[U-Boot] U-boot NAND boot support on Freescale PDK
Hi:
Was going through the patch for uboot posted earlier this year and the related review comments here:
http://www.mail-archive.com/u-boot@lists.denx.de/msg01140.html.
From the thread, it looks like no one is working on the u-boot nand
boot on imx31 pdk. Please correct me if i am wrong.
I was told that if that the mx31 phycore patch should work just fine for most things on the PDK and hence can be used as a base for a port to the PDK. I am ignorant of the hardware details on MX31 Phycore but when i had a look at the low level init code (lowlevelinit.S) for Phycore and compared it to the existing lowlevelinit.S for the MX31 3stack board, i realized that they are a lot different when it comes to the NAND SPL.
I am a newbie to NAND boot. Do i have to rewrite the NAND SPL code for preparing NAND boot images for the 3stack, Or is there a way i can translate some existing code(Maybe redboot code) for the MX31 3stack(PDK). I have modified the top level Makefile to include the sources for getting the SPL code compiled but am stuck at the lowlevelinit.S file.
My final intent is to get the uboot boot out of NAND on MX31- 3stack(PDK) although i have managed to get uboot running from RAM on the platform.
Thanks in Advance! Alfred.
2009/4/2 alfred steele alfred.jaquez@gmail.com:
Hi:
Was going through the patch for uboot posted earlier this year and the related review comments here:
http://www.mail-archive.com/u-boot@lists.denx.de/msg01140.html.
From the thread, it looks like no one is working on the u-boot nand boot on imx31 pdk. Please correct me if i am wrong.
Until yesterday you were correct. I spent some time updating the published patches to the current u-boot and the status now is that I was able to compile the stuff. Haven't tried it on a PDK board yet but I'm hoping to do that in a couple of days. I will post patches when the stuff is running, the patches will most likely not be ready to be merged as there are previous comments to take care of.
/Magnus
Hi Magnus,
Thanks for the response:)
Until yesterday you were correct. I spent some time updating the published patches to the current u-boot and the status now is that I was able to compile the stuff. Haven't tried it on a PDK board yet but I'm hoping to do that in a couple of days. I will post patches when the stuff is running, the patches will most likely not be ready to be merged as there are previous comments to take care of.
Ii had actually embarked myself on modifying the lowlevelinit.S and start.S to create a SPL for uboot although i wasn't sure of finishing it that soo;(. I hope these files apart from the linker script(u-boot.lds) and the config files(config.mk and Makefile) are the ones that need to change for getting NAND boot to work. Please correct me if i don't understand it correctly.
I wasn't actually sure why there is a need to create a separate linker script like u-boot_nand.lds.
So i hope the patch would be ready in a couple of days. If you want me to try them on the PDK before you actually release it, that would be great.
Another thing i noticed while poking around the code is that though MX35 PDK had SDHC(sd card) drivers in it, the MX31 did not. Would it be in a individual GIT , e.g the U-boot mmc git.
Thanks, Alfred.
On Fri, Apr 03, 2009 at 10:21:50AM -0500, alfred steele wrote:
I wasn't actually sure why there is a need to create a separate linker script like u-boot_nand.lds.
The addresses may be different than normal u-boot (though we could take care of that by preprocessing the linker scripts), and you don't need to pull in as many sections (though it would be harmless to do so if those sections are empty in the loader). mpc8313erdb uses the nand_spl linker script to provide a link-time warning if the size limit is exceeded.
A more basic question is why does each board need its own linker script, NAND or not?
-Scott
participants (3)
-
alfred steele -
Magnus Lilja -
Scott Wood