Strange finding (CCS rev 3.186)

Guest · Posted: Tue Sep 21, 2004 1:55 pm

Recently, I ran into a series strange problems with 18F8620 & CCS

When I burn a "good" program my designs, some of them will not run.

If I give pic a reset signal, it will run fine.

so the problem seems to be in the reset, but I confirm good clock/reset/voltage, etc -- I contacted microchip and show them my design and fuse map, and they didn't find anything wrong.

I can reprogram the pic many times and it still will not come out of reset

Yet, if I add a few dummy lines in my source, as simple as "i=i-1; i=i+1':, recompile and program pic again, they will all work fine!

Sad

dyeatman · Joined: 06 Sep 2003 Posts: 1934 Location: Norman, OK

What speed are you running? Are you using 4xPLL?

PCM programmer · Joined: 06 Sep 2003 Posts: 21708

There have been posts on this type of problem before.
http://www.ccsinfo.com/forum/viewtopic.php?t=20195&highlight=pll

Can you post a small test program that demonstrates the problem ?
Be sure to show all pre-processor statements. Then we may
notice something and be able to help.

Guest · Posted: Wed Sep 22, 2004 9:20 am

29.xx MHZ

Because this problem only happens to some but not all systems, and the program is rather big, and any change will take care the problem, it will not be easy to come up with a small test program to demonstrate

:(

asmallri · Joined: 12 Aug 2004 Posts: 1635 Location: Perth, Australia

It may be an interrupt related software problem

PCM programmer · Joined: 06 Sep 2003 Posts: 21708

Study these posts by C-H Wu. Try his recommendation to
reduce the #opt level. See if that helps.
http://www.ccsinfo.com/forum/viewtopic.php?t=19386
http://www.ccsinfo.com/forum/viewtopic.php?t=19439
http://www.ccsinfo.com/forum/viewtopic.php?t=19233
http://www.ccsinfo.com/forum/viewtopic.php?t=19417

Mark · Joined: 07 Sep 2003 Posts: 2838 Location: Atlanta, GA

Check the errata on the PIC. Some of the bigger 18's had problems at over 4MHz. Does the date code start with 03 or 04?