Re: 48GB NUMA-Q boots, with major IO-APIC hassles

From: Anton Blanchard (anton@samba.org)
Date: Wed Jan 15 2003 - 06:24:12 EST


> -- I've already tried and failed on PCI segments.
> -- I would be much obliged for advice on this one, especially
> -- since the workaround cripples the machine.

I was wondering if we could use pci_scan_bus_parented as a start.
Assuming we fixed pci_bus_exists, this should allow us to create
overlapping busses.

What else needs fixing? To start with:

* pci config read/writes
        Should be arch specific, can hide things in struct pci_bus
        which is passed in. (eg on ppc64 we have to pass an identifier
        into the low level config read or write)

* pci IO/memory reads/writes
        No problems on ppc64, it should just work. Are there problems
        on x86?

* /proc/bus/pci/
        As davem pointed out we have a backwards compatibility bit
        where all devices on domain 0 appear as they always have.
        Additionally we create another level of directories to represent
        the domain.

* /proc/pci
        Need to print the domain

* device printing
        We need a macro that drivers can use to print their PCI location
        Then we make that print the domain.

* sysfs
        I havent looked into this yet

Does anything else spring to mind?

Anton
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Wed Jan 15 2003 - 22:00:54 EST