drm: implement experimental render nodes

Render nodes provide an API for userspace to use non-privileged GPU
commands without any running DRM-Master. It is useful for offscreen
rendering, GPGPU clients, and normal render clients which do not perform
modesetting.

Compared to legacy clients, render clients no longer need any
authentication to perform client ioctls. Instead, user-space controls
render/client access to GPUs via filesystem access-modes on the
render-node. Once a render-node was opened, a client has full access to
the client/render operations on the GPU. However, no modesetting or ioctls
that affect global state are allowed on render nodes.

To prevent privilege-escalation, drivers must explicitly state that they
support render nodes. They must mark their render-only ioctls as
DRM_RENDER_ALLOW so render clients can use them. Furthermore, they must
support clients without any attached master.

If filesystem access-modes are not enough for fine-grained access control
to render nodes (very unlikely, considering the versaitlity of FS-ACLs),
you may still fall-back to fd-passing from server to client (which allows
arbitrary access-control). However, note that revoking access is
currently impossible and unlikely to get implemented.

Note: Render clients no longer have any associated DRM-Master as they are
supposed to be independent of any server state. DRM core highly depends on
file_priv->master to be non-NULL for modesetting/ctx/etc. commands.
Therefore, drivers must be very careful to not require DRM-Master if they
support DRIVER_RENDER.

So far render-nodes are protected by "drm_rnodes". As long as this
module-parameter is not set to 1, a driver will not create render nodes.
This allows us to experiment with the API a bit before we stabilize it.

v2: drop insecure GEM_FLINK to force use of dmabuf
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>

Documentation/DocBook/drm.tmpl

@@ -205,6 +205,12 @@
               Driver implements DRM PRIME buffer sharing.
             </para></listitem>
           </varlistentry>
+          <varlistentry>
+            <term>DRIVER_RENDER</term>
+            <listitem><para>
+              Driver supports dedicated render nodes.
+            </para></listitem>
+          </varlistentry>
         </variablelist>
       </sect3>
       <sect3>
@@ -2644,6 +2650,69 @@ int (*resume) (struct drm_device *);</synopsis>
       info, since man pages should cover the rest.
     </para>
 
+  <!-- External: render nodes -->
+
+    <sect1>
+      <title>Render nodes</title>
+      <para>
+        DRM core provides multiple character-devices for user-space to use.
+        Depending on which device is opened, user-space can perform a different
+        set of operations (mainly ioctls). The primary node is always created
+        and called <term>card&lt;num&gt;</term>. Additionally, a currently
+        unused control node, called <term>controlD&lt;num&gt;</term> is also
+        created. The primary node provides all legacy operations and
+        historically was the only interface used by userspace. With KMS, the
+        control node was introduced. However, the planned KMS control interface
+        has never been written and so the control node stays unused to date.
+      </para>
+      <para>
+        With the increased use of offscreen renderers and GPGPU applications,
+        clients no longer require running compositors or graphics servers to
+        make use of a GPU. But the DRM API required unprivileged clients to
+        authenticate to a DRM-Master prior to getting GPU access. To avoid this
+        step and to grant clients GPU access without authenticating, render
+        nodes were introduced. Render nodes solely serve render clients, that
+        is, no modesetting or privileged ioctls can be issued on render nodes.
+        Only non-global rendering commands are allowed. If a driver supports
+        render nodes, it must advertise it via the <term>DRIVER_RENDER</term>
+        DRM driver capability. If not supported, the primary node must be used
+        for render clients together with the legacy drmAuth authentication
+        procedure.
+      </para>
+      <para>
+        If a driver advertises render node support, DRM core will create a
+        separate render node called <term>renderD&lt;num&gt;</term>. There will
+        be one render node per device. No ioctls except  PRIME-related ioctls
+        will be allowed on this node. Especially <term>GEM_OPEN</term> will be
+        explicitly prohibited. Render nodes are designed to avoid the
+        buffer-leaks, which occur if clients guess the flink names or mmap
+        offsets on the legacy interface. Additionally to this basic interface,
+        drivers must mark their driver-dependent render-only ioctls as
+        <term>DRM_RENDER_ALLOW</term> so render clients can use them. Driver
+        authors must be careful not to allow any privileged ioctls on render
+        nodes.
+      </para>
+      <para>
+        With render nodes, user-space can now control access to the render node
+        via basic file-system access-modes. A running graphics server which
+        authenticates clients on the privileged primary/legacy node is no longer
+        required. Instead, a client can open the render node and is immediately
+        granted GPU access. Communication between clients (or servers) is done
+        via PRIME. FLINK from render node to legacy node is not supported. New
+        clients must not use the insecure FLINK interface.
+      </para>
+      <para>
+        Besides dropping all modeset/global ioctls, render nodes also drop the
+        DRM-Master concept. There is no reason to associate render clients with
+        a DRM-Master as they are independent of any graphics server. Besides,
+        they must work without any running master, anyway.
+        Drivers must be able to run without a master object if they support
+        render nodes. If, on the other hand, a driver requires shared state
+        between clients which is visible to user-space and accessible beyond
+        open-file boundaries, they cannot support render nodes.
+      </para>
+    </sect1>
+
   <!-- External: vblank handling -->
 
