Re: [RFA] Fix to handle enums with values above LONG_MAX

["John R. Moore" <jmoore@redhat.com>]

Ok, this went on deaf ears... to illustrate, do the following:

------------------------- example.c ---------------------------
#include <stdio.h>

enum AAA
{
  a = 0x1,
  b = 0xFFFFFFFF,  /* If this is > LONG_MAX we core dump */
};

int
main ()
{
  unsigned int x = a;
  printf ("x = 0x%x\n", x);
}
----------------------- end example.c -------------------------

% gcc -g example.c -o example
% gdb -nw example
(gdb) print a
Segmentation fault (core dumped)
%

John Moore

On Tue, 8 May 2001, John R. Moore wrote:

>
> This fix has been tested on linux 2.4.4 kernel using gcc RedHat-2.97.
> Without this fix, gdb simply core-dumps.
>
>
> 2001-05-08  John Moore  <jmoore@redhat.com>
>
> 	* stabsread.c (read_huge_number): Fix to allow gdb to handle
> 	enums with unsigned long values above LONG_MAX.
>
> Index: gdb/stabsread.c
> ===================================================================
> RCS file: /cvs/cvsfiles/devo/gdb/stabsread.c,v
> retrieving revision 2.178
> diff -p -u -r2.178 stabsread.c
> --- gdb/stabsread.c	2001/03/26 19:54:39	2.178
> +++ gdb/stabsread.c	2001/05/07 22:45:47
> @@ -4433,10 +4433,7 @@ read_huge_number (char **pp, int end, in
>        p++;
>      }
>
> -  if (os9k_stabs)
> -    upper_limit = ULONG_MAX / radix;
> -  else
> -    upper_limit = LONG_MAX / radix;
> +  upper_limit = ULONG_MAX / radix;
>
>    while ((c = *p++) >= '0' && c < ('0' + radix))
>      {
>
>
>