On Sat, 22 Sep 2012, Ming Lei wrote: > On Fri, Sep 21, 2012 at 11:53 PM, Alan Stern <stern@xxxxxxxxxxxxxxxxxxx> wrote: > > > The hub will also be resumed when a child is accessed via libusb, and > > suspended when the child is released. As a result, the hub gets > > suspended and resumed multiple times: once for itself and once for each > > descendant. If the autosuspend timeout is large enough then the hub > > will be suspended and resumed only once. > > True. > > Considered 'lsusb' is used with less frequency and for test/debug > purpose under most of situations, it shouldn't be a big deal. > > A bit large timeout may introduce extra CPU wakeup, so it might be better > to set it as zero, IMO. Well, we can go ahead and try it. The main situation where it would matter would be if two connect or disconnect events occur in quick succession, and I admit this doesn't happen very often. 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
