220 35457 <4c614758-259b-4eeb-b4f7-89c42ed0e4d4@isocpp.org> article Path: news.gmane.org!.POSTED!not-for-mail From: Ben Craig Newsgroups: gmane.comp.lang.c++.isocpp.proposals Subject: Re: On P0829 - a freestanding implementation Date: Wed, 22 Nov 2017 12:09:40 -0800 (PST) Lines: 58 Approved: news@gmane.org Message-ID: <4c614758-259b-4eeb-b4f7-89c42ed0e4d4@isocpp.org> References: <32147114.z48B4pQc9d@tjmaciei-mobl1> Reply-To: std-proposals@isocpp.org NNTP-Posting-Host: blaine.gmane.org Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_Part_11373_142227205.1511381380439" X-Trace: blaine.gmane.org 1511381382 5832 195.159.176.226 (22 Nov 2017 20:09:42 GMT) X-Complaints-To: usenet@blaine.gmane.org NNTP-Posting-Date: Wed, 22 Nov 2017 20:09:42 +0000 (UTC) To: ISO C++ Standard - Future Proposals Original-X-From: std-proposals+bncBDCM7T7A5UMRBBNT27IAKGQEBMA3Y2Q@isocpp.org Wed Nov 22 21:09:36 2017 Return-path: Envelope-to: gclcip-std-proposals@m.gmane.org Original-Received: from mail-ua0-f199.google.com ([209.85.217.199]) by blaine.gmane.org with esmtp (Exim 4.84_2) (envelope-from ) id 1eHbKx-00014k-Qe for gclcip-std-proposals@m.gmane.org; Wed, 22 Nov 2017 21:09:35 +0100 Original-Received: by mail-ua0-f199.google.com with SMTP id p33sf8486529uag.4 for ; Wed, 22 Nov 2017 12:09:43 -0800 (PST) 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 :list-post:list-help:list-archive:list-subscribe:list-unsubscribe; bh=iJnWvPWng9zvo58SCh/V8JB+6vvO/Z/jymeDrDDESTM=; b=pt7Wa6rt6jfxwQzmi/lFJoJOPk0BUdgLSl92yzNaFFLQKf2cv7jRohIW6w3FYLsWky DBl3EDghB+1/a1c8STLfk+13aoIiy8CRzDas7DQGvJxjklL+WwSZWU9JCWlpeW9KqLhn QXLd2ZIY6Iw97o0eHExO+m/IUs9LXCvK7LE9vuDfQ6agKzTn7tu5Nge/ZeIzQ1xw+XeC vPHYbOkiUucXvtcfZy0M9Vd3v+AIA7GSfKXNEmhGJDys+Elq/EA9mg+8COwh/NZcnag7 jHruj/exavpCdxdPC5Ip2TxSm9SAo2bixJ+mki/KI5tYd5H+76yCFC4H/0jmnnHpdVVj iYIw== DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=date:from:to:message-id:in-reply-to:references:subject:mime-version :x-original-sender:reply-to:precedence:mailing-list:list-id :list-post:list-help:list-archive:list-subscribe:list-unsubscribe; bh=iJnWvPWng9zvo58SCh/V8JB+6vvO/Z/jymeDrDDESTM=; b=WUz5fsqaOCEKiqKULaDj6mjc01l3iW/hc9a1nuqqptHb/NpRd/B7BAWJyIi/oP4PR/ aePbAM//Rf8i6xeGeyYWNmjK9jPuv5eX3m5qjrpI14+UbbcF45NbfAMYVfO5I9IA3dgY caM3tTxZnXIPuh8I/xIRTHYEFuQqcoubjNr+viPiHcV6M6lTjYnmgHWxLmxS/r7lRNBC VbsnYQHGMyYequN+y2+2R5+74o96IO//O3m0SLz83Ghw2VO/CwfNerHcVK4539pnhj0n qDlkB8YZsym67v2khNlJuI5O4RaC130EOSVeyZSh248eK2StQBo+uenh0qBJ5LxaeTBU uszg== 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=iJnWvPWng9zvo58SCh/V8JB+6vvO/Z/jymeDrDDESTM=; b=mDHedMh+l5aAhzm6AoRf2KZZS4gNLa17meD9CDUb6GoKkvqUXgCZnYbvnLmldBNX5E AecIblZgGamHkEWIwH345siu7IYPnh4HqiiyPoabwBOxX21IndEqzKwqHuUA5pj0pASD zJfg1rn2FScLiMr3M0QMGkHU9Ds4N+sXXxIlRgMap5KMNtBeI5Hp2GDyxAh5yEv0T1SS JM6e0xOOeuDCSg+kXgSiaaaRCtU4UxXwhOkGaBaxgN2dmKXhY0QgpmUpnQ3A4smD3a4N Yfi5wX6LCIAtvLKjZHzYYa3QhntbKCpQabwAVyhdZugBajQFS478FZL1/1r+W0rTJpHt DB3A== X-Gm-Message-State: AJaThX6ryrGo/OOewDWi+12iW1qn7+G+juoU3WzCC8ZAsFBwBHGtm+V0 