Paul wrote: > Hi Jan > > On Wednesday 17 January 2007 18:19, Jan Kiszka wrote: >>>> But the best is a Xenomai which supports 64-bit AMD processors :) > > I've made a tentative stab at a couple of the kernel sources, but must admit > the assembly code scares the heck out of me....
I think the assembly part should be a smaller issue, certainly a solvable one. You are invited to post your questions regarding this here (or better on xenomai-core) so that we all can try to look at it. /me has no x86-64 assembly experience yet, but I think to remember that it's very similar to well-know i386. The bigger issue with such ports are the testing/debugging cycles which can cost quite some time if you are unlucky (or blind, or both ;)). > >> [Really, it depends on someone feeling enough need to work on this (or >> sponsor the work). A few pieces are already there from previous RTAI >> efforts towards x86-64 support. But now it takes a clean implementation >> over recent Linux/I-pipe/Xenomai.] > > How much value is the 2.6.10 x86_64 patch from RTAI when it comes to working > on an I-pipe port for 2.6.19 ? I would consider it as an inspiration for certain detail problems, also regarding the assembly stuff. But the overall design is of no use because it's old-style Adeos and likely has RTAI hacks included. > > (I may well have an AMD64 socket 939 processor available for loan next week > if > that would help). Thanks, but for me this wouldn't change the situation. Jan
signature.asc
Description: OpenPGP digital signature
_______________________________________________ Xenomai-help mailing list [email protected] https://mail.gna.org/listinfo/xenomai-help
