Merge branch 'work.symlinks' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs

Pull vfs RCU symlink updates from Al Viro:
 "Replacement of ->follow_link/->put_link, allowing to stay in RCU mode
  even if the symlink is not an embedded one.

  No changes since the mailbomb on Jan 1"

* 'work.symlinks' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
  switch ->get_link() to delayed_call, kill ->put_link()
  kill free_page_put_link()
  teach nfs_get_link() to work in RCU mode
  teach proc_self_get_link()/proc_thread_self_get_link() to work in RCU mode
  teach shmem_get_link() to work in RCU mode
  teach page_get_link() to work in RCU mode
  replace ->follow_link() with new method that could stay in RCU mode
  don't put symlink bodies in pagecache into highmem
  namei: page_getlink() and page_follow_link_light() are the same thing
  ufs: get rid of ->setattr() for symlinks
  udf: don't duplicate page_symlink_inode_operations
  logfs: don't duplicate page_symlink_inode_operations
  switch befs long symlinks to page_symlink_operations

Documentation/filesystems/Locking

@@ -50,8 +50,7 @@ prototypes:
 	int (*rename2) (struct inode *, struct dentry *,
 			struct inode *, struct dentry *, unsigned int);
 	int (*readlink) (struct dentry *, char __user *,int);
-	const char *(*follow_link) (struct dentry *, void **);
-	void (*put_link) (struct inode *, void *);
+	const char *(*get_link) (struct dentry *, struct inode *, void **);
 	void (*truncate) (struct inode *);
 	int (*permission) (struct inode *, int, unsigned int);
 	int (*get_acl)(struct inode *, int);
@@ -83,8 +82,7 @@ rmdir:		yes (both)	(see below)
 rename:		yes (all)	(see below)
 rename2:	yes (all)	(see below)
 readlink:	no
-follow_link:	no
-put_link:	no
+get_link:	no
 setattr:	yes
 permission:	no (may not block if called in rcu-walk mode)
 get_acl:	no

Documentation/filesystems/porting

@@ -504,3 +504,20 @@ in your dentry operations instead.
 [mandatory]
 	__fd_install() & fd_install() can now sleep. Callers should not
 	hold a spinlock	or other resources that do not allow a schedule.
+--
+[mandatory]
+	any symlink that might use page_follow_link_light/page_put_link() must
+	have inode_nohighmem(inode) called before anything might start playing with
+	its pagecache.
+--
+[mandatory]
+	->follow_link() is replaced with ->get_link(); same API, except that
+		* ->get_link() gets inode as a separate argument
+		* ->get_link() may be called in RCU mode - in that case NULL
+		  dentry is passed
+--
+[mandatory]
+	->get_link() gets struct delayed_call *done now, and should do
+	set_delayed_call() where it used to set *cookie.
+	->put_link() is gone - just give the destructor to set_delayed_call()
+	in ->get_link().

Documentation/filesystems/vfs.txt

@@ -350,8 +350,8 @@ struct inode_operations {
 	int (*rename2) (struct inode *, struct dentry *,
 			struct inode *, struct dentry *, unsigned int);
 	int (*readlink) (struct dentry *, char __user *,int);
-	const char *(*follow_link) (struct dentry *, void **);
-	void (*put_link) (struct inode *, void *);
+	const char *(*get_link) (struct dentry *, struct inode *,
+				 struct delayed_call *);
 	int (*permission) (struct inode *, int);
 	int (*get_acl)(struct inode *, int);
 	int (*setattr) (struct dentry *, struct iattr *);
@@ -434,20 +434,19 @@ otherwise noted.
   readlink: called by the readlink(2) system call. Only required if
 	you want to support reading symbolic links
 
