reiserfs: use generic readdir for operations across all xattrs

The current reiserfs xattr implementation open codes reiserfs_readdir
and frees the path before calling the filldir function.  Typically, the
filldir function is something that modifies the file system, such as a
chown or an inode deletion that also require reading of an inode
associated with each direntry.  Since the file system is modified, the
path retained becomes invalid for the next run.  In addition, it runs
backwards in attempt to minimize activity.

This is clearly suboptimal from a code cleanliness perspective as well
as performance-wise.

This patch implements a generic reiserfs_for_each_xattr that uses the
generic readdir and a specific filldir routine that simply populates an
array of dentries and then performs a specific operation on them.  When
all files have been operated on, it then calls the operation on the
directory itself.

The result is a noticable code reduction and better performance.
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

fs/reiserfs/dir.c

@@ -41,10 +41,10 @@ static int reiserfs_dir_fsync(struct file *filp, struct dentry *dentry,
 
 #define store_ih(where,what) copy_item_head (where, what)
 
-//
-static int reiserfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
+int reiserfs_readdir_dentry(struct dentry *dentry, void *dirent,
+			   filldir_t filldir, loff_t *pos)
 {
-	struct inode *inode = filp->f_path.dentry->d_inode;
+	struct inode *inode = dentry->d_inode;
 	struct cpu_key pos_key;	/* key of current position in the directory (key of directory entry) */
 	INITIALIZE_PATH(path_to_entry);
 	struct buffer_head *bh;
@@ -64,13 +64,9 @@ static int reiserfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 
 	/* form key for search the next directory entry using f_pos field of
 	   file structure */
-	make_cpu_key(&pos_key, inode,
-		     (filp->f_pos) ? (filp->f_pos) : DOT_OFFSET, TYPE_DIRENTRY,
-		     3);
+	make_cpu_key(&pos_key, inode, *pos ?: DOT_OFFSET, TYPE_DIRENTRY, 3);
 	next_pos = cpu_key_k_offset(&pos_key);
 
-	/*  reiserfs_warning (inode->i_sb, "reiserfs_readdir 1: f_pos = %Ld", filp->f_pos); */
-
 	path_to_entry.reada = PATH_READA;
 	while (1) {
 	      research:
@@ -144,7 +140,7 @@ static int reiserfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 				/* Ignore the .reiserfs_priv entry */
 				if (reiserfs_xattrs(inode->i_sb) &&
 				    !old_format_only(inode->i_sb) &&
-				    filp->f_path.dentry == inode->i_sb->s_root &&
+				    dentry == inode->i_sb->s_root &&
 				    REISERFS_SB(inode->i_sb)->priv_root &&
 				    REISERFS_SB(inode->i_sb)->priv_root->d_inode
 				    && deh_objectid(deh) ==
@@ -156,7 +152,7 @@ static int reiserfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 				}
 
 				d_off = deh_offset(deh);
-				filp->f_pos = d_off;
+				*pos = d_off;
 				d_ino = deh_objectid(deh);
 				if (d_reclen <= 32) {
 					local_buf = small_buf;
@@ -223,15 +219,21 @@ static int reiserfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 
 	}			/* while */
 
-      end:
-	filp->f_pos = next_pos;
+end:
+	*pos = next_pos;
 	pathrelse(&path_to_entry);
 	reiserfs_check_path(&path_to_entry);
-      out:
+out:
 	reiserfs_write_unlock(inode->i_sb);
 	return ret;
 }
 
+static int reiserfs_readdir(struct file *file, void *dirent, filldir_t filldir)
+{
+	struct dentry *dentry = file->f_path.dentry;
+	return reiserfs_readdir_dentry(dentry, dirent, filldir, &file->f_pos);
+}
+
 /* compose directory item containing "." and ".." entries (entries are
    not aligned to 4 byte boundary) */
 /* the last four params are LE */

fs/reiserfs/xattr.c

@@ -167,218 +167,65 @@ static struct dentry *open_xa_dir(const struct inode *inode, int flags)
 
