mptcp: clean-up the rtx path

After the previous patch we can easily avoid invoking
the workqueue to perform the retransmission, if the
msk socket lock is held at rtx timer expiration.

This also simplifies the relevant code.

Co-developed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 1590b9d..171b775 100644 (file)
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -2047,28 +2047,21 @@ out_err:
        return copied;
 }
 
-static void mptcp_retransmit_handler(struct sock *sk)
-{
-       struct mptcp_sock *msk = mptcp_sk(sk);
-
-       set_bit(MPTCP_WORK_RTX, &msk->flags);
-       mptcp_schedule_work(sk);
-}
-
 static void mptcp_retransmit_timer(struct timer_list *t)
 {
        struct inet_connection_sock *icsk = from_timer(icsk, t,
                                                       icsk_retransmit_timer);
        struct sock *sk = &icsk->icsk_inet.sk;
+       struct mptcp_sock *msk = mptcp_sk(sk);
 
        bh_lock_sock(sk);
        if (!sock_owned_by_user(sk)) {
-               mptcp_retransmit_handler(sk);
+               /* we need a process context to retransmit */
+               if (!test_and_set_bit(MPTCP_WORK_RTX, &msk->flags))
+                       mptcp_schedule_work(sk);
        } else {
                /* delegate our work to tcp_release_cb() */
-               if (!test_and_set_bit(TCP_WRITE_TIMER_DEFERRED,
-                                     &sk->sk_tsq_flags))
-                       sock_hold(sk);
+               set_bit(MPTCP_RETRANSMIT, &msk->flags);
        }
        bh_unlock_sock(sk);
        sock_put(sk);
@@ -2958,17 +2951,16 @@ void __mptcp_check_push(struct sock *sk, struct sock *ssk)
        }
 }
 
-#define MPTCP_DEFERRED_ALL (TCPF_WRITE_TIMER_DEFERRED)
-
 /* processes deferred events and flush wmem */
 static void mptcp_release_cb(struct sock *sk)
 {
-       unsigned long flags, nflags;
-
        for (;;) {
-               flags = 0;
+               unsigned long flags = 0;
+
                if (test_and_clear_bit(MPTCP_PUSH_PENDING, &mptcp_sk(sk)->flags))
                        flags |= BIT(MPTCP_PUSH_PENDING);
+               if (test_and_clear_bit(MPTCP_RETRANSMIT, &mptcp_sk(sk)->flags))
+                       flags |= BIT(MPTCP_RETRANSMIT);
                if (!flags)
                        break;
 
@@ -2983,6 +2975,8 @@ static void mptcp_release_cb(struct sock *sk)
                spin_unlock_bh(&sk->sk_lock.slock);
                if (flags & BIT(MPTCP_PUSH_PENDING))
                        __mptcp_push_pending(sk, 0);
+               if (flags & BIT(MPTCP_RETRANSMIT))
+                       __mptcp_retrans(sk);
 
                cond_resched();
                spin_lock_bh(&sk->sk_lock.slock);
@@ -2998,20 +2992,6 @@ static void mptcp_release_cb(struct sock *sk)
         */
        __mptcp_update_wmem(sk);
        __mptcp_update_rmem(sk);
-
-       do {
-               flags = sk->sk_tsq_flags;
-               if (!(flags & MPTCP_DEFERRED_ALL))
-                       return;
-               nflags = flags & ~MPTCP_DEFERRED_ALL;
-       } while (cmpxchg(&sk->sk_tsq_flags, flags, nflags) != flags);
-
-       sock_release_ownership(sk);
-
-       if (flags & TCPF_WRITE_TIMER_DEFERRED) {
-               mptcp_retransmit_handler(sk);
-               __sock_put(sk);
-       }
 }
 
 void mptcp_subflow_process_delegated(struct sock *ssk)

diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 1111a99..0116308 100644 (file)
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -104,6 +104,7 @@
 #define MPTCP_PUSH_PENDING     6
 #define MPTCP_CLEAN_UNA                7
 #define MPTCP_ERROR_REPORT     8
+#define MPTCP_RETRANSMIT       9
 
 static inline bool before64(__u64 seq1, __u64 seq2)
 {