On Mon, 10 Jan 2011 07:57:24 -0800 Greg KH <greg@xxxxxxxxx> wrote: > On Sun, Jan 09, 2011 at 11:58:58AM -0800, Yinghai Lu wrote: > > move quirks for usb handoff early for x86. > > > > So we can get stable result when trying to calibrate apic timer with pm-timer. > > Is this a real problem today? What is the symptom of the issue? Where > was it reported as a problem? Yes, please give us a rationale for all this mess. I really don't like the sleep wrappers or the "early" PCI device; would it be easier to extract some of the core disabling MMIO from the quirks and re-use it for a totally separate set of early quirk functions? Like Ben suggested, making it totally arch specific would be a good option, though presumably most arches don't need this (again if you explained why/when this was needed maybe we could come up with a better solution, or just ignore the problem if you're trying to fix a pre-production board again with all sorts of broken stuff). -- Jesse Barnes, Intel Open Source Technology Center -- 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
