[m-dev.] for review: a big step towards the trace-based debugger (part 2 of 3)

[Andrew Bromage]

G'day all.

I've just noticed that some of the code in my latest diff overlaps
with the trace debugger, so I'm going to hold off that until this is
committed.  (Please review it anyway, though.)

On 20-Mar-1998, Zoltan Somogyi <zs at cs.mu.OZ.AU> wrote:

> Index: runtime/mercury_misc.c
> ===================================================================
> RCS file: /home/mercury1/repository/mercury/runtime/mercury_misc.c,v
> retrieving revision 1.4
> diff -u -r1.4 mercury_misc.c
> --- mercury_misc.c	1998/03/16 12:23:33	1.4
> +++ mercury_misc.c	1998/03/20 08:06:01
> @@ -468,6 +468,7 @@
>  void 
>  fatal_error(const char *message) {
>  	fprintf(stderr, "Mercury runtime: %s\n", message);
> +	MR_trace_report();
>  	exit(1);
>  }

Is this safe?  fatal_error() can be called from init_memory_arena()
which is called before MR_trace_init().  Does MR_trace_report() depend
on that?

Cheers,
Andrew Bromage