path: root/fs/jfs/namei.c

/*
 *   Copyright (C) International Business Machines Corp., 2000-2004
 *   Portions Copyright (C) Christoph Hellwig, 2001-2002
 *
 *   This program is free software;  you can redistribute it and/or modify
 *   it under the terms of the GNU General Public License as published by
 *   the Free Software Foundation; either version 2 of the License, or 
 *   (at your option) any later version.
 * 
 *   This program is distributed in the hope that it will be useful,
 *   but WITHOUT ANY WARRANTY;  without even the implied warranty of
 *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See
 *   the GNU General Public License for more details.
 *
 *   You should have received a copy of the GNU General Public License
 *   along with this program;  if not, write to the Free Software 
 *   Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
 */

#include <linux/fs.h>
#include <linux/ctype.h>
#include <linux/quotaops.h>
#include "jfs_incore.h"
#include "jfs_superblock.h"
#include "jfs_inode.h"
#include "jfs_dinode.h"
#include "jfs_dmap.h"
#include "jfs_unicode.h"
#include "jfs_metapage.h"
#include "jfs_xattr.h"
#include "jfs_acl.h"
#include "jfs_debug.h"

/*
 * forward references
 */
struct dentry_operations jfs_ci_dentry_operations;

static s64 commitZeroLink(tid_t, struct inode *);

/*
 * NAME:	free_ea_wmap(inode)
 *
 * FUNCTION:	free uncommitted extended attributes from working map 
 *
 */
static inline void free_ea_wmap(struct inode *inode)
{
	dxd_t *ea = &JFS_IP(inode)->ea;

	if (ea->flag & DXD_EXTENT) {
		/* free EA pages from cache */
		invalidate_dxd_metapages(inode, *ea);
		dbFree(inode, addressDXD(ea), lengthDXD(ea));
	}
	ea->flag = 0;
}

/*
 * NAME:	jfs_create(dip, dentry, mode)
 *
 * FUNCTION:	create a regular file in the parent directory <dip>
 *		with name = <from dentry> and mode = <mode>
 *
 * PARAMETER:	dip 	- parent directory vnode
 *		dentry	- dentry of new file
 *		mode	- create mode (rwxrwxrwx).
 *		nd- nd struct
 *
 * RETURN:	Errors from subroutines
 *
 */
static int jfs_create(struct inode *dip, struct dentry *dentry, int mode,
		struct nameidata *nd)
{
	int rc = 0;
	tid_t tid;		/* transaction id */
	struct inode *ip = NULL;	/* child directory inode */
	ino_t ino;
	struct component_name dname;	/* child directory name */
	struct btstack btstack;
	struct inode *iplist[2];
	struct tblock *tblk;

	jfs_info("jfs_create: dip:0x%p name:%s", dip, dentry->d_name.name);

	/*
	 * search parent directory for entry/freespace
	 * (dtSearch() returns parent directory page pinned)
	 */
	if ((rc = get_UCSname(&dname, dentry)))
		goto out1;

	/*
	 * Either iAlloc() or txBegin() may block.  Deadlock can occur if we
	 * block there while holding dtree page, so we allocate the inode &
	 * begin the transaction before we search the directory.
	 */
	ip = ialloc(dip, mode);
	if (ip == NULL) {
		rc = -ENOSPC;
		goto out2;
	}

	tid = txBegin(dip->i_sb, 0);

	mutex_lock(&JFS_IP(dip)->commit_mutex);
	mutex_lock(&JFS_IP(ip)->commit_mutex);

	rc = jfs_init_acl(tid, ip, dip);
	if (rc)
		goto out3;

	rc = jfs_init_security(tid, ip, dip);
	if (rc) {
		txAbort(tid, 0);
		goto out3;
	}

	if ((rc = dtSearch(dip, &dname, &ino, &btstack, JFS_CREATE))) {
		jfs_err("jfs_create: dtSearch returned %d", rc);
		txAbort(tid, 0);
		goto out3;
	}

	tblk = tid_to_tblock(tid);
	tblk->xflag |= COMMIT_CREATE;
	tblk->ino = ip->i_ino;
	tblk->u.ixpxd = JFS_IP(ip)->ixpxd;

	iplist[0] = dip;
	iplist[1] = ip;

	/*
	 * initialize the child XAD tree root in-line in inode
	 */
	xtInitRoot(tid, ip);

	/*
	 * create entry in parent directory for child directory
	 * (dtInsert() releases parent directory page)
	 */
	ino = ip->i_ino;
	if ((rc = dtInsert(tid, dip, &dname, &ino, &btstack))) {
		if (rc == -EIO) {
			jfs_err("jfs_create: dtInsert returned -EIO");
			txAbort(tid, 1);	/* Marks Filesystem dirty */
		} else
			txAbort(tid, 0);	/* Filesystem full */
		goto out3;
	}

	ip->i_op = &jfs_file_inode_operations;
	ip->i_fop = &jfs_file_operations;
	ip->i_mapping->a_ops = &jfs_aops;

	insert_inode_hash(ip);
	mark_inode_dirty(ip);

	dip->i_ctime = dip->i_mtime = CURRENT_TIME;

	mark_inode_dirty(dip);

	rc = txCommit(tid, 2, &iplist[0], 0);

      out3:
	txEnd(tid);
	mutex_unlock(&JFS_IP(dip)->commit_mutex);
	mutex_unlock(&JFS_IP(ip)->commit_mutex);
	if (rc) {
		free_ea_wmap(ip);
		ip->i_nlink = 0;
		iput(ip);
	} else
		d_instantiate(dentry, ip);

      out2:
	free_UCSname(&dname);

      out1:

	jfs_info("jfs_create: rc:%d", rc);
	return rc;
}


/*
 * NAME:	jfs_mkdir(dip, dentry, mode)
 *
 * FUNCTION:	create a child directory in the parent directory <dip>
 *		with name = <from dentry> and mode = <mode>
 *
 * PARAMETER:	dip 	- parent directory vnode
 *		dentry	- dentry of child directory
 *		mode	- create mode (rwxrwxrwx).
 *
 * RETURN:	Errors from subroutines
 *
 * note:
 * EACCESS: user needs search+write permission on the parent directory
 */
static int jfs_mkdir(struct inode *dip, struct dentry *dentry, int mode)
{
	int rc = 0;
	tid_t tid;		/* transaction id */
	struct inode *ip = NULL;	/* child directory inode */
	ino_t ino;
	struct component_name dname;	/* child directory name */
	struct btstack btstack;
	struct inode *iplist[2];
	struct tblock *tblk;

	jfs_info("jfs_mkdir: dip:0x%p name:%s", dip, dentry->d_name.name);

	/* link count overflow on parent directory ? */
	if (dip->i_nlink == JFS_LINK_MAX) {
		rc = -EMLINK;
		goto out1;
	}

	/*
	 * search parent directory for entry/freespace
	 * (dtSearch() returns parent directory page pinned)
	 */
	if ((rc = get_UCSname(&dname, dentry)))
		goto out1;

	/*
	 * Either iAlloc() or txBegin() may block.  Deadlock can occur if we
	 * block there while holding dtree page, so we allocate the inode &
	 * begin the transaction before we search the directory.
	 */
	ip = ialloc(dip, S_IFDIR | mode);
	if (ip == NULL) {
		rc = -ENOSPC;
		goto out2;
	}

	tid = txBegin(dip->i_sb, 0);

	mutex_lock(&JFS_IP(dip)->commit_mutex);
	mutex_lock(&JFS_IP(ip)->commit_mutex);

	rc = jfs_init_acl(tid, ip, dip);
	if (rc)
		goto out3;

	rc = jfs_init_security(tid, ip, dip);
	if (rc) {
		txAbort(tid, 0);
		goto out3;
	}

	if ((rc = dtSearch(dip, &dname, &ino, &btstack, JFS_CREATE))) {
		jfs_err("jfs_mkdir: dtSearch returned %d", rc);
		txAbort(tid, 0);
		goto out3;
	}

	tblk = tid_to_tblock(tid);
	tblk->xflag |= COMMIT_CREATE;
	tblk->ino = ip->i_ino;
	tblk->u.ixpxd = JFS_IP(ip)->ixpxd;

	iplist[0] = dip;
	iplist[1] = ip;

	/*
	 * initialize the child directory in-line in inode
	 */
	dtInitRoot(tid, ip, dip->i_ino);

	/*
	 * create entry in parent directory for child directory
	 * (dtInsert() releases parent directory page)
	 */
	ino = ip->i_ino;
	if ((rc = dtInsert(tid, dip, &dname, &ino, &btstack))) {
		if (rc == -EIO) {
			jfs_err("jfs_mkdir: dtInsert returned -EIO");
			txAbort(tid, 1);	/* Marks Filesystem dirty */
		} else
			txAbort(tid, 0);	/* Filesystem full */
		goto out3;
	}

	ip->i_nlink = 2;	/* for '.' */
	ip->i_op = &jfs_dir_inode_operations;
	ip->i_fop = &jfs_dir_operations;

	insert_inode_hash(ip);
	mark_inode_dirty(ip);

	/* update parent directory inode */
	dip->i_nlink++;		/* for '..' from child directory */
	dip->i_ctime = dip->i_mtime = CURRENT_TIME;
	mark_inode_dirty(dip);

	rc = txCommit(tid, 2, &iplist[0], 0);

      out3:
	txEnd(tid);
	mutex_unlock(&JFS_IP(dip)->commit_mutex);
	mutex_unlock(&JFS_IP(ip)->commit_mutex);
	if (rc) {
		free_ea_wmap(ip);
		ip->i_nlink = 0;
		iput(ip);
	} else
		d_instantiate(dentry, ip);

      out2:
	free_UCSname(&dname);


      out1:

	jfs_info("jfs_mkdir: rc:%d", rc);
	return rc;
}

/*
 * NAME:	jfs_rmdir(dip, dentry)
 *
 * FUNCTION:	remove a link to child directory
 *
 * PARAMETER:	dip 	- parent inode
 *		dentry	- child directory dentry
 *
 * RETURN:	-EINVAL	- if name is . or ..
 *		-EINVAL  - if . or .. exist but are invalid.
 *		errors from subroutines
 *
 * note:
 * if other threads have the directory open when the last link 
 * is removed, the "." and ".." entries, if present, are removed before 
 * rmdir() returns and no new entries may be created in the directory, 
 * but the directory is not removed until the last reference to 
 * the directory is released (cf.unlink() of regular file).
 */
static int jfs_rmdir(struct inode *dip, struct dentry *dentry)
{
	int rc;
	tid_t tid;		/* transaction id */
	struct inode *ip = dentry->d_inode;
	ino_t ino;
	struct component_name dname;
	struct inode *iplist[2];
	struct tblock *tblk;

	jfs_info("jfs_rmdir: dip:0x%p name:%s", dip, dentry->d_name.name);

