This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: [PATCH, rs6000, libstdc++] Add baseline_symbols.txt for powerpc64le-linux-gnu
- From: David Edelsohn <dje dot gcc at gmail dot com>
- To: Jakub Jelinek <jakub at redhat dot com>
- Cc: Bill Schmidt <wschmidt at linux dot vnet dot ibm dot com>, Jonathan Wakely <jwakely at redhat dot com>, GCC Patches <gcc-patches at gcc dot gnu dot org>, "libstdc++ at gcc dot gnu dot org" <libstdc++ at gcc dot gnu dot org>
- Date: Tue, 7 Apr 2015 09:49:11 -0400
- Subject: Re: [PATCH, rs6000, libstdc++] Add baseline_symbols.txt for powerpc64le-linux-gnu
- Authentication-results: sourceware.org; auth=none
- References: <1428373888 dot 2816 dot 53 dot camel at gnopaine> <20150407130905 dot GE19273 at tucnak dot redhat dot com> <20150407131858 dot GD9755 at redhat dot com> <1428413977 dot 6000 dot 4 dot camel at oc8801110288 dot ibm dot com> <20150407134538 dot GF19273 at tucnak dot redhat dot com>
On Tue, Apr 7, 2015 at 9:45 AM, Jakub Jelinek <jakub@redhat.com> wrote:
> On Tue, Apr 07, 2015 at 08:39:37AM -0500, Bill Schmidt wrote:
>> Posted below the differences from powerpc64-linux-gnu. A surprising
>> number of additional symbols, but none are missing in the new baseline.
>> (Perhaps powerpc64-linux-gnu needs updating?)
>
> Sure, it needs updating.
>
> So, if I filter the GLIBCXX_3.4.21 and CXXABI_1.3.9 symbols from your diff,
> I get
> grep -v '@@GLIBCXX_3.4.21\|@@CXXABI_1.3.9' /tmp/XX | grep @
>> TLS:8:_ZSt11__once_call@@GLIBCXX_3.4.11
>> TLS:8:_ZSt15__once_callable@@GLIBCXX_3.4.11
>
> which are the two lines you should manually remove.
> But then, it means the powerpc64-linux-gnu and powerpc64le-linux-gnu
> baseline_symbols.txt are identical, so the question is if we really need
> to duplicate it, instead of just telling in the configury that for
> powerpc64le-linux-gnu it should use the powerpc64-linux-gnu
> baseline_symbols.txt.
I thought that they are using the same baseline_symbols.txt at the
moment and Bill's patch explicitly separates them in configure. If
the files are identical, then no need to separate them and no need for
the patch. If/when they diverge for a good reason (IEEE long
double?), we can revisit the patch.
Thanks, David