Home Home > GIT Browse
summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorEric Dumazet <edumazet@google.com>2018-12-06 09:58:24 -0800
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>2018-12-17 09:28:55 +0100
commit4465b31b75490c038777cfccfbdf2ba8e9e5286a (patch)
tree479a879d372a0517bc895331e3000db44961b3d8
parent01a166012cd16b9ce147744c514f228df6cf495e (diff)
tcp: lack of available data can also cause TSO defer
commit f9bfe4e6a9d08d405fe7b081ee9a13e649c97ecf upstream. tcp_tso_should_defer() can return true in three different cases : 1) We are cwnd-limited 2) We are rwnd-limited 3) We are application limited. Neal pointed out that my recent fix went too far, since it assumed that if we were not in 1) case, we must be rwnd-limited Fix this by properly populating the is_cwnd_limited and is_rwnd_limited booleans. After this change, we can finally move the silly check for FIN flag only for the application-limited case. The same move for EOR bit will be handled in net-next, since commit 1c09f7d073b1 ("tcp: do not try to defer skbs with eor mark (MSG_EOR)") is scheduled for linux-4.21 Tested by running 200 concurrent netperf -t TCP_RR -- -r 60000,100 and checking none of them was rwnd_limited in the chrono_stat output from "ss -ti" command. Fixes: 41727549de3e ("tcp: Do not underestimate rwnd_limited") Signed-off-by: Eric Dumazet <edumazet@google.com> Suggested-by: Neal Cardwell <ncardwell@google.com> Reviewed-by: Neal Cardwell <ncardwell@google.com> Acked-by: Soheil Hassas Yeganeh <soheil@google.com> Reviewed-by: Yuchung Cheng <ycheng@google.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-rw-r--r--net/ipv4/tcp_output.c35
1 files changed, 24 insertions, 11 deletions
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 1b31b0a1c7fa..24bad638c2ec 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -1885,7 +1885,9 @@ static int tso_fragment(struct sock *sk, struct sk_buff *skb, unsigned int len,
* This algorithm is from John Heffner.
*/
static bool tcp_tso_should_defer(struct sock *sk, struct sk_buff *skb,
- bool *is_cwnd_limited, u32 max_segs)
+ bool *is_cwnd_limited,
+ bool *is_rwnd_limited,
+ u32 max_segs)
{
const struct inet_connection_sock *icsk = inet_csk(sk);
u32 age, send_win, cong_win, limit, in_flight;
@@ -1893,9 +1895,6 @@ static bool tcp_tso_should_defer(struct sock *sk, struct sk_buff *skb,
struct sk_buff *head;
int win_divisor;
- if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN)
- goto send_now;
-
if (icsk->icsk_ca_state >= TCP_CA_Recovery)
goto send_now;
@@ -1951,10 +1950,27 @@ static bool tcp_tso_should_defer(struct sock *sk, struct sk_buff *skb,
if (age < (tp->srtt_us >> 4))
goto send_now;
- /* Ok, it looks like it is advisable to defer. */
+ /* Ok, it looks like it is advisable to defer.
+ * Three cases are tracked :
+ * 1) We are cwnd-limited
+ * 2) We are rwnd-limited
+ * 3) We are application limited.
+ */
+ if (cong_win < send_win) {
+ if (cong_win <= skb->len) {
+ *is_cwnd_limited = true;
+ return true;
+ }
+ } else {
+ if (send_win <= skb->len) {
+ *is_rwnd_limited = true;
+ return true;
+ }
+ }
- if (cong_win < send_win && cong_win <= skb->len)
- *is_cwnd_limited = true;
+ /* If this packet won't get more data, do not wait. */
+ if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN)
+ goto send_now;
return true;
@@ -2328,11 +2344,8 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle,
} else {
if (!push_one &&
tcp_tso_should_defer(sk, skb, &is_cwnd_limited,
- max_segs)) {
- if (!is_cwnd_limited)
- is_rwnd_limited = true;
+ &is_rwnd_limited, max_segs))
break;
- }
}
limit = mss_now;