[Swanky] Brightness keys (F6 and F7) are not working. |
||||||||||||||
Issue descriptionChrome Version: 61.0.3163.47 OS: 9765.29.0 DUT: Swanky What steps will reproduce the problem? (1)sign in to the device. (2)Try to use brightness keys F6 and F7. What is the expected result? Brightness keys should work. What happens instead? Brightness keys are not working. Please use labels and text to provide additional information. For graphics-related bugs, please copy/paste the contents of the about:gpu page at the end of this report.
,
Aug 16 2017
,
Aug 16 2017
We see a pass result on kb_input test - https://wmatrix.googleplex.com/unfiltered?hide_missing=True&tests=platform_InputBrightness&releases=61&platforms=swanky&days_back=10 The test emulates keys pressed by setting event code for the kb input device. This should be keyboard/kernel issue.
,
Aug 16 2017
snanda@ this is currently a beta blocker. Will you please take a look and let us know who can take a look at this? This doesn't look like a beta blocker. WDYT?
,
Aug 16 2017
Are the brightness keys broken on Swanky only or on other platforms too? Probably a beta blocker if it is indeed broken more widely.
,
Aug 17 2017
Squawks in Chrome OS 9765.30.0, 61.0.3163.51 works fine.
,
Aug 17 2017
Can we try 9765.30.0, 61.0.3163.51 on swanky as well? Also, to be clear on swanky are we trying the brightness up / down keys on the built-in keyboard or F6 / F7 on an external USB keyboard?
,
Aug 17 2017
helenzhang@ can you please update results based on comment #7 by snanda@?
,
Aug 17 2017
@comment #7 Still able to reproduce this issue on swanky 9765.30.0, 61.0.3163.51 Facing same issue with built-in keyboard and USB keyboard. Note: Only level 0 brightness(Display goes balck) is working.
,
Aug 18 2017
This is display/drm related as I can see the backlight requests arriving in the driver when keys are pressed, 2017-08-18T10:08:05.791546-07:00 DEBUG kernel: [ 43.479280] [drm:intel_backlight_device_update_status] updating intel_backlight, brightness=13085/65535 2017-08-18T10:08:05.791571-07:00 DEBUG kernel: [ 43.479292] [drm:intel_panel_actually_set_backlight] set backlight PWM = 13085 2017-08-18T10:08:05.791576-07:00 DEBUG kernel: [ 43.479308] [drm:edp_panel_vdd_on] Turning eDP port C VDD on 2017-08-18T10:08:05.791579-07:00 DEBUG kernel: [ 43.479319] [drm:edp_panel_vdd_on] PP_STATUS: 0x80000008 PP_CONTROL: 0xabcd000f 2017-08-18T10:08:05.811078-07:00 DEBUG kernel: [ 43.499964] [drm:intel_backlight_device_update_status] updating intel_backlight, brightness=12827/65535 2017-08-18T10:08:05.811102-07:00 DEBUG kernel: [ 43.499972] [drm:intel_panel_actually_set_backlight] set backlight PWM = 12827 Sontis also mentioned that the screen stayed black after s2r. The mosaic was still working but backlight no longer does. We could potentially see regression from power_Idle measurement results (keyval level_backlight_max & level_backlight_current) but there's no coverage for R61 https://wmatrix.googleplex.com/unfiltered?suites=power_build&releases=61&platforms=swanky couple of runs on 62, https://wmatrix.googleplex.com/testrun/unfiltered?test_ids=537260333 But the logs aren't available, (500 internal error) https://ubercautotest.corp.google.com/tko/retrieve_logs.cgi?job=/results/135801866-chromeos-test/chromeos4-row9-rack13-host9/power_Idle.AC_OFFLINE ----------------- So back to logs (attached) The one thing I see is recently added controls for backlight over aux, ./messages:2017-08-18T10:07:24.011226-07:00 DEBUG kernel: [ 0.449721] [drm:intel_dp_aux_init_backlight_funcs] AUX Backlight Control Supported! Now that should ONLY activate for panels edp panels supported 1.4 via, https://chromium-review.googlesource.com/c/606779 # landed in 9765.22.0 And reading panel's DPCD[0] its = 12 -> 1.3 via, # define DP_EDP_13 0x02 But then we shouldn't see the debug message. Adding i915.enable_dpcd_backlight=0 to the bootargs solves the problem. Opal can you have a look?
