On Tue, Mar 29, 2022 at 3:25 PM Miklos Szeredi <miklos@xxxxxxxxxx> wrote: > > On Tue, 29 Mar 2022 at 12:35, Christian Brauner <brauner@xxxxxxxxxx> wrote: > > > > From: "Christian Brauner (Microsoft)" <brauner@xxxxxxxxxx> > > > > Hey, > > > > This adds support for mounting overlay on top of idmapped layers. > > Looks good at first glance. > > I'm wondering if there's a way to deduplicate the info stored in > ovl_entry and ovl_inode, but that can come later... > Yes at this point the deduplication of lowerstack elements became a bit odd, but the final step is not obvious. I don't think we want a variable size ovl_inode. That would be suboptimal and make inode cache monitoring hard. I suppose we could have lowerstack or 'stack' allocated and hanging off ovl_inode. At that point, dentry->d_fsdata can become just flat flags. Thanks, Amir.
