less
+ * - bound to a vfio driver
+ * - bound to a whitelisted driver
+ *
+ * We use two methods to determine whether a device is bound to a vfio
+ * driver. The first is to test whether the device exists in the vfio
+ * group. The second is to test if the device exists on the group
+ * unbound_list, indicating it's in the middle of transitioning from
+ * a vfio driver to driver-less.
*/
static int vfio_dev_viable(struct device *dev, void *data)
{
struct vfio_group *group = data;
struct vfio_device *device;
struct device_driver *drv = ACCESS_ONCE(dev->driver);
+ struct vfio_unbound_dev *unbound;
+ int ret = -EINVAL;
- if (!drv || vfio_whitelisted_driver(drv))
+ mutex_lock(&group->unbound_lock);
+ list_for_each_entry(unbound, &group->unbound_list, unbound_next) {
+ if (dev == unbound->dev) {
+ ret = 0;
+ break;
+ }
+ }
+ mutex_unlock(&group->unbound_lock);
+
+ if (!ret || !drv || vfio_whitelisted_driver(drv))
return 0;
device = vfio_group_get_device(group, dev);
return 0;
}
- return -EINVAL;
+ return ret;
}
/**
{
struct vfio_group *group = container_of(nb, struct vfio_group, nb);
struct device *dev = data;
+ struct vfio_unbound_dev *unbound;
/*
* Need to go through a group_lock lookup to get a reference or we
* stop the system to maintain isolation. At a minimum, we'd
* want a toggle to disable driver auto probe for this device.
*/
+
+ mutex_lock(&group->unbound_lock);
+ list_for_each_entry(unbound,
+ &group->unbound_list, unbound_next) {
+ if (dev == unbound->dev) {
+ list_del(&unbound->unbound_next);
+ kfree(unbound);
+ break;
+ }
+ }
+ mutex_unlock(&group->unbound_lock);
break;
}
iommu_group_put(iommu_group);
return PTR_ERR(group);
}
+ } else {
+ /*
+ * A found vfio_group already holds a reference to the
+ * iommu_group. A created vfio_group keeps the reference.
+ */
+ iommu_group_put(iommu_group);
}
device = vfio_group_get_device(group, dev);
dev_name(dev), iommu_group_id(iommu_group));
vfio_device_put(device);
vfio_group_put(group);
- iommu_group_put(iommu_group);
return -EBUSY;
}
device = vfio_group_create_device(group, dev, ops, device_data);
if (IS_ERR(device)) {
vfio_group_put(group);
- iommu_group_put(iommu_group);
return PTR_ERR(device);
}
/*
- * Added device holds reference to iommu_group and vfio_device
- * (which in turn holds reference to vfio_group). Drop extra
- * group reference used while acquiring device.
+ * Drop all but the vfio_device reference. The vfio_device holds
+ * a reference to the vfio_group, which holds a reference to the
+ * iommu_group.
*/
vfio_group_put(group);
{
struct vfio_device *device = dev_get_drvdata(dev);
struct vfio_group *group = device->group;
- struct iommu_group *iommu_group = group->iommu_group;
void *device_data = device->device_data;
+ struct vfio_unbound_dev *unbound;
+ unsigned int i = 0;
/*
* The group exists so long as we have a device reference. Get
*/
vfio_group_get(group);
+ /*
+ * When the device is removed from the group, the group suddenly
+ * becomes non-viable; the device has a driver (until the unbind
+ * completes), but it's not present in the group. This is bad news
+ * for any external users that need to re-acquire a group reference
+ * in order to match and release their existing reference. To
+ * solve this, we track such devices on the unbound_list to bridge
+ * the gap until they're fully unbound.
+ */
+ unbound = kzalloc(sizeof(*unbound), GFP_KERNEL);
+ if (unbound) {
+ unbound->dev = dev;
+ mutex_lock(&group->unbound_lock);
+ list_add(&unbound->unbound_next, &group->unbound_list);
+ mutex_unlock(&group->unbound_lock);
+ }
+ WARN_ON(!unbound);
+
vfio_device_put(device);
- /* TODO send a signal to encourage this to be released */
- wait_event(vfio.release_q, !vfio_dev_present(group, dev));
+ /*
+ * If the device is still present in the group after the above
+ * 'put', then it is in use and we need to request it from the
+ * bus driver. The driver may in turn need to request the
+ * device from the user. We send the request on an arbitrary
+ * interval with counter to allow the driver to take escalating
+ * measures to release the device if it has the ability to do so.
+ */
+ do {
+ device = vfio_group_get_device(group, dev);
+ if (!device)
+ break;
- vfio_group_put(group);
+ if (device->ops->request)
+ device->ops->request(device_data, i++);
- iommu_group_put(iommu_group);
+ vfio_device_put(device);
+
+ } while (wait_event_interruptible_timeout(vfio.release_q,
+ !vfio_dev_present(group, dev),
+ HZ * 10) <= 0);
+
+ vfio_group_put(group);
return device_data;
}
git.openpandora.org / pandora-kernel.git / blobdiff