[PATCH 1/7] vfio: Create vfio_fs_type with inode per device

[Alex Williamson]

By linking all the device fds we provide to userspace to an
address space through a new pseudo fs, we can use tools like
unmap_mapping_range() to zap all vmas associated with a device.

Suggested-by: Jason Gunthorpe <jgg@xxxxxxxxxx>
Signed-off-by: Alex Williamson <alex.williamson@xxxxxxxxxx>
---
 drivers/vfio/vfio.c  |   57 ++++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/vfio.h |    1 +
 2 files changed, 58 insertions(+)

diff --git a/drivers/vfio/vfio.c b/drivers/vfio/vfio.c
index 02cc51ce6891..b88de89bda31 100644
--- a/drivers/vfio/vfio.c
+++ b/drivers/vfio/vfio.c
@@ -21,8 +21,10 @@
 #include <linux/list.h>
 #include <linux/miscdevice.h>
 #include <linux/module.h>
+#include <linux/mount.h>
 #include <linux/mutex.h>
 #include <linux/pci.h>
+#include <linux/pseudo_fs.h>
 #include <linux/rwsem.h>
 #include <linux/sched.h>
 #include <linux/slab.h>
@@ -37,6 +39,14 @@
 #define DRIVER_AUTHOR	"Alex Williamson <alex.williamson@xxxxxxxxxx>"
 #define DRIVER_DESC	"VFIO - User Level meta-driver"
 
+/*
+ * Not exposed via UAPI
+ *
+ * XXX Adopt the following when available:
+ * https://lore.kernel.org/lkml/20210309155348.974875-1-hch@xxxxxx/
+ */
+#define VFIO_MAGIC 0x5646494f /* "VFIO" */
+
 static struct vfio {
 	struct class			*class;
 	struct list_head		iommu_drivers_list;
@@ -46,6 +56,8 @@ static struct vfio {
 	struct mutex			group_lock;
 	struct cdev			group_cdev;
 	dev_t				group_devt;
+	struct vfsmount			*vfio_fs_mnt;
+	int				vfio_fs_cnt;
 } vfio;
 
 struct vfio_iommu_driver {
@@ -519,6 +531,35 @@ static struct vfio_group *vfio_group_get_from_dev(struct device *dev)
 	return group;
 }
 
+static int vfio_fs_init_fs_context(struct fs_context *fc)
+{
+	return init_pseudo(fc, VFIO_MAGIC) ? 0 : -ENOMEM;
+}
+
+static struct file_system_type vfio_fs_type = {
+	.name = "vfio",
+	.owner = THIS_MODULE,
+	.init_fs_context = vfio_fs_init_fs_context,
+	.kill_sb = kill_anon_super,
+};
+
+static struct inode *vfio_fs_inode_new(void)
+{
+	struct inode *inode;
+	int ret;
+
+	ret = simple_pin_fs(&vfio_fs_type,
+			    &vfio.vfio_fs_mnt, &vfio.vfio_fs_cnt);
+	if (ret)
+		return ERR_PTR(ret);
+
+	inode = alloc_anon_inode(vfio.vfio_fs_mnt->mnt_sb);
+	if (IS_ERR(inode))
+		simple_release_fs(&vfio.vfio_fs_mnt, &vfio.vfio_fs_cnt);
+
+	return inode;
+}
+
 /**
  * Device objects - create, release, get, put, search
  */
@@ -783,6 +824,12 @@ int vfio_register_group_dev(struct vfio_device *device)
 		return -EBUSY;
 	}
 
+	device->inode = vfio_fs_inode_new();
+	if (IS_ERR(device->inode)) {
+		vfio_group_put(group);
+		return PTR_ERR(device->inode);
+	}
+
 	/* Our reference on group is moved to the device */
 	device->group = group;
 
@@ -907,6 +954,9 @@ void vfio_unregister_group_dev(struct vfio_device *device)
 	group->dev_counter--;
 	mutex_unlock(&group->device_lock);
 
+	iput(device->inode);
+	simple_release_fs(&vfio.vfio_fs_mnt, &vfio.vfio_fs_cnt);
+
 	/*
 	 * In order to support multiple devices per group, devices can be
 	 * plucked from the group while other devices in the group are still
@@ -1411,6 +1461,13 @@ static int vfio_group_get_device_fd(struct vfio_group *group, char *buf)
 	 */
 	filep->f_mode |= (FMODE_LSEEK | FMODE_PREAD | FMODE_PWRITE);
 
+	/*
+	 * Use the pseudo fs inode on the device to link all mmaps
+	 * to the same address space, allowing us to unmap all vmas
+	 * associated to this device using unmap_mapping_range().
+	 */
+	filep->f_mapping = device->inode->i_mapping;
+
 	atomic_inc(&group->container_users);
 
 	fd_install(ret, filep);
diff --git a/include/linux/vfio.h b/include/linux/vfio.h
index a2c5b30e1763..90bcc2e9c8eb 100644
--- a/include/linux/vfio.h
+++ b/include/linux/vfio.h
@@ -24,6 +24,7 @@ struct vfio_device {
 	refcount_t refcount;
 	struct completion comp;
 	struct list_head group_next;
+	struct inode *inode;
 };
 
 /**