reiserfs: eliminate per-super xattr lock
With the switch to using inode->i_mutex locking during lookups/creation
in the xattr root, the per-super xattr lock is no longer needed.
This patch removes it.
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c
index 57920a4..62c9882 100644
--- a/fs/reiserfs/xattr.c
+++ b/fs/reiserfs/xattr.c
@@ -27,6 +27,12 @@
* these are special cases for filesystem ACLs, they are interpreted by the
* kernel, in addition, they are negatively and positively cached and attached
* to the inode so that unnecessary lookups are avoided.
+ *
+ * Locking works like so:
+ * The xattr root (/.reiserfs_priv/xattrs) is protected by its i_mutex.
+ * The xattr dir (/.reiserfs_priv/xattrs/<oid>.<gen>) is protected by
+ * inode->xattr_sem.
+ * The xattrs themselves are likewise protected by the xattr_sem.
*/
#include <linux/reiserfs_fs.h>
@@ -392,16 +398,17 @@
/* This is called w/ inode->i_mutex downed */
int reiserfs_delete_xattrs(struct inode *inode)
{
- struct dentry *dir, *root;
int err = 0;
+ 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);
/* Skip out, an xattr has no xattrs associated with it */
if (IS_PRIVATE(inode) || get_inode_sd_version(inode) == STAT_DATA_V1)
return 0;
- reiserfs_read_lock_xattrs(inode->i_sb);
dir = open_xa_dir(inode, XATTR_REPLACE);
- reiserfs_read_unlock_xattrs(inode->i_sb);
if (IS_ERR(dir)) {
err = PTR_ERR(dir);
goto out;
@@ -416,18 +423,26 @@
if (err)
goto out_dir;
- /* Leftovers besides . and .. -- that's not good. */
- if (dir->d_inode->i_nlink <= 2) {
- root = open_xa_root(inode->i_sb, XATTR_REPLACE);
- reiserfs_write_lock_xattrs(inode->i_sb);
+ /* 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. */
+ err = journal_begin(&th, inode->i_sb, blocks);
+ if (!err) {
+ int jerror;
+ root = dget(dir->d_parent);
mutex_lock_nested(&root->d_inode->i_mutex, I_MUTEX_XATTR);
err = xattr_rmdir(root->d_inode, dir);
+ jerror = journal_end(&th, inode->i_sb, blocks);
mutex_unlock(&root->d_inode->i_mutex);
- reiserfs_write_unlock_xattrs(inode->i_sb);
dput(root);
- } else {
- reiserfs_warning(inode->i_sb, "jdm-20006",
- "Couldn't remove all entries in directory");
+
+ err = jerror ?: err;
}
out_dir:
@@ -437,6 +452,9 @@
if (!err)
REISERFS_I(inode)->i_flags =
REISERFS_I(inode)->i_flags & ~i_has_xattr_dir;
+ else
+ reiserfs_warning(inode->i_sb, "jdm-20004",
+ "Couldn't remove all xattrs (%d)\n", err);
return err;
}
@@ -485,9 +503,7 @@
if (IS_PRIVATE(inode) || get_inode_sd_version(inode) == STAT_DATA_V1)
return 0;
- reiserfs_read_lock_xattrs(inode->i_sb);
dir = open_xa_dir(inode, XATTR_REPLACE);
- reiserfs_read_unlock_xattrs(inode->i_sb);
if (IS_ERR(dir)) {
if (PTR_ERR(dir) != -ENODATA)
err = PTR_ERR(dir);
@@ -731,6 +747,11 @@
goto out;
}
+ /* protect against concurrent access. xattrs are backed by
+ * regular files, but they're not regular files. The updates
+ * must be atomic from the perspective of the user. */
+ mutex_lock_nested(&dentry->d_inode->i_mutex, I_MUTEX_XATTR);
+
isize = i_size_read(dentry->d_inode);
REISERFS_I(inode)->i_flags |= i_has_xattr_dir;
@@ -798,6 +819,7 @@
}
out_dput:
+ mutex_unlock(&dentry->d_inode->i_mutex);
dput(dentry);
out:
@@ -834,7 +856,6 @@
static struct reiserfs_xattr_handler *find_xattr_handler_prefix(const char *);
/*
* Inode operation getxattr()
- * Preliminary locking: we down dentry->d_inode->i_mutex
*/
ssize_t
reiserfs_getxattr(struct dentry * dentry, const char *name, void *buffer,
@@ -848,9 +869,7 @@
return -EOPNOTSUPP;
reiserfs_read_lock_xattr_i(dentry->d_inode);
- reiserfs_read_lock_xattrs(dentry->d_sb);
err = xah->get(dentry->d_inode, name, buffer, size);
- reiserfs_read_unlock_xattrs(dentry->d_sb);
reiserfs_read_unlock_xattr_i(dentry->d_inode);
return err;
}
@@ -866,23 +885,13 @@
{
struct reiserfs_xattr_handler *xah = find_xattr_handler_prefix(name);
int err;
- int lock;
if (!xah || !reiserfs_xattrs(dentry->d_sb) ||
get_inode_sd_version(dentry->d_inode) == STAT_DATA_V1)
return -EOPNOTSUPP;
reiserfs_write_lock_xattr_i(dentry->d_inode);
- lock = !has_xattr_dir(dentry->d_inode);
- if (lock)
- reiserfs_write_lock_xattrs(dentry->d_sb);
- else
- reiserfs_read_lock_xattrs(dentry->d_sb);
err = xah->set(dentry->d_inode, name, value, size, flags);
- if (lock)
- reiserfs_write_unlock_xattrs(dentry->d_sb);
- else
- reiserfs_read_unlock_xattrs(dentry->d_sb);
reiserfs_write_unlock_xattr_i(dentry->d_inode);
return err;
}
@@ -902,8 +911,6 @@
return -EOPNOTSUPP;
reiserfs_write_lock_xattr_i(dentry->d_inode);
- reiserfs_read_lock_xattrs(dentry->d_sb);
-
/* Deletion pre-operation */
if (xah->del) {
err = xah->del(dentry->d_inode, name);
@@ -917,7 +924,6 @@
mark_inode_dirty(dentry->d_inode);
out:
- reiserfs_read_unlock_xattrs(dentry->d_sb);
reiserfs_write_unlock_xattr_i(dentry->d_inode);
return err;
}
@@ -966,8 +972,6 @@
/*
* Inode operation listxattr()
- *
- * Preliminary locking: we down dentry->d_inode->i_mutex
*/
ssize_t reiserfs_listxattr(struct dentry * dentry, char *buffer, size_t size)
{
@@ -983,9 +987,7 @@
return -EOPNOTSUPP;
reiserfs_read_lock_xattr_i(dentry->d_inode);
- reiserfs_read_lock_xattrs(dentry->d_sb);
dir = open_xa_dir(dentry->d_inode, XATTR_REPLACE);
- reiserfs_read_unlock_xattrs(dentry->d_sb);
if (IS_ERR(dir)) {
err = PTR_ERR(dir);
if (err == -ENODATA)
@@ -1114,11 +1116,9 @@
int error = -EAGAIN; /* do regular unix permission checks by default */
reiserfs_read_lock_xattr_i(inode);
- reiserfs_read_lock_xattrs(inode->i_sb);
acl = reiserfs_get_acl(inode, ACL_TYPE_ACCESS);
- reiserfs_read_unlock_xattrs(inode->i_sb);
reiserfs_read_unlock_xattr_i(inode);
if (acl) {