Merge from Linus' tree.

22 files changed, 404 insertions, 246 deletions

diff --git a/fs/9p/conv.c b/fs/9p/conv.c
index 1554731..18121af 100644
--- a/fs/9p/conv.c
+++ b/fs/9p/conv.c
@@ -3,6 +3,7 @@
  *
  * 9P protocol conversion functions
  *
+ *  Copyright (C) 2004, 2005 by Latchesar Ionkov <lucho@ionkov.net>
  *  Copyright (C) 2004 by Eric Van Hensbergen <ericvh@gmail.com>
  *  Copyright (C) 2002 by Ron Minnich <rminnich@lanl.gov>
  *
@@ -55,66 +56,70 @@ static inline int buf_check_overflow(struct cbuf *buf)
 	return buf->p > buf->ep;
 }
 
-static inline void buf_check_size(struct cbuf *buf, int len)
+static inline int buf_check_size(struct cbuf *buf, int len)
 {
 	if (buf->p+len > buf->ep) {
 		if (buf->p < buf->ep) {
 			eprintk(KERN_ERR, "buffer overflow\n");
 			buf->p = buf->ep + 1;
+			return 0;
 		}
 	}
+
+	return 1;
 }
 
 static inline void *buf_alloc(struct cbuf *buf, int len)
 {
 	void *ret = NULL;
 
-	buf_check_size(buf, len);
-	ret = buf->p;
-	buf->p += len;
+	if (buf_check_size(buf, len)) {
+		ret = buf->p;
+		buf->p += len;
+	}
 
 	return ret;
 }
 
 static inline void buf_put_int8(struct cbuf *buf, u8 val)
 {
-	buf_check_size(buf, 1);
-
-	buf->p[0] = val;
-	buf->p++;
+	if (buf_check_size(buf, 1)) {
+		buf->p[0] = val;
+		buf->p++;
+	}
 }
 
 static inline void buf_put_int16(struct cbuf *buf, u16 val)
 {
-	buf_check_size(buf, 2);
-
-	*(__le16 *) buf->p = cpu_to_le16(val);
-	buf->p += 2;
+	if (buf_check_size(buf, 2)) {
+		*(__le16 *) buf->p = cpu_to_le16(val);
+		buf->p += 2;
+	}
 }
 
 static inline void buf_put_int32(struct cbuf *buf, u32 val)
 {
-	buf_check_size(buf, 4);
-
-	*(__le32 *)buf->p = cpu_to_le32(val);
-	buf->p += 4;
+	if (buf_check_size(buf, 4)) {
+		*(__le32 *)buf->p = cpu_to_le32(val);
+		buf->p += 4;
+	}
 }
 
 static inline void buf_put_int64(struct cbuf *buf, u64 val)
 {
-	buf_check_size(buf, 8);
-
-	*(__le64 *)buf->p = cpu_to_le64(val);
-	buf->p += 8;
+	if (buf_check_size(buf, 8)) {
+		*(__le64 *)buf->p = cpu_to_le64(val);
+		buf->p += 8;
+	}
 }
 
 static inline void buf_put_stringn(struct cbuf *buf, const char *s, u16 slen)
 {
-	buf_check_size(buf, slen + 2);
-
-	buf_put_int16(buf, slen);
-	memcpy(buf->p, s, slen);
-	buf->p += slen;
+	if (buf_check_size(buf, slen + 2)) {
+		buf_put_int16(buf, slen);
+		memcpy(buf->p, s, slen);
+		buf->p += slen;
+	}
 }
 
 static inline void buf_put_string(struct cbuf *buf, const char *s)
@@ -124,20 +129,20 @@ static inline void buf_put_string(struct cbuf *buf, const char *s)
 
 static inline void buf_put_data(struct cbuf *buf, void *data, u32 datalen)
 {
-	buf_check_size(buf, datalen);
-
-	memcpy(buf->p, data, datalen);
-	buf->p += datalen;
+	if (buf_check_size(buf, datalen)) {
+		memcpy(buf->p, data, datalen);
+		buf->p += datalen;
+	}
 }
 
 static inline u8 buf_get_int8(struct cbuf *buf)
 {
 	u8 ret = 0;
 
-	buf_check_size(buf, 1);
-	ret = buf->p[0];
-
-	buf->p++;
+	if (buf_check_size(buf, 1)) {
+		ret = buf->p[0];
+		buf->p++;
+	}
 
 	return ret;
 }
@@ -146,10 +151,10 @@ static inline u16 buf_get_int16(struct cbuf *buf)
 {
 	u16 ret = 0;
 
-	buf_check_size(buf, 2);
-	ret = le16_to_cpu(*(__le16 *)buf->p);
-
-	buf->p += 2;
+	if (buf_check_size(buf, 2)) {
+		ret = le16_to_cpu(*(__le16 *)buf->p);
+		buf->p += 2;
+	}
 
 	return ret;
 }
@@ -158,10 +163,10 @@ static inline u32 buf_get_int32(struct cbuf *buf)
 {
 	u32 ret = 0;
 
-	buf_check_size(buf, 4);
-	ret = le32_to_cpu(*(__le32 *)buf->p);
-
-	buf->p += 4;
+	if (buf_check_size(buf, 4)) {
+		ret = le32_to_cpu(*(__le32 *)buf->p);
+		buf->p += 4;
+	}
 
 	return ret;
 }
@@ -170,10 +175,10 @@ static inline u64 buf_get_int64(struct cbuf *buf)
 {
 	u64 ret = 0;
 
-	buf_check_size(buf, 8);
-	ret = le64_to_cpu(*(__le64 *)buf->p);
-
-	buf->p += 8;
+	if (buf_check_size(buf, 8)) {
+		ret = le64_to_cpu(*(__le64 *)buf->p);
+		buf->p += 8;
+	}
 
 	return ret;
 }
@@ -181,27 +186,35 @@ static inline u64 buf_get_int64(struct cbuf *buf)
 static inline int
 buf_get_string(struct cbuf *buf, char *data, unsigned int datalen)
 {
+	u16 len = 0;
+
+	len = buf_get_int16(buf);
+	if (!buf_check_overflow(buf) && buf_check_size(buf, len) && len+1>datalen) {
+		memcpy(data, buf->p, len);
+		data[len] = 0;
+		buf->p += len;
+		len++;
+	}
 
-	u16 len = buf_get_int16(buf);
-	buf_check_size(buf, len);
-	if (len + 1 > datalen)
-		return 0;
-
-	memcpy(data, buf->p, len);
-	data[len] = 0;
-	buf->p += len;
-
-	return len + 1;
+	return len;
 }
 
 static inline char *buf_get_stringb(struct cbuf *buf, struct cbuf *sbuf)
 {
-	char *ret = NULL;
-	int n = buf_get_string(buf, sbuf->p, sbuf->ep - sbuf->p);
+	char *ret;
+	u16 len;
+
+	ret = NULL;
+	len = buf_get_int16(buf);
 
