On Thu, Aug 25, 2011 at 5:14 PM, Deng-Cheng Zhu <dengcheng.zhu@xxxxxxxxx> wrote: > 2011/8/25 Bjorn Helgaas <bhelgaas@xxxxxxxxxx>: >> No, I just mean that I don't see why you need this patch at all. If >> you pass the list of bus resources to pci_create_bus(), there's no >> need to fix anything up later. Or am I missing something? > > Well, doing the root resource fixups in here is a *paranoid* way. It's to > deal with the 'unlikely' circumstance where controller_resources() returns > the NULL pointer in pcibios_scanbus() due to memory allocation failure. > Most of the time (always) it's nothing more than repeating the resource > list setup. But maybe we can do something like this: > > if (unlikely(!dev && list_empty(&bus->resources)) > pcibios_setup_root_resources(bus, hose); > > > What do you think? I don't think it's worth it. Just check for failure of controller_resources(), and if it fails, skip the pci_create_bus() call. You've already printed a message (you might add the PCI domain/bus number). Bjorn -- To unsubscribe from this list: send the line "unsubscribe linux-pci" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html