This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: C++0x library ABI
- From: Ulrich Drepper <drepper at redhat dot com>
- To: Jason Merrill <jason at redhat dot com>
- Cc: libstdc++ <libstdc++ at gcc dot gnu dot org>
- Date: Fri, 17 Jul 2009 08:36:21 -0700
- Subject: Re: C++0x library ABI
- References: <200907160806.BAA10914@hpcll183.cup.hp.com> <29bd08b70907160741s477a7683w5103cbca501e10@mail.gmail.com> <4A5F554A.40603@codesourcery.com> <29bd08b70907161300w680952d7sdd7e1d24520dac0b@mail.gmail.com> <F11683A7-58DE-4FE0-89B3-A47EA94859B3@edg.com> <206fcf960907161601g244e375fu4cb7b387e2331eeb@mail.gmail.com> <4A607FDF.3080306@redhat.com>
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Jason Merrill wrote:
> 1) Any plans to move to non-refcounted strings at some point?
> 2) LWG is moving this week to require container size() to be O(1), which
> is not true of our std::list.
The second item is certainly something that can be implemented using
namespace tricks or with symbol versioning. The first is optional.
It is a humongous cost to break the ABI. I really don't want it to
happen unless the world comes to an end if it's not done.
If C++ wants to be taken serious its developers must stop regarding ABI
breakage as a cavalier's delict. It isn't. To the contrary.
- --
â Ulrich Drepper â Red Hat, Inc. â 444 Castro St â Mountain View, CA â
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
iEYEARECAAYFAkpgmnUACgkQ2ijCOnn/RHQBGgCgpXnpORTBnyP0qFbrQ+VXxWW3
/b4AoIwHuxEQ+dIoeAN5o6UbEJn03B7o
=yc1R
-----END PGP SIGNATURE-----