You may try.

set coverage to longdelay.log  && In an appropriate place
...
set coverage to && in another appropriate place

then look for loops in the log

>Ok, I'm stumped. I'm working on a problem which only occurs on some machines (can't find anything common between these machines), and is going to drive me nuts.
>
>What happens is, if a certain form is maximized (or sized over a certain amount), and you pull up another certain form (which is a search screen), CPU usage jumps to 100%, and the mouse acts like there's a 1-15 second delay between input and action (depending on the computer, and how large the first form is sized).
>
>BUT, if you give the focus to another windows application, wait a second, and go back, everything's normal (back to 0% CPU). Now, the 100% CPU usage actually occurs on every machine, but the mouse remains responsive. There's no VFP code running (I've checked with SET COVERAGE and SET STEP ON).
>
>The search form is model, the calling form is not. There are several layers of code between the code to respond to the click, and (calls a method which calls a method which calls a method which calls a method, etc). I *really* do not want to try and piece together the different methods into the same piece of code (there's a lot of it), but that's what I was going to try next. (smaller call stack?)
>
>Also, if you pull this form up from another form, even with the first one open, no problem. I'm thinking of using a Timer as a workaround, but I don't really want to.
>
>Any suggestions? Anyone seen this before?