[Ecn-sane] more edge cases for ecn?

[Ecn-sane] more edge cases for ecn?

[Dave Taht]

https://linuxplumbersconf.org/event/11/contributions/943/

-- 
Fixing Starlink's Latencies: https://www.youtube.com/watch?v=c9gLo6Xrwgw

Dave Täht CEO, TekLibre, LLC

Re: [Ecn-sane] more edge cases for ecn?

[Rodney W. Grimes]

> https://linuxplumbersconf.org/event/11/contributions/943/

Dave,
	The people that wrote this need to be informed about
how to use the RFC's and the IETF datatracker.  First and
formost they start off CITING rfc1349 which has been OBSOLETED
by RFC2474, in 1998 no less, so anything and everthing refering
to rfc1349 is refering to the wrong document.  Saying there is
a "conflict" between two documents when one was obsoleted by
the other is just non-sense.

The fact that LINUX still has RFC1349 TOS stuff in it is
what is "conflicted", not the RFC's.   Linux continues
to use and support a now 23 year obsoleted model of the
byte being discussed.  This lack of following the updates
in the standards is infact causing the current ECN usage
some issues.

I'll note that at least FreeBSD, and I believe the other BSD's
have some vestigaes of this stuff around, but it is not
in use as far as I can tell.  Ie, there are defines, but
nothing is using them.  And IIRC the one bit that is defanitly
different no longer even has a define.

Regards,
Rod Grimes

> Dave T?ht CEO, TekLibre, LLC
-- 
Rod Grimes                                                 rgrimes@freebsd.org

Re: [Ecn-sane] more edge cases for ecn?

[Dave Taht]

I am very happy to hear someone other than us, is going to tackle
this, and discuss it in this talk.

Usually the problem is the other way around, standards not keeping up
with linux.

On Sun, Sep 19, 2021 at 9:21 AM Rodney W. Grimes
<4bone@gndrsh.dnsmgr.net> wrote:
>
> > https://linuxplumbersconf.org/event/11/contributions/943/
>
> Dave,
>         The people that wrote this need to be informed about
> how to use the RFC's and the IETF datatracker.  First and
> formost they start off CITING rfc1349 which has been OBSOLETED
> by RFC2474, in 1998 no less, so anything and everthing refering
> to rfc1349 is refering to the wrong document.  Saying there is
> a "conflict" between two documents when one was obsoleted by
> the other is just non-sense.
>
> The fact that LINUX still has RFC1349 TOS stuff in it is
> what is "conflicted", not the RFC's.   Linux continues
> to use and support a now 23 year obsoleted model of the
> byte being discussed.  This lack of following the updates
> in the standards is infact causing the current ECN usage
> some issues.
>
> I'll note that at least FreeBSD, and I believe the other BSD's
> have some vestigaes of this stuff around, but it is not
> in use as far as I can tell.  Ie, there are defines, but
> nothing is using them.  And IIRC the one bit that is defanitly
> different no longer even has a define.
>
> Regards,
> Rod Grimes
>
> > Dave T?ht CEO, TekLibre, LLC
> --
> Rod Grimes                                                 rgrimes@freebsd.org



-- 
Fixing Starlink's Latencies: https://www.youtube.com/watch?v=c9gLo6Xrwgw

Dave Täht CEO, TekLibre, LLC