This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: Status of std::string ABI change
- From: Jonathan Wakely <jwakely dot gcc at gmail dot com>
- To: "Peter A. Bigot" <pab at pabigot dot com>
- Cc: "libstdc++" <libstdc++ at gcc dot gnu dot org>
- Date: Mon, 25 Nov 2013 12:18:48 +0000
- Subject: Re: Status of std::string ABI change
- Authentication-results: sourceware.org; auth=none
- References: <52928551 dot 20607 at pabigot dot com>
On 24 November 2013 23:01, Peter A. Bigot <pab@pabigot.com> wrote:
> TL;DR: When will the underlying implementation of std::string in GCC
> provde a non-copy-on-write implementation by default, through a
> supported configuration-time option, or by selecting alternative
> header/library configurations during application build?
We now have a mechanism to allow two std::string implementations to
co-exist in a single executable, using the abi_tag attribute to use
different name mangling for the newer implementation (and any types
depending on it).
I think the plan is to start using that attribute only when we have
better tools for identifying and measuring ABI changes, to ensure we
get the ABI changes right first time and don't screw things up. I
expect to be working on those tools in the new year.