• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 98 subscribers
  • Views 219 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.

AM3359: UBI file system issue sometimes leads to kernel crash dump and sometimes leads to kernel panic

Vanam Bala Raju
Prodigy 50 points
Part Number: AM3359


Hello team,

We are using nand  flash MT29F4G08ABADAWP-IT:D (8 bit) in our design and using ubi file-system 

Sometimes we are getting crash messages which are related to ubi file-system and sometimes it leads to kernel panic

We are unable to find why these errors are generating, and tried to search in forum for help but couldn't get it.

I am attaching the screenshots which we are getting in the device.

please go through the screenshots and suggest how to come out of this issue .........

Regards

Vanam Bala Raju 

  • 0
    TI__Guru 69847 points

    How often do you get those errors?

    Do they happen during normal operation, or during restart?

    Does your unit implement a clean shutdown (power-off sequence), with a graceful shutdown of Linux Kernel activities?

    Do you see the issue on specific boards, or all boards?

    Regards, Andreas

  • 0 in reply to Andreas Dannenberg
    Prodigy 50 points

    Hi Andreas,

    Out of 10 times,2-3 times we are facing crash issue.

    Crash messages are seen in normal operation,sometimes in restart also.

    we are making clean shutdown with power-off

    command every time.

    In all 5 proto boards,we are having this issue.

    Regards

    Vanam Bala Raju 

     

  • 0 in reply to Vanam Bala Raju
    TI__Guru 69847 points

    Hi!

    thanks for the additional details, and confirming you are using a graceful shutdown sequence. The issue seems to be happen quite often. Also I assume your proto boards are not very old, so the Flash chips should still be in good shape from a wear/erase-level point of view, correct?

    What does your "normal operation" look like? (How much/often data is written).

    Which Kernel version are you using?

    In case you are not using the current TI AM335x production Kernel at https://git.ti.com/cgit/ti-linux-kernel/ti-linux-kernel/?h=ti-linux-5.10.y please give this a try to see if this gives different results.

    Then, since AM335x is actually very well supported in the upstream Kernel, I'd suggest trying the latest tagged Kernel 'master' branch as well for testing purposes, which currently is v6.4, see https://github.com/torvalds/linux

    Those experiments help to ensure you are not affected by what might be a known fixed issue, and will also sometimes help the investigation in case different behavior is shown with the different kernel versions.

    Then, it might be helpful to temporarily enable additional debugging/logging/self-test features to see if this yields any additional helpful information, see http://www.linux-mtd.infradead.org/faq/ubifs.html#L_how_debug

    It's also possible that there is some HW related issue which results in the errors you saw, such as signal integrity related (traces too long, improper routing, EMI, etc.), or power-supply related (power supply too weak?), specifically as it relates to your Flash chip, so those should also all be double-checked.

    Regards, Andreas