[PATCH] relayfs

Here's the latest version of relayfs, against linux-2.6.11-mm2.  I'm hoping
you'll consider putting this version back into your tree - the previous
rounds of comment seem to have shaken out all the API issues and the number
of comments on the code itself have also steadily dwindled.

This patch is essentially the same as the relayfs redux part 5 patch, with
some minor changes based on reviewer comments.  Thanks again to Pekka
Enberg for those.  The patch size without documentation is now a little
smaller at just over 40k.  Here's a detailed list of the changes:

- removed the attribute_flags in relay open and changed it to a
  boolean specifying either overwrite or no-overwrite mode, and removed
  everything referencing the attribute flags.
- added a check for NULL names in relayfs_create_entry()
- got rid of the unnecessary multiple labels in relay_create_buf()
- some minor simplification of relay_alloc_buf() which got rid of a
  couple params
- updated the Documentation

In addition, this version (through code contained in the relay-apps tarball
linked to below, not as part of the relayfs patch) tries to make it as easy
as possible to create the cooperating kernel/user pieces of a typical and
common type of logging application, one where kernel logging is kicked off
when a user space data collection app starts and stops when the collection
app exits, with the data being automatically logged to disk in between.  To
create this type of application, you basically just include a header file
(relay-app.h, included in the relay-apps tarball) in your kernel module,
define a couple of callbacks and call an initialization function, and on
the user side call a single function that sets up and continuously monitors
the buffers, and writes data to files as it becomes available.  Channels
are created when the collection app is started and destroyed when it exits,
not when the kernel module is inserted, so different channel buffer sizes
can be specified for each separate run via command-line options.  See the
README in the relay-apps tarball for details.

Also included in the relay-apps tarball are a couple examples
demonstrating how you can use this to create quick and dirty kernel
logging/debugging applications.  They are:

- tprintk, short for 'tee printk', which temporarily puts a kprobe on
  printk() and writes a duplicate stream of printk output to a relayfs
  channel.  This could be used anywhere there's printk() debugging code
  in the kernel which you'd like to exercise, but would rather not have
  your system logs cluttered with debugging junk.  You'd probably want
  to kill klogd while you do this, otherwise there wouldn't be much
  point (since putting a kprobe on printk() doesn't change the output
  of printk()).  I've used this method to temporarily divert the packet
  logging output of the iptables LOG target from the system logs to
  relayfs files instead, for instance.

- klog, which just provides a printk-like formatted logging function
  on top of relayfs.  Again, you can use this to keep stuff out of your
  system logs if used in place of printk.

The example applications can be found here:

http://prdownloads.sourceforge.net/dprobes/relay-apps.tar.gz?download

From: Christoph Hellwig <hch@lst.de>

  avoid lookup_hash usage in relayfs

Signed-off-by: Tom Zanussi <zanussi@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

6 files changed, 1257 insertions, 0 deletions

diff --git a/fs/relayfs/Makefile b/fs/relayfs/Makefile
new file mode 100644
index 000000000000..e76e182cdb38
--- /dev/null
+++ b/fs/relayfs/Makefile
@@ -0,0 +1,4 @@
+obj-$(CONFIG_RELAYFS_FS) += relayfs.o
+
+relayfs-y := relay.o inode.o buffers.o
+
diff --git a/fs/relayfs/buffers.c b/fs/relayfs/buffers.c
new file mode 100644
index 000000000000..2aa8e2719999
--- /dev/null
+++ b/fs/relayfs/buffers.c
@@ -0,0 +1,189 @@
+/*
+ * RelayFS buffer management code.
+ *
+ * Copyright (C) 2002-2005 - Tom Zanussi (zanussi@us.ibm.com), IBM Corp
+ * Copyright (C) 1999-2005 - Karim Yaghmour (karim@opersys.com)
+ *
+ * This file is released under the GPL.
+ */
+
+#include <linux/module.h>
+#include <linux/vmalloc.h>
+#include <linux/mm.h>
+#include <linux/relayfs_fs.h>
+#include "relay.h"
+#include "buffers.h"
+
+/*
+ * close() vm_op implementation for relayfs file mapping.
+ */
+static void relay_file_mmap_close(struct vm_area_struct *vma)
+{
+	struct rchan_buf *buf = vma->vm_private_data;
+	buf->chan->cb->buf_unmapped(buf, vma->vm_file);
+}
+
+/*
+ * nopage() vm_op implementation for relayfs file mapping.
+ */
+static struct page *relay_buf_nopage(struct vm_area_struct *vma,
+				     unsigned long address,
+				     int *type)
+{
+	struct page *page;
+	struct rchan_buf *buf = vma->vm_private_data;
+	unsigned long offset = address - vma->vm_start;
+
+	if (address > vma->vm_end)
+		return NOPAGE_SIGBUS; /* Disallow mremap */
+	if (!buf)
+		return NOPAGE_OOM;
+
+	page = vmalloc_to_page(buf->start + offset);
+	if (!page)
+		return NOPAGE_OOM;
+	get_page(page);
+
+	if (type)
+		*type = VM_FAULT_MINOR;
+
+	return page;
+}
+
+/*
+ * vm_ops for relay file mappings.
+ */
+static struct vm_operations_struct relay_file_mmap_ops = {
+	.nopage = relay_buf_nopage,
+	.close = relay_file_mmap_close,
+};
+
+/**
+ *	relay_mmap_buf: - mmap channel buffer to process address space
+ *	@buf: relay channel buffer
+ *	@vma: vm_area_struct describing memory to be mapped
+ *
+ *	Returns 0 if ok, negative on error
+ *
+ *	Caller should already have grabbed mmap_sem.
