This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: [v3] make check-abi
- From: Benjamin Kosnik <bkoz at redhat dot com>
- To: Loren James Rittle <rittle at latour dot rsch dot comm dot mot dot com>
- Cc: gcc-patches at gcc dot gnu dot org, libstdc++ at gcc dot gnu dot org
- Date: Fri, 23 Aug 2002 09:47:05 -0700
- Subject: Re: [v3] make check-abi
- Organization: Red Hat / San Francisco
- References: <200208222001.g7MK1Od03578@fillmore.constant.com><200208230336.g7N3aqvJ097906@latour.rsch.comm.mot.com>
> Most sane. As well, check-ins on mainline should have explainable
> changes only. For example, we have three symbol removals that are
> explainable in that no user code should have called them directly or
> indirectly.
The whole weak symbols of templates thing is really a hard issue to deal
with. I think all weak instances have to be taken seriously, if for no
other reason that -fno-implicit-templates has the possibility to change
behavior when linking against libstdc++.so.
The other issue is that 'extern template' usage in some of the standard
include files, which prevents weak implicit instatiations, means that
weak definitions in the library (weak objects in particular, for
vtables, etc) must be able to be found.
-benjamin