mptcp: accurate SIOCOUTQ for fallback socket

The MPTCP SIOCOUTQ implementation is not very accurate in case of fallback: it only measures the data in the MPTCP-level write queue, but it does not take in account the subflow write queue utilization. In case of fallback the first can be empty, while the latter is not. The above produces sporadic self-tests issues and can foul legit user-space application. Fix the issue additionally querying the subflow in case of fallback. Fixes: 644807e ("mptcp: add SIOCINQ, OUTQ and OUTQNSD ioctls") Closes: #260 Reported-by: Matthieu Baerts <[email protected]> Reviewed-by: Mat Martineau <[email protected]> Signed-off-by: Paolo Abeni <[email protected]>
multipath-tcp · Feb 23, 2022 · 7b784a7 · 7b784a7
1 parent 2221b36
commit 7b784a7
Showing 1 changed file with 11 additions and 0 deletions.
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
@@ -3294,6 +3294,17 @@ static int mptcp_ioctl_outq(const struct mptcp_sock *msk, u64 v)
 		return 0;
 
 	delta = msk->write_seq - v;
+	if (__mptcp_check_fallback(msk) && msk->first) {
+		struct tcp_sock *tp = tcp_sk(msk->first);
+
+		/* the first subflow is disconnected after close - see
+		 * __mptcp_close_ssk(). tcp_disconnect() moves the write_seq
+		 * so ignore that status, too.
+		 */
+		if (!((1 << msk->first->sk_state) &
+		      (TCPF_SYN_SENT | TCPF_SYN_RECV | TCPF_CLOSE)))
+			delta += READ_ONCE(tp->write_seq) - tp->snd_una;
+	}
 	if (delta > INT_MAX)
 		delta = INT_MAX;