+ */
+int relay_mmap_buf(struct rchan_buf *buf, struct vm_area_struct *vma)
+{
+	unsigned long length = vma->vm_end - vma->vm_start;
+	struct file *filp = vma->vm_file;
+
+	if (!buf)
+		return -EBADF;
+
+	if (length != (unsigned long)buf->chan->alloc_size)
+		return -EINVAL;
+
+	vma->vm_ops = &relay_file_mmap_ops;
+	vma->vm_private_data = buf;
+	buf->chan->cb->buf_mapped(buf, filp);
+
+	return 0;
+}
+
+/**
+ *	relay_alloc_buf - allocate a channel buffer
+ *	@buf: the buffer struct
+ *	@size: total size of the buffer
+ *
+ *	Returns a pointer to the resulting buffer, NULL if unsuccessful
+ */
+static void *relay_alloc_buf(struct rchan_buf *buf, unsigned long size)
+{
+	void *mem;
+	unsigned int i, j, n_pages;
+
+	size = PAGE_ALIGN(size);
+	n_pages = size >> PAGE_SHIFT;
+
+	buf->page_array = kcalloc(n_pages, sizeof(struct page *), GFP_KERNEL);
+	if (!buf->page_array)
+		return NULL;
+
+	for (i = 0; i < n_pages; i++) {
+		buf->page_array[i] = alloc_page(GFP_KERNEL);
+		if (unlikely(!buf->page_array[i]))
+			goto depopulate;
+	}
+	mem = vmap(buf->page_array, n_pages, GFP_KERNEL, PAGE_KERNEL);
+	if (!mem)
+		goto depopulate;
+
+	memset(mem, 0, size);
+	buf->page_count = n_pages;
+	return mem;
+
+depopulate:
+	for (j = 0; j < i; j++)
+		__free_page(buf->page_array[j]);
+	kfree(buf->page_array);
+	return NULL;
+}
+
+/**
+ *	relay_create_buf - allocate and initialize a channel buffer
+ *	@alloc_size: size of the buffer to allocate
+ *	@n_subbufs: number of sub-buffers in the channel
+ *
+ *	Returns channel buffer if successful, NULL otherwise
+ */
+struct rchan_buf *relay_create_buf(struct rchan *chan)
+{
+	struct rchan_buf *buf = kcalloc(1, sizeof(struct rchan_buf), GFP_KERNEL);
+	if (!buf)
+		return NULL;
+
+	buf->padding = kmalloc(chan->n_subbufs * sizeof(size_t *), GFP_KERNEL);
+	if (!buf->padding)
+		goto free_buf;
+
+	buf->start = relay_alloc_buf(buf, chan->alloc_size);
+	if (!buf->start)
+		goto free_buf;
+
+	buf->chan = chan;
+	kref_get(&buf->chan->kref);
+	return buf;
+
+free_buf:
+	kfree(buf->padding);
+	kfree(buf);
+	return NULL;
+}
+
+/**
+ *	relay_destroy_buf - destroy an rchan_buf struct and associated buffer
+ *	@buf: the buffer struct
+ */
+void relay_destroy_buf(struct rchan_buf *buf)
+{
+	struct rchan *chan = buf->chan;
+	unsigned int i;
+
+	if (likely(buf->start)) {
+		vunmap(buf->start);
+		for (i = 0; i < buf->page_count; i++)
+			__free_page(buf->page_array[i]);
+		kfree(buf->page_array);
+	}
+	kfree(buf->padding);
+	kfree(buf);
+	kref_put(&chan->kref, relay_destroy_channel);
+}
+
+/**
+ *	relay_remove_buf - remove a channel buffer
+ *
+ *	Removes the file from the relayfs fileystem, which also frees the
+ *	rchan_buf_struct and the channel buffer.  Should only be called from
+ *	kref_put().
+ */
+void relay_remove_buf(struct kref *kref)
+{
+	struct rchan_buf *buf = container_of(kref, struct rchan_buf, kref);
+	relayfs_remove(buf->dentry);
+}
diff --git a/fs/relayfs/buffers.h b/fs/relayfs/buffers.h
new file mode 100644
index 000000000000..37a12493f641
--- /dev/null
+++ b/fs/relayfs/buffers.h
@@ -0,0 +1,12 @@
+#ifndef _BUFFERS_H
+#define _BUFFERS_H
+
+/* This inspired by rtai/shmem */
+#define FIX_SIZE(x) (((x) - 1) & PAGE_MASK) + PAGE_SIZE
+
+extern int relay_mmap_buf(struct rchan_buf *buf, struct vm_area_struct *vma);
+extern struct rchan_buf *relay_create_buf(struct rchan *chan);
+extern void relay_destroy_buf(struct rchan_buf *buf);
+extern void relay_remove_buf(struct kref *kref);
+
+#endif/* _BUFFERS_H */
diff --git a/fs/relayfs/inode.c b/fs/relayfs/inode.c
new file mode 100644
index 000000000000..0f7f88d067ad
--- /dev/null
+++ b/fs/relayfs/inode.c
@@ -0,0 +1,609 @@
+/*
+ * VFS-related code for RelayFS, a high-speed data relay filesystem.
+ *
+ * Copyright (C) 2003-2005 - Tom Zanussi <zanussi@us.ibm.com>, IBM Corp
+ * Copyright (C) 2003-2005 - Karim Yaghmour <karim@opersys.com>
+ *
+ * Based on ramfs, Copyright (C) 2002 - Linus Torvalds
+ *
+ * This file is released under the GPL.
