>>> On 07.01.15 at 16:31, <boris.ostrov...@oracle.com> wrote: > On 01/07/2015 10:06 AM, Jan Beulich wrote: >> >>>> Of course an additional question would be whether the node wouldn't >>>> better go into struct arch_pci_dev - that depends on whether we >>>> expect ARM to be using NUMA... >>> Since we have CPU topology in common code I thought this would be >>> arch-independent as well. >> Not sure what you're referring to here: What common piece of data >> stores the node of a particular CPU? cpu_to_node[] clearly is x86- >> specific. > > I meant that we have common interface (sysctl) for querying topology and > PCI is a common IO bus (unlike CPUs) so I figured I'd keep node in > common part as well.
Makes little sense: ARM stubs out cpu_to_node(), so a similar mechanism could be used to hide the fact that some arch-es may have no node associated with a PCI device. Jan _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel