afs: Fix race between post-modification dir edit and readdir/d_revalidate

AFS directories are retained locally as a structured file, with lookup
being effected by a local search of the file contents.  When a modification
(such as mkdir) happens, the dir file content is modified locally rather
than redownloading the directory.

The directory contents are accessed in a number of ways, with a number of
different locks schemes:

 (1) Download of contents - dvnode->validate_lock/write in afs_read_dir().

 (2) Lookup and readdir - dvnode->validate_lock/read in afs_dir_iterate(),
     downgrading from (1) if necessary.

 (3) d_revalidate of child dentry - dvnode->validate_lock/read in
     afs_do_lookup_one() downgrading from (1) if necessary.

 (4) Edit of dir after modification - page locks on individual dir pages.

Unfortunately, because (4) uses different locking scheme to (1) - (3),
nothing protects against the page being scanned whilst the edit is
underway.  Even download is not safe as it doesn't lock the pages - relying
instead on the validate_lock to serialise as a whole (the theory being that
directory contents are treated as a block and always downloaded as a
block).

Fix this by write-locking dvnode->validate_lock around the edits.  Care
must be taken in the rename case as there may be two different dirs - but
they need not be locked at the same time.  In any case, once the lock is
taken, the directory version must be rechecked, and the edit skipped if a
later version has been downloaded by revalidation (there can't have been
any local changes because the VFS holds the inode lock, but there can have
been remote changes).

Fixes: 63a4681ff39c ("afs: Locally edit directory data for mkdir/create/unlink/...")
Signed-off-by: David Howells <dhowells@redhat.com>

1 files changed, 61 insertions, 28 deletions

diff --git a/fs/afs/dir.c b/fs/afs/dir.c
index 31d297e0f765..d6278616fb88 100644
--- a/fs/afs/dir.c
+++ b/fs/afs/dir.c
@@ -1275,6 +1275,7 @@ static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
 	struct afs_fs_cursor fc;
 	struct afs_vnode *dvnode = AFS_FS_I(dir);
 	struct key *key;
+	afs_dataversion_t data_version;
 	int ret;
 
 	mode |= S_IFDIR;
@@ -1295,7 +1296,7 @@ static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
 
 	ret = -ERESTARTSYS;
 	if (afs_begin_vnode_operation(&fc, dvnode, key, true)) {
-		afs_dataversion_t data_version = dvnode->status.data_version + 1;
+		data_version = dvnode->status.data_version + 1;
 
 		while (afs_select_fileserver(&fc)) {
 			fc.cb_break = afs_calc_vnode_cb_break(dvnode);
@@ -1316,10 +1317,14 @@ static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
 		goto error_key;
 	}
 
-	if (ret == 0 &&
-	    test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags))
-		afs_edit_dir_add(dvnode, &dentry->d_name, &iget_data.fid,
-				 afs_edit_dir_for_create);
+	if (ret == 0) {
+		down_write(&dvnode->validate_lock);
+		if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) &&
+		    dvnode->status.data_version == data_version)
+			afs_edit_dir_add(dvnode, &dentry->d_name, &iget_data.fid,
+					 afs_edit_dir_for_create);
+		up_write(&dvnode->validate_lock);
+	}
 
 	key_put(key);
 	kfree(scb);
@@ -1360,6 +1365,7 @@ static int afs_rmdir(struct inode *dir, struct dentry *dentry)
 	struct afs_fs_cursor fc;
 	struct afs_vnode *dvnode = AFS_FS_I(dir), *vnode = NULL;
 	struct key *key;
+	afs_dataversion_t data_version;
 	int ret;
 
 	_enter("{%llx:%llu},{%pd}",
@@ -1391,7 +1397,7 @@ static int afs_rmdir(struct inode *dir, struct dentry *dentry)
 
 	ret = -ERESTARTSYS;
 	if (afs_begin_vnode_operation(&fc, dvnode, key, true)) {
-		afs_dataversion_t data_version = dvnode->status.data_version + 1;
+		data_version = dvnode->status.data_version + 1;
 
 		while (afs_select_fileserver(&fc)) {
 			fc.cb_break = afs_calc_vnode_cb_break(dvnode);
@@ -1404,9 +1410,12 @@ static int afs_rmdir(struct inode *dir, struct dentry *dentry)
 		ret = afs_end_vnode_operation(&fc);
 		if (ret == 0) {
 			afs_dir_remove_subdir(dentry);
-			if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags))
+			down_write(&dvnode->validate_lock);
+			if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) &&
+			    dvnode->status.data_version == data_version)
 				afs_edit_dir_remove(dvnode, &dentry->d_name,
 						    afs_edit_dir_for_rmdir);
+			up_write(&dvnode->validate_lock);
 		}
 	}
 