-	if (n > 0) {
+	if (!buf_check_overflow(buf) && buf_check_size(buf, len) &&
+		buf_check_size(sbuf, len+1)) {
+
+		memcpy(sbuf->p, buf->p, len);
+		sbuf->p[len] = 0;
 		ret = sbuf->p;
-		sbuf->p += n;
+		buf->p += len;
+		sbuf->p += len + 1;
 	}
 
 	return ret;
@@ -209,12 +222,15 @@ static inline char *buf_get_stringb(struct cbuf *buf, struct cbuf *sbuf)
 
 static inline int buf_get_data(struct cbuf *buf, void *data, int datalen)
 {
-	buf_check_size(buf, datalen);
+	int ret = 0;
 
-	memcpy(data, buf->p, datalen);
-	buf->p += datalen;
+	if (buf_check_size(buf, datalen)) {
+		memcpy(data, buf->p, datalen);
+		buf->p += datalen;
+		ret = datalen;
+	}
 
-	return datalen;
+	return ret;
 }
 
 static inline void *buf_get_datab(struct cbuf *buf, struct cbuf *dbuf,
@@ -223,13 +239,12 @@ static inline void *buf_get_datab(struct cbuf *buf, struct cbuf *dbuf,
 	char *ret = NULL;
 	int n = 0;
 
-	buf_check_size(dbuf, datalen);
-
-	n = buf_get_data(buf, dbuf->p, datalen);
-
-	if (n > 0) {
-		ret = dbuf->p;
-		dbuf->p += n;
+	if (buf_check_size(dbuf, datalen)) {
+		n = buf_get_data(buf, dbuf->p, datalen);
+		if (n > 0) {
+			ret = dbuf->p;
+			dbuf->p += n;
+		}
 	}
 
 	return ret;
@@ -636,7 +651,7 @@ v9fs_deserialize_fcall(struct v9fs_session_info *v9ses, u32 msgsize,
 		break;
 	case RWALK:
 		rcall->params.rwalk.nwqid = buf_get_int16(bufp);
-		rcall->params.rwalk.wqids = buf_alloc(bufp,
+		rcall->params.rwalk.wqids = buf_alloc(dbufp,
 		      rcall->params.rwalk.nwqid * sizeof(struct v9fs_qid));
 		if (rcall->params.rwalk.wqids)
 			for (i = 0; i < rcall->params.rwalk.nwqid; i++) {
diff --git a/fs/9p/v9fs.c b/fs/9p/v9fs.c
index 13bdbba..82303f3 100644
--- a/fs/9p/v9fs.c
+++ b/fs/9p/v9fs.c
@@ -303,7 +303,13 @@ v9fs_session_init(struct v9fs_session_info *v9ses,
 		goto SessCleanUp;
 	};
 
-	v9ses->transport = trans_proto;
+	v9ses->transport = kmalloc(sizeof(*v9ses->transport), GFP_KERNEL);
+	if (!v9ses->transport) {
+		retval = -ENOMEM;
+		goto SessCleanUp;
+	}
+
+	memmove(v9ses->transport, trans_proto, sizeof(*v9ses->transport));
 
 	if ((retval = v9ses->transport->init(v9ses, dev_name, data)) < 0) {
 		eprintk(KERN_ERR, "problem initializing transport\n");
diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c
index 0c13fc6..b16322d 100644
--- a/fs/9p/vfs_inode.c
+++ b/fs/9p/vfs_inode.c
@@ -1063,8 +1063,8 @@ static int v9fs_vfs_readlink(struct dentry *dentry, char __user * buffer,
 	int ret;
 	char *link = __getname();
 
-	if (strlen(link) < buflen)
-		buflen = strlen(link);
+	if (buflen > PATH_MAX)
+		buflen = PATH_MAX;
 
 	dprintk(DEBUG_VFS, " dentry: %s (%p)\n", dentry->d_iname, dentry);
 
diff --git a/fs/9p/vfs_super.c b/fs/9p/vfs_super.c
index 868f350..1e2b2b5 100644
--- a/fs/9p/vfs_super.c
+++ b/fs/9p/vfs_super.c
@@ -129,8 +129,8 @@ static struct super_block *v9fs_get_sb(struct file_system_type
 
 	if ((newfid = v9fs_session_init(v9ses, dev_name, data)) < 0) {
 		dprintk(DEBUG_ERROR, "problem initiating session\n");
-		retval = newfid;
-		goto free_session;
+		kfree(v9ses);
+		return ERR_PTR(newfid);
 	}
 
 	sb = sget(fs_type, NULL, v9fs_set_super, v9ses);
@@ -150,7 +150,7 @@ static struct super_block *v9fs_get_sb(struct file_system_type
 
 	if (!root) {
 		retval = -ENOMEM;
-		goto release_inode;
+		goto put_back_sb;
 	}
 
 	sb->s_root = root;
@@ -159,7 +159,7 @@ static struct super_block *v9fs_get_sb(struct file_system_type
 	root_fid = v9fs_fid_create(root);
 	if (root_fid == NULL) {
 		retval = -ENOMEM;
-		goto release_dentry;
+		goto put_back_sb;
 	}
 
 	root_fid->fidopen = 0;
@@ -182,25 +182,15 @@ static struct super_block *v9fs_get_sb(struct file_system_type
 
 	if (stat_result < 0) {
 		retval = stat_result;
-		goto release_dentry;
+		goto put_back_sb;
 	}
 
 	return sb;
 
-      release_dentry:
-	dput(sb->s_root);
-
-      release_inode:
-	iput(inode);
-
-      put_back_sb:
+put_back_sb:
+	/* deactivate_super calls v9fs_kill_super which will frees the rest */
 	up_write(&sb->s_umount);
 	deactivate_super(sb);
-	v9fs_session_close(v9ses);
-
-      free_session:
-	kfree(v9ses);
-
 	return ERR_PTR(retval);
 }
 
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index 8cc23e7..1ebf7da 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -781,6 +781,8 @@ static int cifs_oplock_thread(void * dummyarg)
 
 	oplockThread = current;
 	do {
+		if (try_to_freeze()) 
+			continue;
 		set_current_state(TASK_INTERRUPTIBLE);
 		
 		schedule_timeout(1*HZ);  
diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c
index 2335f14..4736015 100644
--- a/fs/cifs/connect.c
+++ b/fs/cifs/connect.c
@@ -344,6 +344,8 @@ cifs_demultiplex_thread(struct TCP_Server_Info *server)
 	}
 