-  follow_link: called by the VFS to follow a symbolic link to the
+  get_link: called by the VFS to follow a symbolic link to the
 	inode it points to.  Only required if you want to support
 	symbolic links.  This method returns the symlink body
 	to traverse (and possibly resets the current position with
 	nd_jump_link()).  If the body won't go away until the inode
 	is gone, nothing else is needed; if it needs to be otherwise
-	pinned, the data needed to release whatever we'd grabbed
-	is to be stored in void * variable passed by address to
-	follow_link() instance.
-
-  put_link: called by the VFS to release resources allocated by
-	follow_link().  The cookie stored by follow_link() is passed
-	to this method as the last parameter; only called when
-	cookie isn't NULL.
+	pinned, arrange for its release by having get_link(..., ..., done)
+	do set_delayed_call(done, destructor, argument).
+	In that case destructor(argument) will be called once VFS is
+	done with the body you've returned.
+	May be called in RCU mode; that is indicated by NULL dentry
+	argument.  If request can't be handled without leaving RCU mode,
+	have it return ERR_PTR(-ECHILD).
 
   permission: called by the VFS to check for access rights on a POSIX-like
   	filesystem.

drivers/staging/lustre/lustre/llite/symlink.c

@@ -118,12 +118,20 @@ failed:
 	return rc;
 }
 
-static const char *ll_follow_link(struct dentry *dentry, void **cookie)
+static void ll_put_link(void *p)
+{
+	ptlrpc_req_finished(p);
+}
+
+static const char *ll_get_link(struct dentry *dentry,
+			       struct inode *inode,
+			       struct delayed_call *done)
 {
-	struct inode *inode = d_inode(dentry);
 	struct ptlrpc_request *request = NULL;
 	int rc;
 	char *symname = NULL;
+	if (!dentry)
+		return ERR_PTR(-ECHILD);
 
 	CDEBUG(D_VFSTRACE, "VFS Op\n");
 	ll_inode_size_lock(inode);
@@ -135,22 +143,16 @@ static const char *ll_follow_link(struct dentry *dentry, void **cookie)
 	}
 
 	/* symname may contain a pointer to the request message buffer,
-	 * we delay request releasing until ll_put_link then.
+	 * we delay request releasing then.
 	 */
-	*cookie = request;
+	set_delayed_call(done, ll_put_link, request);
 	return symname;
 }
 
-static void ll_put_link(struct inode *unused, void *cookie)
-{
-	ptlrpc_req_finished(cookie);
-}
-
 struct inode_operations ll_fast_symlink_inode_operations = {
 	.readlink	= generic_readlink,
 	.setattr	= ll_setattr,
-	.follow_link	= ll_follow_link,
-	.put_link	= ll_put_link,
+	.get_link	= ll_get_link,
 	.getattr	= ll_getattr,
 	.permission	= ll_inode_permission,
 	.setxattr	= ll_setxattr,

fs/9p/vfs_inode.c

@@ -1223,18 +1223,26 @@ ino_t v9fs_qid2ino(struct p9_qid *qid)
 }
 
 /**
- * v9fs_vfs_follow_link - follow a symlink path
+ * v9fs_vfs_get_link - follow a symlink path
  * @dentry: dentry for symlink
- * @cookie: place to pass the data to put_link()
+ * @inode: inode for symlink
+ * @done: delayed call for when we are done with the return value
  */
 
-static const char *v9fs_vfs_follow_link(struct dentry *dentry, void **cookie)
+static const char *v9fs_vfs_get_link(struct dentry *dentry,
+				     struct inode *inode,
+				     struct delayed_call *done)
 {
-	struct v9fs_session_info *v9ses = v9fs_dentry2v9ses(dentry);
-	struct p9_fid *fid = v9fs_fid_lookup(dentry);
+	struct v9fs_session_info *v9ses;
+	struct p9_fid *fid;
 	struct p9_wstat *st;
 	char *res;
 
+	if (!dentry)
+		return ERR_PTR(-ECHILD);
+
+	v9ses = v9fs_dentry2v9ses(dentry);
+	fid = v9fs_fid_lookup(dentry);
 	p9_debug(P9_DEBUG_VFS, "%pd\n", dentry);
 
 	if (IS_ERR(fid))
@@ -1259,7 +1267,8 @@ static const char *v9fs_vfs_follow_link(struct dentry *dentry, void **cookie)
 
 	p9stat_free(st);
 	kfree(st);
-	return *cookie = res;
+	set_delayed_call(done, kfree_link, res);
+	return res;
 }
 
