Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6: (107 commits)
  vfs: use ERR_CAST for err-ptr tossing in lookup_instantiate_filp
  isofs: Remove global fs lock
  jffs2: fix IN_DELETE_SELF on overwriting rename() killing a directory
  fix IN_DELETE_SELF on overwriting rename() on ramfs et.al.
  mm/truncate.c: fix build for CONFIG_BLOCK not enabled
  fs:update the NOTE of the file_operations structure
  Remove dead code in dget_parent()
  AFS: Fix silly characters in a comment
  switch d_add_ci() to d_splice_alias() in "found negative" case as well
  simplify gfs2_lookup()
  jfs_lookup(): don't bother with . or ..
  get rid of useless dget_parent() in btrfs rename() and link()
  get rid of useless dget_parent() in fs/btrfs/ioctl.c
  fs: push i_mutex and filemap_write_and_wait down into ->fsync() handlers
  drivers: fix up various ->llseek() implementations
  fs: handle SEEK_HOLE/SEEK_DATA properly in all fs's that define their own llseek
  Ext4: handle SEEK_HOLE/SEEK_DATA generically
  Btrfs: implement our own ->llseek
  fs: add SEEK_HOLE and SEEK_DATA flags
  reiserfs: make reiserfs default to barrier=flush
  ...

Fix up trivial conflicts in fs/xfs/linux-2.6/xfs_super.c due to the new
shrinker callout for the inode cache, that clashed with the xfs code to
start the periodic workers later.

Author: torvalds

Documentation/filesystems/Locking

@@ -52,7 +52,7 @@ ata *);
 	void (*put_link) (struct dentry *, struct nameidata *, void *);
 	void (*truncate) (struct inode *);
 	int (*permission) (struct inode *, int, unsigned int);
-	int (*check_acl)(struct inode *, int, unsigned int);
+	int (*check_acl)(struct inode *, int);
 	int (*setattr) (struct dentry *, struct iattr *);
 	int (*getattr) (struct vfsmount *, struct dentry *, struct kstat *);
 	int (*setxattr) (struct dentry *, const char *,const void *,size_t,int);
@@ -412,7 +412,7 @@ prototypes:
 	int (*open) (struct inode *, struct file *);
 	int (*flush) (struct file *);
 	int (*release) (struct inode *, struct file *);
-	int (*fsync) (struct file *, int datasync);
+	int (*fsync) (struct file *, loff_t start, loff_t end, int datasync);
 	int (*aio_fsync) (struct kiocb *, int datasync);
 	int (*fasync) (int, struct file *, int);
 	int (*lock) (struct file *, int, struct file_lock *);
@@ -438,9 +438,7 @@ prototypes:
 
 locking rules:
 	All may block except for ->setlease.
-	No VFS locks held on entry except for ->fsync and ->setlease.
-
-->fsync() has i_mutex on inode.
+	No VFS locks held on entry except for ->setlease.
 
 ->setlease has the file_list_lock held and must not sleep.
 

Documentation/filesystems/porting

@@ -398,12 +398,33 @@ Currently you can only have FALLOC_FL_PUNCH_HOLE with FALLOC_FL_KEEP_SIZE set,
 so the i_size should not change when hole punching, even when puching the end of
 a file off.
 
---
-[mandatory]
-
 --
 [mandatory]
 	->get_sb() is gone.  Switch to use of ->mount().  Typically it's just
 a matter of switching from calling get_sb_... to mount_... and changing the
 function type.  If you were doing it manually, just switch from setting ->mnt_root
 to some pointer to returning that pointer.  On errors return ERR_PTR(...).
+
+--
+[mandatory]
+	->permission(), generic_permission() and ->check_acl() have lost flags
+argument; instead of passing IPERM_FLAG_RCU we add MAY_NOT_BLOCK into mask.
+	generic_permission() has also lost the check_acl argument; if you want
+non-NULL to be used for that inode, put it into ->i_op->check_acl.
+
+--
+[mandatory]
+	If you implement your own ->llseek() you must handle SEEK_HOLE and
+SEEK_DATA.  You can hanle this by returning -EINVAL, but it would be nicer to
+support it in some way.  The generic handler assumes that the entire file is
+data and there is a virtual hole at the end of the file.  So if the provided
+offset is less than i_size and SEEK_DATA is specified, return the same offset.
+If the above is true for the offset and you are given SEEK_HOLE, return the end
+of the file.  If the offset is i_size or greater return -ENXIO in either case.
+
+[mandatory]
+	If you have your own ->fsync() you must make sure to call
+filemap_write_and_wait_range() so that all dirty pages are synced out properly.
+You must also keep in mind that ->fsync() is not called with i_mutex held
+anymore, so if you require i_mutex locking you must make sure to take it and
+release it yourself.

