• State Answer Suggested
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 29 subscribers
  • Views 275 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TM4C123GH6ZRB poor debug performance

Juergen Gaertner
Intellectual 330 points

Hi community,

our prototype has been arrived and so I've switched from TM4F232 evalboard to the real target containing the above mentioned processor. In principle it works.

But for debug, the main difference is that I do not have anymore the small ICDI emulator on board (as it was on evalboard). Instead I am connected via a Blackhawk USB510 JTAG emulator. In principle this works too.

But, every single source step takes about 45 seconds. It seems that the emulator is waiting for some signals expecting from the target which never occur. So a timeout escapes. With this behavior I cannot work.

The interesting thing is that sometimes (in 10% of all tries) a normal stepping is possible but I cannot find out under which circumstances.

Does anyone know what this is and what to do?

CCS version is 5.5.0.00077

  • 0
    Guru 117855 points

    Have you beyond "single" prototype?   (our group's rule)  If so - does issue persist?

    Are external JTAG pull-up R's present/accounted for - and w/in 4K7 - 10K value?

    Might there be "interference" between JTAG connections - an additional post spoke to your desire to "share" JTAG between MCU & FPGA.  Suggest that you "break" any JTAG connection "commonality" - and repeat your tests...

    Update: Dawns now that our group has 3 LX4F232 eval boards - and connected via 10 pin mini header to our JLink.  (via paid IAR)  And - all program/debug operations ran correctly - avoiding your incident.  Suggest that you attach your new debug probe to this LX4F232 board - and see if the same (45S delay issue) persists.  Such would indicate a probe - or probe set-up issue - should the delay remain.  

    Having a new board - and new program/debug probe - demands that you isolate the issue's cause - prior to applying any/all corrections...  Facts/evidence you submit leaves unclear whether the Blackhawk probe, new MCU board - or both - cause/contribute to your issue...

  • 0 in reply to cb1_mobile
    Intellectual 330 points

    Root cause was found but the root's root is still unclear. If anybody wants to know what it was:

    Our target board is connected via USB to PC. A second PC connection comes from USB510 JTAG emulator. As long as the target USB driver is still not active all works fine and stepping takes the expected 1..2 seconds per step. But if the USB driver becomes ready the trouble starts. The CCS window on PC shows in its status line above: (Not Responding).

    So I moved to another PC still running Windows XP - everything worked well.

    I went back to my notebook and changed the one USB connection to another socket down on the docking station. Success.

    Maybe it's enough to have both connections in different hub pathes, I don't know. And most probable you will never see this problem.