+ */
+
+#include <linux/module.h>
+#include <linux/fs.h>
+#include <linux/mount.h>
+#include <linux/pagemap.h>
+#include <linux/init.h>
+#include <linux/string.h>
+#include <linux/backing-dev.h>
+#include <linux/namei.h>
+#include <linux/poll.h>
+#include <linux/relayfs_fs.h>
+#include "relay.h"
+#include "buffers.h"
+
+#define RELAYFS_MAGIC			0xF0B4A981
+
+static struct vfsmount *		relayfs_mount;
+static int				relayfs_mount_count;
+static kmem_cache_t *			relayfs_inode_cachep;
+
+static struct backing_dev_info		relayfs_backing_dev_info = {
+	.ra_pages	= 0,	/* No readahead */
+	.capabilities	= BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_WRITEBACK,
+};
+
+static struct inode *relayfs_get_inode(struct super_block *sb, int mode,
+				       struct rchan *chan)
+{
+	struct rchan_buf *buf = NULL;
+	struct inode *inode;
+
+	if (S_ISREG(mode)) {
+		BUG_ON(!chan);
+		buf = relay_create_buf(chan);
+		if (!buf)
+			return NULL;
+	}
+
+	inode = new_inode(sb);
+	if (!inode) {
+		relay_destroy_buf(buf);
+		return NULL;
+	}
+
+	inode->i_mode = mode;
+	inode->i_uid = 0;
+	inode->i_gid = 0;
+	inode->i_blksize = PAGE_CACHE_SIZE;
+	inode->i_blocks = 0;
+	inode->i_mapping->backing_dev_info = &relayfs_backing_dev_info;
+	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
+	switch (mode & S_IFMT) {
+	case S_IFREG:
+		inode->i_fop = &relayfs_file_operations;
+		RELAYFS_I(inode)->buf = buf;
+		break;
+	case S_IFDIR:
+		inode->i_op = &simple_dir_inode_operations;
+		inode->i_fop = &simple_dir_operations;
+
+		/* directory inodes start off with i_nlink == 2 (for "." entry) */
+		inode->i_nlink++;
+		break;
+	default:
+		break;
+	}
+
+	return inode;
+}
+
+/**
+ *	relayfs_create_entry - create a relayfs directory or file
+ *	@name: the name of the file to create
+ *	@parent: parent directory
+ *	@mode: mode
+ *	@chan: relay channel associated with the file
+ *
+ *	Returns the new dentry, NULL on failure
+ *
+ *	Creates a file or directory with the specifed permissions.
+ */
+static struct dentry *relayfs_create_entry(const char *name,
+					   struct dentry *parent,
+					   int mode,
+					   struct rchan *chan)
+{
+	struct dentry *d;
+	struct inode *inode;
+	int error = 0;
+
+	BUG_ON(!name || !(S_ISREG(mode) || S_ISDIR(mode)));
+
+	error = simple_pin_fs("relayfs", &relayfs_mount, &relayfs_mount_count);
+	if (error) {
+		printk(KERN_ERR "Couldn't mount relayfs: errcode %d\n", error);
+		return NULL;
+	}
+
+	if (!parent && relayfs_mount && relayfs_mount->mnt_sb)
+		parent = relayfs_mount->mnt_sb->s_root;
+
+	if (!parent) {
+		simple_release_fs(&relayfs_mount, &relayfs_mount_count);
+		return NULL;
+	}
+
+	parent = dget(parent);
+	down(&parent->d_inode->i_sem);
+	d = lookup_one_len(name, parent, strlen(name));
+	if (IS_ERR(d)) {
+		d = NULL;
+		goto release_mount;
+	}
+
+	if (d->d_inode) {
+		d = NULL;
+		goto release_mount;
+	}
+
+	inode = relayfs_get_inode(parent->d_inode->i_sb, mode, chan);
+	if (!inode) {
+		d = NULL;
+		goto release_mount;
+	}
+
+	d_instantiate(d, inode);
+	dget(d);	/* Extra count - pin the dentry in core */
+
+	if (S_ISDIR(mode))
+		parent->d_inode->i_nlink++;
+
+	goto exit;
+
+release_mount:
+	simple_release_fs(&relayfs_mount, &relayfs_mount_count);
+
+exit:
+	up(&parent->d_inode->i_sem);
+	dput(parent);
+	return d;
+}
+
+/**
+ *	relayfs_create_file - create a file in the relay filesystem
+ *	@name: the name of the file to create
+ *	@parent: parent directory
+ *	@mode: mode, if not specied the default perms are used
+ *	@chan: channel associated with the file
+ *
+ *	Returns file dentry if successful, NULL otherwise.
+ *
+ *	The file will be created user r on behalf of current user.
+ */
+struct dentry *relayfs_create_file(const char *name, struct dentry *parent,
+				   int mode, struct rchan *chan)
+{
+	if (!mode)
+		mode = S_IRUSR;
+	mode = (mode & S_IALLUGO) | S_IFREG;
+
+	return relayfs_create_entry(name, parent, mode, chan);
+}
+
+/**
+ *	relayfs_create_dir - create a directory in the relay filesystem
+ *	@name: the name of the directory to create
+ *	@parent: parent directory, NULL if parent should be fs root
+ *
+ *	Returns directory dentry if successful, NULL otherwise.
+ *
+ *	The directory will be created world rwx on behalf of current user.
+ */
+struct dentry *relayfs_create_dir(const char *name, struct dentry *parent)
+{
+	int mode = S_IFDIR | S_IRWXU | S_IRUGO | S_IXUGO;
+	return relayfs_create_entry(name, parent, mode, NULL);
+}
+
+/**
+ *	relayfs_remove - remove a file or directory in the relay filesystem
+ *	@dentry: file or directory dentry
+ *
+ *	Returns 0 if successful, negative otherwise.