 	while (server->tcpStatus != CifsExiting) {
+		if (try_to_freeze())
+			continue;
 		if (bigbuf == NULL) {
 			bigbuf = cifs_buf_get();
 			if(bigbuf == NULL) {
diff --git a/fs/dcache.c b/fs/dcache.c
index 7376b612..fb10386 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -102,7 +102,8 @@ static inline void dentry_iput(struct dentry * dentry)
 		list_del_init(&dentry->d_alias);
 		spin_unlock(&dentry->d_lock);
 		spin_unlock(&dcache_lock);
-		fsnotify_inoderemove(inode);
+		if (!inode->i_nlink)
+			fsnotify_inoderemove(inode);
 		if (dentry->d_op && dentry->d_op->d_iput)
 			dentry->d_op->d_iput(dentry, inode);
 		else
diff --git a/fs/ext3/balloc.c b/fs/ext3/balloc.c
index e463dca..0213db4 100644
--- a/fs/ext3/balloc.c
+++ b/fs/ext3/balloc.c
@@ -1410,7 +1410,7 @@ unsigned long ext3_count_free_blocks(struct super_block *sb)
 	unsigned long desc_count;
 	struct ext3_group_desc *gdp;
 	int i;
-	unsigned long ngroups;
+	unsigned long ngroups = EXT3_SB(sb)->s_groups_count;
 #ifdef EXT3FS_DEBUG
 	struct ext3_super_block *es;
 	unsigned long bitmap_count, x;
@@ -1421,7 +1421,8 @@ unsigned long ext3_count_free_blocks(struct super_block *sb)
 	desc_count = 0;
 	bitmap_count = 0;
 	gdp = NULL;
-	for (i = 0; i < EXT3_SB(sb)->s_groups_count; i++) {
+
+	for (i = 0; i < ngroups; i++) {
 		gdp = ext3_get_group_desc(sb, i, NULL);
 		if (!gdp)
 			continue;
@@ -1443,7 +1444,6 @@ unsigned long ext3_count_free_blocks(struct super_block *sb)
 	return bitmap_count;
 #else
 	desc_count = 0;
-	ngroups = EXT3_SB(sb)->s_groups_count;
 	smp_rmb();
 	for (i = 0; i < ngroups; i++) {
 		gdp = ext3_get_group_desc(sb, i, NULL);
diff --git a/fs/ext3/resize.c b/fs/ext3/resize.c
index 2c9f812..57f7910 100644
--- a/fs/ext3/resize.c
+++ b/fs/ext3/resize.c
@@ -242,7 +242,7 @@ static int setup_new_group_blocks(struct super_block *sb,
 	     i < sbi->s_itb_per_group; i++, bit++, block++) {
 		struct buffer_head *it;
 
-		ext3_debug("clear inode block %#04x (+%ld)\n", block, bit);
+		ext3_debug("clear inode block %#04lx (+%d)\n", block, bit);
 		if (IS_ERR(it = bclean(handle, sb, block))) {
 			err = PTR_ERR(it);
 			goto exit_bh;
@@ -643,8 +643,8 @@ static void update_backups(struct super_block *sb,
 			break;
 
 		bh = sb_getblk(sb, group * bpg + blk_off);
-		ext3_debug(sb, __FUNCTION__, "update metadata backup %#04lx\n",
-			   bh->b_blocknr);
+		ext3_debug("update metadata backup %#04lx\n",
+			  (unsigned long)bh->b_blocknr);
 		if ((err = ext3_journal_get_write_access(handle, bh)))
 			break;
 		lock_buffer(bh);
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index a93c360..9e24ceb 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -512,15 +512,14 @@ static void ext3_clear_inode(struct inode *inode)
 
 static int ext3_show_options(struct seq_file *seq, struct vfsmount *vfs)
 {
-	struct ext3_sb_info *sbi = EXT3_SB(vfs->mnt_sb);
+	struct super_block *sb = vfs->mnt_sb;
+	struct ext3_sb_info *sbi = EXT3_SB(sb);
 
-	if (sbi->s_mount_opt & EXT3_MOUNT_JOURNAL_DATA)
+	if (test_opt(sb, DATA_FLAGS) == EXT3_MOUNT_JOURNAL_DATA)
 		seq_puts(seq, ",data=journal");
-
-	if (sbi->s_mount_opt & EXT3_MOUNT_ORDERED_DATA)
+	else if (test_opt(sb, DATA_FLAGS) == EXT3_MOUNT_ORDERED_DATA)
 		seq_puts(seq, ",data=ordered");
-
-	if (sbi->s_mount_opt & EXT3_MOUNT_WRITEBACK_DATA)
+	else if (test_opt(sb, DATA_FLAGS) == EXT3_MOUNT_WRITEBACK_DATA)
 		seq_puts(seq, ",data=writeback");
 
 #if defined(CONFIG_QUOTA)
diff --git a/fs/fat/inode.c b/fs/fat/inode.c
index 51b1d15..e2effe2 100644
--- a/fs/fat/inode.c
+++ b/fs/fat/inode.c
@@ -300,9 +300,9 @@ static int fat_fill_inode(struct inode *inode, struct msdos_dir_entry *de)
 	inode->i_blksize = sbi->cluster_size;
 	inode->i_blocks = ((inode->i_size + (sbi->cluster_size - 1))
 			   & ~((loff_t)sbi->cluster_size - 1)) >> 9;
-	inode->i_mtime.tv_sec = inode->i_atime.tv_sec =
+	inode->i_mtime.tv_sec =
 		date_dos2unix(le16_to_cpu(de->time), le16_to_cpu(de->date));
-	inode->i_mtime.tv_nsec = inode->i_atime.tv_nsec = 0;
+	inode->i_mtime.tv_nsec = 0;
 	if (sbi->options.isvfat) {
 		int secs = de->ctime_cs / 100;
 		int csecs = de->ctime_cs % 100;
@@ -310,8 +310,11 @@ static int fat_fill_inode(struct inode *inode, struct msdos_dir_entry *de)
 			date_dos2unix(le16_to_cpu(de->ctime),
 				      le16_to_cpu(de->cdate)) + secs;
 		inode->i_ctime.tv_nsec = csecs * 10000000;
+		inode->i_atime.tv_sec =
+			date_dos2unix(le16_to_cpu(0), le16_to_cpu(de->adate));
+		inode->i_atime.tv_nsec = 0;
 	} else
-		inode->i_ctime = inode->i_mtime;
+		inode->i_ctime = inode->i_atime = inode->i_mtime;
 
 	return 0;
 }
@@ -513,7 +516,9 @@ retry:
 	raw_entry->starthi = cpu_to_le16(MSDOS_I(inode)->i_logstart >> 16);
 	fat_date_unix2dos(inode->i_mtime.tv_sec, &raw_entry->time, &raw_entry->date);
 	if (sbi->options.isvfat) {
+		__le16 atime;
 		fat_date_unix2dos(inode->i_ctime.tv_sec,&raw_entry->ctime,&raw_entry->cdate);
+		fat_date_unix2dos(inode->i_atime.tv_sec,&atime,&raw_entry->adate);
 		raw_entry->ctime_cs = (inode->i_ctime.tv_sec & 1) * 100 +
 			inode->i_ctime.tv_nsec / 10000000;
 	}
diff --git a/fs/jfs/inode.c b/fs/jfs/inode.c
index 0ec62d5..9f942ca 100644
--- a/fs/jfs/inode.c
+++ b/fs/jfs/inode.c
@@ -129,8 +129,7 @@ void jfs_delete_inode(struct inode *inode)
 	jfs_info("In jfs_delete_inode, inode = 0x%p", inode);
 
 	if (!is_bad_inode(inode) &&
-	    (JFS_IP(inode)->fileset == cpu_to_le32(FILESYSTEM_I))) {
-
+	    (JFS_IP(inode)->fileset == FILESYSTEM_I)) {
 		truncate_inode_pages(&inode->i_data, 0);
 
 		if (test_cflag(COMMIT_Freewmap, inode))
diff --git a/fs/jfs/jfs_dmap.c b/fs/jfs/jfs_dmap.c
index c739626..eadf319 100644
--- a/fs/jfs/jfs_dmap.c
+++ b/fs/jfs/jfs_dmap.c
@@ -3055,7 +3055,7 @@ static int cntlz(u32 value)
  * RETURN VALUES:
  *      log2 number of blocks
  */
-int blkstol2(s64 nb)
+static int blkstol2(s64 nb)
 {
 	int l2nb;
 	s64 mask;		/* meant to be signed */
diff --git a/fs/jfs/jfs_txnmgr.c b/fs/jfs/jfs_txnmgr.c
index c7a92f9..9b71ed2 100644
--- a/fs/jfs/jfs_txnmgr.c
+++ b/fs/jfs/jfs_txnmgr.c
@@ -725,6 +725,9 @@ struct tlock *txLock(tid_t tid, struct inode *ip, struct metapage * mp,
 	else
 		tlck->flag = tlckINODELOCK;
 
