summaryrefslogtreecommitdiff
path: root/include
diff options
context:
space:
mode:
authorEric Dumazet <edumazet@google.com>2018-10-23 21:54:16 +0300
committerDavid S. Miller <davem@davemloft.net>2018-10-24 05:42:44 +0300
commit3f80e08f40cdb308589a49077c87632fa4508b21 (patch)
tree236cff8470641456fb68472d2e17045ee1aafa62 /include
parent68203a67a7024c5d0b8e545d3d370b1fec971551 (diff)
downloadlinux-3f80e08f40cdb308589a49077c87632fa4508b21.tar.xz
tcp: add tcp_reset_xmit_timer() helper
With EDT model, SRTT no longer is inflated by pacing delays. This means that RTO and some other xmit timers might be setup incorrectly. This is particularly visible with either : - Very small enforced pacing rates (SO_MAX_PACING_RATE) - Reduced rto (from the default 200 ms) This can lead to TCP flows aborts in the worst case, or spurious retransmits in other cases. For example, this session gets far more throughput than the requested 80kbit : $ netperf -H 127.0.0.2 -l 100 -- -q 10000 MIGRATED TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 127.0.0.2 () port 0 AF_INET Recv Send Send Socket Socket Message Elapsed Size Size Size Time Throughput bytes bytes bytes secs. 10^6bits/sec 540000 262144 262144 104.00 2.66 With the fix : $ netperf -H 127.0.0.2 -l 100 -- -q 10000 MIGRATED TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 127.0.0.2 () port 0 AF_INET Recv Send Send Socket Socket Message Elapsed Size Size Size Time Throughput bytes bytes bytes secs. 10^6bits/sec 540000 262144 262144 104.00 0.12 EDT allows for better control of rtx timers, since TCP has a better idea of the earliest departure time of each skb in the rtx queue. We only have to eventually add to the timer the difference of the EDT time with current time. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'include')
-rw-r--r--include/net/tcp.h30
1 files changed, 27 insertions, 3 deletions
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 8a61c3e8c15d..a18914d20486 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1245,8 +1245,31 @@ static inline bool tcp_needs_internal_pacing(const struct sock *sk)
return smp_load_acquire(&sk->sk_pacing_status) == SK_PACING_NEEDED;
}
+/* Return in jiffies the delay before one skb is sent.
+ * If @skb is NULL, we look at EDT for next packet being sent on the socket.
+ */
+static inline unsigned long tcp_pacing_delay(const struct sock *sk,
+ const struct sk_buff *skb)
+{
+ s64 pacing_delay = skb ? skb->tstamp : tcp_sk(sk)->tcp_wstamp_ns;
+
+ pacing_delay -= tcp_sk(sk)->tcp_clock_cache;
+
+ return pacing_delay > 0 ? nsecs_to_jiffies(pacing_delay) : 0;
+}
+
+static inline void tcp_reset_xmit_timer(struct sock *sk,
+ const int what,
+ unsigned long when,
+ const unsigned long max_when,
+ const struct sk_buff *skb)
+{
+ inet_csk_reset_xmit_timer(sk, what, when + tcp_pacing_delay(sk, skb),
+ max_when);
+}
+
/* Something is really bad, we could not queue an additional packet,
- * because qdisc is full or receiver sent a 0 window.
+ * because qdisc is full or receiver sent a 0 window, or we are paced.
* We do not want to add fuel to the fire, or abort too early,
* so make sure the timer we arm now is at least 200ms in the future,
* regardless of current icsk_rto value (as it could be ~2ms)
@@ -1268,8 +1291,9 @@ static inline unsigned long tcp_probe0_when(const struct sock *sk,
static inline void tcp_check_probe_timer(struct sock *sk)
{
if (!tcp_sk(sk)->packets_out && !inet_csk(sk)->icsk_pending)
- inet_csk_reset_xmit_timer(sk, ICSK_TIME_PROBE0,
- tcp_probe0_base(sk), TCP_RTO_MAX);
+ tcp_reset_xmit_timer(sk, ICSK_TIME_PROBE0,
+ tcp_probe0_base(sk), TCP_RTO_MAX,
+ NULL);
}
static inline void tcp_init_wl(struct tcp_sock *tp, u32 seq)