On 09/05/2012 01:15 PM, Sascha Hauer wrote:
Maybe you could try an older toolchain. There are some dark memories in my mind, but it could also be that my mind is fooling me...
Ok, i will try this..
The MLO is generated from barebox.bin with the omap_signGP tool. It seems the header is 520byte in size. You could compare these between both images.
Both configuration headers are the same.
I don't know though if this is enough for the OMAP ROM code to detect a valid image.
Meanwhile, i took a look at the MLO. The configuration header is ok, the image size in the GP header is correct, and the loading address is the same like in the working MLO. The first instruction in the image differs from the working one (the others too, of course).
So the MLO should at least get loaded and started.. strange. Norbert _______________________________________________ barebox mailing list barebox@xxxxxxxxxxxxxxxxxxx http://lists.infradead.org/mailman/listinfo/barebox