[zeromq-dev] ZeroMQ for Contiki OS

Pieter Hintjens ph at imatix.com
Wed Jan 20 21:47:23 CET 2016


Try to create the repo now, it should work.

Indeed, zmq.c/h is probably a bad filename to use and zmtp.c/h is
better. However I'd probably advise putting the sources into src and
prefixing them all with something like cz_

On Wed, Jan 20, 2016 at 5:08 PM, Axel Voitier <axel.voitier at gmail.com> wrote:
> Got it.
>
> It seems I don't have sufficient rights to create a repo in the
> organisation.
> Please, could you either add me the right, or create a repo named
> "contiki-zmtp"?
>
> By the way, there are zmq.c/h files in the current implementation in which I
> have put things built on top of the ZMTP implementation (typically socket
> related stuff, module initialisation).
> Should I merge what it contains to zmtp.c/h (for instance), or find another
> name for it?
>
>
> Cheers,
> Axel
>
> --
> Contact: axel.voitier at gmail.com
> Secure: axel.voitier at protonmail.ch
>
> 2016-01-20 16:47 GMT+01:00 Pieter Hintjens <ph at imatix.com>:
>>
>> I've updated the trademark policy since it was slightly out of date.
>>
>> http://zeromq.org/docs:trademark-policy
>>
>> The goal of this policy is to ensure that users know what they're
>> getting. In your case I'd use -zmtp in the name, or call it something
>> else.
>>
>>
>> On Wed, Jan 20, 2016 at 4:20 PM, Axel Voitier <axel.voitier at gmail.com>
>> wrote:
>> > Hi Pieter,
>> >
>> > Neat! :)
>> >
>> > Yes, it would be nice if it spawns interest for other embedded targets.
>> > I am all for the C4.1 policy, and for the move to the zeromq
>> > organization.
>> > Thanks for proposing!
>> >
>> > Could you enlighten me about what can be named zeromq actually? Is it
>> > the
>> > full set of ZMTP+all sockets behaviour, or is it just limited to libzmq
>> > and
>> > its bindings?
>> > I guess the name of the project can then be changed to contiki-zmtp (or
>> > contiki-libzmtp?) when moving to the organisation.
>> >
>> >
>> > Cheers,
>> > Axel
>> >
>> > --
>> > Contact: axel.voitier at gmail.com
>> > Secure: axel.voitier at protonmail.ch
>> >
>> > 2016-01-20 12:47 GMT+01:00 Pieter Hintjens <ph at imatix.com>:
>> >>
>> >> Oh, and I sent you your first pull request. :)
>> >>
>> >> On Wed, Jan 20, 2016 at 12:39 PM, Pieter Hintjens <ph at imatix.com>
>> >> wrote:
>> >> > Hi Axel,
>> >> >
>> >> > This is really cool... could be the basis for other embedded
>> >> > projects.
>> >> > I'd like to move this into the zeromq organization and use our C4.1
>> >> > contribution policy, is that OK?
>> >> >
>> >> > The name suggests you ported libzmq, which is misleading. It'd be
>> >> > better to say "ZMTP" instead of "ZeroMQ".
>> >> >
>> >> > I've invited you to the ZeroMQ core team so you can move this project
>> >> > (if you approve of this idea).
>> >> >
>> >> > -Pieter
>> >> >
>> >> >
>> >> > On Wed, Jan 20, 2016 at 11:47 AM, Axel Voitier
>> >> > <axel.voitier at gmail.com>
>> >> > wrote:
>> >> >> Yes, I actually started from libzmtp. Few things remain from it now
>> >> >> (channels and msgs) as I had to adapt it to whatever Contiki
>> >> >> provides.
>> >> >> Contiki does not have a posix socket interface. But the hardest was
>> >> >> t
>> >> >> adapt
>> >> >> to its threading model (protothreads) which were really
>> >> >> constraining...
>> >> >>
>> >> >> Also, I adapted the memory allocations to do use as much as possible
>> >> >> preallocation (called memb in Contiki). A few malloc remains here
>> >> >> and
>> >> >> there,
>> >> >> but much less than initially.
>> >> >>
>> >> >>
>> >> >> However, if I was to restart this project, I would not pick Contiki
>> >> >> anymore.
>> >> >> But rather give it a try with RIOT, which seems to have something
>> >> >> that
>> >> >> match
>> >> >> a POSIX interface. And a much saner threading model that behaves a
