1 files changed, 139 insertions, 168 deletions

diff --git a/drivers/vfio/vfio.c b/drivers/vfio/vfio.c
index 3c034fe14ccb..08b27b64f0f9 100644
--- a/drivers/vfio/vfio.c
+++ b/drivers/vfio/vfio.c
@@ -32,6 +32,7 @@
 #include <linux/vfio.h>
 #include <linux/wait.h>
 #include <linux/sched/signal.h>
+#include "vfio.h"
 
 #define DRIVER_VERSION	"0.3"
 #define DRIVER_AUTHOR	"Alex Williamson <alex.williamson@redhat.com>"
@@ -83,7 +84,7 @@ struct vfio_group {
 	struct mutex			unbound_lock;
 	atomic_t			opened;
 	wait_queue_head_t		container_q;
-	bool				noiommu;
+	enum vfio_group_type		type;
 	unsigned int			dev_counter;
 	struct kvm			*kvm;
 	struct blocking_notifier_head	notifier;
@@ -169,70 +170,6 @@ static void vfio_release_device_set(struct vfio_device *device)
 	xa_unlock(&vfio_device_set_xa);
 }
 
-/*
- * vfio_iommu_group_{get,put} are only intended for VFIO bus driver probe
- * and remove functions, any use cases other than acquiring the first
- * reference for the purpose of calling vfio_register_group_dev() or removing
- * that symmetric reference after vfio_unregister_group_dev() should use the raw
- * iommu_group_{get,put} functions.  In particular, vfio_iommu_group_put()
- * removes the device from the dummy group and cannot be nested.
- */
-struct iommu_group *vfio_iommu_group_get(struct device *dev)
-{
-	struct iommu_group *group;
-	int __maybe_unused ret;
-
-	group = iommu_group_get(dev);
-
-#ifdef CONFIG_VFIO_NOIOMMU
-	/*
-	 * With noiommu enabled, an IOMMU group will be created for a device
-	 * that doesn't already have one and doesn't have an iommu_ops on their
-	 * bus.  We set iommudata simply to be able to identify these groups
-	 * as special use and for reclamation later.
-	 */
-	if (group || !noiommu || iommu_present(dev->bus))
-		return group;
-
-	group = iommu_group_alloc();
-	if (IS_ERR(group))
-		return NULL;
-
-	iommu_group_set_name(group, "vfio-noiommu");
-	iommu_group_set_iommudata(group, &noiommu, NULL);
-	ret = iommu_group_add_device(group, dev);
-	if (ret) {
-		iommu_group_put(group);
-		return NULL;
-	}
-
-	/*
-	 * Where to taint?  At this point we've added an IOMMU group for a
-	 * device that is not backed by iommu_ops, therefore any iommu_
-	 * callback using iommu_ops can legitimately Oops.  So, while we may
-	 * be about to give a DMA capable device to a user without IOMMU
-	 * protection, which is clearly taint-worthy, let's go ahead and do
-	 * it here.
-	 */
-	add_taint(TAINT_USER, LOCKDEP_STILL_OK);
-	dev_warn(dev, "Adding kernel taint for vfio-noiommu group on device\n");
-#endif
-
-	return group;
-}
-EXPORT_SYMBOL_GPL(vfio_iommu_group_get);
-
-void vfio_iommu_group_put(struct iommu_group *group, struct device *dev)
-{
-#ifdef CONFIG_VFIO_NOIOMMU
-	if (iommu_group_get_iommudata(group) == &noiommu)
-		iommu_group_remove_device(dev);
-#endif
-
-	iommu_group_put(group);
-}
-EXPORT_SYMBOL_GPL(vfio_iommu_group_put);
-
 #ifdef CONFIG_VFIO_NOIOMMU
 static void *vfio_noiommu_open(unsigned long arg)
 {
@@ -258,9 +195,9 @@ static long vfio_noiommu_ioctl(void *iommu_data,
 }
 
