220 31558 <996e889d-5922-4e13-84dd-d914ec44f5a4@isocpp.org> article Path: news.gmane.org!.POSTED!not-for-mail From: dan@soundradix.com Newsgroups: gmane.comp.lang.c++.isocpp.proposals Subject: Re: First draft of [[unreachable]] attribute proposal Date: Sun, 12 Mar 2017 06:19:15 -0700 (PDT) Lines: 180 Approved: news@gmane.org Message-ID: <996e889d-5922-4e13-84dd-d914ec44f5a4@isocpp.org> References: <74928e4c-8f1e-4036-93d8-2560c27b651d@isocpp.org> Reply-To: std-proposals@isocpp.org NNTP-Posting-Host: blaine.gmane.org Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_Part_1670_1281867276.1489324755479" X-Trace: blaine.gmane.org 1489324764 18914 195.159.176.226 (12 Mar 2017 13:19:24 GMT) X-Complaints-To: usenet@blaine.gmane.org NNTP-Posting-Date: Sun, 12 Mar 2017 13:19:24 +0000 (UTC) To: ISO C++ Standard - Future Proposals Original-X-From: std-proposals+bncBCPJTS7U6MJBBVEVSXDAKGQE5YROXWY@isocpp.org Sun Mar 12 14:19:20 2017 Return-path: Envelope-to: gclcip-std-proposals@m.gmane.org Original-Received: from mail-yw0-f198.google.com ([209.85.161.198]) by blaine.gmane.org with esmtp (Exim 4.84_2) (envelope-from ) id 1cn3P2-0003ZI-Dc for gclcip-std-proposals@m.gmane.org; Sun, 12 Mar 2017 14:19:16 +0100 Original-Received: by mail-yw0-f198.google.com with SMTP id v76sf250667790ywg.4 for ; Sun, 12 Mar 2017 06:19:17 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=isocpp-org.20150623.gappssmtp.com; s=20150623; h=date:from:to:message-id:in-reply-to:references:subject:mime-version :x-original-sender:reply-to:precedence:mailing-list:list-id :x-spam-checked-in-group:list-post:list-help:list-archive :list-subscribe:list-unsubscribe; bh=yMVjS/xpgvNRURPB7AFgwAGjL4gu++ChNVTsFyfnhPk=; b=dG3TBbTjyU+LVmhPEl4WeX3mpwZ8xCM6Bh3vzrJLP0vqTq1JuKPBsnpYMVoLVQkVlO ENQvn+NuLboh/j/8pAPHhO0KpUn003bOoUE4+k0GzzCEVj+Vb2d6gfH4e7umpovaVfWB f2Sy/VHFbNUsXEmxmAFKm8Oku/+QaxDfxkfOZnFgrYVS2XjyKWXzSiGRSVJ30mxtvGRW BoR77YcNmSHo2Tip0lw7lyiRNj0+8USs4CDDmCwxJPTGxlhdt1p+SYCTltpVkTTakriN AaidOHiC8ZVQ07/R3n8a8ox/ZzCZwH1EkIUno+sNqFRNODAqmvFcc+/Lie0eJBcTUnET 5kgg== DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=soundradix.com; s=google; h=date:from:to:message-id:in-reply-to:references:subject:mime-version :x-original-sender:reply-to:precedence:mailing-list:list-id :x-spam-checked-in-group:list-post:list-help:list-archive :list-subscribe:list-unsubscribe; bh=yMVjS/xpgvNRURPB7AFgwAGjL4gu++ChNVTsFyfnhPk=; b=NxV5Hsj43ZyIVPIDI6hz+S3nB2OuFmVN2jpSp5H1U7cu59wZ3VO7+DYfqHitsFBs3+ Ipcx4ETMC5mMvrSmS7fRgs7EGO8ktbXeoaPzPnD+Vwdwk1ABuIGwuoNcpwKz9b+55OKJ 69nKn3/GUQ3T4e/sTrfmmJ6XegMqf4nm8DnorMPwWTNtqRgM6p0dIVA02Uus2u6CvySa sVBW5ofKLUS7IhigJSDAPDYf76pZR4K7fTAm0FFIKSGdEevO+y40P1fE1GiNRzScs95G 1zJomJlxz+ngYIFsAE8sbLhMOdOQWQ6SrnwLTByetoipN2rzMYd94nKmTYW8VOjnVxXq EoEg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:date:from:to:message-id:in-reply-to:references :subject:mime-version:x-original-sender:reply-to:precedence :mailing-list:list-id:x-spam-checked-in-group:list-post:list-help :list-archive:list-subscribe:list-unsubscribe; bh=yMVjS/xpgvNRURPB7AFgwAGjL4gu++ChNVTsFyfnhPk=; b=AQcLkibbHpIUNtAcMfL/IaMEv8mf6/a+yjallUFLvgICV3U5WPfW804JcA6n/TgfJ+ XXnMafG3kVWEwZ0kj/mPjUJsm3GxJaik8/v1oCbiAOOkxznFx4l3CoRpX8eG2QCuKclP Sicl3fVCLI9SF3cmzkZvtb0QySuPAfe59dvCGFTi3UHWWKbGLB+vaNTtPKR1kGs+j8ZX kfSILYUcxE9R157gHy9vBbVh5MiggARLaxe2f57JarJOgdx15YBkDkgrjiORJDPrZyoG BKsTTQYaShyrCzYT/vtSuF/LvVbvV1dXR0sScrEUyQNSY/3jM8Htrr+Y3lZcIsyJlSAC VEkQ== X-Gm-Message-State: AMke39n7Ka4Bznv549HaRFpjIauj1miVruj4SaGha7/r3QY1kFbYcUvxGRxxDK8L/yWkhA== X-Received: by 10.129.107.134 with SMTP id g128mr14464687ywc.139.1489324757000; Sun, 12 Mar 2017 06:19:17 -0700 (PDT) X-BeenThere: std-proposals@isocpp.org Original-Received: by 10.157.53.6 with SMTP id o6ls13649825otc.47.gmail; Sun, 12 Mar 2017 06:19:16 -0700 (PDT) X-Received: by 10.157.13.82 with SMTP id 76mr2947770oti.6.1489324756162; Sun, 12 Mar 2017 06:19:16 -0700 (PDT) In-Reply-To: X-Original-Sender: dan@soundradix.