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

31 files changed, 305 insertions, 145 deletions

diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h
index 9c09641ce90728..fca83e28edcf67 100644
--- a/fs/autofs4/autofs_i.h
+++ b/fs/autofs4/autofs_i.h
@@ -92,6 +92,7 @@ struct autofs_wait_queue {
 
 struct autofs_sb_info {
 	u32 magic;
+	struct dentry *root;
 	struct file *pipe;
 	pid_t oz_pgrp;
 	int catatonic;
diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c
index 4bb14cc680407e..0a3c05d101679e 100644
--- a/fs/autofs4/inode.c
+++ b/fs/autofs4/inode.c
@@ -16,6 +16,7 @@
 #include <linux/pagemap.h>
 #include <linux/parser.h>
 #include <linux/bitops.h>
+#include <linux/smp_lock.h>
 #include "autofs_i.h"
 #include <linux/module.h>
 
@@ -76,6 +77,66 @@ void autofs4_free_ino(struct autofs_info *ino)
 	kfree(ino);
 }
 
+/*
+ * Deal with the infamous "Busy inodes after umount ..." message.
+ *
+ * Clean up the dentry tree. This happens with autofs if the user
+ * space program goes away due to a SIGKILL, SIGSEGV etc.
+ */
+static void autofs4_force_release(struct autofs_sb_info *sbi)
+{
+	struct dentry *this_parent = sbi->root;
+	struct list_head *next;
+
+	spin_lock(&dcache_lock);
+repeat:
+	next = this_parent->d_subdirs.next;
+resume:
+	while (next != &this_parent->d_subdirs) {
+		struct dentry *dentry = list_entry(next, struct dentry, d_child);
+
+		/* Negative dentry - don`t care */
+		if (!simple_positive(dentry)) {
+			next = next->next;
+			continue;
+		}
+
+		if (!list_empty(&dentry->d_subdirs)) {
+			this_parent = dentry;
+			goto repeat;
+		}
+
+		next = next->next;
+		spin_unlock(&dcache_lock);
+
+		DPRINTK("dentry %p %.*s",
+			dentry, (int)dentry->d_name.len, dentry->d_name.name);
+
+		dput(dentry);
+		spin_lock(&dcache_lock);
+	}
+
+	if (this_parent != sbi->root) {
+		struct dentry *dentry = this_parent;
+
+		next = this_parent->d_child.next;
+		this_parent = this_parent->d_parent;
+		spin_unlock(&dcache_lock);
+		DPRINTK("parent dentry %p %.*s",
+			dentry, (int)dentry->d_name.len, dentry->d_name.name);
+		dput(dentry);
+		spin_lock(&dcache_lock);
+		goto resume;
+	}
+	spin_unlock(&dcache_lock);
+
+	dput(sbi->root);
+	sbi->root = NULL;
+	shrink_dcache_sb(sbi->sb);
+
+	return;
+}
+
 static void autofs4_put_super(struct super_block *sb)
 {
 	struct autofs_sb_info *sbi = autofs4_sbi(sb);
@@ -85,6 +146,10 @@ static void autofs4_put_super(struct super_block *sb)
 	if ( !sbi->catatonic )
 		autofs4_catatonic_mode(sbi); /* Free wait queues, close pipe */
 
+	/* Clean up and release dangling references */
+	if (sbi)
+		autofs4_force_release(sbi);
+
 	kfree(sbi);
 
 	DPRINTK("shutting down");
@@ -199,6 +264,7 @@ int autofs4_fill_super(struct super_block *s, void *data, int silent)
 
 	s->s_fs_info = sbi;
 	sbi->magic = AUTOFS_SBI_MAGIC;
+	sbi->root = NULL;
 	sbi->catatonic = 0;
 	sbi->exp_timeout = 0;
 	sbi->oz_pgrp = process_group(current);
@@ -267,6 +333,13 @@ int autofs4_fill_super(struct super_block *s, void *data, int silent)
 	sbi->pipe = pipe;
 
 	/*
+	 * Take a reference to the root dentry so we get a chance to
+	 * clean up the dentry tree on umount.
+	 * See autofs4_force_release.
+	 */
+	sbi->root = dget(root);
+
+	/*
 	 * Success! Install the root dentry now to indicate completion.
 	 */
 	s->s_root = root;
diff --git a/fs/bio.c b/fs/bio.c
index ca8f7a850fe30a..249dd6bb66c843 100644
--- a/fs/bio.c
+++ b/fs/bio.c
@@ -261,6 +261,7 @@ inline void __bio_clone(struct bio *bio, struct bio *bio_src)
 	 */
 	bio->bi_vcnt = bio_src->bi_vcnt;
 	bio->bi_size = bio_src->bi_size;
+	bio->bi_idx = bio_src->bi_idx;
 	bio_phys_segments(q, bio);
 	bio_hw_segments(q, bio);
 }
diff --git a/fs/ext2/ialloc.c b/fs/ext2/ialloc.c
index 77e05914921242..161f156d98c863 100644
--- a/fs/ext2/ialloc.c
+++ b/fs/ext2/ialloc.c
@@ -612,6 +612,7 @@ got:
 	err = ext2_init_acl(inode, dir);
 	if (err) {
 		DQUOT_FREE_INODE(inode);
+		DQUOT_DROP(inode);
 		goto fail2;
 	}
 	mark_inode_dirty(inode);
diff --git a/fs/ext2/xattr.c b/fs/ext2/xattr.c
index 27982b500e84a9..0099462d427100 100644
--- a/fs/ext2/xattr.c
+++ b/fs/ext2/xattr.c
@@ -823,7 +823,7 @@ cleanup:
 void
 ext2_xattr_put_super(struct super_block *sb)
 {
-	mb_cache_shrink(ext2_xattr_cache, sb->s_bdev);
+	mb_cache_shrink(sb->s_bdev);
 }
 
 
diff --git a/fs/ext2/xip.c b/fs/ext2/xip.c
index 0aa5ac159c0940..ca7f00312388ef 100644
--- a/fs/ext2/xip.c
+++ b/fs/ext2/xip.c
@@ -36,7 +36,7 @@ __ext2_get_sector(struct inode *inode, sector_t offset, int create,
 	*result = tmp.b_blocknr;
 