	/* Init inode for quota operations. */
	DQUOT_INIT(ip);

	/* directory must be empty to be removed */
	if (!dtEmpty(ip)) {
		rc = -ENOTEMPTY;
		goto out;
	}

	if ((rc = get_UCSname(&dname, dentry))) {
		goto out;
	}

	tid = txBegin(dip->i_sb, 0);

	mutex_lock(&JFS_IP(dip)->commit_mutex);
	mutex_lock(&JFS_IP(ip)->commit_mutex);

	iplist[0] = dip;
	iplist[1] = ip;

	tblk = tid_to_tblock(tid);
	tblk->xflag |= COMMIT_DELETE;
	tblk->u.ip = ip;

	/*
	 * delete the entry of target directory from parent directory
	 */
	ino = ip->i_ino;
	if ((rc = dtDelete(tid, dip, &dname, &ino, JFS_REMOVE))) {
		jfs_err("jfs_rmdir: dtDelete returned %d", rc);
		if (rc == -EIO)
			txAbort(tid, 1);
		txEnd(tid);
		mutex_unlock(&JFS_IP(dip)->commit_mutex);
		mutex_unlock(&JFS_IP(ip)->commit_mutex);

		goto out2;
	}

	/* update parent directory's link count corresponding
	 * to ".." entry of the target directory deleted
	 */
	dip->i_nlink--;
	dip->i_ctime = dip->i_mtime = CURRENT_TIME;
	mark_inode_dirty(dip);

	/*
	 * OS/2 could have created EA and/or ACL
	 */
	/* free EA from both persistent and working map */
	if (JFS_IP(ip)->ea.flag & DXD_EXTENT) {
		/* free EA pages */
		txEA(tid, ip, &JFS_IP(ip)->ea, NULL);
	}
	JFS_IP(ip)->ea.flag = 0;

	/* free ACL from both persistent and working map */
	if (JFS_IP(ip)->acl.flag & DXD_EXTENT) {
		/* free ACL pages */
		txEA(tid, ip, &JFS_IP(ip)->acl, NULL);
	}
	JFS_IP(ip)->acl.flag = 0;

	/* mark the target directory as deleted */
	ip->i_nlink = 0;
	mark_inode_dirty(ip);

	rc = txCommit(tid, 2, &iplist[0], 0);

	txEnd(tid);

	mutex_unlock(&JFS_IP(dip)->commit_mutex);
	mutex_unlock(&JFS_IP(ip)->commit_mutex);

	/*
	 * Truncating the directory index table is not guaranteed.  It
	 * may need to be done iteratively
	 */
	if (test_cflag(COMMIT_Stale, dip)) {
		if (dip->i_size > 1)
			jfs_truncate_nolock(dip, 0);

		clear_cflag(COMMIT_Stale, dip);
	}

      out2:
	free_UCSname(&dname);

      out:
	jfs_info("jfs_rmdir: rc:%d", rc);
	return rc;
}

/*
 * NAME:	jfs_unlink(dip, dentry)
 *
 * FUNCTION:	remove a link to object <vp> named by <name> 
 *		from parent directory <dvp>
 *
 * PARAMETER:	dip 	- inode of parent directory
 *		dentry 	- dentry of object to be removed
 *
 * RETURN:	errors from subroutines
 *
 * note:
 * temporary file: if one or more processes have the file open
 * when the last link is removed, the link will be removed before
 * unlink() returns, but the removal of the file contents will be
 * postponed until all references to the files are closed.
 *
 * JFS does NOT support unlink() on directories.
 *
 */
static int jfs_unlink(struct inode *dip, struct dentry *dentry)
{
	int rc;
	tid_t tid;		/* transaction id */
	struct inode *ip = dentry->d_inode;
	ino_t ino;
	struct component_name dname;	/* object name */
	struct inode *iplist[2];
	struct tblock *tblk;
	s64 new_size = 0;
	int commit_flag;

	jfs_info("jfs_unlink: dip:0x%p name:%s", dip, dentry->d_name.name);

	/* Init inode for quota operations. */
	DQUOT_INIT(ip);

	if ((rc = get_UCSname(&dname, dentry)))
		goto out;

	IWRITE_LOCK(ip);

	tid = txBegin(dip->i_sb, 0);

	mutex_lock(&JFS_IP(dip)->commit_mutex);
	mutex_lock(&JFS_IP(ip)->commit_mutex);

	iplist[0] = dip;
	iplist[1] = ip;

	/*
	 * delete the entry of target file from parent directory
	 */
	ino = ip->i_ino;
	if ((rc = dtDelete(tid, dip, &dname, &ino, JFS_REMOVE))) {
		jfs_err("jfs_unlink: dtDelete returned %d", rc);
		if (rc == -EIO)
			txAbort(tid, 1);	/* Marks FS Dirty */
		txEnd(tid);
		mutex_unlock(&JFS_IP(dip)->commit_mutex);
		mutex_unlock(&JFS_IP(ip)->commit_mutex);
		IWRITE_UNLOCK(ip);
		goto out1;
	}

	ASSERT(ip->i_nlink);

	ip->i_ctime = dip->i_ctime = dip->i_mtime = CURRENT_TIME;
	mark_inode_dirty(dip);

	/* update target's inode */
	ip->i_nlink--;
	mark_inode_dirty(ip);

	/*
	 *      commit zero link count object
	 */
	if (ip->i_nlink == 0) {
		assert(!test_cflag(COMMIT_Nolink, ip));
		/* free block resources */
		if ((new_size = commitZeroLink(tid, ip)) < 0) {
			txAbort(tid, 1);	/* Marks FS Dirty */
			txEnd(tid);
			mutex_unlock(&JFS_IP(dip)->commit_mutex);
			mutex_unlock(&JFS_IP(ip)->commit_mutex);
			IWRITE_UNLOCK(ip);
			rc = new_size;
			goto out1;
		}
		tblk = tid_to_tblock(tid);
		tblk->xflag |= COMMIT_DELETE;
		tblk->u.ip = ip;
	}

	/*
	 * Incomplete truncate of file data can
	 * result in timing problems unless we synchronously commit the
	 * transaction.
	 */
	if (new_size)
		commit_flag = COMMIT_SYNC;
	else
		commit_flag = 0;

	/*
	 * If xtTruncate was incomplete, commit synchronously to avoid
	 * timing complications
	 */
	rc = txCommit(tid, 2, &iplist[0], commit_flag);

	txEnd(tid);

	mutex_unlock(&JFS_IP(dip)->commit_mutex);
	mutex_unlock(&JFS_IP(ip)->commit_mutex);


	while (new_size && (rc == 0)) {
		tid = txBegin(dip->i_sb, 0);
		mutex_lock(&JFS_IP(ip)->commit_mutex);
		new_size = xtTruncate_pmap(tid, ip, new_size);
		if (new_size < 0) {
			txAbort(tid, 1);	/* Marks FS Dirty */
			rc = new_size;
		} else
			rc = txCommit(tid, 2, &iplist[0], COMMIT_SYNC);
		txEnd(tid);
		mutex_unlock(&JFS_IP(ip)->commit_mutex);
	}

	if (ip->i_nlink == 0)
		set_cflag(COMMIT_Nolink, ip);

	IWRITE_UNLOCK(ip);

	/*
	 * Truncating the directory index table is not guaranteed.  It
	 * may need to be done iteratively
	 */
	if (test_cflag(COMMIT_Stale, dip)) {
		if (dip->i_size > 1)
			jfs_truncate_nolock(dip, 0);

		clear_cflag(COMMIT_Stale, dip);
	}

      out1:
	free_UCSname(&dname);
      out:
	jfs_info("jfs_unlink: rc:%d", rc);
	return rc;
}

/*
 * NAME:	commitZeroLink()
 *
 * FUNCTION:    for non-directory, called by jfs_remove(),
 *		truncate a regular file, directory or symbolic
 *		link to zero length. return 0 if type is not 
 *		one of these.
 *
 *		if the file is currently associated with a VM segment
 *		only permanent disk and inode map resources are freed,
 *		and neither the inode nor indirect blocks are modified
 *		so that the resources can be later freed in the work
 *		map by ctrunc1.
 *		if there is no VM segment on entry, the resources are
 *		freed in both work and permanent map.
 *		(? for temporary file - memory object is cached even 
 *		after no reference:
 *		reference count > 0 -   )
 *
 * PARAMETERS:	cd	- pointer to commit data structure.
 *			  current inode is the one to truncate.
 *
 * RETURN:	Errors from subroutines
 */
static s64 commitZeroLink(tid_t tid, struct inode *ip)
{
	int filetype;
	struct tblock *tblk;

	jfs_info("commitZeroLink: tid = %d, ip = 0x%p", tid, ip);

	filetype = ip->i_mode & S_IFMT;
	switch (filetype) {
	case S_IFREG:
		break;
	case S_IFLNK:
		/* fast symbolic link */
		if (ip->i_size < IDATASIZE) {
			ip->i_size = 0;
			return 0;
		}
		break;
	default:
		assert(filetype != S_IFDIR);
		return 0;
	}

	set_cflag(COMMIT_Freewmap, ip);

	/* mark transaction of block map update type */
	tblk = tid_to_tblock(tid);
	tblk->xflag |= COMMIT_PMAP;

	/*
	 * free EA
	 */
	if (JFS_IP(ip)->ea.flag & DXD_EXTENT)
		/* acquire maplock on EA to be freed from block map */
		txEA(tid, ip, &JFS_IP(ip)->ea, NULL);

