Robert Dege <[EMAIL PROTECTED]> writes: > > > I then upgraded to 3.1 (which caused nothing but > > > nightmares there on in). Not only did MPlayer not compile, but other > > > programs that did compile correctly, caused unknown seg faults, and > > > other various disturbances. But I chalked that up to beta releases. > > > > Yes. It's not binary compatible with the old one either, so it will break. > > Binary compatible? care to elaborate? I don't understand what you're > trying to say here.
For C++, no version of gcc has been compatible with another. Same version of gcc, but compiled on different versions of glibc, are also incompatible. > > > So as a last resort, I downgraded to 2.95. I haven't had a problem > > > since. > > > > It's not binary compatible with gcc296 (or other versions) > > wrt. C++. It also has worse performance, more bugs and poorer C++ > > conformance (and of course, doesn't run at at all on IA64) > > I guess that's why ver 2.95 < 2.96. I'd have to question the whole > revision numbering system if 2.96 didn't have some form of improvement > or fix over 2.95. > > I think my whole beef is that 2.96 is a RedHat standard, not a Linux > standard. ie:, gcc-296 is only offered in rpm format (feel free to > correct me here if I'm wrong). The SRPM contains a tarball plus nicely separated patches. Shouldn't be any problem. -- Trond Eivind Glomsrød Red Hat, Inc. _______________________________________________ Redhat-list mailing list [EMAIL PROTECTED] https://listman.redhat.com/mailman/listinfo/redhat-list