 	/* did we get a sparse block (hole in the file)? */
-	if (!(*result)) {
+	if (!tmp.b_blocknr && !rc) {
 		BUG_ON(create);
 		rc = -ENODATA;
 	}
diff --git a/fs/ext3/ialloc.c b/fs/ext3/ialloc.c
index 1e6f3ea2871341..6981bd014ede9f 100644
--- a/fs/ext3/ialloc.c
+++ b/fs/ext3/ialloc.c
@@ -604,12 +604,14 @@ got:
 	err = ext3_init_acl(handle, inode, dir);
 	if (err) {
 		DQUOT_FREE_INODE(inode);
+		DQUOT_DROP(inode);
 		goto fail2;
   	}
 	err = ext3_mark_inode_dirty(handle, inode);
 	if (err) {
 		ext3_std_error(sb, err);
 		DQUOT_FREE_INODE(inode);
+		DQUOT_DROP(inode);
 		goto fail2;
 	}
 
diff --git a/fs/ext3/xattr.c b/fs/ext3/xattr.c
index 3f9dfa643b197d..269c7b92db9ada 100644
--- a/fs/ext3/xattr.c
+++ b/fs/ext3/xattr.c
@@ -1106,7 +1106,7 @@ cleanup:
 void
 ext3_xattr_put_super(struct super_block *sb)
 {
-	mb_cache_shrink(ext3_xattr_cache, sb->s_bdev);
+	mb_cache_shrink(sb->s_bdev);
 }
 
 /*
diff --git a/fs/fcntl.c b/fs/fcntl.c
index 286a9f8f3d4988..6fbc9d8fcc3621 100644
--- a/fs/fcntl.c
+++ b/fs/fcntl.c
@@ -288,7 +288,7 @@ static long do_fcntl(int fd, unsigned int cmd, unsigned long arg,
 		break;
 	case F_SETLK:
 	case F_SETLKW:
-		err = fcntl_setlk(filp, cmd, (struct flock __user *) arg);
+		err = fcntl_setlk(fd, filp, cmd, (struct flock __user *) arg);
 		break;
 	case F_GETOWN:
 		/*
@@ -376,7 +376,8 @@ asmlinkage long sys_fcntl64(unsigned int fd, unsigned int cmd, unsigned long arg
 			break;
 		case F_SETLK64:
 		case F_SETLKW64:
-			err = fcntl_setlk64(filp, cmd, (struct flock64 __user *) arg);
+			err = fcntl_setlk64(fd, filp, cmd,
+					(struct flock64 __user *) arg);
 			break;
 		default:
 			err = do_fcntl(fd, cmd, arg, filp);
diff --git a/fs/hfs/bnode.c b/fs/hfs/bnode.c
index 6ad1211f84edb0..a096c5a5666442 100644
--- a/fs/hfs/bnode.c
+++ b/fs/hfs/bnode.c
@@ -480,6 +480,8 @@ void hfs_bnode_put(struct hfs_bnode *node)
 			return;
 		}
 		for (i = 0; i < tree->pages_per_bnode; i++) {
+			if (!node->page[i])
+				continue;
 			mark_page_accessed(node->page[i]);
 #if REF_PAGES
 			put_page(node->page[i]);
diff --git a/fs/hfs/extent.c b/fs/hfs/extent.c
index cbc8510ad22212..5ea6b3d45eaa60 100644
--- a/fs/hfs/extent.c
+++ b/fs/hfs/extent.c
@@ -482,7 +482,8 @@ void hfs_file_truncate(struct inode *inode)
 		page_cache_release(page);
 		mark_inode_dirty(inode);
 		return;
-	}
+	} else if (inode->i_size == HFS_I(inode)->phys_size)
+		return;
 	size = inode->i_size + HFS_SB(sb)->alloc_blksz - 1;
 	blk_cnt = size / HFS_SB(sb)->alloc_blksz;
 	alloc_cnt = HFS_I(inode)->alloc_blocks;
diff --git a/fs/hfsplus/bnode.c b/fs/hfsplus/bnode.c
index 267872e84d714d..8868d3b766fd46 100644
--- a/fs/hfsplus/bnode.c
+++ b/fs/hfsplus/bnode.c
@@ -643,6 +643,8 @@ void hfs_bnode_put(struct hfs_bnode *node)
 			return;
 		}
 		for (i = 0; i < tree->pages_per_bnode; i++) {
+			if (!node->page[i])
+				continue;
 			mark_page_accessed(node->page[i]);
 #if REF_PAGES
 			put_page(node->page[i]);
diff --git a/fs/hfsplus/extents.c b/fs/hfsplus/extents.c
index 376498cc64fddb..e7235ca79a9528 100644
--- a/fs/hfsplus/extents.c
+++ b/fs/hfsplus/extents.c
@@ -461,7 +461,9 @@ void hfsplus_file_truncate(struct inode *inode)
 		page_cache_release(page);
 		mark_inode_dirty(inode);
 		return;
-	}
+	} else if (inode->i_size == HFSPLUS_I(inode).phys_size)
+		return;
+
 	blk_cnt = (inode->i_size + HFSPLUS_SB(sb).alloc_blksz - 1) >> HFSPLUS_SB(sb).alloc_blksz_shift;
 	alloc_cnt = HFSPLUS_I(inode).alloc_blocks;
 	if (blk_cnt == alloc_cnt)
diff --git a/fs/hostfs/hostfs.h b/fs/hostfs/hostfs.h
index c1516d013bf637..67bca0d4a33ba5 100644
--- a/fs/hostfs/hostfs.h
+++ b/fs/hostfs/hostfs.h
@@ -69,6 +69,7 @@ extern int read_file(int fd, unsigned long long *offset, char *buf, int len);
 extern int write_file(int fd, unsigned long long *offset, const char *buf,
 		      int len);
 extern int lseek_file(int fd, long long offset, int whence);
+extern int fsync_file(int fd, int datasync);
 extern int file_create(char *name, int ur, int uw, int ux, int gr,
 		       int gw, int gx, int or, int ow, int ox);
 extern int set_attr(const char *file, struct hostfs_iattr *attrs);
diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c
index 88e68caa378467..b2d18200a003f0 100644
--- a/fs/hostfs/hostfs_kern.c
+++ b/fs/hostfs/hostfs_kern.c
@@ -382,7 +382,7 @@ int hostfs_file_open(struct inode *ino, struct file *file)
 