	/*
	 * free ACL
	 */
	if (JFS_IP(ip)->acl.flag & DXD_EXTENT)
		/* acquire maplock on EA to be freed from block map */
		txEA(tid, ip, &JFS_IP(ip)->acl, NULL);

	/*
	 * free xtree/data (truncate to zero length):
	 * free xtree/data pages from cache if COMMIT_PWMAP, 
	 * free xtree/data blocks from persistent block map, and
	 * free xtree/data blocks from working block map if COMMIT_PWMAP;
	 */
	if (ip->i_size)
		return xtTruncate_pmap(tid, ip, 0);

	return 0;
}


/*
 * NAME:	jfs_free_zero_link()
 *
 * FUNCTION:    for non-directory, called by iClose(),
 *		free resources of a file from cache and WORKING map 
 *		for a file previously committed with zero link count
 *		while associated with a pager object,
 *
 * PARAMETER:	ip	- pointer to inode of file.
 */
void jfs_free_zero_link(struct inode *ip)
{
	int type;

	jfs_info("jfs_free_zero_link: ip = 0x%p", ip);

	/* return if not reg or symbolic link or if size is
	 * already ok.
	 */
	type = ip->i_mode & S_IFMT;

	switch (type) {
	case S_IFREG:
		break;
	case S_IFLNK:
		/* if its contained in inode nothing to do */
		if (ip->i_size < IDATASIZE)
			return;
		break;
	default:
		return;
	}

	/*
	 * free EA
	 */
	if (JFS_IP(ip)->ea.flag & DXD_EXTENT) {
		s64 xaddr = addressDXD(&JFS_IP(ip)->ea);
		int xlen = lengthDXD(&JFS_IP(ip)->ea);
		struct maplock maplock;	/* maplock for COMMIT_WMAP */
		struct pxd_lock *pxdlock;	/* maplock for COMMIT_WMAP */

		/* free EA pages from cache */
		invalidate_dxd_metapages(ip, JFS_IP(ip)->ea);

		/* free EA extent from working block map */
		maplock.index = 1;
		pxdlock = (struct pxd_lock *) & maplock;
		pxdlock->flag = mlckFREEPXD;
		PXDaddress(&pxdlock->pxd, xaddr);
		PXDlength(&pxdlock->pxd, xlen);
		txFreeMap(ip, pxdlock, NULL, COMMIT_WMAP);
	}

	/*
	 * free ACL
	 */
	if (JFS_IP(ip)->acl.flag & DXD_EXTENT) {
		s64 xaddr = addressDXD(&JFS_IP(ip)->acl);
		int xlen = lengthDXD(&JFS_IP(ip)->acl);
		struct maplock maplock;	/* maplock for COMMIT_WMAP */
		struct pxd_lock *pxdlock;	/* maplock for COMMIT_WMAP */

		invalidate_dxd_metapages(ip, JFS_IP(ip)->acl);

		/* free ACL extent from working block map */
		maplock.index = 1;
		pxdlock = (struct pxd_lock *) & maplock;
		pxdlock->flag = mlckFREEPXD;
		PXDaddress(&pxdlock->pxd, xaddr);
		PXDlength(&pxdlock->pxd, xlen);
		txFreeMap(ip, pxdlock, NULL, COMMIT_WMAP);
	}

	/*
	 * free xtree/data (truncate to zero length):
	 * free xtree/data pages from cache, and
	 * free xtree/data blocks from working block map;
	 */
	if (ip->i_size)
		xtTruncate(0, ip, 0, COMMIT_WMAP);
}

/*
 * NAME:	jfs_link(vp, dvp, name, crp)
 *
 * FUNCTION:	create a link to <vp> by the name = <name>
 *		in the parent directory <dvp>
 *
 * PARAMETER:	vp 	- target object
 *		dvp	- parent directory of new link
 *		name	- name of new link to target object
 *		crp	- credential
 *
 * RETURN:	Errors from subroutines
 *
 * note:
 * JFS does NOT support link() on directories (to prevent circular
 * path in the directory hierarchy);
 * EPERM: the target object is a directory, and either the caller
 * does not have appropriate privileges or the implementation prohibits
 * using link() on directories [XPG4.2].
 *
 * JFS does NOT support links between file systems:
 * EXDEV: target object and new link are on different file systems and
 * implementation does not support links between file systems [XPG4.2].
 */
static int jfs_link(struct dentry *old_dentry,
	     struct inode *dir, struct dentry *dentry)
{
	int rc;
	tid_t tid;
	struct inode *ip = old_dentry->d_inode;
	ino_t ino;
	struct component_name dname;
	struct btstack btstack;
	struct inode *iplist[2];

	jfs_info("jfs_link: %s %s", old_dentry->d_name.name,
		 dentry->d_name.name);

	if (ip->i_nlink == JFS_LINK_MAX)
		return -EMLINK;

	if (ip->i_nlink == 0)
		return -ENOENT;

	tid = txBegin(ip->i_sb, 0);

	mutex_lock(&JFS_IP(dir)->commit_mutex);
	mutex_lock(&JFS_IP(ip)->commit_mutex);

	/*
	 * scan parent directory for entry/freespace
	 */
	if ((rc = get_UCSname(&dname, dentry)))
		goto out;

	if ((rc = dtSearch(dir, &dname, &ino, &btstack, JFS_CREATE)))
		goto free_dname;

	/*
	 * create entry for new link in parent directory
	 */
	ino = ip->i_ino;
	if ((rc = dtInsert(tid, dir, &dname, &ino, &btstack)))
		goto free_dname;

	/* update object inode */
	ip->i_nlink++;		/* for new link */
	ip->i_ctime = CURRENT_TIME;
	dir->i_ctime = dir->i_mtime = CURRENT_TIME;
	mark_inode_dirty(dir);
	atomic_inc(&ip->i_count);

	iplist[0] = ip;
	iplist[1] = dir;
	rc = txCommit(tid, 2, &iplist[0], 0);

	if (rc) {
		ip->i_nlink--;
		iput(ip);
	} else
		d_instantiate(dentry, ip);

      free_dname:
	free_UCSname(&dname);

      out:
	txEnd(tid);

	mutex_unlock(&JFS_IP(dir)->commit_mutex);
	mutex_unlock(&JFS_IP(ip)->commit_mutex);

	jfs_info("jfs_link: rc:%d", rc);
	return rc;
}

/*
 * NAME:	jfs_symlink(dip, dentry, name)
 *
 * FUNCTION:	creates a symbolic link to <symlink> by name <name>
 *		        in directory <dip>
 *
 * PARAMETER:	dip	    - parent directory vnode
 *		        dentry 	- dentry of symbolic link
 *		        name    - the path name of the existing object 
 *			              that will be the source of the link
 *
 * RETURN:	errors from subroutines
 *
 * note:
 * ENAMETOOLONG: pathname resolution of a symbolic link produced
 * an intermediate result whose length exceeds PATH_MAX [XPG4.2]
*/

static int jfs_symlink(struct inode *dip, struct dentry *dentry,
		const char *name)
{
	int rc;
	tid_t tid;
	ino_t ino = 0;
	struct component_name dname;
	int ssize;		/* source pathname size */
	struct btstack btstack;
	struct inode *ip = dentry->d_inode;
	unchar *i_fastsymlink;
	s64 xlen = 0;
	int bmask = 0, xsize;
	s64 extent = 0, xaddr;
	struct metapage *mp;
	struct super_block *sb;
	struct tblock *tblk;

	struct inode *iplist[2];

	jfs_info("jfs_symlink: dip:0x%p name:%s", dip, name);

	ssize = strlen(name) + 1;

	/*
	 * search parent directory for entry/freespace
	 * (dtSearch() returns parent directory page pinned)
	 */

	if ((rc = get_UCSname(&dname, dentry)))
		goto out1;

	/*
	 * allocate on-disk/in-memory inode for symbolic link:
	 * (iAlloc() returns new, locked inode)
	 */
	ip = ialloc(dip, S_IFLNK | 0777);
	if (ip == NULL) {
		rc = -ENOSPC;
		goto out2;
	}

	tid = txBegin(dip->i_sb, 0);

	mutex_lock(&JFS_IP(dip)->commit_mutex);
	mutex_lock(&JFS_IP(ip)->commit_mutex);

	rc = jfs_init_security(tid, ip, dip);
	if (rc)
		goto out3;

	tblk = tid_to_tblock(tid);
	tblk->xflag |= COMMIT_CREATE;
	tblk->ino = ip->i_ino;
	tblk->u.ixpxd = JFS_IP(ip)->ixpxd;

	/* fix symlink access permission
	 * (dir_create() ANDs in the u.u_cmask, 
	 * but symlinks really need to be 777 access)
	 */
	ip->i_mode |= 0777;

	/*
	 * write symbolic link target path name
	 */
	xtInitRoot(tid, ip);

	/*
	 * write source path name inline in on-disk inode (fast symbolic link)
	 */

	if (ssize <= IDATASIZE) {
		ip->i_op = &jfs_symlink_inode_operations;

		i_fastsymlink = JFS_IP(ip)->i_inline;
		memcpy(i_fastsymlink, name, ssize);
		ip->i_size = ssize - 1;

		/*
		 * if symlink is > 128 bytes, we don't have the space to
		 * store inline extended attributes
		 */
		if (ssize > sizeof (JFS_IP(ip)->i_inline))
			JFS_IP(ip)->mode2 &= ~INLINEEA;

		jfs_info("jfs_symlink: fast symlink added  ssize:%d name:%s ",
			 ssize, name);
	}
	/*
	 * write source path name in a single extent
	 */
	else {
		jfs_info("jfs_symlink: allocate extent ip:0x%p", ip);

		ip->i_op = &page_symlink_inode_operations;
		ip->i_mapping->a_ops = &jfs_aops;

