Remove mention of pipeACK #4190
Conversation
ianswett
added
editorial
draft-ietf-quic-recovery.md
Outdated
| determine if the congestion window is sufficiently utilized. | ||
| A sender can use a variety of mechnaisms to determine if the congestion window | ||
| is sufficiently utuilized. For example, the pipeACK method described in | ||
| Section 4.3 of {{?RFC7661}}. |
There was a problem hiding this comment.
I would actually rephrase this part completely (or maybe just remove it entirely as there is another reference two paragraphs down). RFC7661 mainly uses the term "application-limited", however, it does not really propose a mechanism to exactly detect that a connection is application-limited, but rather proposes additional mechanism to even decrease the congestion window when no validated. It further says that the cwnd window MUST NOT be increase when not cwnd-limited and that cwnd-limitation is detected based on FlightSize (or other mechanism, see also section 4.5.3 of RFC7661) as also described in the previous paragraph in the recovery draft:
A sender that is not cwnd-limited MUST NOT increase the cwnd
when ACK packets are received in this phase
So I don't think RFC7661 actually provides any different meachanism to detect if the cwnd is sufficiently utilised.
Please note also that RFC7661 actually says MUST NOT while the QUIC recovery draft has only a SHOULD NOT...
There was a problem hiding this comment.
Good point, I'll remove this paragraph.
Co-authored-by: mirjak <mirja.kuehlewind@ericsson.com>
Co-authored-by: mirjak <mirja.kuehlewind@ericsson.com>
ianswett
changed the title
ianswett
removed
the
editorial
I'm not sure if all of them need to be normative or not?
janaiyengar
added
the
editorial
Fixes #4178 by removing a paragraph mentioning pipeACK and moving RFC3168 to a normative reference.