     <sect1>

drivers/gpu/drm/drm_drv.c

@@ -68,7 +68,7 @@ static const struct drm_ioctl_desc drm_ioctls[] = {
 	DRM_IOCTL_DEF(DRM_IOCTL_GET_MAP, drm_getmap, DRM_UNLOCKED),
 	DRM_IOCTL_DEF(DRM_IOCTL_GET_CLIENT, drm_getclient, DRM_UNLOCKED),
 	DRM_IOCTL_DEF(DRM_IOCTL_GET_STATS, drm_getstats, DRM_UNLOCKED),
-	DRM_IOCTL_DEF(DRM_IOCTL_GET_CAP, drm_getcap, DRM_UNLOCKED),
+	DRM_IOCTL_DEF(DRM_IOCTL_GET_CAP, drm_getcap, DRM_UNLOCKED|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF(DRM_IOCTL_SET_VERSION, drm_setversion, DRM_MASTER),
 
 	DRM_IOCTL_DEF(DRM_IOCTL_SET_UNIQUE, drm_setunique, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
@@ -130,14 +130,14 @@ static const struct drm_ioctl_desc drm_ioctls[] = {
 
 	DRM_IOCTL_DEF(DRM_IOCTL_UPDATE_DRAW, drm_noop, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
 
-	DRM_IOCTL_DEF(DRM_IOCTL_GEM_CLOSE, drm_gem_close_ioctl, DRM_UNLOCKED),
+	DRM_IOCTL_DEF(DRM_IOCTL_GEM_CLOSE, drm_gem_close_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF(DRM_IOCTL_GEM_FLINK, drm_gem_flink_ioctl, DRM_AUTH|DRM_UNLOCKED),
 	DRM_IOCTL_DEF(DRM_IOCTL_GEM_OPEN, drm_gem_open_ioctl, DRM_AUTH|DRM_UNLOCKED),
 
 	DRM_IOCTL_DEF(DRM_IOCTL_MODE_GETRESOURCES, drm_mode_getresources, DRM_CONTROL_ALLOW|DRM_UNLOCKED),
 
-	DRM_IOCTL_DEF(DRM_IOCTL_PRIME_HANDLE_TO_FD, drm_prime_handle_to_fd_ioctl, DRM_AUTH|DRM_UNLOCKED),
-	DRM_IOCTL_DEF(DRM_IOCTL_PRIME_FD_TO_HANDLE, drm_prime_fd_to_handle_ioctl, DRM_AUTH|DRM_UNLOCKED),
+	DRM_IOCTL_DEF(DRM_IOCTL_PRIME_HANDLE_TO_FD, drm_prime_handle_to_fd_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF(DRM_IOCTL_PRIME_FD_TO_HANDLE, drm_prime_fd_to_handle_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
 
 	DRM_IOCTL_DEF(DRM_IOCTL_MODE_GETPLANERESOURCES, drm_mode_getplane_res, DRM_CONTROL_ALLOW|DRM_UNLOCKED),
 	DRM_IOCTL_DEF(DRM_IOCTL_MODE_GETCRTC, drm_mode_getcrtc, DRM_CONTROL_ALLOW|DRM_UNLOCKED),
@@ -420,9 +420,10 @@ long drm_ioctl(struct file *filp,
 		DRM_DEBUG("no function\n");
 		retcode = -EINVAL;
 	} else if (((ioctl->flags & DRM_ROOT_ONLY) && !capable(CAP_SYS_ADMIN)) ||
-		   ((ioctl->flags & DRM_AUTH) && !file_priv->authenticated) ||
+		   ((ioctl->flags & DRM_AUTH) && !drm_is_render_client(file_priv) && !file_priv->authenticated) ||
 		   ((ioctl->flags & DRM_MASTER) && !file_priv->is_master) ||
-		   (!(ioctl->flags & DRM_CONTROL_ALLOW) && (file_priv->minor->type == DRM_MINOR_CONTROL))) {
+		   (!(ioctl->flags & DRM_CONTROL_ALLOW) && (file_priv->minor->type == DRM_MINOR_CONTROL)) ||
+		   (!(ioctl->flags & DRM_RENDER_ALLOW) && drm_is_render_client(file_priv))) {
 		retcode = -EACCES;
 	} else {
 		if (cmd & (IOC_IN | IOC_OUT)) {

drivers/gpu/drm/drm_fops.c

@@ -262,10 +262,10 @@ static int drm_open_helper(struct inode *inode, struct file *filp,
 			goto out_prime_destroy;
 	}
 
