Merge with git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git

15 files changed, 434 insertions, 136 deletions

diff --git a/fs/exec.c b/fs/exec.c
index 3234a0c32d5..4121bb55973 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -665,9 +665,7 @@ static int de_thread(struct task_struct *tsk)
 	 * and to assume its PID:
 	 */
 	if (!thread_group_leader(current)) {
-		struct task_struct *parent;
 		struct dentry *proc_dentry1, *proc_dentry2;
-		unsigned long ptrace;
 
 		/*
 		 * Wait for the thread group leader to be a zombie.
@@ -704,22 +702,6 @@ static int de_thread(struct task_struct *tsk)
 		 * two threads with a switched PID, and release
 		 * the former thread group leader:
 		 */
-		ptrace = leader->ptrace;
-		parent = leader->parent;
-		if (unlikely(ptrace) && unlikely(parent == current)) {
-			/*
-			 * Joker was ptracing his own group leader,
-			 * and now he wants to be his own parent!
-			 * We can't have that.
-			 */
-			ptrace = 0;
-		}
-
-		ptrace_unlink(current);
-		ptrace_unlink(leader);
-		remove_parent(current);
-		remove_parent(leader);
-
 
 		/* Become a process group leader with the old leader's pid.
 		 * Note: The old leader also uses thispid until release_task
@@ -732,8 +714,6 @@ static int de_thread(struct task_struct *tsk)
 		attach_pid(current, PIDTYPE_SID,  current->signal->session);
 		list_add_tail(&current->tasks, &init_task.tasks);
 
-		current->parent = current->real_parent = leader->real_parent;
-		leader->parent = leader->real_parent = child_reaper;
 		current->group_leader = current;
 		leader->group_leader = current;
 
@@ -742,13 +722,6 @@ static int de_thread(struct task_struct *tsk)
 		detach_pid(leader, PIDTYPE_SID);
 		list_del_init(&leader->tasks);
 
-		add_parent(current);
-		add_parent(leader);
-		if (ptrace) {
-			current->ptrace = ptrace;
-			__ptrace_link(current, parent);
-		}
-
 		current->exit_signal = SIGCHLD;
 
 		BUG_ON(leader->exit_state != EXIT_ZOMBIE);
diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c
index 6c740f86066..cc750c68fe7 100644
--- a/fs/fuse/dev.c
+++ b/fs/fuse/dev.c
@@ -92,48 +92,50 @@ struct fuse_req *fuse_get_req(struct fuse_conn *fc)
 {
 	struct fuse_req *req;
 	sigset_t oldset;
+	int intr;
 	int err;
 
+	atomic_inc(&fc->num_waiting);
 	block_sigs(&oldset);
-	err = wait_event_interruptible(fc->blocked_waitq, !fc->blocked);
+	intr = wait_event_interruptible(fc->blocked_waitq, !fc->blocked);
 	restore_sigs(&oldset);
-	if (err)
-		return ERR_PTR(-EINTR);
+	err = -EINTR;
+	if (intr)
+		goto out;
 
 	req = fuse_request_alloc();
+	err = -ENOMEM;
 	if (!req)
-		return ERR_PTR(-ENOMEM);
+		goto out;
 
-	atomic_inc(&fc->num_waiting);
-	fuse_request_init(req);
 	req->in.h.uid = current->fsuid;
 	req->in.h.gid = current->fsgid;
 	req->in.h.pid = current->pid;
+	req->waiting = 1;
 	return req;
+
+ out:
+	atomic_dec(&fc->num_waiting);
+	return ERR_PTR(err);
 }
 
 void fuse_put_request(struct fuse_conn *fc, struct fuse_req *req)
 {
 	if (atomic_dec_and_test(&req->count)) {
-		atomic_dec(&fc->num_waiting);
+		if (req->waiting)
+			atomic_dec(&fc->num_waiting);
 		fuse_request_free(req);
 	}
 }
 
-void fuse_release_background(struct fuse_conn *fc, struct fuse_req *req)
+void fuse_remove_background(struct fuse_conn *fc, struct fuse_req *req)
 {
-	iput(req->inode);
-	iput(req->inode2);
-	if (req->file)
-		fput(req->file);
-	spin_lock(&fc->lock);
-	list_del(&req->bg_entry);
+	list_del_init(&req->bg_entry);
 	if (fc->num_background == FUSE_MAX_BACKGROUND) {
 		fc->blocked = 0;
 		wake_up_all(&fc->blocked_waitq);
 	}
 	fc->num_background--;
-	spin_unlock(&fc->lock);
 }
 
 /*
@@ -163,17 +165,27 @@ static void request_end(struct fuse_conn *fc, struct fuse_req *req)
 		wake_up(&req->waitq);
 		fuse_put_request(fc, req);
 	} else {
+		struct inode *inode = req->inode;
+		struct inode *inode2 = req->inode2;
+		struct file *file = req->file;
 		void (*end) (struct fuse_conn *, struct fuse_req *) = req->end;
 		req->end = NULL;
+		req->inode = NULL;
+		req->inode2 = NULL;
+		req->file = NULL;
+		if (!list_empty(&req->bg_entry))
+			fuse_remove_background(fc, req);
 		spin_unlock(&fc->lock);
-		down_read(&fc->sbput_sem);
-		if (fc->mounted)
-			fuse_release_background(fc, req);
-		up_read(&fc->sbput_sem);
+
 		if (end)
 			end(fc, req);
 		else
 			fuse_put_request(fc, req);
+
+		if (file)
+			fput(file);
+		iput(inode);
+		iput(inode2);
 	}
 }
 
