Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse

Pull fuse updates from Miklos Szeredi:
 "This contains two patchsets from Maxim Patlasov.

  The first reworks the request throttling so that only async requests
  are throttled.  Wakeup of waiting async requests is also optimized.

  The second series adds support for async processing of direct IO which
  optimizes direct IO and enables the use of the AIO userspace
  interface."

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse:
  fuse: add flag to turn on async direct IO
  fuse: truncate file if async dio failed
  fuse: optimize short direct reads
  fuse: enable asynchronous processing direct IO
  fuse: make fuse_direct_io() aware about AIO
  fuse: add support of async IO
  fuse: move fuse_release_user_pages() up
  fuse: optimize wake_up
  fuse: implement exclusive wakeup for blocked_waitq
  fuse: skip blocking on allocations of synchronous requests
  fuse: add flag fc->initialized
  fuse: make request allocations for background processing explicit

fs/fuse/cuse.c

@@ -92,8 +92,9 @@ static ssize_t cuse_read(struct file *file, char __user *buf, size_t count,
 {
 	loff_t pos = 0;
 	struct iovec iov = { .iov_base = buf, .iov_len = count };
+	struct fuse_io_priv io = { .async = 0, .file = file };
 
-	return fuse_direct_io(file, &iov, 1, count, &pos, 0);
+	return fuse_direct_io(&io, &iov, 1, count, &pos, 0);
 }
 
 static ssize_t cuse_write(struct file *file, const char __user *buf,
@@ -101,12 +102,13 @@ static ssize_t cuse_write(struct file *file, const char __user *buf,
 {
 	loff_t pos = 0;
 	struct iovec iov = { .iov_base = (void __user *)buf, .iov_len = count };
+	struct fuse_io_priv io = { .async = 0, .file = file };
 
 	/*
 	 * No locking or generic_write_checks(), the server is
 	 * responsible for locking and sanity checks.
 	 */
-	return fuse_direct_io(file, &iov, 1, count, &pos, 1);
+	return fuse_direct_io(&io, &iov, 1, count, &pos, 1);
 }
 
 static int cuse_open(struct inode *inode, struct file *file)
@@ -422,7 +424,7 @@ static int cuse_send_init(struct cuse_conn *cc)
 
 	BUILD_BUG_ON(CUSE_INIT_INFO_MAX > PAGE_SIZE);
 
-	req = fuse_get_req(fc, 1);
+	req = fuse_get_req_for_background(fc, 1);
 	if (IS_ERR(req)) {
 		rc = PTR_ERR(req);
 		goto err;
@@ -504,7 +506,7 @@ static int cuse_channel_open(struct inode *inode, struct file *file)
 	cc->fc.release = cuse_fc_release;
 
 	cc->fc.connected = 1;
-	cc->fc.blocked = 0;
+	cc->fc.initialized = 1;
 	rc = cuse_send_init(cc);
 	if (rc) {
 		fuse_conn_put(&cc->fc);

fs/fuse/dev.c

@@ -111,7 +111,7 @@ static void restore_sigs(sigset_t *oldset)
 	sigprocmask(SIG_SETMASK, oldset, NULL);
 }
 
-static void __fuse_get_request(struct fuse_req *req)
+void __fuse_get_request(struct fuse_req *req)
 {
 	atomic_inc(&req->count);
 }
@@ -130,20 +130,30 @@ static void fuse_req_init_context(struct fuse_req *req)
 	req->in.h.pid = current->pid;
 }
 
-struct fuse_req *fuse_get_req(struct fuse_conn *fc, unsigned npages)
+static bool fuse_block_alloc(struct fuse_conn *fc, bool for_background)
+{
+	return !fc->initialized || (for_background && fc->blocked);
+}
+
+static struct fuse_req *__fuse_get_req(struct fuse_conn *fc, unsigned npages,
+				       bool for_background)
 {
 	struct fuse_req *req;
+	int err;
+	atomic_inc(&fc->num_waiting);
+
+	if (fuse_block_alloc(fc, for_background)) {
 		sigset_t oldset;
 		int intr;
-	int err;
 
-	atomic_inc(&fc->num_waiting);
 		block_sigs(&oldset);
-	intr = wait_event_interruptible(fc->blocked_waitq, !fc->blocked);
+		intr = wait_event_interruptible_exclusive(fc->blocked_waitq,
+				!fuse_block_alloc(fc, for_background));
 		restore_sigs(&oldset);
 		err = -EINTR;
 		if (intr)
 			goto out;
+	}
 
 	err = -ENOTCONN;
 	if (!fc->connected)
@@ -151,19 +161,35 @@ struct fuse_req *fuse_get_req(struct fuse_conn *fc, unsigned npages)
 
 	req = fuse_request_alloc(npages);
 	err = -ENOMEM;
-	if (!req)
+	if (!req) {
+		if (for_background)
+			wake_up(&fc->blocked_waitq);
 		goto out;
+	}
 
 	fuse_req_init_context(req);
 	req->waiting = 1;
+	req->background = for_background;
 	return req;
 
  out:
 	atomic_dec(&fc->num_waiting);
 	return ERR_PTR(err);
 }
+
+struct fuse_req *fuse_get_req(struct fuse_conn *fc, unsigned npages)
+{
+	return __fuse_get_req(fc, npages, false);
+}
 EXPORT_SYMBOL_GPL(fuse_get_req);
 
