Issue metadata
Sign in to add a comment
|
Rotation and convertibility on cave is broken
Reported by
polarime...@gmail.com,
Jan 10 2017
|
||||||||||||||||||||
Issue descriptionUserAgent: Mozilla/5.0 (X11; CrOS x86_64 9150.0.0) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/57.0.2970.0 Safari/537.36 Platform: 9150.0.0 (Official Build) dev-channel cave Steps to reproduce the problem: 1. Rotate screen on hinge to enter tablet mode; tablet mode does not activate 2. Rotate screen from landscape to portrait mode; screen does not rotate 3. What is the expected behavior? Tablet mode activation and screen rotation; keyboard deactivation in tablet mode What went wrong? Chrome OS no longer responds in any way to the hinge; the keyboard is not deactivated in tablet form factor, tablet mode is not activated, and the screen does not rotate regardless of rotation lock setting Did this work before? Yes 56.0.2924.53 Chrome version: 57.0.2970.0 Channel: dev OS Version: 9150.0.0 Flash Version: Shockwave Flash 24.0 r0
,
Jan 10 2017
polarimetric@ can you file a feedback report (alt+shift+i) when the problem occurs? Include "for abodenha" in the problem description and update here so I can find it. +a couple of people who might know more.
,
Jan 10 2017
abodenha@ I have submitted two feedback reports, one with the device in tent position where tablet mode was not activated, and another with the device in tablet position where tablet mode was activated but the device does not rotate in the normal position and rotating the screen back to laptop position does not disable tablet mode or reactivate the physical keyboard/touchpad.
,
Jan 11 2017
Looking at reports 51447898114 and 51447684754 It looks like at least one sensor is stuck, no data is acquired. #1, is it stated the user needs (refresh+power), while in report 51447898114 a suspend/resume is enough. Can we confirm we need a hard reset? Try to repro locally.
,
Jan 11 2017
gwendal@ : clarification: suspend/resume returns the device to laptop mode restoring usability of physical keyboard and touchpad, but does not fix the issue. When rolling back to beta or stable with powerwash, hard reset is required to fix the sensor permanently (after which everything functions as expected until dev 57.0.2970.0 is reinstalled).
,
Jan 11 2017
,
Jan 11 2017
Here's another report might be related: 51415983660 Description: The latest version of Chrome Dev (57.0.2970.0) breaks rotation and convertibility on Cave (Asus C302CA); keyboard is no longer disabled when screen is placed in tablet mode
,
Jan 11 2017
Could not repro yet. Same EC version: RO: cave_v1.9.282-23fe579 RW: cave_v1.9.333-13498c7 Same Coreboot,Release (although I use a test image). Check with servo: Reboot EC: path RO->RW does not break sensor init. After installing R57-9150, put the device into full tablet mode, does the display ever rotate? Does the acceleration field are updated when browsing http://www.albertosarullo.com/demos/accelerometer/? I am wondering if the sensor(s) never work in R57-9150, or if they stop working after a while.
,
Jan 11 2017
,
Jan 17 2017
One more feedback report: 51721260541.
,
Jan 17 2017
Our partners also tested below versions but could not reproduce yet. OS : 8872.73.0 / 9150.0.0 / 9176.0.0 RO/RW: 197/230 242/242
,
Jan 27 2017
Daisuke saw a problem with rotation on another machine (chrome-os-partner:62342). I wonder if invalid calibration value could be an explanation. A tar of /sys/firmware/vpd/ro directory on a affected machine would help.
,
Jan 27 2017
,
Jan 27 2017
Reproduced. It works at first. Suspend without moving works. But after suspending in tablet mode, clamshell mode stop working: input from keyboard and touchpad are ignored. Rotation is not working either. However, ectool confirms sensors are working, lid angle is sane. Evtest confirm EC is not gating keyboard events, touch pad is working. CHROMEOS_RELEASE_DESCRIPTION=9227.0.2017_01_27_1329 (Test Build - gwendal) developer-build cave Chromium 58.0.2994.0
,
Jan 28 2017
I just received another firmware update today and this behavior seems to be gone now.3 Version 58.0.2994.0 canary (64-bit) Platform 9228.0.0 (Official Build) canary-channel cave ARC Version 3682651 Firmware Google_Cave.7820.242.0
,
Feb 2 2017
Still broken for me on: Google Chrome 58.0.2997.0 (Official Build) canary (64-bit) Platform 9240.0.0 (Official Build) canary-channel cave I can only get it to go into "tablet" mode if I shake the device after tilting the display past 180/all the way back. And by shake, I don't mean a gentle shake; I mean somewhat forcefully shaking the device up and down/side to side. It will then switch to touchview mode.
,
Feb 2 2017
I retract what I previously wrote. After signing out and back in again, forcefully shaking the device no longer trigger touchview mode.
,
Feb 2 2017
Third Update: Sorry for all the posts. If you put the device in "tent" mode (may about 70 degrees), where the display is not all the way back as like a tablet, and then "flip" the device back and forth, touchview mode will turn on, although almost always in a portrait configuration rather than a landscape. If you put the device in full tablet mode, it's much harder to trigger touchview mode by flipping/shaking.
,
Feb 8 2017
One more report: 52882955243 Description: On this build, the keyboard does not "lock" or "disable" when in tablet mode. The screen does not rotate properly either. CHROMEOS_RELEASE_DESCRIPTION: 9202.18.0 vendor | coreboot version | Google_Cave.7820.242.0 ro bios version | Google_Cave.7820.197.0 However, I tried the same OS/bios, but couldn't reproduce it even after suspend/resume.
,
Feb 8 2017
Seems fixed in the past two Canary updates.
,
Feb 8 2017
Can still be issue prone. I had to hard reset my EC just now to get it working.
