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: Default values in #define
Date: Wed, 31 Aug 2016 12:30:16 -0700
Organization: None to speak of
Lines: 21
Message-ID: <lna8fsbut3.fsf@kst-u.example.com>
References: <d6e97054-1a68-418a-a6ae-5807ffccabbe@googlegroups.com> <lnoa4kfmkv.fsf@kst-u.example.com> <1d377167-a33a-402e-89ee-d3efa06d95ab@googlegroups.com> <87twebuwwn.fsf@gmail.com> <36a5bcc3-9e30-468c-a3e1-4b9c9788d395@googlegroups.com> <npjm5q$vho$1@dont-email.me> <1Kgvz.279$6d.174@fx26.iad> <npk7v8$p1c$1@dont-email.me> <8277ffd2-121d-4f3a-9bb0-ad3068696415@googlegroups.com> <nplahs$ngv$1@dont-email.me> <ee1434bc-5a52-4cea-91e3-432d67f2727c@googlegroups.com> <502669cd-f0fc-4b99-8c8a-0d806c8b5f20@googlegroups.com> <aa1be119-a537-4df5-97da-ade93afdba0c@googlegroups.com> <566c947b-91d6-43c1-86aa-772f6cf21e35@googlegroups.com> <nq1dfn$ola$1@dont-email.me> <ln8tvfczff.fsf@kst-u.example.com> <nq277p$4sb$1@dont-email.me> <lninujb133.fsf@kst-u.example.com> <kfnshtmp4nu.fsf@x-alumni2.alumni.caltech.edu> <lneg54c0dt.fsf@kst-u.example.com> <e8bd75c3-14cb-469d-a864-e62d8085cc5f@googlegroups.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Injection-Info: mx02.eternal-september.org; posting-host="945944de09706c9b4e29b53c9d2efdc2"; logging-data="4935"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX1/sCsn5GQwt97VNeDTEqs9v"
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.1 (gnu/linux)
Cancel-Lock: sha1:kXolIlS3WWSn6Kzm82dAiXTvvU0= sha1:Hxbq2RY6dxo7ZdoA1B/mafnD0HQ=
Xref: csiph.com comp.lang.c:93007

supercat@casperkitty.com writes:
> On Wednesday, August 31, 2016 at 11:22:01 AM UTC-5, Keith Thompson wrote:
>> I was (somewhat vaguely) thinking that an implementation is still
>> allowed to optimize away some apparent accesses to volatile objects.
>> Which, now that I think about it, doesn't make a whole lot of sense.
>
> If a compiler knows that code will be running on hardware where volatile
> reads of a certain location will always yield the value 12, it would be
> entitled to replace reads of that location with loads of a constant 12
> *provided* that the read still acted as a barrier to certain kinds of
> optimization.

Not if the read has effects other than yielding 12.

[...]

-- 
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"