Documentation/filesystems/vfs.txt

@@ -229,6 +229,8 @@ struct super_operations {
 
         ssize_t (*quota_read)(struct super_block *, int, char *, size_t, loff_t);
         ssize_t (*quota_write)(struct super_block *, int, const char *, size_t, loff_t);
+	int (*nr_cached_objects)(struct super_block *);
+	void (*free_cached_objects)(struct super_block *, int);
 };
 
 All methods are called without any locks being held, unless otherwise
@@ -301,6 +303,26 @@ or bottom half).
 
   quota_write: called by the VFS to write to filesystem quota file.
 
+  nr_cached_objects: called by the sb cache shrinking function for the
+	filesystem to return the number of freeable cached objects it contains.
+	Optional.
+
+  free_cache_objects: called by the sb cache shrinking function for the
+	filesystem to scan the number of objects indicated to try to free them.
+	Optional, but any filesystem implementing this method needs to also
+	implement ->nr_cached_objects for it to be called correctly.
+
+	We can't do anything with any errors that the filesystem might
+	encountered, hence the void return type. This will never be called if
+	the VM is trying to reclaim under GFP_NOFS conditions, hence this
+	method does not need to handle that situation itself.
+
+	Implementations must include conditional reschedule calls inside any
+	scanning loop that is done. This allows the VFS to determine
+	appropriate scan batch sizes without having to worry about whether
+	implementations will cause holdoff problems due to large scan batch
+	sizes.
+
 Whoever sets up the inode is responsible for filling in the "i_op" field. This
 is a pointer to a "struct inode_operations" which describes the methods that
 can be performed on individual inodes.
@@ -333,8 +355,8 @@ struct inode_operations {
         void * (*follow_link) (struct dentry *, struct nameidata *);
         void (*put_link) (struct dentry *, struct nameidata *, void *);
 	void (*truncate) (struct inode *);
-	int (*permission) (struct inode *, int, unsigned int);
-	int (*check_acl)(struct inode *, int, unsigned int);
+	int (*permission) (struct inode *, int);
+	int (*check_acl)(struct inode *, int);
 	int (*setattr) (struct dentry *, struct iattr *);
 	int (*getattr) (struct vfsmount *mnt, struct dentry *, struct kstat *);
 	int (*setxattr) (struct dentry *, const char *,const void *,size_t,int);
@@ -423,7 +445,7 @@ otherwise noted.
   permission: called by the VFS to check for access rights on a POSIX-like
   	filesystem.
 
-	May be called in rcu-walk mode (flags & IPERM_FLAG_RCU). If in rcu-walk
+	May be called in rcu-walk mode (mask & MAY_NOT_BLOCK). If in rcu-walk
         mode, the filesystem must check the permission without blocking or
 	storing to the inode.
 
@@ -755,7 +777,7 @@ struct file_operations {
 	int (*open) (struct inode *, struct file *);
 	int (*flush) (struct file *);
 	int (*release) (struct inode *, struct file *);
-	int (*fsync) (struct file *, int datasync);
+	int (*fsync) (struct file *, loff_t, loff_t, int datasync);
 	int (*aio_fsync) (struct kiocb *, int datasync);
 	int (*fasync) (int, struct file *, int);
 	int (*lock) (struct file *, int, struct file_lock *);

arch/arm/mach-tegra/clock.c

@@ -585,7 +585,7 @@ static const struct file_operations possible_parents_fops = {
 
 static int clk_debugfs_register_one(struct clk *c)
 {
-	struct dentry *d, *child, *child_tmp;
+	struct dentry *d;
 
 	d = debugfs_create_dir(c->name, clk_debugfs_root);
 	if (!d)
@@ -614,10 +614,7 @@ static int clk_debugfs_register_one(struct clk *c)
 	return 0;
 
 err_out:
-	d = c->dent;
-	list_for_each_entry_safe(child, child_tmp, &d->d_subdirs, d_u.d_child)
-		debugfs_remove(child);
-	debugfs_remove(c->dent);
+	debugfs_remove_recursive(c->dent);
 	return -ENOMEM;
 }
 

arch/arm/mach-ux500/clock.c

@@ -635,16 +635,13 @@ static const struct file_operations set_rate_fops = {
 static struct dentry *clk_debugfs_register_dir(struct clk *c,
 						struct dentry *p_dentry)
 {
-	struct dentry *d, *clk_d, *child, *child_tmp;
-	char s[255];
-	char *p = s;
+	struct dentry *d, *clk_d;
+	const char *p = c->name;
 
-	if (c->name == NULL)
-		p += sprintf(p, "BUG");
-	else
-		p += sprintf(p, "%s", c->name);
+	if (!p)
+		p = "BUG";
 