+ */
+int relayfs_remove(struct dentry *dentry)
+{
+	struct dentry *parent;
+	int error = 0;
+
+	if (!dentry)
+		return -EINVAL;
+	parent = dentry->d_parent;
+	if (!parent)
+		return -EINVAL;
+
+	parent = dget(parent);
+	down(&parent->d_inode->i_sem);
+	if (dentry->d_inode) {
+		if (S_ISDIR(dentry->d_inode->i_mode))
+			error = simple_rmdir(parent->d_inode, dentry);
+		else
+			error = simple_unlink(parent->d_inode, dentry);
+		if (!error)
+			d_delete(dentry);
+	}
+	if (!error)
+		dput(dentry);
+	up(&parent->d_inode->i_sem);
+	dput(parent);
+
+	if (!error)
+		simple_release_fs(&relayfs_mount, &relayfs_mount_count);
+
+	return error;
+}
+
+/**
+ *	relayfs_remove_dir - remove a directory in the relay filesystem
+ *	@dentry: directory dentry
+ *
+ *	Returns 0 if successful, negative otherwise.
+ */
+int relayfs_remove_dir(struct dentry *dentry)
+{
+	return relayfs_remove(dentry);
+}
+
+/**
+ *	relayfs_open - open file op for relayfs files
+ *	@inode: the inode
+ *	@filp: the file
+ *
+ *	Increments the channel buffer refcount.
+ */
+static int relayfs_open(struct inode *inode, struct file *filp)
+{
+	struct rchan_buf *buf = RELAYFS_I(inode)->buf;
+	kref_get(&buf->kref);
+
+	return 0;
+}
+
+/**
+ *	relayfs_mmap - mmap file op for relayfs files
+ *	@filp: the file
+ *	@vma: the vma describing what to map
+ *
+ *	Calls upon relay_mmap_buf to map the file into user space.
+ */
+static int relayfs_mmap(struct file *filp, struct vm_area_struct *vma)
+{
+	struct inode *inode = filp->f_dentry->d_inode;
+	return relay_mmap_buf(RELAYFS_I(inode)->buf, vma);
+}
+
+/**
+ *	relayfs_poll - poll file op for relayfs files
+ *	@filp: the file
+ *	@wait: poll table
+ *
+ *	Poll implemention.
+ */
+static unsigned int relayfs_poll(struct file *filp, poll_table *wait)
+{
+	unsigned int mask = 0;
+	struct inode *inode = filp->f_dentry->d_inode;
+	struct rchan_buf *buf = RELAYFS_I(inode)->buf;
+
+	if (buf->finalized)
+		return POLLERR;
+
+	if (filp->f_mode & FMODE_READ) {
+		poll_wait(filp, &buf->read_wait, wait);
+		if (!relay_buf_empty(buf))
+			mask |= POLLIN | POLLRDNORM;
+	}
+
+	return mask;
+}
+
+/**
+ *	relayfs_release - release file op for relayfs files
+ *	@inode: the inode
+ *	@filp: the file
+ *
+ *	Decrements the channel refcount, as the filesystem is
+ *	no longer using it.
+ */
+static int relayfs_release(struct inode *inode, struct file *filp)
+{
+	struct rchan_buf *buf = RELAYFS_I(inode)->buf;
+	kref_put(&buf->kref, relay_remove_buf);
+
+	return 0;
+}
+
+/**
+ *	relayfs_read_consume - update the consumed count for the buffer
+ */
+static void relayfs_read_consume(struct rchan_buf *buf,
+				 size_t read_pos,
+				 size_t bytes_consumed)
+{
+	size_t subbuf_size = buf->chan->subbuf_size;
+	size_t n_subbufs = buf->chan->n_subbufs;
+	size_t read_subbuf;
+
+	if (buf->bytes_consumed + bytes_consumed > subbuf_size) {
+		relay_subbufs_consumed(buf->chan, buf->cpu, 1);
+		buf->bytes_consumed = 0;
+	}
+
+	buf->bytes_consumed += bytes_consumed;
+	read_subbuf = read_pos / buf->chan->subbuf_size;
+	if (buf->bytes_consumed + buf->padding[read_subbuf] == subbuf_size) {
+		if ((read_subbuf == buf->subbufs_produced % n_subbufs) &&
+		    (buf->offset == subbuf_size))
+			return;
+		relay_subbufs_consumed(buf->chan, buf->cpu, 1);
+		buf->bytes_consumed = 0;
+	}
+}
+
+/**
+ *	relayfs_read_avail - boolean, are there unconsumed bytes available?
+ */
+static int relayfs_read_avail(struct rchan_buf *buf, size_t read_pos)
+{
+	size_t bytes_produced, bytes_consumed, write_offset;
+	size_t subbuf_size = buf->chan->subbuf_size;
+	size_t n_subbufs = buf->chan->n_subbufs;
+	size_t produced = buf->subbufs_produced % n_subbufs;
+	size_t consumed = buf->subbufs_consumed % n_subbufs;
+
+	write_offset = buf->offset > subbuf_size ? subbuf_size : buf->offset;
+
+	if (consumed > produced) {
+		if ((produced > n_subbufs) &&
+		    (produced + n_subbufs - consumed <= n_subbufs))
+			produced += n_subbufs;
+	} else if (consumed == produced) {
+		if (buf->offset > subbuf_size) {
+			produced += n_subbufs;
+			if (buf->subbufs_produced == buf->subbufs_consumed)
+				consumed += n_subbufs;
+		}
+	}
+
+	if (buf->offset > subbuf_size)
+		bytes_produced = (produced - 1) * subbuf_size + write_offset;
+	else
+		bytes_produced = produced * subbuf_size + write_offset;
+	bytes_consumed = consumed * subbuf_size + buf->bytes_consumed;
+
+	if (bytes_produced == bytes_consumed)
+		return 0;
+
+	relayfs_read_consume(buf, read_pos, 0);
+
+	return 1;
+}
+
+/**
+ *	relayfs_read_subbuf_avail - return bytes available in sub-buffer
+ */
+static size_t relayfs_read_subbuf_avail(size_t read_pos,
+					struct rchan_buf *buf)
+{
+	size_t padding, avail = 0;
+	size_t read_subbuf, read_offset, write_subbuf, write_offset;
+	size_t subbuf_size = buf->chan->subbuf_size;
+
+	write_subbuf = (buf->data - buf->start) / subbuf_size;
+	write_offset = buf->offset > subbuf_size ? subbuf_size : buf->offset;
+	read_subbuf = read_pos / subbuf_size;
+	read_offset = read_pos % subbuf_size;
+	padding = buf->padding[read_subbuf];
+
+	if (read_subbuf == write_subbuf) {
+		if (read_offset + padding < write_offset)
+			avail = write_offset - (read_offset + padding);
+	} else
+		avail = (subbuf_size - padding) - read_offset;
+
+	return avail;
+}
+
+/**
+ *	relayfs_read_start_pos - find the first available byte to read
+ *
+ *	If the read_pos is in the middle of padding, return the
+ *	position of the first actually available byte, otherwise
+ *	return the original value.
