diff options
| author | David S. Miller <davem@davemloft.net> | 2023-09-16 13:42:34 +0100 |
|---|---|---|
| committer | David S. Miller <davem@davemloft.net> | 2023-09-16 13:42:34 +0100 |
| commit | fff755e76e64f0274e4d40c8ff273e3bc0ffc8a2 (patch) | |
| tree | 5cd1330b779f689ecfc3e24b6fb8a5baa8600799 /net/ipv4/tcp_timer.c | |
| parent | 50675d84e3995f2606306b5e23e6847273a730e9 (diff) | |
| parent | 3868ab0f192581eff978501a05f3dc2e01541d77 (diff) | |
Merge branch 'TCP_INFO-RTO'
Aananth V says:
====================
tcp: new TCP_INFO stats for RTO events
The 2023 SIGCOMM paper "Improving Network Availability with Protective
ReRoute" has indicated Linux TCP's RTO-triggered txhash rehashing can
effectively reduce application disruption during outages. To better
measure the efficacy of this feature, this patch set adds three more
detailed stats during RTO recovery and exports via TCP_INFO.
Applications and monitoring systems can leverage this data to measure
the network path diversity and end-to-end repair latency during network
outages to improve their network infrastructure.
Patch 1 fixes a bug in TFO SYNACK that we encountered while testing
these new metrics.
Patch 2 adds the new metrics to tcp_sock and tcp_info.
v2: Addressed feedback from a check bot in patch 2 by removing the
inline keyword from the tcp_update_rto_time and tcp_update_rto_stats
functions. Changed a comment in include/net/tcp.h to fit under 80 words.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net/ipv4/tcp_timer.c')
| -rw-r--r-- | net/ipv4/tcp_timer.c | 17 |
1 files changed, 15 insertions, 2 deletions
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index d7d64682b068..3f61c6a70a1f 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -415,6 +415,19 @@ abort: tcp_write_err(sk); } } +static void tcp_update_rto_stats(struct sock *sk) +{ + struct inet_connection_sock *icsk = inet_csk(sk); + struct tcp_sock *tp = tcp_sk(sk); + + if (!icsk->icsk_retransmits) { + tp->total_rto_recoveries++; + tp->rto_stamp = tcp_time_stamp(tp); + } + icsk->icsk_retransmits++; + tp->total_rto++; +} + /* * Timer for Fast Open socket to retransmit SYNACK. Note that the * sk here is the child socket, not the parent (listener) socket. @@ -447,7 +460,7 @@ static void tcp_fastopen_synack_timer(struct sock *sk, struct request_sock *req) */ inet_rtx_syn_ack(sk, req); req->num_timeout++; - icsk->icsk_retransmits++; + tcp_update_rto_stats(sk); if (!tp->retrans_stamp) tp->retrans_stamp = tcp_time_stamp(tp); inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS, @@ -575,7 +588,7 @@ void tcp_retransmit_timer(struct sock *sk) tcp_enter_loss(sk); - icsk->icsk_retransmits++; + tcp_update_rto_stats(sk); if (tcp_retransmit_skb(sk, tcp_rtx_queue_head(sk), 1) > 0) { /* Retransmission failed because of local congestion, * Let senders fight for local resources conservatively. |