@@ -1544,10 +1553,15 @@ static int afs_unlink(struct inode *dir, struct dentry *dentry)
 		ret = afs_end_vnode_operation(&fc);
 		if (ret == 0 && !(scb[1].have_status || scb[1].have_error))
 			ret = afs_dir_remove_link(dvnode, dentry, key);
-		if (ret == 0 &&
-		    test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags))
-			afs_edit_dir_remove(dvnode, &dentry->d_name,
-					    afs_edit_dir_for_unlink);
+
+		if (ret == 0) {
+			down_write(&dvnode->validate_lock);
+			if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) &&
+			    dvnode->status.data_version == data_version)
+				afs_edit_dir_remove(dvnode, &dentry->d_name,
+						    afs_edit_dir_for_unlink);
+			up_write(&dvnode->validate_lock);
+		}
 	}
 
 	if (need_rehash && ret < 0 && ret != -ENOENT)
@@ -1573,6 +1587,7 @@ static int afs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
 	struct afs_status_cb *scb;
 	struct afs_vnode *dvnode = AFS_FS_I(dir);
 	struct key *key;
+	afs_dataversion_t data_version;
 	int ret;
 
 	mode |= S_IFREG;
@@ -1597,7 +1612,7 @@ static int afs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
 
 	ret = -ERESTARTSYS;
 	if (afs_begin_vnode_operation(&fc, dvnode, key, true)) {
-		afs_dataversion_t data_version = dvnode->status.data_version + 1;
+		data_version = dvnode->status.data_version + 1;
 
 		while (afs_select_fileserver(&fc)) {
 			fc.cb_break = afs_calc_vnode_cb_break(dvnode);
@@ -1618,9 +1633,12 @@ static int afs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
 		goto error_key;
 	}
 
-	if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags))
+	down_write(&dvnode->validate_lock);
+	if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) &&
+	    dvnode->status.data_version == data_version)
 		afs_edit_dir_add(dvnode, &dentry->d_name, &iget_data.fid,
 				 afs_edit_dir_for_create);
+	up_write(&dvnode->validate_lock);
 
 	kfree(scb);
 	key_put(key);
@@ -1648,6 +1666,7 @@ static int afs_link(struct dentry *from, struct inode *dir,
 	struct afs_vnode *dvnode = AFS_FS_I(dir);
 	struct afs_vnode *vnode = AFS_FS_I(d_inode(from));
 	struct key *key;
+	afs_dataversion_t data_version;
 	int ret;
 
 	_enter("{%llx:%llu},{%llx:%llu},{%pd}",
@@ -1672,7 +1691,7 @@ static int afs_link(struct dentry *from, struct inode *dir,
 
 	ret = -ERESTARTSYS;
 	if (afs_begin_vnode_operation(&fc, dvnode, key, true)) {
-		afs_dataversion_t data_version = dvnode->status.data_version + 1;
+		data_version = dvnode->status.data_version + 1;
 
 		if (mutex_lock_interruptible_nested(&vnode->io_lock, 1) < 0) {
 			afs_end_vnode_operation(&fc);
@@ -1702,9 +1721,12 @@ static int afs_link(struct dentry *from, struct inode *dir,
 		goto error_key;
 	}
 
-	if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags))
+	down_write(&dvnode->validate_lock);
+	if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) &&
+	    dvnode->status.data_version == data_version)
 		afs_edit_dir_add(dvnode, &dentry->d_name, &vnode->fid,
 				 afs_edit_dir_for_link);
+	up_write(&dvnode->validate_lock);
 
 	key_put(key);
 	kfree(scb);
@@ -1732,6 +1754,7 @@ static int afs_symlink(struct inode *dir, struct dentry *dentry,
 	struct afs_status_cb *scb;
 	struct afs_vnode *dvnode = AFS_FS_I(dir);
 	struct key *key;
+	afs_dataversion_t data_version;
 	int ret;
 