+ */
+static size_t relayfs_read_start_pos(size_t read_pos,
+				     struct rchan_buf *buf)
+{
+	size_t read_subbuf, padding, padding_start, padding_end;
+	size_t subbuf_size = buf->chan->subbuf_size;
+	size_t n_subbufs = buf->chan->n_subbufs;
+
+	read_subbuf = read_pos / subbuf_size;
+	padding = buf->padding[read_subbuf];
+	padding_start = (read_subbuf + 1) * subbuf_size - padding;
+	padding_end = (read_subbuf + 1) * subbuf_size;
+	if (read_pos >= padding_start && read_pos < padding_end) {
+		read_subbuf = (read_subbuf + 1) % n_subbufs;
+		read_pos = read_subbuf * subbuf_size;
+	}
+
+	return read_pos;
+}
+
+/**
+ *	relayfs_read_end_pos - return the new read position
+ */
+static size_t relayfs_read_end_pos(struct rchan_buf *buf,
+				   size_t read_pos,
+				   size_t count)
+{
+	size_t read_subbuf, padding, end_pos;
+	size_t subbuf_size = buf->chan->subbuf_size;
+	size_t n_subbufs = buf->chan->n_subbufs;
+
+	read_subbuf = read_pos / subbuf_size;
+	padding = buf->padding[read_subbuf];
+	if (read_pos % subbuf_size + count + padding == subbuf_size)
+		end_pos = (read_subbuf + 1) * subbuf_size;
+	else
+		end_pos = read_pos + count;
+	if (end_pos >= subbuf_size * n_subbufs)
+		end_pos = 0;
+
+	return end_pos;
+}
+
+/**
+ *	relayfs_read - read file op for relayfs files
+ *	@filp: the file
+ *	@buffer: the userspace buffer
+ *	@count: number of bytes to read
+ *	@ppos: position to read from
+ *
+ *	Reads count bytes or the number of bytes available in the
+ *	current sub-buffer being read, whichever is smaller.
+ */
+static ssize_t relayfs_read(struct file *filp,
+			    char __user *buffer,
+			    size_t count,
+			    loff_t *ppos)
+{
+	struct inode *inode = filp->f_dentry->d_inode;
+	struct rchan_buf *buf = RELAYFS_I(inode)->buf;
+	size_t read_start, avail;
+	ssize_t ret = 0;
+	void *from;
+
+	down(&inode->i_sem);
+	if(!relayfs_read_avail(buf, *ppos))
+		goto out;
+
+	read_start = relayfs_read_start_pos(*ppos, buf);
+	avail = relayfs_read_subbuf_avail(read_start, buf);
+	if (!avail)
+		goto out;
+
+	from = buf->start + read_start;
+	ret = count = min(count, avail);
+	if (copy_to_user(buffer, from, count)) {
+		ret = -EFAULT;
+		goto out;
+	}
+	relayfs_read_consume(buf, read_start, count);
+	*ppos = relayfs_read_end_pos(buf, read_start, count);
+out:
+	up(&inode->i_sem);
+	return ret;
+}
+
+/**
+ *	relayfs alloc_inode() implementation
+ */
+static struct inode *relayfs_alloc_inode(struct super_block *sb)
+{
+	struct relayfs_inode_info *p = kmem_cache_alloc(relayfs_inode_cachep, SLAB_KERNEL);
+	if (!p)
+		return NULL;
+	p->buf = NULL;
+
+	return &p->vfs_inode;
+}
+
+/**
+ *	relayfs destroy_inode() implementation
+ */
+static void relayfs_destroy_inode(struct inode *inode)
+{
+	if (RELAYFS_I(inode)->buf)
+		relay_destroy_buf(RELAYFS_I(inode)->buf);
+
+	kmem_cache_free(relayfs_inode_cachep, RELAYFS_I(inode));
+}
+
+static void init_once(void *p, kmem_cache_t *cachep, unsigned long flags)
+{
+	struct relayfs_inode_info *i = p;
+	if ((flags & (SLAB_CTOR_VERIFY | SLAB_CTOR_CONSTRUCTOR)) == SLAB_CTOR_CONSTRUCTOR)
+		inode_init_once(&i->vfs_inode);
+}
+
+struct file_operations relayfs_file_operations = {
+	.open		= relayfs_open,
+	.poll		= relayfs_poll,
+	.mmap		= relayfs_mmap,
+	.read		= relayfs_read,
+	.llseek		= no_llseek,
+	.release	= relayfs_release,
+};
+
+static struct super_operations relayfs_ops = {
+	.statfs		= simple_statfs,
+	.drop_inode	= generic_delete_inode,
+	.alloc_inode	= relayfs_alloc_inode,
+	.destroy_inode	= relayfs_destroy_inode,
+};
+
+static int relayfs_fill_super(struct super_block * sb, void * data, int silent)
+{
+	struct inode *inode;
+	struct dentry *root;
+	int mode = S_IFDIR | S_IRWXU | S_IRUGO | S_IXUGO;
+
+	sb->s_blocksize = PAGE_CACHE_SIZE;
+	sb->s_blocksize_bits = PAGE_CACHE_SHIFT;
+	sb->s_magic = RELAYFS_MAGIC;
+	sb->s_op = &relayfs_ops;
+	inode = relayfs_get_inode(sb, mode, NULL);
+
+	if (!inode)
+		return -ENOMEM;
+
+	root = d_alloc_root(inode);
