This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] Update baseline_symbols.txt for {x86_64,i?86,aarch64,powerpc64,s390{,x}}-linux (PR libstdc++/79348)
- From: Jonathan Wakely <jwakely at redhat dot com>
- To: Jakub Jelinek <jakub at redhat dot com>
- Cc: gcc-patches at gcc dot gnu dot org, libstdc++ at gcc dot gnu dot org
- Date: Mon, 13 Feb 2017 12:51:34 +0000
- Subject: Re: [PATCH] Update baseline_symbols.txt for {x86_64,i?86,aarch64,powerpc64,s390{,x}}-linux (PR libstdc++/79348)
- Authentication-results: sourceware.org; auth=none
- References: <20170213120925.GG1849@tucnak>
On 13/02/17 13:09 +0100, Jakub Jelinek wrote:
Hi!
This patch updates baseline_symbols.txt mostly from our latest rpm builds,
x86_64 and i686 have been also compared to my local bootstrap abi lists
and s390 (31-bit) comes from my bootstrap (and 64-bit s390x compared against
the rpm builds too).
s390{,x}-linux weren't updated the last time (for GCC 6), so it now even
fails the abi check without this patch, the rest has been.
s390x 64-bit also is identical to powerpc64 64-bit (which is right, both
have the same mangling of important types and also the long double situation
is the same (well, powerpc64 also has the even newer long double, but I
think we don't instantiate in libstdc++ anything for that yet).
Ok for trunk?
OK, thanks.
The only weird thing I've noticed in the changes is that
_ZNSbIwSt11char_traitsIwESaIwEEC1ERKS2_mRKS1_
_ZNSbIwSt11char_traitsIwESaIwEEC2ERKS2_mRKS1_
is added on all arches but i?86, any explanation for that?
Huh, no. That symbol comes from the instantiations in
src/c++11/cow-wstring-inst.cc but I don't know why it would be
different for i?86.