https://bugzilla.kernel.org/show_bug.cgi?id=84651

            Bug ID: 84651
           Summary: Powerbutton not detected/handled on Surface Pro 3
           Product: ACPI
           Version: 2.5
    Kernel Version: 3.16.2
          Hardware: All
                OS: Linux
              Tree: Mainline
            Status: NEW
          Severity: normal
          Priority: P1
         Component: Power-Other
          Assignee: [email protected]
          Reporter: [email protected]
        Regression: No

Created attachment 150461
  --> https://bugzilla.kernel.org/attachment.cgi?id=150461&action=edit
full dmesg output

Summary:
-------------------------

Power button (and other hardware buttons including volume up/down and
capacitive home button) on the Microsoft Surface Pro 3 fail to function under
3.16. This prevents wake from suspend using the power button. I am using usb
plug/unplug events to wake the device, but I note that power consumption is
very high in suspend and am not sure if this is also a related ACPI issue or
simply due to my use of USB to wake from suspend. Either way, a working power
button would be a step in the right direction.

I'm including as much detail as I feel is relevant here based on review of
other bug reports and my own reading on ACPI. I apologize if I have missed
anything critical and will be happy to amend this report as necessary.



Details:
-------------------------

Under Linux, button presses (power, volume, capacitive home button)
successfully generate System Control Interrupts (both power and volume buttons
increment the interrupt counter). Test method: monitoring /proc/interrupts and
/sys/firmware/acpi/interrupts. Attachments include pre/post interrupts output.

However, they are either not exposed as ACPI devices or are not properly
handing the event off to be handled by acpid or systemd.



Tests:
-------------------------

Buttons are confirmed physically working under dual boot into Windows. Power
button successfully powers on device when booting to Linux.

Running acpid and acpi_listen generates no events when pressing the power (or
volume) hardware button.

xinput shows no input device for the power button.

Running evtest on /dev/input/event0 (associated with
/proc/acpi/button/lid/LID0) shows no events either (this may be a lid sensor
unassociated with power button). Directly monitoring
/proc/acpi/button/lid/LID0/state also shows no change on power button presses.

There are no other items under /proc/acpi/button.

As mentioned, there is a button device under /proc/acpi/button/lid/LID0
associated with /dev/input/event0 but I’m not sure/convinced this is the
hardware powerbutton or if it is a laptop close sensor for the detachable
keyboard cover for this unit (which I do not own). Regardless, it does not
change state when pressing the power button (as monitored by watching
/proc/acpi/button/lid/LID0/state and pressing button).



dmesg ACPI power notices (full dmesg in attachments)
-------------------------
# dmesg -t | grep -i power
ACPI: Power Resource [PRWF] (on)
ACPI: Power Resource [CAMP] (on)
ACPI: Power Resource [PAUD] (on)
ACPI: Power Resource [TPWR] (on)



Attachments:
-------------------------
I am attaching relevant output including:
# dmesg
# cat /proc/acpi/wakeup
# grep . /sys/firmware/acpi/interrupts/* (pre and post power button depress)
# grep . /sys/bus/acpi/devices/*/hid
# ls /sys/bus/acpi/drivers/*/



Kernel / Hardware summary
-------------------------
Linux palimpsest 3.16.2-1-ARCH #1 SMP PREEMPT Sat Sep 6 13:12:51 CEST 2014
x86_64 GNU/Linux
description: Laptop
product: Surface Pro 3 (Surface_Pro_3)
vendor: Microsoft Corporation
version: 1

-- 
You are receiving this mail because:
You are watching the assignee of the bug.
------------------------------------------------------------------------------
Want excitement?
Manually upgrade your production database.
When you want reliability, choose Perforce.
Perforce version control. Predictably reliable.
http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
_______________________________________________
acpi-bugzilla mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/acpi-bugzilla

Reply via email to