+	if (!root) {
+		iput(inode);
+		return -ENOMEM;
+	}
+	sb->s_root = root;
+
+	return 0;
+}
+
+static struct super_block * relayfs_get_sb(struct file_system_type *fs_type,
+					   int flags, const char *dev_name,
+					   void *data)
+{
+	return get_sb_single(fs_type, flags, data, relayfs_fill_super);
+}
+
+static struct file_system_type relayfs_fs_type = {
+	.owner		= THIS_MODULE,
+	.name		= "relayfs",
+	.get_sb		= relayfs_get_sb,
+	.kill_sb	= kill_litter_super,
+};
+
+static int __init init_relayfs_fs(void)
+{
+	int err;
+
+	relayfs_inode_cachep = kmem_cache_create("relayfs_inode_cache",
+				sizeof(struct relayfs_inode_info), 0,
+				0, init_once, NULL);
+	if (!relayfs_inode_cachep)
+		return -ENOMEM;
+
+	err = register_filesystem(&relayfs_fs_type);
+	if (err)
+		kmem_cache_destroy(relayfs_inode_cachep);
+
+	return err;
+}
+
+static void __exit exit_relayfs_fs(void)
+{
+	unregister_filesystem(&relayfs_fs_type);
+	kmem_cache_destroy(relayfs_inode_cachep);
+}
+
+module_init(init_relayfs_fs)
+module_exit(exit_relayfs_fs)
+
+EXPORT_SYMBOL_GPL(relayfs_file_operations);
+EXPORT_SYMBOL_GPL(relayfs_create_dir);
+EXPORT_SYMBOL_GPL(relayfs_remove_dir);
+
+MODULE_AUTHOR("Tom Zanussi <zanussi@us.ibm.com> and Karim Yaghmour <karim@opersys.com>");
+MODULE_DESCRIPTION("Relay Filesystem");
+MODULE_LICENSE("GPL");
+
diff --git a/fs/relayfs/relay.c b/fs/relayfs/relay.c
new file mode 100644
index 000000000000..16446a15c96d
--- /dev/null
+++ b/fs/relayfs/relay.c
@@ -0,0 +1,431 @@
+/*
+ * Public API and common code for RelayFS.
+ *
+ * See Documentation/filesystems/relayfs.txt for an overview of relayfs.
+ *
+ * Copyright (C) 2002-2005 - Tom Zanussi (zanussi@us.ibm.com), IBM Corp
+ * Copyright (C) 1999-2005 - Karim Yaghmour (karim@opersys.com)
+ *
+ * This file is released under the GPL.
+ */
+
+#include <linux/errno.h>
+#include <linux/stddef.h>
+#include <linux/slab.h>
+#include <linux/module.h>
+#include <linux/string.h>
+#include <linux/relayfs_fs.h>
+#include "relay.h"
+#include "buffers.h"
+
+/**
+ *	relay_buf_empty - boolean, is the channel buffer empty?
+ *	@buf: channel buffer
+ *
+ *	Returns 1 if the buffer is empty, 0 otherwise.
+ */
+int relay_buf_empty(struct rchan_buf *buf)
+{
+	return (buf->subbufs_produced - buf->subbufs_consumed) ? 0 : 1;
+}
+
+/**
+ *	relay_buf_full - boolean, is the channel buffer full?
+ *	@buf: channel buffer
+ *
+ *	Returns 1 if the buffer is full, 0 otherwise.
+ */
+int relay_buf_full(struct rchan_buf *buf)
+{
+	size_t ready = buf->subbufs_produced - buf->subbufs_consumed;
+	return (ready >= buf->chan->n_subbufs) ? 1 : 0;
+}
+
+/*
+ * High-level relayfs kernel API and associated functions.
+ */
+
+/*
+ * rchan_callback implementations defining default channel behavior.  Used
+ * in place of corresponding NULL values in client callback struct.
+ */
+
+/*
+ * subbuf_start() default callback.  Does nothing.
+ */
+static int subbuf_start_default_callback (struct rchan_buf *buf,
+					  void *subbuf,
+					  void *prev_subbuf,
+					  size_t prev_padding)
+{
+	if (relay_buf_full(buf))
+		return 0;
+
+	return 1;
+}
+
+/*
+ * buf_mapped() default callback.  Does nothing.
+ */
+static void buf_mapped_default_callback(struct rchan_buf *buf,
+					struct file *filp)
+{
+}
+
+/*
+ * buf_unmapped() default callback.  Does nothing.
+ */
+static void buf_unmapped_default_callback(struct rchan_buf *buf,
+					  struct file *filp)
+{
+}
+
+/* relay channel default callbacks */
+static struct rchan_callbacks default_channel_callbacks = {
+	.subbuf_start = subbuf_start_default_callback,
+	.buf_mapped = buf_mapped_default_callback,
+	.buf_unmapped = buf_unmapped_default_callback,
+};
+
+/**
+ *	wakeup_readers - wake up readers waiting on a channel
+ *	@private: the channel buffer
+ *
+ *	This is the work function used to defer reader waking.  The
+ *	reason waking is deferred is that calling directly from write
+ *	causes problems if you're writing from say the scheduler.
