Hi Hendrik Boom,

I have not followed this thread very much, but I am just curious about the
length of your ackerman function. 483 lines seem too lengthy to me.
My frontend produces around 270 lines including I/O, but the main function
takes less than 70 lines (see
http://ttic.uchicago.edu/~wchae/files/MLPolyRforC--.pdf). If you don't mind,
I would like to see your C-- code and compare it with mine until the C--
team can be available.

Regards,
Wonseok Chae
http://ttic.uchicago.edu/~wchae/


On Thu, Feb 12, 2009 at 2:43 PM, Hendrik Boom <[email protected]>wrote:

> On Mon, 26 Jan 2009 22:01:42 -0500, Norman Ramsey wrote:
>
> >> I gather this isn't supposed to happen:
> >  >
> >  >
> >  > qc-- -globals -o ACKERMAN ACKERMAN.c-- This can't happen: Asked for
> >  > temporary in space `general-purpose temporaries' with unsupported
> >  > width 64 Fatal error: exception Impossible.Impossible("Asked for
> >  > temporary in space `general-purpose temporaries' with unsupported
> >  > width 64") make: *** [ACKERMAN] Error 2
> >  > hend...@lovesong:~/dv/lang/a68h/workspace$
> >  >
> >  > It might be useful if a this-can't-happen were to produce the source
> >  > line number of the construct occasioning the problem.
> >
> > this-can't-happen means that the compiler has become so sick that it
> > can't continue.  source coordinates are but a distant memory. We usually
> > get a stack trace, which is useful...
> >
> >  > ACKERMAN.c-- is 483 lines.  Do you want a copy?
> >
> > Yes.  Though I can't see doing anything about it until after the ICFP
> > deadline (March 2).   I can guess where this message is coming from, and
> > the code is foul.  Very foul.
>
> Well, I have more than enough urgent, important, other things to do to
> keep me busy until then;  I can wait.
>
> But there's a part of me that's itching to go on.  I can't promise to
> have the time to go through it, but is there any chance you could direct
> me as to where to look in the very foul source code?  References to any
> available documents explaining how the code is intended to work would be
> helpful, too.  If nothing else I'll learn something while waiting, even
> if only how difficult it still is to read other people's programs.
>
> -- hendrik
>
>
> _______________________________________________
> Cminusminus mailing list
> [email protected]
> https://cminusminus.org/mailman/listinfo/cminusminus
>
_______________________________________________
Cminusminus mailing list
[email protected]
https://cminusminus.org/mailman/listinfo/cminusminus

Reply via email to