Merge branch 'cuse' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse

* 'cuse' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse:
  CUSE: implement CUSE - Character device in Userspace
  fuse: export symbols to be used by CUSE
  fuse: update fuse_conn_init() and separate out fuse_conn_kill()
  fuse: don't use inode in fuse_file_poll
  fuse: don't use inode in fuse_do_ioctl() helper
  fuse: don't use inode in fuse_sync_release()
  fuse: create fuse_do_open() helper for CUSE
  fuse: clean up args in fuse_finish_open() and fuse_release_fill()
  fuse: don't use inode in helpers called by fuse_direct_io()
  fuse: add members to struct fuse_file
  fuse: prepare fuse_direct_io() for CUSE
  fuse: clean up fuse_write_fill()
  fuse: use struct path in release structure
  fuse: misc cleanups

fs/Kconfig

@@ -62,6 +62,16 @@ source "fs/autofs/Kconfig"
 source "fs/autofs4/Kconfig"
 source "fs/fuse/Kconfig"
 
+config CUSE
+	tristate "Character device in Userpace support"
+	depends on FUSE_FS
+	help
+	  This FUSE extension allows character devices to be
+	  implemented in userspace.
+
+	  If you want to develop or use userspace character device
+	  based on CUSE, answer Y or M.
+
 config GENERIC_ACL
 	bool
 	select FS_POSIX_ACL