 int hostfs_fsync(struct file *file, struct dentry *dentry, int datasync)
 {
-	return(0);
+	return fsync_file(HOSTFS_I(dentry->d_inode)->fd, datasync);
 }
 
 static struct file_operations hostfs_file_fops = {
diff --git a/fs/hostfs/hostfs_user.c b/fs/hostfs/hostfs_user.c
index 4796e8490f7d47..b97809deba66ce 100644
--- a/fs/hostfs/hostfs_user.c
+++ b/fs/hostfs/hostfs_user.c
@@ -153,10 +153,24 @@ int lseek_file(int fd, long long offset, int whence)
 	int ret;
 
 	ret = lseek64(fd, offset, whence);
-	if(ret < 0) return(-errno);
+	if(ret < 0)
+		return(-errno);
 	return(0);
 }
 
+int fsync_file(int fd, int datasync)
+{
+	int ret;
+	if (datasync)
+		ret = fdatasync(fd);
+	else
+		ret = fsync(fd);
+
+	if (ret < 0)
+		return -errno;
+	return 0;
+}
+
 void close_file(void *stream)
 {
 	close(*((int *) stream));
diff --git a/fs/inotify.c b/fs/inotify.c
index 54757be888b679..27ebcac5e07ff8 100644
--- a/fs/inotify.c
+++ b/fs/inotify.c
@@ -62,8 +62,8 @@ int inotify_max_queued_events;
  * Lifetimes of the three main data structures--inotify_device, inode, and
  * inotify_watch--are managed by reference count.
  *
- * inotify_device: Lifetime is from open until release.  Additional references
- * can bump the count via get_inotify_dev() and drop the count via
+ * inotify_device: Lifetime is from inotify_init() until release.  Additional
+ * references can bump the count via get_inotify_dev() and drop the count via
  * put_inotify_dev().
  *
  * inotify_watch: Lifetime is from create_watch() to destory_watch().
@@ -75,7 +75,7 @@ int inotify_max_queued_events;
  */
 
 /*
- * struct inotify_device - represents an open instance of an inotify device
+ * struct inotify_device - represents an inotify instance
  *
  * This structure is protected by the semaphore 'sem'.
  */
@@ -90,6 +90,7 @@ struct inotify_device {
 	unsigned int		queue_size;	/* size of the queue (bytes) */
 	unsigned int		event_count;	/* number of pending events */
 	unsigned int		max_events;	/* maximum number of events */
+	u32			last_wd;	/* the last wd allocated */
 };
 
 /*
@@ -352,7 +353,7 @@ static int inotify_dev_get_wd(struct inotify_device *dev,
 	do {
 		if (unlikely(!idr_pre_get(&dev->idr, GFP_KERNEL)))
 			return -ENOSPC;
-		ret = idr_get_new(&dev->idr, watch, &watch->wd);
+		ret = idr_get_new_above(&dev->idr, watch, dev->last_wd, &watch->wd);
 	} while (ret == -EAGAIN);
 
 	return ret;
@@ -371,7 +372,7 @@ static int find_inode(const char __user *dirname, struct nameidata *nd)
 	/* you can only watch an inode if you have read permissions on it */
 	error = permission(nd->dentry->d_inode, MAY_READ, NULL);
 	if (error) 
-		path_release (nd);
+		path_release(nd);
 	return error;
 }
 
@@ -387,7 +388,8 @@ static struct inotify_watch *create_watch(struct inotify_device *dev,
 	struct inotify_watch *watch;
 	int ret;
 
-	if (atomic_read(&dev->user->inotify_watches) >= inotify_max_user_watches)
+	if (atomic_read(&dev->user->inotify_watches) >=
+			inotify_max_user_watches)
 		return ERR_PTR(-ENOSPC);
 
 	watch = kmem_cache_alloc(watch_cachep, GFP_KERNEL);
@@ -400,6 +402,7 @@ static struct inotify_watch *create_watch(struct inotify_device *dev,
 		return ERR_PTR(ret);
 	}
 
+	dev->last_wd = ret;
 	watch->mask = mask;
 	atomic_set(&watch->count, 0);
 	INIT_LIST_HEAD(&watch->d_list);
@@ -783,15 +786,14 @@ static int inotify_release(struct inode *ignored, struct file *file)
 		inotify_dev_event_dequeue(dev);
 	up(&dev->sem);
 