 static int vfio_noiommu_attach_group(void *iommu_data,
-				     struct iommu_group *iommu_group)
+		struct iommu_group *iommu_group, enum vfio_group_type type)
 {
-	return iommu_group_get_iommudata(iommu_group) == &noiommu ? 0 : -EINVAL;
+	return 0;
 }
 
 static void vfio_noiommu_detach_group(void *iommu_data,
@@ -277,8 +214,23 @@ static const struct vfio_iommu_driver_ops vfio_noiommu_ops = {
 	.attach_group = vfio_noiommu_attach_group,
 	.detach_group = vfio_noiommu_detach_group,
 };
-#endif
 
+/*
+ * Only noiommu containers can use vfio-noiommu and noiommu containers can only
+ * use vfio-noiommu.
+ */
+static inline bool vfio_iommu_driver_allowed(struct vfio_container *container,
+		const struct vfio_iommu_driver *driver)
+{
+	return container->noiommu == (driver->ops == &vfio_noiommu_ops);
+}
+#else
+static inline bool vfio_iommu_driver_allowed(struct vfio_container *container,
+		const struct vfio_iommu_driver *driver)
+{
+	return true;
+}
+#endif /* CONFIG_VFIO_NOIOMMU */
 
 /**
  * IOMMU driver registration
@@ -384,7 +336,8 @@ static void vfio_group_unlock_and_free(struct vfio_group *group)
 /**
  * Group objects - create, release, get, put, search
  */
-static struct vfio_group *vfio_create_group(struct iommu_group *iommu_group)
+static struct vfio_group *vfio_create_group(struct iommu_group *iommu_group,
+		enum vfio_group_type type)
 {
 	struct vfio_group *group, *tmp;
 	struct device *dev;
@@ -403,9 +356,7 @@ static struct vfio_group *vfio_create_group(struct iommu_group *iommu_group)
 	atomic_set(&group->opened, 0);
 	init_waitqueue_head(&group->container_q);
 	group->iommu_group = iommu_group;
-#ifdef CONFIG_VFIO_NOIOMMU
-	group->noiommu = (iommu_group_get_iommudata(iommu_group) == &noiommu);
-#endif
+	group->type = type;
 	BLOCKING_INIT_NOTIFIER_HEAD(&group->notifier);
 
 	group->nb.notifier_call = vfio_iommu_group_notifier;
@@ -441,8 +392,8 @@ static struct vfio_group *vfio_create_group(struct iommu_group *iommu_group)
 	}
 
 	dev = device_create(vfio.class, NULL,
-			    MKDEV(MAJOR(vfio.group_devt), minor),
-			    group, "%s%d", group->noiommu ? "noiommu-" : "",
+			    MKDEV(MAJOR(vfio.group_devt), minor), group, "%s%d",
+			    group->type == VFIO_NO_IOMMU ? "noiommu-" : "",
 			    iommu_group_id(iommu_group));
 	if (IS_ERR(dev)) {
 		vfio_free_group_minor(minor);
@@ -828,11 +779,85 @@ void vfio_uninit_group_dev(struct vfio_device *device)
 }
 EXPORT_SYMBOL_GPL(vfio_uninit_group_dev);
 