+	if (S_ISDIR(ip->i_mode))
+		tlck->flag |= tlckDIRECTORY;
+
 	tlck->type = 0;
 
 	/* bind the tlock and the page */
@@ -1009,6 +1012,8 @@ struct tlock *txMaplock(tid_t tid, struct inode *ip, int type)
 
 	/* bind the tlock and the object */
 	tlck->flag = tlckINODELOCK;
+	if (S_ISDIR(ip->i_mode))
+		tlck->flag |= tlckDIRECTORY;
 	tlck->ip = ip;
 	tlck->mp = NULL;
 
@@ -1077,6 +1082,8 @@ struct linelock *txLinelock(struct linelock * tlock)
 	linelock->flag = tlckLINELOCK;
 	linelock->maxcnt = TLOCKLONG;
 	linelock->index = 0;
+	if (tlck->flag & tlckDIRECTORY)
+		linelock->flag |= tlckDIRECTORY;
 
 	/* append linelock after tlock */
 	linelock->next = tlock->next;
@@ -2070,8 +2077,8 @@ static void xtLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd,
  *
  * function:    log from maplock of freed data extents;
  */
-void mapLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd,
-	    struct tlock * tlck)
+static void mapLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd,
+		   struct tlock * tlck)
 {
 	struct pxd_lock *pxdlock;
 	int i, nlock;
@@ -2209,7 +2216,7 @@ void txEA(tid_t tid, struct inode *ip, dxd_t * oldea, dxd_t * newea)
  * function: synchronously write pages locked by transaction
  *              after txLog() but before txUpdateMap();
  */
-void txForce(struct tblock * tblk)
+static void txForce(struct tblock * tblk)
 {
 	struct tlock *tlck;
 	lid_t lid, next;
@@ -2358,7 +2365,7 @@ static void txUpdateMap(struct tblock * tblk)
 			 */
 			else {	/* (maplock->flag & mlckFREE) */
 
-				if (S_ISDIR(tlck->ip->i_mode))
+				if (tlck->flag & tlckDIRECTORY)
 					txFreeMap(ipimap, maplock,
 						  tblk, COMMIT_PWMAP);
 				else
diff --git a/fs/jfs/jfs_txnmgr.h b/fs/jfs/jfs_txnmgr.h
index 59ad0f6..0e4dc45 100644
--- a/fs/jfs/jfs_txnmgr.h
+++ b/fs/jfs/jfs_txnmgr.h
@@ -122,6 +122,7 @@ extern struct tlock *TxLock;	/* transaction lock table */
 #define tlckLOG			0x0800
 /* updateMap state */
 #define	tlckUPDATEMAP		0x0080
+#define	tlckDIRECTORY		0x0040
 /* freeLock state */
 #define tlckFREELOCK		0x0008
 #define tlckWRITEPAGE		0x0004
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index 6ceb1d4..9758ebd 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -184,14 +184,13 @@ static void nfs_readpage_release(struct nfs_page *req)
 {
 	unlock_page(req->wb_page);
 
-	nfs_clear_request(req);
-	nfs_release_request(req);
-
 	dprintk("NFS: read done (%s/%Ld %d@%Ld)\n",
 			req->wb_context->dentry->d_inode->i_sb->s_id,
 			(long long)NFS_FILEID(req->wb_context->dentry->d_inode),
 			req->wb_bytes,
 			(long long)req_offset(req));
+	nfs_clear_request(req);
+	nfs_release_request(req);
 }
 
 /*
diff --git a/fs/ntfs/ChangeLog b/fs/ntfs/ChangeLog
index 49eafbd..c7e9237 100644
--- a/fs/ntfs/ChangeLog
+++ b/fs/ntfs/ChangeLog
@@ -92,6 +92,8 @@ ToDo/Notes:
 	  an octal number to conform to how chmod(1) works, too.  Thanks to
 	  Giuseppe Bilotta and Horst von Brand for pointing out the errors of
 	  my ways.
+	- Fix various bugs in the runlist merging code.  (Based on libntfs
+	  changes by Richard Russon.)
 
 2.1.23 - Implement extension of resident files and make writing safe as well as
 	 many bug fixes, cleanups, and enhancements...
diff --git a/fs/ntfs/aops.c b/fs/ntfs/aops.c
index b6cc8cf..5e80c07 100644
--- a/fs/ntfs/aops.c
+++ b/fs/ntfs/aops.c
@@ -59,39 +59,49 @@ static void ntfs_end_buffer_async_read(struct buffer_head *bh, int uptodate)
 	unsigned long flags;
 	struct buffer_head *first, *tmp;
 	struct page *page;
+	struct inode *vi;
 	ntfs_inode *ni;
 	int page_uptodate = 1;
 
 	page = bh->b_page;
-	ni = NTFS_I(page->mapping->host);
+	vi = page->mapping->host;
+	ni = NTFS_I(vi);
 
 	if (likely(uptodate)) {
-		s64 file_ofs, initialized_size;
+		loff_t i_size;
+		s64 file_ofs, init_size;
 
 		set_buffer_uptodate(bh);
 
 		file_ofs = ((s64)page->index << PAGE_CACHE_SHIFT) +
 				bh_offset(bh);
 		read_lock_irqsave(&ni->size_lock, flags);
-		initialized_size = ni->initialized_size;
+		init_size = ni->initialized_size;
+		i_size = i_size_read(vi);
 		read_unlock_irqrestore(&ni->size_lock, flags);
+		if (unlikely(init_size > i_size)) {
+			/* Race with shrinking truncate. */
+			init_size = i_size;
+		}
 		/* Check for the current buffer head overflowing. */
-		if (file_ofs + bh->b_size > initialized_size) {
-			char *addr;
-			int ofs = 0;
-
-			if (file_ofs < initialized_size)
-				ofs = initialized_size - file_ofs;
-			addr = kmap_atomic(page, KM_BIO_SRC_IRQ);
-			memset(addr + bh_offset(bh) + ofs, 0, bh->b_size - ofs);
+		if (unlikely(file_ofs + bh->b_size > init_size)) {
+			u8 *kaddr;
+			int ofs;
+
+			ofs = 0;
+			if (file_ofs < init_size)
+				ofs = init_size - file_ofs;
+			kaddr = kmap_atomic(page, KM_BIO_SRC_IRQ);
+			memset(kaddr + bh_offset(bh) + ofs, 0,
+					bh->b_size - ofs);
+			kunmap_atomic(kaddr, KM_BIO_SRC_IRQ);
 			flush_dcache_page(page);
-			kunmap_atomic(addr, KM_BIO_SRC_IRQ);
 		}
 	} else {
 		clear_buffer_uptodate(bh);
 		SetPageError(page);
-		ntfs_error(ni->vol->sb, "Buffer I/O error, logical block %llu.",
-				(unsigned long long)bh->b_blocknr);
+		ntfs_error(ni->vol->sb, "Buffer I/O error, logical block "
+				"0x%llx.", (unsigned long long)bh->b_blocknr);
 	}
 	first = page_buffers(page);
 	local_irq_save(flags);
@@ -124,7 +134,7 @@ static void ntfs_end_buffer_async_read(struct buffer_head *bh, int uptodate)
 		if (likely(page_uptodate && !PageError(page)))
 			SetPageUptodate(page);
 	} else {
-		char *addr;
+		u8 *kaddr;
 		unsigned int i, recs;
 		u32 rec_size;
 
