In message 5E974ACF-3EBE-405A-92AC-BFC2FD391E43@kernel.crashing.org you wrote:
Right, lots of work :). The roadmap also depends on kernel driver of RapidIO.
Sure, I guess my point is that until there is something beyond fsl_rio_quirk that would use the framework I don't see why we should put it into u-boot.
Why we do not start the job with a formal framework?
I'll leave the decision up to Wolfgang. Just giving my two cents about it.
Always all difficult decisions are left up to me :-(
I'd appreciate to see a formal framework for this, but then I also think that this needs some generalization which may be difficult or even impossible with the initial implementation. So I will neither protest against nor insist on such a formal framework.
Best regards,
Wolfgang Denk