🚀 go-pugleaf

I wrote some image convolution code on a straight MacOS MACII, using
5MB, system 5, using MPW C v2.0.  Because this uses a lot of floating
point multiplication, I used the following compiler directives to
directly generate instructions for the coprocessor and 68020:

	C -mc68020 -mc68881 -elems881 imagefuns.c

As required, I linked with CLib881.o, CSANELib881.o, and Math881.o .

The code runs well and very fast as described (45 seconds versus 24
minutes).

Here is the problem.  When I run the code on the A/UX at work, the
machine freezes up.  This only happens when I specify the faster
compiler options.  It seems to especially freeze up on floating point
operations. 

Finally, the question:  Is there a known problem with A/UX configured
machines using MPW V2.0 and the faster compiler options?

thanks,
phil...