-	/* free this device: the put matching the get in inotify_open() */
+	/* free this device: the put matching the get in inotify_init() */
 	put_inotify_dev(dev);
 
 	return 0;
 }
 
 /*
- * inotify_ignore - handle the INOTIFY_IGNORE ioctl, asking that a given wd be
- * removed from the device.
+ * inotify_ignore - remove a given wd from this inotify instance.
  *
  * Can sleep.
  */
@@ -856,42 +858,40 @@ asmlinkage long sys_inotify_init(void)
 {
 	struct inotify_device *dev;
 	struct user_struct *user;
-	int ret = -ENOTTY;
-	int fd;
-	struct file *filp;
+	struct file *filp;	
+	int fd, ret;
 
 	fd = get_unused_fd();
-	if (fd < 0) {
-		ret = fd;
-		goto out;
-	}
+	if (fd < 0)
+		return fd;
 
 	filp = get_empty_filp();
 	if (!filp) {
-		put_unused_fd(fd);
 		ret = -ENFILE;
-		goto out;
+		goto out_put_fd;
 	}
-	filp->f_op = &inotify_fops;
-	filp->f_vfsmnt = mntget(inotify_mnt);
-	filp->f_dentry = dget(inotify_mnt->mnt_root);
-	filp->f_mapping = filp->f_dentry->d_inode->i_mapping;
-	filp->f_mode = FMODE_READ;
-	filp->f_flags = O_RDONLY;
 
 	user = get_uid(current->user);
-
-	if (unlikely(atomic_read(&user->inotify_devs) >= inotify_max_user_instances)) {
+	if (unlikely(atomic_read(&user->inotify_devs) >=
+			inotify_max_user_instances)) {
 		ret = -EMFILE;
-		goto out_err;
+		goto out_free_uid;
 	}
 
 	dev = kmalloc(sizeof(struct inotify_device), GFP_KERNEL);
 	if (unlikely(!dev)) {
 		ret = -ENOMEM;
-		goto out_err;
+		goto out_free_uid;
 	}
 
+	filp->f_op = &inotify_fops;
+	filp->f_vfsmnt = mntget(inotify_mnt);
+	filp->f_dentry = dget(inotify_mnt->mnt_root);
+	filp->f_mapping = filp->f_dentry->d_inode->i_mapping;
+	filp->f_mode = FMODE_READ;
+	filp->f_flags = O_RDONLY;
+	filp->private_data = dev;
+
 	idr_init(&dev->idr);
 	INIT_LIST_HEAD(&dev->events);
 	INIT_LIST_HEAD(&dev->watches);
@@ -901,50 +901,55 @@ asmlinkage long sys_inotify_init(void)
 	dev->queue_size = 0;
 	dev->max_events = inotify_max_queued_events;
 	dev->user = user;
+	dev->last_wd = 0;
 	atomic_set(&dev->count, 0);
 
 	get_inotify_dev(dev);
 	atomic_inc(&user->inotify_devs);
+	fd_install(fd, filp);
 
-	filp->private_data = dev;
-	fd_install (fd, filp);
 	return fd;
-out_err:
-	put_unused_fd (fd);
-	put_filp (filp);
+out_free_uid:
 	free_uid(user);
-out:
+	put_filp(filp);
+out_put_fd:
+	put_unused_fd(fd);
 	return ret;
 }
 
-asmlinkage long sys_inotify_add_watch(int fd, const char *path, u32 mask)
+asmlinkage long sys_inotify_add_watch(int fd, const char __user *path, u32 mask)
 {
 	struct inotify_watch *watch, *old;
 	struct inode *inode;
 	struct inotify_device *dev;
 	struct nameidata nd;
 	struct file *filp;
-	int ret;
+	int ret, fput_needed;
 
-	filp = fget(fd);
-	if (!filp)
+	filp = fget_light(fd, &fput_needed);
+	if (unlikely(!filp))
 		return -EBADF;
 
-	dev = filp->private_data;
+	/* verify that this is indeed an inotify instance */
+	if (unlikely(filp->f_op != &inotify_fops)) {
+		ret = -EINVAL;
+		goto fput_and_out;
+	}
 
-	ret = find_inode((const char __user*) path, &nd);
-	if (ret)
+	ret = find_inode(path, &nd);
+	if (unlikely(ret))
 		goto fput_and_out;
 
-	/* Held in place by reference in nd */
+	/* inode held in place by reference to nd; dev by fget on fd */
 	inode = nd.dentry->d_inode;
+	dev = filp->private_data;
 
 	down(&inode->inotify_sem);
 	down(&dev->sem);
 
 	/* don't let user-space set invalid bits: we don't want flags set */
 	mask &= IN_ALL_EVENTS;
-	if (!mask) {
+	if (unlikely(!mask)) {
 		ret = -EINVAL;
 		goto out;
 	}
@@ -971,11 +976,11 @@ asmlinkage long sys_inotify_add_watch(int fd, const char *path, u32 mask)
 	list_add(&watch->i_list, &inode->inotify_watches);
 	ret = watch->wd;
 out:
-	path_release (&nd);
 	up(&dev->sem);
 	up(&inode->inotify_sem);
+	path_release(&nd);
 fput_and_out:
-	fput(filp);
+	fput_light(filp, fput_needed);
 	return ret;
 }
 
@@ -983,15 +988,23 @@ asmlinkage long sys_inotify_rm_watch(int fd, u32 wd)
 {
 	struct file *filp;
 	struct inotify_device *dev;
-	int ret;
+	int ret, fput_needed;
 
-	filp = fget(fd);
-	if (!filp)
+	filp = fget_light(fd, &fput_needed);
+	if (unlikely(!filp))
 		return -EBADF;
+
+	/* verify that this is indeed an inotify instance */
+	if (unlikely(filp->f_op != &inotify_fops)) {
+		ret = -EINVAL;
+		goto out;
+	}
+
 	dev = filp->private_data;
 	ret = inotify_ignore(dev, wd);
-	fput(filp);
 
+out:
+	fput_light(filp, fput_needed);
 	return ret;
 }
 
@@ -1009,17 +1022,24 @@ static struct file_system_type inotify_fs_type = {
 };
 
 /*
- * inotify_init - Our initialization function.  Note that we cannnot return
+ * inotify_setup - Our initialization function.  Note that we cannnot return
  * error because we have compiled-in VFS hooks.  So an (unlikely) failure here
  * must result in panic().
  */
-static int __init inotify_init(void)
+static int __init inotify_setup(void)
 {
-	register_filesystem(&inotify_fs_type);
+	int ret;
+
+	ret = register_filesystem(&inotify_fs_type);
+	if (unlikely(ret))
+		panic("inotify: register_filesystem returned %d!\n", ret);
+
 	inotify_mnt = kern_mount(&inotify_fs_type);
+	if (IS_ERR(inotify_mnt))
+		panic("inotify: kern_mount ret %ld!\n", PTR_ERR(inotify_mnt));
 
