Content-Type: multipart/mixed; boundary="===============5917360850985843965==" MIME-Version: 1.0 From: Paolo Abeni To: mptcp at lists.01.org Subject: [MPTCP] Re: [RFC PATCH 6/8] mptcp: refactor shutdown and close Date: Mon, 28 Sep 2020 16:47:46 +0200 Message-ID: <8685bde51729292e11779df5dce99ce66b497cce.camel@redhat.com> In-Reply-To: 6ba1cefb2825480dc5b5c3a29a14b63373e44dc6.1600955407.git.pabeni@redhat.com X-Status: X-Keywords: X-UID: 6074 --===============5917360850985843965== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable On Thu, 2020-09-24 at 15:59 +0200, Paolo Abeni wrote: > @@ -417,9 +435,8 @@ static void mptcp_check_data_fin(struct sock *sk) > inet_sk_state_store(sk, TCP_CLOSING); > break; > case TCP_FIN_WAIT2: > - inet_sk_state_store(sk, TCP_CLOSE); > - // @@ Close subflows now? > - break; > + inet_sk_state_store(sk, TCP_TIME_WAIT); > + return; the 'return;' statement slipped-in unintentionally here: this does not send data_fin ack, if the fin in received in fin_wait2 state. I have quite a few pending fixes, so I'll send a v1 soon, even if there are likely more bugs pending. /P --===============5917360850985843965==--