-
-	/* if there is no current master make this fd it */
+	/* if there is no current master make this fd it, but do not create
+	 * any master object for render clients */
 	mutex_lock(&dev->struct_mutex);
-	if (!priv->minor->master) {
+	if (!priv->minor->master && !drm_is_render_client(priv)) {
 		/* create a new master */
 		priv->minor->master = drm_master_create(priv->minor);
 		if (!priv->minor->master) {
@@ -303,12 +303,11 @@ static int drm_open_helper(struct inode *inode, struct file *filp,
 				goto out_close;
 			}
 		}
-		mutex_unlock(&dev->struct_mutex);
-	} else {
+	} else if (!drm_is_render_client(priv)) {
 		/* get a reference to the master */
 		priv->master = drm_master_get(priv->minor->master);
-		mutex_unlock(&dev->struct_mutex);
 	}
+	mutex_unlock(&dev->struct_mutex);
 
 	mutex_lock(&dev->struct_mutex);
 	list_add(&priv->lhead, &dev->filelist);
@@ -478,7 +477,8 @@ int drm_release(struct inode *inode, struct file *filp)
 	iput(container_of(dev->dev_mapping, struct inode, i_data));
 
 	/* drop the reference held my the file priv */
-	drm_master_put(&file_priv->master);
+	if (file_priv->master)
+		drm_master_put(&file_priv->master);
 	file_priv->is_master = 0;
 	list_del(&file_priv->lhead);
 	mutex_unlock(&dev->struct_mutex);

drivers/gpu/drm/drm_pci.c

@@ -354,6 +354,12 @@ int drm_get_pci_dev(struct pci_dev *pdev, const struct pci_device_id *ent,
 			goto err_g2;
 	}
 
+	if (drm_core_check_feature(dev, DRIVER_RENDER) && drm_rnodes) {
+		ret = drm_get_minor(dev, &dev->render, DRM_MINOR_RENDER);
+		if (ret)
+			goto err_g21;
+	}
+
 	if ((ret = drm_get_minor(dev, &dev->primary, DRM_MINOR_LEGACY)))
 		goto err_g3;
 
@@ -383,6 +389,9 @@ int drm_get_pci_dev(struct pci_dev *pdev, const struct pci_device_id *ent,
 err_g4:
 	drm_put_minor(&dev->primary);
 err_g3:
+	if (dev->render)
+		drm_put_minor(&dev->render);
+err_g21:
 	if (drm_core_check_feature(dev, DRIVER_MODESET))
 		drm_put_minor(&dev->control);
 err_g2:

drivers/gpu/drm/drm_platform.c

@@ -69,6 +69,12 @@ static int drm_get_platform_dev(struct platform_device *platdev,
 			goto err_g1;
 	}
 
+	if (drm_core_check_feature(dev, DRIVER_RENDER) && drm_rnodes) {
+		ret = drm_get_minor(dev, &dev->render, DRM_MINOR_RENDER);
+		if (ret)
+			goto err_g11;
+	}
+
 	ret = drm_get_minor(dev, &dev->primary, DRM_MINOR_LEGACY);
 	if (ret)
 		goto err_g2;
@@ -100,6 +106,9 @@ static int drm_get_platform_dev(struct platform_device *platdev,
 err_g3:
 	drm_put_minor(&dev->primary);
 err_g2:
+	if (dev->render)
+		drm_put_minor(&dev->render);
+err_g11:
 	if (drm_core_check_feature(dev, DRIVER_MODESET))
 		drm_put_minor(&dev->control);
 err_g1:

drivers/gpu/drm/drm_stub.c

@@ -40,6 +40,9 @@
 unsigned int drm_debug = 0;	/* 1 to enable debug output */
 EXPORT_SYMBOL(drm_debug);
 
+unsigned int drm_rnodes = 0;	/* 1 to enable experimental render nodes API */
+EXPORT_SYMBOL(drm_rnodes);
+
 unsigned int drm_vblank_offdelay = 5000;    /* Default to 5000 msecs. */
 EXPORT_SYMBOL(drm_vblank_offdelay);
 
