From: Yuchung Cheng <ycheng@google.com>
To: davem@davemloft.net, edumazet@google.com
Cc: netdev@vger.kernel.org, ncardwell@google.com, soheil@google.com,
Yuchung Cheng <ycheng@google.com>
Subject: [PATCH net-next 6/8] tcp: simplify window probe aborting on USER_TIMEOUT
Date: Wed, 16 Jan 2019 15:05:33 -0800 [thread overview]
Message-ID: <20190116230535.162758-7-ycheng@google.com> (raw)
In-Reply-To: <20190116230535.162758-1-ycheng@google.com>
Previously we use the next unsent skb's timestamp to determine
when to abort a socket stalling on window probes. This no longer
works as skb timestamp reflects the last instead of the first
transmission.
Instead we can estimate how long the socket has been stalling
with the probe count and the exponential backoff behavior.
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Neal Cardwell <ncardwell@google.com>
Reviewed-by: Soheil Hassas Yeganeh <soheil@google.com>
---
net/ipv4/tcp_timer.c | 14 +++++++-------
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index bcc2f5783e57..c36089aa3515 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -333,7 +333,6 @@ static void tcp_probe_timer(struct sock *sk)
struct sk_buff *skb = tcp_send_head(sk);
struct tcp_sock *tp = tcp_sk(sk);
int max_probes;
- u32 start_ts;
if (tp->packets_out || !skb) {
icsk->icsk_probes_out = 0;
@@ -348,12 +347,13 @@ static void tcp_probe_timer(struct sock *sk)
* corresponding system limit. We also implement similar policy when
* we use RTO to probe window in tcp_retransmit_timer().
*/
- start_ts = tcp_skb_timestamp(skb);
- if (!start_ts)
- skb->skb_mstamp_ns = tp->tcp_clock_cache;
- else if (icsk->icsk_user_timeout &&
- (s32)(tcp_time_stamp(tp) - start_ts) > icsk->icsk_user_timeout)
- goto abort;
+ if (icsk->icsk_user_timeout) {
+ u32 elapsed = tcp_model_timeout(sk, icsk->icsk_probes_out,
+ tcp_probe0_base(sk));
+
+ if (elapsed >= icsk->icsk_user_timeout)
+ goto abort;
+ }
max_probes = sock_net(sk)->ipv4.sysctl_tcp_retries2;
if (sock_flag(sk, SOCK_DEAD)) {
--
2.20.1.97.g81188d93c3-goog
next prev parent reply other threads:[~2019-01-16 23:06 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-01-16 23:05 [PATCH net-next 0/8] improving TCP behavior on host congestion Yuchung Cheng
2019-01-16 23:05 ` [PATCH net-next 1/8] tcp: exit if nothing to retransmit on RTO timeout Yuchung Cheng
2019-01-16 23:05 ` [PATCH net-next 2/8] tcp: always timestamp on every skb transmission Yuchung Cheng
2019-01-16 23:05 ` [PATCH net-next 3/8] tcp: always set retrans_stamp on recovery Yuchung Cheng
2019-01-16 23:05 ` [PATCH net-next 4/8] tcp: properly track retry time on passive Fast Open Yuchung Cheng
2019-01-16 23:05 ` [PATCH net-next 5/8] tcp: create a helper to model exponential backoff Yuchung Cheng
2019-01-16 23:05 ` Yuchung Cheng [this message]
2019-01-16 23:05 ` [PATCH net-next 7/8] tcp: retry more conservatively on local congestion Yuchung Cheng
2019-01-16 23:05 ` [PATCH net-next 8/8] tcp: less aggressive window probing " Yuchung Cheng
2019-01-17 23:12 ` [PATCH net-next 0/8] improving TCP behavior on host congestion David Miller
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20190116230535.162758-7-ycheng@google.com \
--to=ycheng@google.com \
--cc=davem@davemloft.net \
--cc=edumazet@google.com \
--cc=ncardwell@google.com \
--cc=netdev@vger.kernel.org \
--cc=soheil@google.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).