From: Tejun Heo <tj@kernel.org>
Date: Thu, 6 Jan 2022 11:02:28 -1000
Subject: cgroup: Use open-time credentials for process migraton perm checks
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Git-commit: 1756d7994ad85c2479af6ae5a9750b92324685af
Patch-mainline: v5.16
References: bsc#1194302 CVE-2021-4197

cgroup process migration permission checks are performed at write time as
whether a given operation is allowed or not is dependent on the content of
the write - the PID. This currently uses current's credentials which is a
potential security weakness as it may allow scenarios where a less
privileged process tricks a more privileged one into writing into a fd that
it created.

This patch makes both cgroup2 and cgroup1 process migration interfaces to
use the credentials saved at the time of open (file->f_cred) instead of
current's.

Reported-by: "Eric W. Biederman" <ebiederm@xmission.com>
Suggested-by: Linus Torvalds <torvalds@linuxfoundation.org>
Fixes: 187fe84067bd ("cgroup: require write perm on common ancestor when moving processes on the default hierarchy")
Reviewed-by: Michal Koutný <mkoutny@suse.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Michal Koutný <mkoutny@suse.com>
---
 kernel/cgroup/cgroup-v1.c | 7 ++++---
 kernel/cgroup/cgroup.c    | 9 ++++++++-
 2 files changed, 12 insertions(+), 4 deletions(-)

diff --git a/kernel/cgroup/cgroup-v1.c b/kernel/cgroup/cgroup-v1.c
index 81c9e0685948..0e7369103ba6 100644
--- a/kernel/cgroup/cgroup-v1.c
+++ b/kernel/cgroup/cgroup-v1.c
@@ -504,10 +504,11 @@ static ssize_t __cgroup1_procs_write(struct kernfs_open_file *of,
 		goto out_unlock;
 
 	/*
-	 * Even if we're attaching all tasks in the thread group, we only
-	 * need to check permissions on one of them.
+	 * Even if we're attaching all tasks in the thread group, we only need
+	 * to check permissions on one of them. Check permissions using the
+	 * credentials from file open to protect against inherited fd attacks.
 	 */
-	cred = current_cred();
+	cred = of->file->f_cred;
 	tcred = get_task_cred(task);
 	if (!uid_eq(cred->euid, GLOBAL_ROOT_UID) &&
 	    !uid_eq(cred->euid, tcred->uid) &&
diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
index 919194de39c8..2632e46da1d4 100644
--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -4892,6 +4892,7 @@ static ssize_t __cgroup_procs_write(struct kernfs_open_file *of, char *buf,
 {
 	struct cgroup *src_cgrp, *dst_cgrp;
 	struct task_struct *task;
+	const struct cred *saved_cred;
 	ssize_t ret;
 	bool locked;
 
@@ -4909,9 +4910,15 @@ static ssize_t __cgroup_procs_write(struct kernfs_open_file *of, char *buf,
 	src_cgrp = task_cgroup_from_root(task, &cgrp_dfl_root);
 	spin_unlock_irq(&css_set_lock);
 
-	/* process and thread migrations follow same delegation rule */
+	/*
+	 * Process and thread migrations follow same delegation rule. Check
+	 * permissions using the credentials from file open to protect against
+	 * inherited fd attacks.
+	 */
+	saved_cred = override_creds(of->file->f_cred);
 	ret = cgroup_attach_permissions(src_cgrp, dst_cgrp,
 					of->file->f_path.dentry->d_sb, threadgroup);
+	revert_creds(saved_cred);
 	if (ret)
 		goto out_finish;