Skip to contentSkip to conversation

vfs: stop d_splice_alias creating directory aliases

PATCH In review J. Bruce Fields 2 participants 3 messages May 23, 2012 1 message not in the archive

Subsystems (3)
VFS NFS client NFS server
lore
Share & tools
lore

J. Bruce Fields

author
#0 lore raw

PATCH 1/2 vfs: stop d_splice_alias creating directory aliases

From: "J. Bruce Fields" <[email protected]>

A directory should never have more than one dentry pointing to it.

But d_splice_alias() will add one if it finds a directory with an already-existing non-DISCONNECTED dentry.

I can't find an obvious reproducer, but I also can't see what prevents d_splice_alias() from encountering such a case.

It therefore seems safest to allow d_splice_alias to use any dentry it finds.

(Prior to the removal of dentry_unhash() from vfs_rmdir(), around v3.0, this could cause an nfsd deadlock like this:

- Somebody attempts to remove a non-empty directory.
- The dentry_unhash() in vfs_rmdir() unhashes the dentry pointing to the non-empty directory.
- ->rmdir() then fails with -ENOTEMPTY
- Before the vfs_rmdir() caller reaches dput(), an nfsd process in rename looks up the directory by filehandle; at the end of that lookup, this dentry is found by d_alloc_anon(), and a reference is taken on it, preventing dput() from removing it.
- A regular lookup of the directory calls d_splice_alias(), finds only an unhashed (not a DISCONNECTED) dentry, and insteads adds a new one, so the directory now has two dentries.
- The nfsd process in rename, which was previously looking up the source directory of the rename, now looks up the target directory (which is the same), and gets the dentry newly created by the previous lookup.
- The rename, seeing two different dentries, assumes this is a cross-directory rename and attempts to take the i_mutex on the directory twice.

That reproducer no longer exists, but I don't think there was anything fundamentally incorrect about the vfs_rmdir() behavior there, so I think the real fault was here in d_splice_alias().)

Signed-off-by J. Bruce Fields <[email protected]>
---
 fs/dcache.c |    3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
fs/dcache.c +1 -2
index b60ddc4..2434c1e 100644
@@ -1606,9 +1606,8 @@ struct dentry *d_splice_alias(struct inode *inode, struct dentry *dentry)
 
 	if (inode && S_ISDIR(inode->i_mode)) {
 		spin_lock(&inode->i_lock);
-		new = __d_find_alias(inode, 1);
+		new = __d_find_any_alias(inode);
 		if (new) {
-			BUG_ON(!(new->d_flags & DCACHE_DISCONNECTED));
 			spin_unlock(&inode->i_lock);
 			security_d_instantiate(new, inode);
 			d_move(new, dentry);
Signature · 1 line
1.7.9.5
author
#1 lore raw

PATCH 2/2 vfs: remove unused __d_splice_alias argument

From: "J. Bruce Fields" <[email protected]>

Nobody sets want_disconn any more.

Reported-by Peng Tao <[email protected]>
Signed-off-by J. Bruce Fields <[email protected]>
---
 fs/dcache.c |   13 +++++--------
 1 file changed, 5 insertions(+), 8 deletions(-)
fs/dcache.c +5 -8
index 2434c1e..8e34e91 100644
@@ -649,8 +649,6 @@ EXPORT_SYMBOL(dget_parent);
 /**
  * d_find_alias - grab a hashed alias of inode
  * @inode: inode in question
- * @want_discon:  flag, used by d_splice_alias, to request
- *          that only a DISCONNECTED alias be returned.
  *
  * If inode has a hashed alias, or is a directory and has any alias,
  * acquire the reference to alias and return it. Otherwise return NULL.

