This is the mail archive of the
mailing list for the GCC project.
Re: Make some comdats implicitly hidden
- From: Jan Hubicka <hubicka at ucw dot cz>
- To: Paolo Carlini <paolo dot carlini at oracle dot com>
- Cc: Jan Hubicka <hubicka at ucw dot cz>, gcc-patches at gcc dot gnu dot org, Jakub Jelinek <jakub at redhat dot com>
- Date: Thu, 29 Aug 2013 14:54:05 +0200
- Subject: Re: Make some comdats implicitly hidden
- Authentication-results: sourceware.org; auth=none
- References: <20130826195724 dot GB13069 at kam dot mff dot cuni dot cz> <521BFB17 dot 5000902 at redhat dot com> <20130828172318 dot GB3513 at kam dot mff dot cuni dot cz> <521E6B5F dot 1070305 at redhat dot com> <20130829081148 dot GA11983 at kam dot mff dot cuni dot cz> <521F10C2 dot 4040001 at oracle dot com> <20130829121907 dot GA20627 at kam dot mff dot cuni dot cz> <521F42F2 dot 4060804 at oracle dot com>
> On 08/29/2013 02:19 PM, Jan Hubicka wrote:
> >So my belief is that it is safe to drop those symbols from
> >libstdc++. Every program/DSO using them have to define its own
> >copy of those symbols, so I believe removing them from libstdc++
> >won't cause issues.
> Really, you should check with Jakub before proceeding. I the change
> it's Ok with him, it's Ok with me too (the other library maintainers
> should be in CC however). At minimum the baselines would need
> If I can ask you a personal courtesy, please don't commit anything
> causing regressions, like check_abi failing, either a complete
> change, or nothing.
No worries. I do not intend to commit the patch until the check_abi issue is
discussed and solved.
(check_abi tends to be easy to miss when one checks test results by hand, but I
am trying to force myself to always use compare_tests, so hopefully I won't do
that in future)
Note that currently you will get the same abi_check failure if you try to LTO
libstdc++.so. It would be really nice if we started to do that eventually.