@@ -277,6 +289,10 @@ static void queue_request(struct fuse_conn *fc, struct fuse_req *req)
 		len_args(req->in.numargs, (struct fuse_arg *) req->in.args);
 	list_add_tail(&req->list, &fc->pending);
 	req->state = FUSE_REQ_PENDING;
+	if (!req->waiting) {
+		req->waiting = 1;
+		atomic_inc(&fc->num_waiting);
+	}
 	wake_up(&fc->waitq);
 	kill_fasync(&fc->fasync, SIGIO, POLL_IN);
 }
diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index e4f041a11bb..fc342cf7c2c 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -1,6 +1,6 @@
 /*
   FUSE: Filesystem in Userspace
-  Copyright (C) 2001-2005  Miklos Szeredi <miklos@szeredi.hu>
+  Copyright (C) 2001-2006  Miklos Szeredi <miklos@szeredi.hu>
 
   This program can be distributed under the terms of the GNU GPL.
   See the file COPYING.
@@ -565,8 +565,12 @@ static ssize_t fuse_direct_io(struct file *file, const char __user *buf,
 		buf += nres;
 		if (nres != nbytes)
 			break;
-		if (count)
-			fuse_reset_request(req);
+		if (count) {
+			fuse_put_request(fc, req);
+			req = fuse_get_req(fc);
+			if (IS_ERR(req))
+				break;
+		}
 	}
 	fuse_put_request(fc, req);
 	if (res > 0) {
diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h
index 19c7185a754..59661c481d9 100644
--- a/fs/fuse/fuse_i.h
+++ b/fs/fuse/fuse_i.h
@@ -159,6 +159,9 @@ struct fuse_req {
 	/** Data is being copied to/from the request */
 	unsigned locked:1;
 
+	/** Request is counted as "waiting" */
+	unsigned waiting:1;
+
 	/** State of the request */
 	enum fuse_req_state state;
 
@@ -255,15 +258,9 @@ struct fuse_conn {
 	/** waitq for blocked connection */
 	wait_queue_head_t blocked_waitq;
 
-	/** RW semaphore for exclusion with fuse_put_super() */
-	struct rw_semaphore sbput_sem;
-
 	/** The next unique request id */
 	u64 reqctr;
 
-	/** Mount is active */
-	unsigned mounted;
-
 	/** Connection established, cleared on umount, connection
 	    abort and device release */
 	unsigned connected;
@@ -474,11 +471,11 @@ void request_send_noreply(struct fuse_conn *fc, struct fuse_req *req);
 void request_send_background(struct fuse_conn *fc, struct fuse_req *req);
 
 /**
- * Release inodes and file associated with background request
+ * Remove request from the the background list
  */
-void fuse_release_background(struct fuse_conn *fc, struct fuse_req *req);
+void fuse_remove_background(struct fuse_conn *fc, struct fuse_req *req);
 
-/* Abort all requests */
+/** Abort all requests */
 void fuse_abort_conn(struct fuse_conn *fc);
 
 /**
diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c
index fd34037b058..43a6fc0db8a 100644
--- a/fs/fuse/inode.c
+++ b/fs/fuse/inode.c
@@ -204,17 +204,26 @@ static void fuse_put_super(struct super_block *sb)
 {
 	struct fuse_conn *fc = get_fuse_conn_super(sb);
 
-	down_write(&fc->sbput_sem);
-	while (!list_empty(&fc->background))
-		fuse_release_background(fc,
-					list_entry(fc->background.next,
-						   struct fuse_req, bg_entry));
-
 	spin_lock(&fc->lock);
-	fc->mounted = 0;
 	fc->connected = 0;
+	while (!list_empty(&fc->background)) {
+		struct fuse_req *req = list_entry(fc->background.next,
+						  struct fuse_req, bg_entry);
+		struct inode *inode = req->inode;
+		struct inode *inode2 = req->inode2;
+
+		/* File would hold a reference to vfsmount */
+		BUG_ON(req->file);
+		req->inode = NULL;
+		req->inode2 = NULL;
+		fuse_remove_background(fc, req);
+
+		spin_unlock(&fc->lock);
+		iput(inode);
+		iput(inode2);
+		spin_lock(&fc->lock);
+	}
 	spin_unlock(&fc->lock);
-	up_write(&fc->sbput_sem);
 	/* Flush all readers on this fs */
 	kill_fasync(&fc->fasync, SIGIO, POLL_IN);
 	wake_up_all(&fc->waitq);
@@ -386,7 +395,6 @@ static struct fuse_conn *new_conn(void)
 		INIT_LIST_HEAD(&fc->processing);
 		INIT_LIST_HEAD(&fc->io);
 		INIT_LIST_HEAD(&fc->background);
-		init_rwsem(&fc->sbput_sem);
 		kobj_set_kset_s(fc, connections_subsys);
 		kobject_init(&fc->kobj);
 		atomic_set(&fc->num_waiting, 0);
@@ -541,7 +549,6 @@ static int fuse_fill_super(struct super_block *sb, void *data, int silent)
 		goto err_free_req;
 
 	sb->s_root = root_dentry;
-	fc->mounted = 1;
 	fc->connected = 1;
 	kobject_get(&fc->kobj);
 	file->private_data = fc;
diff --git a/fs/partitions/check.c b/fs/partitions/check.c
index af0cb4b9e78..f3b6af07172 100644
--- a/fs/partitions/check.c
+++ b/fs/partitions/check.c
@@ -331,7 +331,9 @@ void delete_partition(struct gendisk *disk, int part)
 	devfs_remove("%s/part%d", disk->devfs_name, part);
 	if (p->holder_dir)
 		kobject_unregister(p->holder_dir);
-	kobject_unregister(&p->kobj);
+	kobject_uevent(&p->kobj, KOBJ_REMOVE);
+	kobject_del(&p->kobj);
+	kobject_put(&p->kobj);
 }
 
 void add_partition(struct gendisk *disk, int part, sector_t start, sector_t len)
@@ -357,7 +359,10 @@ void add_partition(struct gendisk *disk, int part, sector_t start, sector_t len)
 		snprintf(p->kobj.name,KOBJ_NAME_LEN,"%s%d",disk->kobj.name,part);
 	p->kobj.parent = &disk->kobj;
 	p->kobj.ktype = &ktype_part;
-	kobject_register(&p->kobj);
+	kobject_init(&p->kobj);
+	kobject_add(&p->kobj);
+	if (!disk->part_uevent_suppress)
+		kobject_uevent(&p->kobj, KOBJ_ADD);
 	partition_sysfs_add_subdir(p);
 	disk->part[part-1] = p;
 }
@@ -395,6 +400,8 @@ void register_disk(struct gendisk *disk)
 {
 	struct block_device *bdev;
 	char *s;
+	int i;
+	struct hd_struct *p;
 	int err;
 
 	strlcpy(disk->kobj.name,disk->disk_name,KOBJ_NAME_LEN);
@@ -406,13 +413,12 @@ void register_disk(struct gendisk *disk)
 		return;
 	disk_sysfs_symlinks(disk);
  	disk_sysfs_add_subdirs(disk);
-	kobject_uevent(&disk->kobj, KOBJ_ADD);
 