fs/fuse/Makefile

@@ -3,5 +3,6 @@
 #
 
 obj-$(CONFIG_FUSE_FS) += fuse.o
+obj-$(CONFIG_CUSE) += cuse.o
 
 fuse-objs := dev.o dir.o file.o inode.o control.o

fs/fuse/cuse.c

+/*
+ * CUSE: Character device in Userspace
+ *
+ * Copyright (C) 2008-2009  SUSE Linux Products GmbH
+ * Copyright (C) 2008-2009  Tejun Heo <tj@kernel.org>
+ *
+ * This file is released under the GPLv2.
+ *
+ * CUSE enables character devices to be implemented from userland much
+ * like FUSE allows filesystems.  On initialization /dev/cuse is
+ * created.  By opening the file and replying to the CUSE_INIT request
+ * userland CUSE server can create a character device.  After that the
+ * operation is very similar to FUSE.
+ *
+ * A CUSE instance involves the following objects.
+ *
+ * cuse_conn	: contains fuse_conn and serves as bonding structure
+ * channel	: file handle connected to the userland CUSE server
+ * cdev		: the implemented character device
+ * dev		: generic device for cdev
+ *
+ * Note that 'channel' is what 'dev' is in FUSE.  As CUSE deals with
+ * devices, it's called 'channel' to reduce confusion.
+ *
+ * channel determines when the character device dies.  When channel is
+ * closed, everything begins to destruct.  The cuse_conn is taken off
+ * the lookup table preventing further access from cdev, cdev and
+ * generic device are removed and the base reference of cuse_conn is
+ * put.
+ *
+ * On each open, the matching cuse_conn is looked up and if found an
+ * additional reference is taken which is released when the file is
+ * closed.
+ */
+
+#include <linux/fuse.h>
+#include <linux/cdev.h>
+#include <linux/device.h>
+#include <linux/file.h>
+#include <linux/fs.h>
+#include <linux/kdev_t.h>
+#include <linux/kthread.h>
+#include <linux/list.h>
+#include <linux/magic.h>
+#include <linux/miscdevice.h>
+#include <linux/mutex.h>
+#include <linux/spinlock.h>
+#include <linux/stat.h>
+
+#include "fuse_i.h"
+
+#define CUSE_CONNTBL_LEN	64
+
+struct cuse_conn {
+	struct list_head	list;	/* linked on cuse_conntbl */
+	struct fuse_conn	fc;	/* fuse connection */
+	struct cdev		*cdev;	/* associated character device */
+	struct device		*dev;	/* device representing @cdev */
+
+	/* init parameters, set once during initialization */
+	bool			unrestricted_ioctl;
+};
+
+static DEFINE_SPINLOCK(cuse_lock);		/* protects cuse_conntbl */
+static struct list_head cuse_conntbl[CUSE_CONNTBL_LEN];
+static struct class *cuse_class;
+
+static struct cuse_conn *fc_to_cc(struct fuse_conn *fc)
+{
+	return container_of(fc, struct cuse_conn, fc);
+}
+
+static struct list_head *cuse_conntbl_head(dev_t devt)
+{
+	return &cuse_conntbl[(MAJOR(devt) + MINOR(devt)) % CUSE_CONNTBL_LEN];
+}
+
+
+/**************************************************************************
+ * CUSE frontend operations
+ *
+ * These are file operations for the character device.
+ *
+ * On open, CUSE opens a file from the FUSE mnt and stores it to
+ * private_data of the open file.  All other ops call FUSE ops on the
+ * FUSE file.
+ */
+
+static ssize_t cuse_read(struct file *file, char __user *buf, size_t count,
+			 loff_t *ppos)
+{
+	loff_t pos = 0;
+
+	return fuse_direct_io(file, buf, count, &pos, 0);
+}
+
+static ssize_t cuse_write(struct file *file, const char __user *buf,
+			  size_t count, loff_t *ppos)
+{
+	loff_t pos = 0;
+	/*
+	 * No locking or generic_write_checks(), the server is
+	 * responsible for locking and sanity checks.
+	 */
+	return fuse_direct_io(file, buf, count, &pos, 1);
+}
+
+static int cuse_open(struct inode *inode, struct file *file)
+{
+	dev_t devt = inode->i_cdev->dev;
+	struct cuse_conn *cc = NULL, *pos;
+	int rc;
+
+	/* look up and get the connection */
+	spin_lock(&cuse_lock);
+	list_for_each_entry(pos, cuse_conntbl_head(devt), list)
+		if (pos->dev->devt == devt) {
+			fuse_conn_get(&pos->fc);
+			cc = pos;
+			break;
+		}
+	spin_unlock(&cuse_lock);
+
+	/* dead? */
+	if (!cc)
+		return -ENODEV;
+
+	/*
+	 * Generic permission check is already done against the chrdev
+	 * file, proceed to open.
+	 */
+	rc = fuse_do_open(&cc->fc, 0, file, 0);
+	if (rc)
+		fuse_conn_put(&cc->fc);
+	return rc;
+}
+
+static int cuse_release(struct inode *inode, struct file *file)
+{
+	struct fuse_file *ff = file->private_data;
+	struct fuse_conn *fc = ff->fc;
+
+	fuse_sync_release(ff, file->f_flags);
+	fuse_conn_put(fc);
+
+	return 0;
+}
+
+static long cuse_file_ioctl(struct file *file, unsigned int cmd,
+			    unsigned long arg)
+{
+	struct fuse_file *ff = file->private_data;
+	struct cuse_conn *cc = fc_to_cc(ff->fc);
+	unsigned int flags = 0;
+
+	if (cc->unrestricted_ioctl)
+		flags |= FUSE_IOCTL_UNRESTRICTED;
+
+	return fuse_do_ioctl(file, cmd, arg, flags);
+}
+
+static long cuse_file_compat_ioctl(struct file *file, unsigned int cmd,
+				   unsigned long arg)
+{
+	struct fuse_file *ff = file->private_data;
+	struct cuse_conn *cc = fc_to_cc(ff->fc);
+	unsigned int flags = FUSE_IOCTL_COMPAT;
+
+	if (cc->unrestricted_ioctl)
+		flags |= FUSE_IOCTL_UNRESTRICTED;
+
+	return fuse_do_ioctl(file, cmd, arg, flags);
+}
+
+static const struct file_operations cuse_frontend_fops = {
+	.owner			= THIS_MODULE,
+	.read			= cuse_read,
+	.write			= cuse_write,
+	.open			= cuse_open,
+	.release		= cuse_release,
