Steve Chen <[email protected]> writes:

> On Wed, 2009-01-21 at 10:01 -0800, Kevin Hilman wrote:
>> Steve Chen <[email protected]> writes:
>> 
>> > On Wed, 2009-01-21 at 15:12 +0300, Sergei Shtylyov wrote:
>> >
>> >> >
>> >> > - We (or at least *I*) had no desire to have the same kernel binary
>> >> >   run on both a da8xx and a davinci.  So, cutting out the davinci
>> >> >   runtime code & data that was wasting memory was "A Good Thing (tm)".
>> >> >   
>> >> 
>> >>    Kevin seems the only person interested in ahving the same kernel 
>> >> binary, not clear why though...
>> >> 
>> >
>> > Personally, I would love to see single binary that is capable of running
>> > on all DaVinci + OMAPLx/DA8xx EVMs.  It really does make testing
>> > integration much easier.   
>> 
>> Phew, so I'm not completely alone. ;)
>> 
>> > I also believe that it is worth the extra efforts.  However, I also
>> > find that not able to have a usable uImage for DA8xx very difficult
>> > to swallow.  Actually, that is the only real sticking point for me.
>> > I'll investigate how to get around this problem when time permits.
>> 
>> I agree, I don't like that option very much either.  But that being
>> the most significant issue, I think between us we can figure out
>> something that will work.
>> 
>> If nothing else, a simple patch to the da8xx uboot to replace any load
>> addresss that start with 0x8xxxxxx with 0xcxxxxxx should work just
>> fine, since the 0x8xxxxxxx region is obviously not a place where an
>> image could be loaded anyways.
>> 
>> BTW, did you try my idea of manually loading a zImage linked at
>> 0x8xxxxxxx on the da8xx at 0xCxxxxxx?  I'm pretty sure that should
>> work. If so, then my u-boot hack idea would work as well.
>
> I have not try it yet.  For one thing, I don't know how to start the
> kernel zImage under U-boot.
>

Loading and running a zImage can be done like this:

  tftp 0xc0008000 zImage
  go 0xc0008000

That's the equivalent of the 'bootm' on a uImage since the uImage has
a small header with the entry point in it.
 
Try the above with you current zImage (where Makefile.boot has the
0xCxxxxxxx addresses.)  If that works, Then undo the changes to
Makefile.boot and do the same again.

I just took a quick glance at arch/arm/boot/compressed/head.S and I'm
pretty sure it can detect if the zImage has been relocated to a
location that it has not been linked at.  If it has been relocated, it
fixes up various pointers and continues on.

BTW, I'm haging out in the #davinci channel on Freenode if you want to
continue discussions there.

Kevin


_______________________________________________
Davinci-linux-open-source mailing list
[email protected]
http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source

Reply via email to