@@ -56,11 +59,13 @@ MODULE_AUTHOR(CORE_AUTHOR);
 MODULE_DESCRIPTION(CORE_DESC);
 MODULE_LICENSE("GPL and additional rights");
 MODULE_PARM_DESC(debug, "Enable debug output");
+MODULE_PARM_DESC(rnodes, "Enable experimental render nodes API");
 MODULE_PARM_DESC(vblankoffdelay, "Delay until vblank irq auto-disable [msecs]");
 MODULE_PARM_DESC(timestamp_precision_usec, "Max. error on timestamps [usecs]");
 MODULE_PARM_DESC(timestamp_monotonic, "Use monotonic timestamps");
 
 module_param_named(debug, drm_debug, int, 0600);
+module_param_named(rnodes, drm_rnodes, int, 0600);
 module_param_named(vblankoffdelay, drm_vblank_offdelay, int, 0600);
 module_param_named(timestamp_precision_usec, drm_timestamp_precision, int, 0600);
 module_param_named(timestamp_monotonic, drm_timestamp_monotonic, int, 0600);
@@ -446,6 +451,9 @@ void drm_put_dev(struct drm_device *dev)
 	if (drm_core_check_feature(dev, DRIVER_MODESET))
 		drm_put_minor(&dev->control);
 
+	if (dev->render)
+		drm_put_minor(&dev->render);
+
 	if (driver->driver_features & DRIVER_GEM)
 		drm_gem_destroy(dev);
 
@@ -462,6 +470,8 @@ void drm_unplug_dev(struct drm_device *dev)
 	/* for a USB device */
 	if (drm_core_check_feature(dev, DRIVER_MODESET))
 		drm_unplug_minor(dev->control);
+	if (dev->render)
+		drm_unplug_minor(dev->render);
 	drm_unplug_minor(dev->primary);
 
 	mutex_lock(&drm_global_mutex);

drivers/gpu/drm/drm_usb.c

@@ -33,6 +33,12 @@ int drm_get_usb_dev(struct usb_interface *interface,
 	if (ret)
 		goto err_g1;
 
+	if (drm_core_check_feature(dev, DRIVER_RENDER) && drm_rnodes) {
+		ret = drm_get_minor(dev, &dev->render, DRM_MINOR_RENDER);
+		if (ret)
+			goto err_g11;
+	}
+
 	ret = drm_get_minor(dev, &dev->primary, DRM_MINOR_LEGACY);
 	if (ret)
 		goto err_g2;
@@ -62,6 +68,9 @@ int drm_get_usb_dev(struct usb_interface *interface,
 err_g3:
 	drm_put_minor(&dev->primary);
 err_g2:
+	if (dev->render)
+		drm_put_minor(&dev->render);
+err_g11:
 	drm_put_minor(&dev->control);
 err_g1:
 	kfree(dev);

include/drm/drmP.h

@@ -145,6 +145,7 @@ int drm_err(const char *func, const char *format, ...);
 #define DRIVER_GEM         0x1000
 #define DRIVER_MODESET     0x2000
 #define DRIVER_PRIME       0x4000
+#define DRIVER_RENDER      0x8000
 
 #define DRIVER_BUS_PCI 0x1
 #define DRIVER_BUS_PLATFORM 0x2
@@ -290,6 +291,7 @@ typedef int drm_ioctl_compat_t(struct file *filp, unsigned int cmd,
 #define DRM_ROOT_ONLY	0x4
 #define DRM_CONTROL_ALLOW 0x8
 #define DRM_UNLOCKED	0x10
+#define DRM_RENDER_ALLOW 0x20
 
 struct drm_ioctl_desc {
 	unsigned int cmd;
@@ -1204,6 +1206,7 @@ struct drm_device {
 	unsigned int agp_buffer_token;
 	struct drm_minor *control;		/**< Control node for card */
 	struct drm_minor *primary;		/**< render type primary screen head */
+	struct drm_minor *render;		/**< render node for card */
 
         struct drm_mode_config mode_config;	/**< Current mode config */
 
@@ -1251,6 +1254,11 @@ static inline bool drm_modeset_is_locked(struct drm_device *dev)
 	return mutex_is_locked(&dev->mode_config.mutex);
 }
 
+static inline bool drm_is_render_client(struct drm_file *file_priv)
+{
+	return file_priv->minor->type == DRM_MINOR_RENDER;
+}
+
 /******************************************************************/
 /** \name Internal function definitions */
 /*@{*/
@@ -1450,6 +1458,7 @@ extern void drm_put_dev(struct drm_device *dev);
 extern int drm_put_minor(struct drm_minor **minor);
 extern void drm_unplug_dev(struct drm_device *dev);
 extern unsigned int drm_debug;
+extern unsigned int drm_rnodes;
 
 extern unsigned int drm_vblank_offdelay;
 extern unsigned int drm_timestamp_precision;