NAND Formatting

1258021

Genius 5120 points

Other Parts Discussed in Thread: OMAP-L138

Hi,

I am using BSP_WINCE_ARM9_01_00_02 for OMAP-L138.

We have a NAND Chip mounted onto the customized HW and it uses OMAP-L138 SOM.

We have basically verified NAND Read and Write from CCS Level-0 Code and it works fine.

We have flashed the UBL and EBOOT using SFH Utility version (2.30)

The function responsible for flashing NK.bin is WriteNKToNAND, It is calling IsBlockBad.

It able to flash till 50th block. From 51st block to till the end of the flash it reports the blocks are bad.

So, my understanding is SFH Erases the first 50 blocks of the NAND.

Any suggestions on how to handle this?

Thank You & Regards,

GSR

over 14 years ago

0 WMK over 14 years ago

TI__Intellectual 1640 points

Yes sfh only erases the first 50 blocks.

DEVICE_NAND_UBL_SEARCH_END_BLOCK in Common\include\device_nand.h needs to be modified to the last block to be erased and recompile the tool.

sfh source can be downloaded from here

Regards

0 1258021 over 14 years ago in reply to WMK

Genius 5120 points

Hi WMK,

Thank You for the reply.

I am trying to understand how this was achieved during development of the OMAP-L138 BSP.

Also, I don't want to modify the TI code unless TI is planning to do same changes.

Regards,

GSR

0 1258021 over 14 years ago in reply to 1258021

Genius 5120 points

Hi,

I have modified NAND_eraseBlocks function under switch case UBL_MAGIC_ERASE to the following

if (NAND_eraseBlocks(hNandInfo, DEVICE_NAND_RBL_SEARCH_START_BLOCK, hNandInfo->numBlocks) != E_PASS)

with this change the code never returns and it doesn't come out from there. So, I have hardcoded the endblocks to 4000.

The code I have modified is in the following souce file OMAP-L138_FlashAndBootUtils_2_30\Common\sft\src\uartboot.c

Thank You & Regards,

GSR

0 WMK over 14 years ago in reply to 1258021

TI__Intellectual 1640 points

If you have any issue regarding the sfh tool, please refer to the SUPPORT page on the SourceForge web page where you download the sfh package.

Regards

0 1258021 over 14 years ago in reply to WMK

Genius 5120 points

Hi WM,

Thank You for the reply.

By the way do you have any information on why TI didn't come across this problem during development of the BSP?

Regards,

GSR

0 WMK over 14 years ago in reply to 1258021

TI__Intellectual 1640 points

Does your custom board have the same parts?

0 1258021 over 14 years ago in reply to WMK

Genius 5120 points

Hi WM,

Thank You for the reply...

MT29F4G08ABADAWP being used in our custom hardware where as in LogicPD schematics I do see MT29F4G08AACWC:C

Regards,

GSR

Processors

Processors forum

NAND Formatting