+	.unlocked_ioctl		= cuse_file_ioctl,
+	.compat_ioctl		= cuse_file_compat_ioctl,
+	.poll			= fuse_file_poll,
+};
+
+
+/**************************************************************************
+ * CUSE channel initialization and destruction
+ */
+
+struct cuse_devinfo {
+	const char		*name;
+};
+
+/**
+ * cuse_parse_one - parse one key=value pair
+ * @pp: i/o parameter for the current position
+ * @end: points to one past the end of the packed string
+ * @keyp: out parameter for key
+ * @valp: out parameter for value
+ *
+ * *@pp points to packed strings - "key0=val0\0key1=val1\0" which ends
+ * at @end - 1.  This function parses one pair and set *@keyp to the
+ * start of the key and *@valp to the start of the value.  Note that
+ * the original string is modified such that the key string is
+ * terminated with '\0'.  *@pp is updated to point to the next string.
+ *
+ * RETURNS:
+ * 1 on successful parse, 0 on EOF, -errno on failure.
+ */
+static int cuse_parse_one(char **pp, char *end, char **keyp, char **valp)
+{
+	char *p = *pp;
+	char *key, *val;
+
+	while (p < end && *p == '\0')
+		p++;
+	if (p == end)
+		return 0;
+
+	if (end[-1] != '\0') {
+		printk(KERN_ERR "CUSE: info not properly terminated\n");
+		return -EINVAL;
+	}
+
+	key = val = p;
+	p += strlen(p);
+
+	if (valp) {
+		strsep(&val, "=");
+		if (!val)
+			val = key + strlen(key);
+		key = strstrip(key);
+		val = strstrip(val);
+	} else
+		key = strstrip(key);
+
+	if (!strlen(key)) {
+		printk(KERN_ERR "CUSE: zero length info key specified\n");
+		return -EINVAL;
+	}
+
+	*pp = p;
+	*keyp = key;
+	if (valp)
+		*valp = val;
+
+	return 1;
+}
+
+/**
+ * cuse_parse_dev_info - parse device info
+ * @p: device info string
+ * @len: length of device info string
+ * @devinfo: out parameter for parsed device info
+ *
+ * Parse @p to extract device info and store it into @devinfo.  String
+ * pointed to by @p is modified by parsing and @devinfo points into
+ * them, so @p shouldn't be freed while @devinfo is in use.
+ *
+ * RETURNS:
+ * 0 on success, -errno on failure.
+ */
+static int cuse_parse_devinfo(char *p, size_t len, struct cuse_devinfo *devinfo)
+{
+	char *end = p + len;
+	char *key, *val;
+	int rc;
+
+	while (true) {
+		rc = cuse_parse_one(&p, end, &key, &val);
+		if (rc < 0)
+			return rc;
+		if (!rc)
+			break;
+		if (strcmp(key, "DEVNAME") == 0)
+			devinfo->name = val;
+		else
+			printk(KERN_WARNING "CUSE: unknown device info \"%s\"\n",
+			       key);
+	}
+
+	if (!devinfo->name || !strlen(devinfo->name)) {
+		printk(KERN_ERR "CUSE: DEVNAME unspecified\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static void cuse_gendev_release(struct device *dev)
+{
+	kfree(dev);
+}
+
+/**
+ * cuse_process_init_reply - finish initializing CUSE channel
+ *
+ * This function creates the character device and sets up all the
+ * required data structures for it.  Please read the comment at the
+ * top of this file for high level overview.
+ */
+static void cuse_process_init_reply(struct fuse_conn *fc, struct fuse_req *req)
+{
+	struct cuse_conn *cc = fc_to_cc(fc);
+	struct cuse_init_out *arg = &req->misc.cuse_init_out;
+	struct page *page = req->pages[0];
+	struct cuse_devinfo devinfo = { };
+	struct device *dev;
+	struct cdev *cdev;
+	dev_t devt;
+	int rc;
+
+	if (req->out.h.error ||
+	    arg->major != FUSE_KERNEL_VERSION || arg->minor < 11) {
+		goto err;
+	}
+
+	fc->minor = arg->minor;
+	fc->max_read = max_t(unsigned, arg->max_read, 4096);
+	fc->max_write = max_t(unsigned, arg->max_write, 4096);
+
+	/* parse init reply */
+	cc->unrestricted_ioctl = arg->flags & CUSE_UNRESTRICTED_IOCTL;
+
+	rc = cuse_parse_devinfo(page_address(page), req->out.args[1].size,
+				&devinfo);
+	if (rc)
+		goto err;
+
+	/* determine and reserve devt */
+	devt = MKDEV(arg->dev_major, arg->dev_minor);
+	if (!MAJOR(devt))
+		rc = alloc_chrdev_region(&devt, MINOR(devt), 1, devinfo.name);
+	else
+		rc = register_chrdev_region(devt, 1, devinfo.name);
+	if (rc) {
+		printk(KERN_ERR "CUSE: failed to register chrdev region\n");
+		goto err;
+	}
+
+	/* devt determined, create device */
+	rc = -ENOMEM;
+	dev = kzalloc(sizeof(*dev), GFP_KERNEL);
+	if (!dev)
+		goto err_region;
+
+	device_initialize(dev);
+	dev_set_uevent_suppress(dev, 1);
+	dev->class = cuse_class;
+	dev->devt = devt;
+	dev->release = cuse_gendev_release;
+	dev_set_drvdata(dev, cc);
+	dev_set_name(dev, "%s", devinfo.name);
+
+	rc = device_add(dev);
+	if (rc)
+		goto err_device;
+
+	/* register cdev */
+	rc = -ENOMEM;
+	cdev = cdev_alloc();
+	if (!cdev)
+		goto err_device;
+
+	cdev->owner = THIS_MODULE;
+	cdev->ops = &cuse_frontend_fops;
+
+	rc = cdev_add(cdev, devt, 1);
+	if (rc)
+		goto err_cdev;
+
+	cc->dev = dev;
+	cc->cdev = cdev;
+
+	/* make the device available */
+	spin_lock(&cuse_lock);
+	list_add(&cc->list, cuse_conntbl_head(devt));
+	spin_unlock(&cuse_lock);
+
+	/* announce device availability */
+	dev_set_uevent_suppress(dev, 0);
+	kobject_uevent(&dev->kobj, KOBJ_ADD);
+out:
+	__free_page(page);
+	return;
+
+err_cdev:
+	cdev_del(cdev);
+err_device:
+	put_device(dev);
+err_region:
+	unregister_chrdev_region(devt, 1);
+err:
+	fc->conn_error = 1;
+	goto out;
+}
+
+static int cuse_send_init(struct cuse_conn *cc)
+{
+	int rc;
