New issue
Advanced search Search tips

Issue 908618 link

Starred by 1 user

Issue metadata

Status: Assigned
Owner:
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Chrome
Pri: 2
Type: Bug



Sign in to add a comment

[nocturne pre-pvt] device will not power on after suspend from idle, and connected to peripherals

Project Member Reported by matthewjoseph@chromium.org, Nov 26

Issue description

Chrome Version: R72-11298.0.0

Nocturne A70 pre-pvt device is unable to power on after the device was left idle while connected to usb-c hub (CZHOON) with usb keyboard, mouse, and external display plugged in

What is the expected result?
Device should wake from suspend state

What happens instead?
Device appears to have lost power and will not power on.  holding the power button for > 10 seconds does not power the device on.  Also unable to enter recovery mode.

Additionally, it appears the device is not charging (the LED on the side is not illuminated) when connected to known-good usb-c charger.

 
Cc: tbroch@chromium.org kmshelton@chromium.org hungte@chromium.org
Components: OS>Firmware
+some more folks
Some more details:

- Recovered the device around 8AM (Device had ~10% charge)
- Performed testing on it until ~11AM 
- Device performed 1st idle-suspend
- Woke device successfully and continued testing
- Stepped away from device ~1:30PM
- Attempted to wake device ~2:30PM

Cc: dchan@chromium.org
Summary: [nocturne pre-pvt] device will not power on after suspend from idle, and connected to peripherals (was: [nocturne pre-pvt] device will not power on after suspend)
Cc: aaboagye@chromium.org puthik@chromium.org
Are you able to debug / collect logs via CCD?  Would be best to proceed there if device has had CCD setup.

If not, EC reset would likely fix things but also destroy any chance of debug in all likelihood.

+Opal & Aseda for thoughts/visibility
The device was not setup for CCD prior to this happening.  Additionally, the device was recovered and tested *without* the whiskers KB attached. 
The device has recovered on its own.  It was left on the desk no connected to power and still unable to power on as of early this morning.  Without any intervention on my part, the device powered on.

I was able to generate_logs after this occurred.  Logs here:

https://pantheon.corp.google.com/storage/browser/chromiumos-test-logs/bugfiles/cr/908618
Also, the battery is reporting 100% charge after it recovered itself. (Again, not plugged into power since the issue occurred yesterday afternoon)
The eventlog shows a couple kernel panics happened and then perhaps got stuck in the FSP while booting?

285 | 2018-11-26 11:06:42 | S0ix Enter
286 | 2018-11-26 11:58:58 | S0ix Exit
287 | 2018-11-26 11:58:58 | Wake Source | Power Button | 0
288 | 2018-11-26 11:58:58 | EC Event | Power Button
289 | 2018-11-26 12:24:34 | Kernel Event | Panic
290 | 2018-11-26 12:24:38 | System boot | 9
291 | 2018-11-26 12:24:38 | System Reset
292 | 2018-11-26 12:24:38 | Chrome OS Developer Mode
293 | 2018-11-26 12:34:58 | Kernel Event | Oops
294 | 2018-11-26 12:34:58 | Kernel Event | Panic
295 | 2018-11-26 12:35:03 | System boot | 10
296 | 2018-11-26 12:35:03 | System Reset
297 | 2018-11-26 12:35:03 | Chrome OS Developer Mode
298 | 2018-11-26 13:02:49 | System boot | 12
299 | 2018-11-26 13:02:49 | Last post code in previous boot | 0x93 | FSP-S Enter
300 | 2018-11-26 13:02:49 | Extra info from previous boot | CPU Cluster | 0x0000
301 | 2018-11-26 13:02:49 | ACPI Wake | Deep S5
302 | 2018-11-26 13:02:49 | Wake Source | Power Button | 0
303 | 2018-11-26 13:02:49 | Chrome OS Developer Mode
re c#8

The kernel panics were manually induced, as part of the logging / crash-reporting test cases.


Owner: aaboagye@chromium.org
Status: Assigned (was: Untriaged)
Do we think the failure is specific to this one nocturne or this one particular usb-c hub (CZHOON)?

Aseda can you follow-up with Matt? 
The device is back in this state.  Unable to power on.  LMK if someone wants to take a look at it.
I took a look at the device yesterday and it seemed totally dead. It was consuming ~2.5W though somewhere, but this device will need to be unpressed to be debugged further. I think it's the failure is specific to this device and has nothing to do with peripherals.

Sign in to add a comment