-	inotify_max_queued_events = 8192;
-	inotify_max_user_instances = 8;
+	inotify_max_queued_events = 16384;
+	inotify_max_user_instances = 128;
 	inotify_max_user_watches = 8192;
 
 	atomic_set(&inotify_cookie, 0);
@@ -1034,4 +1054,4 @@ static int __init inotify_init(void)
 	return 0;
 }
 
-module_init(inotify_init);
+module_init(inotify_setup);
diff --git a/fs/jffs/intrep.c b/fs/jffs/intrep.c
index fc589ddd0762d6..456d7e6e29c268 100644
--- a/fs/jffs/intrep.c
+++ b/fs/jffs/intrep.c
@@ -3397,6 +3397,9 @@ jffs_garbage_collect_thread(void *ptr)
 			siginfo_t info;
 			unsigned long signr = 0;
 
+			if (try_to_freeze())
+				continue;
+
 			spin_lock_irq(&current->sighand->siglock);
 			signr = dequeue_signal(current, &current->blocked, &info);
 			spin_unlock_irq(&current->sighand->siglock);
diff --git a/fs/jfs/jfs_dmap.c b/fs/jfs/jfs_dmap.c
index 0732f206ca60aa..c739626f5bf118 100644
--- a/fs/jfs/jfs_dmap.c
+++ b/fs/jfs/jfs_dmap.c
@@ -75,7 +75,7 @@ static void dbAllocBits(struct bmap * bmp, struct dmap * dp, s64 blkno,
 			int nblocks);
 static void dbSplit(dmtree_t * tp, int leafno, int splitsz, int newval);
 static void dbBackSplit(dmtree_t * tp, int leafno);
-static void dbJoin(dmtree_t * tp, int leafno, int newval);
+static int dbJoin(dmtree_t * tp, int leafno, int newval);
 static void dbAdjTree(dmtree_t * tp, int leafno, int newval);
 static int dbAdjCtl(struct bmap * bmp, s64 blkno, int newval, int alloc,
 		    int level);
@@ -98,8 +98,8 @@ static int dbExtend(struct inode *ip, s64 blkno, s64 nblocks, s64 addnblocks);
 static int dbFindBits(u32 word, int l2nb);
 static int dbFindCtl(struct bmap * bmp, int l2nb, int level, s64 * blkno);
 static int dbFindLeaf(dmtree_t * tp, int l2nb, int *leafidx);
-static void dbFreeBits(struct bmap * bmp, struct dmap * dp, s64 blkno,
-		       int nblocks);
+static int dbFreeBits(struct bmap * bmp, struct dmap * dp, s64 blkno,
+		      int nblocks);
 static int dbFreeDmap(struct bmap * bmp, struct dmap * dp, s64 blkno,
 		      int nblocks);
 static int dbMaxBud(u8 * cp);
@@ -378,6 +378,7 @@ int dbFree(struct inode *ip, s64 blkno, s64 nblocks)
 
 		/* free the blocks. */
 		if ((rc = dbFreeDmap(bmp, dp, blkno, nb))) {
+			jfs_error(ip->i_sb, "dbFree: error in block map\n");
 			release_metapage(mp);
 			IREAD_UNLOCK(ipbmap);
 			return (rc);
@@ -2020,7 +2021,7 @@ static int dbFreeDmap(struct bmap * bmp, struct dmap * dp, s64 blkno,
 		      int nblocks)
 {
 	s8 oldroot;
-	int rc, word;
+	int rc = 0, word;
 
 	/* save the current value of the root (i.e. maximum free string)
 	 * of the dmap tree.
@@ -2028,11 +2029,11 @@ static int dbFreeDmap(struct bmap * bmp, struct dmap * dp, s64 blkno,
 	oldroot = dp->tree.stree[ROOT];
 
 	/* free the specified (blocks) bits */
-	dbFreeBits(bmp, dp, blkno, nblocks);
+	rc = dbFreeBits(bmp, dp, blkno, nblocks);
 
-	/* if the root has not changed, done. */
-	if (dp->tree.stree[ROOT] == oldroot)
-		return (0);
+	/* if error or the root has not changed, done. */
+	if (rc || (dp->tree.stree[ROOT] == oldroot))
+		return (rc);
 
 	/* root changed. bubble the change up to the dmap control pages.
 	 * if the adjustment of the upper level control pages fails,
@@ -2221,15 +2222,16 @@ static void dbAllocBits(struct bmap * bmp, struct dmap * dp, s64 blkno,
  *      blkno	-  starting block number of the bits to be freed.
  *      nblocks	-  number of bits to be freed.
  *
- * RETURN VALUES: none
+ * RETURN VALUES: 0 for success
  *
  * serialization: IREAD_LOCK(ipbmap) or IWRITE_LOCK(ipbmap) held on entry/exit;
  */
-static void dbFreeBits(struct bmap * bmp, struct dmap * dp, s64 blkno,
+static int dbFreeBits(struct bmap * bmp, struct dmap * dp, s64 blkno,
 		       int nblocks)
 {
 	int dbitno, word, rembits, nb, nwords, wbitno, nw, agno;
 	dmtree_t *tp = (dmtree_t *) & dp->tree;
+	int rc = 0;
 	int size;
 
 	/* determine the bit number and word within the dmap of the
@@ -2278,8 +2280,10 @@ static void dbFreeBits(struct bmap * bmp, struct dmap * dp, s64 blkno,
 
