On Thursday 10 April 2008, Benjamin Herrenschmidt wrote: > On Thu, 2008-04-10 at 12:21 +0200, Stefan Roese wrote: > > Sure, this would be optimal. But frankly, I currently have no need for > > this > > non U-Boot mode, and therefore I didn't implement it. This can be > > added later > > when really needed don't you think? > > It can... but on the other hand, it's not hard :-)
But it takes time. And time is as always short. :) > > Good. But again, I would really prefer to first include this U-Boot > > style > > endpoint support and later add this independent device tree endpoint > > configuration when really needed. > > Well, you are the one with a use case... but I do think we should > differenciate a PCI in host mode vs. endpoint mode right away. We don't > want the PCI stack from kicking in at all in endpoint mode, we don't > want to bother with creating the PCI infrastructure and config space > accessors etc... Understood. But for this "dynamic" root-complex/endpoint configuration as implemented with the current version of the patch, I can't change the PCIe node in the dts from root-complex mode to endpoint mode, since I would loose root-complex functionality. Perhaps I'm missing something here. Best regards, Stefan _______________________________________________ Linuxppc-dev mailing list Linuxppc-dev@ozlabs.org https://ozlabs.org/mailman/listinfo/linuxppc-dev
