On Wed, 27 Apr 2005, David Brownell wrote: > > Why should USB core's auto-suspend depend on CONFIG_PM? In > > particular, if this results in (at least partial) > > duplication of the auto-suspend functionality in HCDs. > > It could be argued two different ways: that "hub autosuspend" > should exist in usbcore, and cover root hubs; or alternatively > that drivers should handle their own suspend logic, and so root > hub autosuspend should continue to driven by the HCDs.
Um, I would interpret that second argument differently. To me, the part of an HCD that drives the root hub is more akin to the firmware in an external hub. It's an "emulator", if you like. The actual "driver" (in the conventional sense of software that tells the device what to do and interprets the results) is the regular hub driver. By this way of thinking, HCDs shouldn't implement autosuspend any more than external hubs do. Pragmatically, HCDs do need to autosuspend their root hubs because otherwise the controller causes a large and unnecessary drain on system resources -- and also because the hub driver doesn't do autosuspend yet. > It'd also be easy to argue that without CONFIG_PM there's no > need for any autosuspend ... Yes, although host controllers may be a particularly egregious case. Not only do they consume power, they use up plenty of PCI bus cycles. (Maybe not too many though, if no devices are plugged in.) Alan Stern ------------------------------------------------------- SF.Net email is sponsored by: Tell us your software development plans! Take this survey and enter to win a one-year sub to SourceForge.net Plus IDC's 2005 look-ahead and a copy of this survey Click here to start! http://www.idcswdc.com/cgi-bin/survey?id=105hix _______________________________________________ linux-usb-devel@lists.sourceforge.net To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