 			/* update the leaf for this dmap word.
 			 */
-			dbJoin(tp, word,
-			       dbMaxBud((u8 *) & dp->wmap[word]));
+			rc = dbJoin(tp, word,
+				    dbMaxBud((u8 *) & dp->wmap[word]));
+			if (rc)
+				return rc;
 
 			word += 1;
 		} else {
@@ -2310,7 +2314,9 @@ static void dbFreeBits(struct bmap * bmp, struct dmap * dp, s64 blkno,
 
 				/* update the leaf.
 				 */
-				dbJoin(tp, word, size);
+				rc = dbJoin(tp, word, size);
+				if (rc)
+					return rc;
 
 				/* get the number of dmap words handled.
 				 */
@@ -2357,6 +2363,8 @@ static void dbFreeBits(struct bmap * bmp, struct dmap * dp, s64 blkno,
 	}
 
 	BMAP_UNLOCK(bmp);
+
+	return 0;
 }
 
 
@@ -2464,7 +2472,9 @@ dbAdjCtl(struct bmap * bmp, s64 blkno, int newval, int alloc, int level)
 		}
 		dbSplit((dmtree_t *) dcp, leafno, dcp->budmin, newval);
 	} else {
-		dbJoin((dmtree_t *) dcp, leafno, newval);
+		rc = dbJoin((dmtree_t *) dcp, leafno, newval);
+		if (rc)
+			return rc;
 	}
 
 	/* check if the root of the current dmap control page changed due
@@ -2689,7 +2699,7 @@ static void dbBackSplit(dmtree_t * tp, int leafno)
  *
  * RETURN VALUES: none
  */
-static void dbJoin(dmtree_t * tp, int leafno, int newval)
+static int dbJoin(dmtree_t * tp, int leafno, int newval)
 {
 	int budsz, buddy;
 	s8 *leaf;
@@ -2729,7 +2739,9 @@ static void dbJoin(dmtree_t * tp, int leafno, int newval)
 			if (newval > leaf[buddy])
 				break;
 
-			assert(newval == leaf[buddy]);
+			/* It shouldn't be less */
+			if (newval < leaf[buddy])
+				return -EIO;
 
 			/* check which (leafno or buddy) is the left buddy.
 			 * the left buddy gets to claim the blocks resulting
@@ -2761,6 +2773,8 @@ static void dbJoin(dmtree_t * tp, int leafno, int newval)
 	/* update the leaf value.
 	 */
 	dbAdjTree(tp, leafno, newval);
+
+	return 0;
 }
 
 
diff --git a/fs/jfs/jfs_dtree.c b/fs/jfs/jfs_dtree.c
index 73b5fc7eda80f4..404f33eae50792 100644
--- a/fs/jfs/jfs_dtree.c
+++ b/fs/jfs/jfs_dtree.c
@@ -381,9 +381,12 @@ static u32 add_index(tid_t tid, struct inode *ip, s64 bn, int slot)
 		 * It's time to move the inline table to an external
 		 * page and begin to build the xtree
 		 */
-		if (DQUOT_ALLOC_BLOCK(ip, sbi->nbperpage) ||
-		    dbAlloc(ip, 0, sbi->nbperpage, &xaddr))
-			goto clean_up;	/* No space */
+		if (DQUOT_ALLOC_BLOCK(ip, sbi->nbperpage))
+			goto clean_up;
+		if (dbAlloc(ip, 0, sbi->nbperpage, &xaddr)) {
+			DQUOT_FREE_BLOCK(ip, sbi->nbperpage);
+			goto clean_up;
+		}
 
 		/*
 		 * Save the table, we're going to overwrite it with the
@@ -397,13 +400,15 @@ static u32 add_index(tid_t tid, struct inode *ip, s64 bn, int slot)
 		xtInitRoot(tid, ip);
 
 		/*
-		 * Allocate the first block & add it to the xtree
+		 * Add the first block to the xtree
 		 */
 		if (xtInsert(tid, ip, 0, 0, sbi->nbperpage, &xaddr, 0)) {
 			/* This really shouldn't fail */
 			jfs_warn("add_index: xtInsert failed!");
 			memcpy(&jfs_ip->i_dirtable, temp_table,
 			       sizeof (temp_table));
+			dbFree(ip, xaddr, sbi->nbperpage);
+			DQUOT_FREE_BLOCK(ip, sbi->nbperpage);
 			goto clean_up;
 		}
 		ip->i_size = PSIZE;
diff --git a/fs/jfs/jfs_logmgr.c b/fs/jfs/jfs_logmgr.c
index 79d07624bfe178..22815e88e7cc0c 100644
--- a/fs/jfs/jfs_logmgr.c
+++ b/fs/jfs/jfs_logmgr.c
@@ -1030,7 +1030,8 @@ static int lmLogSync(struct jfs_log * log, int nosyncwait)
 	 * starting until all current transactions are completed
 	 * by setting syncbarrier flag.
 	 */
-	if (written > LOGSYNC_BARRIER(logsize) && logsize > 32 * LOGPSIZE) {
+	if (!test_bit(log_SYNCBARRIER, &log->flag) &&
+	    (written > LOGSYNC_BARRIER(logsize)) && log->active) {
 		set_bit(log_SYNCBARRIER, &log->flag);
 		jfs_info("log barrier on: lsn=0x%x syncpt=0x%x", lsn,
 			 log->syncpt);
diff --git a/fs/jfs/jfs_metapage.c b/fs/jfs/jfs_metapage.c
index 6c5485d16c3916..13d7e3f1feb4e0 100644
--- a/fs/jfs/jfs_metapage.c
+++ b/fs/jfs/jfs_metapage.c
@@ -561,7 +561,6 @@ static int metapage_releasepage(struct page *page, int gfp_mask)
 			dump_mem("page", page, sizeof(struct page));
 			dump_stack();
 		}
-		WARN_ON(mp->lsn);
 		if (mp->lsn)
 			remove_from_logsync(mp);
 		remove_metapage(page, mp);
@@ -641,7 +640,7 @@ struct metapage *__get_metapage(struct inode *inode, unsigned long lblock,
 	} else {
 		page = read_cache_page(mapping, page_index,
 			    (filler_t *)mapping->a_ops->readpage, NULL);
-		if (IS_ERR(page)) {
+		if (IS_ERR(page) || !PageUptodate(page)) {
 			jfs_err("read_cache_page failed!");
 			return NULL;
 		}
@@ -783,14 +782,6 @@ void release_metapage(struct metapage * mp)
 	if (test_bit(META_discard, &mp->flag) && !mp->count) {
 		clear_page_dirty(page);
 		ClearPageUptodate(page);
-#ifdef _NOT_YET
-		if (page->mapping) {
-		/* Remove from page cache and page cache reference */
-			remove_from_page_cache(page);
-			page_cache_release(page);
-			metapage_releasepage(page, 0);
-		}
-#endif
 	}
 #else
 	/* Try to keep metapages from using up too much memory */
diff --git a/fs/locks.c b/fs/locks.c
index 29fa5da6c1170c..11956b6179ff81 100644
--- a/fs/locks.c
+++ b/fs/locks.c
@@ -1591,7 +1591,8 @@ out:
 /* 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(struct file *filp, unsigned int cmd, struct flock __user *l)
+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;
@@ -1620,6 +1621,7 @@ int fcntl_setlk(struct file *filp, unsigned int cmd, struct flock __user *l)
 		goto out;
 	}
 
