On Sat, 9 Sep 2017, Mike Simms wrote: > > Did you do this? > > > > Alan Stern > > Yes, the information was already in the message quoted in your reply: > > > > 4.11.11-300.fc26.x86_64 and 4.12.9-300.fc26.x86_64 > > > are showing identical settings - autosuspend definitely on for > > > device /sys/bus/usb/devices/2-1.3/2-1.3:1.0/0003:046D:C52B.0001/power This is not what I asked you to do. I asked you to check the contents of /sys/bus/usb/devices/2-1.3/2-1.3:1.0/power. Not /sys/bus/usb/devices/2-1.3/2-1.3:1.0/0003:046D:C52B.0001/power. > > > it shows in the powertop UI as good for both kernels and > > > > > > Folder contents are autosuspend_delay_ms (can't open), control > > > (auto), runtime_active_time (0), runtime_status (unsupported), > > Frustratingly I can't CC the linux-usb or linux-input mailing lists, my > emails are rejected as spam otherwise it would be a heck of a lot > easier to share all this information with you. You are in my reply > to all list, so you should get any further communications with Jiri too. The mailing lists reject email in HTML format, but that's not the case here. Maybe there's something about outlook.com they don't like. You could try signing up for a gmail account and sending from it. > Did you get the last email containing usbmon traces? You should have but > please check your spam or junk folder just in case. I did get it. Nothing unusual jumped out. The trace showed that when you started using the trackball again after it had been suspended, the wake-up procedure took a little under 0.1 seconds. That's long enough to be perceptible, and it's understandable that the delay could be annoying. (Incidentally, the output format is more condensed if instead of running the usbmon program, you do: sudo cat /sys/kernel/debug/usb/usbmon/2u >/tmp/mon.out or the equivalent.) When you collected the data for the /dev/bus/.../power files, was that while the trackball was active or while it was suspended? The default setting is to suspend it after a few seconds of inactivity, so you can force it to remain at full power by clicking it and moving it around while you copy the file data. It would be worthwhile to compare this with the file contents collected while the device was suspended. You know, this change in behavior could easily be the result of a bug fix. It's possible that a bug in the 4.11 kernel was preventing the trackball from suspending, and when the bug was fixed in 4.12, the kernel was then able to suspend the device automatically. Perhaps Jiri can find a change having that effect. Anyway, if you don't want the trackball to be suspended while it's not in use, you can simply turn off power-saving for that device. Either in powertop or in a shell script or in a udev script. Alan Stern -- To unsubscribe from this list: send the line "unsubscribe linux-input" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
