tcp: GSO should be TSQ friendly

I noticed that TSQ (TCP Small queues) was less effective when TSO is
turned off, and GSO is on. If BQL is not enabled, TSQ has then no
effect.

It turns out the GSO engine frees the original gso_skb at the time the
fragments are generated and queued to the NIC.

We should instead call the tcp_wfree() destructor for the last fragment,
to keep the flow control as intended in TSQ. This effectively limits
the number of queued packets on qdisc + NIC layers.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Tom Herbert <therbert@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Cc: Nandita Dukkipati <nanditad@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

2 files changed, 13 insertions, 1 deletions

diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index a96f7b586277..963bda18486f 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -2885,6 +2885,7 @@ struct sk_buff *tcp_tso_segment(struct sk_buff *skb,
         __be32 delta;
         unsigned int oldlen;
         unsigned int mss;
+        struct sk_buff *gso_skb = skb;
 
         if (!pskb_may_pull(skb, sizeof(*th)))
                 goto out;
@@ -2953,6 +2954,17 @@ struct sk_buff *tcp_tso_segment(struct sk_buff *skb,
                 th->cwr = 0;
         } while (skb->next);
 
+        /* Following permits TCP Small Queues to work well with GSO :
+         * The callback to TCP stack will be called at the time last frag
+         * is freed at TX completion, and not right now when gso_skb
+         * is freed by GSO engine
+         */
+        if (gso_skb->destructor == tcp_wfree) {
+                swap(gso_skb->sk, skb->sk);
+                swap(gso_skb->destructor, skb->destructor);
+                swap(gso_skb->truesize, skb->truesize);
+        }
+
         delta = htonl(oldlen + (skb->tail - skb->transport_header) +
                       skb->data_len);
         th->check = ~csum_fold((__force __wsum)((__force u32)th->check +
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index af354c98fdb5..d12694353540 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -787,7 +787,7 @@ void __init tcp_tasklet_init(void)
  * We cant xmit new skbs from this context, as we might already
  * hold qdisc lock.
  */
-static void tcp_wfree(struct sk_buff *skb)
+void tcp_wfree(struct sk_buff *skb)
 {
         struct sock *sk = skb->sk;
         struct tcp_sock *tp = tcp_sk(sk);