 }
 
-/*
- * this is very similar to fs/reiserfs/dir.c:reiserfs_readdir, but
- * we need to drop the path before calling the filldir struct.  That
- * would be a big performance hit to the non-xattr case, so I've copied
- * the whole thing for now. --clm
- *
- * the big difference is that I go backwards through the directory,
- * and don't mess with f->f_pos, but the idea is the same.  Do some
- * action on each and every entry in the directory.
- *
- * we're called with i_mutex held, so there are no worries about the directory
- * changing underneath us.
- */
-static int __xattr_readdir(struct inode *inode, void *dirent, filldir_t filldir)
-{
-	struct cpu_key pos_key;	/* key of current position in the directory (key of directory entry) */
-	INITIALIZE_PATH(path_to_entry);
-	struct buffer_head *bh;
-	int entry_num;
-	struct item_head *ih, tmp_ih;
-	int search_res;
-	char *local_buf;
-	loff_t next_pos;
-	char small_buf[32];	/* avoid kmalloc if we can */
-	struct reiserfs_de_head *deh;
-	int d_reclen;
-	char *d_name;
-	off_t d_off;
-	ino_t d_ino;
-	struct reiserfs_dir_entry de;
-
-	/* form key for search the next directory entry using f_pos field of
-	   file structure */
-	next_pos = max_reiserfs_offset(inode);
-
-	while (1) {
-	      research:
-		if (next_pos <= DOT_DOT_OFFSET)
-			break;
-		make_cpu_key(&pos_key, inode, next_pos, TYPE_DIRENTRY, 3);
-
-		search_res =
-		    search_by_entry_key(inode->i_sb, &pos_key, &path_to_entry,
-					&de);
-		if (search_res == IO_ERROR) {
-			// FIXME: we could just skip part of directory which could
-			// not be read
-			pathrelse(&path_to_entry);
-			return -EIO;
-		}
-
-		if (search_res == NAME_NOT_FOUND)
-			de.de_entry_num--;
-
-		set_de_name_and_namelen(&de);
-		entry_num = de.de_entry_num;
-		deh = &(de.de_deh[entry_num]);
-
-		bh = de.de_bh;
-		ih = de.de_ih;
-
-		if (!is_direntry_le_ih(ih)) {
-			reiserfs_error(inode->i_sb, "jdm-20000",
-				       "not direntry %h", ih);
-			break;
-		}
-		copy_item_head(&tmp_ih, ih);
-
-		/* we must have found item, that is item of this directory, */
-		RFALSE(COMP_SHORT_KEYS(&(ih->ih_key), &pos_key),
-		       "vs-9000: found item %h does not match to dir we readdir %K",
-		       ih, &pos_key);
-
-		if (deh_offset(deh) <= DOT_DOT_OFFSET) {
-			break;
-		}
-
-		/* look for the previous entry in the directory */
-		next_pos = deh_offset(deh) - 1;
-
-		if (!de_visible(deh))
-			/* it is hidden entry */
-			continue;
-
-		d_reclen = entry_length(bh, ih, entry_num);
-		d_name = B_I_DEH_ENTRY_FILE_NAME(bh, ih, deh);
-		d_off = deh_offset(deh);
-		d_ino = deh_objectid(deh);
-
-		if (!d_name[d_reclen - 1])
-			d_reclen = strlen(d_name);
-
-		if (d_reclen > REISERFS_MAX_NAME(inode->i_sb->s_blocksize)) {
-			/* too big to send back to VFS */
-			continue;
-		}
-
-		/* Ignore the .reiserfs_priv entry */
-		if (reiserfs_xattrs(inode->i_sb) &&
-		    !old_format_only(inode->i_sb) &&
-		    deh_objectid(deh) ==
-		    le32_to_cpu(INODE_PKEY
-				(REISERFS_SB(inode->i_sb)->priv_root->d_inode)->
-				k_objectid))
-			continue;
-
-		if (d_reclen <= 32) {
-			local_buf = small_buf;
-		} else {
-			local_buf = kmalloc(d_reclen, GFP_NOFS);
-			if (!local_buf) {
-				pathrelse(&path_to_entry);
-				return -ENOMEM;
-			}
-			if (item_moved(&tmp_ih, &path_to_entry)) {
-				kfree(local_buf);
-
-				/* sigh, must retry.  Do this same offset again */
-				next_pos = d_off;
-				goto research;
-			}
-		}
-
-		// Note, that we copy name to user space via temporary
-		// buffer (local_buf) because filldir will block if
-		// user space buffer is swapped out. At that time
-		// entry can move to somewhere else
-		memcpy(local_buf, d_name, d_reclen);
-
-		/* the filldir function might need to start transactions,
-		 * or do who knows what.  Release the path now that we've
-		 * copied all the important stuff out of the deh
-		 */
-		pathrelse(&path_to_entry);
-
-		if (filldir(dirent, local_buf, d_reclen, d_off, d_ino,
-			    DT_UNKNOWN) < 0) {
-			if (local_buf != small_buf) {
-				kfree(local_buf);
-			}
-			goto end;
-		}
-		if (local_buf != small_buf) {
-			kfree(local_buf);
-		}
-	}			/* while */
-
-      end:
-	pathrelse(&path_to_entry);
-	return 0;
-}
-
-/*
- * this could be done with dedicated readdir ops for the xattr files,
- * but I want to get something working asap
- * this is stolen from vfs_readdir
- *
- */
-static
-int xattr_readdir(struct inode *inode, filldir_t filler, void *buf)
-{
-	int res = -ENOENT;
-	if (!IS_DEADDIR(inode)) {
-		lock_kernel();
-		res = __xattr_readdir(inode, buf, filler);
-		unlock_kernel();
-	}
-	return res;
-}
-
 /* The following are side effects of other operations that aren't explicitly
  * modifying extended attributes. This includes operations such as permissions
  * or ownership changes, object deletions, etc. */
+struct reiserfs_dentry_buf {
+	struct dentry *xadir;
+	int count;
+	struct dentry *dentries[8];
+};
 