 	_enter("{%llx:%llu},{%pd},%s",
@@ -1759,7 +1782,7 @@ static int afs_symlink(struct inode *dir, struct dentry *dentry,
 
 	ret = -ERESTARTSYS;
 	if (afs_begin_vnode_operation(&fc, dvnode, key, true)) {
-		afs_dataversion_t data_version = dvnode->status.data_version + 1;
+		data_version = dvnode->status.data_version + 1;
 
 		while (afs_select_fileserver(&fc)) {
 			fc.cb_break = afs_calc_vnode_cb_break(dvnode);
@@ -1780,9 +1803,12 @@ static int afs_symlink(struct inode *dir, struct dentry *dentry,
 		goto error_key;
 	}
 
-	if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags))
+	down_write(&dvnode->validate_lock);
+	if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) &&
+	    dvnode->status.data_version == data_version)
 		afs_edit_dir_add(dvnode, &dentry->d_name, &iget_data.fid,
 				 afs_edit_dir_for_symlink);
+	up_write(&dvnode->validate_lock);
 
 	key_put(key);
 	kfree(scb);
@@ -1812,6 +1838,8 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	struct dentry *tmp = NULL, *rehash = NULL;
 	struct inode *new_inode;
 	struct key *key;
+	afs_dataversion_t orig_data_version;
+	afs_dataversion_t new_data_version;
 	bool new_negative = d_is_negative(new_dentry);
 	int ret;
 
@@ -1890,9 +1918,6 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry,
 
 	ret = -ERESTARTSYS;
 	if (afs_begin_vnode_operation(&fc, orig_dvnode, key, true)) {
-		afs_dataversion_t orig_data_version;
-		afs_dataversion_t new_data_version;
-
 		orig_data_version = orig_dvnode->status.data_version + 1;
 
 		if (orig_dvnode != new_dvnode) {
@@ -1928,18 +1953,25 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	if (ret == 0) {
 		if (rehash)
 			d_rehash(rehash);
-		if (test_bit(AFS_VNODE_DIR_VALID, &orig_dvnode->flags))
-		    afs_edit_dir_remove(orig_dvnode, &old_dentry->d_name,
-					afs_edit_dir_for_rename_0);
+		down_write(&orig_dvnode->validate_lock);
+		if (test_bit(AFS_VNODE_DIR_VALID, &orig_dvnode->flags) &&
+		    orig_dvnode->status.data_version == orig_data_version)
+			afs_edit_dir_remove(orig_dvnode, &old_dentry->d_name,
+					    afs_edit_dir_for_rename_0);
+		if (orig_dvnode != new_dvnode) {
+			up_write(&orig_dvnode->validate_lock);
 
-		if (!new_negative &&
-		    test_bit(AFS_VNODE_DIR_VALID, &new_dvnode->flags))
-			afs_edit_dir_remove(new_dvnode, &new_dentry->d_name,
-					    afs_edit_dir_for_rename_1);
+			down_write(&new_dvnode->validate_lock);
+		}
+		if (test_bit(AFS_VNODE_DIR_VALID, &new_dvnode->flags) &&
+		    orig_dvnode->status.data_version == new_data_version) {
+			if (!new_negative)
+				afs_edit_dir_remove(new_dvnode, &new_dentry->d_name,
+						    afs_edit_dir_for_rename_1);
 
-		if (test_bit(AFS_VNODE_DIR_VALID, &new_dvnode->flags))
 			afs_edit_dir_add(new_dvnode, &new_dentry->d_name,
 					 &vnode->fid, afs_edit_dir_for_rename_2);
+		}
 
 		new_inode = d_inode(new_dentry);
 		if (new_inode) {
@@ -1958,6 +1990,7 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		afs_update_dentry_version(&fc, old_dentry, &scb[1]);
 		afs_update_dentry_version(&fc, new_dentry, &scb[1]);
 		d_move(old_dentry, new_dentry);
+		up_write(&new_dvnode->validate_lock);
 		goto error_tmp;
 	}