Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
From:   Lorenz Bauer <lmb@cloudflare.com>
To:     kafai@fb.com, Alexei Starovoitov <ast@kernel.org>,
        Daniel Borkmann <daniel@iogearbox.net>,
        "David S. Miller" <davem@davemloft.net>,
        Lorenz Bauer <lmb@cloudflare.com>,
        Joe Stringer <joe@isovalent.com>, netdev@vger.kernel.org,
        bpf@vger.kernel.org, linux-kernel@vger.kernel.org
Cc:     kernel-team@cloudflare.com
Subject: [PATCH bpf v2] net: bpf: don't leak time wait and request sockets
Date:   Fri, 10 Jan 2020 13:23:36 +0000
Message-Id: <20200110132336.26099-1-lmb@cloudflare.com>
In-Reply-To: <20200109115749.12283-1-lmb@cloudflare.com>
References: <20200109115749.12283-1-lmb@cloudflare.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

It's possible to leak time wait and request sockets via the following
BPF pseudo code:
 
  sk = bpf_skc_lookup_tcp(...)
  if (sk)
    bpf_sk_release(sk)

If sk->sk_state is TCP_NEW_SYN_RECV or TCP_TIME_WAIT the refcount taken
by bpf_skc_lookup_tcp is not undone by bpf_sk_release. This is because
sk_flags is re-used for other data in both kinds of sockets. The check

  !sock_flag(sk, SOCK_RCU_FREE)

therefore returns a bogus result. Check that sk_flags is valid by calling
sk_fullsock. Skip checking SOCK_RCU_FREE if we already know that sk is
not a full socket.

Fixes: edbf8c01de5a ("bpf: add skc_lookup_tcp helper")
Fixes: f7355a6c0497 ("bpf: Check sk_fullsock() before returning from bpf_sk_lookup()")
Signed-off-by: Lorenz Bauer <lmb@cloudflare.com>
---
 net/core/filter.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/net/core/filter.c b/net/core/filter.c
index 42fd17c48c5f..41820ba0774c 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -5277,8 +5277,7 @@ __bpf_sk_lookup(struct sk_buff *skb, struct bpf_sock_tuple *tuple, u32 len,
 	if (sk) {
 		sk = sk_to_full_sk(sk);
 		if (!sk_fullsock(sk)) {
-			if (!sock_flag(sk, SOCK_RCU_FREE))
-				sock_gen_put(sk);
+			sock_gen_put(sk);
 			return NULL;
 		}
 	}
@@ -5315,8 +5314,7 @@ bpf_sk_lookup(struct sk_buff *skb, struct bpf_sock_tuple *tuple, u32 len,
 	if (sk) {
 		sk = sk_to_full_sk(sk);
 		if (!sk_fullsock(sk)) {
-			if (!sock_flag(sk, SOCK_RCU_FREE))
-				sock_gen_put(sk);
+			sock_gen_put(sk);
 			return NULL;
 		}
 	}
@@ -5383,7 +5381,8 @@ static const struct bpf_func_proto bpf_sk_lookup_udp_proto = {
 
 BPF_CALL_1(bpf_sk_release, struct sock *, sk)
 {
-	if (!sock_flag(sk, SOCK_RCU_FREE))
+	/* Only full sockets have sk->sk_flags. */
+	if (!sk_fullsock(sk) || !sock_flag(sk, SOCK_RCU_FREE))
 		sock_gen_put(sk);
 	return 0;
 }
-- 
2.20.1