+	struct fuse_req *req;
+	struct page *page;
+	struct fuse_conn *fc = &cc->fc;
+	struct cuse_init_in *arg;
+
+	BUILD_BUG_ON(CUSE_INIT_INFO_MAX > PAGE_SIZE);
+
+	req = fuse_get_req(fc);
+	if (IS_ERR(req)) {
+		rc = PTR_ERR(req);
+		goto err;
+	}
+
+	rc = -ENOMEM;
+	page = alloc_page(GFP_KERNEL | __GFP_ZERO);
+	if (!page)
+		goto err_put_req;
+
+	arg = &req->misc.cuse_init_in;
+	arg->major = FUSE_KERNEL_VERSION;
+	arg->minor = FUSE_KERNEL_MINOR_VERSION;
+	arg->flags |= CUSE_UNRESTRICTED_IOCTL;
+	req->in.h.opcode = CUSE_INIT;
+	req->in.numargs = 1;
+	req->in.args[0].size = sizeof(struct cuse_init_in);
+	req->in.args[0].value = arg;
+	req->out.numargs = 2;
+	req->out.args[0].size = sizeof(struct cuse_init_out);
+	req->out.args[0].value = &req->misc.cuse_init_out;
+	req->out.args[1].size = CUSE_INIT_INFO_MAX;
+	req->out.argvar = 1;
+	req->out.argpages = 1;
+	req->pages[0] = page;
+	req->num_pages = 1;
+	req->end = cuse_process_init_reply;
+	fuse_request_send_background(fc, req);
+
+	return 0;
+
+err_put_req:
+	fuse_put_request(fc, req);
+err:
+	return rc;
+}
+
+static void cuse_fc_release(struct fuse_conn *fc)
+{
+	struct cuse_conn *cc = fc_to_cc(fc);
+	kfree(cc);
+}
+
+/**
+ * cuse_channel_open - open method for /dev/cuse
+ * @inode: inode for /dev/cuse
+ * @file: file struct being opened
+ *
+ * Userland CUSE server can create a CUSE device by opening /dev/cuse
+ * and replying to the initilaization request kernel sends.  This
+ * function is responsible for handling CUSE device initialization.
+ * Because the fd opened by this function is used during
+ * initialization, this function only creates cuse_conn and sends
+ * init.  The rest is delegated to a kthread.
+ *
+ * RETURNS:
+ * 0 on success, -errno on failure.
+ */
+static int cuse_channel_open(struct inode *inode, struct file *file)
+{
+	struct cuse_conn *cc;
+	int rc;
+
+	/* set up cuse_conn */
+	cc = kzalloc(sizeof(*cc), GFP_KERNEL);
+	if (!cc)
+		return -ENOMEM;
+
+	fuse_conn_init(&cc->fc);
+
+	INIT_LIST_HEAD(&cc->list);
+	cc->fc.release = cuse_fc_release;
+
+	cc->fc.connected = 1;
+	cc->fc.blocked = 0;
+	rc = cuse_send_init(cc);
+	if (rc) {
+		fuse_conn_put(&cc->fc);
+		return rc;
+	}
+	file->private_data = &cc->fc;	/* channel owns base reference to cc */
+
+	return 0;
+}
+
+/**
+ * cuse_channel_release - release method for /dev/cuse
+ * @inode: inode for /dev/cuse
+ * @file: file struct being closed
+ *
+ * Disconnect the channel, deregister CUSE device and initiate
+ * destruction by putting the default reference.
+ *
+ * RETURNS:
+ * 0 on success, -errno on failure.
+ */
+static int cuse_channel_release(struct inode *inode, struct file *file)
+{
+	struct cuse_conn *cc = fc_to_cc(file->private_data);
+	int rc;
+
+	/* remove from the conntbl, no more access from this point on */
+	spin_lock(&cuse_lock);
+	list_del_init(&cc->list);
+	spin_unlock(&cuse_lock);
+
+	/* remove device */
+	if (cc->dev)
+		device_unregister(cc->dev);
+	if (cc->cdev) {
+		unregister_chrdev_region(cc->cdev->dev, 1);
+		cdev_del(cc->cdev);
+	}
+
+	/* kill connection and shutdown channel */
+	fuse_conn_kill(&cc->fc);
+	rc = fuse_dev_release(inode, file);	/* puts the base reference */
+
+	return rc;
+}
+
+static struct file_operations cuse_channel_fops; /* initialized during init */
+
+
+/**************************************************************************
+ * Misc stuff and module initializatiion
+ *
+ * CUSE exports the same set of attributes to sysfs as fusectl.
+ */
+
+static ssize_t cuse_class_waiting_show(struct device *dev,
+				       struct device_attribute *attr, char *buf)
+{
+	struct cuse_conn *cc = dev_get_drvdata(dev);
+
+	return sprintf(buf, "%d\n", atomic_read(&cc->fc.num_waiting));
+}
+
+static ssize_t cuse_class_abort_store(struct device *dev,
+				      struct device_attribute *attr,
+				      const char *buf, size_t count)
+{
+	struct cuse_conn *cc = dev_get_drvdata(dev);
+
+	fuse_abort_conn(&cc->fc);
+	return count;
+}
+
+static struct device_attribute cuse_class_dev_attrs[] = {
+	__ATTR(waiting, S_IFREG | 0400, cuse_class_waiting_show, NULL),
+	__ATTR(abort, S_IFREG | 0200, NULL, cuse_class_abort_store),
+	{ }
+};
+
+static struct miscdevice cuse_miscdev = {
+	.minor		= MISC_DYNAMIC_MINOR,
+	.name		= "cuse",
+	.fops		= &cuse_channel_fops,
+};
+
+static int __init cuse_init(void)
+{
+	int i, rc;
+
+	/* init conntbl */
+	for (i = 0; i < CUSE_CONNTBL_LEN; i++)
+		INIT_LIST_HEAD(&cuse_conntbl[i]);
+
+	/* inherit and extend fuse_dev_operations */
+	cuse_channel_fops		= fuse_dev_operations;
+	cuse_channel_fops.owner		= THIS_MODULE;
+	cuse_channel_fops.open		= cuse_channel_open;
+	cuse_channel_fops.release	= cuse_channel_release;
+
+	cuse_class = class_create(THIS_MODULE, "cuse");
+	if (IS_ERR(cuse_class))
+		return PTR_ERR(cuse_class);
+
+	cuse_class->dev_attrs = cuse_class_dev_attrs;
+
+	rc = misc_register(&cuse_miscdev);
+	if (rc) {
+		class_destroy(cuse_class);
+		return rc;
+	}
+
+	return 0;
+}
+
+static void __exit cuse_exit(void)
+{
+	misc_deregister(&cuse_miscdev);
+	class_destroy(cuse_class);
+}
+
+module_init(cuse_init);
+module_exit(cuse_exit);
+
+MODULE_AUTHOR("Tejun Heo <tj@kernel.org>");
+MODULE_DESCRIPTION("Character device in Userspace");
+MODULE_LICENSE("GPL");