 /**
@@ -1452,8 +1461,7 @@ static const struct inode_operations v9fs_file_inode_operations = {
 
 static const struct inode_operations v9fs_symlink_inode_operations = {
 	.readlink = generic_readlink,
-	.follow_link = v9fs_vfs_follow_link,
-	.put_link = kfree_put_link,
+	.get_link = v9fs_vfs_get_link,
 	.getattr = v9fs_vfs_getattr,
 	.setattr = v9fs_vfs_setattr,
 };

fs/9p/vfs_inode_dotl.c

@@ -899,26 +899,34 @@ error:
 }
 
 /**
- * v9fs_vfs_follow_link_dotl - follow a symlink path
+ * v9fs_vfs_get_link_dotl - follow a symlink path
  * @dentry: dentry for symlink
- * @cookie: place to pass the data to put_link()
+ * @inode: inode for symlink
+ * @done: destructor for return value
  */
 
 static const char *
-v9fs_vfs_follow_link_dotl(struct dentry *dentry, void **cookie)
+v9fs_vfs_get_link_dotl(struct dentry *dentry,
+		       struct inode *inode,
+		       struct delayed_call *done)
 {
-	struct p9_fid *fid = v9fs_fid_lookup(dentry);
+	struct p9_fid *fid;
 	char *target;
 	int retval;
 
+	if (!dentry)
+		return ERR_PTR(-ECHILD);
+
 	p9_debug(P9_DEBUG_VFS, "%pd\n", dentry);
 
+	fid = v9fs_fid_lookup(dentry);
 	if (IS_ERR(fid))
 		return ERR_CAST(fid);
 	retval = p9_client_readlink(fid, &target);
 	if (retval)
 		return ERR_PTR(retval);
-	return *cookie = target;
+	set_delayed_call(done, kfree_link, target);
+	return target;
 }
 
 int v9fs_refresh_inode_dotl(struct p9_fid *fid, struct inode *inode)
@@ -984,8 +992,7 @@ const struct inode_operations v9fs_file_inode_operations_dotl = {
 
 const struct inode_operations v9fs_symlink_inode_operations_dotl = {
 	.readlink = generic_readlink,
-	.follow_link = v9fs_vfs_follow_link_dotl,
-	.put_link = kfree_put_link,
+	.get_link = v9fs_vfs_get_link_dotl,
 	.getattr = v9fs_vfs_getattr_dotl,
 	.setattr = v9fs_vfs_setattr_dotl,
 	.setxattr = generic_setxattr,

fs/affs/inode.c

@@ -140,6 +140,7 @@ struct inode *affs_iget(struct super_block *sb, unsigned long ino)
 		break;
 	case ST_SOFTLINK:
 		inode->i_mode |= S_IFLNK;
+		inode_nohighmem(inode);
 		inode->i_op = &affs_symlink_inode_operations;
 		inode->i_data.a_ops = &affs_symlink_aops;
 		break;

fs/affs/namei.c

@@ -344,6 +344,7 @@ affs_symlink(struct inode *dir, struct dentry *dentry, const char *symname)
 		return -ENOSPC;
 
 	inode->i_op = &affs_symlink_inode_operations;
+	inode_nohighmem(inode);
 	inode->i_data.a_ops = &affs_symlink_aops;
 	inode->i_mode = S_IFLNK | 0777;
 	mode_to_prot(inode);

fs/affs/symlink.c

@@ -14,13 +14,13 @@ static int affs_symlink_readpage(struct file *file, struct page *page)
 {
 	struct buffer_head *bh;
 	struct inode *inode = page->mapping->host;
-	char *link = kmap(page);
+	char *link = page_address(page);
 	struct slink_front *lf;
 	int			 i, j;
 	char			 c;
 	char			 lc;
 
