site_archiver@lists.apple.com Delivered-To: darwin-kernel@lists.apple.com File a request for a general API and post the radar # here. -- Terry Hi Garth, Thank you very much for your help in advance. Best regards Michael On Oct 29, 2007, at 5:46 PM, Garth Cummings wrote: Hi Andreas, On Oct 27, 2007, at 7:43 AM, Alexander von Below wrote: --gc ____________________________________________________________________ Garth Cummings gcummings@apple.com Sr. Software Engineer Apple Developer Technical Support <http://developer.apple.com/technicalsupport> _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/michael.tuexen%40lurchi... This email sent to michael.tuexen@lurchi.franken.de _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/tlambert%40apple.com _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.a... On Oct 29, 2007, at 10:47 AM, Michael Tuexen <Michael.Tuexen@lurchi.franken.de

wrote: I had hope that mailing to DTS would be a way to get some header files of

the Leopard kernel sources. We can not even compile the SCTP NKE without them, no to think about testing. I know that we should be using a supported API and we have to accept some consequences when not doing so. But there is currently no supported API which would allow us to build a transport layer on top of IPv[46] and below the socket layer. And I do understand Apple that it might not want to spend the time/effort for providing this functionality, since I would see currently only SCTP and DCCP implementations using this interface. And the current unsupported interface of Panther and Tiger was good enough. Your suggestion on filing an enhancement request is interesting. I filed one a long time ago to add #define IPPROTO_SCTP 132 in /usr/include/netinet/in.h This change is trivial, part of the FreeBSD code base and I see no way how this could result in a problem. However, this change was never accepted for Tiger or Leopard. So I doubt that the above API extension would happen. Maybe I'm too pessimistic. So how can we progress getting the SCTP stack running on Leopard? Can't we do more than just waiting until someday the complete sources are released? I'm willing to sign additional NDAs or think about what Apple think is necessary. While we are at it: SCTP supports a very nice and attractive mobility concept: Transport Layer Mobility. So it would be *very* interesting to have an SCTP stack on the iPhone which supports this mobility concept. Is there any way to get the possibility on porting the NKE to the iPhone? And there is a last question I have: Can we use DTrace for function boundary tracing of an NKE? I tried to build a dummy SCTP stack (just a loadable NKE which registers a dummy sctp_input() function). But I can not see it in sudo dtrace -l My development is 99% blocked because of this. Sources are a MUST in this case. My suggestion would be to file a DTS incident in this case. If the idea is to try another channel to get the sources, please don't submit a DTS incident. There is nothing DTS can do to help get you the sources. If you suspect there's a supported KPI to do what you want, DTS can definitely help with that. If you need the sources because your product relies on private APIs, I trust you've filed an enhancement request asking for a supported way to do what you need. This email sent to tlambert@apple.com This email sent to site_archiver@lists.apple.com