Revert "ACPI / hotplug / PCI: Avoid doing too much for spurious notifies"
authorMika Westerberg <mika.westerberg@linux.intel.com>
Wed, 30 Oct 2013 12:40:36 +0000 (14:40 +0200)
committerRafael J. Wysocki <rafael.j.wysocki@intel.com>
Wed, 30 Oct 2013 14:28:52 +0000 (15:28 +0100)
commitab1225901da2d4cd2dcbae6840e93abbef417064
tree1d7e39205fb8be851f19bff36fed83c7edf42f7c
parent59612d187912750f416fbffe0c00bc0811c54ab5
Revert "ACPI / hotplug / PCI: Avoid doing too much for spurious notifies"

Commit 2dc4128 (ACPI / hotplug / PCI: Avoid doing too much for
spurious notifies) changed the enable_slot() to check return value of
pci_scan_slot() and if it is zero return early from the function. It
means that there were no new devices in this particular slot.

However, if a device appeared deeper in the hierarchy the code now
ignores it causing things like Thunderbolt chaining fail to recognize
new devices.

The problem with Alex Williamson's machine was solved with commit
a47d8c8 (ACPI / hotplug / PCI: Avoid parent bus rescans on spurious
device checks) and hence we should be able to restore the original
functionality that we always rescan on bus check notification.

On a device check notification we still check what acpiphp_rescan_slot()
returns and on zero bail out early.

Fixes: 2dc41281b1d1 (ACPI / hotplug / PCI: Avoid doing too much for spurious notifies)
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Tested-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
drivers/pci/hotplug/acpiphp_glue.c