Sorry to jump into this late -- yes, opal/util/if.c is the exact place for this stuff.
Ralph is exactly correct that this code has been touched by multiple people over a few years, so it's possible that it's a little krufty. I certainly hope it isn't working by accident -- but given the context of his statement, he could have meant "by accident" only for FreeBSD. Shrug. Aleksej's comments indicate that he was going to look at this issue next week -- that would be fabulous. Since he has direct access to and assumedly much deeper knowledge of these platforms than we do, it might be best to see if he simply has an updated patch that we can apply. We can double check it against the platforms that we know / test / etc. On Dec 2, 2009, at 10:40 AM, Ralph Castain wrote: > Given that it is working for us at the moment, and my current priorities, I > doubt I'll get to this over the next 2-3 weeks. So if you have time and care > to look at it before then, please do! > > Thanks > > On Dec 1, 2009, at 8:45 PM, kevin.buck...@ecs.vuw.ac.nz wrote: > > >> Interesting - especially since the existing code works quite well over a > >> wide range of platforms. So I'm not quite so eager to declare it incorrect > >> and only working by accident. > >> > >> However, I would welcome a proposed patch so we can look at it. This is > >> always an important area for us, so the more discussion over how to do it > >> the better. > >> > >> Thanks > >> Ralph > > > > > > Just reading through all this again whilst looking at the source code > > for 1.3.3 as well, and I think what Aleskej was getting at was that > > there's already a block of: > > > > > > #if defined( __NetBSD__) || defined(__OpenBSD__) || defined(__FreeBSD__) || > > \ > > defined(__386BSD__) || defined(__bsdi__) || defined(__APPLE__) > > > > in there for the IPv6 (inet6) interfaces but NOT for the IPv4 (inet) > > ones. > > > > > > Indeed, the IPv6 stuff (the #if OPAL_WANT_IPV6 block) seems to be broken > > down into > > > > #ifdef __linux__ > > > > #ifdef __sun__ > > > > and the 6 OSes above. > > > > There would seem to be no such distinction for the IPv4 interface block. > > > > > > > > -- > > Kevin M. Buckley Room: CO327 > > School of Engineering and Phone: +64 4 463 5971 > > Computer Science > > Victoria University of Wellington > > New Zealand > > > > _______________________________________________ > > users mailing list > > us...@open-mpi.org > > http://www.open-mpi.org/mailman/listinfo.cgi/users > > > _______________________________________________ > users mailing list > us...@open-mpi.org > http://www.open-mpi.org/mailman/listinfo.cgi/users > -- Jeff Squyres jsquy...@cisco.com