-	clk_d = debugfs_create_dir(s, p_dentry);
+	clk_d = debugfs_create_dir(p, p_dentry);
 	if (!clk_d)
 		return NULL;
 
@@ -666,24 +663,10 @@ static struct dentry *clk_debugfs_register_dir(struct clk *c,
 	return clk_d;
 
 err_out:
-	d = clk_d;
-	list_for_each_entry_safe(child, child_tmp, &d->d_subdirs, d_u.d_child)
-		debugfs_remove(child);
-	debugfs_remove(clk_d);
+	debugfs_remove_recursive(clk_d);
 	return NULL;
 }
 
-static void clk_debugfs_remove_dir(struct dentry *cdentry)
-{
-	struct dentry *d, *child, *child_tmp;
-
-	d = cdentry;
-	list_for_each_entry_safe(child, child_tmp, &d->d_subdirs, d_u.d_child)
-		debugfs_remove(child);
-	debugfs_remove(cdentry);
-	return ;
-}
-
 static int clk_debugfs_register_one(struct clk *c)
 {
 	struct clk *pa = c->parent_periph;
@@ -700,7 +683,7 @@ static int clk_debugfs_register_one(struct clk *c)
 		c->dent_bus = clk_debugfs_register_dir(c,
 				bpa->dent_bus ? bpa->dent_bus : bpa->dent);
 		if ((!c->dent_bus) &&  (c->dent)) {
-			clk_debugfs_remove_dir(c->dent);
+			debugfs_remove_recursive(c->dent);
 			c->dent = NULL;
 			return -ENOMEM;
 		}

arch/arm/plat-omap/clock.c

@@ -480,13 +480,10 @@ static struct dentry *clk_debugfs_root;
 static int clk_debugfs_register_one(struct clk *c)
 {
 	int err;
-	struct dentry *d, *child, *child_tmp;
+	struct dentry *d;
 	struct clk *pa = c->parent;
-	char s[255];
-	char *p = s;
 
-	p += sprintf(p, "%s", c->name);
-	d = debugfs_create_dir(s, pa ? pa->dent : clk_debugfs_root);
+	d = debugfs_create_dir(c->name, pa ? pa->dent : clk_debugfs_root);
 	if (!d)
 		return -ENOMEM;
 	c->dent = d;
@@ -509,10 +506,7 @@ static int clk_debugfs_register_one(struct clk *c)
 	return 0;
 
 err_out:
-	d = c->dent;
-	list_for_each_entry_safe(child, child_tmp, &d->d_subdirs, d_u.d_child)
-		debugfs_remove(child);
-	debugfs_remove(c->dent);
+	debugfs_remove_recursive(c->dent);
 	return err;
 }
 

arch/arm/plat-samsung/clock.c

@@ -458,7 +458,7 @@ static struct dentry *clk_debugfs_root;
 static int clk_debugfs_register_one(struct clk *c)
 {
 	int err;
-	struct dentry *d, *child, *child_tmp;
+	struct dentry *d;
 	struct clk *pa = c->parent;
 	char s[255];
 	char *p = s;
@@ -488,10 +488,7 @@ static int clk_debugfs_register_one(struct clk *c)
 	return 0;
 
 err_out:
-	d = c->dent;
-	list_for_each_entry_safe(child, child_tmp, &d->d_subdirs, d_u.d_child)
-		debugfs_remove(child);
-	debugfs_remove(c->dent);
+	debugfs_remove_recursive(c->dent);
 	return err;
 }
 

arch/arm/plat-spear/clock.c

@@ -916,7 +916,7 @@ static struct dentry *clk_debugfs_root;
 static int clk_debugfs_register_one(struct clk *c)
 {
 	int err;
-	struct dentry *d, *child;
+	struct dentry *d;
 	struct clk *pa = c->pclk;
 	char s[255];
 	char *p = s;
@@ -951,10 +951,7 @@ static int clk_debugfs_register_one(struct clk *c)
 	return 0;
 
 err_out:
-	d = c->dent;
-	list_for_each_entry(child, &d->d_subdirs, d_u.d_child)
-		debugfs_remove(child);
-	debugfs_remove(c->dent);
+	debugfs_remove_recursive(c->dent);
 	return err;
 }
 

arch/powerpc/platforms/cell/spufs/file.c

@@ -1850,9 +1850,16 @@ static int spufs_mfc_flush(struct file *file, fl_owner_t id)
 	return ret;
 }
 