		/*
		 * even though the data of symlink object (source 
		 * path name) is treated as non-journaled user data,
		 * it is read/written thru buffer cache for performance.
		 */
		sb = ip->i_sb;
		bmask = JFS_SBI(sb)->bsize - 1;
		xsize = (ssize + bmask) & ~bmask;
		xaddr = 0;
		xlen = xsize >> JFS_SBI(sb)->l2bsize;
		if ((rc = xtInsert(tid, ip, 0, 0, xlen, &xaddr, 0))) {
			txAbort(tid, 0);
			rc = -ENOSPC;
			goto out3;
		}
		extent = xaddr;
		ip->i_size = ssize - 1;
		while (ssize) {
			/* This is kind of silly since PATH_MAX == 4K */
			int copy_size = min(ssize, PSIZE);

			mp = get_metapage(ip, xaddr, PSIZE, 1);

			if (mp == NULL) {
				xtTruncate(tid, ip, 0, COMMIT_PWMAP);
				rc = -EIO;
				txAbort(tid, 0);
				goto out3;
			}
			memcpy(mp->data, name, copy_size);
			flush_metapage(mp);
			ssize -= copy_size;
			name += copy_size;
			xaddr += JFS_SBI(sb)->nbperpage;
		}
	}

	/*
	 * create entry for symbolic link in parent directory
	 */
	rc = dtSearch(dip, &dname, &ino, &btstack, JFS_CREATE);
	if (rc == 0) {
		ino = ip->i_ino;
		rc = dtInsert(tid, dip, &dname, &ino, &btstack);
	}
	if (rc) {
		if (xlen)
			xtTruncate(tid, ip, 0, COMMIT_PWMAP);
		txAbort(tid, 0);
		/* discard new inode */
		goto out3;
	}

	insert_inode_hash(ip);
	mark_inode_dirty(ip);

	dip->i_ctime = dip->i_mtime = CURRENT_TIME;
	mark_inode_dirty(dip);
	/*
	 * commit update of parent directory and link object
	 */

	iplist[0] = dip;
	iplist[1] = ip;
	rc = txCommit(tid, 2, &iplist[0], 0);

      out3:
	txEnd(tid);
	mutex_unlock(&JFS_IP(dip)->commit_mutex);
	mutex_unlock(&JFS_IP(ip)->commit_mutex);
	if (rc) {
		free_ea_wmap(ip);
		ip->i_nlink = 0;
		iput(ip);
	} else
		d_instantiate(dentry, ip);

      out2:
	free_UCSname(&dname);

      out1:
	jfs_info("jfs_symlink: rc:%d", rc);
	return rc;
}


/*
 * NAME:        jfs_rename
 *
 * FUNCTION:    rename a file or directory
 */
static int jfs_rename(struct inode *old_dir, struct dentry *old_dentry,
	       struct inode *new_dir, struct dentry *new_dentry)
{
	struct btstack btstack;
	ino_t ino;
	struct component_name new_dname;
	struct inode *new_ip;
	struct component_name old_dname;
	struct inode *old_ip;
	int rc;
	tid_t tid;
	struct tlock *tlck;
	struct dt_lock *dtlck;
	struct lv *lv;
	int ipcount;
	struct inode *iplist[4];
	struct tblock *tblk;
	s64 new_size = 0;
	int commit_flag;


	jfs_info("jfs_rename: %s %s", old_dentry->d_name.name,
		 new_dentry->d_name.name);

	old_ip = old_dentry->d_inode;
	new_ip = new_dentry->d_inode;

	if ((rc = get_UCSname(&old_dname, old_dentry)))
		goto out1;

	if ((rc = get_UCSname(&new_dname, new_dentry)))
		goto out2;

	/*
	 * Make sure source inode number is what we think it is
	 */
	rc = dtSearch(old_dir, &old_dname, &ino, &btstack, JFS_LOOKUP);
	if (rc || (ino != old_ip->i_ino)) {
		rc = -ENOENT;
		goto out3;
	}

	/*
	 * Make sure dest inode number (if any) is what we think it is
	 */
	rc = dtSearch(new_dir, &new_dname, &ino, &btstack, JFS_LOOKUP);
	if (rc == 0) {
		if ((new_ip == 0) || (ino != new_ip->i_ino)) {
			rc = -ESTALE;
			goto out3;
		}
	} else if (rc != -ENOENT)
		goto out3;
	else if (new_ip) {
		/* no entry exists, but one was expected */
		rc = -ESTALE;
		goto out3;
	}

	if (S_ISDIR(old_ip->i_mode)) {
		if (new_ip) {
			if (!dtEmpty(new_ip)) {
				rc = -ENOTEMPTY;
				goto out3;
			}
		} else if ((new_dir != old_dir) &&
			   (new_dir->i_nlink == JFS_LINK_MAX)) {
			rc = -EMLINK;
			goto out3;
		}
	} else if (new_ip) {
		IWRITE_LOCK(new_ip);
		/* Init inode for quota operations. */
		DQUOT_INIT(new_ip);
	}

	/*
	 * The real work starts here
	 */
	tid = txBegin(new_dir->i_sb, 0);

	mutex_lock(&JFS_IP(new_dir)->commit_mutex);
	mutex_lock(&JFS_IP(old_ip)->commit_mutex);
	if (old_dir != new_dir)
		mutex_lock(&JFS_IP(old_dir)->commit_mutex);

	if (new_ip) {
		mutex_lock(&JFS_IP(new_ip)->commit_mutex);
		/*
		 * Change existing directory entry to new inode number
		 */
		ino = new_ip->i_ino;
		rc = dtModify(tid, new_dir, &new_dname, &ino,
			      old_ip->i_ino, JFS_RENAME);
		if (rc)
			goto out4;
		new_ip->i_nlink--;
		if (S_ISDIR(new_ip->i_mode)) {
			new_ip->i_nlink--;
			if (new_ip->i_nlink) {
				mutex_unlock(&JFS_IP(new_dir)->commit_mutex);
				mutex_unlock(&JFS_IP(old_ip)->commit_mutex);
				if (old_dir != new_dir)
					mutex_unlock(&JFS_IP(old_dir)->commit_mutex);
				if (!S_ISDIR(old_ip->i_mode) && new_ip)
					IWRITE_UNLOCK(new_ip);
				jfs_error(new_ip->i_sb,
					  "jfs_rename: new_ip->i_nlink != 0");
				return -EIO;
			}
			tblk = tid_to_tblock(tid);
			tblk->xflag |= COMMIT_DELETE;
			tblk->u.ip = new_ip;
		} else if (new_ip->i_nlink == 0) {
			assert(!test_cflag(COMMIT_Nolink, new_ip));
			/* free block resources */
			if ((new_size = commitZeroLink(tid, new_ip)) < 0) {
				txAbort(tid, 1);	/* Marks FS Dirty */
				rc = new_size;		
				goto out4;
			}
			tblk = tid_to_tblock(tid);
			tblk->xflag |= COMMIT_DELETE;
			tblk->u.ip = new_ip;
		} else {
			new_ip->i_ctime = CURRENT_TIME;
			mark_inode_dirty(new_ip);
		}
	} else {
		/*
		 * Add new directory entry
		 */
		rc = dtSearch(new_dir, &new_dname, &ino, &btstack,
			      JFS_CREATE);
		if (rc) {
			jfs_err("jfs_rename didn't expect dtSearch to fail "
				"w/rc = %d", rc);
			goto out4;
		}

		ino = old_ip->i_ino;
		rc = dtInsert(tid, new_dir, &new_dname, &ino, &btstack);
		if (rc) {
			if (rc == -EIO)
				jfs_err("jfs_rename: dtInsert returned -EIO");
			goto out4;
		}
		if (S_ISDIR(old_ip->i_mode))
			new_dir->i_nlink++;
	}
	/*
	 * Remove old directory entry
	 */

	ino = old_ip->i_ino;
	rc = dtDelete(tid, old_dir, &old_dname, &ino, JFS_REMOVE);
	if (rc) {
		jfs_err("jfs_rename did not expect dtDelete to return rc = %d",
			rc);
		txAbort(tid, 1);	/* Marks Filesystem dirty */
		goto out4;
	}
	if (S_ISDIR(old_ip->i_mode)) {
		old_dir->i_nlink--;
		if (old_dir != new_dir) {
			/*
			 * Change inode number of parent for moved directory
			 */

			JFS_IP(old_ip)->i_dtroot.header.idotdot =
				cpu_to_le32(new_dir->i_ino);

			/* Linelock header of dtree */
			tlck = txLock(tid, old_ip,
				    (struct metapage *) &JFS_IP(old_ip)->bxflag,
				      tlckDTREE | tlckBTROOT | tlckRELINK);
			dtlck = (struct dt_lock *) & tlck->lock;
			ASSERT(dtlck->index == 0);
			lv = & dtlck->lv[0];
			lv->offset = 0;
			lv->length = 1;
			dtlck->index++;
		}
	}

	/*
	 * Update ctime on changed/moved inodes & mark dirty
	 */
	old_ip->i_ctime = CURRENT_TIME;
	mark_inode_dirty(old_ip);

	new_dir->i_ctime = new_dir->i_mtime = current_fs_time(new_dir->i_sb);
	mark_inode_dirty(new_dir);

	/* Build list of inodes modified by this transaction */
	ipcount = 0;
	iplist[ipcount++] = old_ip;
	if (new_ip)
		iplist[ipcount++] = new_ip;
	iplist[ipcount++] = old_dir;

	if (old_dir != new_dir) {
		iplist[ipcount++] = new_dir;
		old_dir->i_ctime = old_dir->i_mtime = CURRENT_TIME;
		mark_inode_dirty(old_dir);
	}

