From: cwr AT cts DOT com (Will Rose)
Newsgroups: comp.lang.c,comp.os.msdos.djgpp
Subject: Re: Casting void pointers
Followup-To: comp.lang.c,comp.os.msdos.djgpp
Date: 23 Jun 1998 00:19:17 GMT
Organization: CTS Network Services (CTSNET), San Diego, CA
Lines: 33
Message-ID: <898561284.92944@optional.cts.com>
References: <6mkaos$k7o AT dfw-ixnews6 DOT ix DOT netcom DOT com> <358DC1FA DOT 443E297D AT cs DOT com> <898511636snz AT genesis DOT demon DOT co DOT uk>
NNTP-Posting-Host: optional.cts.com
Cache-Post-Path: optional.cts.com!cwr AT crash-i2 DOT cts DOT com
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Precedence: bulk

Lawrence Kirby (fred AT genesis DOT demon DOT co DOT uk) wrote:
: In article <358DC1FA DOT 443E297D AT cs DOT com> fighteer AT cs DOT com "John M. Aldrich" writes:

: ...

: >#include <stdio.h>
: >
: >int foo( int x )
: >{
: >    return x + 10;
: >}
: >
: >int main( void )
: >{
: >    void *vp = foo;

: Your compiler should have generated a diagnostic for this. The C language
: requires a cast to convert between void * and function pointers (in both
: directions). Note that C doesn't guarantee that a void * object can properly
: hold a funciton pointer. You should avoid doing this.

OTOH, I think:
  int main(void)
  {
    void (*p)(void) = (void (*)(void)) &foo;

should work.  I think the & operator is un-necessary on modern compilers.


Will
cwr AT crash DOT cts DOT com