Posts by Festus

  • Watchdog Reset

    I'm nearly (but not fully) 100% sure that Kernel V2.00 was already installed on Monday.


    No, I never tested Audio because we do not use Audio, but Rev 1.00 is printed on the board.


    There are different first exceptions:


    Code
    1. Exception 'Data Abort'(4) Thread-Id=03a70002(pth=bb461d00) PC=802626c8 BVA=00001000, dwInfo = 00000807
    2. R0=00000000 R1=eee5e5f8 R2=cc7d0000 R3=60860000
    3. R4=bb461d00 R5=cc7efce0 R6=00000000 R7=ffffc800
    4. R8=00000000 R9=00000001 R10=ffffc888 R11=ffffc824
    5. R12=ffffc800 SP=cc7efbd4 Lr=802626c8 PC=802626c8, Psr=2000001f


    Code
    1. Exception 'Undefined Instruction'(1) Thread-Id=07400152(pth=bb5e7750) PC=802626c8 BVA=0126effc, dwInfo = 00000807
    2. R0=00000000 R1=00000002 R2=6000011f R3=ffffc800
    3. R4=bb5e7750 R5=d6c9fca8 R6=00000000 R7=ffffc800
    4. R8=00000000 R9=00000002 R10=ffffc888 R11=ffffc824
    5. R12=ffffc800 SP=d6c9fb9c Lr=802626c8 PC=802626c8, Psr=2000011f


    Code
    1. Exception 'Prefetch Abort'(3) Thread-Id=021c0006(pth=bb46d7e0) PC=802626c8 BVA=d7e9efcc, dwInfo = 00000807
    2. R0=00000000 R1=00000002 R2=6000001f R3=ffffc800
    3. R4=bb46d7e0 R5=cc9bfcdc R6=00000000 R7=ffffc800
    4. R8=00000000 R9=00000001 R10=ffffc888 R11=ffffc824
    5. R12=ffffc800 SP=cc9bfbd0 Lr=802626c8 PC=802626c8, Psr=2000001f


    But the program Counter is equal in all cases.

  • Watchdog Reset

    If you Need more Information, here are the registry Settings of our can bus:



    By the way:
    I now made a test with an external power supply of 3.3V and I had teh same watchdog reset. So I assume that the Problem is not caused by a weak power supply.

  • Watchdog Reset

    Ok thanks,


    no hint yet, sorry. It seems that code 8 Comes right after the can port was opened but that it takes a while until the other exceptions are raised.

  • Watchdog Reset

    Did you already had a Chance to look into this?


    Yesterday afternoon, I again had a watchdog reset and again there was a debug Output CID: [0] (RX) Unexpected MB code 0x8.


    But this time, there where a ot of


    Code
    1. Exception 'Undefined Instruction'(1) Thread-Id=07400152(pth=bb5e7750) PC=802626c8 BVA=0126effc, dwInfo = 00000807
    2. R0=00000000 R1=00000002 R2=6000011f R3=ffffc800
    3. R4=bb5e7750 R5=d6c9fca8 R6=00000000 R7=ffffc800
    4. R8=00000000 R9=00000002 R10=ffffc888 R11=ffffc824
    5. R12=ffffc800 SP=d6c9fb9c Lr=802626c8 PC=802626c8, Psr=2000011f


    messages. I fount 26 Messages with "Undefined instruction" and 4 with "Data Abort" in the debug Output, the stack pointer (SP) Switches between d6c9fb9c and ffffxxxx (xxxx is different in every log message)


    I do not think it will help when I append the full debug Output again, what do you think?


    I'm not 100% sure, but I think that reading CAN Messages causes this Problem. If I run our application while the PicoMODA9 is not connected to our CAN bus, I didn't encounter a watchdog reset yet.


    I really need help here.

  • Watchdog Reset

    Some additional Information: I just saw that I set the priority of the can Driver to 50, so it seems to be possible that the can Driver can cause the watchdog reset.

  • USB (Virtual COM Port) and RS232 Problems

    I tested RS232 RTS/CTS handshake with Kernel 2.00 but it still does not work.


    Now I found another entry in the Roadmap for V2.10:
    - 0002890: [Serial Driver/Interface] RTS/CTS does not work. - new


    I think, this will solve my RS232 Problems... nor not?


    I still would like to know if Roadmap entry 2772 (see above) is related to my Problems.

  • Watchdog Reset

    (Continued because more than 10000 characters)



    As you can see, thare a a lot of exceptions in Thread 065002ee (No idea what this thread does or if that thread is one of out application threads)

  • Watchdog Reset

    I think I have to correct myself.


    I also did a test over the Weekend with a PicoMODA9 without any application running (Explorer was shown) and didn't had a watchdog reset.


    But today, I again had a watchdog reset while our applicacation was running.
    I now think that the Problem is caused by CAN bus communication.


    Here is the bootlog (fetched by DCU-Termi.exe):


  • Language support

    I just saw that V2.00 was released but I think there is a Problem with the zip file (when I try to unzip it, I get an error message that the zipped foler is invalid).
    The zip file of the new kernel also is only 1,1MB large. The last one had ~26MB.

  • Watchdog Reset

    I didn't has any application running when the reset occured.
    Just the Explorer was active.


    No I didn't has a chance to reproduce it and no I do not have any Serial debug Output (I didn't expect that I have to log the debug Output because I did not have any running application).

  • Language support

    Sounds good, but actually the Roadmap entry (0002626) is listed for Version 2.10 (although the scheduled release of 2.10 is only one day later as 2.00)