Editorial suggestions on the Streams section #2730

ianswett · 2019-05-20T21:49:52Z

Feel free to take the suggestions you like and revert or fix the others.

martinthomson

A few comments, some of which are important enough to fix.

martinthomson · 2019-05-22T12:50:21Z

draft-ietf-quic-transport.md

@@ -187,8 +187,9 @@ QUIC:

 QUIC packet:

-: The smallest unit of QUIC that can be encapsulated in a UDP datagram. Multiple
-  QUIC packets can be encapsulated in a single UDP datagram.
+: A complete processable payload with a packet type, encryption level and


I don't understand this. I agree that the existing definition isn't good, but "payload" means something else.

Payload was my best suggestion here, so I'm open to others.

Structure? Sequence of octets? Or drop it entirely and define a packet as a header, payload, and authentication tag in that order.

Thanks Mike, I updated it to header, payload, and authentication tag.

martinthomson · 2019-05-22T12:50:56Z

draft-ietf-quic-transport.md

-abstraction.
+application. Streams can be unidirectional or bidirecational.  An alternative
+view of QUIC unidirectional streams is a "message" abstraction of unlimited
+length.


Streams aren't unlimited length.

practically unlimited?

martinthomson · 2019-05-22T12:53:05Z

draft-ietf-quic-transport.md

-has all stream data.  Any STREAM or STREAM_DATA_BLOCKED frames it receives for
-the stream can be discarded.
+frame that causes the transition to "Size Known".  Any STREAM or
+STREAM_DATA_BLOCKED frames it receives for the stream can be discarded.


I realize here that the last sentence here, if taken out of context, might be misleading.

Updated, PTAL

draft-ietf-quic-transport.md

martinthomson · 2019-05-22T12:53:53Z

draft-ietf-quic-transport.md

-If the STOP_SENDING frame is received on a stream that is already in the
-"Data Sent" state, an endpoint that wishes to cease retransmission of
-previously-sent STREAM frames on that stream MUST first send a RESET_STREAM
-frame.


Upon reading, I thought it was redundant with the surrounding text, but I'll restore it for this PR.

draft-ietf-quic-transport.md

martinthomson · 2019-05-22T12:59:37Z

draft-ietf-quic-transport.md

@@ -861,13 +855,12 @@ If either is received, the connection MUST be closed immediately with a
 connection error of type STREAM_LIMIT_ERROR (see {{immediate-close}}).

 Endpoints MUST NOT exceed the limit set by their peer.  An endpoint that
-receives a STREAM frame with a stream ID exceeding the limit it has sent MUST
+receives a frame with a stream ID exceeding the limit it has sent MUST


You are talking here about trying to include STREAM_BLOCKED and RESET_STREAM as well as STREAM? But by making it generic you might accidentally capture STOP_SENDING.

That's true, but I think that's a feature, not a bug.

Perhaps, but wouldn't this then be a design change? This is an editorial PR.

Fair point, I'll revert that and if we think it's worth changing, I'll open a design issue for it.

Martin's suggestions

ianswett

Thanks Martin, I made some updates based on your suggestions.

ianswett · 2019-06-19T01:27:57Z

draft-ietf-quic-transport.md

@@ -187,8 +187,9 @@ QUIC:

 QUIC packet:

-: The smallest unit of QUIC that can be encapsulated in a UDP datagram. Multiple
-  QUIC packets can be encapsulated in a single UDP datagram.
+: A complete processable payload with a packet type, encryption level and


Payload was my best suggestion here, so I'm open to others.

ianswett · 2019-06-19T01:47:31Z

draft-ietf-quic-transport.md

-abstraction.
+application. Streams can be unidirectional or bidirecational.  An alternative
+view of QUIC unidirectional streams is a "message" abstraction of unlimited
+length.


practically unlimited?

ianswett · 2019-06-19T01:50:00Z

draft-ietf-quic-transport.md

-has all stream data.  Any STREAM or STREAM_DATA_BLOCKED frames it receives for
-the stream can be discarded.
+frame that causes the transition to "Size Known".  Any STREAM or
+STREAM_DATA_BLOCKED frames it receives for the stream can be discarded.


Updated, PTAL

draft-ietf-quic-transport.md

ianswett · 2019-06-19T02:06:23Z

draft-ietf-quic-transport.md

@@ -861,13 +855,12 @@ If either is received, the connection MUST be closed immediately with a
 connection error of type STREAM_LIMIT_ERROR (see {{immediate-close}}).

 Endpoints MUST NOT exceed the limit set by their peer.  An endpoint that
-receives a STREAM frame with a stream ID exceeding the limit it has sent MUST
+receives a frame with a stream ID exceeding the limit it has sent MUST


That's true, but I think that's a feature, not a bug.

ianswett · 2019-06-19T02:09:36Z

draft-ietf-quic-transport.md

-If the STOP_SENDING frame is received on a stream that is already in the
-"Data Sent" state, an endpoint that wishes to cease retransmission of
-previously-sent STREAM frames on that stream MUST first send a RESET_STREAM
-frame.


Upon reading, I thought it was redundant with the surrounding text, but I'll restore it for this PR.

mikkelfj · 2019-06-19T06:28:04Z

draft-ietf-quic-transport.md

-  QUIC packets can be encapsulated in a single UDP datagram.
+: A complete processable payload with a packet type, encryption level and
+  packet number. Multiple QUIC packets can be encapsulated in a single UDP
+  datagram.