-	pr_debug("follow_link(ino=%lu)\n", inode->i_ino);
+	pr_debug("get_link(ino=%lu)\n", inode->i_ino);
 
 	bh = affs_bread(inode->i_sb, inode->i_ino);
 	if (!bh)
@@ -57,12 +57,10 @@ static int affs_symlink_readpage(struct file *file, struct page *page)
 	link[i] = '\0';
 	affs_brelse(bh);
 	SetPageUptodate(page);
-	kunmap(page);
 	unlock_page(page);
 	return 0;
 fail:
 	SetPageError(page);
-	kunmap(page);
 	unlock_page(page);
 	return -EIO;
 }
@@ -73,7 +71,6 @@ const struct address_space_operations affs_symlink_aops = {
 
 const struct inode_operations affs_symlink_inode_operations = {
 	.readlink	= generic_readlink,
-	.follow_link	= page_follow_link_light,
-	.put_link	= page_put_link,
+	.get_link	= page_get_link,
 	.setattr	= affs_notify_change,
 };

fs/afs/inode.c

@@ -56,6 +56,7 @@ static int afs_inode_map_status(struct afs_vnode *vnode, struct key *key)
 	case AFS_FTYPE_SYMLINK:
 		inode->i_mode	= S_IFLNK | vnode->status.mode;
 		inode->i_op	= &page_symlink_inode_operations;
+		inode_nohighmem(inode);
 		break;
 	default:
 		printk("kAFS: AFS vnode with undefined type\n");

fs/autofs4/symlink.c

@@ -12,10 +12,16 @@
 
 #include "autofs_i.h"
 
-static const char *autofs4_follow_link(struct dentry *dentry, void **cookie)
+static const char *autofs4_get_link(struct dentry *dentry,
+				    struct inode *inode,
+				    struct delayed_call *done)
 {
-	struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
-	struct autofs_info *ino = autofs4_dentry_ino(dentry);
+	struct autofs_sb_info *sbi;
+	struct autofs_info *ino;
+	if (!dentry)
+		return ERR_PTR(-ECHILD);
+	sbi = autofs4_sbi(dentry->d_sb);
+	ino = autofs4_dentry_ino(dentry);
 	if (ino && !autofs4_oz_mode(sbi))
 		ino->last_used = jiffies;
 	return d_inode(dentry)->i_private;
@@ -23,5 +29,5 @@ static const char *autofs4_follow_link(struct dentry *dentry, void **cookie)
 
 const struct inode_operations autofs4_symlink_inode_operations = {
 	.readlink	= generic_readlink,
-	.follow_link	= autofs4_follow_link
+	.get_link	= autofs4_get_link
 };

fs/befs/linuxvfs.c

@@ -42,7 +42,7 @@ static struct inode *befs_iget(struct super_block *, unsigned long);
 static struct inode *befs_alloc_inode(struct super_block *sb);
 static void befs_destroy_inode(struct inode *inode);
 static void befs_destroy_inodecache(void);
-static const char *befs_follow_link(struct dentry *, void **);
+static int befs_symlink_readpage(struct file *, struct page *);
 static int befs_utf2nls(struct super_block *sb, const char *in, int in_len,
 			char **out, int *out_len);
 static int befs_nls2utf(struct super_block *sb, const char *in, int in_len,
@@ -79,10 +79,8 @@ static const struct address_space_operations befs_aops = {
 	.bmap		= befs_bmap,
 };
 
-static const struct inode_operations befs_symlink_inode_operations = {
-	.readlink	= generic_readlink,
-	.follow_link	= befs_follow_link,
-	.put_link	= kfree_put_link,
+static const struct address_space_operations befs_symlink_aops = {
+	.readpage	= befs_symlink_readpage,
 };
 
 /* 
@@ -398,7 +396,9 @@ static struct inode *befs_iget(struct super_block *sb, unsigned long ino)
 		inode->i_fop = &befs_dir_operations;
 	} else if (S_ISLNK(inode->i_mode)) {
 		if (befs_ino->i_flags & BEFS_LONG_SYMLINK) {
-			inode->i_op = &befs_symlink_inode_operations;
+			inode->i_op = &page_symlink_inode_operations;
+			inode_nohighmem(inode);
+			inode->i_mapping->a_ops = &befs_symlink_aops;
 		} else {
 			inode->i_link = befs_ino->i_data.symlink;
 			inode->i_op = &simple_symlink_inode_operations;
@@ -463,31 +463,33 @@ befs_destroy_inodecache(void)
  * The data stream become link name. Unless the LONG_SYMLINK
  * flag is set.
  */
-static const char *
-befs_follow_link(struct dentry *dentry, void **cookie)
+static int befs_symlink_readpage(struct file *unused, struct page *page)
 {
-	struct super_block *sb = dentry->d_sb;
-	struct befs_inode_info *befs_ino = BEFS_I(d_inode(dentry));
+	struct inode *inode = page->mapping->host;
+	struct super_block *sb = inode->i_sb;
+	struct befs_inode_info *befs_ino = BEFS_I(inode);
 	befs_data_stream *data = &befs_ino->i_data.ds;
 	befs_off_t len = data->size;
-	char *link;
+	char *link = page_address(page);
 