fs/fuse/dev.c

@@ -46,6 +46,7 @@ struct fuse_req *fuse_request_alloc(void)
 		fuse_request_init(req);
 	return req;
 }
+EXPORT_SYMBOL_GPL(fuse_request_alloc);
 
 struct fuse_req *fuse_request_alloc_nofs(void)
 {
@@ -124,6 +125,7 @@ struct fuse_req *fuse_get_req(struct fuse_conn *fc)
 	atomic_dec(&fc->num_waiting);
 	return ERR_PTR(err);
 }
+EXPORT_SYMBOL_GPL(fuse_get_req);
 
 /*
  * Return request in fuse_file->reserved_req.  However that may
@@ -208,6 +210,7 @@ void fuse_put_request(struct fuse_conn *fc, struct fuse_req *req)
 			fuse_request_free(req);
 	}
 }
+EXPORT_SYMBOL_GPL(fuse_put_request);
 
 static unsigned len_args(unsigned numargs, struct fuse_arg *args)
 {
@@ -282,7 +285,7 @@ __releases(&fc->lock)
 			wake_up_all(&fc->blocked_waitq);
 		}
 		if (fc->num_background == FUSE_CONGESTION_THRESHOLD &&
-		    fc->connected) {
+		    fc->connected && fc->bdi_initialized) {
 			clear_bdi_congested(&fc->bdi, READ);
 			clear_bdi_congested(&fc->bdi, WRITE);
 		}
@@ -400,6 +403,7 @@ void fuse_request_send(struct fuse_conn *fc, struct fuse_req *req)
 	}
 	spin_unlock(&fc->lock);
 }
+EXPORT_SYMBOL_GPL(fuse_request_send);
 
 static void fuse_request_send_nowait_locked(struct fuse_conn *fc,
 					    struct fuse_req *req)
@@ -408,7 +412,8 @@ static void fuse_request_send_nowait_locked(struct fuse_conn *fc,
 	fc->num_background++;
 	if (fc->num_background == FUSE_MAX_BACKGROUND)
 		fc->blocked = 1;
-	if (fc->num_background == FUSE_CONGESTION_THRESHOLD) {
+	if (fc->num_background == FUSE_CONGESTION_THRESHOLD &&
+	    fc->bdi_initialized) {
 		set_bdi_congested(&fc->bdi, READ);
 		set_bdi_congested(&fc->bdi, WRITE);
 	}
@@ -439,6 +444,7 @@ void fuse_request_send_background(struct fuse_conn *fc, struct fuse_req *req)
 	req->isreply = 1;
 	fuse_request_send_nowait(fc, req);
 }
+EXPORT_SYMBOL_GPL(fuse_request_send_background);
 
 /*
  * Called under fc->lock
@@ -1105,8 +1111,9 @@ void fuse_abort_conn(struct fuse_conn *fc)
 	}
 	spin_unlock(&fc->lock);
 }
+EXPORT_SYMBOL_GPL(fuse_abort_conn);
 
-static int fuse_dev_release(struct inode *inode, struct file *file)
+int fuse_dev_release(struct inode *inode, struct file *file)
 {
 	struct fuse_conn *fc = fuse_get_conn(file);
 	if (fc) {
@@ -1120,6 +1127,7 @@ static int fuse_dev_release(struct inode *inode, struct file *file)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(fuse_dev_release);
 
 static int fuse_dev_fasync(int fd, struct file *file, int on)
 {
@@ -1142,6 +1150,7 @@ const struct file_operations fuse_dev_operations = {
 	.release	= fuse_dev_release,
 	.fasync		= fuse_dev_fasync,
 };
+EXPORT_SYMBOL_GPL(fuse_dev_operations);
 
 static struct miscdevice fuse_miscdevice = {
 	.minor = FUSE_MINOR,

fs/fuse/dir.c

@@ -361,19 +361,6 @@ static struct dentry *fuse_lookup(struct inode *dir, struct dentry *entry,
 	return ERR_PTR(err);
 }
 
-/*
- * Synchronous release for the case when something goes wrong in CREATE_OPEN
- */
-static void fuse_sync_release(struct fuse_conn *fc, struct fuse_file *ff,
-			      u64 nodeid, int flags)
-{
-	fuse_release_fill(ff, nodeid, flags, FUSE_RELEASE);
-	ff->reserved_req->force = 1;
-	fuse_request_send(fc, ff->reserved_req);
-	fuse_put_request(fc, ff->reserved_req);
-	kfree(ff);
-}
-
 /*
  * Atomic create+open operation
  *
@@ -445,12 +432,14 @@ static int fuse_create_open(struct inode *dir, struct dentry *entry, int mode,
 		goto out_free_ff;
 