 static int
-reiserfs_delete_xattrs_filler(void *buf, const char *name, int namelen,
-			      loff_t offset, u64 ino, unsigned int d_type)
+fill_with_dentries(void *buf, const char *name, int namelen, loff_t offset,
+		    u64 ino, unsigned int d_type)
 {
-	struct dentry *xadir = (struct dentry *)buf;
+	struct reiserfs_dentry_buf *dbuf = buf;
 	struct dentry *dentry;
-	int err = 0;
 
-	dentry = lookup_one_len(name, xadir, namelen);
+	if (dbuf->count == ARRAY_SIZE(dbuf->dentries))
+		return -ENOSPC;
+
+	if (name[0] == '.' && (name[1] == '\0' ||
+			       (name[1] == '.' && name[2] == '\0')))
+		return 0;
+
+	dentry = lookup_one_len(name, dbuf->xadir, namelen);
 	if (IS_ERR(dentry)) {
-		err = PTR_ERR(dentry);
-		goto out;
+		return PTR_ERR(dentry);
 	} else if (!dentry->d_inode) {
-		err = -ENODATA;
-		goto out_file;
+		/* A directory entry exists, but no file? */
+		reiserfs_error(dentry->d_sb, "xattr-20003",
+			       "Corrupted directory: xattr %s listed but "
+			       "not found for file %s.\n",
+			       dentry->d_name.name, dbuf->xadir->d_name.name);
+		dput(dentry);
+		return -EIO;
 	}
 
-	/* Skip directories.. */
-	if (S_ISDIR(dentry->d_inode->i_mode))
-		goto out_file;
-
-	err = xattr_unlink(xadir->d_inode, dentry);
-
-out_file:
-	dput(dentry);
+	dbuf->dentries[dbuf->count++] = dentry;
+	return 0;
+}
 
-out:
-	return err;
+static void
+cleanup_dentry_buf(struct reiserfs_dentry_buf *buf)
+{
+	int i;
+	for (i = 0; i < buf->count; i++)
+		if (buf->dentries[i])
+			dput(buf->dentries[i]);
 }
 
-/* This is called w/ inode->i_mutex downed */
-int reiserfs_delete_xattrs(struct inode *inode)
+static int reiserfs_for_each_xattr(struct inode *inode,
+				   int (*action)(struct dentry *, void *),
+				   void *data)
 {
-	int err = -ENODATA;
-	struct dentry *dir, *root;
-	struct reiserfs_transaction_handle th;
-	int blocks = JOURNAL_PER_BALANCE_CNT * 2 + 2 +
-		     4 * REISERFS_QUOTA_TRANS_BLOCKS(inode->i_sb);
+	struct dentry *dir;
+	int i, err = 0;
+	loff_t pos = 0;
+	struct reiserfs_dentry_buf buf = {
+		.count = 0,
+	};
 