>> >> >> bit
>> >> >> like
>> >> >> an event loop with a hub (and reusing features from the underlying
>> >> >> MCU
>> >> >> to do
>> >> >> context switching, like the SVC instruction with ARM).
>> >> >>
>> >> >>
>> >> >> Cheers,
>> >> >> Axel
>> >> >>
>> >> >> --
>> >> >> Contact: axel.voitier at gmail.com
>> >> >> Secure: axel.voitier at protonmail.ch
>> >> >>
>> >> >> 2016-01-20 9:31 GMT+01:00 Benjamin Henrion <zoobab at gmail.com>:
>> >> >>>
>> >> >>> On Wed, Jan 20, 2016 at 4:01 AM, Axel Voitier
>> >> >>> <axel.voitier at gmail.com>
>> >> >>> wrote:
>> >> >>> > Hello,
>> >> >>> >
>> >> >>> > I have published an attempt at porting zeromq to Contiki OS:
>> >> >>> > https://github.com/Alidron/contiki-zeromq
>> >> >>> >
>> >> >>> > It has a few limitations still. However, I covered most basics of
>> >> >>> > ZMTP
>> >> >>> > and a
>> >> >>> > handful of sockets. It is not entirely compliant with the RFCs
>> >> >>> > for
>> >> >>> > the
>> >> >>> > moment though (some parts prove to be harder to do than I
>> >> >>> > expected).
>> >> >>> >
>> >> >>> > I have been able to run it on a constrained setup: ARM Cortext M3
>> >> >>> > running at
>> >> >>> > 32MHz, with 512KB of flash (the demo programs weight about 130KB)
>> >> >>> > and
>> >> >>> > 80KB
>> >> >>> > of RAM. And with a sub-1GHz RF link.
>> >> >>> >
>> >> >>> > The implementation makes use of Contiki network stack, which can
>> >> >>> > use
>> >> >>> > TCP,
>> >> >>> > IPv4, IPv6 and 6LoWPAN. I actually tested it over 6LoWPAN on the
>> >> >>> > sub-GHz
>> >> >>> > link :).
>> >> >>> >
>> >> >>> > It is not has featured and fast as running zeromq between two
>> >> >>> > usual
>> >> >>> > computers. But I hope this implementation prove it is feasible to
>> >> >>> > have
>> >> >>> > ZeroMQ on such constrained hardware :).
>> >> >>>
>> >> >>> Have you tried to compile libzmtp?
>> >> >>>
>> >> >>> One of my short-term goal is to make libzmptp running on esp8266
>> >> >>> and
>> >> >>> some cortex m3 board.
>> >> >>>
>> >> >>> --
>> >> >>> Benjamin Henrion <bhenrion at ffii.org>
>> >> >>> FFII Brussels - +32-484-566109 - +32-2-3500762
>> >> >>> "In July 2005, after several failed attempts to legalise software
>> >> >>> patents in Europe, the patent establishment changed its strategy.
>> >> >>> Instead of explicitly seeking to sanction the patentability of
>> >> >>> software, they are now seeking to create a central European patent
>> >> >>> court, which would establish and enforce patentability rules in
>> >> >>> their
>> >> >>> favor, without any possibility of correction by competing courts or
>> >> >>> democratically elected legislators."
>> >> >>> _______________________________________________
>> >> >>> zeromq-dev mailing list
>> >> >>> zeromq-dev at lists.zeromq.org
>> >> >>> http://lists.zeromq.org/mailman/listinfo/zeromq-dev
>> >> >>
>> >> >>
>> >> >>
>> >> >> _______________________________________________
>> >> >> zeromq-dev mailing list
>> >> >> zeromq-dev at lists.zeromq.org
>> >> >> http://lists.zeromq.org/mailman/listinfo/zeromq-dev
>> >> >>
>> >> _______________________________________________
>> >> zeromq-dev mailing list
>> >> zeromq-dev at lists.zeromq.org
>> >> http://lists.zeromq.org/mailman/listinfo/zeromq-dev
>> >
>> >
>> >
>> > _______________________________________________
>> > zeromq-dev mailing list
>> > zeromq-dev at lists.zeromq.org
>> > http://lists.zeromq.org/mailman/listinfo/zeromq-dev
>> >
>> _______________________________________________
>> zeromq-dev mailing list
>> zeromq-dev at lists.zeromq.org
>> http://lists.zeromq.org/mailman/listinfo/zeromq-dev
>
>
>
> _______________________________________________
> zeromq-dev mailing list
> zeromq-dev at lists.zeromq.org
> http://lists.zeromq.org/mailman/listinfo/zeromq-dev
>



More information about the zeromq-dev mailing list