Content-Type: multipart/mixed; boundary="===============3704181833984626075==" MIME-Version: 1.0 From: Florian Westphal To: mptcp at lists.01.org Subject: [MPTCP] Re: [PATCH mptcp-next v2] mptcp: track window announced to peer Date: Thu, 22 Oct 2020 22:52:41 +0200 Message-ID: <20201022205241.GI16895@breakpoint.cc> In-Reply-To: d6c955dacd0fd80adc1e09845ed3f8e6fb11cb0b.camel@redhat.com X-Status: X-Keywords: X-UID: 6408 --===============3704181833984626075== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Paolo Abeni wrote: > On Thu, 2020-10-22 at 12:32 +0200, Florian Westphal wrote: > > OoO handling attemtps to detect when packet is out-of-window by testing > > current ack sequence and remaining space vs. sequence number. > > = > > This doesn't work reliably. Store the highest allowed sequence number > > that we've announced and use it to detect oow packets. > > = > > Do this when mptcp options get written to the packet (wire format). > > For this to work we need to move the write_options call until after sta= ck > > selected a new tcp window. > > = > > Signed-off-by: Florian Westphal > = > I've only a couple of minor comments, see below... > = > > @@ -445,11 +445,11 @@ struct tcp_out_options { > > struct mptcp_out_options mptcp; > > }; > > = > > -static void mptcp_options_write(__be32 *ptr, struct tcp_out_options *o= pts) > > +static void mptcp_options_write(__be32 *ptr, const struct tcp_sock *tp= , struct tcp_out_options *opts) > = > Nit picking: this looks above the 80 char limits ;) Ok, i will break the line. > > @@ -1147,5 +1163,7 @@ void mptcp_write_options(__be32 *ptr, struct mptc= p_out_options *opts) > > put_unaligned_be32(mpext->data_len << 16 | > > TCPOPT_NOP << 8 | TCPOPT_NOP, ptr); > > } > > + > > + mptcp_set_rwin(tp); > = > If I read correctly, we will not update the receive window for pure ack > carrying ipv6 add addr option (and nothing else). Likely not > particulary relevant, but perhaps we could call the above helper for > all egress packet? rx side only updates end of window if MPTCP_DSS_HAS_ACK flag is set. I'll move it outside of the conditional, perhaps its less confusing that way. --===============3704181833984626075==--