220 31802 article Path: news.gmane.org!.POSTED!not-for-mail From: Alberto Barbati Newsgroups: gmane.comp.lang.c++.isocpp.proposals Subject: Re: This variable should not be named: an identifier (not) to remember Date: Tue, 28 Mar 2017 11:09:54 +0200 Lines: 166 Approved: news@gmane.org Message-ID: References: <985b9b2a-c734-45eb-95f4-db4dc0d309a1@isocpp.org> <6096248.pKJ4QPQkKS@tjmaciei-mobl1> <4E0805FE-B97D-4A93-956D-9281B3207F39@gmail.com> <3421677.yAxhhm3ryH@tjmaciei-mobl1> Reply-To: std-proposals@isocpp.org NNTP-Posting-Host: blaine.gmane.org Mime-Version: 1.0 Content-Type: multipart/alternative; boundary=001a114e5364ba3ed3054bc6d233 X-Trace: blaine.gmane.org 1490692207 26837 195.159.176.226 (28 Mar 2017 09:10:07 GMT) X-Complaints-To: usenet@blaine.gmane.org NNTP-Posting-Date: Tue, 28 Mar 2017 09:10:07 +0000 (UTC) To: std-proposals@isocpp.org Original-X-From: std-proposals+bncBCPY5DV6RIMBBY6Q5DDAKGQEPWMM7TY@isocpp.org Tue Mar 28 11:10:03 2017 Return-path: Envelope-to: gclcip-std-proposals@m.gmane.org Original-Received: from mail-pg0-f71.google.com ([74.125.83.71]) by blaine.gmane.org with esmtp (Exim 4.84_2) (envelope-from ) id 1csn8Q-0005MN-IK for gclcip-std-proposals@m.gmane.org; Tue, 28 Mar 2017 11:09:50 +0200 Original-Received: by mail-pg0-f71.google.com with SMTP id n5sf96762870pgd.19 for ; Tue, 28 Mar 2017 02:09:56 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=isocpp-org.20150623.gappssmtp.com; s=20150623; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :x-original-sender:x-original-authentication-results:reply-to :precedence:mailing-list:list-id:x-spam-checked-in-group:list-post :list-help:list-archive:list-subscribe:list-unsubscribe; bh=Vm23YswxiExz9ZtlycZ0L2BNe3QLlUZ7z5imGZkwnG4=; b=NxQnfmUj58dhfYd71kBVbHHZfjyEID8Il1eRE9JH/GFvltoZvJHUsKemFEMMvmr6RZ YlxvoJZaYlKexu5J/J3M6J/lK0YqJ2/Mkxsj8/FJquT7ruUlYmZL6CK9MgdBwNVrqVAv Xj66IEJBkATKtiY0xxjoxFcokUgi0TLW0jJEQb0Tmd19BxQQCjyaZeCY5Dilk8F9XHIM juE0UMeIaCuOe3a31zePXs0gd/2yqkJbesdAbslk+S5PLFdSGbtaeOe3aF+yZ1cfclwj OsZM5BtUnsm/8E6bhWoWQ/eBKVL736uZgpXBWiIsX/gVK71igPp6rU9H4K1t5uOlzEIa ItrQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:x-original-sender :x-original-authentication-results:reply-to:precedence:mailing-list :list-id:x-spam-checked-in-group:list-post:list-help:list-archive :list-subscribe:list-unsubscribe; bh=Vm23YswxiExz9ZtlycZ0L2BNe3QLlUZ7z5imGZkwnG4=; b=KEcbyB+E9pzuataIqX9rj7lyZZZRwhN9WfB2PfSWlQNhlOh5S6Bqk3XJOXDdcQp95P riNQPuu46YYg2ozI0IKCTgpXcbozJtfrMG46gON4meCvs4+m9V+pV0B7hLkYLgWPSpzJ kVJJMT48XecxPjykXH7y1YzzNTdnPiTG9I2fYrnUIa9SLxNud1MuhPhORB5csHFlgPcL IldVXpz2Du8JcKR7ZCenQJZE/LxJ6k1cHx7cBTmA6bdOTK3bVG8osenB3sJY3W1kFFd1 6DrT1DrSS38CStRmGkRPGrBtoSrObxzTbjdBZKactW2/SCjB5IL08qz99eX+4r9cmjZ4 te4g== X-Gm-Message-State: AFeK/H3P6PHpEqDQA2h1EINu9JRKeJotUSa5uB5I97dAss/fKexkUhw+x6E3UobfftCZ3g== X-Received: by 10.99.188.2 with SMTP id q2mr9720343pge.36.1490692196350; Tue, 28 Mar 2017 02:09:56 -0700 (PDT) X-BeenThere: std-proposals@isocpp.org Original-Received: by 10.157.56.80 with SMTP id r16ls2120263otd.15.gmail; Tue, 28 Mar 2017 02:09:55 -0700 (PDT) X-Received: by 10.176.2.9 with SMTP id 9mr381466uas.155.1490692195189; Tue, 28 Mar 2017 02:09:55 -0700 (PDT) Original-Received: from mail-vk0-x244.google.com (mail-vk0-x244.google.com. [2607:f8b0:400c:c05::244]) by mx.google.com with ESMTPS id f139si1436771vkd.180.2017.03.28.02.09.55 for (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Tue, 28 Mar 2017 02:09:55 -0700 (PDT) Received-SPF: pass (google.com: domain of albertobarbati@gmail.com designates 2607:f8b0:400c:c05::244 as permitted sender) client-ip=2607:f8b0:400c:c05::244; Original-Received: by mail-vk0-x244.google.com with SMTP id d188so11931717vka.3 for ; Tue, 28 Mar 2017 02:09:55 -0700 (PDT) X-Received: by 10.31.88.1 with SMTP id m1mr1670904vkb.140.1490692194580; Tue, 28 Mar 2017 02:09:54 -0700 (PDT) Original-Received: by 10.176.74.93 with HTTP; Tue, 28 Mar 2017 