220 16341 article Path: news.gmane.org!not-for-mail From: David Krauss Newsgroups: gmane.comp.lang.c++.isocpp.proposals Subject: Re: Re: Pointers to members of unknown class Date: Wed, 18 Feb 2015 01:41:31 +0800 Lines: 130 Approved: news@gmane.org Message-ID: References: <9b03a820-0ac5-4931-9825-b2fc9ef893f9@isocpp.org> <885c1fbd-8701-4f70-b7ad-89ee9104f42d@isocpp.org> Reply-To: std-proposals@isocpp.org NNTP-Posting-Host: plane.gmane.org Mime-Version: 1.0 (Mac OS X Mail 8.2 \(2070.6\)) Content-Type: multipart/alternative; boundary="Apple-Mail=_746C61F7-C9C0-4B75-BA2A-8B2E3D4AC91D" X-Trace: ger.gmane.org 1424194913 14321 80.91.229.3 (17 Feb 2015 17:41:53 GMT) X-Complaints-To: usenet@ger.gmane.org NNTP-Posting-Date: Tue, 17 Feb 2015 17:41:53 +0000 (UTC) To: std-proposals@isocpp.org Original-X-From: std-proposals+bncBCW25A7E3QCRBVP2RWTQKGQEHRCC34I@isocpp.org Tue Feb 17 18:41:45 2015 Return-path: Envelope-to: gclcip-std-proposals@m.gmane.org Original-Received: from mail-vc0-f197.google.com ([209.85.220.197]) by plane.gmane.org with esmtp (Exim 4.69) (envelope-from ) id 1YNm9X-0006Jb-2D for gclcip-std-proposals@m.gmane.org; Tue, 17 Feb 2015 18:41:43 +0100 Original-Received: by mail-vc0-f197.google.com with SMTP id hy4sf124028771vcb.0 for ; Tue, 17 Feb 2015 09:41:42 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:from:content-type:message-id:mime-version :subject:date:references:to:in-reply-to:x-original-sender :x-original-authentication-results:reply-to:precedence:mailing-list :list-id:list-post:list-help:list-archive:list-subscribe :list-unsubscribe; bh=9MgnwwQH/i0FQje8klBRqw8ZqDzNMO4Kb5xCbGLzjbw=; b=CDxrvdRq/ja9jUhHGnu5xqt8ZcACrZZhgboxY8x4sH1LJt2qa6WuSqF9wi//wqREdZ kEQNKZ2pd3Jgs/Oir5+c/rVpUbHtKumbm3D7GUbvJxHpwr7JQxiMau6/NXI97HxuE6Fv lgB7jiSfEvvXaZxsuyjpHjnAlSEzqCjRfIj4q8ODDIr1+o5zaqxGuqUctG9AJTuCJxbk UB1D8rl6UfGA2ABMQZ5XaCug0eC2E3P02INx4GMZtI5rpEjDq6+YMIZa9545rfvikeQ4 HpZAGqPYOAWuHNfwsrmT3tXrrmIIJo6ZJCK2CIhIa5zoHzk7Qm1B/SoJB8UD5ybvVb7P myXA== X-Gm-Message-State: ALoCoQlBfJrZ+cFsCkg0wyLV410O0N++Q21peRGFV6R88eAyHRBAKy9uHJymKpo349xYE6bLNhBv X-Received: by 10.236.202.207 with SMTP id d55mr27293967yho.4.1424194902339; Tue, 17 Feb 2015 09:41:42 -0800 (PST) X-BeenThere: std-proposals@isocpp.org Original-Received: by 10.50.8.98 with SMTP id q2ls585262iga.33.gmail; Tue, 17 Feb 2015 09:41:41 -0800 (PST) X-Received: by 10.68.139.98 with SMTP id qx2mr51288213pbb.39.1424194901552; Tue, 17 Feb 2015 09:41:41 -0800 (PST) Original-Received: from mail-pa0-f47.google.com (mail-pa0-f47.google.com. [209.85.220.47]) by mx.google.com with ESMTPS id ix2si3819532pac.158.2015.02.17.09.41.41 for (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Tue, 17 Feb 2015 09:41:41 -0800 (PST) Received-SPF: pass (google.com: domain of potswa@gmail.com designates 209.85.220.47 as permitted sender) client-ip=209.85.220.47; Original-Received: by padbj1 with SMTP id bj1so7864707pad.5 for ; Tue, 17 Feb 2015 09:41:41 -0800 (PST) X-Received: by 10.70.103.162 with SMTP id fx2mr51910021pdb.24.1424194901412; Tue, 17 Feb 2015 09:41:41 -0800 (PST) Original-Received: from [172.20.10.2] ([121.54.44.90]) by mx.google.com with ESMTPSA id qa1sm10470019pdb.84.2015.02.17.09.41.38 for (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Tue, 17 Feb 2015 09:41:40 -0800 (PST) In-Reply-To: X-Mailer: Apple Mail (2.2070.6) X-Original-Sender: potswa@gmail.com X-Original-Authentication-Results: mx.google.com; spf=pass (google.com: domain of potswa@gmail.com designates 209.85.220.47 as permitted sender) smtp.mail=potswa@gmail.com; dkim=pass header.i=@gmail.com; dmarc=pass (p=NONE dis=NONE) header.from=gmail.com Precedence: list Mailing-list: list std-proposals@isocpp.org; contact std-proposals+owners@isocpp.org List-ID: X-Google-Group-Id: 399137483710 List-Post: , List-Help: , List-Archive: List-Subscribe: , List-Unsubscribe: , Xref: news.gmane.org gmane.comp.lang.c++.isocpp.proposals:16341 Archived-At: --Apple-Mail=_746C61F7-C9C0-4B75-BA2A-8B2E3D4AC91D Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=UTF-8 > On 2015=E2=80=9302=E2=80=9317, at 9:48 PM, David Rodr=C3=ADguez Ibeas wrote: >=20 > Sorry for looking like the bad guy here, but I am still unsure of the fea= ture. Little added value and increased dangers as people don't quite unders= tand pointers to members. >=20 > To be clear, the Itanium ABI does not have the pointer-to-member sizes de= scribed in the document to adhere to the requirement of 'reinterpret_cast',= but rather as the result of the design of dynamic dispatch in that ABI (wh= at value is 'this' on entry to the final overrider?). Virtual functions and virtual bases set requirements for certain PTM types,= causing increased size. The reinterpret_cast requirement causes that size = to apply to all other PTM types, even for classes without virtual anything. Looking at the proposal again, I find the incomplete classes troubling. Isn= =E2=80=99t it inconsistent to retroactively change the size of a class=E2= =80=99 PTMs at the time it gets completed? Shouldn=E2=80=99t pointers to me= mbers of incomplete classes be incomplete too? For that matter, MSVC doesn=E2=80=99t go far enough with this optimization.= Most classes are less than 256 bytes, and then ordinary PTMs only need one= byte. > I am fine changing the standard to make VS implementation conformant. For= that a single change needs to be done: remove the requirement that roundtr= ip conversions of pointer-to-member are guaranteed (becomes implementation = defined, still works where it works today, doesn't where it fails today). A= nother change would then be useful: provide a mechanism to replace the inte= rmediate point of the roundtrip. For that you can either extend the gramma= r to allow for a "pointer to member of unknown type" or provide a different= facility. The simplest would be to have a way of creating a buffer that i= s "large enough".=20 >=20 > With such a buffer, expressions in the form 'T void::*dst =3D src;' becom= e 'memcpy(&dst, &src, sizeof src)' (where 'src' is a pointer to member of t= ype 'T' in some class 'C'). The conversions in the opposite direction are = the inverse: 'memcpy(&ptr, buffer, sizeof ptr)=E2=80=99. Sure, any POD type can be constructed from its object representation, and P= TMs are POD. No special buffer provision is required. Just let the user cal= l memcpy directly. Taken to its bare essentials, the proposal is simple indeed! --=20 ---=20 You received this message because you are subscribed to the Google Groups "= ISO C++ Standard - Future Proposals" group. To unsubscribe from this group and stop receiving emails from it, send an e= mail to std-proposals+unsubscribe@isocpp.org. To post to this group, send email to std-proposals@isocpp.org. Visit this group at http://groups.google.com/a/isocpp.org/group/std-proposa= ls/. --Apple-Mail=_746C61F7-C9C0-4B75-BA2A-8B2E3D4AC91D Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=UTF-8
On 2015=E2=80=9302= =E2=80=9317, at 9:48 PM, David Rodr=C3=ADguez Ibeas <dibeas@ieee.org> wrote:

