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.

How to replace problematic NAND flash IC on evmdm365 board?

Expert 2315 points

I received a batch of evmdm365 boards last year and one of them was labelled as having a dodgey NAND when we initially fired them up and put our evaluation code on.  I am busy with this specific board and the seocnd time now since we got it it does not want to boot from NAND - it loads kernel but running it up gets to a point of kernel panic.  I want to put a new NAND chip in but all the documentation are just too confusing for me wanting to operate the whole story without having to dive too deep if not necessary.  Is there someone who can give an N-point plan on what to do to get it to boot from the new flash as I am struggling to figure out exactly how RBL, UBL and u-boot fits together with the kernel image and filesystem I load onto NAND and get it to boot from there like the out-of-the-box board.  I am using dvsdk_2_10_01_18 and does not have CCS running yet and would prefer to get this going without it.

thanks, hijn.

  • After discovering that "something" wrote an extra meg of a directory, specifically modules/2.6.18_pro500-davinci_evm-arm-v5t-le/ into my filesystems lib/ directory and that this pushed my ramdisk size over the limit of what I was writing to NAND I got it going with the bigger ramdisk.gz file by increaing the size.  It was happy to run from RAM but not happy to boot from NAND after power-up.  Removing the suspicious directory from my filesystem and using the previous ranges writing to NAND is also working again.

    I think that "something" wrote the extra directory into my filesystem on a powerfailure we had.  Any ideas if it is so and how I can prevent it other than an ups on my development host?

    Thanks, Jinh T.