Re: FORTRAN :PATCH FOR PR 7179

[Toon Moene <toon at moene dot indiv dot nluug dot nl>]

bdavis wrote:

> The problem:
>
>      dimension  a(250000,250000)
>      a(1,1)= 0.
>      write(*,*) ' a',a(1,1)
>      end
>
> gives no error message.
>
>
> Here is what it should do:
>
>
> big.f: In program `MAIN__':
> big.f:1:
>         dimension  a(250000,250000)
>                    ^
> Array `a' at (^) is too large to handle.
>
> Patch tested on i386 / linux with no new failures. 
> The above snippet can be added to the "non-compile" directory of the 
> testsuite.
>    (Toon, what is the best way to post new test files?)
>
> --bud davis
>
> ---------------------------------------------------------------------------------------------------------------------------------------- 
>
> Index: gcc/gcc/f/com.c
> ===================================================================
> RCS file: /cvs/gcc/gcc/gcc/f/com.c,v
> retrieving revision 1.195
> diff -c -3 -p -r1.195 com.c
> *** gcc/gcc/f/com.c     13 May 2003 18:06:48 -0000      1.195
> --- gcc/gcc/f/com.c     18 May 2003 21:57:36 -0000
> *************** ffecom_sym_transform_ (ffesymbol s)
> *** 7407,7420 ****
>          {
>            ffestorag st = ffesymbol_storage (s);
>            tree type;
> -
> -           if ((st != NULL)
> -               && (ffestorag_size (st) == 0))
> -             {
> -               t = error_mark_node;
> -               break;
> -             }
> -
>            type = ffecom_type_localvar_ (s, bt, kt);
>
>            if (type == error_mark_node)
> --- 7407,7412 ----
Hmmmm, can we be sure that `s' being "too large" is the only reason that 
ffestorag_size (st) is zero ?

--
Toon Moene - mailto:toon@moene.indiv.nluug.nl - phoneto: +31 346 214290
Saturnushof 14, 3738 XG  Maartensdijk, The Netherlands
Maintainer, GNU Fortran 77: http://gcc.gnu.org/onlinedocs/g77_news.html
GNU Fortran 95: http://gcc-g95.sourceforge.net/ (under construction)