On Tue, Sep 03, 2024 at 04:20:41PM +0200, Christian Brauner wrote: > However, I came up with an alternative using _Generic() to create a > compatibility layer that will call the correct variant of > kmem_cache_create() depending on whether struct kmem_cache_args is > passed or not. That compatibility layer can stay in place until we > updated all calls to be based on struct kmem_cache_args. Very nice! Yeah, this looks good to me. I'd been long pondering something like this for kmalloc itself (though codetags have taken that in a different direction). All the patches look like the right way forward to me: Reviewed-by: Kees Cook <kees@xxxxxxxxxx> -- Kees Cook