-static int spufs_mfc_fsync(struct file *file, int datasync)
+static int spufs_mfc_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 {
-	return spufs_mfc_flush(file, NULL);
+	struct inode *inode = file->f_path.dentry->d_inode;
+	int err = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (!err) {
+		mutex_lock(&inode->i_mutex);
+		err = spufs_mfc_flush(file, NULL);
+		mutex_unlock(&inode->i_mutex);
+	}
+	return err;
 }
 
 static int spufs_mfc_fasync(int fd, struct file *file, int on)

arch/powerpc/platforms/cell/spufs/inode.c

@@ -611,49 +611,42 @@ static int spufs_create_gang(struct inode *inode,
 
 static struct file_system_type spufs_type;
 
-long spufs_create(struct nameidata *nd, unsigned int flags, mode_t mode,
-							struct file *filp)
+long spufs_create(struct path *path, struct dentry *dentry,
+		unsigned int flags, mode_t mode, struct file *filp)
 {
-	struct dentry *dentry;
 	int ret;
 
 	ret = -EINVAL;
 	/* check if we are on spufs */
-	if (nd->path.dentry->d_sb->s_type != &spufs_type)
+	if (path->dentry->d_sb->s_type != &spufs_type)
 		goto out;
 
 	/* don't accept undefined flags */
 	if (flags & (~SPU_CREATE_FLAG_ALL))
 		goto out;
 
 	/* only threads can be underneath a gang */
-	if (nd->path.dentry != nd->path.dentry->d_sb->s_root) {
+	if (path->dentry != path->dentry->d_sb->s_root) {
 		if ((flags & SPU_CREATE_GANG) ||
-		    !SPUFS_I(nd->path.dentry->d_inode)->i_gang)
+		    !SPUFS_I(path->dentry->d_inode)->i_gang)
 			goto out;
 	}
 
-	dentry = lookup_create(nd, 1);
-	ret = PTR_ERR(dentry);
-	if (IS_ERR(dentry))
-		goto out_dir;
-
 	mode &= ~current_umask();
 
 	if (flags & SPU_CREATE_GANG)
-		ret = spufs_create_gang(nd->path.dentry->d_inode,
-					 dentry, nd->path.mnt, mode);
+		ret = spufs_create_gang(path->dentry->d_inode,
+					 dentry, path->mnt, mode);
 	else
-		ret = spufs_create_context(nd->path.dentry->d_inode,
-					    dentry, nd->path.mnt, flags, mode,
+		ret = spufs_create_context(path->dentry->d_inode,
+					    dentry, path->mnt, flags, mode,
 					    filp);
 	if (ret >= 0)
-		fsnotify_mkdir(nd->path.dentry->d_inode, dentry);
+		fsnotify_mkdir(path->dentry->d_inode, dentry);
 	return ret;
 
-out_dir:
-	mutex_unlock(&nd->path.dentry->d_inode->i_mutex);
 out:
+	mutex_unlock(&path->dentry->d_inode->i_mutex);
 	return ret;
 }
 

arch/powerpc/platforms/cell/spufs/spufs.h

@@ -248,7 +248,7 @@ extern const struct spufs_tree_descr spufs_dir_debug_contents[];
 /* system call implementation */
 extern struct spufs_calls spufs_calls;
 long spufs_run_spu(struct spu_context *ctx, u32 *npc, u32 *status);
-long spufs_create(struct nameidata *nd, unsigned int flags,
+long spufs_create(struct path *nd, struct dentry *dentry, unsigned int flags,
 			mode_t mode, struct file *filp);
 /* ELF coredump callbacks for writing SPU ELF notes */
 extern int spufs_coredump_extra_notes_size(void);

arch/powerpc/platforms/cell/spufs/syscalls.c

@@ -62,21 +62,17 @@ static long do_spu_run(struct file *filp,
 static long do_spu_create(const char __user *pathname, unsigned int flags,
 		mode_t mode, struct file *neighbor)
 {
-	char *tmp;
+	struct path path;
+	struct dentry *dentry;
 	int ret;
 
-	tmp = getname(pathname);
-	ret = PTR_ERR(tmp);
-	if (!IS_ERR(tmp)) {
-		struct nameidata nd;
-
-		ret = kern_path_parent(tmp, &nd);
-		if (!ret) {
-			nd.flags |= LOOKUP_OPEN | LOOKUP_CREATE;
-			ret = spufs_create(&nd, flags, mode, neighbor);
-			path_put(&nd.path);
-		}
-		putname(tmp);
+	dentry = user_path_create(AT_FDCWD, pathname, &path, 1);
+	ret = PTR_ERR(dentry);
+	if (!IS_ERR(dentry)) {
+		ret = spufs_create(&path, dentry, flags, mode, neighbor);
+		mutex_unlock(&path.dentry->d_inode->i_mutex);
+		dput(dentry);
+		path_put(&path);
 	}
 
 	return ret;