usb_autosuspend should not be enabled on desktop systems
Bug #814289 reported by
Michael Hudson-Doyle
This bug affects 3 people
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
powernap |
Won't Fix
|
Undecided
|
Andres Rodriguez |
Bug Description
usb autosuspend interacts very unpleasantly with using a mouse (the pointer freezes when usb is suspended, requiring a button click or other usb activity to wake it up).
How a 'desktop' system is determined, I don't really have an idea for. Maybe just a system with a pointer attached?
Changed in powernap: | |
status: | New → Confirmed |
Changed in powernap: | |
assignee: | nobody → Andres Rodriguez (andreserl) |
Changed in powernap: | |
status: | Confirmed → Won't Fix |
To post a comment you must log in.
There's enough data in the /sys/bus/ usb/devices/ directory to see that a particular device is an input device (eg by the presence of an 'input/' subdirectory, or by looking at the bInterfaceClass. Perhaps in that case powernap should avoid putting it to sleep, or at least it should turn up the timeout substantially?
It's not clear to me yet whether it's enough to suppress autosuspend only for that device or if it also needs to be manually done up the device tree so that it stays reachable. (But I thought the kernel took care of the latter.)
(Perhaps there are other input devices that can be asleep, and that save substantial energy, without any down side?)