 	/* No minors to use for partitions */
 	if (disk->minors == 1) {
 		if (disk->devfs_name[0] != '\0')
 			devfs_add_disk(disk);
-		return;
+		goto exit;
 	}
 
 	/* always add handle for the whole disk */
@@ -420,16 +426,32 @@ void register_disk(struct gendisk *disk)
 
 	/* No such device (e.g., media were just removed) */
 	if (!get_capacity(disk))
-		return;
+		goto exit;
 
 	bdev = bdget_disk(disk, 0);
 	if (!bdev)
-		return;
+		goto exit;
 
+	/* scan partition table, but suppress uevents */
 	bdev->bd_invalidated = 1;
-	if (blkdev_get(bdev, FMODE_READ, 0) < 0)
-		return;
+	disk->part_uevent_suppress = 1;
+	err = blkdev_get(bdev, FMODE_READ, 0);
+	disk->part_uevent_suppress = 0;
+	if (err < 0)
+		goto exit;
 	blkdev_put(bdev);
+
+exit:
+	/* announce disk after possible partitions are already created */
+	kobject_uevent(&disk->kobj, KOBJ_ADD);
+
+	/* announce possible partitions */
+	for (i = 1; i < disk->minors; i++) {
+		p = disk->part[i-1];
+		if (!p || !p->nr_sects)
+			continue;
+		kobject_uevent(&p->kobj, KOBJ_ADD);
+	}
 }
 
 int rescan_partitions(struct gendisk *disk, struct block_device *bdev)
diff --git a/fs/pipe.c b/fs/pipe.c
index e984beb93a0..7fefb10db8d 100644
--- a/fs/pipe.c
+++ b/fs/pipe.c
@@ -131,12 +131,19 @@ static int anon_pipe_buf_steal(struct pipe_inode_info *pipe,
 	return 0;
 }
 
+static void anon_pipe_buf_get(struct pipe_inode_info *info,
+			      struct pipe_buffer *buf)
+{
+	page_cache_get(buf->page);
+}
+
 static struct pipe_buf_operations anon_pipe_buf_ops = {
 	.can_merge = 1,
 	.map = anon_pipe_buf_map,
 	.unmap = anon_pipe_buf_unmap,
 	.release = anon_pipe_buf_release,
 	.steal = anon_pipe_buf_steal,
+	.get = anon_pipe_buf_get,
 };
 
 static ssize_t
diff --git a/fs/splice.c b/fs/splice.c
index e50a460239d..8d57e89924a 100644
--- a/fs/splice.c
+++ b/fs/splice.c
@@ -125,12 +125,19 @@ static void page_cache_pipe_buf_unmap(struct pipe_inode_info *info,
 	kunmap(buf->page);
 }
 
+static void page_cache_pipe_buf_get(struct pipe_inode_info *info,
+				    struct pipe_buffer *buf)
+{
+	page_cache_get(buf->page);
+}
+
 static struct pipe_buf_operations page_cache_pipe_buf_ops = {
 	.can_merge = 0,
 	.map = page_cache_pipe_buf_map,
 	.unmap = page_cache_pipe_buf_unmap,
 	.release = page_cache_pipe_buf_release,
 	.steal = page_cache_pipe_buf_steal,
+	.get = page_cache_pipe_buf_get,
 };
 
 /*
@@ -231,8 +238,9 @@ static ssize_t move_to_pipe(struct pipe_inode_info *pipe, struct page **pages,
 }
 
 static int
-__generic_file_splice_read(struct file *in, struct pipe_inode_info *pipe,
-			   size_t len, unsigned int flags)
+__generic_file_splice_read(struct file *in, loff_t *ppos,
+			   struct pipe_inode_info *pipe, size_t len,
+			   unsigned int flags)
 {
 	struct address_space *mapping = in->f_mapping;
 	unsigned int offset, nr_pages;
@@ -241,8 +249,8 @@ __generic_file_splice_read(struct file *in, struct pipe_inode_info *pipe,
 	pgoff_t index;
 	int i, error;
 
-	index = in->f_pos >> PAGE_CACHE_SHIFT;
-	offset = in->f_pos & ~PAGE_CACHE_MASK;
+	index = *ppos >> PAGE_CACHE_SHIFT;
+	offset = *ppos & ~PAGE_CACHE_MASK;
 	nr_pages = (len + offset + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
 
 	if (nr_pages > PIPE_BUFFERS)
@@ -348,8 +356,9 @@ fill_it:
  *
  * Will read pages from given file and fill them into a pipe.
  */
-ssize_t generic_file_splice_read(struct file *in, struct pipe_inode_info *pipe,
-				 size_t len, unsigned int flags)
+ssize_t generic_file_splice_read(struct file *in, loff_t *ppos,
+				 struct pipe_inode_info *pipe, size_t len,
+				 unsigned int flags)
 {
 	ssize_t spliced;
 	int ret;
@@ -358,12 +367,12 @@ ssize_t generic_file_splice_read(struct file *in, struct pipe_inode_info *pipe,
 	spliced = 0;
 
 	while (len) {
-		ret = __generic_file_splice_read(in, pipe, len, flags);
+		ret = __generic_file_splice_read(in, ppos, pipe, len, flags);
 
 		if (ret <= 0)
 			break;
 