QYn9ZpPov/o9Q83KtDJURpzpmA== X-Google-Smtp-Source: AGs4zMaQV4kp391EHkBn2m+UFhrmyI0qtdd6kIK/AcU7rQrKCOZ4yJM66/PAyh1oLLRoXYPGnw9bug== X-Received: by 10.31.69.149 with SMTP id s143mr10320831vka.11.1511381383006; Wed, 22 Nov 2017 12:09:43 -0800 (PST) X-BeenThere: std-proposals@isocpp.org Original-Received: by 10.31.151.69 with SMTP id z66ls1190205vkd.16.gmail; Wed, 22 Nov 2017 12:09:41 -0800 (PST) X-Received: by 10.31.161.87 with SMTP id k84mr14670vke.10.1511381381040; Wed, 22 Nov 2017 12:09:41 -0800 (PST) In-Reply-To: X-Original-Sender: ben.craig@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:35457 Archived-At: ------=_Part_11373_142227205.1511381380439 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Re: Thiago I disagree with the assertion that the whole point of freestanding is that = there is no library. Even in the c11 spec, 5.1.2.1, there is wording about= the lack of an operating system, but in the same clause it talks about lib= rary facilities. Why does it matter if memcpy comes from a header or an ob= ject file? Some non-functions even come from an object file (e.g. 64-bit * = 64-bit multiplication on i386). Thiago mentioned Zephyr in an earlier post. It appears to allow the hosted = newlib c library to be used, with the caveat that filesystem operations wil= l error at runtime. This kind of support for existing c libraries is very c= ommon. Re: Mathias I do not know of a standards conforming way to make throw trap without incu= rring the space overhead of exception tables. I can set the terminate handl= er or the unexpected handler, but both of those happen too late. I'm not tr= ying to argue the speed benefits of exceptions vs. no exceptions, I'm argui= ng the size difference, particularly when my error handling strategy is "re= boot". I don't make that very clear in r0 of my paper though. For floating point, it isn't about the size or speed of floating point. My = concern is that in kernels, you can't use floating point without making ext= ra, non portable calls. If you do use floating point without the non-portab= le save/restore calls, you corrupt user mode. There are different heap arguments depending on the environment. In kernel = mode, the issue isn't the lack of a heap, but what the default heap should = be. Do you choose the plentiful, but unsafe paged pool? Or do you choose th= e scarce but safe nonpaged pool? For embedded, the concern is fragmentation. These devices are supposed to r= un for years at a time without restarting. To accomplish that, you need to = avoid all transient allocation. Allocations at startup are fine. If you g= et rid of all transient allocations, you basically get rid of all stl alloc= ations. Also note that the standard library does not permit allocators to signal fa= ilure by returning nullptr. That means that using stl allocators requires e= xceptions. --=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/4c614758-259b-4eeb-b4f7-89c42ed0e4d4%40isocpp.or= g. ------=_Part_11373_142227205.1511381380439-- .