This is the mail archive of the
libc-alpha@sources.redhat.com
mailing list for the glibc project.
Re: [PATCH] Fixup PPC symbol versioning for libgcc symbols
- From: Roland McGrath <roland at frob dot com>
- To: Jack Howarth <howarth at bromo dot med dot uc dot edu>
- Cc: libc-alpha at sources dot redhat dot com, drepper at redhat dot com
- Date: Mon, 19 Aug 2002 22:40:58 -0400 (EDT)
- Subject: Re: [PATCH] Fixup PPC symbol versioning for libgcc symbols
> Roland,
> I don't quite see where the SHLIB_COMPAT test would ever disable
> this code. The libgcc-compat code is wrappered by...
>
> +#if SHLIB_COMPAT(libc, GLIBC_2_0, GLIBC_2_2_6)
>
> which to me seems to indicate that --enable-oldest-abi=2.3 will enable
> these symbols.
Then you do not understand the macro.
> Besides these symbols don't have anything to do with
> glibc 2.2.5 vs 2.2.90 or 2.3. Rather they are required because of
> the .hidden added to the libgcc routines starting with gcc 3.1.
You are confusing the origin of the bug with the ABIs it leaked into. The
lack of that .hidden is the original bug, and it meant that a libc compiled
with such a gcc inadvertently exported these symbols. Anything linked
against such a libc might require libc to define them. Any newly-compiled
libc will take care to avoid that bug happening again, so you can be sure
that if you don't need support for previous libc ABIs then you don't need
the ABI support for binaries linked with this bug.