,
Aug 18 2017
Panel dpcd is lying, it claim that PWM pin is not supported so it didn't fallback to that even panel is eDP 1.3 At dpcd[0x702] == 0xf6 which mean bit0 == 0 and bit1 == 1 From the spec (see picture), this mean 1. Panel can't use PWM PIN to adjust brightness 2. Panel can use AUX to adjust brightness But the reality is the opposite of what it claim using DPCD. # cat /sys/kernel/debug/dri/0/eDP-1/i915_dpcd 0000: 12 0a c2 41 00 00 01 c0 02 00 00 00 1f 0b 00 0070: 01 00 0080: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0100: 0a 82 00 00 00 00 00 00 01 08 00 0200: 01 00 77 00 01 01 00 00 0600: 01 0700: 02 0701: b7 f6 00 00 0720: 01 1a 2a 5b 0a 04 0a 00 84 00 00 01 00 ff ff 03 0732: 00 14 Upload https://crrev.com/c/622047 + merge request
,
Aug 18 2017
This bug requires manual review: M61 has already been promoted to the beta branch, so this requires manual review Please contact the milestone owner if you have questions. Owners: amineer@(Android), cmasso@(iOS), ketakid@(ChromeOS), govind@(Desktop) For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
Aug 19 2017
Approving merge to M61 Chrome OS.
,
Aug 19 2017
This bug requires manual review: M61 has already been promoted to the beta branch, so this requires manual review Please contact the milestone owner if you have questions. Owners: amineer@(Android), cmasso@(iOS), ketakid@(ChromeOS), govind@(Desktop) For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
Aug 19 2017
Approving merge to M61 Chrome OS.
,
Aug 19 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromiumos/third_party/kernel/+/b7e444d5698fa25b271471cba61e2dedf61a44e0 commit b7e444d5698fa25b271471cba61e2dedf61a44e0 Author: Puthikorn Voravootivat <puthik@chromium.org> Date: Sat Aug 19 01:53:25 2017 CHROMIUM: Don't trust DPCD for panel earlier than eDP 1.4 Swanky panel has DPCD 702h = 0xf6 which means panel only support AUX to adjust brightness not the PWM pin. But actually it is opposite of what it is claiming in DPCD register. This CL make driver to always use PWM pin to adjust brightness for panel eariler than eDP 1.4 even when the panel DPCD said that PWM pin is not supported. BUG= chromium:755752 TEST=Swanky brightness settings work again. Change-Id: I305b72fc9c7f19581cc350bd3475a18cacef1da8 Signed-off-by: Puthikorn Voravootivat <puthik@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/622047 Reviewed-by: Todd Broch <tbroch@chromium.org> [modify] https://crrev.com/b7e444d5698fa25b271471cba61e2dedf61a44e0/drivers/gpu/drm/i915/intel_dp_aux_backlight.c
,
Aug 19 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromiumos/third_party/kernel/+/117a2358b43e1084a47e8ecc2079cebebd951499 commit 117a2358b43e1084a47e8ecc2079cebebd951499 Author: Puthikorn Voravootivat <puthik@chromium.org> Date: Sat Aug 19 02:11:56 2017 CHROMIUM: Don't trust DPCD for panel earlier than eDP 1.4 Swanky panel has DPCD 702h = 0xf6 which means panel only support AUX to adjust brightness not the PWM pin. But actually it is opposite of what it is claiming in DPCD register. This CL make driver to always use PWM pin to adjust brightness for panel eariler than eDP 1.4 even when the panel DPCD said that PWM pin is not supported. BUG= chromium:755752 TEST=Swanky brightness settings work again. Change-Id: I305b72fc9c7f19581cc350bd3475a18cacef1da8 Signed-off-by: Puthikorn Voravootivat <puthik@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/622047 Reviewed-by: Todd Broch <tbroch@chromium.org> (cherry picked from commit b7e444d5698fa25b271471cba61e2dedf61a44e0) Reviewed-on: https://chromium-review.googlesource.com/621276 [modify] https://crrev.com/117a2358b43e1084a47e8ecc2079cebebd951499/drivers/gpu/drm/i915/intel_dp_aux_backlight.c
,
Aug 19 2017
,
Aug 24 2017
Verified in Chrome OS 9765.38.0, 61.0.3163.61.
,
Aug 29 2017
,
Aug 29 2017
Issue 759939 has been merged into this issue.
,
Aug 30 2017
Verified - this problem also occurs on my Toshiba CB 2 Platform 9765.31.0 (Official Build) beta-channel swanky Firmware Google_Swanky.5216.238.5
,
Aug 30 2017
This is fixed in R61-9765.33.0 / R62-9857.0.0
,
Oct 9 2017
Issue 758637 has been merged into this issue. |
||||||||||||||
►
Sign in to add a comment |
||||||||||||||
Comment 1 by sontis@chromium.org
, Aug 15 2017