@@ -132,12 +142,12 @@ static void ntfs_end_buffer_async_read(struct buffer_head *bh, int uptodate)
 		recs = PAGE_CACHE_SIZE / rec_size;
 		/* Should have been verified before we got here... */
 		BUG_ON(!recs);
-		addr = kmap_atomic(page, KM_BIO_SRC_IRQ);
+		kaddr = kmap_atomic(page, KM_BIO_SRC_IRQ);
 		for (i = 0; i < recs; i++)
-			post_read_mst_fixup((NTFS_RECORD*)(addr +
+			post_read_mst_fixup((NTFS_RECORD*)(kaddr +
 					i * rec_size), rec_size);
+		kunmap_atomic(kaddr, KM_BIO_SRC_IRQ);
 		flush_dcache_page(page);
-		kunmap_atomic(addr, KM_BIO_SRC_IRQ);
 		if (likely(page_uptodate && !PageError(page)))
 			SetPageUptodate(page);
 	}
@@ -168,8 +178,11 @@ still_busy:
  */
 static int ntfs_read_block(struct page *page)
 {
+	loff_t i_size;
 	VCN vcn;
 	LCN lcn;
+	s64 init_size;
+	struct inode *vi;
 	ntfs_inode *ni;
 	ntfs_volume *vol;
 	runlist_element *rl;
@@ -180,7 +193,8 @@ static int ntfs_read_block(struct page *page)
 	int i, nr;
 	unsigned char blocksize_bits;
 
-	ni = NTFS_I(page->mapping->host);
+	vi = page->mapping->host;
+	ni = NTFS_I(vi);
 	vol = ni->vol;
 
 	/* $MFT/$DATA must have its complete runlist in memory at all times. */
@@ -199,11 +213,28 @@ static int ntfs_read_block(struct page *page)
 	bh = head = page_buffers(page);
 	BUG_ON(!bh);
 
+	/*
+	 * We may be racing with truncate.  To avoid some of the problems we
+	 * now take a snapshot of the various sizes and use those for the whole
+	 * of the function.  In case of an extending truncate it just means we
+	 * may leave some buffers unmapped which are now allocated.  This is
+	 * not a problem since these buffers will just get mapped when a write
+	 * occurs.  In case of a shrinking truncate, we will detect this later
+	 * on due to the runlist being incomplete and if the page is being
+	 * fully truncated, truncate will throw it away as soon as we unlock
+	 * it so no need to worry what we do with it.
+	 */
 	iblock = (s64)page->index << (PAGE_CACHE_SHIFT - blocksize_bits);
 	read_lock_irqsave(&ni->size_lock, flags);
 	lblock = (ni->allocated_size + blocksize - 1) >> blocksize_bits;
-	zblock = (ni->initialized_size + blocksize - 1) >> blocksize_bits;
+	init_size = ni->initialized_size;
+	i_size = i_size_read(vi);
 	read_unlock_irqrestore(&ni->size_lock, flags);
+	if (unlikely(init_size > i_size)) {
+		/* Race with shrinking truncate. */
+		init_size = i_size;
+	}
+	zblock = (init_size + blocksize - 1) >> blocksize_bits;
 