-		in->f_pos += ret;
+		*ppos += ret;
 		len -= ret;
 		spliced += ret;
 
@@ -561,7 +570,7 @@ typedef int (splice_actor)(struct pipe_inode_info *, struct pipe_buffer *,
  * to the wanted destination. See pipe_to_file/pipe_to_sendpage above.
  */
 static ssize_t move_from_pipe(struct pipe_inode_info *pipe, struct file *out,
-			      size_t len, unsigned int flags,
+			      loff_t *ppos, size_t len, unsigned int flags,
 			      splice_actor *actor)
 {
 	int ret, do_wakeup, err;
@@ -573,7 +582,7 @@ static ssize_t move_from_pipe(struct pipe_inode_info *pipe, struct file *out,
 	sd.total_len = len;
 	sd.flags = flags;
 	sd.file = out;
-	sd.pos = out->f_pos;
+	sd.pos = *ppos;
 
 	if (pipe->inode)
 		mutex_lock(&pipe->inode->i_mutex);
@@ -656,9 +665,7 @@ static ssize_t move_from_pipe(struct pipe_inode_info *pipe, struct file *out,
 		kill_fasync(&pipe->fasync_writers, SIGIO, POLL_OUT);
 	}
 
-	out->f_pos = sd.pos;
 	return ret;
-
 }
 
 /**
@@ -674,12 +681,12 @@ static ssize_t move_from_pipe(struct pipe_inode_info *pipe, struct file *out,
  */
 ssize_t
 generic_file_splice_write(struct pipe_inode_info *pipe, struct file *out,
-			  size_t len, unsigned int flags)
+			  loff_t *ppos, size_t len, unsigned int flags)
 {
 	struct address_space *mapping = out->f_mapping;
 	ssize_t ret;
 
-	ret = move_from_pipe(pipe, out, len, flags, pipe_to_file);
+	ret = move_from_pipe(pipe, out, ppos, len, flags, pipe_to_file);
 
 	/*
 	 * If file or inode is SYNC and we actually wrote some data, sync it.
@@ -715,9 +722,9 @@ EXPORT_SYMBOL(generic_file_splice_write);
  *
  */
 ssize_t generic_splice_sendpage(struct pipe_inode_info *pipe, struct file *out,
-				size_t len, unsigned int flags)
+				loff_t *ppos, size_t len, unsigned int flags)
 {
-	return move_from_pipe(pipe, out, len, flags, pipe_to_sendpage);
+	return move_from_pipe(pipe, out, ppos, len, flags, pipe_to_sendpage);
 }
 
 EXPORT_SYMBOL(generic_splice_sendpage);
@@ -726,9 +733,8 @@ EXPORT_SYMBOL(generic_splice_sendpage);
  * Attempt to initiate a splice from pipe to file.
  */
 static long do_splice_from(struct pipe_inode_info *pipe, struct file *out,
-			   size_t len, unsigned int flags)
+			   loff_t *ppos, size_t len, unsigned int flags)
 {
-	loff_t pos;
 	int ret;
 
 	if (unlikely(!out->f_op || !out->f_op->splice_write))
@@ -737,22 +743,21 @@ static long do_splice_from(struct pipe_inode_info *pipe, struct file *out,
 	if (unlikely(!(out->f_mode & FMODE_WRITE)))
 		return -EBADF;
 
-	pos = out->f_pos;
-
-	ret = rw_verify_area(WRITE, out, &pos, len);
+	ret = rw_verify_area(WRITE, out, ppos, len);
 	if (unlikely(ret < 0))
 		return ret;
 
-	return out->f_op->splice_write(pipe, out, len, flags);
+	return out->f_op->splice_write(pipe, out, ppos, len, flags);
 }
 
 /*
  * Attempt to initiate a splice from a file to a pipe.
  */
-static long do_splice_to(struct file *in, struct pipe_inode_info *pipe,
-			 size_t len, unsigned int flags)
+static long do_splice_to(struct file *in, loff_t *ppos,
+			 struct pipe_inode_info *pipe, size_t len,
+			 unsigned int flags)
 {
-	loff_t pos, isize, left;
+	loff_t isize, left;
 	int ret;
 
 	if (unlikely(!in->f_op || !in->f_op->splice_read))
@@ -761,28 +766,27 @@ static long do_splice_to(struct file *in, struct pipe_inode_info *pipe,
 	if (unlikely(!(in->f_mode & FMODE_READ)))
 		return -EBADF;
 
-	pos = in->f_pos;
-
-	ret = rw_verify_area(READ, in, &pos, len);
+	ret = rw_verify_area(READ, in, ppos, len);
 	if (unlikely(ret < 0))
 		return ret;
 
 	isize = i_size_read(in->f_mapping->host);
-	if (unlikely(in->f_pos >= isize))
+	if (unlikely(*ppos >= isize))
 		return 0;
 	
-	left = isize - in->f_pos;
+	left = isize - *ppos;
 	if (unlikely(left < len))
 		len = left;
 
-	return in->f_op->splice_read(in, pipe, len, flags);
+	return in->f_op->splice_read(in, ppos, pipe, len, flags);
 }
 
-long do_splice_direct(struct file *in, struct file *out, size_t len,
-		      unsigned int flags)
+long do_splice_direct(struct file *in, loff_t *ppos, struct file *out,
+		      size_t len, unsigned int flags)
 {
 	struct pipe_inode_info *pipe;
 	long ret, bytes;
+	loff_t out_off;
 	umode_t i_mode;
 	int i;
 
@@ -820,6 +824,7 @@ long do_splice_direct(struct file *in, struct file *out, size_t len,
 	 */
 	ret = 0;
 	bytes = 0;
+	out_off = 0;
 
 	while (len) {
 		size_t read_len, max_read_len;
@@ -829,7 +834,7 @@ long do_splice_direct(struct file *in, struct file *out, size_t len,
 		 */
 		max_read_len = min(len, (size_t)(PIPE_BUFFERS*PAGE_SIZE));
 
