Chrome crashes if I suspend and resume with stylus touching the screen
Reported by
rajatxj...@gmail.com,
Feb 21 2018
|
|||||||||||||||
Issue descriptionUserAgent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/64.0.3282.140 Safari/537.36 Steps to reproduce the problem: 1. Login, and open any random website on chrome 2. USe stylus to open a site, and then have the stylus in contact with the screen. 3. Go to VT2, and "powerd_dbus_suspend" (stylus should stay in contact with screen). 4. Resume and check the chrome crash in /var/spool/crash, also on the screen, the websites are closed and a new chrome session is asking whether you'd like to resume those websites. What is the expected behavior? chrome should not crash What went wrong? Chrome crashed Crashed report ID: No, but I have crash logs (core files) available How much crashed? Whole browser Is it a problem with a plugin? N/A Did this work before? N/A Chrome version: 65.0.3325.85 Channel: dev OS Version: 10323.38.0 Flash Version: I couldn't attach log files and crash core files because of the attachment size restrictions, please refer to https://b.corp.google.com/issues/73419154#comment3 for logs and attachments.
,
Feb 21 2018
Forgot to mention, I could reproduce this on 2 platforms that I tried - Soraka and Eve (could be reproducible with others).
,
Feb 21 2018
I also tried this on older releases (10032.0.0 and 10323.0.0) and the latest release 10421.0.0 and the issue exists.
,
Feb 21 2018
,
Feb 21 2018
Issue 814067 has been merged into this issue.
,
Feb 21 2018
,
Feb 21 2018
,
Feb 21 2018
+Tom, is there anyone from the ink/stylus team who should take a look too? This also affects Eve
,
Feb 21 2018
I was able to repro and got: https://crash.corp.google.com/browse?q=reportid=%271bbdc8de9191de74%27 Which looks like a service crash of some sort but doesn't give a lot of info. adlr@ is your team still the right folks to look at low-level issues with stylus?
,
Feb 21 2018
@Abodenha, if you try it out in a dev mode, you can get chrome core dumps. I have also attached 3 chrome core dumps at https://b.corp.google.com/issues/73419154#comment3 Do they help? It doesn't look like a low level crash IMHO, but we can only find out once some one looks at the core dumps.
,
Feb 21 2018
Looks like adlr@ hasn't been checking bugs. Definitely agree that core dumps are worth looking at. I'm not the person to do that. Let me see if I can find someone who can.
,
Feb 21 2018
xdai@ can someone on your team take a look at this? Maybe sammiequon@?
,
Feb 21 2018
Sammie, could you try to get the core dump to see if it's a UI issue?
,
Feb 27 2018
Sammie, did you get a chance to try this out?
,
Mar 13 2018
Any updates?
,
Mar 14 2018
Why is this restricted?
,
Mar 14 2018
This happens because we re-enter SetEnabled() infinitely when disabling the pen device if the pen is near the screen.