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:31558 Archived-At: ------=_Part_1670_1281867276.1489324755479 Content-Type: multipart/alternative; boundary="----=_Part_1671_340757506.1489324755479" ------=_Part_1671_340757506.1489324755479 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable I think the opposite is correct WRT tooling; if you have a separate=20 [[unreachable]] attribute, not only does it convey the intention better to= =20 the reader, but it also does to any tool which sees the attribute, so for= =20 example a static analyzer can show exactly the path in which=20 [[unreachable]] code becomes reachable, but not necessarily show it for=20 failing [[assert]] attributes. Cheers, Dan On Saturday, March 11, 2017 at 4:24:35 PM UTC+2, Andrzej Krzemie=C5=84ski w= rote: > > > > W dniu pi=C4=85tek, 10 marca 2017 13:41:05 UTC+1 u=C5=BCytkownik Morwenn = napisa=C5=82: >> >> Le vendredi 10 mars 2017 13:04:37 UTC+1, Andrzej Krzemie=C5=84ski a =C3= =A9crit : >>> >>> >>> >>> W dniu czwartek, 9 marca 2017 04:49:35 UTC+1 u=C5=BCytkownik Myriachan= =20 >>> napisa=C5=82: >>>> >>>> I wrote up a proposal document for a new attribute, [[unreachable]],= =20 >>>> which lets you mark locations in code as being unreachable for optimiz= ation=20 >>>> purposes, or to suppress certain warnings. >>>> >>>> This has been discussed on the mailing list before; there's a link to= =20 >>>> that thread in the .pdf. >>>> >>>> https://1drv.ms/b/s!Agwy_QljRwlGgcoJeZMiimuq3H36Ww=20 >>>> >>>> >>>> I have no idea what to do with it now. If you have any comments on th= e=20 >>>> content, or how to actually proceed, let me know! =3D^-^=3D >>>> >>>> >>>> Under the contract proposal, anywhere where the statement is expected,= =20 >>>> you can type: >>>> >>>> [[assert: false]]; >>>> >>>> While [[assert axiom: false]] is probably semantically equivalent to= =20 >>>> the proposed [[unreachable]], the former doesn't really convey the=20 >>>> meaning. >>>> >>>> I would be more confortable adding [[unreachable]] to the default=20 >>>> branch of a switch than the equivalent [[assert axiom: false]]. >>>> >>> > Given that the expression in the assertion is a compile-time constant onl= y=20 > [[assert: false]] is fine. But the interaction goes beyond only syntax. I= f=20 > we have attribute [[unreachable]] maybe we want it to behave like [[asser= t:=20 > false]] in every detail; that is: allow calling contract violation handle= rs=20 > in test builds, if this place in the code is reached. This way the hint c= an=20 > be used not only for optimization, but also for instrumentation. > > Regards, > &rzej; > --=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/996e889d-5922-4e13-84dd-d914ec44f5a4%40isocpp.or= g. ------=_Part_1671_340757506.1489324755479 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
I think the opposite is correct WRT tooling; if you have a= separate [[unreachable]] attribute, not only does it convey the intention = better to the reader, but it also does to any tool which sees the attribute= , so for example a static analyzer can show exactly the path in which [[unr= eachable]] code becomes reachable, but not necessarily show it for failing = [[assert]] attributes.

