Path: ...!eternal-september.org!feeder3.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: Tim Rentsch <tr.17687@z991.linuxsc.com>
Newsgroups: comp.lang.c
Subject: Re: Casting the return value of ...
Date: Sat, 30 Mar 2024 10:33:31 -0700
Organization: A noiseless patient Spider
Lines: 57
Message-ID: <86le5zipzo.fsf@linuxsc.com>
References: <uu416t$33u55$1@news.xmission.com> <20240328105203.773@kylheku.com> <K0jNN.158579$zF_1.100106@fx18.iad>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Injection-Date: Sat, 30 Mar 2024 17:33:34 +0100 (CET)
Injection-Info: dont-email.me; posting-host="8552662d97416c03078d7b971fbb1163";
	logging-data="1197312"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX1+yY/TmzqbZGgOGSWzhTxMYyIOj9F9/rm8="
User-Agent: Gnus/5.11 (Gnus v5.11) Emacs/22.4 (gnu/linux)
Cancel-Lock: sha1:ySb8D4zaRWuTnUgL4wfWsrgU/s0=
	sha1:5LHXmmhlhgaQC48+znPAxoaiIXI=
Bytes: 3285

scott@slp53.sl.home (Scott Lurndal) writes:

> Kaz Kylheku <433-929-6894@kylheku.com> writes:
>
>> On 2024-03-28, Kenny McCormack <gazelle@shell.xmission.com> wrote:
>>
>>> But here's the thing.  Is the casting of the result of dlsym()
>>> [to a function pointer type] necessary?
>>> Isn't this the same as "casting the return value of malloc()", where you
>>
>> Conversions between function pointers and data pointers are an
>> extension;  it is not well-defined behavior in ISO C.
>>
>> I also seem to remember something like this, but I cannot trust this
>> documentation without a chapter-and-verse citation.
>
> https://pubs.opengroup.org/onlinepubs/9699919799/functions/dlsym.html
>
>> Assuming it is true, there you have it;  if you're on POSIX, the compiler
>> is required to have the extension and it is connected to casting,
>> in which case the cast is required.
>
> I've used this form for the last two decades, with gcc, with no
> issues:
>
> /**
>  * Each shared object that simulates a Data Link Processor (DLP), will
>  * contain a single namespace-scope function <b>get_dlp</b> which constructs
>  * a DLP object of the specified type (for example, a #c_uniline_dlp,
>  * #c_card_reader_dlp, et alia).   <b>get_dlp</b> returns the constructed
>  * object as a #c_dlp object to the #channel function, which is then used
>  * by the I/O subsystem to request services of the DLP at the MLI level.
>  */
> typedef c_dlp* (*get_dlp_t)(const char *, uint64_t, c_logger *);
> ...
>     get_dlp_t   sym;
> ...
>
>     sym = (get_dlp_t)dlsym(handle, "get_dlp");
>     if (sym == NULL) {
>         lp->log("Invalid DLP shared object format: %s\n", dlerror());
>         unregister_handle(channel);
>         dlclose(handle);
>         return 1;
>     }
>
>
>> I think there was a time in the development of GCC when there was
>> a warning even with the cast.  I don't think it's enabled by default
>> now?
>
> We compile with -Wall -Werror and have never seen any warnings related to
> casting the result of dlsym(), and we build with GCC[4..13].

Do you use -pedantic?  Compiling with -pedantic using gcc 8.4.0
gives a warning diagnostic (and a fatal error if -pedantic-errors
is specified in place of -pedantic).