[U-Boot] fastboot boot base address behaviour

[Maxime Ripard]

On Thu, Apr 23, 2015 at 02:00:01PM +0200, Marek Vasut wrote:
> On Thursday, April 23, 2015 at 09:41:04 AM, Maxime Ripard wrote:
> > On Wed, Apr 22, 2015 at 05:56:23PM +0200, Marek Vasut wrote:
> > > > I've been trying to use fastboot (and especially the boot command) on
> > > > sunxi recently, and got it to work pretty fine (apart from PSCI, but
> > > > that's another story).
> > > > 
> > > > The only thing that worries me a bit is that by default, both the
> > > > fastboot tool and mkbootimg will generate an image with the kernel
> > > > address set to 0x10008000.
> > > 
> > > Looks like MX6 DRAM base address, so this should definitelly not be
> > > fixed to this address here. The +0x8000 offset is the kernel load
> > > offset.
> > > 
> > > > While it might work on some targets, it obviously doesn't on the
> > > > Allwinner SoCs that most of the time have the RAM mapped to 0x4000000,
> > > > which result in the kernel being relocated to some address that is not
> > > > in RAM, failing badly.
> > > 
> > > Yep.
> > > 
> > > > I would expect U-Boot to relocate the kernel to some reasonable
> > > > address, and not try to do something dumb by actually trusting
> > > > completely the boot image.
> > > 
> > > I'd expect the image to be correct in the first place though ;-)
> > > 
> > > > I guess one way to solve this would be to really treat 0x10008000 as
> > > > the default, and relocate the kernel to whatever value make sense on
> > > > the current platform (even though that needs to be defined).
> > > > 
> > > > That way, "fastboot boot zImage" would actually work out of the box,
> > > > without requiring to set the optional "-b" option to set the kernel
> > > > base address to some decent value.
> > > 
> > > Then I'd say such "default" address should be something like 0xffffffff .
> > 
> > I'd probably agree if we were on a perfect world :)
> > 
> > But the fact is that the tools that are creating these boot.img in
> > Android all have this default, and *will* fill the kernel address
> > header to this value.
> > 
> > https://android.googlesource.com/platform/system/core/+/master/fastboot/fas
> > tboot.cpp, line 72
> > https://android.googlesource.com/platform/system/core/+/master/mkbootimg/m
> > kbootimg.c, line 112
> > 
> > So even if we send a patch to Android itself, that won't fix the tools
> > that are already in the source code of published Android versions, or
> > the packages bundled in the distribution.
> > 
> > So I think that instead of blindly trusting these tools, U-Boot should
> > handle that address has the de facto default, even if it doesn't make
> > any sense :)
> 
> OK, now I agree. Maybe we should fix both -- the tools and U-Boot.

I'll try to cook up a patch, send it to AOSP and see how it goes.

Maxime

-- 
Maxime Ripard, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20150425/ecee03b1/attachment.sig>