+ */
+static void wakeup_readers(void *private)
+{
+	struct rchan_buf *buf = private;
+	wake_up_interruptible(&buf->read_wait);
+}
+
+/**
+ *	__relay_reset - reset a channel buffer
+ *	@buf: the channel buffer
+ *	@init: 1 if this is a first-time initialization
+ *
+ *	See relay_reset for description of effect.
+ */
+static inline void __relay_reset(struct rchan_buf *buf, unsigned int init)
+{
+	size_t i;
+
+	if (init) {
+		init_waitqueue_head(&buf->read_wait);
+		kref_init(&buf->kref);
+		INIT_WORK(&buf->wake_readers, NULL, NULL);
+	} else {
+		cancel_delayed_work(&buf->wake_readers);
+		flush_scheduled_work();
+	}
+
+	buf->subbufs_produced = 0;
+	buf->subbufs_consumed = 0;
+	buf->bytes_consumed = 0;
+	buf->finalized = 0;
+	buf->data = buf->start;
+	buf->offset = 0;
+
+	for (i = 0; i < buf->chan->n_subbufs; i++)
+		buf->padding[i] = 0;
+
+	buf->chan->cb->subbuf_start(buf, buf->data, NULL, 0);
+}
+
+/**
+ *	relay_reset - reset the channel
+ *	@chan: the channel
+ *
+ *	This has the effect of erasing all data from all channel buffers
+ *	and restarting the channel in its initial state.  The buffers
+ *	are not freed, so any mappings are still in effect.
+ *
+ *	NOTE: Care should be taken that the channel isn't actually
+ *	being used by anything when this call is made.
+ */
+void relay_reset(struct rchan *chan)
+{
+	unsigned int i;
+
+	if (!chan)
+		return;
+
+	for (i = 0; i < NR_CPUS; i++) {
+		if (!chan->buf[i])
+			continue;
+		__relay_reset(chan->buf[i], 0);
+	}
+}
+
+/**
+ *	relay_open_buf - create a new channel buffer in relayfs
+ *
+ *	Internal - used by relay_open().
+ */
+static struct rchan_buf *relay_open_buf(struct rchan *chan,
+					const char *filename,
+					struct dentry *parent)
+{
+	struct rchan_buf *buf;
+	struct dentry *dentry;
+
+	/* Create file in fs */
+	dentry = relayfs_create_file(filename, parent, S_IRUSR, chan);
+	if (!dentry)
+		return NULL;
+
+	buf = RELAYFS_I(dentry->d_inode)->buf;
+	buf->dentry = dentry;
+	__relay_reset(buf, 1);
+
+	return buf;
+}
+
+/**
+ *	relay_close_buf - close a channel buffer
+ *	@buf: channel buffer
+ *
+ *	Marks the buffer finalized and restores the default callbacks.
+ *	The channel buffer and channel buffer data structure are then freed
+ *	automatically when the last reference is given up.
+ */
+static inline void relay_close_buf(struct rchan_buf *buf)
+{
+	buf->finalized = 1;
+	buf->chan->cb = &default_channel_callbacks;
+	cancel_delayed_work(&buf->wake_readers);
+	flush_scheduled_work();
+	kref_put(&buf->kref, relay_remove_buf);
+}
+
+static inline void setup_callbacks(struct rchan *chan,
+				   struct rchan_callbacks *cb)
+{
+	if (!cb) {
+		chan->cb = &default_channel_callbacks;
+		return;
+	}
+
+	if (!cb->subbuf_start)
+		cb->subbuf_start = subbuf_start_default_callback;
+	if (!cb->buf_mapped)
+		cb->buf_mapped = buf_mapped_default_callback;
+	if (!cb->buf_unmapped)
+		cb->buf_unmapped = buf_unmapped_default_callback;
+	chan->cb = cb;
+}
+
+/**
+ *	relay_open - create a new relayfs channel
+ *	@base_filename: base name of files to create
+ *	@parent: dentry of parent directory, NULL for root directory
+ *	@subbuf_size: size of sub-buffers
+ *	@n_subbufs: number of sub-buffers
+ *	@cb: client callback functions
+ *
+ *	Returns channel pointer if successful, NULL otherwise.
+ *
+ *	Creates a channel buffer for each cpu using the sizes and
+ *	attributes specified.  The created channel buffer files
+ *	will be named base_filename0...base_filenameN-1.  File
+ *	permissions will be S_IRUSR.
+ */
+struct rchan *relay_open(const char *base_filename,
+			 struct dentry *parent,
+			 size_t subbuf_size,
+			 size_t n_subbufs,
+			 struct rchan_callbacks *cb)
+{
+	unsigned int i;
+	struct rchan *chan;
+	char *tmpname;
+
+	if (!base_filename)
+		return NULL;
+
+	if (!(subbuf_size && n_subbufs))
+		return NULL;
+
+	chan = kcalloc(1, sizeof(struct rchan), GFP_KERNEL);
+	if (!chan)
+		return NULL;
+
+	chan->version = RELAYFS_CHANNEL_VERSION;
+	chan->n_subbufs = n_subbufs;
+	chan->subbuf_size = subbuf_size;
+	chan->alloc_size = FIX_SIZE(subbuf_size * n_subbufs);
+	setup_callbacks(chan, cb);
+	kref_init(&chan->kref);
+
+	tmpname = kmalloc(NAME_MAX + 1, GFP_KERNEL);
+	if (!tmpname)
+		goto free_chan;
+
+	for_each_online_cpu(i) {
+		sprintf(tmpname, "%s%d", base_filename, i);
+		chan->buf[i] = relay_open_buf(chan, tmpname, parent);
+		chan->buf[i]->cpu = i;
+		if (!chan->buf[i])
+			goto free_bufs;
+	}
+
+	kfree(tmpname);
+	return chan;
+
+free_bufs:
+	for (i = 0; i < NR_CPUS; i++) {
+		if (!chan->buf[i])
+			break;
+		relay_close_buf(chan->buf[i]);
+	}
+	kfree(tmpname);
+
+free_chan:
+	kref_put(&chan->kref, relay_destroy_channel);
+	return NULL;
+}
+
+/**
+ *	relay_switch_subbuf - switch to a new sub-buffer
+ *	@buf: channel buffer
+ *	@length: size of current event
+ *
+ *	Returns either the length passed in or 0 if full.
