On Jul 13, 2007, at 6:36 AM, Joerg Sonnenberger wrote:

> I still think that putting the parser in the kernel is the wrong
> approach. But having a small header that can be parsed by firmload  
> would
> be fine with me. A few restrictions are IMO perfectly fine like

Just jumping in the middle here without the background, but I've  
written s-record and intel-hex parsers for other embedded systems and  
they can be written with very little code which begets the question  
why you'd support a proprietary parser over an industry standard parser?