On Fri, 9 Nov 2012, Lan Tianyu wrote: > When pm qos flags is changing, the pm core will keep device not > in RPM_SUSPEND via pm_runtime_get_sync/put(dev). When the flags are > changed, this will affect usb device status. If NO_POWER_OFF flag was > set when the device was power off, it would need to be resumed and > suspended again without power off. If NO_POWER_OFF flag was cleared > when the device was suspended without power off, it would need to be > resumed and suspended again with power off inorder to save more power. > > This patch is to add runtime pm callback for usb port device. In the callback, > usb device attached to the port will be resumed and suspended. No, this is completely the wrong way around. The device's resume routine should force the port to resume, not the reverse. Think of the port device as a parent of the USB device (even though it's not represented that way in the device hierarchy) and you'll see why. Also, all this port stuff is making hub.c a real mess. Before going any farther, can you split it all out into a separate source file? Alan Stern -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
