21.2 on Irix 6.5 segfault on startup; patch.

21.2 on Irix 6.5 segfault on startup; patch.

[Judah Milgram]

X-Mailer: mh-e 5.0.90; nmh 1.0.4; Emacs 20.4

In GNU Emacs 21.2.15 (mips-sgi-irix6.5, X toolkit)
 of 2002-10-04
configured using `configure  --prefix=/usr/local/gnu'
Important settings:
  value of $LC_ALL: nil
  value of $LC_COLLATE: nil
  value of $LC_CTYPE: nil
  value of $LC_MESSAGES: nil
  value of $LC_MONETARY: nil
  value of $LC_NUMERIC: nil
  value of $LC_TIME: nil
  value of $LANG: C
  locale-coding-system: nil
  default-enable-multibyte-characters: t

Compiled with gcc version egcs-2.91.66 19990314 (egcs-1.1.2 release)
Problem seen with binutils 2.13, then I completely uninstalled
binutils and re-compiled gcc (thus configuring to use system ld etc.)
and rebuilt emacs. No change.

Please describe exactly what actions triggered the bug
and the precise symptoms of the bug:

Seg fault & core dump on startup.

Believe it or not, the following patch fixes things. Don't  ask me why.


*** gmalloc.c.dist	Sun Nov  4 12:35:43 2001
--- gmalloc.c	Fri Oct  4 17:03:41 2002
***************
*** 987,992 ****
--- 987,994 ----
      = FINAL_FREE_BLOCKS + 2 * __malloc_extra_blocks;
  
    register struct alignlist *l;
+   char gronked[7]="gronked";     /* On Irix 6.5, prevents seg
+                                     fault on startup. Go figure. */
  
    if (ptr == NULL)
      return;


- Judah Milgram
  milgramjh@nswccd.navy.mil

Re: 21.2 on Irix 6.5 segfault on startup; patch.

[Richard Stallman]

    Compiled with gcc version egcs-2.91.66 19990314 (egcs-1.1.2 release)
    Problem seen with binutils 2.13, then I completely uninstalled
    binutils and re-compiled gcc (thus configuring to use system ld etc.)
    and rebuilt emacs. No change.

    Please describe exactly what actions triggered the bug
    and the precise symptoms of the bug:

    Seg fault & core dump on startup.

    Believe it or not, the following patch fixes things. Don't  ask me why.

I suspect that is due to a bug in the old GCC version you are using.
It is not worth debugging, but does the latest GCC work right?