From patchwork Wed Jun 16 13:22:18 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Paolo Abeni X-Patchwork-Id: 12325211 X-Patchwork-Delegate: mat@martineau.name Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 7253D71 for ; Wed, 16 Jun 2021 13:22:56 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1623849775; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding; bh=UhG2CF7WNmhW6ZGSfscPGsf2qBQTgTXrJ/YejNVexfs=; b=hjaHd2833fTtEytVWmfVQISAmeAsNK6d2uqCzBQslhTU6izuVYYW9k9I5D8a1G9c0nCT+7 zUpRf6c/MlId3Kg0itcwI1pk2ZnXhKuvFN4YW3GGD8maYKSV3PpmOLxsVjbGLT+Ar+Vo0C /T03D9UDR78HhFyxEctOvcJMOal5074= Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-454-u2QvjNZTO3Ktzfj47a6bgA-1; Wed, 16 Jun 2021 09:22:47 -0400 X-MC-Unique: u2QvjNZTO3Ktzfj47a6bgA-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id 584E0192D796 for ; Wed, 16 Jun 2021 13:22:24 +0000 (UTC) Received: from gerbillo.redhat.com (ovpn-115-78.ams2.redhat.com [10.36.115.78]) by smtp.corp.redhat.com (Postfix) with ESMTP id C6FBB60BF1 for ; Wed, 16 Jun 2021 13:22:23 +0000 (UTC) From: Paolo Abeni To: mptcp@lists.linux.dev Subject: [PATCH mptcp-net] mptcp: avoid race on msk state changes Date: Wed, 16 Jun 2021 15:22:18 +0200 Message-Id: X-Mailing-List: mptcp@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 Authentication-Results: relay.mimecast.com; auth=pass smtp.auth=CUSA124A263 smtp.mailfrom=pabeni@redhat.com X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: redhat.com The msk socket state is currently updated in a few spots without owning the msk socket lock itself. Some of such operations are safe, as they happens before exposing the msk socket to user-space and can't race with other changes. A couple of them, at connect time, can actually race with close() or shutdown(), leaving breaking the socket state machine. This change addresses the issue moving such update under the msk socket lock with the usual: scheme. Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/56 Fixes: 8fd738049ac3 ("mptcp: fallback in case of simultaneous connect") Fixes: c3c123d16c0e ("net: mptcp: don't hang in mptcp_sendmsg() after TCP fallbac") Signed-off-by: Paolo Abeni --- net/mptcp/protocol.c | 2 ++ net/mptcp/protocol.h | 2 ++ net/mptcp/subflow.c | 30 ++++++++++++++++++++++-------- 3 files changed, 26 insertions(+), 8 deletions(-) diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 05c8382aafef..15c3b75516fb 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -2906,6 +2906,8 @@ static void mptcp_release_cb(struct sock *sk) __mptcp_clean_una_wakeup(sk); if (test_and_clear_bit(MPTCP_ERROR_REPORT, &mptcp_sk(sk)->flags)) __mptcp_error_report(sk); + if (test_and_clear_bit(MPTCP_CONNECTED, &mptcp_sk(sk)->flags)) + __mptcp_set_connected(sk); /* push_pending may touch wmem_reserved, ensure we do the cleanup * later diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index 2480db50cbd2..515bb1e6acec 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -112,6 +112,7 @@ #define MPTCP_ERROR_REPORT 8 #define MPTCP_RETRANSMIT 9 #define MPTCP_WORK_SYNC_SETSOCKOPT 10 +#define MPTCP_CONNECTED 11 static inline bool before64(__u64 seq1, __u64 seq2) { @@ -600,6 +601,7 @@ void mptcp_get_options(const struct sock *sk, struct mptcp_options_received *mp_opt); void mptcp_finish_connect(struct sock *sk); +void __mptcp_set_connected(struct sock *sk); static inline bool mptcp_is_fully_established(struct sock *sk) { return inet_sk_state_load(sk) == TCP_ESTABLISHED && diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index 6772802737e4..0b5d4a3eadcd 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -373,6 +373,24 @@ static bool subflow_use_different_dport(struct mptcp_sock *msk, const struct soc return inet_sk(sk)->inet_dport != inet_sk((struct sock *)msk)->inet_dport; } +void __mptcp_set_connected(struct sock *sk) +{ + if (sk->sk_state == TCP_SYN_SENT) { + inet_sk_state_store(sk, TCP_ESTABLISHED); + sk->sk_state_change(sk); + } +} + +static void mptcp_set_connected(struct sock *sk) +{ + mptcp_data_lock(sk); + if (!sock_owned_by_user(sk)) + __mptcp_set_connected(sk); + else + set_bit(MPTCP_CONNECTED, &mptcp_sk(sk)->flags); + mptcp_data_unlock(sk); +} + static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb) { struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk); @@ -381,10 +399,6 @@ static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb) subflow->icsk_af_ops->sk_rx_dst_set(sk, skb); - if (inet_sk_state_load(parent) == TCP_SYN_SENT) { - inet_sk_state_store(parent, TCP_ESTABLISHED); - parent->sk_state_change(parent); - } /* be sure no special action on any packet other than syn-ack */ if (subflow->conn_finished) @@ -417,6 +431,7 @@ static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb) subflow->remote_key); MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_MPCAPABLEACTIVEACK); mptcp_finish_connect(sk); + mptcp_set_connected(parent); } else if (subflow->request_join) { u8 hmac[SHA256_DIGEST_SIZE]; @@ -457,6 +472,7 @@ static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb) } else if (mptcp_check_fallback(sk)) { fallback: mptcp_rcv_space_init(mptcp_sk(parent), sk); + mptcp_set_connected(parent); } return; @@ -564,6 +580,7 @@ static void mptcp_sock_destruct(struct sock *sk) static void mptcp_force_close(struct sock *sk) { + /* the msk is not yet exposed to user-space */ inet_sk_state_store(sk, TCP_CLOSE); sk_common_release(sk); } @@ -1577,10 +1594,7 @@ static void subflow_state_change(struct sock *sk) mptcp_rcv_space_init(mptcp_sk(parent), sk); pr_fallback(mptcp_sk(parent)); subflow->conn_finished = 1; - if (inet_sk_state_load(parent) == TCP_SYN_SENT) { - inet_sk_state_store(parent, TCP_ESTABLISHED); - parent->sk_state_change(parent); - } + mptcp_set_connected(parent); } /* as recvmsg() does not acquire the subflow socket for ssk selection