Sor= ry for looking like the bad guy here, but I am still unsure of the feature.= Little added value and increased dangers as people don't quite understand = pointers to members.

To be clear, the Itanium = ABI does not have the pointer-to-member sizes described in the document to = adhere to the requirement of 'reinterpret_cast', but rather as the result o= f the design of dynamic dispatch in that ABI (what value is 'this' on entry= to the final overrider?).

Virtual functions and virtual bases set requirements = for certain PTM types, causing increased size. The reinterpret_cast requirement causes that size to apply to = all other PTM types, even for classes without virtual anything.
<= br class=3D"">
Looking at the proposal again, I find the incomple= te classes troubling. Isn=E2=80=99t it inconsistent to retroactively change= the size of a class=E2=80=99 PTMs at the time it gets completed? Shouldn= =E2=80=99t pointers to members of incomplete classes be incomplete too?

For that matter, MSVC doesn=E2=80=99t go f= ar enough with this optimization. Most classes are less than 256 bytes, and= then ordinary PTMs only need one byte.

I am fine changing the standard to make VS implementation conformant. For= that a single change needs to be done: remove the requirement that roundtr= ip conversions of pointer-to-member are guaranteed (becomes implementation = defined, still works where it works today, doesn't where it fails today). A= nother change would then be useful: provide a mechanism to replace the inte= rmediate point of the roundtrip.  For that you can either extend the g= rammar to allow for a "pointer to member of unknown type" or provide a diff= erent facility.  The simplest would be to have a way of creating a buf= fer that is "large enough". 

With such a = buffer, expressions in the form 'T void::*dst =3D src;' become 'memcpy(&= ;dst, &src, sizeof src)' (where 'src' is a pointer to member of type 'T= ' in some class 'C').  The conversions in the opposite direction are t= he inverse: 'memcpy(&ptr, buffer, sizeof ptr)=E2=80=99.
<= /div>

Sure, any POD type c= an be constructed from its object representation, and PTMs are POD. No spec= ial buffer provision is required. Just let the user call memcpy directly.

Taken to its bare essentials, the proposal is simple indeed!

--

---
You received this message because you are subscribed to the Google Groups &= quot;ISO C++ Standard - Future Proposals" group.
To unsubscribe from this group and stop receiving emails from it, send an e= mail to std-proposa= ls+unsubscribe@isocpp.org.
To post to this group, send email to std-proposals@isocpp.org.
Visit this group at http://groups.google.com/a/isocpp.org/group/std-proposals/<= /a>.
--Apple-Mail=_746C61F7-C9C0-4B75-BA2A-8B2E3D4AC91D-- .