	/*
	 * Incomplete truncate of file data can
	 * result in timing problems unless we synchronously commit the
	 * transaction.
	 */
	if (new_size)
		commit_flag = COMMIT_SYNC;
	else
		commit_flag = 0;

	rc = txCommit(tid, ipcount, iplist, commit_flag);

      out4:
	txEnd(tid);

	mutex_unlock(&JFS_IP(new_dir)->commit_mutex);
	mutex_unlock(&JFS_IP(old_ip)->commit_mutex);
	if (old_dir != new_dir)
		mutex_unlock(&JFS_IP(old_dir)->commit_mutex);
	if (new_ip)
		mutex_unlock(&JFS_IP(new_ip)->commit_mutex);

	while (new_size && (rc == 0)) {
		tid = txBegin(new_ip->i_sb, 0);
		mutex_lock(&JFS_IP(new_ip)->commit_mutex);
		new_size = xtTruncate_pmap(tid, new_ip, new_size);
		if (new_size < 0) {
			txAbort(tid, 1);
			rc = new_size;		
		} else
			rc = txCommit(tid, 1, &new_ip, COMMIT_SYNC);
		txEnd(tid);
		mutex_unlock(&JFS_IP(new_ip)->commit_mutex);
	}
	if (new_ip && (new_ip->i_nlink == 0))
		set_cflag(COMMIT_Nolink, new_ip);
      out3:
	free_UCSname(&new_dname);
      out2:
	free_UCSname(&old_dname);
      out1:
	if (new_ip && !S_ISDIR(new_ip->i_mode))
		IWRITE_UNLOCK(new_ip);
	/*
	 * Truncating the directory index table is not guaranteed.  It
	 * may need to be done iteratively
	 */
	if (test_cflag(COMMIT_Stale, old_dir)) {
		if (old_dir->i_size > 1)
			jfs_truncate_nolock(old_dir, 0);

		clear_cflag(COMMIT_Stale, old_dir);
	}

	jfs_info("jfs_rename: returning %d", rc);
	return rc;
}


/*
 * NAME:        jfs_mknod
 *
 * FUNCTION:    Create a special file (device)
 */
static int jfs_mknod(struct inode *dir, struct dentry *dentry,
		int mode, dev_t rdev)
{
	struct jfs_inode_info *jfs_ip;
	struct btstack btstack;
	struct component_name dname;
	ino_t ino;
	struct inode *ip;
	struct inode *iplist[2];
	int rc;
	tid_t tid;
	struct tblock *tblk;

	if (!new_valid_dev(rdev))
		return -EINVAL;

	jfs_info("jfs_mknod: %s", dentry->d_name.name);

	if ((rc = get_UCSname(&dname, dentry)))
		goto out;

	ip = ialloc(dir, mode);
	if (ip == NULL) {
		rc = -ENOSPC;
		goto out1;
	}
	jfs_ip = JFS_IP(ip);

	tid = txBegin(dir->i_sb, 0);

	mutex_lock(&JFS_IP(dir)->commit_mutex);
	mutex_lock(&JFS_IP(ip)->commit_mutex);

	rc = jfs_init_acl(tid, ip, dir);
	if (rc)
		goto out3;

	rc = jfs_init_security(tid, ip, dir);
	if (rc) {
		txAbort(tid, 0);
		goto out3;
	}

	if ((rc = dtSearch(dir, &dname, &ino, &btstack, JFS_CREATE))) {
		txAbort(tid, 0);
		goto out3;
	}

	tblk = tid_to_tblock(tid);
	tblk->xflag |= COMMIT_CREATE;
	tblk->ino = ip->i_ino;
	tblk->u.ixpxd = JFS_IP(ip)->ixpxd;

	ino = ip->i_ino;
	if ((rc = dtInsert(tid, dir, &dname, &ino, &btstack))) {
		txAbort(tid, 0);
		goto out3;
	}

	ip->i_op = &jfs_file_inode_operations;
	jfs_ip->dev = new_encode_dev(rdev);
	init_special_inode(ip, ip->i_mode, rdev);

	insert_inode_hash(ip);
	mark_inode_dirty(ip);

	dir->i_ctime = dir->i_mtime = CURRENT_TIME;

	mark_inode_dirty(dir);

	iplist[0] = dir;
	iplist[1] = ip;
	rc = txCommit(tid, 2, iplist, 0);

      out3:
	txEnd(tid);
	mutex_unlock(&JFS_IP(ip)->commit_mutex);
	mutex_unlock(&JFS_IP(dir)->commit_mutex);
	if (rc) {
		free_ea_wmap(ip);
		ip->i_nlink = 0;
		iput(ip);
	} elselass="hl com">			 * changed the permissions behind our back.
			 */
			if (__mandatory_lock(inode))
				continue;
		}

		locks_delete_block(&fl);
		break;
	}

	return error;
}

EXPORT_SYMBOL(locks_mandatory_area);

static void lease_clear_pending(struct file_lock *fl, int arg)
{
	switch (arg) {
	case F_UNLCK:
		fl->fl_flags &= ~FL_UNLOCK_PENDING;
		/* fall through: */
	case F_RDLCK:
		fl->fl_flags &= ~FL_DOWNGRADE_PENDING;
	}
}

/* We already had a lease on this file; just change its type */
int lease_modify(struct file_lock **before, int arg)
{
	struct file_lock *fl = *before;
	int error = assign_type(fl, arg);

	if (error)
		return error;
	lease_clear_pending(fl, arg);
	locks_wake_up_blocks(fl);
	if (arg == F_UNLCK)
		locks_delete_lock(before);
	return 0;
}

EXPORT_SYMBOL(lease_modify);

static bool past_time(unsigned long then)
{
	if (!then)
		/* 0 is a special value meaning "this never expires": */
		return false;
	return time_after(jiffies, then);
}

static void time_out_leases(struct inode *inode)
{
	struct file_lock **before;
	struct file_lock *fl;

	before = &inode->i_flock;
	while ((fl = *before) && IS_LEASE(fl) && lease_breaking(fl)) {
		if (past_time(fl->fl_downgrade_time))
			lease_modify(before, F_RDLCK);
		if (past_time(fl->fl_break_time))
			lease_modify(before, F_UNLCK);
		if (fl == *before)	/* lease_modify may have freed fl */
			before = &fl->fl_next;
	}
}

/**
 *	__break_lease	-	revoke all outstanding leases on file
 *	@inode: the inode of the file to return
 *	@mode: the open mode (read or write)
 *
 *	break_lease (inlined for speed) has checked there already is at least
 *	some kind of lock (maybe a lease) on this file.  Leases are broken on
 *	a call to open() or truncate().  This function can sleep unless you
 *	specified %O_NONBLOCK to your open().
 */
int __break_lease(struct inode *inode, unsigned int mode)
{
	int error = 0;
	struct file_lock *new_fl, *flock;
	struct file_lock *fl;
	unsigned long break_time;
	int i_have_this_lease = 0;
	int want_write = (mode & O_ACCMODE) != O_RDONLY;

	new_fl = lease_alloc(NULL, want_write ? F_WRLCK : F_RDLCK);
	if (IS_ERR(new_fl))
		return PTR_ERR(new_fl);

	lock_flocks();

	time_out_leases(inode);

	flock = inode->i_flock;
	if ((flock == NULL) || !IS_LEASE(flock))
		goto out;

	if (!locks_conflict(flock, new_fl))
		goto out;

	for (fl = flock; fl && IS_LEASE(fl); fl = fl->fl_next)
		if (fl->fl_owner == current->files)
			i_have_this_lease = 1;

	break_time = 0;
	if (lease_break_time > 0) {
		break_time = jiffies + lease_break_time * HZ;
		if (break_time == 0)
			break_time++;	/* so that 0 means no break time */
	}

	for (fl = flock; fl && IS_LEASE(fl); fl = fl->fl_next) {
		if (want_write) {
			if (fl->fl_flags & FL_UNLOCK_PENDING)
				continue;
			fl->fl_flags |= FL_UNLOCK_PENDING;
			fl->fl_break_time = break_time;
		} else {
			if (lease_breaking(flock))
				continue;
			fl->fl_flags |= FL_DOWNGRADE_PENDING;
			fl->fl_downgrade_time = break_time;
		}
		fl->fl_lmops->lm_break(fl);
	}

	if (i_have_this_lease || (mode & O_NONBLOCK)) {
		error = -EWOULDBLOCK;
		goto out;
	}

restart:
	break_time = flock->fl_break_time;
	if (break_time != 0) {
		break_time -= jiffies;
		if (break_time == 0)
			break_time++;
	}
	locks_insert_block(flock, new_fl);
	unlock_flocks();
	error = wait_event_interruptible_timeout(new_fl->fl_wait,
						!new_fl->fl_next, break_time);
	lock_flocks();
	__locks_delete_block(new_fl);
	if (error >= 0) {
		if (error == 0)
			time_out_leases(inode);
		/*
		 * Wait for the next conflicting lease that has not been
		 * broken yet
		 */
		for (flock = inode->i_flock; flock && IS_LEASE(flock);
				flock = flock->fl_next) {
			if (locks_conflict(new_fl, flock))
				goto restart;
		}
		error = 0;
	}

out:
	unlock_flocks();
	locks_free_lock(new_fl);
	return error;
}

EXPORT_SYMBOL(__break_lease);

/**
 *	lease_get_mtime - get the last modified time of an inode
 *	@inode: the inode
 *      @time:  pointer to a timespec which will contain the last modified time
 *
 * This is to force NFS clients to flush their caches for files with
 * exclusive leases.  The justification is that if someone has an
 * exclusive lease, then they could be modifying it.
 */
void lease_get_mtime(struct inode *inode, struct timespec *time)
{
	struct file_lock *flock = inode->i_flock;
	if (flock && IS_LEASE(flock) && (flock->fl_type & F_WRLCK))
		*time = current_fs_time(inode->i_sb);
	else
		*time = inode->i_mtime;
}

EXPORT_SYMBOL(lease_get_mtime);

/**
 *	fcntl_getlease - Enquire what lease is currently active
 *	@filp: the file
 *
 *	The value returned by this function will be one of
 *	(if no lease break is pending):
 *
 *	%F_RDLCK to indicate a shared lease is held.
 *
 *	%F_WRLCK to indicate an exclusive lease is held.
 *
 *	%F_UNLCK to indicate no lease is held.
 *
 *	(if a lease break is pending):
 *
 *	%F_RDLCK to indicate an exclusive lease needs to be
 *		changed to a shared lease (or removed).
 *
 *	%F_UNLCK to indicate the lease needs to be removed.
 *
 *	XXX: sfr & willy disagree over whether F_INPROGRESS
 *	should be returned to userspace.
 */
int fcntl_getlease(struct file *filp)
{
	struct file_lock *fl;
	int type = F_UNLCK;

	lock_flocks();
	time_out_leases(filp->f_path.dentry->d_inode);
	for (fl = filp->f_path.dentry->d_inode->i_flock; fl && IS_LEASE(fl);
			fl = fl->fl_next) {
		if (fl->fl_file == filp) {
			type = target_leasetype(fl);
			break;
		}
	}
	unlock_flocks();
	return type;
}

int generic_add_lease(struct file *filp, long arg, struct file_lock **flp)
{
	struct file_lock *fl, **before, **my_before = NULL, *lease;
	struct dentry *dentry = filp->f_path.dentry;
	struct inode *inode = dentry->d_inode;
	int error;

	lease = *flp;

	error = -EAGAIN;
	if ((arg == F_RDLCK) && (atomic_read(&inode->i_writecount) > 0))
		goto out;
	if ((arg == F_WRLCK)
	    && ((dentry->d_count > 1)
		|| (atomic_read(&inode->i_count) > 1)))
		goto out;

