[PATCH v2 net-next 2/2] mptcp: move recbuf adjustment to recvmsg path

[Florian Westphal]

From: Paolo Abeni <pab...@redhat.com>

Place receive window tuning in the recvmsg path.
This makes sure the size is only increased when userspace consumes data.

Previously we would grow the sk receive buffer towards tcp_rmem[2], now we
so only if userspace reads data.

Simply adjust the msk rcvbuf size to the largest receive buffer of any of
the existing subflows.

Signed-off-by: Paolo Abeni <pab...@redhat.com>
Signed-off-by: Florian Westphal <f...@strlen.de>
---
 This patch is new in v2.

 net/mptcp/protocol.c | 32 ++++++++++++++++++++++++--------
 1 file changed, 24 insertions(+), 8 deletions(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index dbb86cbb9e77..89a35c3fc499 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -190,13 +190,6 @@ static bool __mptcp_move_skbs_from_subflow(struct 
mptcp_sock *msk,
                return false;
        }
 
-       if (!(sk->sk_userlocks & SOCK_RCVBUF_LOCK)) {
-               int rcvbuf = max(ssk->sk_rcvbuf, sk->sk_rcvbuf);
-
-               if (rcvbuf > sk->sk_rcvbuf)
-                       sk->sk_rcvbuf = rcvbuf;
-       }
-
        tp = tcp_sk(ssk);
        do {
                u32 map_remaining, offset;
@@ -933,6 +926,25 @@ static bool __mptcp_move_skbs(struct mptcp_sock *msk)
        return moved > 0;
 }
 
+static void mptcp_rcv_space_adjust(struct mptcp_sock *msk)
+{
+       const struct mptcp_subflow_context *subflow;
+       struct sock *sk = (struct sock *)msk;
+       const struct sock *ssk;
+       int rcvbuf = 0;
+
+       if (sk->sk_userlocks & SOCK_RCVBUF_LOCK)
+               return;
+
+       mptcp_for_each_subflow(msk, subflow) {
+               ssk = mptcp_subflow_tcp_sock(subflow);
+               rcvbuf = max(ssk->sk_rcvbuf, rcvbuf);
+       }
+
+       if (rcvbuf)
+               sk->sk_rcvbuf = rcvbuf;
+}
+
 static int mptcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
                         int nonblock, int flags, int *addr_len)
 {
@@ -962,6 +974,8 @@ static int mptcp_recvmsg(struct sock *sk, struct msghdr 
*msg, size_t len,
        target = sock_rcvlowat(sk, flags & MSG_WAITALL, len);
        __mptcp_flush_join_list(msk);
 
+       mptcp_rcv_space_adjust(msk);
+
        while (len > (size_t)copied) {
                int bytes_read;
 
@@ -975,8 +989,10 @@ static int mptcp_recvmsg(struct sock *sk, struct msghdr 
*msg, size_t len,
                copied += bytes_read;
 
                if (skb_queue_empty(&sk->sk_receive_queue) &&
-                   __mptcp_move_skbs(msk))
+                   __mptcp_move_skbs(msk)) {
+                       mptcp_rcv_space_adjust(msk);
                        continue;
+               }
 
                /* only the master socket status is relevant here. The exit
                 * conditions mirror closely tcp_recvmsg()
-- 
2.26.2