Ok, it seems that recent changes really did made /proc/acpi/wakeup
unnecessary.
Let me explain from beginning the current situation:
1. Indeed if I enable manually wakeup on both usb controller, wakeup
works.
usb controller = /sys/devices/pci0000:00/0000:00:1a.0/power/wakeup
usb device = /sys/devices/pci0000:00/0000:00:1a.0/usb2/2-2/power/wakeup
UDEV rule to set wakeup on usb controller works:
SUBSYSTEM=="pci", DRIVER=="uhci_hcd", ATTR{power/wakeup}="enabled"
SUBSYSTEM=="pci", DRIVER=="ehci_hcd", ATTR{power/wakeup}="enabled"
UDEV rule to set wakeup on device itself just doesn't work, even though
udev executes it:
ACTION=="add", ENV{DEVTYPE}=="usb_device", ATTR{power/wakeup}="enabled"
I did try to add my usb id for mouse like you did. It of course doesn't
help.
I also found that above rule, *sometimes* works, but very rarely, so
indeed something else sets it. I strongly suspect its kernel.
I also think that I shouldn't need to set wakeup to enabled on usb
controller. This should be done automaticly as soon as I set 'enabled'
on usb device.
Best regards,
Maxim Levitsky
_______________________________________________
linux-pm mailing list
linux-pm@xxxxxxxxxxxxxxxxxxxxxxxxxx
https://lists.linux-foundation.org/mailman/listinfo/linux-pm