	/*
	 * At this point, we know that if there is an exclusive
	 * lease on this file, then we hold it on this filp
	 * (otherwise our open of this file would have blocked).
	 * And if we are trying to acquire an exclusive lease,
	 * then the file is not open by anyone (including us)
	 * except for this filp.
	 */
	error = -EAGAIN;
	for (before = &inode->i_flock;
			((fl = *before) != NULL) && IS_LEASE(fl);
			before = &fl->fl_next) {
		if (fl->fl_file == filp) {
			my_before = before;
			continue;
		}
		/*
		 * No exclusive leases if someone else has a lease on
		 * this file:
		 */
		if (arg == F_WRLCK)
			goto out;
		/*
		 * Modifying our existing lease is OK, but no getting a
		 * new lease if someone else is opening for write:
		 */
		if (fl->fl_flags & FL_UNLOCK_PENDING)
			goto out;
	}

	if (my_before != NULL) {
		error = lease->fl_lmops->lm_change(my_before, arg);
		if (!error)
			*flp = *my_before;
		goto out;
	}

	error = -EINVAL;
	if (!leases_enable)
		goto out;

	locks_insert_lock(before, lease);
	return 0;

out:
	return error;
}

int generic_delete_lease(struct file *filp, struct file_lock **flp)
{
	struct file_lock *fl, **before;
	struct dentry *dentry = filp->f_path.dentry;
	struct inode *inode = dentry->d_inode;

	for (before = &inode->i_flock;
			((fl = *before) != NULL) && IS_LEASE(fl);
			before = &fl->fl_next) {
		if (fl->fl_file != filp)
			continue;
		return (*flp)->fl_lmops->lm_change(before, F_UNLCK);
	}
	return -EAGAIN;
}

/**
 *	generic_setlease	-	sets a lease on an open file
 *	@filp: file pointer
 *	@arg: type of lease to obtain
 *	@flp: input - file_lock to use, output - file_lock inserted
 *
 *	The (input) flp->fl_lmops->lm_break function is required
 *	by break_lease().
 *
 *	Called with file_lock_lock held.
 */
int generic_setlease(struct file *filp, long arg, struct file_lock **flp)
{
	struct dentry *dentry = filp->f_path.dentry;
	struct inode *inode = dentry->d_inode;
	int error;

	if ((current_fsuid() != inode->i_uid) && !capable(CAP_LEASE))
		return -EACCES;
	if (!S_ISREG(inode->i_mode))
		return -EINVAL;
	error = security_file_lock(filp, arg);
	if (error)
		return error;

	time_out_leases(inode);

	BUG_ON(!(*flp)->fl_lmops->lm_break);

	switch (arg) {
	case F_UNLCK:
		return generic_delete_lease(filp, flp);
	case F_RDLCK:
	case F_WRLCK:
		return generic_add_lease(filp, arg, flp);
	default:
		BUG();
	}
}
EXPORT_SYMBOL(generic_setlease);

static int __vfs_setlease(struct file *filp, long arg, struct file_lock **lease)
{
	if (filp->f_op && filp->f_op->setlease)
		return filp->f_op->setlease(filp, arg, lease);
	else
		return generic_setlease(filp, arg, lease);
}

/**
 *	vfs_setlease        -       sets a lease on an open file
 *	@filp: file pointer
 *	@arg: type of lease to obtain
 *	@lease: file_lock to use
 *
 *	Call this to establish a lease on the file.
 *	The (*lease)->fl_lmops->lm_break operation must be set; if not,
 *	break_lease will oops!
 *
 *	This will call the filesystem's setlease file method, if
 *	defined.  Note that there is no getlease method; instead, the
 *	filesystem setlease method should call back to setlease() to
 *	add a lease to the inode's lease list, where fcntl_getlease() can
 *	find it.  Since fcntl_getlease() only reports whether the current
 *	task holds a lease, a cluster filesystem need only do this for
 *	leases held by processes on this node.
 *
 *	There is also no break_lease method; filesystems that
 *	handle their own leases should break leases themselves from the
 *	filesystem's open, create, and (on truncate) setattr methods.
 *
 *	Warning: the only current setlease methods exist only to disable
 *	leases in certain cases.  More vfs changes may be required to
 *	allow a full filesystem lease implementation.
 */

int vfs_setlease(struct file *filp, long arg, struct file_lock **lease)
{
	int error;

	lock_flocks();
	error = __vfs_setlease(filp, arg, lease);
	unlock_flocks();

	return error;
}
EXPORT_SYMBOL_GPL(vfs_setlease);

static int do_fcntl_delete_lease(struct file *filp)
{
	struct file_lock fl, *flp = &fl;

	lease_init(filp, F_UNLCK, flp);

	return vfs_setlease(filp, F_UNLCK, &flp);
}

static int do_fcntl_add_lease(unsigned int fd, struct file *filp, long arg)
{
	struct file_lock *fl, *ret;
	struct fasync_struct *new;
	int error;

	fl = lease_alloc(filp, arg);
	if (IS_ERR(fl))
		return PTR_ERR(fl);

	new = fasync_alloc();
	if (!new) {
		locks_free_lock(fl);
		return -ENOMEM;
	}
	ret = fl;
	lock_flocks();
	error = __vfs_setlease(filp, arg, &ret);
	if (error) {
		unlock_flocks();
		locks_free_lock(fl);
		goto out_free_fasync;
	}
	if (ret != fl)
		locks_free_lock(fl);

	/*
	 * fasync_insert_entry() returns the old entry if any.
	 * If there was no old entry, then it used 'new' and
	 * inserted it into the fasync list. Clear new so that
	 * we don't release it here.
	 */
	if (!fasync_insert_entry(fd, filp, &ret->fl_fasync, new))
		new = NULL;

	error = __f_setown(filp, task_pid(current), PIDTYPE_PID, 0);
	unlock_flocks();

out_free_fasync:
	if (new)
		fasync_free(new);
	return error;
}

/**
 *	fcntl_setlease	-	sets a lease on an open file
 *	@fd: open file descriptor
 *	@filp: file pointer
 *	@arg: type of lease to obtain
 *
 *	Call this fcntl to establish a lease on the file.
 *	Note that you also need to call %F_SETSIG to
 *	receive a signal when the lease is broken.
 */
int fcntl_setlease(unsigned int fd, struct file *filp, long arg)
{
	if (arg == F_UNLCK)
		return do_fcntl_delete_lease(filp);
	return do_fcntl_add_lease(fd, filp, arg);
}

/**
 * flock_lock_file_wait - Apply a FLOCK-style lock to a file
 * @filp: The file to apply the lock to
 * @fl: The lock to be applied
 *
 * Add a FLOCK style lock to a file.
 */
int flock_lock_file_wait(struct file *filp, struct file_lock *fl)
{
	int error;
	might_sleep();
	for (;;) {
		error = flock_lock_file(filp, fl);
		if (error != FILE_LOCK_DEFERRED)
			break;
		error = wait_event_interruptible(fl->fl_wait, !fl->fl_next);
		if (!error)
			continue;

		locks_delete_block(fl);
		break;
	}
	return error;
}

EXPORT_SYMBOL(flock_lock_file_wait);

/**
 *	sys_flock: - flock() system call.
 *	@fd: the file descriptor to lock.
 *	@cmd: the type of lock to apply.
 *
 *	Apply a %FL_FLOCK style lock to an open file descriptor.
 *	The @cmd can be one of
 *
 *	%LOCK_SH -- a shared lock.
 *
 *	%LOCK_EX -- an exclusive lock.
 *
 *	%LOCK_UN -- remove an existing lock.
 *
 *	%LOCK_MAND -- a `mandatory' flock.  This exists to emulate Windows Share Modes.
 *
 *	%LOCK_MAND can be combined with %LOCK_READ or %LOCK_WRITE to allow other
 *	processes read and write access respectively.
 */
SYSCALL_DEFINE2(flock, unsigned int, fd, unsigned int, cmd)
{
	struct file *filp;
	struct file_lock *lock;
	int can_sleep, unlock;
	int error;

	error = -EBADF;
	filp = fget(fd);
	if (!filp)
		goto out;

	can_sleep = !(cmd & LOCK_NB);
	cmd &= ~LOCK_NB;
	unlock = (cmd == LOCK_UN);

	if (!unlock && !(cmd & LOCK_MAND) &&
	    !(filp->f_mode & (FMODE_READ|FMODE_WRITE)))
		goto out_putf;

	error = flock_make_lock(filp, &lock, cmd);
	if (error)
		goto out_putf;
	if (can_sleep)
		lock->fl_flags |= FL_SLEEP;

	error = security_file_lock(filp, lock->fl_type);
	if (error)
		goto out_free;

	if (filp->f_op && filp->f_op->flock)
		error = filp->f_op->flock(filp,
					  (can_sleep) ? F_SETLKW : F_SETLK,
					  lock);
	else
		error = flock_lock_file_wait(filp, lock);

 out_free:
	locks_free_lock(lock);

 out_putf:
	fput(filp);
 out:
	return error;
}

/**
 * vfs_test_lock - test file byte range lock
 * @filp: The file to test lock for
 * @fl: The lock to test; also used to hold result
 *
 * Returns -ERRNO on failure.  Indicates presence of conflicting lock by
 * setting conf->fl_type to something other than F_UNLCK.
 */
int vfs_test_lock(struct file *filp, struct file_lock *fl)
{
	if (filp->f_op && filp->f_op->lock)
		return filp->f_op->lock(filp, F_GETLK, fl);
	posix_test_lock(filp, fl);
	return 0;
}
EXPORT_SYMBOL_GPL(vfs_test_lock);

static int posix_lock_to_flock(struct flock *flock, struct file_lock *fl)
{
	flock->l_pid = fl->fl_pid;
#if BITS_PER_LONG == 32
	/*
	 * Make sure we can represent the posix lock via
	 * legacy 32bit flock.
	 */
	if (fl->fl_start > OFFT_OFFSET_MAX)
		return -EOVERFLOW;
	if (fl->fl_end != OFFSET_MAX && fl->fl_end > OFFT_OFFSET_MAX)
		return -EOVERFLOW;
#endif
	flock->l_start = fl->fl_start;
	flock->l_len = fl->fl_end == OFFSET_MAX ? 0 :
		fl->fl_end - fl->fl_start + 1;
	flock->l_whence = 0;
	flock->l_type = fl->fl_type;
	return 0;
}