@@ -659,10 +657,9 @@ EXPORT_SYMBOL(dget_parent);
  * of a filesystem.
  *
  * If the inode has an IS_ROOT, DCACHE_DISCONNECTED alias, then prefer
- * any other hashed alias over that one unless @want_discon is set,
- * in which case only return an IS_ROOT, DCACHE_DISCONNECTED alias.
+ * any other hashed alias over that.
  */
-static struct dentry *__d_find_alias(struct inode *inode, int want_discon)
+static struct dentry *__d_find_alias(struct inode *inode)
 {
 	struct dentry *alias, *discon_alias;
 

@@ -674,7 +671,7 @@ again:
 			if (IS_ROOT(alias) &&
 			    (alias->d_flags & DCACHE_DISCONNECTED)) {
 				discon_alias = alias;
-			} else if (!want_discon) {
+			} else {
 				__dget_dlock(alias);
 				spin_unlock(&alias->d_lock);
 				return alias;

@@ -705,7 +702,7 @@ struct dentry *d_find_alias(struct inode *inode)
 
 	if (!list_empty(&inode->i_dentry)) {
 		spin_lock(&inode->i_lock);
-		de = __d_find_alias(inode, 0);
+		de = __d_find_alias(inode);
 		spin_unlock(&inode->i_lock);
 	}
 	return de;

@@ -2395,7 +2392,7 @@ struct dentry *d_materialise_unique(struct dentry *dentry, struct inode *inode)
 		struct dentry *alias;
 
 		/* Does an aliased dentry already exist? */
-		alias = __d_find_alias(inode, 0);
+		alias = __d_find_alias(inode);
 		if (alias) {
 			actual = alias;
 			write_seqlock(&rename_lock);
Signature · 1 line
1.7.9.5

A message that is not in the archive (sent off-list or not yet received): 1337810746-16240-1-git-send-email-bfields-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org

#3 lore raw

These two patches are a resend. I think they're appropriate for 3.5.

--b.

Quoted 67 lines · J. Bruce Fields

From: "J. Bruce Fields" <[email protected]>

A directory should never have more than one dentry pointing to it.

But d_splice_alias() will add one if it finds a directory with an already-existing non-DISCONNECTED dentry.

I can't find an obvious reproducer, but I also can't see what prevents d_splice_alias() from encountering such a case.

It therefore seems safest to allow d_splice_alias to use any dentry it finds.

(Prior to the removal of dentry_unhash() from vfs_rmdir(), around v3.0, this could cause an nfsd deadlock like this:

- Somebody attempts to remove a non-empty directory.
- The dentry_unhash() in vfs_rmdir() unhashes the dentry pointing to the non-empty directory.
- ->rmdir() then fails with -ENOTEMPTY
- Before the vfs_rmdir() caller reaches dput(), an nfsd process in rename looks up the directory by filehandle; at the end of that lookup, this dentry is found by d_alloc_anon(), and a reference is taken on it, preventing dput() from removing it.
- A regular lookup of the directory calls d_splice_alias(), finds only an unhashed (not a DISCONNECTED) dentry, and insteads adds a new one, so the directory now has two dentries.
- The nfsd process in rename, which was previously looking up the source directory of the rename, now looks up the target directory (which is the same), and gets the dentry newly created by the previous lookup.
- The rename, seeing two different dentries, assumes this is a cross-directory rename and attempts to take the i_mutex on the directory twice.

That reproducer no longer exists, but I don't think there was anything fundamentally incorrect about the vfs_rmdir() behavior there, so I think the real fault was here in d_splice_alias().)

Signed-off-by J. Bruce Fields <[email protected]>
---
 fs/dcache.c |    3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
fs/dcache.c +1 -2
index b60ddc4..2434c1e 100644
@@ -1606,9 +1606,8 @@ struct dentry *d_splice_alias(struct inode *inode, struct dentry *dentry)
 
 	if (inode && S_ISDIR(inode->i_mode)) {
 		spin_lock(&inode->i_lock);
-		new = __d_find_alias(inode, 1);
+		new = __d_find_any_alias(inode);
 		if (new) {
-			BUG_ON(!(new->d_flags & DCACHE_DISCONNECTED));
 			spin_unlock(&inode->i_lock);
 			security_d_instantiate(new, inode);
 			d_move(new, dentry);
Signature · 6 lines
1.7.9.5

--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html