 	fuse_put_request(fc, req);
+	ff->fh = outopen.fh;
+	ff->nodeid = outentry.nodeid;
+	ff->open_flags = outopen.open_flags;
 	inode = fuse_iget(dir->i_sb, outentry.nodeid, outentry.generation,
 			  &outentry.attr, entry_attr_timeout(&outentry), 0);
 	if (!inode) {
 		flags &= ~(O_CREAT | O_EXCL | O_TRUNC);
-		ff->fh = outopen.fh;
-		fuse_sync_release(fc, ff, outentry.nodeid, flags);
+		fuse_sync_release(ff, flags);
 		fuse_send_forget(fc, forget_req, outentry.nodeid, 1);
 		return -ENOMEM;
 	}
@@ -460,11 +449,11 @@ static int fuse_create_open(struct inode *dir, struct dentry *entry, int mode,
 	fuse_invalidate_attr(dir);
 	file = lookup_instantiate_filp(nd, entry, generic_file_open);
 	if (IS_ERR(file)) {
-		ff->fh = outopen.fh;
-		fuse_sync_release(fc, ff, outentry.nodeid, flags);
+		fuse_sync_release(ff, flags);
 		return PTR_ERR(file);
 	}
-	fuse_finish_open(inode, file, ff, &outopen);
+	file->private_data = fuse_file_get(ff);
+	fuse_finish_open(inode, file);
 	return 0;
 
  out_free_ff:
@@ -1035,7 +1024,7 @@ static int fuse_readdir(struct file *file, void *dstbuf, filldir_t filldir)
 	req->out.argpages = 1;
 	req->num_pages = 1;
 	req->pages[0] = page;
-	fuse_read_fill(req, file, inode, file->f_pos, PAGE_SIZE, FUSE_READDIR);
+	fuse_read_fill(req, file, file->f_pos, PAGE_SIZE, FUSE_READDIR);
 	fuse_request_send(fc, req);
 	nbytes = req->out.args[0].size;
 	err = req->out.h.error;
@@ -1101,12 +1090,14 @@ static void fuse_put_link(struct dentry *dentry, struct nameidata *nd, void *c)
 
 static int fuse_dir_open(struct inode *inode, struct file *file)
 {
-	return fuse_open_common(inode, file, 1);
+	return fuse_open_common(inode, file, true);
 }
 
 static int fuse_dir_release(struct inode *inode, struct file *file)
 {
-	return fuse_release_common(inode, file, 1);
+	fuse_release_common(file, FUSE_RELEASEDIR);
+
+	return 0;
 }
 
 static int fuse_dir_fsync(struct file *file, struct dentry *de, int datasync)

fs/fuse/fuse_i.h

@@ -97,8 +97,13 @@ struct fuse_inode {
 	struct list_head writepages;
 };
 
+struct fuse_conn;
+
 /** FUSE specific file data */
 struct fuse_file {
+	/** Fuse connection for this file */
+	struct fuse_conn *fc;
+
 	/** Request reserved for flush and release */
 	struct fuse_req *reserved_req;
 
@@ -108,9 +113,15 @@ struct fuse_file {
 	/** File handle used by userspace */
 	u64 fh;
 
+	/** Node id of this file */
+	u64 nodeid;
+
 	/** Refcount */
 	atomic_t count;
 
+	/** FOPEN_* flags returned by open */
+	u32 open_flags;
+
 	/** Entry on inode's write_files list */
 	struct list_head write_entry;
 
@@ -185,8 +196,6 @@ enum fuse_req_state {
 	FUSE_REQ_FINISHED
 };
 
-struct fuse_conn;
-
 /**
  * A request to the client
  */
@@ -248,11 +257,12 @@ struct fuse_req {
 		struct fuse_forget_in forget_in;
 		struct {
 			struct fuse_release_in in;
-			struct vfsmount *vfsmount;
-			struct dentry *dentry;
+			struct path path;
 		} release;
 		struct fuse_init_in init_in;
 		struct fuse_init_out init_out;
+		struct cuse_init_in cuse_init_in;
+		struct cuse_init_out cuse_init_out;
 		struct {
 			struct fuse_read_in in;
 			u64 attr_ver;
@@ -386,6 +396,9 @@ struct fuse_conn {
 	/** Filesystem supports NFS exporting.  Only set in INIT */
 	unsigned export_support:1;
 
+	/** Set if bdi is valid */
+	unsigned bdi_initialized:1;
+
 	/*
 	 * The following bitfields are only for optimization purposes
 	 * and hence races in setting them will not cause malfunction
@@ -515,25 +528,24 @@ void fuse_send_forget(struct fuse_conn *fc, struct fuse_req *req,
  * Initialize READ or READDIR request
  */
 void fuse_read_fill(struct fuse_req *req, struct file *file,
-		    struct inode *inode, loff_t pos, size_t count, int opcode);
+		    loff_t pos, size_t count, int opcode);
 
 /**
  * Send OPEN or OPENDIR request
  */
-int fuse_open_common(struct inode *inode, struct file *file, int isdir);
+int fuse_open_common(struct inode *inode, struct file *file, bool isdir);
 
 struct fuse_file *fuse_file_alloc(struct fuse_conn *fc);
+struct fuse_file *fuse_file_get(struct fuse_file *ff);
 void fuse_file_free(struct fuse_file *ff);
-void fuse_finish_open(struct inode *inode, struct file *file,
-		      struct fuse_file *ff, struct fuse_open_out *outarg);
+void fuse_finish_open(struct inode *inode, struct file *file);
 
