Hi Michael!Great.
OK, the RR-SCTP thesis is attached. Please give it a try. I will send my when I finish it, in a couple of months (I hope!).
Hmmmm.
I've made some modifications to the code, so that some parameters in some functions are no more "void *" pointers, instead they have a proper parameter type. This is something I disliked a lot! and why I fixed it.
I can add a configure option to do that.
Beside that, all the code concerning RR-SCTP is enclosed in "#ifdef" and you can enable it through a #define on the config.h file.
Fine. I'll integrate it.
So, I will update my patch to sctplib-1.0.4 version and I will send it to you (as soon as I can).
Regards,
Julio
On 11/22/05, Michael Tuexen <Michael.Tuexen AT micmac.franken.de> wrote: Hi Julio,
see my comments in-line.
Best regards
Michael
On Nov 22, 2005, at 6:57 PM, Julio Kriger wrote:
> Hi Michael!
> Well, thanks for your help. I think the chapter about sctplib will
> be a short one... Thanks anyway.
>
> I would to tell you about RR-SCTP.
>
> RR-SCTP is a proposed algorithm that make SCTP more robust to
> abrupt changes on RTT and packages reordering. This highly improves
> the performance of SCTP, specilly on wireless and satellital
> networks. I already have the algorithm implemented and working, and
> the results are very promising.
>
> I can send you the RR-SCTP thesis if you like (but it's in spanish).
Just send it. Maybe I can get some idea, maybe not. I can give it a try.
>
> I already have a patch for sctplib-1.0.3, that I could update to
> sctplib-1.0.4, regarding RR-SCTP.
> I would like to know if you would consider add my patch to the main
> sctplib develpment tree?
The situation is as follows:
The sctplib-1.0.4 is in bugfix mode... So I'm very hesitant to put it
in... We had a developer branch for new features, but it is currently
not maintained. So this is not a good idea.
I do not know if it is applicable, but if you can put in the changes
#ifdefed out by default and provide a configure option to enable your
stuff,
it would be OK. I would only check that it is really ifdefed out and
compiles
also when it is enabled. This would not introduce any instability for
people
not interested in RR-SCTP but gives people the possibility to play with
it, if they want. What do you think?
>
> Regards,
> Julio
>
>
> On 11/22/05, Michael Tuexen
<Michael.Tuexen AT micmac.franken.de>
> wrote: Hi Julio,
>
> see my comments in-line.
>
> Best regards
> Michael
>
>
> On Nov 22, 2005, at 3:50 PM, Julio Kriger wrote:
>
> > Hi!
> >
> > I'm writing my master thesys (an implementation of RR-SCTP over
> > sctplib) and need some documentation. I've read the documentation
> > that comes with the software. That is very good, but I need more.
> > But just about sctplib, the internal working, if it use IP or UDP,
> > etc.. I've enough information about the SCTP protocol.
> Well, you got the source. That is all we can provide. The sctplib uses
> a raw IP socket per default. UDP Tunneling is a compile option.
> > Any line you could write to me will be welcome!
> Three...
> > Thanks in advance.
> > Regards,
> > Julio
> >
> >
> > --
> > --
> > Stewie: [After Lois tries to feed Stewie his broccoli "airplane
> > style"] Damn you, Damn the Broccoli, and Damn the Wright Brothers.
> > ----------------------------
> > Julio Kriger
> >
mailto:juliokriger AT gmail.com
>
>
>
>
> --
> --
> Stewie: [After Lois tries to feed Stewie his broccoli "airplane
> style"] Damn you, Damn the Broccoli, and Damn the Wright Brothers.
> ----------------------------
> Julio Kriger
>
mailto:juliokriger AT gmail.com
--
--
Stewie: [After Lois tries to feed Stewie his broccoli "airplane style"] Damn you, Damn the Broccoli, and Damn the Wright Brothers.
----------------------------
Julio Kriger
mailto:juliokriger AT gmail.com
<RR-SCTP.pdf>
Archive powered by MHonArc 2.6.19+.