From: Theodore A. R. <tr...@op...> - 2004-11-30 15:15:35
|
On Tue, 30 Nov 2004, Torsten Hahn wrote: > > Am 29.11.2004 um 23:37 schrieb E. Weddington: > > > Theodore A. Roth wrote: > > > >> On Mon, 29 Nov 2004, E. Weddington wrote: > >> > >> > >>>> I seriously doubt it. It sounds like the problem you are seeing > >>>> is API skew in libiberty from binutils to gcc. If we where to > >>>> include only libiberty source in avarice, we'd be up against the > >>>> same wall, but on all systems, not just cygwin. We have to either > >>>> include the source for both libbfd and libiberty or neither. > >>>> > >>>> > >>>> > >>> Really? Does libbfd require libiberty? > >>> > >> > >> Yes. It's the only reason that we are linking to it. Besides, if it > >> didn't, I don't think that you would have the API skew issues. > >> > >> > >> > > > Hi there, > > i just followed your discussion. The bfd/libiberty trouble is not only > lineked to the Cygwin environment. On MacOS X there is the same > problem. I never managed to get avarice to build at this platform > because of this trouble (ok, i have to say, i have not tried too much, > because i have also a debian box here). It would be very nice to have > this problem solved finally. :-) Thanks for the info Torsten. I had a private discussion with Brian Dean about this last week, we thought it was because Brian had done a mass install of avr tools into ${prefix}=/usr/local/avr. Looks like it is a broader problem than I thought. Unfortunately, I can't duplicate the problem on my systems yet. I'm going to try digging in some debian packages, but if that fails, would there be anyone with a Mac OSX system willing to give me temporary shell access (via ssh)? Preferably a system that can build avarice except for this libiberty failure mode. I'd like to try to get this resolved quickly so a release can go out well before friday. --- Ted Roth PGP Key ID: 0x18F846E9 Jabber ID: tr...@ja... |