Lpc2000

OK, wow have I messed up.  They say as you get old, the mind is the
second thing to go.  Yesterday when I finally stated my conclusion
that the gcc-3.4 snapshot was working OK, well, err, I was wrong.
Between the mail list activity, meetings, and other work, I had
removed the snapshot and I was actually compiling and testing the code
with the GNUpro distribution that has always worked for me.

This morning I reloaded the gcc-3.4 snapshot and recompiled from
scratch.  Using an older version of Insight, I could not get the code
to run on either my eval board or in the simulator.  In the simulator,
I could single-step (asm level) it seems for as long as I wanted.  But
anything greater than that (C lang step) would cause the simulator to
start executing code where there was no code.  I could not get even
that far with Insight and a JTAG connection to my eval board (running
in RAM on an EB40A).  The code jumps into never-never land during the
port initialization routine well before interrupts are enabled.

I'll try to spend some more time on the problem in an attempt to
isolate it.  Sorry for the repeated erroneous reports.

-Bill Knight
R O SoftWare