-	if (len == 0) {
+	if (len == 0 || len > PAGE_SIZE) {
 		befs_error(sb, "Long symlink with illegal length");
-		return ERR_PTR(-EIO);
+		goto fail;
 	}
 	befs_debug(sb, "Follow long symlink");
 
-	link = kmalloc(len, GFP_NOFS);
-	if (!link)
-		return ERR_PTR(-ENOMEM);
 	if (befs_read_lsymlink(sb, data, link, len) != len) {
-		kfree(link);
 		befs_error(sb, "Failed to read entire long symlink");
-		return ERR_PTR(-EIO);
+		goto fail;
 	}
 	link[len - 1] = '\0';
-	return *cookie = link;
+	SetPageUptodate(page);
+	unlock_page(page);
+	return 0;
+fail:
+	SetPageError(page);
+	unlock_page(page);
+	return -EIO;
 }
 
 /*

fs/btrfs/inode.c

@@ -3774,6 +3774,7 @@ cache_acl:
 		break;
 	case S_IFLNK:
 		inode->i_op = &btrfs_symlink_inode_operations;
+		inode_nohighmem(inode);
 		inode->i_mapping->a_ops = &btrfs_symlink_aops;
 		break;
 	default:
@@ -9705,6 +9706,7 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry,
 	btrfs_free_path(path);
 
 	inode->i_op = &btrfs_symlink_inode_operations;
+	inode_nohighmem(inode);
 	inode->i_mapping->a_ops = &btrfs_symlink_aops;
 	inode_set_bytes(inode, name_len);
 	btrfs_i_size_write(inode, name_len);
@@ -10094,8 +10096,7 @@ static const struct inode_operations btrfs_special_inode_operations = {
 };
 static const struct inode_operations btrfs_symlink_inode_operations = {
 	.readlink	= generic_readlink,
-	.follow_link	= page_follow_link_light,
-	.put_link	= page_put_link,
+	.get_link	= page_get_link,
 	.getattr	= btrfs_getattr,
 	.setattr	= btrfs_setattr,
 	.permission	= btrfs_permission,

fs/ceph/inode.c

@@ -1756,7 +1756,7 @@ retry:
  */
 static const struct inode_operations ceph_symlink_iops = {
 	.readlink = generic_readlink,
-	.follow_link = simple_follow_link,
+	.get_link = simple_get_link,
 	.setattr = ceph_setattr,
 	.getattr = ceph_getattr,
 	.setxattr = ceph_setxattr,

fs/cifs/cifsfs.c

@@ -900,8 +900,7 @@ const struct inode_operations cifs_file_inode_ops = {
 
 const struct inode_operations cifs_symlink_inode_ops = {
 	.readlink = generic_readlink,
-	.follow_link = cifs_follow_link,
-	.put_link = kfree_put_link,
+	.get_link = cifs_get_link,
 	.permission = cifs_permission,
 	/* BB add the following two eventually */
 	/* revalidate: cifs_revalidate,

fs/cifs/cifsfs.h

@@ -120,9 +120,8 @@ extern struct vfsmount *cifs_dfs_d_automount(struct path *path);
 #endif
 
 /* Functions related to symlinks */
-extern const char *cifs_follow_link(struct dentry *direntry, void **cookie);
-extern int cifs_readlink(struct dentry *direntry, char __user *buffer,
-			 int buflen);
+extern const char *cifs_get_link(struct dentry *, struct inode *,
+			struct delayed_call *);
 extern int cifs_symlink(struct inode *inode, struct dentry *direntry,
 			const char *symname);
 extern int	cifs_removexattr(struct dentry *, const char *);

fs/cifs/link.c

@@ -627,9 +627,9 @@ cifs_hl_exit:
 }
 
 const char *
-cifs_follow_link(struct dentry *direntry, void **cookie)
+cifs_get_link(struct dentry *direntry, struct inode *inode,
+	      struct delayed_call *done)
 {
-	struct inode *inode = d_inode(direntry);
 	int rc = -ENOMEM;
 	unsigned int xid;
 	char *full_path = NULL;
@@ -639,6 +639,9 @@ cifs_follow_link(struct dentry *direntry, void **cookie)
 	struct cifs_tcon *tcon;
 	struct TCP_Server_Info *server;
 