-		ret = do_splice_to(in, pipe, max_read_len, flags);
+		ret = do_splice_to(in, ppos, pipe, max_read_len, flags);
 		if (unlikely(ret < 0))
 			goto out_release;
 
@@ -840,7 +845,7 @@ long do_splice_direct(struct file *in, struct file *out, size_t len,
 		 * must not do the output in nonblocking mode as then we
 		 * could get stuck data in the internal pipe:
 		 */
-		ret = do_splice_from(pipe, out, read_len,
+		ret = do_splice_from(pipe, out, &out_off, read_len,
 				     flags & ~SPLICE_F_NONBLOCK);
 		if (unlikely(ret < 0))
 			goto out_release;
@@ -898,6 +903,7 @@ static long do_splice(struct file *in, loff_t __user *off_in,
 		      size_t len, unsigned int flags)
 {
 	struct pipe_inode_info *pipe;
+	loff_t offset, *off;
 
 	pipe = in->f_dentry->d_inode->i_pipe;
 	if (pipe) {
@@ -906,12 +912,13 @@ static long do_splice(struct file *in, loff_t __user *off_in,
 		if (off_out) {
 			if (out->f_op->llseek == no_llseek)
 				return -EINVAL;
-			if (copy_from_user(&out->f_pos, off_out,
-					   sizeof(loff_t)))
+			if (copy_from_user(&offset, off_out, sizeof(loff_t)))
 				return -EFAULT;
-		}
+			off = &offset;
+		} else
+			off = &out->f_pos;
 
-		return do_splice_from(pipe, out, len, flags);
+		return do_splice_from(pipe, out, off, len, flags);
 	}
 
 	pipe = out->f_dentry->d_inode->i_pipe;
@@ -921,11 +928,13 @@ static long do_splice(struct file *in, loff_t __user *off_in,
 		if (off_in) {
 			if (in->f_op->llseek == no_llseek)
 				return -EINVAL;
-			if (copy_from_user(&in->f_pos, off_in, sizeof(loff_t)))
+			if (copy_from_user(&offset, off_in, sizeof(loff_t)))
 				return -EFAULT;
-		}
+			off = &offset;
+		} else
+			off = &in->f_pos;
 
-		return do_splice_to(in, pipe, len, flags);
+		return do_splice_to(in, off, pipe, len, flags);
 	}
 
 	return -EINVAL;
@@ -961,3 +970,182 @@ asmlinkage long sys_splice(int fd_in, loff_t __user *off_in,
 
 	return error;
 }
+
+/*
+ * Link contents of ipipe to opipe.
+ */
+static int link_pipe(struct pipe_inode_info *ipipe,
+		     struct pipe_inode_info *opipe,
+		     size_t len, unsigned int flags)
+{
+	struct pipe_buffer *ibuf, *obuf;
+	int ret = 0, do_wakeup = 0, i;
+
+	/*
+	 * Potential ABBA deadlock, work around it by ordering lock
+	 * grabbing by inode address. Otherwise two different processes
+	 * could deadlock (one doing tee from A -> B, the other from B -> A).
+	 */
+	if (ipipe->inode < opipe->inode) {
+		mutex_lock(&ipipe->inode->i_mutex);
+		mutex_lock(&opipe->inode->i_mutex);
+	} else {
+		mutex_lock(&opipe->inode->i_mutex);
+		mutex_lock(&ipipe->inode->i_mutex);
+	}
+
+	for (i = 0;; i++) {
+		if (!opipe->readers) {
+			send_sig(SIGPIPE, current, 0);
+			if (!ret)
+				ret = -EPIPE;
+			break;
+		}
+		if (ipipe->nrbufs - i) {
+			ibuf = ipipe->bufs + ((ipipe->curbuf + i) & (PIPE_BUFFERS - 1));
+
+			/*
+			 * If we have room, fill this buffer
+			 */
+			if (opipe->nrbufs < PIPE_BUFFERS) {
+				int nbuf = (opipe->curbuf + opipe->nrbufs) & (PIPE_BUFFERS - 1);
+
+				/*
+				 * Get a reference to this pipe buffer,
+				 * so we can copy the contents over.
+				 */
+				ibuf->ops->get(ipipe, ibuf);
+
+				obuf = opipe->bufs + nbuf;
+				*obuf = *ibuf;
+
+				if (obuf->len > len)
+					obuf->len = len;
+
+				opipe->nrbufs++;
+				do_wakeup = 1;
+				ret += obuf->len;
+				len -= obuf->len;
+
+				if (!len)
+					break;
+				if (opipe->nrbufs < PIPE_BUFFERS)
+					continue;
+			}
+
+			/*
+			 * We have input available, but no output room.
+			 * If we already copied data, return that.
+			 */
+			if (flags & SPLICE_F_NONBLOCK) {
+				if (!ret)
+					ret = -EAGAIN;
+				break;
+			}
+			if (signal_pending(current)) {
+				if (!ret)
+					ret = -ERESTARTSYS;
+				break;
+			}
+			if (do_wakeup) {
+				smp_mb();
+				if (waitqueue_active(&opipe->wait))
+					wake_up_interruptible(&opipe->wait);
+				kill_fasync(&opipe->fasync_readers, SIGIO, POLL_IN);
+				do_wakeup = 0;
+			}
+
+			opipe->waiting_writers++;
+			pipe_wait(opipe);
+			opipe->waiting_writers--;
+			continue;
+		}
+
+		/*
+		 * No input buffers, do the usual checks for available
+		 * writers and blocking and wait if necessary
+		 */
+		if (!ipipe->writers)
+			break;
+		if (!ipipe->waiting_writers) {
+			if (ret)
+				break;
+		}
+		if (flags & SPLICE_F_NONBLOCK) {
+			if (!ret)
+				ret = -EAGAIN;
+			break;
+		}
+		if (signal_pending(current)) {
+			if (!ret)
+				ret = -ERESTARTSYS;
+			break;
+		}
+
+		if (waitqueue_active(&ipipe->wait))
+			wake_up_interruptible_sync(&ipipe->wait);
+		kill_fasync(&ipipe->fasync_writers, SIGIO, POLL_OUT);
+
+		pipe_wait(ipipe);
+	}
+
+	mutex_unlock(&ipipe->inode->i_mutex);
+	mutex_unlock(&opipe->inode->i_mutex);
+
+	if (do_wakeup) {
+		smp_mb();
+		if (waitqueue_active(&opipe->wait))
+			wake_up_interruptible(&opipe->wait);
+		kill_fasync(&opipe->fasync_readers, SIGIO, POLL_IN);
+	}
+
+	return ret;
+}
+
+/*
+ * This is a tee(1) implementation that works on pipes. It doesn't copy
+ * any data, it simply references the 'in' pages on the 'out' pipe.
+ * The 'flags' used are the SPLICE_F_* variants, currently the only
+ * applicable one is SPLICE_F_NONBLOCK.
+ */
+static long do_tee(struct file *in, struct file *out, size_t len,
+		   unsigned int flags)
+{
+	struct pipe_inode_info *ipipe = in->f_dentry->d_inode->i_pipe;
+	struct pipe_inode_info *opipe = out->f_dentry->d_inode->i_pipe;
+
+	/*
+	 * Link ipipe to the two output pipes, consuming as we go along.
+	 */
+	if (ipipe && opipe)
+		return link_pipe(ipipe, opipe, len, flags);
+
+	return -EINVAL;
+}
+
+asmlinkage long sys_tee(int fdin, int fdout, size_t len, unsigned int flags)
+{
+	struct file *in;
+	int error, fput_in;
+
+	if (unlikely(!len))
+		return 0;
+
+	error = -EBADF;
+	in = fget_light(fdin, &fput_in);
+	if (in) {
+		if (in->f_mode & FMODE_READ) {
+			int fput_out;
+			struct file *out = fget_light(fdout, &fput_out);
+
+			if (out) {
+				if (out->f_mode & FMODE_WRITE)
+					error = do_tee(in, out, len, flags);
+				fput_light(out, fput_out);
+			}
+		}
+ 		fput_light(in, fput_in);
+ 	}
+
+	return error;
+}
diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c
index 6cfdc9a8777..610b5bdbe75 100644
--- a/fs/sysfs/dir.c
+++ b/fs/sysfs/dir.c
@@ -43,6 +43,7 @@ static struct sysfs_dirent * sysfs_new_dirent(struct sysfs_dirent * parent_sd,
 