#if BITS_PER_LONG == 32
static void posix_lock_to_flock64(struct flock64 *flock, struct file_lock *fl)
{
	flock->l_pid = fl->fl_pid;
	flock->l_start = fl->fl_start;
	flock->l_len = fl->fl_end == OFFSET_MAX ? 0 :
		fl->fl_end - fl->fl_start + 1;
	flock->l_whence = 0;
	flock->l_type = fl->fl_type;
}
#endif

/* Report the first existing lock that would conflict with l.
 * This implements the F_GETLK command of fcntl().
 */
int fcntl_getlk(struct file *filp, struct flock __user *l)
{
	struct file_lock file_lock;
	struct flock flock;
	int error;

	error = -EFAULT;
	if (copy_from_user(&flock, l, sizeof(flock)))
		goto out;
	error = -EINVAL;
	if ((flock.l_type != F_RDLCK) && (flock.l_type != F_WRLCK))
		goto out;

	error = flock_to_posix_lock(filp, &file_lock, &flock);
	if (error)
		goto out;

	error = vfs_test_lock(filp, &file_lock);
	if (error)
		goto out;
 
	flock.l_type = file_lock.fl_type;
	if (file_lock.fl_type != F_UNLCK) {
		error = posix_lock_to_flock(&flock, &file_lock);
		if (error)
			goto out;
	}
	error = -EFAULT;
	if (!copy_to_user(l, &flock, sizeof(flock)))
		error = 0;
out:
	return error;
}

/**
 * vfs_lock_file - file byte range lock
 * @filp: The file to apply the lock to
 * @cmd: type of locking operation (F_SETLK, F_GETLK, etc.)
 * @fl: The lock to be applied
 * @conf: Place to return a copy of the conflicting lock, if found.
 *
 * A caller that doesn't care about the conflicting lock may pass NULL
 * as the final argument.
 *
 * If the filesystem defines a private ->lock() method, then @conf will
 * be left unchanged; so a caller that cares should initialize it to
 * some acceptable default.
 *
 * To avoid blocking kernel daemons, such as lockd, that need to acquire POSIX
 * locks, the ->lock() interface may return asynchronously, before the lock has
 * been granted or denied by the underlying filesystem, if (and only if)
 * lm_grant is set. Callers expecting ->lock() to return asynchronously
 * will only use F_SETLK, not F_SETLKW; they will set FL_SLEEP if (and only if)
 * the request is for a blocking lock. When ->lock() does return asynchronously,
 * it must return FILE_LOCK_DEFERRED, and call ->lm_grant() when the lock
 * request completes.
 * If the request is for non-blocking lock the file system should return
 * FILE_LOCK_DEFERRED then try to get the lock and call the callback routine
 * with the result. If the request timed out the callback routine will return a
 * nonzero return code and the file system should release the lock. The file
 * system is also responsible to keep a corresponding posix lock when it
 * grants a lock so the VFS can find out which locks are locally held and do
 * the correct lock cleanup when required.
 * The underlying filesystem must not drop the kernel lock or call
 * ->lm_grant() before returning to the caller with a FILE_LOCK_DEFERRED
 * return code.
 */
int vfs_lock_file(struct file *filp, unsigned int cmd, struct file_lock *fl, struct file_lock *conf)
{
	if (filp->f_op && filp->f_op->lock)
		return filp->f_op->lock(filp, cmd, fl);
	else
		return posix_lock_file(filp, fl, conf);
}
EXPORT_SYMBOL_GPL(vfs_lock_file);

static int do_lock_file_wait(struct file *filp, unsigned int cmd,
			     struct file_lock *fl)
{
	int error;

	error = security_file_lock(filp, fl->fl_type);
	if (error)
		return error;

	for (;;) {
		error = vfs_lock_file(filp, cmd, fl, NULL);
		if (error != FILE_LOCK_DEFERRED)
			break;
		error = wait_event_interruptible(fl->fl_wait, !fl->fl_next);
		if (!error)
			continue;

		locks_delete_block(fl);
		break;
	}

	return error;
}

/* Apply the lock described by l to an open file descriptor.
 * This implements both the F_SETLK and F_SETLKW commands of fcntl().
 */
int fcntl_setlk(unsigned int fd, struct file *filp, unsigned int cmd,
		struct flock __user *l)
{
	struct file_lock *file_lock = locks_alloc_lock();
	struct flock flock;
	struct inode *inode;
	struct file *f;
	int error;

	if (file_lock == NULL)
		return -ENOLCK;

	/*
	 * This might block, so we do it before checking the inode.
	 */
	error = -EFAULT;
	if (copy_from_user(&flock, l, sizeof(flock)))
		goto out;

	inode = filp->f_path.dentry->d_inode;

	/* Don't allow mandatory locks on files that may be memory mapped
	 * and shared.
	 */
	if (mandatory_lock(inode) && mapping_writably_mapped(filp->f_mapping)) {
		error = -EAGAIN;
		goto out;
	}

again:
	error = flock_to_posix_lock(filp, file_lock, &flock);
	if (error)
		goto out;
	if (cmd == F_SETLKW) {
		file_lock->fl_flags |= FL_SLEEP;
	}
	
	error = -EBADF;
	switch (flock.l_type) {
	case F_RDLCK:
		if (!(filp->f_mode & FMODE_READ))
			goto out;
		break;
	case F_WRLCK:
		if (!(filp->f_mode & FMODE_WRITE))
			goto out;
		break;
	case F_UNLCK:
		break;
	default:
		error = -EINVAL;
		goto out;
	}

	error = do_lock_file_wait(filp, cmd, file_lock);

	/*
	 * Attempt to detect a close/fcntl race and recover by
	 * releasing the lock that was just acquired.
	 */
	/*
	 * we need that spin_lock here - it prevents reordering between
	 * update of inode->i_flock and check for it done in close().
	 * rcu_read_lock() wouldn't do.
	 */
	spin_lock(&current->files->file_lock);
	f = fcheck(fd);
	spin_unlock(&current->files->file_lock);
	if (!error && f != filp && flock.l_type != F_UNLCK) {
		flock.l_type = F_UNLCK;
		goto again;
	}

out:
	locks_free_lock(file_lock);
	return error;
}

#if BITS_PER_LONG == 32
/* Report the first existing lock that would conflict with l.
 * This implements the F_GETLK command of fcntl().
 */
int fcntl_getlk64(struct file *filp, struct flock64 __user *l)
{
	struct file_lock file_lock;
	struct flock64 flock;
	int error;

	error = -EFAULT;
	if (copy_from_user(&flock, l, sizeof(flock)))
		goto out;
	error = -EINVAL;
	if ((flock.l_type != F_RDLCK) && (flock.l_type != F_WRLCK))
		goto out;

	error = flock64_to_posix_lock(filp, &file_lock, &flock);
	if (error)
		goto out;

	error = vfs_test_lock(filp, &file_lock);
	if (error)
		goto out;

	flock.l_type = file_lock.fl_type;
	if (file_lock.fl_type != F_UNLCK)
		posix_lock_to_flock64(&flock, &file_lock);

	error = -EFAULT;
	if (!copy_to_user(l, &flock, sizeof(flock)))
		error = 0;
  
out:
	return error;
}

/* Apply the lock described by l to an open file descriptor.
 * This implements both the F_SETLK and F_SETLKW commands of fcntl().
 */
int fcntl_setlk64(unsigned int fd, struct file *filp, unsigned int cmd,
		struct flock64 __user *l)
{
	struct file_lock *file_lock = locks_alloc_lock();
	struct flock64 flock;
	struct inode *inode;
	struct file *f;
	int error;

	if (file_lock == NULL)
		return -ENOLCK;

	/*
	 * This might block, so we do it before checking the inode.
	 */
	error = -EFAULT;
	if (copy_from_user(&flock, l, sizeof(flock)))
		goto out;

	inode = filp->f_path.dentry->d_inode;

	/* Don't allow mandatory locks on files that may be memory mapped
	 * and shared.
	 */
	if (mandatory_lock(inode) && mapping_writably_mapped(filp->f_mapping)) {
		error = -EAGAIN;
		goto out;
	}

again:
	error = flock64_to_posix_lock(filp, file_lock, &flock);
	if (error)
		goto out;
	if (cmd == F_SETLKW64) {
		file_lock->fl_flags |= FL_SLEEP;
	}
	
	error = -EBADF;
	switch (flock.l_type) {
	case F_RDLCK:
		if (!(filp->f_mode & FMODE_READ))
			goto out;
		break;
	case F_WRLCK:
		if (!(filp->f_mode & FMODE_WRITE))
			goto out;
		break;
	case F_UNLCK:
		break;
	default:
		error = -EINVAL;
		goto out;
	}

	error = do_lock_file_wait(filp, cmd, file_lock);

	/*
	 * Attempt to detect a close/fcntl race and recover by
	 * releasing the lock that was just acquired.
	 */
	spin_lock(&current->files->file_lock);
	f = fcheck(fd);
	spin_unlock(&current->files->file_lock);
	if (!error && f != filp && flock.l_type != F_UNLCK) {
		flock.l_type = F_UNLCK;
		goto again;
	}

out:
	locks_free_lock(file_lock);
	return error;
}
#endif /* BITS_PER_LONG == 32 */

/*
 * This function is called when the file is being removed
 * from the task's fd array.  POSIX locks belonging to this task
 * are deleted at this time.
 */
void locks_remove_posix(struct file *filp, fl_owner_t owner)
{
	struct file_lock lock;