+again:
 	error = flock_to_posix_lock(filp, file_lock, &flock);
 	if (error)
 		goto out;
@@ -1648,25 +1650,33 @@ int fcntl_setlk(struct file *filp, unsigned int cmd, struct flock __user *l)
 	if (error)
 		goto out;
 
-	if (filp->f_op && filp->f_op->lock != NULL) {
+	if (filp->f_op && filp->f_op->lock != NULL)
 		error = filp->f_op->lock(filp, cmd, file_lock);
-		goto out;
-	}
+	else {
+		for (;;) {
+			error = __posix_lock_file(inode, file_lock);
+			if ((error != -EAGAIN) || (cmd == F_SETLK))
+				break;
+			error = wait_event_interruptible(file_lock->fl_wait,
+					!file_lock->fl_next);
+			if (!error)
+				continue;
 
-	for (;;) {
-		error = __posix_lock_file(inode, file_lock);
-		if ((error != -EAGAIN) || (cmd == F_SETLK))
+			locks_delete_block(file_lock);
 			break;
-		error = wait_event_interruptible(file_lock->fl_wait,
-				!file_lock->fl_next);
-		if (!error)
-			continue;
+		}
+	}
 
-		locks_delete_block(file_lock);
-		break;
+	/*
+	 * Attempt to detect a close/fcntl race and recover by
+	 * releasing the lock that was just acquired.
+	 */
+	if (!error && fcheck(fd) != filp && flock.l_type != F_UNLCK) {
+		flock.l_type = F_UNLCK;
+		goto again;
 	}
 
- out:
+out:
 	locks_free_lock(file_lock);
 	return error;
 }
@@ -1724,7 +1734,8 @@ out:
 /* 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(struct file *filp, unsigned int cmd, struct flock64 __user *l)
+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;
@@ -1753,6 +1764,7 @@ int fcntl_setlk64(struct file *filp, unsigned int cmd, struct flock64 __user *l)
 		goto out;
 	}
 
+again:
 	error = flock64_to_posix_lock(filp, file_lock, &flock);
 	if (error)
 		goto out;
@@ -1781,22 +1793,30 @@ int fcntl_setlk64(struct file *filp, unsigned int cmd, struct flock64 __user *l)
 	if (error)
 		goto out;
 
-	if (filp->f_op && filp->f_op->lock != NULL) {
+	if (filp->f_op && filp->f_op->lock != NULL)
 		error = filp->f_op->lock(filp, cmd, file_lock);
-		goto out;
-	}
+	else {
+		for (;;) {
+			error = __posix_lock_file(inode, file_lock);
+			if ((error != -EAGAIN) || (cmd == F_SETLK64))
+				break;
+			error = wait_event_interruptible(file_lock->fl_wait,
+					!file_lock->fl_next);
+			if (!error)
+				continue;
 
-	for (;;) {
-		error = __posix_lock_file(inode, file_lock);
-		if ((error != -EAGAIN) || (cmd == F_SETLK64))
+			locks_delete_block(file_lock);
 			break;
-		error = wait_event_interruptible(file_lock->fl_wait,
-				!file_lock->fl_next);
-		if (!error)
-			continue;
+		}
+	}
 
-		locks_delete_block(file_lock);
-		break;
+	/*
+	 * Attempt to detect a close/fcntl race and recover by
+	 * releasing the lock that was just acquired.
+	 */
+	if (!error && fcheck(fd) != filp && flock.l_type != F_UNLCK) {
+		flock.l_type = F_UNLCK;
+		goto again;
 	}
 
 out:
@@ -1888,12 +1908,7 @@ void locks_remove_flock(struct file *filp)
 