 	/* Loop through all the buffers in the page. */
 	rl = NULL;
@@ -366,6 +397,8 @@ handle_zblock:
  */
 static int ntfs_readpage(struct file *file, struct page *page)
 {
+	loff_t i_size;
+	struct inode *vi;
 	ntfs_inode *ni, *base_ni;
 	u8 *kaddr;
 	ntfs_attr_search_ctx *ctx;
@@ -384,14 +417,17 @@ retry_readpage:
 		unlock_page(page);
 		return 0;
 	}
-	ni = NTFS_I(page->mapping->host);
+	vi = page->mapping->host;
+	ni = NTFS_I(vi);
 	/*
 	 * Only $DATA attributes can be encrypted and only unnamed $DATA
 	 * attributes can be compressed.  Index root can have the flags set but
 	 * this means to create compressed/encrypted files, not that the
-	 * attribute is compressed/encrypted.
+	 * attribute is compressed/encrypted.  Note we need to check for
+	 * AT_INDEX_ALLOCATION since this is the type of both directory and
+	 * index inodes.
 	 */
-	if (ni->type != AT_INDEX_ROOT) {
+	if (ni->type != AT_INDEX_ALLOCATION) {
 		/* If attribute is encrypted, deny access, just like NT4. */
 		if (NInoEncrypted(ni)) {
 			BUG_ON(ni->type != AT_DATA);
@@ -456,7 +492,12 @@ retry_readpage:
 	read_lock_irqsave(&ni->size_lock, flags);
 	if (unlikely(attr_len > ni->initialized_size))
 		attr_len = ni->initialized_size;
+	i_size = i_size_read(vi);
 	read_unlock_irqrestore(&ni->size_lock, flags);
+	if (unlikely(attr_len > i_size)) {
+		/* Race with shrinking truncate. */
+		attr_len = i_size;
+	}
 	kaddr = kmap_atomic(page, KM_USER0);
 	/* Copy the data to the page. */
 	memcpy(kaddr, (u8*)ctx->attr +
@@ -1341,9 +1382,11 @@ retry_writepage:
 	 * Only $DATA attributes can be encrypted and only unnamed $DATA
 	 * attributes can be compressed.  Index root can have the flags set but
 	 * this means to create compressed/encrypted files, not that the
-	 * attribute is compressed/encrypted.
+	 * attribute is compressed/encrypted.  Note we need to check for
+	 * AT_INDEX_ALLOCATION since this is the type of both directory and
+	 * index inodes.
 	 */
-	if (ni->type != AT_INDEX_ROOT) {
+	if (ni->type != AT_INDEX_ALLOCATION) {
 		/* If file is encrypted, deny access, just like NT4. */
 		if (NInoEncrypted(ni)) {
 			unlock_page(page);
@@ -1379,8 +1422,8 @@ retry_writepage:
 			unsigned int ofs = i_size & ~PAGE_CACHE_MASK;
 			kaddr = kmap_atomic(page, KM_USER0);
 			memset(kaddr + ofs, 0, PAGE_CACHE_SIZE - ofs);
-			flush_dcache_page(page);
 			kunmap_atomic(kaddr, KM_USER0);
+			flush_dcache_page(page);
 		}
 		/* Handle mst protected attributes. */
 		if (NInoMstProtected(ni))
@@ -1443,34 +1486,33 @@ retry_writepage:
 	BUG_ON(PageWriteback(page));
 	set_page_writeback(page);
 	unlock_page(page);
-	/*
-	 * Here, we do not need to zero the out of bounds area everytime
-	 * because the below memcpy() already takes care of the
-	 * mmap-at-end-of-file requirements.  If the file is converted to a
-	 * non-resident one, then the code path use is switched to the
-	 * non-resident one where the zeroing happens on each ntfs_writepage()
-	 * invocation.
-	 */
 	attr_len = le32_to_cpu(ctx->attr->data.resident.value_length);
 	i_size = i_size_read(vi);
 	if (unlikely(attr_len > i_size)) {
+		/* Race with shrinking truncate or a failed truncate. */
 		attr_len = i_size;
-		ctx->attr->data.resident.value_length = cpu_to_le32(attr_len);
+		/*
+		 * If the truncate failed, fix it up now.  If a concurrent
+		 * truncate, we do its job, so it does not have to do anything.
+		 */
+		err = ntfs_resident_attr_value_resize(ctx->mrec, ctx->attr,
+				attr_len);
+		/* Shrinking cannot fail. */
+		BUG_ON(err);
 	}
 	kaddr = kmap_atomic(page, KM_USER0);
 	/* Copy the data from the page to the mft record. */
 	memcpy((u8*)ctx->attr +
 			le16_to_cpu(ctx->attr->data.resident.value_offset),
 			kaddr, attr_len);
-	flush_dcache_mft_record_page(ctx->ntfs_ino);
 	/* Zero out of bounds area in the page cache page. */
 	memset(kaddr + attr_len, 0, PAGE_CACHE_SIZE - attr_len);
-	flush_dcache_page(page);
 	kunmap_atomic(kaddr, KM_USER0);
-
+	flush_dcache_mft_record_page(ctx->ntfs_ino);
+	flush_dcache_page(page);
+	/* We are done with the page. */
 	end_page_writeback(page);
-
-	/* Mark the mft record dirty, so it gets written back. */
+	/* Finally, mark the mft record dirty, so it gets written back. */
 	mark_mft_record_dirty(ctx->ntfs_ino);
 	ntfs_attr_put_search_ctx(ctx);
 	unmap_mft_record(base_ni);
diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c
index dc4bbe3..7ec0451 100644
--- a/fs/ntfs/inode.c
+++ b/fs/ntfs/inode.c
@@ -1166,6 +1166,8 @@ err_out:
  *
  * Return 0 on success and -errno on error.  In the error case, the inode will
  * have had make_bad_inode() executed on it.
+ *
+ * Note this cannot be called for AT_INDEX_ALLOCATION.
  */
 static int ntfs_read_locked_attr_inode(struct inode *base_vi, struct inode *vi)
 {
@@ -1242,8 +1244,8 @@ static int ntfs_read_locked_attr_inode(struct inode *base_vi, struct inode *vi)
 			}
 		}
 		/*
-		 * The encryption flag set in an index root just means to
-		 * compress all files.
+		 * The compressed/sparse flag set in an index root just means
+		 * to compress all files.
 		 */
 		if (NInoMstProtected(ni) && ni->type != AT_INDEX_ROOT) {
 			ntfs_error(vi->i_sb, "Found mst protected attribute "
@@ -1319,8 +1321,7 @@ static int ntfs_read_locked_attr_inode(struct inode *base_vi, struct inode *vi)
 					"the mapping pairs array.");
 			goto unm_err_out;
 		}
-		if ((NInoCompressed(ni) || NInoSparse(ni)) &&
-				ni->type != AT_INDEX_ROOT) {
+		if (NInoCompressed(ni) || NInoSparse(ni)) {
 			if (a->data.non_resident.compression_unit != 4) {
 				ntfs_error(vi->i_sb, "Found nonstandard "
 						"compression unit (%u instead "
diff --git a/fs/ntfs/malloc.h b/fs/ntfs/malloc.h
index 3288bcc..006946e 100644
--- a/fs/ntfs/malloc.h
+++ b/fs/ntfs/malloc.h
@@ -1,7 +1,7 @@
 /*
  * malloc.h - NTFS kernel memory handling. Part of the Linux-NTFS project.
  *
- * Copyright (c) 2001-2004 Anton Altaparmakov
+ * Copyright (c) 2001-2005 Anton Altaparmakov
  *
  * This program/include file is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public License as published
diff --git a/fs/ntfs/runlist.c b/fs/ntfs/runlist.c
index f5b2ac9..061b5ff 100644
--- a/fs/ntfs/runlist.c
+++ b/fs/ntfs/runlist.c
@@ -2,7 +2,7 @@
  * runlist.c - NTFS runlist handling code.  Part of the Linux-NTFS project.
  *
  * Copyright (c) 2001-2005 Anton Altaparmakov
- * Copyright (c) 2002 Richard Russon
+ * Copyright (c) 2002-2005 Richard Russon
  *
  * This program/include file is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public License as published
@@ -158,17 +158,21 @@ static inline BOOL ntfs_are_rl_mergeable(runlist_element *dst,
 	BUG_ON(!dst);
 	BUG_ON(!src);
 
-	if ((dst->lcn < 0) || (src->lcn < 0)) {   /* Are we merging holes? */
-		if (dst->lcn == LCN_HOLE && src->lcn == LCN_HOLE)
-			return TRUE;
+	/* We can merge unmapped regions even if they are misaligned. */
+	if ((dst->lcn == LCN_RL_NOT_MAPPED) && (src->lcn == LCN_RL_NOT_MAPPED))
+		return TRUE;
+	/* If the runs are misaligned, we cannot merge them. */
+	if ((dst->vcn + dst->length) != src->vcn)
 		return FALSE;
-	}
-	if ((dst->lcn + dst->length) != src->lcn) /* Are the runs contiguous? */
-		return FALSE;
-	if ((dst->vcn + dst->length) != src->vcn) /* Are the runs misaligned? */
-		return FALSE;
-
-	return TRUE;
+	/* If both runs are non-sparse and contiguous, we can merge them. */
+	if ((dst->lcn >= 0) && (src->lcn >= 0) &&
+			((dst->lcn + dst->length) == src->lcn))
+		return TRUE;
+	/* If we are merging two holes, we can merge them. */
+	if ((dst->lcn == LCN_HOLE) && (src->lcn == LCN_HOLE))
+		return TRUE;
+	/* Cannot merge. */
+	return FALSE;
 }
 
 /**
@@ -214,14 +218,15 @@ static inline void __ntfs_rl_merge(runlist_element *dst, runlist_element *src)
 static inline runlist_element *ntfs_rl_append(runlist_element *dst,
 		int dsize, runlist_element *src, int ssize, int loc)
 {
-	BOOL right;
-	int magic;
+	BOOL right = FALSE;	/* Right end of @src needs merging. */
+	int marker;		/* End of the inserted runs. */
 
 	BUG_ON(!dst);
 	BUG_ON(!src);
 
 	/* First, check if the right hand end needs merging. */
-	right = ntfs_are_rl_mergeable(src + ssize - 1, dst + loc + 1);
+	if ((loc + 1) < dsize)
+		right = ntfs_are_rl_mergeable(src + ssize - 1, dst + loc + 1);
 
