This is the mail archive of the libstdc++@gcc.gnu.org mailing list for the libstdc++ project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: [v3] Add hash<shared_ptr> and hash<unique_ptr>

From: Jonathan Wakely <jwakely dot gcc at gmail dot com>
To: Paolo Carlini <paolo dot carlini at oracle dot com>
Cc: "gcc-patches at gcc dot gnu dot org" <gcc-patches at gcc dot gnu dot org>, "libstdc++" <libstdc++ at gcc dot gnu dot org>
Date: Sat, 12 Jun 2010 01:58:28 +0100
Subject: Re: [v3] Add hash<shared_ptr> and hash<unique_ptr>
References: <4C125D99.4090109@oracle.com> <AANLkTikcIm66mJGzxxXyz-EB4EjDUERHqJ23w4_8fmcg@mail.gmail.com> <4C126146.5060405@oracle.com>

On 11 June 2010 17:16, Paolo Carlini wrote:
> On 06/11/2010 06:06 PM, Jonathan Wakely wrote:
>> Does our tuple benefit from the EBO if the deleter is empty?
>>
> You are right, I think it does now (it didn't at first) and that's a big
> advantage of tuple over pair. We should use a compressed_pair, really...
> We even had it contributed, at some point, quite similar to the boost
> version, I think. Would it be *much* simpler than <tuple>? Probably not
> worth the trouble and breaking the "experimental ABI"...

Hmm, I don't like that the 0x ABI is semi-frozen, so maybe I'm not the
right person to ask, because I say break it early and often while it's
experimental.

> Anyway, I can work immediately on changing <memory> to not include the
> whole <functional>, even if <tuple> remains.

I think that would be a nice improvement.

References:
- [v3] Add hash<shared_ptr> and hash<unique_ptr>
  - From: Paolo Carlini
- Re: [v3] Add hash<shared_ptr> and hash<unique_ptr>
  - From: Jonathan Wakely
- Re: [v3] Add hash<shared_ptr> and hash<unique_ptr>
  - From: Paolo Carlini

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]