-/** Fill in ff->reserved_req with a RELEASE request */
-void fuse_release_fill(struct fuse_file *ff, u64 nodeid, int flags, int opcode);
+void fuse_sync_release(struct fuse_file *ff, int flags);
 
 /**
  * Send RELEASE or RELEASEDIR request
  */
-int fuse_release_common(struct inode *inode, struct file *file, int isdir);
+void fuse_release_common(struct file *file, int opcode);
 
 /**
  * Send FSYNC or FSYNCDIR request
@@ -652,10 +664,12 @@ void fuse_invalidate_entry_cache(struct dentry *entry);
  */
 struct fuse_conn *fuse_conn_get(struct fuse_conn *fc);
 
+void fuse_conn_kill(struct fuse_conn *fc);
+
 /**
  * Initialize fuse_conn
  */
-int fuse_conn_init(struct fuse_conn *fc, struct super_block *sb);
+void fuse_conn_init(struct fuse_conn *fc);
 
 /**
  * Release reference to fuse_conn
@@ -694,4 +708,13 @@ void fuse_release_nowrite(struct inode *inode);
 
 u64 fuse_get_attr_version(struct fuse_conn *fc);
 
+int fuse_do_open(struct fuse_conn *fc, u64 nodeid, struct file *file,
+		 bool isdir);
+ssize_t fuse_direct_io(struct file *file, const char __user *buf,
+		       size_t count, loff_t *ppos, int write);
+long fuse_do_ioctl(struct file *file, unsigned int cmd, unsigned long arg,
+		   unsigned int flags);
+unsigned fuse_file_poll(struct file *file, poll_table *wait);
+int fuse_dev_release(struct inode *inode, struct file *file);
+
 #endif /* _FS_FUSE_I_H */

fs/fuse/inode.c

@@ -277,11 +277,14 @@ static void fuse_send_destroy(struct fuse_conn *fc)
 	}
 }
 
-static void fuse_put_super(struct super_block *sb)
+static void fuse_bdi_destroy(struct fuse_conn *fc)
 {
-	struct fuse_conn *fc = get_fuse_conn_super(sb);
+	if (fc->bdi_initialized)
+		bdi_destroy(&fc->bdi);
+}
 
-	fuse_send_destroy(fc);
+void fuse_conn_kill(struct fuse_conn *fc)
+{
 	spin_lock(&fc->lock);
 	fc->connected = 0;
 	fc->blocked = 0;
@@ -295,7 +298,16 @@ static void fuse_put_super(struct super_block *sb)
 	list_del(&fc->entry);
 	fuse_ctl_remove_conn(fc);
 	mutex_unlock(&fuse_mutex);
-	bdi_destroy(&fc->bdi);
+	fuse_bdi_destroy(fc);
+}
+EXPORT_SYMBOL_GPL(fuse_conn_kill);
+
+static void fuse_put_super(struct super_block *sb)
+{
+	struct fuse_conn *fc = get_fuse_conn_super(sb);
+
+	fuse_send_destroy(fc);
+	fuse_conn_kill(fc);
 	fuse_conn_put(fc);
 }
 
@@ -466,10 +478,8 @@ static int fuse_show_options(struct seq_file *m, struct vfsmount *mnt)
 	return 0;
 }
 
-int fuse_conn_init(struct fuse_conn *fc, struct super_block *sb)
+void fuse_conn_init(struct fuse_conn *fc)
 {
-	int err;
-
 	memset(fc, 0, sizeof(*fc));
 	spin_lock_init(&fc->lock);
 	mutex_init(&fc->inst_mutex);
@@ -484,49 +494,12 @@ int fuse_conn_init(struct fuse_conn *fc, struct super_block *sb)
 	INIT_LIST_HEAD(&fc->bg_queue);
 	INIT_LIST_HEAD(&fc->entry);
 	atomic_set(&fc->num_waiting, 0);
-	fc->bdi.ra_pages = (VM_MAX_READAHEAD * 1024) / PAGE_CACHE_SIZE;
-	fc->bdi.unplug_io_fn = default_unplug_io_fn;
-	/* fuse does it's own writeback accounting */
-	fc->bdi.capabilities = BDI_CAP_NO_ACCT_WB;
 	fc->khctr = 0;
 	fc->polled_files = RB_ROOT;
-	fc->dev = sb->s_dev;
-	err = bdi_init(&fc->bdi);
-	if (err)
-		goto error_mutex_destroy;
-	if (sb->s_bdev) {
-		err = bdi_register(&fc->bdi, NULL, "%u:%u-fuseblk",
-				   MAJOR(fc->dev), MINOR(fc->dev));
-	} else {
-		err = bdi_register_dev(&fc->bdi, fc->dev);
-	}
-	if (err)
-		goto error_bdi_destroy;
-	/*
-	 * For a single fuse filesystem use max 1% of dirty +
-	 * writeback threshold.
-	 *
-	 * This gives about 1M of write buffer for memory maps on a
-	 * machine with 1G and 10% dirty_ratio, which should be more
-	 * than enough.
-	 *
-	 * Privileged users can raise it by writing to
-	 *
-	 *    /sys/class/bdi/<bdi>/max_ratio
-	 */
-	bdi_set_max_ratio(&fc->bdi, 1);
 	fc->reqctr = 0;
 	fc->blocked = 1;
 	fc->attr_version = 1;
 	get_random_bytes(&fc->scramble_key, sizeof(fc->scramble_key));
-
-	return 0;
-
- error_bdi_destroy:
-	bdi_destroy(&fc->bdi);
- error_mutex_destroy:
-	mutex_destroy(&fc->inst_mutex);
-	return err;
 }
 EXPORT_SYMBOL_GPL(fuse_conn_init);
 
