restore pinning the victim dentry in vfs_rmdir()/vfs_rename_dir() (1d2ef590) · Commits · 戴 / test

We used to get the victim pinned by dentry_unhash() prior to commit ("vfs: remove dget() from dentry_unhash()") and ->rmdir() and ->rename() instances relied on that; most of them don't care, but ones that used d_delete() themselves do. As the result, we are getting rmdir() oopses on NFS now. Just grab the reference before locking the victim and drop it explicitly after unlocking, same as vfs_rename_other() does. Signed-off-by:

Al Viro <viro@zeniv.linux.org.uk> Tested-by:

Simon Kirby <sim@hostway.ca> Cc: stable@kernel.org (3.0.x) Signed-off-by:

Linus Torvalds <torvalds@linux-foundation.org>

fs/namei.c

+4 −0

Original line number	Diff line number	Diff line
		@@ -2616,6 +2616,7 @@ int vfs_rmdir(struct inode dir, struct dentry dentry)
		if (!dir->i_op->rmdir)
		return -EPERM;

		dget(dentry);
		mutex_lock(&dentry->d_inode->i_mutex);

		error = -EBUSY;
		@@ -2636,6 +2637,7 @@ int vfs_rmdir(struct inode dir, struct dentry dentry)

		out:
		mutex_unlock(&dentry->d_inode->i_mutex);
		dput(dentry);
		if (!error)
		d_delete(dentry);
		return error;
		@@ -3025,6 +3027,7 @@ static int vfs_rename_dir(struct inode old_dir, struct dentry old_dentry,
		if (error)
		return error;

		dget(new_dentry);
		if (target)
		mutex_lock(&target->i_mutex);

		@@ -3045,6 +3048,7 @@ static int vfs_rename_dir(struct inode old_dir, struct dentry old_dentry,
		out:
		if (target)
		mutex_unlock(&target->i_mutex);
		dput(new_dentry);
		if (!error)
		if (!(old_dir->i_sb->s_type->fs_flags & FS_RENAME_DOES_D_MOVE))
		d_move(old_dentry,new_dentry);

Admin message