 	memset(sd, 0, sizeof(*sd));
 	atomic_set(&sd->s_count, 1);
+	atomic_set(&sd->s_event, 0);
 	INIT_LIST_HEAD(&sd->s_children);
 	list_add(&sd->s_sibling, &parent_sd->s_children);
 	sd->s_element = element;
diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c
index f1cb1ddde51..cf3786625bf 100644
--- a/fs/sysfs/file.c
+++ b/fs/sysfs/file.c
@@ -6,6 +6,7 @@
 #include <linux/fsnotify.h>
 #include <linux/kobject.h>
 #include <linux/namei.h>
+#include <linux/poll.h>
 #include <asm/uaccess.h>
 #include <asm/semaphore.h>
 
@@ -57,6 +58,7 @@ struct sysfs_buffer {
 	struct sysfs_ops	* ops;
 	struct semaphore	sem;
 	int			needs_read_fill;
+	int			event;
 };
 
 
@@ -72,6 +74,7 @@ struct sysfs_buffer {
  */
 static int fill_read_buffer(struct dentry * dentry, struct sysfs_buffer * buffer)
 {
+	struct sysfs_dirent * sd = dentry->d_fsdata;
 	struct attribute * attr = to_attr(dentry);
 	struct kobject * kobj = to_kobj(dentry->d_parent);
 	struct sysfs_ops * ops = buffer->ops;
@@ -83,6 +86,7 @@ static int fill_read_buffer(struct dentry * dentry, struct sysfs_buffer * buffer
 	if (!buffer->page)
 		return -ENOMEM;
 
+	buffer->event = atomic_read(&sd->s_event);
 	count = ops->show(kobj,attr,buffer->page);
 	buffer->needs_read_fill = 0;
 	BUG_ON(count > (ssize_t)PAGE_SIZE);
@@ -348,12 +352,84 @@ static int sysfs_release(struct inode * inode, struct file * filp)
 	return 0;
 }
 
+/* Sysfs attribute files are pollable.  The idea is that you read
+ * the content and then you use 'poll' or 'select' to wait for
+ * the content to change.  When the content changes (assuming the
+ * manager for the kobject supports notification), poll will
+ * return POLLERR|POLLPRI, and select will return the fd whether
+ * it is waiting for read, write, or exceptions.
+ * Once poll/select indicates that the value has changed, you
+ * need to close and re-open the file, as simply seeking and reading
+ * again will not get new data, or reset the state of 'poll'.
+ * Reminder: this only works for attributes which actively support
+ * it, and it is not possible to test an attribute from userspace
+ * to see if it supports poll (Nether 'poll' or 'select' return
+ * an appropriate error code).  When in doubt, set a suitable timeout value.
+ */
+static unsigned int sysfs_poll(struct file *filp, poll_table *wait)
+{
+	struct sysfs_buffer * buffer = filp->private_data;
+	struct kobject * kobj = to_kobj(filp->f_dentry->d_parent);
+	struct sysfs_dirent * sd = filp->f_dentry->d_fsdata;
+	int res = 0;
+
+	poll_wait(filp, &kobj->poll, wait);
+
+	if (buffer->event != atomic_read(&sd->s_event)) {
+		res = POLLERR|POLLPRI;
+		buffer->needs_read_fill = 1;
+	}
+
+	return res;
+}
+
+
+static struct dentry *step_down(struct dentry *dir, const char * name)
+{
+	struct dentry * de;
+
+	if (dir == NULL || dir->d_inode == NULL)
+		return NULL;
+
+	mutex_lock(&dir->d_inode->i_mutex);
+	de = lookup_one_len(name, dir, strlen(name));
+	mutex_unlock(&dir->d_inode->i_mutex);
+	dput(dir);
+	if (IS_ERR(de))
+		return NULL;
+	if (de->d_inode == NULL) {
+		dput(de);
+		return NULL;
+	}
+	return de;
+}
+
+void sysfs_notify(struct kobject * k, char *dir, char *attr)
+{
+	struct dentry *de = k->dentry;
+	if (de)
+		dget(de);
+	if (de && dir)
+		de = step_down(de, dir);
+	if (de && attr)
+		de = step_down(de, attr);
+	if (de) {
+		struct sysfs_dirent * sd = de->d_fsdata;
+		if (sd)
+			atomic_inc(&sd->s_event);
+		wake_up_interruptible(&k->poll);
+		dput(de);
+	}
+}
+EXPORT_SYMBOL_GPL(sysfs_notify);
+
 const struct file_operations sysfs_file_operations = {
 	.read		= sysfs_read_file,
 	.write		= sysfs_write_file,
 	.llseek		= generic_file_llseek,
 	.open		= sysfs_open_file,
 	.release	= sysfs_release,
+	.poll		= sysfs_poll,
 };
 