@@ -539,12 +512,14 @@ void fuse_conn_put(struct fuse_conn *fc)
 		fc->release(fc);
 	}
 }
+EXPORT_SYMBOL_GPL(fuse_conn_put);
 
 struct fuse_conn *fuse_conn_get(struct fuse_conn *fc)
 {
 	atomic_inc(&fc->count);
 	return fc;
 }
+EXPORT_SYMBOL_GPL(fuse_conn_get);
 
 static struct inode *fuse_get_root_inode(struct super_block *sb, unsigned mode)
 {
@@ -797,6 +772,48 @@ static void fuse_free_conn(struct fuse_conn *fc)
 	kfree(fc);
 }
 
+static int fuse_bdi_init(struct fuse_conn *fc, struct super_block *sb)
+{
+	int err;
+
+	fc->bdi.ra_pages = (VM_MAX_READAHEAD * 1024) / PAGE_CACHE_SIZE;
+	fc->bdi.unplug_io_fn = default_unplug_io_fn;
+	/* fuse does it's own writeback accounting */
+	fc->bdi.capabilities = BDI_CAP_NO_ACCT_WB;
+
+	err = bdi_init(&fc->bdi);
+	if (err)
+		return err;
+
+	fc->bdi_initialized = 1;
+
+	if (sb->s_bdev) {
+		err =  bdi_register(&fc->bdi, NULL, "%u:%u-fuseblk",
+				    MAJOR(fc->dev), MINOR(fc->dev));
+	} else {
+		err = bdi_register_dev(&fc->bdi, fc->dev);
+	}
+
+	if (err)
+		return err;
+
+	/*
+	 * For a single fuse filesystem use max 1% of dirty +
+	 * writeback threshold.
+	 *
+	 * This gives about 1M of write buffer for memory maps on a
+	 * machine with 1G and 10% dirty_ratio, which should be more
+	 * than enough.
+	 *
+	 * Privileged users can raise it by writing to
+	 *
+	 *    /sys/class/bdi/<bdi>/max_ratio
+	 */
+	bdi_set_max_ratio(&fc->bdi, 1);
+
+	return 0;
+}
+
 static int fuse_fill_super(struct super_block *sb, void *data, int silent)
 {
 	struct fuse_conn *fc;
@@ -843,11 +860,12 @@ static int fuse_fill_super(struct super_block *sb, void *data, int silent)
 	if (!fc)
 		goto err_fput;
 
-	err = fuse_conn_init(fc, sb);
-	if (err) {
-		kfree(fc);
-		goto err_fput;
-	}
+	fuse_conn_init(fc);
+
+	fc->dev = sb->s_dev;
+	err = fuse_bdi_init(fc, sb);
+	if (err)
+		goto err_put_conn;
 
 	fc->release = fuse_free_conn;
 	fc->flags = d.flags;
@@ -911,7 +929,7 @@ static int fuse_fill_super(struct super_block *sb, void *data, int silent)
  err_put_root:
 	dput(root_dentry);
  err_put_conn:
-	bdi_destroy(&fc->bdi);
+	fuse_bdi_destroy(fc);
 	fuse_conn_put(fc);
  err_fput:
 	fput(file);

include/linux/fuse.h

@@ -120,6 +120,13 @@ struct fuse_file_lock {
 #define FUSE_EXPORT_SUPPORT	(1 << 4)
 #define FUSE_BIG_WRITES		(1 << 5)
 
+/**
+ * CUSE INIT request/reply flags
+ *
+ * CUSE_UNRESTRICTED_IOCTL:  use unrestricted ioctl
+ */
+#define CUSE_UNRESTRICTED_IOCTL	(1 << 0)
+
 /**
  * Release flags
  */
@@ -210,6 +217,9 @@ enum fuse_opcode {
 	FUSE_DESTROY       = 38,
 	FUSE_IOCTL         = 39,
 	FUSE_POLL          = 40,
+
+	/* CUSE specific operations */
+	CUSE_INIT          = 4096,
 };
 
 enum fuse_notify_code {
@@ -401,6 +411,27 @@ struct fuse_init_out {
 	__u32	max_write;
 };
 
+#define CUSE_INIT_INFO_MAX 4096
+
+struct cuse_init_in {
+	__u32	major;
+	__u32	minor;
+	__u32	unused;
+	__u32	flags;
+};
+
+struct cuse_init_out {
+	__u32	major;
+	__u32	minor;
+	__u32	unused;
+	__u32	flags;
+	__u32	max_read;
+	__u32	max_write;
+	__u32	dev_major;		/* chardev major */
+	__u32	dev_minor;		/* chardev minor */
+	__u32	spare[10];
+};
+
 struct fuse_interrupt_in {
 	__u64	unique;
 };