Hi, The device wakeup code in pci-acpi.c currently has a (theoretical) problem that it can disable wakeup for a bridge prematurely in some obscure conditions. This is described in some more detail in the changelog of patch [3/3] and it has been there for quite a while, so it's nothing new, but it would be good to fix it at last and hence this series. Patches [1/3] and [2/3] are mostly preparatory, although the former may be regarded as a fix by itself and the latter is a cleanup. Thanks, Rafael