 	/* Skip out, an xattr has no xattrs associated with it */
 	if (IS_PRIVATE(inode) || get_inode_sd_version(inode) == STAT_DATA_V1)
@@ -389,117 +236,97 @@ int reiserfs_delete_xattrs(struct inode *inode)
 		err = PTR_ERR(dir);
 		goto out;
 	} else if (!dir->d_inode) {
-		dput(dir);
-		goto out;
+		err = 0;
+		goto out_dir;
 	}
 
 	mutex_lock_nested(&dir->d_inode->i_mutex, I_MUTEX_XATTR);
-	err = xattr_readdir(dir->d_inode, reiserfs_delete_xattrs_filler, dir);
-	mutex_unlock(&dir->d_inode->i_mutex);
-	if (err) {
-		dput(dir);
-		goto out;
+	buf.xadir = dir;
+	err = reiserfs_readdir_dentry(dir, &buf, fill_with_dentries, &pos);
+	while ((err == 0 || err == -ENOSPC) && buf.count) {
+		err = 0;
+
+		for (i = 0; i < buf.count && buf.dentries[i]; i++) {
+			int lerr = 0;
+			struct dentry *dentry = buf.dentries[i];
+
+			if (err == 0 && !S_ISDIR(dentry->d_inode->i_mode))
+				lerr = action(dentry, data);
+
+			dput(dentry);
+			buf.dentries[i] = NULL;
+			err = lerr ?: err;
 		}
+		buf.count = 0;
+		if (!err)
+			err = reiserfs_readdir_dentry(dir, &buf,
+						      fill_with_dentries, &pos);
+	}
+	mutex_unlock(&dir->d_inode->i_mutex);
 
-	root = dget(dir->d_parent);
-	dput(dir);
+	/* Clean up after a failed readdir */
+	cleanup_dentry_buf(&buf);
 
+	if (!err) {
 		/* We start a transaction here to avoid a ABBA situation
 		 * between the xattr root's i_mutex and the journal lock.
-	 * Inode creation will inherit an ACL, which requires a
-	 * lookup. The lookup locks the xattr root i_mutex with a
-	 * transaction open.  Inode deletion takes teh xattr root
-	 * i_mutex to delete the directory and then starts a
-	 * transaction inside it. Boom. This doesn't incur much
-	 * additional overhead since the reiserfs_rmdir transaction
-	 * will just nest inside the outer transaction. */
+		 * This doesn't incur much additional overhead since the
+		 * new transaction will just nest inside the
+		 * outer transaction. */
+		int blocks = JOURNAL_PER_BALANCE_CNT * 2 + 2 +
+			     4 * REISERFS_QUOTA_TRANS_BLOCKS(inode->i_sb);
+		struct reiserfs_transaction_handle th;
 		err = journal_begin(&th, inode->i_sb, blocks);
 		if (!err) {
 			int jerror;
-		mutex_lock_nested(&root->d_inode->i_mutex, I_MUTEX_XATTR);
-		err = xattr_rmdir(root->d_inode, dir);
+			mutex_lock_nested(&dir->d_parent->d_inode->i_mutex,
+					  I_MUTEX_XATTR);
+			err = action(dir, data);
 			jerror = journal_end(&th, inode->i_sb, blocks);
-		mutex_unlock(&root->d_inode->i_mutex);
+			mutex_unlock(&dir->d_parent->d_inode->i_mutex);
 			err = jerror ?: err;
 		}
-
-	dput(root);
+	}
+out_dir:
+	dput(dir);
 out:
-	if (err)
-		reiserfs_warning(inode->i_sb, "jdm-20004",
-				 "Couldn't remove all xattrs (%d)\n", err);
+	/* -ENODATA isn't an error */
+	if (err == -ENODATA)
+		err = 0;
 	return err;
 }
 
-struct reiserfs_chown_buf {
-	struct inode *inode;
-	struct dentry *xadir;
-	struct iattr *attrs;
-};
-
-/* XXX: If there is a better way to do this, I'd love to hear about it */
-static int
-reiserfs_chown_xattrs_filler(void *buf, const char *name, int namelen,
-			     loff_t offset, u64 ino, unsigned int d_type)
+static int delete_one_xattr(struct dentry *dentry, void *data)
 {
-	struct reiserfs_chown_buf *chown_buf = (struct reiserfs_chown_buf *)buf;
-	struct dentry *xafile, *xadir = chown_buf->xadir;
-	struct iattr *attrs = chown_buf->attrs;
-	int err = 0;
+	struct inode *dir = dentry->d_parent->d_inode;
 
