From: Yuchung Cheng <ycheng@google.com>
To: davem@davemloft.net
Cc: netdev@vger.kernel.org, edumazet@google.com,
ncardwell@google.com, nanditad@google.com,
Yuchung Cheng <ycheng@google.com>
Subject: [PATCH net-next v2 00/13] tcp: RACK fast recovery
Date: Thu, 12 Jan 2017 22:11:29 -0800 [thread overview]
Message-ID: <20170113061142.127344-1-ycheng@google.com> (raw)
The patch set enables RACK loss detection (draft-ietf-tcpm-rack-01)
to trigger fast recovery with a reordering timer.
Previously RACK has been running in auxiliary mode where it is
used to detect packet losses once the recovery has triggered by
other algorithms (e.g., FACK). By inspecting packet timestamps,
RACK can start ACK-driven repairs timely. A few similar heuristics
are no longer needed and are either removed or disabled to reduce
the complexity of the Linux TCP loss recovery engine:
1. FACK (Forward Acknowledgement)
2. Early Retransmit (RFC5827)
3. thin_dupack (fast recovery on single DUPACK for thin-streams)
4. NCR (Non-Congestion Robustness RFC4653) (RFC4653)
5. Forward Retransmit
After this change, Linux's loss recovery algorithms consist of
1. Conventional DUPACK threshold approach (RFC6675)
2. RACK and Tail Loss Probe (draft-ietf-tcpm-rack-01)
3. RTO plus F-RTO extension (RFC5682)
The patch set has been tested on Google servers extensively and
presented in several IETF meetings. The data suggests that RACK
successfully improves recovery performance:
https://www.ietf.org/proceedings/97/slides/slides-97-tcpm-draft-ietf-tcpm-rack-01.pdf
https://www.ietf.org/proceedings/96/slides/slides-96-tcpm-3.pdf
Yuchung Cheng (13):
tcp: new helper function for RACK loss detection
tcp: new helper for RACK to detect loss
tcp: record most recent RTT in RACK loss detection
tcp: add reordering timer in RACK loss detection
tcp: use sequence to break TS ties for RACK loss detection
tcp: check undo conditions before detecting losses
tcp: enable RACK loss detection to trigger recovery
tcp: extend F-RTO to catch more spurious timeouts
tcp: remove forward retransmit feature
tcp: remove early retransmit
tcp: remove RFC4653 NCR
tcp: remove thin_dupack feature
tcp: disable fack by default
Documentation/networking/ip-sysctl.txt | 31 +----
include/linux/tcp.h | 8 +-
include/net/inet_connection_sock.h | 4 +-
include/net/tcp.h | 40 ++----
net/ipv4/inet_diag.c | 2 +-
net/ipv4/sysctl_net_ipv4.c | 7 --
net/ipv4/tcp.c | 9 +-
net/ipv4/tcp_input.c | 224 +++++++++++++--------------------
net/ipv4/tcp_ipv4.c | 2 +-
net/ipv4/tcp_metrics.c | 1 -
net/ipv4/tcp_minisocks.c | 1 -
net/ipv4/tcp_output.c | 75 ++---------
net/ipv4/tcp_recovery.c | 148 ++++++++++++++++------
net/ipv4/tcp_timer.c | 4 +-
net/ipv6/tcp_ipv6.c | 2 +-
15 files changed, 237 insertions(+), 321 deletions(-)
--
2.11.0.483.g087da7b7c-goog
next reply other threads:[~2017-01-13 6:12 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-01-13 6:11 Yuchung Cheng [this message]
2017-01-13 6:11 ` [PATCH net-next v2 01/13] tcp: new helper function for RACK loss detection Yuchung Cheng
2017-01-13 6:11 ` [PATCH net-next v2 02/13] tcp: new helper for RACK to detect loss Yuchung Cheng
2017-01-13 6:11 ` [PATCH net-next v2 03/13] tcp: record most recent RTT in RACK loss detection Yuchung Cheng
2017-01-13 6:11 ` [PATCH net-next v2 04/13] tcp: add reordering timer " Yuchung Cheng
2017-01-13 6:11 ` [PATCH net-next v2 05/13] tcp: use sequence to break TS ties for " Yuchung Cheng
2017-01-13 6:11 ` [PATCH net-next v2 06/13] tcp: check undo conditions before detecting losses Yuchung Cheng
2017-01-13 6:11 ` [PATCH net-next v2 07/13] tcp: enable RACK loss detection to trigger recovery Yuchung Cheng
2017-01-13 6:11 ` [PATCH net-next v2 08/13] tcp: extend F-RTO to catch more spurious timeouts Yuchung Cheng
2017-01-13 6:11 ` [PATCH net-next v2 09/13] tcp: remove forward retransmit feature Yuchung Cheng
2017-01-13 6:11 ` [PATCH net-next v2 10/13] tcp: remove early retransmit Yuchung Cheng
2017-01-13 6:11 ` [PATCH net-next v2 11/13] tcp: remove RFC4653 NCR Yuchung Cheng
2017-01-13 6:11 ` [PATCH net-next v2 12/13] tcp: remove thin_dupack feature Yuchung Cheng
2017-01-13 6:11 ` [PATCH net-next v2 13/13] tcp: disable fack by default Yuchung Cheng
2017-01-14 4:19 ` [PATCH net-next v2 00/13] tcp: RACK fast recovery 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=20170113061142.127344-1-ycheng@google.com \
--to=ycheng@google.com \
--cc=davem@davemloft.net \
--cc=edumazet@google.com \
--cc=nanditad@google.com \
--cc=ncardwell@google.com \
--cc=netdev@vger.kernel.org \
/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).