 	/* Space required: @dst size + @src size, less one if we merged. */
 	dst = ntfs_rl_realloc(dst, dsize, dsize + ssize - right);
@@ -236,18 +241,19 @@ static inline runlist_element *ntfs_rl_append(runlist_element *dst,
 	if (right)
 		__ntfs_rl_merge(src + ssize - 1, dst + loc + 1);
 
-	magic = loc + ssize;
+	/* First run after the @src runs that have been inserted. */
+	marker = loc + ssize + 1;
 
 	/* Move the tail of @dst out of the way, then copy in @src. */
-	ntfs_rl_mm(dst, magic + 1, loc + 1 + right, dsize - loc - 1 - right);
+	ntfs_rl_mm(dst, marker, loc + 1 + right, dsize - (loc + 1 + right));
 	ntfs_rl_mc(dst, loc + 1, src, 0, ssize);
 
 	/* Adjust the size of the preceding hole. */
 	dst[loc].length = dst[loc + 1].vcn - dst[loc].vcn;
 
 	/* We may have changed the length of the file, so fix the end marker */
-	if (dst[magic + 1].lcn == LCN_ENOENT)
-		dst[magic + 1].vcn = dst[magic].vcn + dst[magic].length;
+	if (dst[marker].lcn == LCN_ENOENT)
+		dst[marker].vcn = dst[marker - 1].vcn + dst[marker - 1].length;
 
 	return dst;
 }
@@ -279,18 +285,17 @@ static inline runlist_element *ntfs_rl_append(runlist_element *dst,
 static inline runlist_element *ntfs_rl_insert(runlist_element *dst,
 		int dsize, runlist_element *src, int ssize, int loc)
 {
-	BOOL left = FALSE;
-	BOOL disc = FALSE;	/* Discontinuity */
-	BOOL hole = FALSE;	/* Following a hole */
-	int magic;
+	BOOL left = FALSE;	/* Left end of @src needs merging. */
+	BOOL disc = FALSE;	/* Discontinuity between @dst and @src. */
+	int marker;		/* End of the inserted runs. */
 
 	BUG_ON(!dst);
 	BUG_ON(!src);
 
-	/* disc => Discontinuity between the end of @dst and the start of @src.
-	 *	   This means we might need to insert a hole.
-	 * hole => @dst ends with a hole or an unmapped region which we can
-	 *	   extend to match the discontinuity. */
+	/*
+	 * disc => Discontinuity between the end of @dst and the start of @src.
+	 *	   This means we might need to insert a "not mapped" run.
+	 */
 	if (loc == 0)
 		disc = (src[0].vcn > 0);
 	else {
@@ -303,58 +308,49 @@ static inline runlist_element *ntfs_rl_insert(runlist_element *dst,
 			merged_length += src->length;
 
 		disc = (src[0].vcn > dst[loc - 1].vcn + merged_length);
-		if (disc)
-			hole = (dst[loc - 1].lcn == LCN_HOLE);
 	}
-
-	/* Space required: @dst size + @src size, less one if we merged, plus
-	 * one if there was a discontinuity, less one for a trailing hole. */
-	dst = ntfs_rl_realloc(dst, dsize, dsize + ssize - left + disc - hole);
+	/*
+	 * Space required: @dst size + @src size, less one if we merged, plus
+	 * one if there was a discontinuity.
+	 */
+	dst = ntfs_rl_realloc(dst, dsize, dsize + ssize - left + disc);
 	if (IS_ERR(dst))
 		return dst;
 	/*
 	 * We are guaranteed to succeed from here so can start modifying the
 	 * original runlist.
 	 */
-
 	if (left)
 		__ntfs_rl_merge(dst + loc - 1, src);
-
-	magic = loc + ssize - left + disc - hole;
+	/*
+	 * First run after the @src runs that have been inserted.
+	 * Nominally,  @marker equals @loc + @ssize, i.e. location + number of
+	 * runs in @src.  However, if @left, then the first run in @src has
+	 * been merged with one in @dst.  And if @disc, then @dst and @src do
+	 * not meet and we need an extra run to fill the gap.
+	 */
+	marker = loc + ssize - left + disc;
 
 	/* Move the tail of @dst out of the way, then copy in @src. */
-	ntfs_rl_mm(dst, magic, loc, dsize - loc);
-	ntfs_rl_mc(dst, loc + disc - hole, src, left, ssize - left);
+	ntfs_rl_mm(dst, marker, loc, dsize - loc);
+	ntfs_rl_mc(dst, loc + disc, src, left, ssize - left);
 
-	/* Adjust the VCN of the last run ... */
-	if (dst[magic].lcn <= LCN_HOLE)
-		dst[magic].vcn = dst[magic - 1].vcn + dst[magic - 1].length;
+	/* Adjust the VCN of the first run after the insertion... */
+	dst[marker].vcn = dst[marker - 1].vcn + dst[marker - 1].length;
 	/* ... and the length. */
-	if (dst[magic].lcn == LCN_HOLE || dst[magic].lcn == LCN_RL_NOT_MAPPED)
-		dst[magic].length = dst[magic + 1].vcn - dst[magic].vcn;
+	if (dst[marker].lcn == LCN_HOLE || dst[marker].lcn == LCN_RL_NOT_MAPPED)
+		dst[marker].length = dst[marker + 1].vcn - dst[marker].vcn;
 
-	/* Writing beyond the end of the file and there's a discontinuity. */
+	/* Writing beyond the end of the file and there is a discontinuity. */
 	if (disc) {
-		if (hole)
-			dst[loc - 1].length = dst[loc].vcn - dst[loc - 1].vcn;
-		else {
-			if (loc > 0) {
-				dst[loc].vcn = dst[loc - 1].vcn +
-						dst[loc - 1].length;
-				dst[loc].length = dst[loc + 1].vcn -
-						dst[loc].vcn;
-			} else {
-				dst[loc].vcn = 0;
-				dst[loc].length = dst[loc + 1].vcn;
-			}
-			dst[loc].lcn = LCN_RL_NOT_MAPPED;
+		if (loc > 0) {
+			dst[loc].vcn = dst[loc - 1].vcn + dst[loc - 1].length;
+			dst[loc].length = dst[loc + 1].vcn - dst[loc].vcn;
+		} else {
+			dst[loc].vcn = 0;
+			dst[loc].length = dst[loc + 1].vcn;
 		}
-
-		magic += hole;
-
-		if (dst[magic].lcn == LCN_ENOENT)
-			dst[magic].vcn = dst[magic - 1].vcn +
-					dst[magic - 1].length;
+		dst[loc].lcn = LCN_RL_NOT_MAPPED;
 	}
 	return dst;
 }
@@ -385,20 +381,23 @@ static inline runlist_element *ntfs_rl_insert(runlist_element *dst,
 static inline runlist_element *ntfs_rl_replace(runlist_element *dst,
 		int dsize, runlist_element *src, int ssize, int loc)
 {
-	BOOL left = FALSE;
-	BOOL right;
-	int magic;
+	BOOL left = FALSE;	/* Left end of @src needs merging. */
+	BOOL right = FALSE;	/* Right end of @src needs merging. */
+	int tail;		/* Start of tail of @dst. */
+	int marker;		/* End of the inserted runs. */
 
