Path: csiph.com!eternal-september.org!feeder.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: Keith Thompson <kst-u@mib.org>
Newsgroups: comp.lang.c
Subject: Re: Is this code considered to transgress the signed overflow purism?
Date: Fri, 28 Apr 2017 09:22:24 -0700
Organization: None to speak of
Lines: 17
Message-ID: <lnk264lcxr.fsf@kst-u.example.com>
References: <11ea6682-cdee-4513-97c2-6c90b585716a@googlegroups.com> <87o9vi6rql.fsf@bsb.me.uk> <26aee5e9-ed2f-4c6e-af68-b85418c7e641@googlegroups.com> <ln1ssdmrc5.fsf@kst-u.example.com> <odtsrt$t90$1@dont-email.me> <lnwpa5l8kh.fsf@kst-u.example.com> <oduono$pn3$1@dont-email.me> <1f9a4820-c10f-4130-90c2-464b4bd48c5d@googlegroups.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Injection-Info: mx02.eternal-september.org; posting-host="2e5ec5e1f0fe87e81c65a0e293e55e1b"; logging-data="28983"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX18NZYIhMaNvTJZqlmwGOxAR"
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.1 (gnu/linux)
Cancel-Lock: sha1:ZLOkL3tkEuMkdiZocR9/BqLFWqs= sha1:CEbOiIYBe+QHFeUmrs9qKpkRryE=
Xref: csiph.com comp.lang.c:108670

supercat@casperkitty.com writes:
[...]
> Incidentally, if a platform with 32-bit "int" defines 1<<31 as -1 (likely
> on platforms that also handle C++, as C++ would require such behavior),
> then if "x" is "long long", then a compiler would likely have to generate
> code for x*(1<<y) that was equivalent to (y!=31) ? x<<y : -(x<<y), which
> would almost certainly be less efficient than x<<y.

No, 1<<31 has undefined behavior in C++ if 2147483648 cannot be
represented as an int.  (C++11 standard, section 5.8 [expr.shift],
unchanged in N4618.)

-- 
Keith Thompson (The_Other_Keith) kst-u@mib.org  <http://www.ghoti.net/~kst>
Working, but not speaking, for JetHead Development, Inc.
"We must do something.  This is something.  Therefore, we must do this."
    -- Antony Jay and Jonathan Lynn, "Yes Minister"