This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [libstdc++ PATCH] Implement observer_ptr
- From: Jonathan Wakely <jwakely at redhat dot com>
- To: Ed Smith-Rowland <3dw4rd at verizon dot net>
- Cc: Ville Voutilainen <ville dot voutilainen at gmail dot com>, "gcc-patches at gcc dot gnu dot org" <gcc-patches at gcc dot gnu dot org>, libstdc++ at gcc dot gnu dot org
- Date: Sat, 2 May 2015 14:01:50 +0100
- Subject: Re: [libstdc++ PATCH] Implement observer_ptr
- Authentication-results: sourceware.org; auth=none
- References: <CAFk2RUaT=BzmzQ=e42OjQz4KT9A1dW+xu+5_gkFMX-DGARJQoQ at mail dot gmail dot com> <20150501210123 dot GN3618 at redhat dot com> <5544303A dot 3010607 at verizon dot net> <20150502094002 dot GQ3618 at redhat dot com> <20150502094224 dot GR3618 at redhat dot com> <5544C7F5 dot 2080305 at verizon dot net>
On 02/05/15 08:49 -0400, Ed Smith-Rowland wrote:
OK. Thanks.
I do remember an SD-6 discussion about how annoying the
define-the-macro-in-all-relevant-headers was.
I didn't know there was a resolution. I need to reeducate myself.
Meanwhile I'll rollback my patch.
Reverted in 222722.
I'll ask next time.
Sorry for the noise.
OK, no problem. I wasn't going to ask you to revert it, as it didn't
really do any harm, I just don't think it is required to be in every
header.
I've been wondering if it would be better to just put all the
feature-test macros in a central place, like <bits/c++config.h> (and
maybe somewhre separate for the "experimental" ones). Stephen Kelly's
complaints about Boost having to include loads of large std::lib
headers to test the macros is a valid complaint about the current SD-6
scheme.
I *think* that would also still be conforming, because the macro would
be defined when you include the right header ... it would just be
defined when you include any other headers too :-)