 
diff --git a/fs/sysfs/sysfs.h b/fs/sysfs/sysfs.h
index 32958a7c50e..3651ffb5ec0 100644
--- a/fs/sysfs/sysfs.h
+++ b/fs/sysfs/sysfs.h
@@ -11,6 +11,7 @@ extern int sysfs_make_dirent(struct sysfs_dirent *, struct dentry *, void *,
 
 extern int sysfs_add_file(struct dentry *, const struct attribute *, int);
 extern void sysfs_hash_and_remove(struct dentry * dir, const char * name);
+extern struct sysfs_dirent *sysfs_find(struct sysfs_dirent *dir, const char * name);
 
 extern int sysfs_create_subdir(struct kobject *, const char *, struct dentry **);
 extern void sysfs_remove_subdir(struct dentry *);
diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c
index 269721af02f..c847416f6d1 100644
--- a/fs/xfs/linux-2.6/xfs_file.c
+++ b/fs/xfs/linux-2.6/xfs_file.c
@@ -252,6 +252,7 @@ xfs_file_sendfile_invis(
 STATIC ssize_t
 xfs_file_splice_read(
 	struct file		*infilp,
+	loff_t			*ppos,
 	struct pipe_inode_info	*pipe,
 	size_t			len,
 	unsigned int		flags)
@@ -259,13 +260,14 @@ xfs_file_splice_read(
 	vnode_t			*vp = vn_from_inode(infilp->f_dentry->d_inode);
 	ssize_t			rval;
 
-	VOP_SPLICE_READ(vp, infilp, pipe, len, flags, 0, NULL, rval);
+	VOP_SPLICE_READ(vp, infilp, ppos, pipe, len, flags, 0, NULL, rval);
 	return rval;
 }
 
 STATIC ssize_t
 xfs_file_splice_read_invis(
 	struct file		*infilp,
+	loff_t			*ppos,
 	struct pipe_inode_info	*pipe,
 	size_t			len,
 	unsigned int		flags)
@@ -273,7 +275,7 @@ xfs_file_splice_read_invis(
 	vnode_t			*vp = vn_from_inode(infilp->f_dentry->d_inode);
 	ssize_t			rval;
 
-	VOP_SPLICE_READ(vp, infilp, pipe, len, flags, IO_INVIS, NULL, rval);
+	VOP_SPLICE_READ(vp, infilp, ppos, pipe, len, flags, IO_INVIS, NULL, rval);
 	return rval;
 }
 
@@ -281,13 +283,14 @@ STATIC ssize_t
 xfs_file_splice_write(
 	struct pipe_inode_info	*pipe,
 	struct file		*outfilp,
+	loff_t			*ppos,
 	size_t			len,
 	unsigned int		flags)
 {
 	vnode_t			*vp = vn_from_inode(outfilp->f_dentry->d_inode);
 	ssize_t			rval;
 
-	VOP_SPLICE_WRITE(vp, pipe, outfilp, len, flags, 0, NULL, rval);
+	VOP_SPLICE_WRITE(vp, pipe, outfilp, ppos, len, flags, 0, NULL, rval);
 	return rval;
 }
 
@@ -295,13 +298,14 @@ STATIC ssize_t
 xfs_file_splice_write_invis(
 	struct pipe_inode_info	*pipe,
 	struct file		*outfilp,
+	loff_t			*ppos,
 	size_t			len,
 	unsigned int		flags)
 {
 	vnode_t			*vp = vn_from_inode(outfilp->f_dentry->d_inode);
 	ssize_t			rval;
 
-	VOP_SPLICE_WRITE(vp, pipe, outfilp, len, flags, IO_INVIS, NULL, rval);
+	VOP_SPLICE_WRITE(vp, pipe, outfilp, ppos, len, flags, IO_INVIS, NULL, rval);
 	return rval;
 }
 
diff --git a/fs/xfs/linux-2.6/xfs_lrw.c b/fs/xfs/linux-2.6/xfs_lrw.c
index 74a52937f20..67efe330898 100644
--- a/fs/xfs/linux-2.6/xfs_lrw.c
+++ b/fs/xfs/linux-2.6/xfs_lrw.c
@@ -338,6 +338,7 @@ ssize_t
 xfs_splice_read(
 	bhv_desc_t		*bdp,
 	struct file		*infilp,
+	loff_t			*ppos,
 	struct pipe_inode_info	*pipe,
 	size_t			count,
 	int			flags,
@@ -360,7 +361,7 @@ xfs_splice_read(
 		int error;
 
 		error = XFS_SEND_DATA(mp, DM_EVENT_READ, BHV_TO_VNODE(bdp),
-					infilp->f_pos, count,
+					*ppos, count,
 					FILP_DELAY_FLAG(infilp), &locktype);
 		if (error) {
 			xfs_iunlock(ip, XFS_IOLOCK_SHARED);
@@ -368,8 +369,8 @@ xfs_splice_read(
 		}
 	}
 	xfs_rw_enter_trace(XFS_SPLICE_READ_ENTER, &ip->i_iocore,
-			   pipe, count, infilp->f_pos, ioflags);
-	ret = generic_file_splice_read(infilp, pipe, count, flags);
+			   pipe, count, *ppos, ioflags);
+	ret = generic_file_splice_read(infilp, ppos, pipe, count, flags);
 	if (ret > 0)
 		XFS_STATS_ADD(xs_read_bytes, ret);
 