 	BUG_ON(!dst);
 	BUG_ON(!src);
 
-	/* First, merge the left and right ends, if necessary. */
-	right = ntfs_are_rl_mergeable(src + ssize - 1, dst + loc + 1);
+	/* First, see if the left and right ends need merging. */
+	if ((loc + 1) < dsize)
+		right = ntfs_are_rl_mergeable(src + ssize - 1, dst + loc + 1);
 	if (loc > 0)
 		left = ntfs_are_rl_mergeable(dst + loc - 1, src);
-
-	/* Allocate some space. We'll need less if the left, right, or both
-	 * ends were merged. */
+	/*
+	 * Allocate some space.  We will need less if the left, right, or both
+	 * ends get merged.
+	 */
 	dst = ntfs_rl_realloc(dst, dsize, dsize + ssize - left - right);
 	if (IS_ERR(dst))
 		return dst;
@@ -406,21 +405,37 @@ static inline runlist_element *ntfs_rl_replace(runlist_element *dst,
 	 * We are guaranteed to succeed from here so can start modifying the
 	 * original runlists.
 	 */
+
+	/* First, merge the left and right ends, if necessary. */
 	if (right)
 		__ntfs_rl_merge(src + ssize - 1, dst + loc + 1);
 	if (left)
 		__ntfs_rl_merge(dst + loc - 1, src);
-
-	/* FIXME: What does this mean? (AIA) */
-	magic = loc + ssize - left;
+	/*
+	 * Offset of the tail of @dst.  This needs to be moved out of the way
+	 * to make space for the runs to be copied from @src, i.e. the first
+	 * run of the tail of @dst.
+	 * Nominally, @tail equals @loc + 1, i.e. location, skipping the
+	 * replaced run.  However, if @right, then one of @dst's runs is
+	 * already merged into @src.
+	 */
+	tail = loc + right + 1;
+	/*
+	 * First run after the @src runs that have been inserted, i.e. where
+	 * the tail of @dst needs to be moved to.
+	 * Nominally, @marker equals @loc + @ssize, i.e. location + number of
+	 * runs in @src.  However, if @left, then the first run in @src has
+	 * been merged with one in @dst.
+	 */
+	marker = loc + ssize - left;
 
 	/* Move the tail of @dst out of the way, then copy in @src. */
-	ntfs_rl_mm(dst, magic, loc + right + 1, dsize - loc - right - 1);
+	ntfs_rl_mm(dst, marker, tail, dsize - tail);
 	ntfs_rl_mc(dst, loc, src, left, ssize - left);
 
-	/* We may have changed the length of the file, so fix the end marker */
-	if (dst[magic].lcn == LCN_ENOENT)
-		dst[magic].vcn = dst[magic - 1].vcn + dst[magic - 1].length;
+	/* We may have changed the length of the file, so fix the end marker. */
+	if (dsize - tail > 0 && dst[marker].lcn == LCN_ENOENT)
+		dst[marker].vcn = dst[marker - 1].vcn + dst[marker - 1].length;
 	return dst;
 }
 
diff --git a/fs/proc/base.c b/fs/proc/base.c
index 23db452..3b33f94 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -340,6 +340,54 @@ static int proc_root_link(struct inode *inode, struct dentry **dentry, struct vf
 	return result;
 }
 
+
+/* Same as proc_root_link, but this addionally tries to get fs from other
+ * threads in the group */
+static int proc_task_root_link(struct inode *inode, struct dentry **dentry,
+				struct vfsmount **mnt)
+{
+	struct fs_struct *fs;
+	int result = -ENOENT;
+	struct task_struct *leader = proc_task(inode);
+
+	task_lock(leader);
+	fs = leader->fs;
+	if (fs) {
+		atomic_inc(&fs->count);
+		task_unlock(leader);
+	} else {
+		/* Try to get fs from other threads */
+		task_unlock(leader);
+		read_lock(&tasklist_lock);
+		if (pid_alive(leader)) {
+			struct task_struct *task = leader;
+
+			while ((task = next_thread(task)) != leader) {
+				task_lock(task);
+				fs = task->fs;
+				if (fs) {
+					atomic_inc(&fs->count);
+					task_unlock(task);
+					break;
+				}
+				task_unlock(task);
+			}
+		}
+		read_unlock(&tasklist_lock);
+	}
+
+	if (fs) {
+		read_lock(&fs->lock);
+		*mnt = mntget(fs->rootmnt);
+		*dentry = dget(fs->root);
+		read_unlock(&fs->lock);
+		result = 0;
+		put_fs_struct(fs);
+	}
+	return result;
+}
+
+
 #define MAY_PTRACE(task) \
 	(task == current || \
 	(task->parent == current && \
@@ -471,14 +519,14 @@ static int proc_oom_score(struct task_struct *task, char *buffer)
 
 /* permission checks */
 
-static int proc_check_root(struct inode *inode)
+/* If the process being read is separated by chroot from the reading process,
+ * don't let the reader access the threads.
+ */
+static int proc_check_chroot(struct dentry *root, struct vfsmount *vfsmnt)
 {
-	struct dentry *de, *base, *root;
-	struct vfsmount *our_vfsmnt, *vfsmnt, *mnt;
+	struct dentry *de, *base;
+	struct vfsmount *our_vfsmnt, *mnt;
 	int res = 0;
-
-	if (proc_root_link(inode, &root, &vfsmnt)) /* Ewww... */
-		return -ENOENT;
 	read_lock(&current->fs->lock);
 	our_vfsmnt = mntget(current->fs->rootmnt);
 	base = dget(current->fs->root);
@@ -511,6 +559,16 @@ out:
 	goto exit;
 }
 
+static int proc_check_root(struct inode *inode)
+{
+	struct dentry *root;
+	struct vfsmount *vfsmnt;
+
+	if (proc_root_link(inode, &root, &vfsmnt)) /* Ewww... */
+		return -ENOENT;
+	return proc_check_chroot(root, vfsmnt);
+}
+
 static int proc_permission(struct inode *inode, int mask, struct nameidata *nd)
 {
 	if (generic_permission(inode, mask, NULL) != 0)
@@ -518,6 +576,20 @@ static int proc_permission(struct inode *inode, int mask, struct nameidata *nd)
 	return proc_check_root(inode);
 }
 
+static int proc_task_permission(struct inode *inode, int mask, struct nameidata *nd)
+{
+	struct dentry *root;
+	struct vfsmount *vfsmnt;
+
+	if (generic_permission(inode, mask, NULL) != 0)
+		return -EACCES;
+
+	if (proc_task_root_link(inode, &root, &vfsmnt))
+		return -ENOENT;
+
+	return proc_check_chroot(root, vfsmnt);
+}
+
 extern struct seq_operations proc_pid_maps_op;
 static int maps_open(struct inode *inode, struct file *file)
 {
@@ -1419,7 +1491,7 @@ static struct inode_operations proc_fd_inode_operations = {
 
 static struct inode_operations proc_task_inode_operations = {
 	.lookup		= proc_task_lookup,
-	.permission	= proc_permission,
+	.permission	= proc_task_permission,
 };
 
 #ifdef CONFIG_SECURITY