On Saturday, August 01, 2015 at 06:08:56 PM, Simon Glass wrote:
Hi Marek,
Hi!
On 28 July 2015 at 09:57, Marek Vasut marex@denx.de wrote:
On Tuesday, July 28, 2015 at 05:53:10 PM, Masahiro Yamada wrote:
[...]
$ git describe v2015.07-307-g2647394 $ make vpac270_ond_256_defconfig # # configuration written to .config # $ make -s CROSS_COMPILE=arm-unknown-linux-gnueabi- ===================== WARNING ====================== Please convert this board to generic board. Otherwise it will be removed by the end of 2014. See doc/README.generic-board for further information ==================================================== $ ls -l spl/u-boot-spl.bin -rwxrwxr-x 1 masahiro masahiro 2828 Jul 28 00:53 spl/u-boot-spl.bin
It is over 2KB.
So the hardware spilt already? Aww :-(
I wonder, maybe converting this to TPL might be the correct solution ?
Maybe, if you could split it into two images.
I think I might need TPL and SPL:
TPL: Load the SDRAM init code into CPU's SRAM from the OneNAND SPL: Load big U-Boot from OneNAND into the main SDRAM
... or maybe we should just sacrifice the machine in the name of progress and be done with it. I don't know when I might have time to fix it afterall.
Well it sounds like it is already broken. It worries me that SPL has somehow got larger though - any reason for that?
I think the machine was always quite borderline and stuffing the SPL into it was always a bit hackish. The changes to the SPL probably just exposed this hackishness.
Best regards, Marek Vasut