6 files changed, 87 insertions, 27 deletions

diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 5c3dd6267ed3..8976ca423a07 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -564,6 +564,40 @@ int inet_rtx_syn_ack(struct sock *parent, struct request_sock *req)
 }
 EXPORT_SYMBOL(inet_rtx_syn_ack);
 
+/* return true if req was found in the syn_table[] */
+static bool reqsk_queue_unlink(struct request_sock_queue *queue,
+                               struct request_sock *req)
+{
+        struct listen_sock *lopt = queue->listen_opt;
+        struct request_sock **prev;
+        bool found = false;
+
+        spin_lock(&queue->syn_wait_lock);
+
+        for (prev = &lopt->syn_table[req->rsk_hash]; *prev != NULL;
+             prev = &(*prev)->dl_next) {
+                if (*prev == req) {
+                        *prev = req->dl_next;
+                        found = true;
+                        break;
+                }
+        }
+
+        spin_unlock(&queue->syn_wait_lock);
+        if (del_timer(&req->rsk_timer))
+                reqsk_put(req);
+        return found;
+}
+
+void inet_csk_reqsk_queue_drop(struct sock *sk, struct request_sock *req)
+{
+        if (reqsk_queue_unlink(&inet_csk(sk)->icsk_accept_queue, req)) {
+                reqsk_queue_removed(&inet_csk(sk)->icsk_accept_queue, req);
+                reqsk_put(req);
+        }
+}
+EXPORT_SYMBOL(inet_csk_reqsk_queue_drop);
+
 static void reqsk_timer_handler(unsigned long data)
 {
         struct request_sock *req = (struct request_sock *)data;
diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c
index a93f260cf24c..05ff44b758df 100644
--- a/net/ipv4/ping.c
+++ b/net/ipv4/ping.c
@@ -158,6 +158,7 @@ void ping_unhash(struct sock *sk)
         if (sk_hashed(sk)) {
                 write_lock_bh(&ping_table.lock);
                 hlist_nulls_del(&sk->sk_nulls_node);
+                sk_nulls_node_init(&sk->sk_nulls_node);
                 sock_put(sk);
                 isk->inet_num = 0;
                 isk->inet_sport = 0;
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index a78540f28276..bff62fc87b8e 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -962,10 +962,7 @@ static void __ip_rt_update_pmtu(struct rtable *rt, struct flowi4 *fl4, u32 mtu)
         if (dst_metric_locked(dst, RTAX_MTU))
                 return;
 
-        if (dst->dev->mtu < mtu)
-                return;
-
-        if (rt->rt_pmtu && rt->rt_pmtu < mtu)
+        if (ipv4_mtu(dst) < mtu)
                 return;
 
         if (mtu < ip_rt_min_pmtu)
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 3571f2be4470..fc1c658ec6c1 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1348,7 +1348,8 @@ static struct sock *tcp_v4_hnd_req(struct sock *sk, struct sk_buff *skb)
         req = inet_csk_search_req(sk, th->source, iph->saddr, iph->daddr);
         if (req) {
                 nsk = tcp_check_req(sk, skb, req, false);
-                reqsk_put(req);
+                if (!nsk)
+                        reqsk_put(req);
                 return nsk;
         }
 
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index 63d6311b5365..e5d7649136fc 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -755,10 +755,11 @@ struct sock *tcp_check_req(struct sock *sk, struct sk_buff *skb,
         if (!child)
                 goto listen_overflow;
 
-        inet_csk_reqsk_queue_unlink(sk, req);
-        inet_csk_reqsk_queue_removed(sk, req);
-
+        inet_csk_reqsk_queue_drop(sk, req);
         inet_csk_reqsk_queue_add(sk, req, child);
+        /* Warning: caller must not call reqsk_put(req);
+         * child stole last reference on it.
+         */
         return child;
 
 listen_overflow:
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 8c8d7e06b72f..a369e8a70b2c 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -2812,39 +2812,65 @@ begin_fwd:
         }
 }
 
-/* Send a fin.  The caller locks the socket for us.  This cannot be
- * allowed to fail queueing a FIN frame under any circumstances.
+/* We allow to exceed memory limits for FIN packets to expedite
+ * connection tear down and (memory) recovery.
+ * Otherwise tcp_send_fin() could be tempted to either delay FIN
+ * or even be forced to close flow without any FIN.
+ */
+static void sk_forced_wmem_schedule(struct sock *sk, int size)
+{
+        int amt, status;
+
+        if (size <= sk->sk_forward_alloc)
+                return;
+        amt = sk_mem_pages(size);
+        sk->sk_forward_alloc += amt * SK_MEM_QUANTUM;
+        sk_memory_allocated_add(sk, amt, &status);
+}
+
+/* Send a FIN. The caller locks the socket for us.
+ * We should try to send a FIN packet really hard, but eventually give up.
  */
 void tcp_send_fin(struct sock *sk)
 {
+        struct sk_buff *skb, *tskb = tcp_write_queue_tail(sk);
         struct tcp_sock *tp = tcp_sk(sk);
-        struct sk_buff *skb = tcp_write_queue_tail(sk);
-        int mss_now;
 
-        /* Optimization, tack on the FIN if we have a queue of
-         * unsent frames.  But be careful about outgoing SACKS
-         * and IP options.
+        /* Optimization, tack on the FIN if we have one skb in write queue and
+         * this skb was not yet sent, or we are under memory pressure.
+         * Note: in the latter case, FIN packet will be sent after a timeout,
+         * as TCP stack thinks it has already been transmitted.
          */
-        mss_now = tcp_current_mss(sk);
-
-        if (tcp_send_head(sk)) {
-                TCP_SKB_CB(skb)->tcp_flags |= TCPHDR_FIN;
-                TCP_SKB_CB(skb)->end_seq++;
+        if (tskb && (tcp_send_head(sk) || sk_under_memory_pressure(sk))) {
+coalesce:
+                TCP_SKB_CB(tskb)->tcp_flags |= TCPHDR_FIN;
+                TCP_SKB_CB(tskb)->end_seq++;
                 tp->write_seq++;
+                if (!tcp_send_head(sk)) {
+                        /* This means tskb was already sent.
+                         * Pretend we included the FIN on previous transmit.
+                         * We need to set tp->snd_nxt to the value it would have
+                         * if FIN had been sent. This is because retransmit path
+                         * does not change tp->snd_nxt.
+                         */
+                        tp->snd_nxt++;
+                        return;
+                }
         } else {
-                /* Socket is locked, keep trying until memory is available. */
-                for (;;) {
-                        skb = sk_stream_alloc_skb(sk, 0, sk->sk_allocation);
-                        if (skb)
-                                break;
-                        yield();
+                skb = alloc_skb_fclone(MAX_TCP_HEADER, sk->sk_allocation);
+                if (unlikely(!skb)) {
+                        if (tskb)
+                                goto coalesce;
+                        return;
                 }
+                skb_reserve(skb, MAX_TCP_HEADER);
+                sk_forced_wmem_schedule(sk, skb->truesize);
                 /* FIN eats a sequence byte, write_seq advanced by tcp_queue_skb(). */
                 tcp_init_nondata_skb(skb, tp->write_seq,
                                      TCPHDR_ACK | TCPHDR_FIN);
                 tcp_queue_skb(sk, skb);
         }
-        __tcp_push_pending_frames(sk, mss_now, TCP_NAGLE_OFF);
+        __tcp_push_pending_frames(sk, tcp_current_mss(sk), TCP_NAGLE_OFF);
 }
 
 /* We get here when a process closes a file descriptor (either due to