This is the mail archive of the libc-hacker@sources.redhat.com mailing list for the glibc project.
Note that libc-hacker is a closed list. You may look at the archives of this list, but subscription and posting are not open.
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
Roland McGrath <roland@frob.com> writes: > But this is categorically less robust than letting the old version talk to > the old version and leaving it at that. Even if you could be sure that > your compatibility code would be perfect, it doesn't seem worth the effort > and complexity to have the client worry about the server's version. No, there is no compatibility code, that's the point. We just use exactly the same types we had before the change. Nothing changes. For Alpha and 32bit platforms. The problem Thorsten and Jakub saw are for SPARC64 when on a machine with SPARC32 binaries and vice versa. This is handled by breaking compatibility with SPARC64 since there are no users. Leave alone MIPS64 which is probably not even building. What HJ saw is the same problem as for SPARC (here compatibility with x86 is wanted). Again, there are no users. HJ is most probably the only one who's building glibc himself. Everybody else is using a glibc 2.1 variant and will soon have to replace it with a completely new installation since there is no compatibility between the old and new installation. If you change the socket name etc you unnecessarily break the use for 32bit platforms and Alpha. -- ---------------. ,-. 1325 Chesapeake Terrace Ulrich Drepper \ ,-------------------' \ Sunnyvale, CA 94089 USA Red Hat `--' drepper at redhat.com `------------------------
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |