On Fri, Jun 14, 2013 at 06:36:06PM +0100, Peter Maydell wrote:
> On 14 June 2013 18:14, Sören Brinkmann <soren.brinkm...@xilinx.com> wrote:
> > On Fri, Jun 14, 2013 at 05:56:31PM +0100, Peter Maydell wrote:
> >> If we're providing separate functions for kernel and
> >> ramdisk loading shouldn't they check that the uimage
> >> has the appropriate type?
> > I'm not sure I understand what you mean here. Moving the check for the
> > appropriate u-boot header type down here? I tried to find some
> > reasonable way to further split the load_uboot_image() routine to move
> > some of it's functionality into the two functions down here. But it all
> > seemed pretty messy and I left pretty much all functionality in
> > load_uboot_image() and just pass errors through.
>
> I had in mind that these functions should pass the "expected"
> type byte in to the load_uboot_image() function so it can
> fail if the blob it is passed isn't actually the right type
> (eg you pass the kernel to -initrd and vice-versa).
Thanks for the clarification. I agree. I'll add such a check.
Sören