+struct fuse_req *fuse_get_req_for_background(struct fuse_conn *fc,
+					     unsigned npages)
+{
+	return __fuse_get_req(fc, npages, true);
+}
+EXPORT_SYMBOL_GPL(fuse_get_req_for_background);
+
 /*
  * Return request in fuse_file->reserved_req.  However that may
  * currently be in use.  If that is the case, wait for it to become
@@ -225,19 +251,31 @@ struct fuse_req *fuse_get_req_nofail_nopages(struct fuse_conn *fc,
 	struct fuse_req *req;
 
 	atomic_inc(&fc->num_waiting);
-	wait_event(fc->blocked_waitq, !fc->blocked);
+	wait_event(fc->blocked_waitq, fc->initialized);
 	req = fuse_request_alloc(0);
 	if (!req)
 		req = get_reserved_req(fc, file);
 
 	fuse_req_init_context(req);
 	req->waiting = 1;
+	req->background = 0;
 	return req;
 }
 
 void fuse_put_request(struct fuse_conn *fc, struct fuse_req *req)
 {
 	if (atomic_dec_and_test(&req->count)) {
+		if (unlikely(req->background)) {
+			/*
+			 * We get here in the unlikely case that a background
+			 * request was allocated but not sent
+			 */
+			spin_lock(&fc->lock);
+			if (!fc->blocked)
+				wake_up(&fc->blocked_waitq);
+			spin_unlock(&fc->lock);
+		}
+
 		if (req->waiting)
 			atomic_dec(&fc->num_waiting);
 
@@ -335,10 +373,15 @@ __releases(fc->lock)
 	list_del(&req->intr_entry);
 	req->state = FUSE_REQ_FINISHED;
 	if (req->background) {
-		if (fc->num_background == fc->max_background) {
+		req->background = 0;
+
+		if (fc->num_background == fc->max_background)
 			fc->blocked = 0;
-			wake_up_all(&fc->blocked_waitq);
-		}
+
+		/* Wake up next waiter, if any */
+		if (!fc->blocked && waitqueue_active(&fc->blocked_waitq))
+			wake_up(&fc->blocked_waitq);
+
 		if (fc->num_background == fc->congestion_threshold &&
 		    fc->connected && fc->bdi_initialized) {
 			clear_bdi_congested(&fc->bdi, BLK_RW_SYNC);
@@ -442,6 +485,7 @@ __acquires(fc->lock)
 
 static void __fuse_request_send(struct fuse_conn *fc, struct fuse_req *req)
 {
+	BUG_ON(req->background);
 	spin_lock(&fc->lock);
 	if (!fc->connected)
 		req->out.h.error = -ENOTCONN;
@@ -469,7 +513,7 @@ EXPORT_SYMBOL_GPL(fuse_request_send);
 static void fuse_request_send_nowait_locked(struct fuse_conn *fc,
 					    struct fuse_req *req)
 {
-	req->background = 1;
+	BUG_ON(!req->background);
 	fc->num_background++;
 	if (fc->num_background == fc->max_background)
 		fc->blocked = 1;
@@ -2071,6 +2115,7 @@ void fuse_abort_conn(struct fuse_conn *fc)
 	if (fc->connected) {
 		fc->connected = 0;
 		fc->blocked = 0;
+		fc->initialized = 1;
 		end_io_requests(fc);
 		end_queued_requests(fc);
 		end_polls(fc);
@@ -2089,6 +2134,7 @@ int fuse_dev_release(struct inode *inode, struct file *file)
 		spin_lock(&fc->lock);
 		fc->connected = 0;
 		fc->blocked = 0;
+		fc->initialized = 1;
 		end_queued_requests(fc);
 		end_polls(fc);
 		wake_up_all(&fc->blocked_waitq);

fs/fuse/dir.c

@@ -1562,10 +1562,9 @@ void fuse_release_nowrite(struct inode *inode)
  * vmtruncate() doesn't allow for this case, so do the rlimit checking
  * and the actual truncation by hand.
  */
-static int fuse_do_setattr(struct dentry *entry, struct iattr *attr,
+int fuse_do_setattr(struct inode *inode, struct iattr *attr,
 		    struct file *file)
 {
-	struct inode *inode = entry->d_inode;
 	struct fuse_conn *fc = get_fuse_conn(inode);
 	struct fuse_req *req;
 	struct fuse_setattr_in inarg;
@@ -1574,9 +1573,6 @@ static int fuse_do_setattr(struct dentry *entry, struct iattr *attr,
 	loff_t oldsize;
 	int err;
 
