On Friday 20 August 2004 7:38 pm, Alan Stern wrote: > On Fri, 20 Aug 2004, David Brownell wrote: > > > > What about hubs, which are after all a very special sort of device? You > > > have structured the code so that children get suspended when the hub's > > > interface is suspended, even though the hub device may be left active. > > > > I sometimes thing of that as a missing feature: when all its children are > > suspended, the device itself should probably suspend too. That'd be > > a useful policy even if it were just internal to the hub driver. > > But it would have visible consequences in userspace. A user could suspend > all the children of a hub, then try to wake one of them up, only to find > that the resume fails because the hub is asleep!
As someone pointed out in one of the power management threads, that particular problem needs to be solved anyway: resuming a device has to resume an access path to the device. In fact I've been waiting for someone to notice that the current usb resume code doesn't actually do that. (Except with remote wakeup.) - Dave ------------------------------------------------------- SF.Net email is sponsored by Shop4tech.com-Lowest price on Blank Media 100pk Sonic DVD-R 4x for only $29 -100pk Sonic DVD+R for only $33 Save 50% off Retail on Ink & Toner - Free Shipping and Free Gift. http://www.shop4tech.com/z/Inkjet_Cartridges/9_108_r285 _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
