Problems during Debug
Anders Eliasson
Posts: 1
Quote "KenM" from the thread "trouble programming":
>I used to have a problem single stepping where the program would jump to a random address with no instructions, and for some unknown reason I no >longer have that problem. The only thing different between when the problem used to happen and now is the date. ??? (all equipment same)
I have exactly the same problem (except that I still have the problem)
I have tried several SX28, two different SXKey (rev. F), two different host computers and two different target boards, and end up with the same results.
Sooner or later (while single-stepping, Walk and Run) the CPU jumps to some random FFF location and runs like crazy.
The two computers run Win 2k and 98SE.
Changing FIFO buffer size doesn't help.
Any ideas??
>I used to have a problem single stepping where the program would jump to a random address with no instructions, and for some unknown reason I no >longer have that problem. The only thing different between when the problem used to happen and now is the date. ??? (all equipment same)
I have exactly the same problem (except that I still have the problem)
I have tried several SX28, two different SXKey (rev. F), two different host computers and two different target boards, and end up with the same results.
Sooner or later (while single-stepping, Walk and Run) the CPU jumps to some random FFF location and runs like crazy.
The two computers run Win 2k and 98SE.
Changing FIFO buffer size doesn't help.
Any ideas??
Comments
If I start to have the problem again (luckily I haven't) I am going to plug the SX key directly into the lap top serial port, no cable.
Another alternative is to use Gunethers SX Simulator. The latest version is in an attachment to one of the posts in this forum.
The simulator obviously does not replace exact hardware debugging, but for me it has been very useful to step through my code to confirm the logic and flow.
Ken
from time to time, I have the same trouble with the SX-Key when single stepping. My assumption is that somehow the communication between the Debugger-IDE and SX-Key gets messed up somehow (maybe due to a spike). This seems to bring them out of sync, letting the IDE "believe" that the program counter actually points into unprogrammed "nowhere land".
I also noticed that the SX-Key (or the SX-Key in combination with a specific SX chip) seems to be quite sensitive on the supply voltage. When you have a power supply with variable output, try to increase the voltage slightly above 5, or slightly below 5 Volts.
I remember that I even had problems programming SX chips when the supply voltage was a bit out of range (but this only happended with the first SXes, the so called "old silicon" versions).
Greetings,
G