Details

Reviewers

rgrimes
tuexen
jtl
cc
bcr
pauamma_gundo.com

Group Reviewers

transport
manpages

Summary

RFC3168 only allows regular data segments to be sent as
ECN-capable transport (ECT). Control (RST, FIN), pure ACK
and retransmissions are explicitly not to be sent with ECT.

However, it has been shown over the years, that this
approach stifles the usefulness of ECN, in particular when
deploying DCTCP in controlled environments.

Current work in the IETF (draft-generalized-ecn, also known
as ECN++) aims to allow all control and retransmission
packets to be marked ECT. In the case of retransmissions,
the only standardized recourse when one of those is dropped,
is to wait for the RTO timer to expire - including the
penalty imposed by having to suffer an RTO.

Furthermore, allowing control and pure ACKs to convey
additional congestion signals on the return path enables
features like AckCC (RFC5690).

Also, adding code to make non-session related control packets
(eg. RST to non-listening ports) appear idential with those in
response to recently used sessions, to prevent sidechannel
information leakage.

Test Plan

Attaching packetdrill scripts for full coverage

Diff Detail

Lint

Lint Passed

Unit

No Test Coverage

Build Status

Buildable 32878
Build 30281: arc lint + arc unit

Event Timeline

There are a very large number of changes, so older changes are hidden. Show Older Changes

Harbormaster completed remote builds in B36805: Diff 83546.Feb 8 2021, 3:02 PM

rscheff retitled this revision from Implement ECN+ (RFC5562) and ECN++ feature (draft-generalized-ecn) to Implement ECN++ (draft-generalized-ecn) .Feb 8 2021, 3:05 PM

rscheff edited the summary of this revision. (Show Details)

updated timestamp in man file

Harbormaster completed remote builds in B37381: Diff 84707.Feb 25 2021, 7:38 PM

rebase
streamline fastpath check
update man date

Harbormaster completed remote builds in B38429: Diff 87067.Apr 8 2021, 7:49 PM

have to explicitly check for ECN SYN

Harbormaster completed remote builds in B38433: Diff 87071.Apr 8 2021, 8:15 PM

add TF2_ECN_PLUSPLUS flag to reduce cacheline churn in fastpath when checking V_ecn_generalized

Harbormaster completed remote builds in B38645: Diff 87615.Apr 17 2021, 2:44 PM

rebase

Harbormaster completed remote builds in B39093: Diff 88943.May 10 2021, 1:39 PM

rscheff mentioned this in D32373: tcp: Add hystart-plus to cc_newreno and rack..Oct 19 2021, 5:48 PM

rebase
bump tcp.4 datestamp

Herald added a subscriber: glebius. · View Herald TranscriptJan 28 2022, 8:43 AM

Harbormaster completed remote builds in B44136: Diff 102035.Jan 28 2022, 8:44 AM

pauamma_gundo.com added a subscriber: pauamma_gundo.com.Jan 28 2022, 11:38 PM

pauamma_gundo.com added inline comments.

share/man/man4/tcp.4
671
677–678	Typo and move full stop to a more natural place.

Current work in the IETF (draft-generalized-ecn, also known as ECN++)

http://datatracker.ietf.org/doc/draft-generalized-ecn does not return a document.
The IETF TSV WG is where I would expect any such work to be done, but
https://datatracker.ietf.org/wg/tsvwg/documents/ does not appear to show
any documents that might be given the moniker "ECN++". Could you provide
a better reference or link for the draft specification that is being implemented?

Interesting. I could find this right away: ECN++: Adding Explicit Congestion Notification (ECN) to TCP Control Packets

Here is the history

fix man page typos

Harbormaster completed remote builds in B44185: Diff 102150.Jan 31 2022, 7:26 AM

Thanks!

In D23230#771332, @rscheff wrote:

Interesting. I could find this right away: ECN++: Adding Explicit Congestion Notification (ECN) to TCP Control Packets

Here is the history

Ah, a TCP-specific draft, not generic to all ECN usage. So I was looking in the wrong place :)
Thanks for the pointer.

LGTM now.

In D23230#771360, @pauamma_gundo.com wrote:

LGTM now.

By which I mean, the language in the manual page change does. Can't say more.

Typo in man page.

share/man/man4/tcp.4
683	s/identially/identically/

fix typo in man page

Harbormaster completed remote builds in B44190: Diff 102159.Jan 31 2022, 2:50 PM

OK for the manpages part of the change.

rebase
simplify the fastpath check if ECT should be set.
streamline fastpath check
update man date
have to explicitly check for ECN SYN
add TF2_ECN_PLUSPLUS flag to reduce cacheline churn in fastpath when checking V_ecn_generalized
bump tcp.4 datestamp
fix man page typos
fix typo in man page
update to use the common tcp_ecn.c sourcefile
fix types

Harbormaster completed remote builds in B44574: Diff 103223.Feb 25 2022, 1:07 PM

pauamma_gundo.com added inline comments.Feb 25 2022, 4:57 PM

share/man/man4/tcp.4
682–683	Can you explain what you mean by that? I'm arguably not the indended audience, but if I don't understand it, there's a chance some of the intended audience and or some interested readers won't either.

rscheff added inline comments.Feb 25 2022, 5:13 PM

share/man/man4/tcp.4
682–683	When a host receives a TCP packet to a port which is not listening, or no connection exists, or the header information is in some other way not acceptable, the host may respond with a RST (reset) packet. Some of these RST packets are sent from "regular" TCP processing (e.g. outside the sequence window) and others from non-open ports. Making these distinctable by carrying different markings - depending which code path was sending them - would give clues as to what ports/services may be reachable, and give rise to more targetted attacks. As part if this change, when introducing this new capability (modified TCP header markings for sessions), the irregular codepath is addressed too, such that a 3rd party can not deduct which of the two different paths was sending out the RST, reducing an information side-channel.

pauamma_gundo.com added inline comments.Feb 26 2022, 12:09 AM

share/man/man4/tcp.4
682–683	Gotcha, thanks. Would "This value also uses ECN for RST replies to probes of non-open ports." mean the same? It looks clearer to me.

rebase
make tcp_ecn_get_ace static inline
include accecn for ecn++
fix ecn bits in timewait
fix syn-sent ip_ecn marking

Harbormaster completed remote builds in B47606: Diff 111188.Sep 29 2022, 2:58 PM

rscheff marked 3 inline comments as done.Sep 29 2022, 3:07 PM

Mnor wording nit, can be fixed on commit.

share/man/man4/tcp.4
503–504

update man page wording

Harbormaster completed remote builds in B47665: Diff 111294.Oct 1 2022, 8:55 AM

pauamma_gundo.com accepted this revision.Oct 5 2022, 4:56 PM

include ect0 for tcp_respond()

Harbormaster completed remote builds in B47828: Diff 111754.Oct 13 2022, 8:43 AM

remove logging

Harbormaster completed remote builds in B47829: Diff 111756.Oct 13 2022, 8:51 AM

Manual page unchanged besides the bumped Dd.

rebase
bump man date
remove duplicate code

Harbormaster completed remote builds in B48231: Diff 112774.Nov 8 2022, 12:11 PM

pauamma_gundo.com added inline comments.Nov 9 2022, 12:58 AM

share/man/man4/tcp.4
534–535	Was removing this intended as part of this review?

rscheff added inline comments.Nov 9 2022, 9:12 AM

share/man/man4/tcp.4
534–535	I'm confused, there is no change here on reviews.freebsd.org; it may be that this diff was uploaded just prior to a flurry of tcp related (including man page) changes, which I have not yet rebased the patch to...