This is a note to let you know that I've just added the patch titled
af_unix: fix races in sk_peer_pid and sk_peer_cred accesses
to the 5.4-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
The filename of the patch is:
af_unix-fix-races-in-sk_peer_pid-and-sk_peer_cred-ac.patch
and it can be found in the queue-5.4 subdirectory.
If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@xxxxxxxxxxxxxxx> know about it.
commit e20576e0edb24cc440225fb48ce7bc5117f820d2
Author: Eric Dumazet <edumazet@xxxxxxxxxx>
Date: Wed Sep 29 15:57:50 2021 -0700
af_unix: fix races in sk_peer_pid and sk_peer_cred accesses
[ Upstream commit 35306eb23814444bd4021f8a1c3047d3cb0c8b2b ]
Jann Horn reported that SO_PEERCRED and SO_PEERGROUPS implementations
are racy, as af_unix can concurrently change sk_peer_pid and sk_peer_cred.
In order to fix this issue, this patch adds a new spinlock that needs
to be used whenever these fields are read or written.
Jann also pointed out that l2cap_sock_get_peer_pid_cb() is currently
reading sk->sk_peer_pid which makes no sense, as this field
is only possibly set by AF_UNIX sockets.
We will have to clean this in a separate patch.
This could be done by reverting b48596d1dc25 "Bluetooth: L2CAP: Add get_peer_pid callback"
or implementing what was truly expected.
Fixes: 109f6e39fa07 ("af_unix: Allow SO_PEERCRED to work across namespaces.")
Signed-off-by: Eric Dumazet <edumazet@xxxxxxxxxx>
Reported-by: Jann Horn <jannh@xxxxxxxxxx>
Cc: Eric W. Biederman <ebiederm@xxxxxxxxxxxx>
Cc: Luiz Augusto von Dentz <luiz.von.dentz@xxxxxxxxx>
Cc: Marcel Holtmann <marcel@xxxxxxxxxxxx>
Signed-off-by: David S. Miller <davem@xxxxxxxxxxxxx>
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
diff --git a/include/net/sock.h b/include/net/sock.h
index d3dd89b6e2cb..079b5f6f13d8 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -470,8 +470,10 @@ struct sock {
u32 sk_ack_backlog;
u32 sk_max_ack_backlog;
kuid_t sk_uid;
+ spinlock_t sk_peer_lock;
struct pid *sk_peer_pid;
const struct cred *sk_peer_cred;
+
long sk_rcvtimeo;
ktime_t sk_stamp;
#if BITS_PER_LONG==32
diff --git a/net/core/sock.c b/net/core/sock.c
index 452883b28aba..57b7a10703c3 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1181,6 +1181,16 @@ int sock_setsockopt(struct socket *sock, int level, int optname,
}
EXPORT_SYMBOL(sock_setsockopt);
+static const struct cred *sk_get_peer_cred(struct sock *sk)
+{
+ const struct cred *cred;
+
+ spin_lock(&sk->sk_peer_lock);
+ cred = get_cred(sk->sk_peer_cred);
+ spin_unlock(&sk->sk_peer_lock);
+
+ return cred;
+}
static void cred_to_ucred(struct pid *pid, const struct cred *cred,
struct ucred *ucred)
@@ -1355,7 +1365,11 @@ int sock_getsockopt(struct socket *sock, int level, int optname,
struct ucred peercred;
if (len > sizeof(peercred))
len = sizeof(peercred);
+
+ spin_lock(&sk->sk_peer_lock);
cred_to_ucred(sk->sk_peer_pid, sk->sk_peer_cred, &peercred);
+ spin_unlock(&sk->sk_peer_lock);
+
if (copy_to_user(optval, &peercred, len))
return -EFAULT;
goto lenout;
@@ -1363,20 +1377,23 @@ int sock_getsockopt(struct socket *sock, int level, int optname,
case SO_PEERGROUPS:
{
+ const struct cred *cred;
int ret, n;
- if (!sk->sk_peer_cred)
+ cred = sk_get_peer_cred(sk);
+ if (!cred)
return -ENODATA;
- n = sk->sk_peer_cred->group_info->ngroups;
+ n = cred->group_info->ngroups;
if (len < n * sizeof(gid_t)) {
len = n * sizeof(gid_t);
+ put_cred(cred);
return put_user(len, optlen) ? -EFAULT : -ERANGE;
}
len = n * sizeof(gid_t);
- ret = groups_to_user((gid_t __user *)optval,
- sk->sk_peer_cred->group_info);
+ ret = groups_to_user((gid_t __user *)optval, cred->group_info);
+ put_cred(cred);
if (ret)
return ret;
goto lenout;
@@ -1714,9 +1731,10 @@ static void __sk_destruct(struct rcu_head *head)
sk->sk_frag.page = NULL;
}
- if (sk->sk_peer_cred)
- put_cred(sk->sk_peer_cred);
+ /* We do not need to acquire sk->sk_peer_lock, we are the last user. */
+ put_cred(sk->sk_peer_cred);
put_pid(sk->sk_peer_pid);
+
if (likely(sk->sk_net_refcnt))
put_net(sock_net(sk));
sk_prot_free(sk->sk_prot_creator, sk);
@@ -2915,6 +2933,8 @@ void sock_init_data(struct socket *sock, struct sock *sk)
sk->sk_peer_pid = NULL;
sk->sk_peer_cred = NULL;
+ spin_lock_init(&sk->sk_peer_lock);
+
sk->sk_write_pending = 0;
sk->sk_rcvlowat = 1;
sk->sk_rcvtimeo = MAX_SCHEDULE_TIMEOUT;
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 3098710c9c34..05470ca91bd9 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -595,20 +595,42 @@ static void unix_release_sock(struct sock *sk, int embrion)
static void init_peercred(struct sock *sk)
{
- put_pid(sk->sk_peer_pid);
- if (sk->sk_peer_cred)
- put_cred(sk->sk_peer_cred);
+ const struct cred *old_cred;
+ struct pid *old_pid;
+
+ spin_lock(&sk->sk_peer_lock);
+ old_pid = sk->sk_peer_pid;
+ old_cred = sk->sk_peer_cred;
sk->sk_peer_pid = get_pid(task_tgid(current));
sk->sk_peer_cred = get_current_cred();
+ spin_unlock(&sk->sk_peer_lock);
+
+ put_pid(old_pid);
+ put_cred(old_cred);
}
static void copy_peercred(struct sock *sk, struct sock *peersk)
{
- put_pid(sk->sk_peer_pid);
- if (sk->sk_peer_cred)
- put_cred(sk->sk_peer_cred);
+ const struct cred *old_cred;
+ struct pid *old_pid;
+
+ if (sk < peersk) {
+ spin_lock(&sk->sk_peer_lock);
+ spin_lock_nested(&peersk->sk_peer_lock, SINGLE_DEPTH_NESTING);
+ } else {
+ spin_lock(&peersk->sk_peer_lock);
+ spin_lock_nested(&sk->sk_peer_lock, SINGLE_DEPTH_NESTING);
+ }
+ old_pid = sk->sk_peer_pid;
+ old_cred = sk->sk_peer_cred;
sk->sk_peer_pid = get_pid(peersk->sk_peer_pid);
sk->sk_peer_cred = get_cred(peersk->sk_peer_cred);
+
+ spin_unlock(&sk->sk_peer_lock);
+ spin_unlock(&peersk->sk_peer_lock);
+
+ put_pid(old_pid);
+ put_cred(old_cred);
}
static int unix_listen(struct socket *sock, int backlog)
