linaro-release team mailing list archive

["Tixy \(Jon Medhurst\)" <879385@xxxxxxxxxxxxxxxxxx>]

On Fri, 2011-10-21 at 14:32 +0100, Jon Medhurst (Tixy) wrote:
> On Fri, 2011-10-21 at 12:46 +0000, Mattias Backman wrote:
> > I'm afraid that the patch may cause problems with all the other
> > boards, since it uses the load_addr for something that current hwpacks
> > do not expect. I'm not certain that the other boards will have
> > problems, but to be confident we should test this very carefully.
> 
> Actually there isn't any bug, it was a misunderstanding on my part. I
> now see that:
> 
>   load_addr is the address the kernel executes from.
>   kernel_addr is the address to load the image to.
> 
> I thought it was the other way around.
> 

Actually, that can't be it, because it uses kernel_addr for bootm, that
would need to be load_addr. I'll do some more investigation...

-- 
Tixy

-- 
You received this bug notification because you are a member of linaro-
infrastructure-drivers, which is the registrant for Linaro Image Tools.
https://bugs.launchpad.net/bugs/879385

Title:
  boot.scr contains wrong kernel load address

Status in Linaro Image Tools:
  Invalid

Bug description:
  The U-Boot bootcmd environment created by BoardConfig._get_bootcmd()
  uses 'kernel_addr' for the address to load the uImage, it should use
  'load_addr' instead. The current use of 'kernel_addr ' has the effect
  of loading the kernel to the address it will be run from, which U-Boot
  interprets as meaning the image is XIP and will end up trying to boot
  Linux by execute the image header and therefore crashing.

  This affects booting Versatile Express with U-Boot.

  I have attached a patch which fixes this.

To manage notifications about this bug go to:
https://bugs.launchpad.net/linaro-image-tools/+bug/879385/+subscriptions