Path: csiph.com!eternal-september.org!feeder.eternal-september.org!reader02.eternal-september.org!.POSTED!not-for-mail
From: Keith Thompson <kst-u@mib.org>
Newsgroups: comp.lang.c
Subject: Re: lvalue types
Date: Wed, 18 Apr 2018 11:01:57 -0700
Organization: None to speak of
Lines: 45
Message-ID: <ln604ol7fu.fsf@kst-u.example.com>
References: <2b687d74-f559-43a9-9aec-8a1067467ce2@googlegroups.com> <a4ad0855-8098-4bb3-b499-ca1ca457df99@googlegroups.com> <83589df8-fa61-4ffb-8fef-09ab50c8fcf4@googlegroups.com> <b9ba615a-a637-4c93-bd44-3f42d2578b2c@googlegroups.com> <646ca446-53a3-4006-a476-6f52caacd8c9@googlegroups.com> <uqcBC.238$o43.209@fx33.iad> <c2d54052-e0c6-4eef-b7c6-ff9839664bc7@googlegroups.com> <lnzi21ltpy.fsf@kst-u.example.com> <108b8472-481e-43b4-9988-2c7e4648449d@googlegroups.com> <lnvacplooa.fsf@kst-u.example.com> <2bf95ece-0d26-49aa-8629-87d9c47ef2ad@googlegroups.com> <lnmuy1lldy.fsf@kst-u.example.com> <97b9de3e-f3f8-45c1-8850-a29ca57069d5@googlegroups.com> <0411fdf5-9ca7-48b2-b7d8-23dd874e95d9@googlegroups.com> <b83e7af9-fbf9-4eb0-8178-808e1df46075@googlegroups.com> <f0cc8e46-f957-44a6-a6f0-b96697f33684@googlegroups.com> <bd5137e0-47d1-4555-9773-12b853f751f6@googlegroups.com> <lnefjclci2.fsf@kst-u.example.com> <fd77df70-1361-4c60-8214-f4e74059281c@googlegroups.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Injection-Info: reader02.eternal-september.org; posting-host="b3081569afdc5f121008704081152326"; logging-data="14773"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX18KhB2BcfYJvIhGbmW6MdZX"
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.1 (gnu/linux)
Cancel-Lock: sha1:tnP0vIrC8xmN0QiuBv9h7+0fZ+w= sha1:jd2u1/MBZyGgQAZ3qGLLY72ENSY=
Xref: csiph.com comp.lang.c:129420

supercat@casperkitty.com writes:
> On Wednesday, April 18, 2018 at 11:12:49 AM UTC-5, Keith Thompson wrote:
>> supercat@casperkitty.com writes:
>> [...]
>> > The ability to pass the address of an object to functions that can use
>> > the object, at least until the next time the object is addressed via other
>> > means, is no less fundamental than the ability to use the structure
>> > directly.  Both clang and gcc, however, *invent* a distinction between
>> > direct and indirect use of objects.
>> 
>> I'm probably going to regret asking this, but can you support the
>> claim in your last sentence?  (I request that you do so without
>> asking rhetorical questions.)
>
> I've posted code many times where both gcc and clang create such a
> distinction.
>
>     struct s1 { int x; };
>     struct s2 { int x; };
>     union u { struct s1 v1; struct s2 v2; } uarr[10];
>
>     int readS1x(struct s1 *p) { return p->x; }
>     void writeS2x(struct s2 *p) { p->x = 5; }
>
>     int test(union u *uarr2, int i, int j)
>     {
>       if (readS1x(&uarr[i].v1))
>           writeS2x(&uarr[j].v2);
>       return readS1x(&uarr[i].v1);
>     }
>
> Both gcc and clang will assume that there's no way an operation on a struct
> s2 can affect a struct s1, even if all operations on the storage using
> different types are separated by actions which derive pointers of those
> types from the same lvalue "uarr".

Can you substantiate your claim about what assumptions gcc and
clang make, by showing us a self-contained program whose output
clearly demonstrates those assumptions?

-- 
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"