svn commit: r277054 - in head: share/man/man4 sys/modules/cc sys/modules/cc/cc_dctcp sys/netinet sys/netinet/cc

[Pedro Giffuni]

Midori-san;

I am aware of the patent issue as we spoke about it with George (gnn@) 
at BSDCan.
It's not really an issue for us, and if you find more patent under the 
same free-use
conditions, feel free to send patches :).

The linux guys are aware of the patent but I agree that the issues linux
might have are none of our business.

Thank you all involved in the DCTCP implementation: it's great to see 
FreeBSD's
networking implementation getting better!

Best regards,

Pedro.


On 01/13/15 06:49, Midori Kato wrote:
> Hi Pedro,
>
> Thank you for your mention about the patent problem.
> Thanks to Hiren and Lars support, we have already discuss patent with 
> microsoft people. But I am not sure that we must care about Linux 
> because we worked on our implementation without linux. Let's wait and 
> see if redhat people say about it.
>
> Again, thank you for your interest to dctcp implementation :)
>
> Regards,
> -- Midori
>
>
> 2015-01-13 1:06 GMT+09:00 Pedro Giffuni <pfg at freebsd.org 
> <mailto:pfg at freebsd.org>>:
>
>
>     On 12/01/2015 03:33 a.m., Hiren Panchasara wrote:
>
>         Author: hiren
>         Date: Mon Jan 12 08:33:04 2015
>         New Revision: 277054
>         URL: https://svnweb.freebsd.org/changeset/base/277054
>
>         Log:
>            DCTCP (Data Center TCP) implementation.
>               DCTCP congestion control algorithm aims to maximise
>         throughput and minimise
>            latency in data center networks by utilising the proportion
>         of Explicit
>            Congestion Notification (ECN) marked packets received from
>         capable hardware as a
>            congestion signal.
>               Highlights:
>            Implemented as a mod_cc(4) module.
>            ECN (Explicit congestion notification) processing is done
>         differently from
>            RFC3168.
>            Takes one-sided DCTCP into consideration where only one of
>         the sides is using
>            DCTCP and other is using standard ECN.
>               IETF draft:
>         http://tools.ietf.org/html/draft-bensley-tcpm-dctcp-00
>            Thesis report by Midori Kato:
>         https://eggert.org/students/kato-thesis.pdf
>               Submitted by:     Midori Kato <katoon at sfc.wide.ad.jp
>         <mailto:katoon at sfc.wide.ad.jp>> and
>                         Lars Eggert <lars at netapp.com
>         <mailto:lars at netapp.com>>
>                         with help and modifications from
>                         hiren
>            Differential Revision: https://reviews.freebsd.org/D604
>            Reviewed by: gnn
>
>
>     Missing
>
>     RelNotes=    yes
>
>     Huge thank you!
>
>     FWIW, I thought the linux guys wouldn't take it over some GPL vs
>     patent issue but apparently redhat pushed some muscle and just did it:
>
>     http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=e3118e8359bb7c59555aca60c725106e6d78c5ce
>
>     Not that I care much what they do ;).
>
>     Pedro.
>
>