-	if (!fuse_allow_current_process(fc))
-		return -EACCES;
-
 	if (!(fc->flags & FUSE_DEFAULT_PERMISSIONS))
 		attr->ia_valid |= ATTR_FORCE;
 
@@ -1671,10 +1667,15 @@ static int fuse_do_setattr(struct dentry *entry, struct iattr *attr,
 
 static int fuse_setattr(struct dentry *entry, struct iattr *attr)
 {
+	struct inode *inode = entry->d_inode;
+
+	if (!fuse_allow_current_process(get_fuse_conn(inode)))
+		return -EACCES;
+
 	if (attr->ia_valid & ATTR_FILE)
-		return fuse_do_setattr(entry, attr, attr->ia_file);
+		return fuse_do_setattr(inode, attr, attr->ia_file);
 	else
-		return fuse_do_setattr(entry, attr, NULL);
+		return fuse_do_setattr(inode, attr, NULL);
 }
 
 static int fuse_getattr(struct vfsmount *mnt, struct dentry *entry,

fs/fuse/fuse_i.h

@@ -228,6 +228,20 @@ enum fuse_req_state {
 	FUSE_REQ_FINISHED
 };
 
+/** The request IO state (for asynchronous processing) */
+struct fuse_io_priv {
+	int async;
+	spinlock_t lock;
+	unsigned reqs;
+	ssize_t bytes;
+	size_t size;
+	__u64 offset;
+	bool write;
+	int err;
+	struct kiocb *iocb;
+	struct file *file;
+};
+
 /**
  * A request to the client
  */
@@ -332,6 +346,9 @@ struct fuse_req {
 	/** Inode used in the request or NULL */
 	struct inode *inode;
 
+	/** AIO control block */
+	struct fuse_io_priv *io;
+
 	/** Link on fi->writepages */
 	struct list_head writepages_entry;
 
@@ -417,6 +434,10 @@ struct fuse_conn {
 	/** Batching of FORGET requests (positive indicates FORGET batch) */
 	int forget_batch;
 
+	/** Flag indicating that INIT reply has been received. Allocating
+	 * any fuse request will be suspended until the flag is set */
+	int initialized;
+
 	/** Flag indicating if connection is blocked.  This will be
 	    the case before the INIT reply is received, and if there
 	    are too many outstading backgrounds requests */
@@ -520,6 +541,9 @@ struct fuse_conn {
 	/** Does the filesystem want adaptive readdirplus? */
 	unsigned readdirplus_auto:1;
 
+	/** Does the filesystem support asynchronous direct-IO submission? */
+	unsigned async_dio:1;
+
 	/** The number of requests waiting for completion */
 	atomic_t num_waiting;
 
@@ -708,6 +732,13 @@ void fuse_request_free(struct fuse_req *req);
  * caller should specify # elements in req->pages[] explicitly
  */
 struct fuse_req *fuse_get_req(struct fuse_conn *fc, unsigned npages);
+struct fuse_req *fuse_get_req_for_background(struct fuse_conn *fc,
+					     unsigned npages);
+
+/*
+ * Increment reference count on request
+ */
+void __fuse_get_request(struct fuse_req *req);
 
 /**
  * Get a request, may fail with -ENOMEM,
@@ -823,7 +854,7 @@ int fuse_reverse_inval_entry(struct super_block *sb, u64 parent_nodeid,
 
 int fuse_do_open(struct fuse_conn *fc, u64 nodeid, struct file *file,
 		 bool isdir);
-ssize_t fuse_direct_io(struct file *file, const struct iovec *iov,
+ssize_t fuse_direct_io(struct fuse_io_priv *io, const struct iovec *iov,
 		       unsigned long nr_segs, size_t count, loff_t *ppos,
 		       int write);
 long fuse_do_ioctl(struct file *file, unsigned int cmd, unsigned long arg,
@@ -835,4 +866,7 @@ int fuse_dev_release(struct inode *inode, struct file *file);
 
 void fuse_write_update_size(struct inode *inode, loff_t pos);
 
