On Wednesday 14 June 2006 12:25 pm, Alan Stern wrote: > On Tue, 13 Jun 2006, David Brownell wrote: > > > > 9. Tie together suspend/resume operations on a device and on its > > > interfaces. > > > > Another way to put this is that you've reverted some of the changes > > from patch seven-of-nine. Any chance of not actually making those > > changes in the first place, and thereby simplifying both patches? > > Or maybe just re-ordering things so this goes first. > > Although you might think of this patch in that way, it's not really a > reversion.
In terms of functionality, it is. You were talking about the code more than the functionality. > > Along the lines of what we said earlier, I think the "right" approach > > would be to have a single sysfs file like usb3/3-1/usb_state rather than > > have all of usb3/3-1/power/state, usb3/3-1/3-1:1.0/power/state, and so > > forth. That could be defined as "suspended" or "active" (or whatever), > > writable from userspace, etc. So when power/state files vanish, a good > > solution would be in place (for USB). Comments? :) > > That's not a bad idea, although right now it would be redundant. Fodder > for a later patch... Just watching for interesting opportunities as you throw out ideas about what needs to change! ;) In this case, thinking about what should should replace those bogus power/state files. Maybe a "usb_device_driver" would also be able to associate its own power governor, to help throttle down VBUS current usage differently on laptops (be clever) vs desktops (dumb is safest). - Dave _______________________________________________ linux-usb-devel@lists.sourceforge.net To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
