Re: Application modernization & re-engineering (was: What should I say to a *nix community?)

Terrence Enger wrote:

Or if you've ever debugged one with an oscilloscope.

You win Joe. I concede.

It's a true story. That was when I wrote standalone operating systems for communication devices. We had a problem where the entire OS ran fine in debug mode (debugging involved something called an In-Circuit Emulator, or ICE, a big honking machine that intercepted all the leads from the CPU and allowed you to basically step through your code one clock cycle at a time).

Unfortunately, it didn't run so fine when plugged into the actual production box. Turns out the timing diagram on the USART chip didn't match the actual timing of when the chip presented an interrupt. The early interrupt caused a jump to an "unpredicted" location. The ICE slowed everything down enough that the interrupt didn't occur before we were ready for it <grin>.

Joe