+int fuse_do_setattr(struct inode *inode, struct iattr *attr,
+		    struct file *file);
+
 #endif /* _FS_FUSE_I_H */

fs/fuse/inode.c

@@ -346,6 +346,7 @@ static void fuse_send_destroy(struct fuse_conn *fc)
 		fc->destroy_req = NULL;
 		req->in.h.opcode = FUSE_DESTROY;
 		req->force = 1;
+		req->background = 0;
 		fuse_request_send(fc, req);
 		fuse_put_request(fc, req);
 	}
@@ -362,6 +363,7 @@ void fuse_conn_kill(struct fuse_conn *fc)
 	spin_lock(&fc->lock);
 	fc->connected = 0;
 	fc->blocked = 0;
+	fc->initialized = 1;
 	spin_unlock(&fc->lock);
 	/* Flush all readers on this fs */
 	kill_fasync(&fc->fasync, SIGIO, POLL_IN);
@@ -581,7 +583,8 @@ void fuse_conn_init(struct fuse_conn *fc)
 	fc->khctr = 0;
 	fc->polled_files = RB_ROOT;
 	fc->reqctr = 0;
-	fc->blocked = 1;
+	fc->blocked = 0;
+	fc->initialized = 0;
 	fc->attr_version = 1;
 	get_random_bytes(&fc->scramble_key, sizeof(fc->scramble_key));
 }
@@ -868,6 +871,8 @@ static void process_init_reply(struct fuse_conn *fc, struct fuse_req *req)
 				fc->do_readdirplus = 1;
 			if (arg->flags & FUSE_READDIRPLUS_AUTO)
 				fc->readdirplus_auto = 1;
+			if (arg->flags & FUSE_ASYNC_DIO)
+				fc->async_dio = 1;
 		} else {
 			ra_pages = fc->max_read / PAGE_CACHE_SIZE;
 			fc->no_lock = 1;
@@ -880,7 +885,7 @@ static void process_init_reply(struct fuse_conn *fc, struct fuse_req *req)
 		fc->max_write = max_t(unsigned, 4096, fc->max_write);
 		fc->conn_init = 1;
 	}
-	fc->blocked = 0;
+	fc->initialized = 1;
 	wake_up_all(&fc->blocked_waitq);
 }
 
@@ -895,7 +900,7 @@ static void fuse_send_init(struct fuse_conn *fc, struct fuse_req *req)
 		FUSE_EXPORT_SUPPORT | FUSE_BIG_WRITES | FUSE_DONT_MASK |
 		FUSE_SPLICE_WRITE | FUSE_SPLICE_MOVE | FUSE_SPLICE_READ |
 		FUSE_FLOCK_LOCKS | FUSE_IOCTL_DIR | FUSE_AUTO_INVAL_DATA |
-		FUSE_DO_READDIRPLUS | FUSE_READDIRPLUS_AUTO;
+		FUSE_DO_READDIRPLUS | FUSE_READDIRPLUS_AUTO | FUSE_ASYNC_DIO;
 	req->in.h.opcode = FUSE_INIT;
 	req->in.numargs = 1;
 	req->in.args[0].size = sizeof(*arg);
@@ -1043,6 +1048,7 @@ static int fuse_fill_super(struct super_block *sb, void *data, int silent)
 	init_req = fuse_request_alloc(0);
 	if (!init_req)
 		goto err_put_root;
+	init_req->background = 1;
 
 	if (is_bdev) {
 		fc->destroy_req = fuse_request_alloc(0);

include/uapi/linux/fuse.h

@@ -90,6 +90,9 @@
  * 7.21
  *  - add FUSE_READDIRPLUS
  *  - send the requested events in POLL request
+ *
+ * 7.22
+ *  - add FUSE_ASYNC_DIO
  */
 
 #ifndef _LINUX_FUSE_H
@@ -125,7 +128,7 @@
 #define FUSE_KERNEL_VERSION 7
 
 /** Minor version number of this interface */
-#define FUSE_KERNEL_MINOR_VERSION 21
+#define FUSE_KERNEL_MINOR_VERSION 22
 
 /** The node ID of the root inode */
 #define FUSE_ROOT_ID 1
@@ -215,6 +218,7 @@ struct fuse_file_lock {
  * FUSE_AUTO_INVAL_DATA: automatically invalidate cached pages
  * FUSE_DO_READDIRPLUS: do READDIRPLUS (READDIR+LOOKUP in one)
  * FUSE_READDIRPLUS_AUTO: adaptive readdirplus
+ * FUSE_ASYNC_DIO: asynchronous direct I/O submission
  */
 #define FUSE_ASYNC_READ		(1 << 0)
 #define FUSE_POSIX_LOCKS	(1 << 1)
@@ -231,6 +235,7 @@ struct fuse_file_lock {
 #define FUSE_AUTO_INVAL_DATA	(1 << 12)
 #define FUSE_DO_READDIRPLUS	(1 << 13)
 #define FUSE_READDIRPLUS_AUTO	(1 << 14)
+#define FUSE_ASYNC_DIO		(1 << 15)
 
 /**
  * CUSE INIT request/reply flags