-int vfio_register_group_dev(struct vfio_device *device)
+static struct vfio_group *vfio_noiommu_group_alloc(struct device *dev,
+		enum vfio_group_type type)
 {
-	struct vfio_device *existing_device;
 	struct iommu_group *iommu_group;
 	struct vfio_group *group;
+	int ret;
+
+	iommu_group = iommu_group_alloc();
+	if (IS_ERR(iommu_group))
+		return ERR_CAST(iommu_group);
+
+	iommu_group_set_name(iommu_group, "vfio-noiommu");
+	ret = iommu_group_add_device(iommu_group, dev);
+	if (ret)
+		goto out_put_group;
+
+	group = vfio_create_group(iommu_group, type);
+	if (IS_ERR(group)) {
+		ret = PTR_ERR(group);
+		goto out_remove_device;
+	}
+
+	return group;
+
+out_remove_device:
+	iommu_group_remove_device(dev);
+out_put_group:
+	iommu_group_put(iommu_group);
+	return ERR_PTR(ret);
+}
+
+static struct vfio_group *vfio_group_find_or_alloc(struct device *dev)
+{
+	struct iommu_group *iommu_group;
+	struct vfio_group *group;
+
+	iommu_group = iommu_group_get(dev);
+#ifdef CONFIG_VFIO_NOIOMMU
+	if (!iommu_group && noiommu && !iommu_present(dev->bus)) {
+		/*
+		 * With noiommu enabled, create an IOMMU group for devices that
+		 * don't already have one and don't have an iommu_ops on their
+		 * bus.  Taint the kernel because we're about to give a DMA
+		 * capable device to a user without IOMMU protection.
+		 */
+		group = vfio_noiommu_group_alloc(dev, VFIO_NO_IOMMU);
+		if (!IS_ERR(group)) {
+			add_taint(TAINT_USER, LOCKDEP_STILL_OK);
+			dev_warn(dev, "Adding kernel taint for vfio-noiommu group on device\n");
+		}
+		return group;
+	}
+#endif
+	if (!iommu_group)
+		return ERR_PTR(-EINVAL);
+
+	/* a found vfio_group already holds a reference to the iommu_group */
+	group = vfio_group_get_from_iommu(iommu_group);
+	if (group)
+		goto out_put;
+
+	/* a newly created vfio_group keeps the reference. */
+	group = vfio_create_group(iommu_group, VFIO_IOMMU);
+	if (IS_ERR(group))
+		goto out_put;
+	return group;
+
+out_put:
+	iommu_group_put(iommu_group);
+	return group;
+}
+
+static int __vfio_register_dev(struct vfio_device *device,
+		struct vfio_group *group)
+{
+	struct vfio_device *existing_device;
+
+	if (IS_ERR(group))
+		return PTR_ERR(group);
 
 	/*
 	 * If the driver doesn't specify a set then the device is added to a
@@ -841,30 +866,14 @@ int vfio_register_group_dev(struct vfio_device *device)
 	if (!device->dev_set)
 		vfio_assign_device_set(device, device);
 
-	iommu_group = iommu_group_get(device->dev);
-	if (!iommu_group)
-		return -EINVAL;
-
-	group = vfio_group_get_from_iommu(iommu_group);
-	if (!group) {
-		group = vfio_create_group(iommu_group);
-		if (IS_ERR(group)) {
-			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);
-	}
-
 	existing_device = vfio_group_get_device(group, device->dev);
 	if (existing_device) {
 		dev_WARN(device->dev, "Device already exists on group %d\n",
-			 iommu_group_id(iommu_group));
+			 iommu_group_id(group->iommu_group));
 		vfio_device_put(existing_device);
+		if (group->type == VFIO_NO_IOMMU ||
+		    group->type == VFIO_EMULATED_IOMMU)
+			iommu_group_remove_device(device->dev);
 		vfio_group_put(group);
 		return -EBUSY;
 	}
@@ -882,8 +891,25 @@ int vfio_register_group_dev(struct vfio_device *device)
 
 	return 0;
 }
+
+int vfio_register_group_dev(struct vfio_device *device)
+{
+	return __vfio_register_dev(device,
+		vfio_group_find_or_alloc(device->dev));
+}
 EXPORT_SYMBOL_GPL(vfio_register_group_dev);
 