,
Mar 14 2018
There's a 2nd case via Reinitialize() -> ReleaseButtons() -> ReportEvents() -> Reinitialize(). I have a fix at https://chromium-review.googlesource.com/c/chromium/src/+/961386
,
Mar 14 2018
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/c642ba4572937977b6338393942f6781688ac02c commit c642ba4572937977b6338393942f6781688ac02c Author: Michael Spang <spang@chromium.org> Date: Wed Mar 14 20:34:14 2018 ozone: evdev: Fix pen-related infinite recursion If the pen device is disabled while the pen is near the screen it leads to infinite recursion. This is because we make a call to disable palm suppression, which revalidates device enablement. Post a task instead of disabling inline to fix this. The same thing happens if we lose events since Reinitialize() gained a call to ReleaseButtons() when pen support was added, which calls ReportEvents(), which can again call Reinitialize(). To fix that one, kill ReleaseButtons() as ReleaseTouches() should be sufficient. BUG= 814071 TEST=powerd_dbus_suspend with pen touching screen Change-Id: I8bb66feddc7961f1fa28d062133671ad1c46948c Reviewed-on: https://chromium-review.googlesource.com/961386 Reviewed-by: Kevin Schoedel <kpschoedel@chromium.org> Commit-Queue: Michael Spang <spang@chromium.org> Cr-Commit-Position: refs/heads/master@{#543185} [modify] https://crrev.com/c642ba4572937977b6338393942f6781688ac02c/ui/events/ozone/evdev/input_device_factory_evdev.cc [modify] https://crrev.com/c642ba4572937977b6338393942f6781688ac02c/ui/events/ozone/evdev/input_device_factory_evdev.h [modify] https://crrev.com/c642ba4572937977b6338393942f6781688ac02c/ui/events/ozone/evdev/touch_evdev_types.h [modify] https://crrev.com/c642ba4572937977b6338393942f6781688ac02c/ui/events/ozone/evdev/touch_event_converter_evdev.cc [modify] https://crrev.com/c642ba4572937977b6338393942f6781688ac02c/ui/events/ozone/evdev/touch_event_converter_evdev.h
,
Mar 19 2018
,
Mar 19 2018
This bug requires manual review: M66 has already been promoted to the beta branch, so this requires manual review Please contact the milestone owner if you have questions. Owners: cmasso@(Android), cmasso@(iOS), josafat@(ChromeOS), abdulsyed@(Desktop) For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
Mar 19 2018
,
Mar 19 2018
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/151db57a9189b8a5bc27c9a5b97fa21c5b32728d commit 151db57a9189b8a5bc27c9a5b97fa21c5b32728d Author: Michael Spang <spang@chromium.org> Date: Mon Mar 19 19:03:36 2018 ozone: evdev: Fix pen-related infinite recursion If the pen device is disabled while the pen is near the screen it leads to infinite recursion. This is because we make a call to disable palm suppression, which revalidates device enablement. Post a task instead of disabling inline to fix this. The same thing happens if we lose events since Reinitialize() gained a call to ReleaseButtons() when pen support was added, which calls ReportEvents(), which can again call Reinitialize(). To fix that one, kill ReleaseButtons() as ReleaseTouches() should be sufficient. BUG= 814071 TEST=powerd_dbus_suspend with pen touching screen Change-Id: I8bb66feddc7961f1fa28d062133671ad1c46948c Reviewed-on: https://chromium-review.googlesource.com/961386 Reviewed-by: Kevin Schoedel <kpschoedel@chromium.org> Commit-Queue: Michael Spang <spang@chromium.org> Cr-Original-Commit-Position: refs/heads/master@{#543185}(cherry picked from commit c642ba4572937977b6338393942f6781688ac02c) Reviewed-on: https://chromium-review.googlesource.com/969283 Reviewed-by: Michael Spang <spang@chromium.org> Cr-Commit-Position: refs/branch-heads/3359@{#319} Cr-Branched-From: 66afc5e5d10127546cc4b98b9117aff588b5e66b-refs/heads/master@{#540276} [modify] https://crrev.com/151db57a9189b8a5bc27c9a5b97fa21c5b32728d/ui/events/ozone/evdev/input_device_factory_evdev.cc [modify] https://crrev.com/151db57a9189b8a5bc27c9a5b97fa21c5b32728d/ui/events/ozone/evdev/input_device_factory_evdev.h [modify] https://crrev.com/151db57a9189b8a5bc27c9a5b97fa21c5b32728d/ui/events/ozone/evdev/touch_evdev_types.h [modify] https://crrev.com/151db57a9189b8a5bc27c9a5b97fa21c5b32728d/ui/events/ozone/evdev/touch_event_converter_evdev.cc [modify] https://crrev.com/151db57a9189b8a5bc27c9a5b97fa21c5b32728d/ui/events/ozone/evdev/touch_event_converter_evdev.h
,
Mar 19 2018
|
|||||||||||||||
►
Sign in to add a comment |
|||||||||||||||
Comment 1 by rajatja@google.com
, Feb 21 2018Components: IO>Stylus
Labels: -OS-Linux -Pri-2 Proj-Soraka OS-Chrome Pri-1
Owner: abodenha@google.com