Hi Walter, LONG time no see/talk.  Anyway, as you may be aware, Comeau is in the process of supporting additional C backends for Comeau C/C++.  Till now, one needed VC++ in order to use Comeau under MS-Windows, and we're finally getting around to changing that.

We have a number of backend in various stages of beta including Borland, MinGW, lcc-win32 and DigitalMars.   One thing I've come across with dm is that it seems too strict with tentative definitions. Here's an example:

  static int j;
  static int j = -99;

  static int k;
  extern int k = 99;

sc reports:

  static int j = -99;
               ^
  tendef.c(5) : Error: 'j' is already defined
  extern int k = 99;
               ^
  tendef.c(8) : Error: 'k' is already defined

And although that's true, C's tentative definitions rules
should accomodate it, and hence no diagnostic.  Is there
an option to change this behavior?
-- 
Greg Comeau   What's next: additional Windows backends and 'export'! Comeau C/C++ ONLINE ==>     http://www.comeaucomputing.com/tryitout World Class Compilers:  Breathtaking C++, Amazing C99, Fabulous C90. Comeau C/C++ with Dinkumware's Libraries... Have you tried it?

Hi Greg! I think it's way cool that you're supporting DMC with Comeau. Of course, you should forget about those other compilers and concentrate on DMC <g>.

I'll look into the problems you identified.

-Walter

"Greg Comeau" <comeau@panix.com> wrote in message news:a1ht5s$n3q$1@panix3.panix.com...
>
> Hi Walter, LONG time no see/talk.  Anyway, as you may be aware, Comeau is in the process of supporting additional C backends for Comeau C/C++.  Till now, one needed VC++ in order to use Comeau under MS-Windows, and we're finally getting around to changing that.
>
> We have a number of backend in various stages of beta including Borland, MinGW, lcc-win32 and DigitalMars.   One thing I've come across with dm is that it seems too strict with tentative definitions. Here's an example:
>
>   static int j;
>   static int j = -99;
>
>   static int k;
>   extern int k = 99;
>
> sc reports:
>
>   static int j = -99;
>                ^
>   tendef.c(5) : Error: 'j' is already defined
>   extern int k = 99;
>                ^
>   tendef.c(8) : Error: 'k' is already defined
>
> And although that's true, C's tentative definitions rules
> should accomodate it, and hence no diagnostic.  Is there
> an option to change this behavior?
> --
> Greg Comeau   What's next: additional Windows backends and 'export'! Comeau C/C++ ONLINE ==>     http://www.comeaucomputing.com/tryitout World Class Compilers:  Breathtaking C++, Amazing C99, Fabulous C90. Comeau C/C++ with Dinkumware's Libraries... Have you tried it?

In article <a1j524$29uc$2@digitaldaemon.com>,
Walter <walter@digitalmars.com> wrote:
>Hi Greg! I think it's way cool that you're supporting DMC with Comeau. Of course, you should forget about those other compilers and concentrate on DMC <g>.

Of course :)

>I'll look into the problems you identified.

Thanks.
-- 
Greg Comeau   What's next: additional Windows backends and 'export'! Comeau C/C++ ONLINE ==>     http://www.comeaucomputing.com/tryitout World Class Compilers:  Breathtaking C++, Amazing C99, Fabulous C90. Comeau C/C++ with Dinkumware's Libraries... Have you tried it?