]> git.ipfire.org Git - thirdparty/kernel/stable.git/commitdiff
ksmbd: Fix user namespace mapping
authorMickaël Salaün <mic@digikod.net>
Thu, 29 Sep 2022 10:04:47 +0000 (12:04 +0200)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Mon, 24 Oct 2022 07:56:52 +0000 (09:56 +0200)
commit 7c88c1e0ab1704bacb751341ee6431c3be34b834 upstream.

A kernel daemon should not rely on the current thread, which is unknown
and might be malicious.  Before this security fix,
ksmbd_override_fsids() didn't correctly override FS UID/GID which means
that arbitrary user space threads could trick the kernel to impersonate
arbitrary users or groups for file system access checks, leading to
file system access bypass.

This was found while investigating truncate support for Landlock:
https://lore.kernel.org/r/CAKYAXd8fpMJ7guizOjHgxEyyjoUwPsx3jLOPZP=wPYcbhkVXqA@mail.gmail.com

Fixes: e2f34481b24d ("cifsd: add server-side procedures for SMB3")
Cc: Hyunchul Lee <hyc.lee@gmail.com>
Cc: Steve French <smfrench@gmail.com>
Cc: stable@vger.kernel.org
Signed-off-by: Mickaël Salaün <mic@digikod.net>
Link: https://lore.kernel.org/r/20220929100447.108468-1-mic@digikod.net
Acked-by: Christian Brauner (Microsoft) <brauner@kernel.org>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
fs/ksmbd/smb_common.c

index 7f8ab14fb8ec147d2212948d2fd1770d31b2e67a..d96da872d70a1701b2e99b962f82ac6b83fd6351 100644 (file)
@@ -4,6 +4,8 @@
  *   Copyright (C) 2018 Namjae Jeon <linkinjeon@kernel.org>
  */
 
+#include <linux/user_namespace.h>
+
 #include "smb_common.h"
 #include "server.h"
 #include "misc.h"
@@ -625,8 +627,8 @@ int ksmbd_override_fsids(struct ksmbd_work *work)
        if (!cred)
                return -ENOMEM;
 
-       cred->fsuid = make_kuid(current_user_ns(), uid);
-       cred->fsgid = make_kgid(current_user_ns(), gid);
+       cred->fsuid = make_kuid(&init_user_ns, uid);
+       cred->fsgid = make_kgid(&init_user_ns, gid);
 
        gi = groups_alloc(0);
        if (!gi) {