On Thu, 2015-05-28 at 15:08 -0700, Yinghai Lu wrote: > On Wed, May 27, 2015 at 6:30 PM, Rafael J. Wysocki <rjw@xxxxxxxxxxxxx> wrote: > > On Wednesday, May 27, 2015 02:31:49 PM Bjorn Helgaas wrote: > > For debug you can always use pcie_ports=compat and that will disable > > pciehp too. > > That will disable AER at the same time, right?. For ppc machines under either our hypervisor or running bare metal with our EEH-enabled bridges, both generic AER and hotplug code are going to interfere. EEH "subsumes" AER, and we have firmware interfaces to do hotplug that know more than what the generic code can know about (such as on-board GPIOs that can control PERST, it's not always via the hotplug registers or hotplug from top-level slots which isn't the same as hotplug from switch slots on our platforms). So basically, at this point, we must not have the PCIe port drivers at all, just just interfere. We should replace that with the appropriate "hooks" for the platform to disable selected functions (hotplug and AER) from the generic port drivers. Cheers, Ben. -- 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
