[MLton] power pc "port"

[Filip Pizlo]

> > Weird.  fixed-integer isn't doing anything too exotic, although it is
> > using the non-standard integer sizes.  When a regression fails, usually
> > the best thing to do is to compile and run it outside the regression
> > script and try to match up where the output deviates from the .ok file.
> > Also, you can try compiling with -exn-history true which might tell you
> > more about where the exception was raised.
> 
> Will play with that.

I rebuilt with -exn-history true, and here's what I got.

with history:
        <basis>/arrays-and-vectors/sequence.fun 107.16

(By the way this is the sub function that's failing on a bounds check.)

My first question would be, is there any way to get a full stack trace?

Anyway, I investigated further by adding some tracing to
fixed-integer.sml.  Turns out that this exception is occurring on the very
first test (Int2), in the very beginning where it tries to do
Int2.toString.  From there, I've been trying to figure out exactly how
toString gets to sub - with not very much luck.

Is there a document I can look at for hints on debugging this kind of
stuff?  My gut instinct is to slap print statements into the basis, in
places where I think that things are going wrong... is this the best way
to go?  Or is there a better way?

--
Filip Pizlo
http://bocks.psych.purdue.edu/
pizlo@purdue.edu