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.

DM365 NAND EXCHANGE ISSUE.

 

Hello everybody.

Now I'm using DM365 and I have used the NAND(samsung K9F1G08UC) but it has been kill off the produce.
so I decided to change the nand part to K9F1G08U0D–SCB0 .
However, we are currently facing serious problems.
current symptoms is that Nand writing is no problem. all files - UBL,Bootloader, kernel, ramdisk - writing without error.
After reboot, UBL and u-boot works fine.
But, when I read the Nand section on the u-boot prompt, every Nand section are recognized bad section include ubl, u-boot and can't read the data.
I have no idea what's going on.

When I compare to both datasheet, read/write timing is a bit changed and ECC detection is changed from 4bit to 1bit.
Current nand write use 4bit ECC. if it's changed, does it possible to occurs above error?
I searched the e2e, but it seems no problem.( http://e2e.ti.com/support/dsp/davinci_digital_media_processors/f/100/p/98648/345060.aspx#345060 )

What do I need to check?
We should go into mass production right now. Please give a goot hint.

Regards.
Jay.

  • It's likely the ROM is using 1bit ECC based on the device ID, however, the u-boot is probably still using 4bit ECC scheme and layout.  Therefore you may need to modify u-boot to use 1 bit ECC. 

     

  • hello jay,

    I am facing the same problem like you described. Have you been successfull to solve this?

     

    Josef

  • Hi:

              Which uboot you are using, are you using the uboot from dvsdk? I have some problems with the nand flash, which after switching  from K9F1G08U0C to K9F1G08U0D , the uboot version was 1.3.4, after switch to uboot from dvsdk, my problem is fixed.

    Patrick

  • Jay

    There are timing difference in the NAND. Refer to the AC timing characteristics. This you can try by changing the timing parameters. But I think the timing parameters may not help you directly. I feel that the busy pin has to be used properly. I'm not really sure in your u-boot its implemented properly. I have seen issues before in busy/wait pin support in u-boot.

  •  

    I missed to check the verify.

    I notify to the development team and there's a no more issue.

    Marcus gave the clue and it may fixed on newer version of dvsdk as Patrick said.

    I appriciate your interest.

    Thank you.

    Jay

  • I tested New u-boot source code. (DaVinci-PSP-SDK-03.21.00.04)

    Samsung nand (SEC_K9F1G08U0D) is normal behavior.

    Samsung nand, I know that there are differences in the timing. (tR)

    so, I want to see source code about Samsung nand's patch.

    1) source code path ,  function name

     I think, The kernel driver must also be considered.

    now kernel error log)
    nand_davinci_4bit_compare_ecc Too many errors to be corrected!
    nand_davinci_4bit_compare_ecc Too many errors to be corrected!
    nand_davinci_4bit_compare_ecc Too many errors to be corrected!
    nand_davinci_4bit_compare_ecc Too many errors to be corrected!

    If you know please respond.

     

     

     

  • Hi Jay,

        can you confirm the problem was only on the1 bit ecc and not on the timings?

    Did you have to modify the kernel too?

    Best Regards,

    mario