Cheers,
Dan


On Saturday, = March 11, 2017 at 4:24:35 PM UTC+2, Andrzej Krzemie=C5=84ski wrote:


W dniu pi=C4= =85tek, 10 marca 2017 13:41:05 UTC+1 u=C5=BCytkownik Morwenn napisa=C5=82:<= blockquote class=3D"gmail_quote" style=3D"margin:0;margin-left:0.8ex;border= -left:1px #ccc solid;padding-left:1ex">
Le vendredi 10 mars= 2017 13:04:37 UTC+1, Andrzej Krzemie=C5=84ski a =C3=A9crit=C2=A0:


W dniu czwartek, 9 = marca 2017 04:49:35 UTC+1 u=C5=BCytkownik Myriachan napisa=C5=82:
I wrote up a proposal document for a new= attribute, [[unreachable]], which lets you mark locations in code as being= unreachable for optimization purposes, or to suppress certain warnings.
This has been discussed on the mailing list before; there's a link= to that thread in the .pdf.

https://1= drv.ms/b/s!Agwy_QljRwlGgcoJeZMiimuq3H36Ww

I have no idea wh= at to do with it now.=C2=A0 If you have any comments on the content, or how= to actually proceed, let me know! =3D^-^=3D


Under the contract = proposal, anywhere where the statement is expected, you can type:

[[assert: fal= se]];

While = [[assert axiom: false]] is probably semantically equivalent to the proposed [[unreachable]], the former doesn&#= 39;t really convey the meaning.

I would be more confortable adding <= span style=3D"font-family:courier new,monospace">[[unreachable]] to = the default branch= of a switch than = the equivalent [[assert a= xiom: false]].

Given that the expressio= n in the assertion is a compile-time constant only [[assert: false]] is fin= e. But the interaction goes beyond only syntax. If we have attribute [[unre= achable]] maybe we want it to behave like [[assert: false]] in every detail= ; that is: allow calling contract violation handlers in test builds, if thi= s place in the code is reached. This way the hint can be used not only for = optimization, but also for instrumentation.

Regards,
&rzej;

--
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 https://groups.google.= com/a/isocpp.org/d/msgid/std-proposals/996e889d-5922-4e13-84dd-d914ec44f5a4= %40isocpp.org.
------=_Part_1671_340757506.1489324755479-- ------=_Part_1670_1281867276.1489324755479-- .