	/*
	 * If there are no locks held on this file, we don't need to call
	 * posix_lock_file().  Another process could be setting a lock on this
	 * file at the same time, but we wouldn't remove that lock anyway.
	 */
	if (!filp->f_path.dentry->d_inode->i_flock)
		return;

	lock.fl_type = F_UNLCK;
	lock.fl_flags = FL_POSIX | FL_CLOSE;
	lock.fl_start = 0;
	lock.fl_end = OFFSET_MAX;
	lock.fl_owner = owner;
	lock.fl_pid = current->tgid;
	lock.fl_file = filp;
	lock.fl_ops = NULL;
	lock.fl_lmops = NULL;

	vfs_lock_file(filp, F_SETLK, &lock, NULL);

	if (lock.fl_ops && lock.fl_ops->fl_release_private)
		lock.fl_ops->fl_release_private(&lock);
}

EXPORT_SYMBOL(locks_remove_posix);

/*
 * This function is called on the last close of an open file.
 */
void locks_remove_flock(struct file *filp)
{
	struct inode * inode = filp->f_path.dentry->d_inode;
	struct file_lock *fl;
	struct file_lock **before;

	if (!inode->i_flock)
		return;

	if (filp->f_op && filp->f_op->flock) {
		struct file_lock fl = {
			.fl_pid = current->tgid,
			.fl_file = filp,
			.fl_flags = FL_FLOCK,
			.fl_type = F_UNLCK,
			.fl_end = OFFSET_MAX,
		};
		filp->f_op->flock(filp, F_SETLKW, &fl);
		if (fl.fl_ops && fl.fl_ops->fl_release_private)
			fl.fl_ops->fl_release_private(&fl);
	}

	lock_flocks();
	before = &inode->i_flock;

	while ((fl = *before) != NULL) {
		if (fl->fl_file == filp) {
			if (IS_FLOCK(fl)) {
				locks_delete_lock(before);
				continue;
			}
			if (IS_LEASE(fl)) {
				lease_modify(before, F_UNLCK);
				continue;
			}
			/* What? */
			BUG();
 		}
		before = &fl->fl_next;
	}
	unlock_flocks();
}

/**
 *	posix_unblock_lock - stop waiting for a file lock
 *      @filp:   how the file was opened
 *	@waiter: the lock which was waiting
 *
 *	lockd needs to block waiting for locks.
 */
int
posix_unblock_lock(struct file *filp, struct file_lock *waiter)
{
	int status = 0;

	lock_flocks();
	if (waiter->fl_next)
		__locks_delete_block(waiter);
	else
		status = -ENOENT;
	unlock_flocks();
	return status;
}

EXPORT_SYMBOL(posix_unblock_lock);

/**
 * vfs_cancel_lock - file byte range unblock lock
 * @filp: The file to apply the unblock to
 * @fl: The lock to be unblocked
 *
 * Used by lock managers to cancel blocked requests
 */
int vfs_cancel_lock(struct file *filp, struct file_lock *fl)
{
	if (filp->f_op && filp->f_op->lock)
		return filp->f_op->lock(filp, F_CANCELLK, fl);
	return 0;
}

EXPORT_SYMBOL_GPL(vfs_cancel_lock);

#ifdef CONFIG_PROC_FS
#include <linux/proc_fs.h>
#include <linux/seq_file.h>

static void lock_get_status(struct seq_file *f, struct file_lock *fl,
			    loff_t id, char *pfx)
{
	struct inode *inode = NULL;
	unsigned int fl_pid;

	if (fl->fl_nspid)
		fl_pid = pid_vnr(fl->fl_nspid);
	else
		fl_pid = fl->fl_pid;

	if (fl->fl_file != NULL)
		inode = fl->fl_file->f_path.dentry->d_inode;

	seq_printf(f, "%lld:%s ", id, pfx);
	if (IS_POSIX(fl)) {
		seq_printf(f, "%6s %s ",
			     (fl->fl_flags & FL_ACCESS) ? "ACCESS" : "POSIX ",
			     (inode == NULL) ? "*NOINODE*" :
			     mandatory_lock(inode) ? "MANDATORY" : "ADVISORY ");
	} else if (IS_FLOCK(fl)) {
		if (fl->fl_type & LOCK_MAND) {
			seq_printf(f, "FLOCK  MSNFS     ");
		} else {
			seq_printf(f, "FLOCK  ADVISORY  ");
		}
	} else if (IS_LEASE(fl)) {
		seq_printf(f, "LEASE  ");
		if (lease_breaking(fl))
			seq_printf(f, "BREAKING  ");
		else if (fl->fl_file)
			seq_printf(f, "ACTIVE    ");
		else
			seq_printf(f, "BREAKER   ");
	} else {
		seq_printf(f, "UNKNOWN UNKNOWN  ");
	}
	if (fl->fl_type & LOCK_MAND) {
		seq_printf(f, "%s ",
			       (fl->fl_type & LOCK_READ)
			       ? (fl->fl_type & LOCK_WRITE) ? "RW   " : "READ "
			       : (fl->fl_type & LOCK_WRITE) ? "WRITE" : "NONE ");
	} else {
		seq_printf(f, "%s ",
			       (lease_breaking(fl))
			       ? (fl->fl_type & F_UNLCK) ? "UNLCK" : "READ "
			       : (fl->fl_type & F_WRLCK) ? "WRITE" : "READ ");
	}
	if (inode) {
#ifdef WE_CAN_BREAK_LSLK_NOW
		seq_printf(f, "%d %s:%ld ", fl_pid,
				inode->i_sb->s_id, inode->i_ino);
#else
		/* userspace relies on this representation of dev_t ;-( */
		seq_printf(f, "%d %02x:%02x:%ld ", fl_pid,
				MAJOR(inode->i_sb->s_dev),
				MINOR(inode->i_sb->s_dev), inode->i_ino);
#endif
	} else {
		seq_printf(f, "%d <none>:0 ", fl_pid);
	}
	if (IS_POSIX(fl)) {
		if (fl->fl_end == OFFSET_MAX)
			seq_printf(f, "%Ld EOF\n", fl->fl_start);
		else
			seq_printf(f, "%Ld %Ld\n", fl->fl_start, fl->fl_end);
	} else {
		seq_printf(f, "0 EOF\n");
	}
}

static int locks_show(struct seq_file *f, void *v)
{
	struct file_lock *fl, *bfl;

	fl = list_entry(v, struct file_lock, fl_link);

	lock_get_status(f, fl, *((loff_t *)f->private), "");

	list_for_each_entry(bfl, &fl->fl_block, fl_block)
		lock_get_status(f, bfl, *((loff_t *)f->private), " ->");

	return 0;
}

static void *locks_start(struct seq_file *f, loff_t *pos)
{
	loff_t *p = f->private;

	lock_flocks();
	*p = (*pos + 1);
	return seq_list_start(&file_lock_list, *pos);
}

static void *locks_next(struct seq_file *f, void *v, loff_t *pos)
{
	loff_t *p = f->private;
	++*p;
	return seq_list_next(v, &file_lock_list, pos);
}

static void locks_stop(struct seq_file *f, void *v)
{
	unlock_flocks();
}

static const struct seq_operations locks_seq_operations = {
	.start	= locks_start,
	.next	= locks_next,
	.stop	= locks_stop,
	.show	= locks_show,
};

static int locks_open(struct inode *inode, struct file *filp)
{
	return seq_open_private(filp, &locks_seq_operations, sizeof(loff_t));
}

static const struct file_operations proc_locks_operations = {
	.open		= locks_open,
	.read		= seq_read,
	.llseek		= seq_lseek,
	.release	= seq_release_private,
};

static int __init proc_locks_init(void)
{
	proc_create("locks", 0, NULL, &proc_locks_operations);
	return 0;
}
module_init(proc_locks_init);
#endif

/**
 *	lock_may_read - checks that the region is free of locks
 *	@inode: the inode that is being read
 *	@start: the first byte to read
 *	@len: the number of bytes to read
 *
 *	Emulates Windows locking requirements.  Whole-file
 *	mandatory locks (share modes) can prohibit a read and
 *	byte-range POSIX locks can prohibit a read if they overlap.
 *
 *	N.B. this function is only ever called
 *	from knfsd and ownership of locks is never checked.
 */
int lock_may_read(struct inode *inode, loff_t start, unsigned long len)
{
	struct file_lock *fl;
	int result = 1;
	lock_flocks();
	for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) {
		if (IS_POSIX(fl)) {
			if (fl->fl_type == F_RDLCK)
				continue;
			if ((fl->fl_end < start) || (fl->fl_start > (start + len)))
				continue;
		} else if (IS_FLOCK(fl)) {
			if (!(fl->fl_type & LOCK_MAND))
				continue;
			if (fl->fl_type & LOCK_READ)
				continue;
		} else
			continue;
		result = 0;
		break;
	}
	unlock_flocks();
	return result;
}

EXPORT_SYMBOL(lock_may_read);

/**
 *	lock_may_write - checks that the region is free of locks
 *	@inode: the inode that is being written
 *	@start: the first byte to write
 *	@len: the number of bytes to write
 *
 *	Emulates Windows locking requirements.  Whole-file
 *	mandatory locks (share modes) can prohibit a write and
 *	byte-range POSIX locks can prohibit a write if they overlap.
 *
 *	N.B. this function is only ever called
 *	from knfsd and ownership of locks is never checked.
 */
int lock_may_write(struct inode *inode, loff_t start, unsigned long len)
{
	struct file_lock *fl;
	int result = 1;
	lock_flocks();
	for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) {
		if (IS_POSIX(fl)) {
			if ((fl->fl_end < start) || (fl->fl_start > (start + len)))
				continue;
		} else if (IS_FLOCK(fl)) {
			if (!(fl->fl_type & LOCK_MAND))
				continue;
			if (fl->fl_type & LOCK_WRITE)
				continue;
		} else
			continue;
		result = 0;
		break;
	}
	unlock_flocks();
	return result;
}

EXPORT_SYMBOL(lock_may_write);

static int __init filelock_init(void)
{
	filelock_cache = kmem_cache_create("file_lock_cache",
			sizeof(struct file_lock), 0, SLAB_PANIC, NULL);

	return 0;
}

core_initcall(filelock_init);