Hi Brian,
From: Enric Balletbo Serra [mailto:eballetbo@gmail.com]
2014/1/6 Stefan Roese sr@denx.de:
...
As there were parallel set of patches running between u-boot and kernel. hence, some patch-sets caused regression for OMAP3x platforms when booting using u-boot specifically for ecc-schemes (like BCH4_SW).
Hence this patch series fixes those regressions, and tests complete NAND boot sequence for multiple ecc-schemes on AM335x-EVM board. (following configurations are required for building u-boot driver which is compatible to kernel ecclayout for selected ecc-scheme)
(BCH8_HW) (HAM1_HW) (HAM1_HW) (HAM1_HW) (UBIFS) ROM ---------> SPL ---------> U-Boot ---------> Kernel ---------> File-System
(BCH8_HW) (BCH8_SW) (BCH8_SW) (BCH8_SW) (UBIFS) ROM ---------> SPL ---------> U-Boot ---------> Kernel ---------> File-System
(BCH8_HW) (BCH8_HW) (BCH8_HW) (BCH8_HW) (UBIFS) ROM ---------> SPL ---------> U-Boot ---------> Kernel ---------> File-System
*Configurations used to build u-boot and kernel for end-to-end NAND boot* +------------+--------------------------------------------+------------------+ | ecc-scheme | u-boot/SPL configs | kernel DTS | +------------+--------------------------------------------+------------------+ | | | | | HAM1_HW | #define CONFIG_NAND_OMAP_ECCSCHEME \ |ti,nand-ecc-opts= | | | OMAP_ECC_HAM1_CODE_HW | "ham1" | | (1-bit | #define CONFIG_SYS_NAND_ECCBYTES 3 | | | Hamming | #define CONFIG_SYS_NAND_ECCPOS \ | | | using h/w) | { 1, 2, 3, 4, 5, 6, 7, 8, 9, \ | | | | 10, 11, 12 } | | | | (for NAND page-size=2048) | | | | | | +------------+--------------------------------------------+------------------+ | | | | | BCH8_SW | #define CONFIG_NAND_OMAP_ECCSCHEME \ |ti,nand-ecc-opts= | | | OMAP_ECC_BCH8_CODE_HW_DETECTION_SW | "bch8" | |(8-bit BCH | #define CONFIG_SYS_NAND_ECCBYTES 13 |(without ELM node)| | using s/w | #define CONFIG_BCH | | |library for | #undef CONFIG_SPL_NAND_AM33XX_BCH | | |for ECC | #define CONFIG_SPL_NAND_SIMPLE | | | error | #define CONFIG_SYS_NAND_ECCPOS \ | | |correction) | {2, 3, 4, 5, 6, 7, 8, 9, 10, \ | | | | 11, 12, 13, 14, \ | | | | 16, 17, 18, 19, 20, 21, 22, 23, 24, \ | | | | 25, 26, 27, 28, \ | | | | 30, 31, 32, 33, 34, 35, 36, 37, 38, \ | | | | 39, 40, 41, 42, \ | | | | 44, 45, 46, 47, 48, 49, 50, 51, 52, \ | | | | 53, 54, 55, 56, } | | | | (for NAND page-size=2048) | | | | #define CONFIG_SYS_NAND_ECCSIZE 512 | | | | | | +------------+--------------------------------------------+------------------+ | | | | | BCH8_HW | #define CONFIG_NAND_OMAP_ECCSCHEME \ |ti,nand-ecc-opts= | | | OMAP_ECC_BCH8_CODE_HW | "bch8" | |(8-bit BCH | #define CONFIG_SYS_NAND_ECCBYTES 14 | | | using ELM | #define CONFIG_SPL_NAND_AM33XX_BCH |(with ELM node) | | h/w engine | #define CONFIG_SYS_NAND_ECCPOS \ |ti,elm-id=<&elm> | |for ECC | {2, 3, 4, 5, 6, 7, 8, 9, \ | | | error | 10, 11, 12, 13, 14, 15, 16, 17, \ | | |correction) | 18, 19, 20, 21, 22, 23, 24, 25, \ | | | | 26, 27, 28, 29, 30, 31, 32, 33, \ | | | | 34, 35, 36, 37, 38, 39, 40, 41, \ | | | | 42, 43, 44, 45, 46, 47, 48, 49, \ | | | | 50, 51, 52, 53, 54, 55, 56, 57, } | | | | (for NAND page-size=2048) | | | | #define CONFIG_SYS_NAND_ECCSIZE 512 | | | | | | +------------+--------------------------------------------+------------------+
#* In addition following patches need to be pulled for u-boot: http://lists.denx.de/pipermail/u-boot/2013-December/168506.html http://lists.denx.de/pipermail/u-boot/2013-December/169021.html
Test1: flash ubi image from u-boot and boot the kernel U-boot> mw 0x82000000 0xff <u-boot.img size aligned to NAND block boundary> U-boot> fatload mmc 0 0x82000000 u-boot.img U-boot> nand erase <u-boot_offset> <u-boot.img size> U-boot> nand write 0x82000000 <u-boot_offset> <u-boot.img size> U-boot> setenv bootargs 'console=ttyO0,115200n8 noinitrd mem=256M \ root=ubi0 rw rootfstype=ubifs ubi.mtd=<mtdpart-of-rootfs>,\ <page-size> ip=off init=/init' U-boot> bootm <kernel_offset>
Test2: update u-boot.img from kernel and re-boot Kernel> flash_erase /dev/<mtdpart-of-u-boot> 0 0 Kernel> nandwrite -s 0 /dev/<mtdpart-of-u-boot> u-boot.img Kernel> reboot
Signed-off-by: Pekon Gupta pekon@ti.com
Though I have done initial level of testing on AM335x as mentioned above, But will it be possible for you to test and confirm if these set of patches solve regressions on your OMAP3 boards ?
Those patches work fine on our custom AM335x board. So:
Tested-by: Stefan Roese sr@denx.de
Thanks, Stefan
Sorry for this long delay, I'm doing this in my free time and sometimes it's difficult to find the time. Those patches also worked on my OMAP3 boards so
Tested-by: Enric Balletbo i Serra eballetbo@gmail.com
Many thanks for your great work.
Best Regards, Enric
This patch-set fixes pending regression on OMAP3 platforms and now it has received Tested-by from: - Stefan Roese sr@denx.de for AM335x platform - Enric Balletbo i Serra eballetbo@gmail.com for OMAP3 platform
So request you to please queue this on priority. Also, please add following as I missed this while preparing the patches. Reported-by: Enric Balletbo i Serra eballetbo@gmail.com
Thank you. with regards, pekon