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.

linux-omap-2.6.32 watchdog timer not resetting

Other Parts Discussed in Thread: OMAP3503, AM3505

Hi,

      I'm seeing an issue with the linux-omap-2.6.32 kernel (TI's arago psp for omap3) where the watchdog timer is not resetting. I have enabled OMAP_WDT in the kernel config and ran the Documentation/watchdog/watchdog-test.c example
./watchdog-test -e; ./watchdog-test -e
But it doesn't reset the board after 60 seconds.
Has anyone else faced a similar issue? I'm using a custom OMAP3503 processor-based beagleboard design.
Elvis Dowson

  • Hi Elvis,

    I am quite not sure whether this is linked to your issue,

    if I remember correctly, on OMAP3EVM (with OMAP35x), we have seen that some times watchdog fails to reset the system cleanly. Please refer to old forum post on the same issue - http://e2e.ti.com/support/dsp/omap_applications_processors/int_omap35x_dsps/f/45/p/60021/219015.aspx

     

    Thanks,

    Vaibhav

  • Hi Vaibhav,

                           I cannot access that link.

     

    Elvis Dowson

  • I am able to access it, not sure what is wrong at your end. Can you try coping the link directly to your browser OR try searching with subject line "OMAP35xx evm REV G , watchdog reset failiure"?

     

    Thanks,

    Vaibhav

  • I also have the problem to access this page.

     

    In my situation, the watchdog timer did reset the system. However, got the following error and hanged.

     

    Texas Instruments X-Loader 1.45 (Oct 13 2010 - 09:00:15)

    ECC Failed, page 0x00080000

    35 28 0 e3 14 f0 9f e5 14 f0 9f e5 14 f0 9f e5 14 f0 9f e5 14 f0 9f e5 14 f0 9f

    e5 14 f0 9f e5 40 1 e8 80 a0 1 e8 80 0 2 e8 80 60 2 e8 80 c0 2 e8 80 20 3 e8 80

    80 3 e8 80 78 56 34 12 0 0 e8 80 0 0 e8 80 2c 3c eb 80 6c 2e ef 80 0 0 f e1 1f 0

     c0 e3 d3 0 80 e3 0 f0 29 e1 68 0 4f e2 4 0 80 e2 40 20 a0 e3 2 20 80 e0 1 11 a0

     e3 2 36 a0 e3 3 10 81 e0 3e 3b a0 e3 3 10 81 e0 f8 7 b0 e8 f8 7 a1 e8 2 0 50 e1

     fb ff ff 1a 24 7 0 eb 19 0 0 eb a4 0 4f e2 68 10 1f e5 1 0 50 e1 7 0 0 a 70 20

    1f e5 70 30 1f e5 2 20 43 e0 2 20 80 e0 f8 7 b0 e8 f8 7 a1 e8 2 0 50 e1 fb ff ff

     da 94 0 1f e5 1 7 40 e2 80 0 40 e2 c d0 40 e2 7 d0 cd e3 a0 0 1f e5 a0 10 1f e5

     0 20 a0 e3 0 20 80 e5 1 0 50 e1 4 0 80 e2 fb ff ff 1a 4 f0 1f e5

     

  • Does this happen consistently at your end?

     

    Thanks,

    Vaibhav

  • The chance becomes pretty hight (more than 50%) after we enable more drivers. We narrow down the problem to the NAND driver in x-loader. When reboot, the driver is not reliable. The data returned from the driver sometimes is not correct and will be failed on ECC check. But if read again, the data might be correct.

    I am not sure whether it is becuase more drivers are eanbled, or some drivers have a conflict.

     

  • Hi,

    I am having a reset issue with my AM3505 processor where the omap_wdt driver doesn't reset the CPU after stopping the watchdog pinging + timeout.

    This issue description felt quite similar but I have failed to open the suggested link or have success with the suggested phrase.

    Please advise if there is a known resolution / workaround for that,

    Best regards,

    Yakir