02:09:54 -0700 (PDT) In-Reply-To: <3421677.yAxhhm3ryH@tjmaciei-mobl1> X-Original-Sender: AlbertoBarbati@gmail.com X-Original-Authentication-Results: mx.google.com; dkim=pass header.i=@gmail.com; spf=pass (google.com: domain of albertobarbati@gmail.com designates 2607:f8b0:400c:c05::244 as permitted sender) smtp.mailfrom=albertobarbati@gmail.com; dmarc=pass (p=NONE sp=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:31802 Archived-At: --001a114e5364ba3ed3054bc6d233 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable 2017-03-28 9:04 GMT+02:00 Thiago Macieira : > Em segunda-feira, 27 de mar=C3=A7o de 2017, =C3=A0s 23:47:55 PDT, Alberto= Barbati > escreveu: > > > And yet it's in use. > > > https://github.com/v8/v8/blob/master/src/x64/codegen-x64.cc#L33 > > > > Thanks Thiago for the pointer. According to my (possibly incorrect) > > interpretation this code raises undefined behavior because __ is reserv= ed > > "for any use", which I would expect includes the use as a macro name. I > may > > be wrong, but I'd expect little sympathy in the committee about > potentially > > breaking it. By the way, this code might possibly not break, because th= ey > > were clever to put #undefs in it. Incidentally, the code would indeed > break > > if we defined __ to be a macro... > > Right. We can say it's V8's own fault for using a reserved identifier. It= 's > like when Qt (qlalr) had an identifier called _Nullable, but that got > repurposed by a recent Clang change. > In this particular instance, V8's programmer could and probably should have used a non-reserved name (for example "m_" would have been a better choice regardless). Even the choice of _Nullable was ill-conceived, since identifiers beginning with an underscore and an uppercase letter are also reserved. > I'm just saying it's known to be in use in a fairly major project. You > know, > the one that is behind Node.js and a whole ecosystem. The argument in the > paper that says it's little used is incorrect: it's not like pre-C++11 > "auto", > which had exactly one use in the entire Microsoft source code. > I said that it's little used in the two librariesI have checked (VS2015 and libc++). For user code such as this, all I said is that "the proposal should not break code", which is correct: the identifier __ should not have been used in the first place, so this is a case of code that someone might say "deserves to be broken". > I guess we could convince GCC and Clang to print a warning about it if it > looks like this is going forward, so the V8 developers start fixing it. > Maybe > we'll get some of these fixed too: > > http://lxr.free-electrons.com/source/include/linux/compiler.h#L115 > http://lxr.free-electrons.com/source/include/asm-generic/fixmap.h#L73 The first one, that #defines "if", is particularly disturbing. Cheers, Alberto --=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. To view this discussion on the web visit https://groups.google.com/a/isocpp= ..org/d/msgid/std-proposals/CAF3xnJRtyCRSqc5dh7fpf1pSQem_Lw4SbePQxY3dBC_xEEi= sEg%40mail.gmail.com. --001a114e5364ba3ed3054bc6d233 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
2017= -03-28 9:04 GMT+02:00 Thiago Macieira <thiago@macieira.org>:
Em segunda-feira, 27 de mar=C3=A7o de = 2017, =C3=A0s 23:47:55 PDT, Alberto Barbati
escreveu:
> > And yet it's in use.
> > https://github.com/v8/v8/bl= ob/master/src/x64/codegen-x64.cc#L33
>
> Thanks Thiago for the pointer. According to my (possibly incorrect) > interpretation this code raises undefined behavior because __ is reser= ved
> "for any use", which I would expect includes the use as a ma= cro name. I may
> be wrong, but I'd expect little sympathy in the committee about po= tentially
> breaking it. By the way, this code might possibly not break, because t= hey
> were clever to put #undefs in it. Incidentally, the code would indeed = break
> if we defined __ to be a macro...