-	xafile = lookup_one_len(name, xadir, namelen);
-	if (IS_ERR(xafile))
-		return PTR_ERR(xafile);
-	else if (!xafile->d_inode) {
-		dput(xafile);
-		return -ENODATA;
-	}
+	/* This is the xattr dir, handle specially. */
+	if (S_ISDIR(dentry->d_inode->i_mode))
+		return xattr_rmdir(dir, dentry);
 
-	if (!S_ISDIR(xafile->d_inode->i_mode)) {
-		mutex_lock_nested(&xafile->d_inode->i_mutex, I_MUTEX_CHILD);
-		err = reiserfs_setattr(xafile, attrs);
-		mutex_unlock(&xafile->d_inode->i_mutex);
-	}
-	dput(xafile);
+	return xattr_unlink(dir, dentry);
+}
+
+static int chown_one_xattr(struct dentry *dentry, void *data)
+{
+	struct iattr *attrs = data;
+	return reiserfs_setattr(dentry, attrs);
+}
 
+/* No i_mutex, but the inode is unconnected. */
+int reiserfs_delete_xattrs(struct inode *inode)
+{
+	int err = reiserfs_for_each_xattr(inode, delete_one_xattr, NULL);
+	if (err)
+		reiserfs_warning(inode->i_sb, "jdm-20004",
+				 "Couldn't delete all xattrs (%d)\n", err);
 	return err;
 }
 
+/* inode->i_mutex: down */
 int reiserfs_chown_xattrs(struct inode *inode, struct iattr *attrs)
 {
-	struct dentry *dir;
-	int err = 0;
-	struct reiserfs_chown_buf buf;
-	unsigned int ia_valid = attrs->ia_valid;
-
-	/* Skip out, an xattr has no xattrs associated with it */
-	if (IS_PRIVATE(inode) || get_inode_sd_version(inode) == STAT_DATA_V1)
-		return 0;
-
-	dir = open_xa_dir(inode, XATTR_REPLACE);
-	if (IS_ERR(dir)) {
-		if (PTR_ERR(dir) != -ENODATA)
-			err = PTR_ERR(dir);
-		goto out;
-	} else if (!dir->d_inode)
-		goto out_dir;
-
-	attrs->ia_valid &= (ATTR_UID | ATTR_GID | ATTR_CTIME);
-	buf.xadir = dir;
-	buf.attrs = attrs;
-	buf.inode = inode;
-
-	mutex_lock_nested(&dir->d_inode->i_mutex, I_MUTEX_XATTR);
-	err = xattr_readdir(dir->d_inode, reiserfs_chown_xattrs_filler, &buf);
-
-	if (!err)
-		err = reiserfs_setattr(dir, attrs);
-	mutex_unlock(&dir->d_inode->i_mutex);
-
-	attrs->ia_valid = ia_valid;
-out_dir:
-	dput(dir);
-out:
+	int err = reiserfs_for_each_xattr(inode, chown_one_xattr, attrs);
 	if (err)
 		reiserfs_warning(inode->i_sb, "jdm-20007",
 				 "Couldn't chown all xattrs (%d)\n", err);
@@ -1004,6 +831,7 @@ ssize_t reiserfs_listxattr(struct dentry * dentry, char *buffer, size_t size)
 {
 	struct dentry *dir;
 	int err = 0;
+	loff_t pos = 0;
 	struct listxattr_buf buf = {
 		.inode = dentry->d_inode,
 		.buf = buffer,
@@ -1026,7 +854,7 @@ ssize_t reiserfs_listxattr(struct dentry * dentry, char *buffer, size_t size)
 	}
 
 	mutex_lock_nested(&dir->d_inode->i_mutex, I_MUTEX_XATTR);
-	err = xattr_readdir(dir->d_inode, listxattr_filler, &buf);
+	err = reiserfs_readdir_dentry(dir, &buf, listxattr_filler, &pos);
 	mutex_unlock(&dir->d_inode->i_mutex);
 
 	if (!err)

include/linux/reiserfs_fs.h

@@ -1984,6 +1984,7 @@ extern const struct inode_operations reiserfs_dir_inode_operations;
 extern const struct inode_operations reiserfs_symlink_inode_operations;
 extern const struct inode_operations reiserfs_special_inode_operations;
 extern const struct file_operations reiserfs_dir_operations;
+int reiserfs_readdir_dentry(struct dentry *, void *, filldir_t, loff_t *);
 
 /* tail_conversion.c */
 int direct2indirect(struct reiserfs_transaction_handle *, struct inode *,