 	while ((fl = *before) != NULL) {
 		if (fl->fl_file == filp) {
-			/*
-			 * We might have a POSIX lock that was created at the same time
-			 * the filp was closed for the last time. Just remove that too,
-			 * regardless of ownership, since nobody can own it.
-			 */
-			if (IS_FLOCK(fl) || IS_POSIX(fl)) {
+			if (IS_FLOCK(fl)) {
 				locks_delete_lock(before);
 				continue;
 			}
diff --git a/fs/mbcache.c b/fs/mbcache.c
index c7170b9221a306..b002a088857da5 100644
--- a/fs/mbcache.c
+++ b/fs/mbcache.c
@@ -316,11 +316,10 @@ fail:
  * currently in use cannot be freed, and thus remain in the cache. All others
  * are freed.
  *
- * @cache: which cache to shrink
  * @bdev: which device's cache entries to shrink
  */
 void
-mb_cache_shrink(struct mb_cache *cache, struct block_device *bdev)
+mb_cache_shrink(struct block_device *bdev)
 {
 	LIST_HEAD(free_list);
 	struct list_head *l, *ltmp;
diff --git a/fs/namei.c b/fs/namei.c
index 02a824cd3c5c01..4a27eb79811895 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -2218,7 +2218,7 @@ int vfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		error = vfs_rename_other(old_dir,old_dentry,new_dir,new_dentry);
 	if (!error) {
 		const char *new_name = old_dentry->d_name.name;
-		fsnotify_move(old_dir, new_dir, old_name, new_name, is_dir);
+		fsnotify_move(old_dir, new_dir, old_name, new_name, is_dir, new_dentry->d_inode);
 	}
 	fsnotify_oldname_free(old_name);
 
diff --git a/fs/ntfs/sysctl.h b/fs/ntfs/sysctl.h
index df749cc0aac828..c8064cae8f17ed 100644
--- a/fs/ntfs/sysctl.h
+++ b/fs/ntfs/sysctl.h
@@ -26,7 +26,7 @@
 
 #include <linux/config.h>
 
-#if (DEBUG && CONFIG_SYSCTL)
+#if defined(DEBUG) && defined(CONFIG_SYSCTL)
 
 extern int ntfs_sysctl(int add);
 
diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c
index 1aaf2c7d44e69e..d9f614a57731bd 100644
--- a/fs/reiserfs/inode.c
+++ b/fs/reiserfs/inode.c
@@ -1980,7 +1980,17 @@ int reiserfs_new_inode(struct reiserfs_transaction_handle *th,
       out_inserted_sd:
 	inode->i_nlink = 0;
 	th->t_trans_id = 0;	/* so the caller can't use this handle later */
-	iput(inode);
+
+	/* If we were inheriting an ACL, we need to release the lock so that
+	 * iput doesn't deadlock in reiserfs_delete_xattrs. The locking
+	 * code really needs to be reworked, but this will take care of it
+	 * for now. -jeffm */
+	if (REISERFS_I(dir)->i_acl_default) {
+		reiserfs_write_unlock_xattrs(dir->i_sb);
+		iput(inode);
+		reiserfs_write_lock_xattrs(dir->i_sb);
+	} else
+		iput(inode);
 	return err;
 }
 
diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c
index c66c27ec410056..ca7989b04be349 100644
--- a/fs/reiserfs/journal.c
+++ b/fs/reiserfs/journal.c
@@ -556,14 +556,14 @@ static inline void insert_journal_hash(struct reiserfs_journal_cnode **table,
 }
 
 /* lock the current transaction */
-inline static void lock_journal(struct super_block *p_s_sb)
+static inline void lock_journal(struct super_block *p_s_sb)
 {
 	PROC_INFO_INC(p_s_sb, journal.lock_journal);
 	down(&SB_JOURNAL(p_s_sb)->j_lock);
 }
 
 /* unlock the current transaction */
-inline static void unlock_journal(struct super_block *p_s_sb)
+static inline void unlock_journal(struct super_block *p_s_sb)
 {
 	up(&SB_JOURNAL(p_s_sb)->j_lock);
 }
diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c
index e386d3db30513c..87ac9dc8b381e8 100644
--- a/fs/reiserfs/xattr.c
+++ b/fs/reiserfs/xattr.c
@@ -39,7 +39,6 @@
 #include <linux/xattr.h>
 #include <linux/reiserfs_xattr.h>
 #include <linux/reiserfs_acl.h>
-#include <linux/mbcache.h>
 #include <asm/uaccess.h>
 #include <asm/checksum.h>
 #include <linux/smp_lock.h>
diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c
index 335288b9be0f13..4013d7905e84a6 100644
--- a/fs/sysfs/file.c
+++ b/fs/sysfs/file.c
@@ -437,8 +437,8 @@ int sysfs_chmod_file(struct kobject *kobj, struct attribute *attr, mode_t mode)
 {
 	struct dentry *dir = kobj->dentry;
 	struct dentry *victim;
-	struct sysfs_dirent *sd;
-	umode_t umode = (mode & S_IALLUGO) | S_IFREG;
+	struct inode * inode;
+	struct iattr newattrs;
 	int res = -ENOENT;
 
 	down(&dir->d_inode->i_sem);
@@ -446,13 +446,15 @@ int sysfs_chmod_file(struct kobject *kobj, struct attribute *attr, mode_t mode)
 	if (!IS_ERR(victim)) {
 		if (victim->d_inode &&
 		    (victim->d_parent->d_inode == dir->d_inode)) {
-			sd = victim->d_fsdata;
-			attr->mode = mode;
-			sd->s_mode = umode;
-			victim->d_inode->i_mode = umode;
-			dput(victim);
-			res = 0;
+			inode = victim->d_inode;
+			down(&inode->i_sem);
+			newattrs.ia_mode = (mode & S_IALLUGO) |
+						(inode->i_mode & ~S_IALLUGO);
+			newattrs.ia_valid = ATTR_MODE | ATTR_CTIME;
+			res = notify_change(victim, &newattrs);
+			up(&inode->i_sem);
 		}
+		dput(victim);
 	}
 	up(&dir->d_inode->i_sem);
 
diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c
index 8de13bafaa7673..d727dc960634a2 100644
--- a/fs/sysfs/inode.c
+++ b/fs/sysfs/inode.c
@@ -85,7 +85,7 @@ int sysfs_setattr(struct dentry * dentry, struct iattr * iattr)
 
 		if (!in_group_p(inode->i_gid) && !capable(CAP_FSETID))
 			mode &= ~S_ISGID;
-		sd_iattr->ia_mode = mode;
+		sd_iattr->ia_mode = sd->s_mode = mode;
 	}
 
 	return error;