Right. We can say it's V8's own fault for using a reserved i= dentifier. It's
like when Qt (qlalr) had an identifier called _Nullable, but that got
repurposed by a recent Clang change.

In= this particular instance, V8's programmer could and probably should ha= ve used a non-reserved name (for example "m_" would have been a b= etter choice regardless). Even the choice of _Nullable was ill-conceived, s= ince identifiers beginning with an underscore and an uppercase letter are a= lso reserved.
=C2=A0
I'm just saying it's known to be in use in a fairly major project. = You know,
the one that is behind Node.js and a whole ecosystem. The argument in the paper that says it's little used is incorrect: it's not like pre-C+= +11 "auto",
which had exactly one use in the entire Microsoft source code.

I said that it's little used in the two librari= esI have checked (VS2015 and libc++). For user code such as this, all I sai= d is that "the proposal should not break code", which is correct:= the identifier __ should not have been used in the first place, so this is= a case of code that someone might say "deserves to be broken".=C2=A0
I guess we could convince GCC and Clang to print a warning about it if it looks like this is going forward, so the V8 developers start fixing it. May= be
we'll get some of these fixed too:

http://lxr.free-electrons.com/source/include/linux/compiler.h#L115
http://lxr.free-electrons.com/<= wbr>source/include/asm-generic/fixmap.h#L73

<= /div>
The first one, that #defines "if", is particularly dist= urbing.

Cheers,

Alberto
=

--
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.
To view this discussion on the web visit htt= ps://groups.google.com/a/isocpp.org/d/msgid/std-proposals/CAF3xnJRtyCRSqc5d= h7fpf1pSQem_Lw4SbePQxY3dBC_xEEisEg%40mail.gmail.com.
--001a114e5364ba3ed3054bc6d233-- .