+/*
+ * Register a virtual device without IOMMU backing.  The user of this
+ * device must not be able to directly trigger unmediated DMA.
+ */
+int vfio_register_emulated_iommu_dev(struct vfio_device *device)
+{
+	return __vfio_register_dev(device,
+		vfio_noiommu_group_alloc(device->dev, VFIO_EMULATED_IOMMU));
+}
+EXPORT_SYMBOL_GPL(vfio_register_emulated_iommu_dev);
+
 /**
  * Get a reference to the vfio_device for a device.  Even if the
  * caller thinks they own the device, they could be racing with a
@@ -1010,6 +1036,9 @@ void vfio_unregister_group_dev(struct vfio_device *device)
 	if (list_empty(&group->device_list))
 		wait_event(group->container_q, !group->container);
 
+	if (group->type == VFIO_NO_IOMMU || group->type == VFIO_EMULATED_IOMMU)
+		iommu_group_remove_device(device->dev);
+
 	/* Matches the get in vfio_register_group_dev() */
 	vfio_group_put(group);
 }
@@ -1042,13 +1071,10 @@ static long vfio_ioctl_check_extension(struct vfio_container *container,
 			list_for_each_entry(driver, &vfio.iommu_drivers_list,
 					    vfio_next) {
 
-#ifdef CONFIG_VFIO_NOIOMMU
 				if (!list_empty(&container->group_list) &&
-				    (container->noiommu !=
-				     (driver->ops == &vfio_noiommu_ops)))
+				    !vfio_iommu_driver_allowed(container,
+							       driver))
 					continue;
-#endif
-
 				if (!try_module_get(driver->ops->owner))
 					continue;
 
@@ -1079,7 +1105,8 @@ static int __vfio_container_attach_groups(struct vfio_container *container,
 	int ret = -ENODEV;
 
 	list_for_each_entry(group, &container->group_list, container_next) {
-		ret = driver->ops->attach_group(data, group->iommu_group);
+		ret = driver->ops->attach_group(data, group->iommu_group,
+						group->type);
 		if (ret)
 			goto unwind;
 	}
@@ -1120,15 +1147,8 @@ static long vfio_ioctl_set_iommu(struct vfio_container *container,
 	list_for_each_entry(driver, &vfio.iommu_drivers_list, vfio_next) {
 		void *data;
 
-#ifdef CONFIG_VFIO_NOIOMMU
-		/*
-		 * Only noiommu containers can use vfio-noiommu and noiommu
-		 * containers can only use vfio-noiommu.
-		 */
-		if (container->noiommu != (driver->ops == &vfio_noiommu_ops))
+		if (!vfio_iommu_driver_allowed(container, driver))
 			continue;
-#endif
-
 		if (!try_module_get(driver->ops->owner))
 			continue;
 
@@ -1234,62 +1254,12 @@ static int vfio_fops_release(struct inode *inode, struct file *filep)
 	return 0;
 }
 
-/*
- * Once an iommu driver is set, we optionally pass read/write/mmap
- * on to the driver, allowing management interfaces beyond ioctl.
- */
-static ssize_t vfio_fops_read(struct file *filep, char __user *buf,
-			      size_t count, loff_t *ppos)
-{
-	struct vfio_container *container = filep->private_data;
-	struct vfio_iommu_driver *driver;
-	ssize_t ret = -EINVAL;
-
-	driver = container->iommu_driver;
-	if (likely(driver && driver->ops->read))
-		ret = driver->ops->read(container->iommu_data,
-					buf, count, ppos);
-
-	return ret;
-}
-
-static ssize_t vfio_fops_write(struct file *filep, const char __user *buf,
-			       size_t count, loff_t *ppos)
-{
-	struct vfio_container *container = filep->private_data;
-	struct vfio_iommu_driver *driver;
-	ssize_t ret = -EINVAL;
-
-	driver = container->iommu_driver;
-	if (likely(driver && driver->ops->write))
-		ret = driver->ops->write(container->iommu_data,
-					 buf, count, ppos);
-
-	return ret;
-}
-
-static int vfio_fops_mmap(struct file *filep, struct vm_area_struct *vma)
-{
-	struct vfio_container *container = filep->private_data;
-	struct vfio_iommu_driver *driver;
-	int ret = -EINVAL;
-
-	driver = container->iommu_driver;
-	if (likely(driver && driver->ops->mmap))
-		ret = driver->ops->mmap(container->iommu_data, vma);
-
-	return ret;
-}
-
 static const struct file_operations vfio_fops = {
 	.owner		= THIS_MODULE,
 	.open		= vfio_fops_open,
 	.release	= vfio_fops_release,
-	.read		= vfio_fops_read,
-	.write		= vfio_fops_write,
 	.unlocked_ioctl	= vfio_fops_unl_ioctl,
 	.compat_ioctl	= compat_ptr_ioctl,
-	.mmap		= vfio_fops_mmap,
 };
 
