Re: overlayfs: BUG in ovl_whiteout

[Andy Whitcroft <apw@canonical.com>]

On Mon, Oct 04, 2010 at 10:16:29PM +0200, Miklos Szeredi wrote:
> 
> OK, my guess is that it's a 'rename to self' which was not properly
> implemented.
> 
> Does the following patch make a difference?

That one panic'd immediatly.  Seems that we can have missmatched lowers
as well.  I applied the below over the top, and that seems to pass
testing without tripping the WARN_ON.

-apw

commit 03375858c62f617846b5dc4968fe8178eda51a7c
Author: Andy Whitcroft <apw@canonical.com>
Date:   Tue Oct 5 14:54:08 2010 +0100

    overlayfs: handle missing lower inodes in ovl_is_same_inode
    
    Signed-off-by: Andy Whitcroft <apw@canonical.com>

diff --git a/fs/overlayfs/overlayfs.c b/fs/overlayfs/overlayfs.c
index c10cc7b..f596222 100644
--- a/fs/overlayfs/overlayfs.c
+++ b/fs/overlayfs/overlayfs.c
@@ -1964,19 +1964,25 @@ static int ovl_statfs(struct dentry *dentry, struct kstatfs *buf)
 
 static bool ovl_is_same_inode(struct dentry *d1, struct dentry *d2)
 {
-	struct dentry *upperd1;
-	struct dentry *upperd2;
+	struct dentry *od1;
+	struct dentry *od2;
 
-	upperd1 = ovl_dentry_upper(d1);
-	upperd2 = ovl_dentry_upper(d2);
+	od1 = ovl_dentry_upper(d1);
+	od2 = ovl_dentry_upper(d2);
 
-	if (upperd1 && upperd2)
-		return vfs_is_same_inode(upperd1, upperd2);
+	if (od1 && od2)
+		return vfs_is_same_inode(od1, od2);
 
-	if (upperd1 || upperd2)
+	if (od1 || od2)
 		return false;
 
-	return vfs_is_same_inode(ovl_dentry_lower(d1), ovl_dentry_lower(d2));
+	od1 = ovl_dentry_lower(d1);
+	od2 = ovl_dentry_lower(d2);
+
+	if (od1 && od2)
+		return vfs_is_same_inode(od1, od2);
+
+	return false;
 }
 
 static const struct super_operations ovl_super_operations = {