Hi, It seems powertop is double counting the number of wakeups derived from USB devices. Since USB devices do not generate interrupts directly, should we only count the number of wakeups come from the host controller?
In the example capture below, powertop shows wakeup counts from both EHCI HCD interrupts and the USB Ethernet adapter. But the real interrupts only come from EHCI HCD. Am i missing something? Wakeups-from-idle per second : 26.9 interval: 15.0s no ACPI power usage estimate available Top causes for wakeups: 32.6% ( 8.5) USB device 1-2 : SMC2209USB/ETH (SMC) 32.1% ( 8.4) [ehci_hcd:usb1] <interrupt> _______________________________________________ Power mailing list [email protected] http://www.bughost.org/mailman/listinfo/power