 /**
@@ -1366,7 +1336,7 @@ static int vfio_group_set_container(struct vfio_group *group, int container_fd)
 	if (atomic_read(&group->container_users))
 		return -EINVAL;
 
-	if (group->noiommu && !capable(CAP_SYS_RAWIO))
+	if (group->type == VFIO_NO_IOMMU && !capable(CAP_SYS_RAWIO))
 		return -EPERM;
 
 	f = fdget(container_fd);
@@ -1386,7 +1356,7 @@ static int vfio_group_set_container(struct vfio_group *group, int container_fd)
 
 	/* Real groups and fake groups cannot mix */
 	if (!list_empty(&container->group_list) &&
-	    container->noiommu != group->noiommu) {
+	    container->noiommu != (group->type == VFIO_NO_IOMMU)) {
 		ret = -EPERM;
 		goto unlock_out;
 	}
@@ -1394,13 +1364,14 @@ static int vfio_group_set_container(struct vfio_group *group, int container_fd)
 	driver = container->iommu_driver;
 	if (driver) {
 		ret = driver->ops->attach_group(container->iommu_data,
-						group->iommu_group);
+						group->iommu_group,
+						group->type);
 		if (ret)
 			goto unlock_out;
 	}
 
 	group->container = container;
-	container->noiommu = group->noiommu;
+	container->noiommu = (group->type == VFIO_NO_IOMMU);
 	list_add(&group->container_next, &container->group_list);
 
 	/* Get a reference on the container and mark a user within the group */
@@ -1424,7 +1395,7 @@ static int vfio_group_add_container_user(struct vfio_group *group)
 	if (!atomic_inc_not_zero(&group->container_users))
 		return -EINVAL;
 
-	if (group->noiommu) {
+	if (group->type == VFIO_NO_IOMMU) {
 		atomic_dec(&group->container_users);
 		return -EPERM;
 	}
@@ -1449,7 +1420,7 @@ static int vfio_group_get_device_fd(struct vfio_group *group, char *buf)
 	    !group->container->iommu_driver || !vfio_group_viable(group))
 		return -EINVAL;
 
-	if (group->noiommu && !capable(CAP_SYS_RAWIO))
+	if (group->type == VFIO_NO_IOMMU && !capable(CAP_SYS_RAWIO))
 		return -EPERM;
 
 	device = vfio_device_get_from_name(group, buf);
@@ -1496,7 +1467,7 @@ static int vfio_group_get_device_fd(struct vfio_group *group, char *buf)
 
 	fd_install(fdno, filep);
 
-	if (group->noiommu)
+	if (group->type == VFIO_NO_IOMMU)
 		dev_warn(device->dev, "vfio-noiommu device opened by user "
 			 "(%s:%d)\n", current->comm, task_pid_nr(current));
 	return fdno;
@@ -1592,7 +1563,7 @@ static int vfio_group_fops_open(struct inode *inode, struct file *filep)
 	if (!group)
 		return -ENODEV;
 
-	if (group->noiommu && !capable(CAP_SYS_RAWIO)) {
+	if (group->type == VFIO_NO_IOMMU && !capable(CAP_SYS_RAWIO)) {
 		vfio_group_put(group);
 		return -EPERM;
 	}