• State
  • Locked Locked
  • Replies 9 replies
  • Subscribers 102 subscribers
  • Views 976 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

File corruption issue on SD card in omap3503 with 2.6.32 linux

shital gandhi1
Prodigy 140 points
Other Parts Discussed in Thread: OMAP3503

Hi All,

My board boots up from SD card, I often see SD card corruption issue on Linux 2.6.32 kernel with omap3503 processor, this will create problem on boot up or on some applications to run due to some file corruption.

I ran SD card stress test on the board and will get data mismatch error on average every 30 cycle.

Can any one help me to find a root cause?

What could be the reason, is it related to SD card driver, I performed this test on brand new 4 GB SD card.

Thanks in advance.

  • 0
    TI__Guru** 116100 points
    Hi,

    My board boots up from SD card, I often see SD card corruption issue on Linux 2.6.32 kernel with omap3503 processor, this will create problem on boot up or on some applications to run due to some file corruption.

    What you have done before corruption ?
    Any abrupt power OFF for testing etc., ?

    What is your SDK version ?
    Is it released by TI ?
  • 0 in reply to Titusrathinaraj Stalin
    Prodigy 140 points
    I have upgraded my device,
    Yes it is release byTI.
  • 0 in reply to shital gandhi1
    TI__Guru** 116100 points
    What stuff you are doing in up gradation ?
    Is that recommended tool ?
    Able to reproduce the problem ?
    How many board having this problem ?
  • 0 in reply to Titusrathinaraj Stalin
    Prodigy 140 points

    Hi Titusrathinaraj Stalin,

    Thank you for your prompt response.

    I am doing firmware upgrade (upgrade mechanism is developed by us which will create and write 6 partition of SD card) in every board and I am facing 1 time failure due to file corruption of SD card out of 10 upgrade.

    When i ran SD card stress test (write pattern and read same pattern) on ext4 partition of SD card i often got data mismatch error.

    The other strange problem I am facing is whenever I delete files using rm or find, that command hangs (i.e. CPU utilization of rm command reaches to 99% and it is not killed by applying kill -9 <pid>) then only option is left is reboot. (This problem occurred 1 out of 50 times)

    I am also applying sync after each rm operation , then e2fsck but got the same result.

    Even I am facing same problem for umount command.

    Tell me if you need more information.

  • 0 in reply to shital gandhi1
    Genius 3720 points

    Hi Shital,

    Is it your custom board ?


    In that case does the electrical characteristics of the SD socket is verified ? Check for voltage fluctuations and also the mechanical fit is compact !!

    The file corruption occurs in SD generally due to the power supply issue or any noise the track signals.

  • 0 in reply to lyf sci
    Prodigy 140 points
    Hi,

    Yes i had checked all electrical characteristics of the SD socket, ans its working fine.
    All this problems occured radomly after some iterations.
    Does we need to do anything related SD card driver or file system (currently we are using EXT4 as ordered data mode mount), earlier in EXT3 we faced more problems.
  • 0 in reply to shital gandhi1
    Prodigy 140 points

    Hi All,

    My SD card host controller driver uses dma and high speed configuration, registered in platform data of board file.
    Is there anything to do with that?

    Regards,
    Shital

  • 0 in reply to shital gandhi1
    Genius 3720 points
    Hi Shital,

    I had faced similar issues over a period of time and it is majorly due to the signal integrity issue which could be due to the driver or mechanical contacts.

    How did you verify the electrical characteristics ? Have checked for any Noise in the MMC signals with any analyzer ?

    Also try SD cards from other vendors and check for signal integrity.

    Regards,
    Lyf
  • 0 in reply to lyf sci
    Prodigy 140 points

    Hi Lyf,

    I have verified all IO signals and timing parameters of SD card using CRO, and all are OK.

    I am using Netlist Micro SD HC 4 GB.

    Actually the problem is very often,  I am getting SD card data blocks values as zero even if using sync.

    Thanks.