+
+ *	Performs sub-buffer-switch tasks such as invoking callbacks,
+ *	updating padding counts, waking up readers, etc.
+ */
+size_t relay_switch_subbuf(struct rchan_buf *buf, size_t length)
+{
+	void *old, *new;
+	size_t old_subbuf, new_subbuf;
+
+	if (unlikely(length > buf->chan->subbuf_size))
+		goto toobig;
+
+	if (buf->offset != buf->chan->subbuf_size + 1) {
+		buf->prev_padding = buf->chan->subbuf_size - buf->offset;
+		old_subbuf = buf->subbufs_produced % buf->chan->n_subbufs;
+		buf->padding[old_subbuf] = buf->prev_padding;
+		buf->subbufs_produced++;
+		if (waitqueue_active(&buf->read_wait)) {
+			PREPARE_WORK(&buf->wake_readers, wakeup_readers, buf);
+			schedule_delayed_work(&buf->wake_readers, 1);
+		}
+	}
+
+	old = buf->data;
+	new_subbuf = buf->subbufs_produced % buf->chan->n_subbufs;
+	new = buf->start + new_subbuf * buf->chan->subbuf_size;
+	buf->offset = 0;
+	if (!buf->chan->cb->subbuf_start(buf, new, old, buf->prev_padding)) {
+		buf->offset = buf->chan->subbuf_size + 1;
+		return 0;
+	}
+	buf->data = new;
+	buf->padding[new_subbuf] = 0;
+
+	if (unlikely(length + buf->offset > buf->chan->subbuf_size))
+		goto toobig;
+
+	return length;
+
+toobig:
+	printk(KERN_WARNING "relayfs: event too large (%Zd)\n", length);
+	WARN_ON(1);
+	return 0;
+}
+
+/**
+ *	relay_subbufs_consumed - update the buffer's sub-buffers-consumed count
+ *	@chan: the channel
+ *	@cpu: the cpu associated with the channel buffer to update
+ *	@subbufs_consumed: number of sub-buffers to add to current buf's count
+ *
+ *	Adds to the channel buffer's consumed sub-buffer count.
+ *	subbufs_consumed should be the number of sub-buffers newly consumed,
+ *	not the total consumed.
+ *
+ *	NOTE: kernel clients don't need to call this function if the channel
+ *	mode is 'overwrite'.
+ */
+void relay_subbufs_consumed(struct rchan *chan,
+			    unsigned int cpu,
+			    size_t subbufs_consumed)
+{
+	struct rchan_buf *buf;
+
+	if (!chan)
+		return;
+
+	if (cpu >= NR_CPUS || !chan->buf[cpu])
+		return;
+
+	buf = chan->buf[cpu];
+	buf->subbufs_consumed += subbufs_consumed;
+	if (buf->subbufs_consumed > buf->subbufs_produced)
+		buf->subbufs_consumed = buf->subbufs_produced;
+}
+
+/**
+ *	relay_destroy_channel - free the channel struct
+ *
+ *	Should only be called from kref_put().
+ */
+void relay_destroy_channel(struct kref *kref)
+{
+	struct rchan *chan = container_of(kref, struct rchan, kref);
+	kfree(chan);
+}
+
+/**
+ *	relay_close - close the channel
+ *	@chan: the channel
+ *
+ *	Closes all channel buffers and frees the channel.
+ */
+void relay_close(struct rchan *chan)
+{
+	unsigned int i;
+
+	if (!chan)
+		return;
+
+	for (i = 0; i < NR_CPUS; i++) {
+		if (!chan->buf[i])
+			continue;
+		relay_close_buf(chan->buf[i]);
+	}
+
+	kref_put(&chan->kref, relay_destroy_channel);
+}
+
+/**
+ *	relay_flush - close the channel
+ *	@chan: the channel
+ *
+ *	Flushes all channel buffers i.e. forces buffer switch.
+ */
+void relay_flush(struct rchan *chan)
+{
+	unsigned int i;
+
+	if (!chan)
+		return;
+
+	for (i = 0; i < NR_CPUS; i++) {
+		if (!chan->buf[i])
+			continue;
+		relay_switch_subbuf(chan->buf[i], 0);
+	}
+}
+
+EXPORT_SYMBOL_GPL(relay_open);
+EXPORT_SYMBOL_GPL(relay_close);
+EXPORT_SYMBOL_GPL(relay_flush);
+EXPORT_SYMBOL_GPL(relay_reset);
+EXPORT_SYMBOL_GPL(relay_subbufs_consumed);
+EXPORT_SYMBOL_GPL(relay_switch_subbuf);
+EXPORT_SYMBOL_GPL(relay_buf_full);
diff --git a/fs/relayfs/relay.h b/fs/relayfs/relay.h
new file mode 100644
index 000000000000..703503fa22b6
--- /dev/null
+++ b/fs/relayfs/relay.h
@@ -0,0 +1,12 @@
+#ifndef _RELAY_H
+#define _RELAY_H
+
+struct dentry *relayfs_create_file(const char *name,
+				   struct dentry *parent,
+				   int mode,
+				   struct rchan *chan);
+extern int relayfs_remove(struct dentry *dentry);
+extern int relay_buf_empty(struct rchan_buf *buf);
+extern void relay_destroy_channel(struct kref *kref);
+
+#endif /* _RELAY_H */