+	if (!direntry)
+		return ERR_PTR(-ECHILD);
+
 	xid = get_xid();
 
 	tlink = cifs_sb_tlink(cifs_sb);
@@ -678,7 +681,8 @@ cifs_follow_link(struct dentry *direntry, void **cookie)
 		kfree(target_path);
 		return ERR_PTR(rc);
 	}
-	return *cookie = target_path;
+	set_delayed_call(done, kfree_link, target_path);
+	return target_path;
 }
 
 int

fs/coda/cnode.c

@@ -8,6 +8,7 @@
 
 #include <linux/coda.h>
 #include <linux/coda_psdev.h>
+#include <linux/pagemap.h>
 #include "coda_linux.h"
 
 static inline int coda_fideq(struct CodaFid *fid1, struct CodaFid *fid2)
@@ -17,8 +18,7 @@ static inline int coda_fideq(struct CodaFid *fid1, struct CodaFid *fid2)
 
 static const struct inode_operations coda_symlink_inode_operations = {
 	.readlink	= generic_readlink,
-	.follow_link	= page_follow_link_light,
-	.put_link	= page_put_link,
+	.get_link	= page_get_link,
 	.setattr	= coda_setattr,
 };
 
@@ -35,6 +35,7 @@ static void coda_fill_inode(struct inode *inode, struct coda_vattr *attr)
                 inode->i_fop = &coda_dir_operations;
         } else if (S_ISLNK(inode->i_mode)) {
 		inode->i_op = &coda_symlink_inode_operations;
+		inode_nohighmem(inode);
 		inode->i_data.a_ops = &coda_symlink_aops;
 		inode->i_mapping = &inode->i_data;
 	} else

fs/coda/symlink.c

@@ -26,7 +26,7 @@ static int coda_symlink_filler(struct file *file, struct page *page)
 	int error;
 	struct coda_inode_info *cii;
 	unsigned int len = PAGE_SIZE;
-	char *p = kmap(page);
+	char *p = page_address(page);
 
 	cii = ITOC(inode);
 
@@ -34,13 +34,11 @@ static int coda_symlink_filler(struct file *file, struct page *page)
 	if (error)
 		goto fail;
 	SetPageUptodate(page);
-	kunmap(page);
 	unlock_page(page);
 	return 0;
 
 fail:
 	SetPageError(page);
-	kunmap(page);
 	unlock_page(page);
 	return error;
 }

fs/configfs/symlink.c

@@ -279,27 +279,33 @@ static int configfs_getlink(struct dentry *dentry, char * path)
 
 }
 
-static const char *configfs_follow_link(struct dentry *dentry, void **cookie)
+static const char *configfs_get_link(struct dentry *dentry,
+				     struct inode *inode,
+				     struct delayed_call *done)
 {
-	unsigned long page = get_zeroed_page(GFP_KERNEL);
+	char *body;
 	int error;
 
-	if (!page)
+	if (!dentry)
+		return ERR_PTR(-ECHILD);
+
+	body = kzalloc(PAGE_SIZE, GFP_KERNEL);
+	if (!body)
 		return ERR_PTR(-ENOMEM);
 
-	error = configfs_getlink(dentry, (char *)page);
+	error = configfs_getlink(dentry, body);
 	if (!error) {
-		return *cookie = (void *)page;
+		set_delayed_call(done, kfree_link, body);
+		return body;
 	}
 
-	free_page(page);
+	kfree(body);
 	return ERR_PTR(error);
 }
 
 const struct inode_operations configfs_symlink_inode_operations = {
-	.follow_link = configfs_follow_link,
+	.get_link = configfs_get_link,
 	.readlink = generic_readlink,
-	.put_link = free_page_put_link,
 	.setattr = configfs_setattr,
 };