,
Feb 14 2017
Recently reported in CBC - https://goo.gl/3ngpfz #CBC-RS/TC-watchlist
,
Feb 16 2017
I tried the steps on #14 but I can't reproduce on my Cave 9280.0.0 dev channel. Here's the report ID: 53397839470 for reference. Re#14, Gwendal, do you mean the EC/sensors were all working as expected? Probably touchview/UI was confused?
,
Feb 16 2017
Another feedback report ID: 53333717931 on 9202.18.0.
,
Feb 16 2017
I can consistently reproduce this by doing a powerwash. I lose the "fix" and need to do a hard reset in order to restore proper function.
,
Feb 18 2017
Can confirm with #25. Currently on the latest Dev release. After powerwashing this morning, rotation is broken again.
,
Feb 18 2017
Thanks #25 and #26, we will try to reproduce early next week!
,
Feb 20 2017
I can reproduce the tablet mode broken issue after power washing the device (M58 9280.0.0). EC reset can recover it. We'll check if it's a regression. Thanks.
,
Feb 22 2017
Issue 626938 has been merged into this issue.
,
Mar 1 2017
After working as intended on dev M58 9280.0.0, this is broken again (in the exact same way as described in the original report) on dev M58 9313.1.0. EC reset does NOT recover touchview functionality. I've submitted feedback on with description "for abodenha regarding bug 679622, device hinge rotated, touchview not activated."
,
Mar 1 2017
On M58 9313.1.0, the workaround method described in comment #18 works: rotating the hinge (tablet mode) and then flipping the device over clockwise (so the keyboard is facing you) and then flipping it back activates touchview. Then putting the device back in laptop mode and repeating the same motion (flip the device clockwise away from you and then back) turns off touchview. Supports the idea that it is a sensor calibration issue.
,
Mar 1 2017
#31: it sounds like it might be calibration. Chrome provides the lid accelerometer data via the html device orientation api. Could you go to this demo site and let us know what the x,y,g axes of the device are? http://wellcaffeinated.net/demos/device-orientation While in normal laptop orientation the arrow that is rendered should point down.
,
Mar 1 2017
#32: I've attached two screenshots, one with the device in laptop mode set on a level surface and another while I was holding the device in tablet mode (hinge completely rotated) while standing up.
,
Mar 1 2017
Yeah that shows that the lid sensor is off. From the workaround involving flipping the device I wouldn't be surprised if the base accelerometer was inverted as well. I've attached an example from a Minnie whose sensors are working fine. You can see the gravity axis is off on #33
,
Mar 1 2017
Post feedback report on dev-channel 9313.1.0: https://feedback.corp.google.com/#/Report/54222079311 Please note that there is an internal bug (35774911) to track the tablet mode/rotation issue happens after performing a powerwash. Potential fixes are on the way for further validation. However, we are not sure if some failures may have other causes.
,
Mar 3 2017
Appears to be fixed (no EC reset necessary) on dev M58 9331.0.0.
,
Mar 20 2017
Thanks for the feedback! Latest beta should also include the fix. Let us know if any further issue. Thanks.
,
Aug 10 2017
I've been suffering from this issue intermittently for the past month . I am on Version 60.0.3112.80 (Official Build) beta (64-bit) Doing a refresh button + power reset clears the issue most times but sometimes have to repeat it. only just found this thread ..
,
Aug 10 2017
,
Aug 24 2017
Hi all. I was wondering if this bug is still active? It seems to be getting worse (more an more frequent) for me - Acer R11 - Version 60.0.3112.80 (Official Build) beta (64-bit) Google_Cyan.7287.57.125 Most times I have to resort to turning on the on-screen keyboard from the acessibility menu when I need it. Thanks
,
Sep 19 2017
gwendal@ there are several reports in this bug, and the other that is listed as a duplicate, of this issue still occurring on 60. Reports mention that the EC refresh fixes it, so it sounds like some failure in the EC level. Could you take another look?
,
Nov 27 2017
Hi there, I believe I have a related issue; Platform 9901.77.0 (Official Build) stable-channel cave Firmware Google_Cave.7820.329.0 When folding the keyboard back into a clamshell/laptop, the UI remains in tablet mode and the mouse and keyboard are deactivated. Tilting the laptop screen towards me then reactivates the keyboard and mouse. Suspending the machine also reactivates them. Occasionally the UI returns to laptop-mode, but the mouse is unresponsive. The keyboard is active, but without the backlight. Again, tilting the screen or suspending the laptop solves this. Powerwashing and hard resets have not solved the issue. If this is a calibration bug, I've attached a screenshot from the device-orientation website listed above. This was from the device being in laptop mode, on a level surface. Happy to take some more screenshots, or provide further feedback. Thanks
,
Jan 10 2018
,
Jun 4 2018
(Bulk Edit) Adding the new conops Chrome OS hotlist to all open issues with the "#CBC-RS/TC-watchlist" tag, our former tracking tag.
,
Aug 28
Hi, my chromebook has been working great until today when I noticed my chromebook no longer recognizes tablet mode. Even past 180 degrees, the chromebook physical keyboard is still active. The on screen keyboard never appears either. I tried the power and refresh method but I am still having the problem. I am also updated to the latest firmware. Is there anything more I can do? Is this something that is being worked on, or is it a hardware issue?
,
Sep 7
,
Sep 7
nicksteele3010@gmail.com do you happen to have a USB mouse or other device plugged into your device? If so can you try removing it, rebooting, and seeing if rotation works without that device present?
,
Sep 14
Physical keyboard shouldn't be active even if mouse is attached. +Sammie to make sure this is the case. |
|||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||
Comment 1 by polarime...@gmail.com
, Jan 10 2017