@@ -382,6 +383,7 @@ xfs_splice_write(
 	bhv_desc_t		*bdp,
 	struct pipe_inode_info	*pipe,
 	struct file		*outfilp,
+	loff_t			*ppos,
 	size_t			count,
 	int			flags,
 	int			ioflags,
@@ -403,7 +405,7 @@ xfs_splice_write(
 		int error;
 
 		error = XFS_SEND_DATA(mp, DM_EVENT_WRITE, BHV_TO_VNODE(bdp),
-					outfilp->f_pos, count,
+					*ppos, count,
 					FILP_DELAY_FLAG(outfilp), &locktype);
 		if (error) {
 			xfs_iunlock(ip, XFS_IOLOCK_EXCL);
@@ -411,8 +413,8 @@ xfs_splice_write(
 		}
 	}
 	xfs_rw_enter_trace(XFS_SPLICE_WRITE_ENTER, &ip->i_iocore,
-			   pipe, count, outfilp->f_pos, ioflags);
-	ret = generic_file_splice_write(pipe, outfilp, count, flags);
+			   pipe, count, *ppos, ioflags);
+	ret = generic_file_splice_write(pipe, outfilp, ppos, count, flags);
 	if (ret > 0)
 		XFS_STATS_ADD(xs_write_bytes, ret);
 
diff --git a/fs/xfs/linux-2.6/xfs_lrw.h b/fs/xfs/linux-2.6/xfs_lrw.h
index 55c689a86ad..8f453995235 100644
--- a/fs/xfs/linux-2.6/xfs_lrw.h
+++ b/fs/xfs/linux-2.6/xfs_lrw.h
@@ -93,11 +93,11 @@ extern ssize_t xfs_write(struct bhv_desc *, struct kiocb *,
 extern ssize_t xfs_sendfile(struct bhv_desc *, struct file *,
 				loff_t *, int, size_t, read_actor_t,
 				void *, struct cred *);
-extern ssize_t xfs_splice_read(struct bhv_desc *, struct file *,
+extern ssize_t xfs_splice_read(struct bhv_desc *, struct file *, loff_t *,
 				struct pipe_inode_info *, size_t, int, int,
 				struct cred *);
 extern ssize_t xfs_splice_write(struct bhv_desc *, struct pipe_inode_info *,
-				struct file *, size_t, int, int,
+				struct file *, loff_t *, size_t, int, int,
 				struct cred *);
 
 #endif	/* __XFS_LRW_H__ */
diff --git a/fs/xfs/linux-2.6/xfs_vnode.h b/fs/xfs/linux-2.6/xfs_vnode.h
index 88b09f18628..2a8e16c2235 100644
--- a/fs/xfs/linux-2.6/xfs_vnode.h
+++ b/fs/xfs/linux-2.6/xfs_vnode.h
@@ -173,11 +173,11 @@ typedef ssize_t (*vop_write_t)(bhv_desc_t *, struct kiocb *,
 typedef ssize_t (*vop_sendfile_t)(bhv_desc_t *, struct file *,
 				loff_t *, int, size_t, read_actor_t,
 				void *, struct cred *);
-typedef ssize_t (*vop_splice_read_t)(bhv_desc_t *, struct file *,
+typedef ssize_t (*vop_splice_read_t)(bhv_desc_t *, struct file *, loff_t *,
 				struct pipe_inode_info *, size_t, int, int,
 				struct cred *);
 typedef ssize_t (*vop_splice_write_t)(bhv_desc_t *, struct pipe_inode_info *,
-				struct file *, size_t, int, int,
+				struct file *, loff_t *, size_t, int, int,
 				struct cred *);
 typedef int	(*vop_ioctl_t)(bhv_desc_t *, struct inode *, struct file *,
 				int, unsigned int, void __user *);
@@ -284,10 +284,10 @@ typedef struct vnodeops {
 	rv = _VOP_(vop_write, vp)((vp)->v_fbhv,file,iov,segs,offset,ioflags,cr)
 #define VOP_SENDFILE(vp,f,off,ioflags,cnt,act,targ,cr,rv)		\
 	rv = _VOP_(vop_sendfile, vp)((vp)->v_fbhv,f,off,ioflags,cnt,act,targ,cr)
-#define VOP_SPLICE_READ(vp,f,pipe,cnt,fl,iofl,cr,rv)			\
-	rv = _VOP_(vop_splice_read, vp)((vp)->v_fbhv,f,pipe,cnt,fl,iofl,cr)
-#define VOP_SPLICE_WRITE(vp,f,pipe,cnt,fl,iofl,cr,rv)			\
-	rv = _VOP_(vop_splice_write, vp)((vp)->v_fbhv,f,pipe,cnt,fl,iofl,cr)
+#define VOP_SPLICE_READ(vp,f,o,pipe,cnt,fl,iofl,cr,rv)			\
+	rv = _VOP_(vop_splice_read, vp)((vp)->v_fbhv,f,o,pipe,cnt,fl,iofl,cr)
+#define VOP_SPLICE_WRITE(vp,f,o,pipe,cnt,fl,iofl,cr,rv)			\
+	rv = _VOP_(vop_splice_write, vp)((vp)->v_fbhv,f,o,pipe,cnt,fl,iofl,cr)
 #define VOP_BMAP(vp,of,sz,rw,b,n,rv)					\
 	rv = _VOP_(vop_bmap, vp)((vp)->v_fbhv,of,sz,rw,b,n)
 #define VOP_OPEN(vp, cr, rv)						\