An authenticated and encrypted container of payload usually carrying a packet number and sometimes a connection identifier and version information in its header.

draft-ietf-quic-transport.md

mikkelfj · 2019-06-19T06:32:27Z

draft-ietf-quic-transport.md

-it relies on receiving priority information from the application that uses QUIC.
+QUIC does not provide a mechanism for exchanging prioritization information.
+Instead it relies on receiving priority information from the application that
+uses QUIC.


Suggested change

uses QUIC.

uses QUIC in an implementation specific manner.

implementation-specific, with the hyphen

I added this, then removed it because it read somewhat redundantly.

MikeBishop · 2019-06-19T16:24:56Z

draft-ietf-quic-transport.md

@@ -187,8 +187,9 @@ QUIC:

 QUIC packet:

-: The smallest unit of QUIC that can be encapsulated in a UDP datagram. Multiple
-  QUIC packets can be encapsulated in a single UDP datagram.
+: A complete processable payload with a packet type, encryption level and


Structure? Sequence of octets? Or drop it entirely and define a packet as a header, payload, and authentication tag in that order.

draft-ietf-quic-transport.md

MikeBishop · 2019-06-19T16:26:03Z

draft-ietf-quic-transport.md

-it relies on receiving priority information from the application that uses QUIC.
+QUIC does not provide a mechanism for exchanging prioritization information.
+Instead it relies on receiving priority information from the application that
+uses QUIC.


implementation-specific, with the hyphen

draft-ietf-quic-transport.md

ianswett

Thanks Mike, I addressed or accepted most of your comments.

draft-ietf-quic-transport.md

ianswett · 2019-06-21T17:37:27Z

draft-ietf-quic-transport.md

-it relies on receiving priority information from the application that uses QUIC.
+QUIC does not provide a mechanism for exchanging prioritization information.
+Instead it relies on receiving priority information from the application that
+uses QUIC.


I added this, then removed it because it read somewhat redundantly.

martinthomson · 2019-06-25T22:24:18Z

draft-ietf-quic-transport.md

-  packet number. Multiple QUIC packets can be encapsulated in a single UDP
-  datagram.
+: A complete processable unit of QUIC consisting of a header, payload, and
+  authentication tag.  Multiple QUIC packets can be encapsulated in a single


Not all packets have an authentication tag. A single atomic unit is really the only thing we can say.

Editorial suggestions on the Streams section

fed5f29

ianswett added the -transport label May 20, 2019

ianswett requested review from martinthomson and janaiyengar May 20, 2019 21:49

martinthomson requested changes May 22, 2019

View reviewed changes

Update draft-ietf-quic-transport.md

483a757

Martin's suggestions

ianswett commented Jun 19, 2019

View reviewed changes

ianswett added 2 commits June 18, 2019 22:20

Update draft-ietf-quic-transport.md

e879587

Update draft-ietf-quic-transport.md

f546d20

mikkelfj reviewed Jun 19, 2019

View reviewed changes

draft-ietf-quic-transport.md Show resolved Hide resolved

mikkelfj reviewed Jun 19, 2019

View reviewed changes

MikeBishop reviewed Jun 19, 2019

View reviewed changes

martinthomson approved these changes Jun 20, 2019

View reviewed changes

Mike's comments

4c32ea3

ianswett commented Jun 21, 2019

View reviewed changes

Update draft-ietf-quic-transport.md

fea52b2

ianswett mentioned this pull request Jun 21, 2019

Exceeding Max STREAM ID should be a connection error #2825

Closed

martinthomson reviewed Jun 25, 2019

View reviewed changes

Update draft-ietf-quic-transport.md

bbe2b80

martinthomson merged commit dd947bc into master Jun 25, 2019

martinthomson added the editorial An issue that does not affect the design of the protocol; does not require consensus. label Jun 25, 2019

martinthomson deleted the ianswett-streams-editorial branch June 26, 2019 03:44

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Editorial suggestions on the Streams section #2730

Editorial suggestions on the Streams section #2730

ianswett commented May 20, 2019 •

edited

martinthomson left a comment

martinthomson May 22, 2019

ianswett Jun 19, 2019

MikeBishop Jun 19, 2019

ianswett Jun 21, 2019

martinthomson May 22, 2019

ianswett Jun 19, 2019

martinthomson May 22, 2019

ianswett Jun 19, 2019

martinthomson May 22, 2019

ianswett Jun 19, 2019

martinthomson May 22, 2019

ianswett Jun 19, 2019

MikeBishop Jun 21, 2019

ianswett Jun 21, 2019

ianswett left a comment

ianswett Jun 19, 2019

ianswett Jun 19, 2019

ianswett Jun 19, 2019

ianswett Jun 19, 2019

ianswett Jun 19, 2019

mikkelfj Jun 19, 2019

mikkelfj Jun 19, 2019

MikeBishop Jun 19, 2019

ianswett Jun 21, 2019

MikeBishop Jun 19, 2019

MikeBishop Jun 19, 2019

ianswett left a comment

ianswett Jun 21, 2019

martinthomson Jun 25, 2019

Editorial suggestions on the Streams section #2730

Editorial suggestions on the Streams section #2730

Conversation

ianswett commented May 20, 2019 • edited

martinthomson left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

ianswett left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

ianswett left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

ianswett commented May 20, 2019 •

edited