Crash in i965_dri.so
Reported by
athos.ba...@bandlab.com,
Aug 3
|
||||||||
Issue descriptionUserAgent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.13; rv:61.0) Gecko/20100101 Firefox/61.0 Platform: 10575.58.0 (Official Build) stable-channel cyan Steps to reproduce the problem: We have an android app on google play, and we received hundreds of crash reports from users on chromebooks, whose stack trace reported by Google Developer Console is just one line: #00 pc 0000000000406567 /vendor/lib/dri/i965_dri.so We are not sure how our users got those crashes, but we were able to reproduce a crash on our development device (the logs for that are attached), but we are not sure whether it's related to the ones reported. It happened while switching the chromebook (with our app open) from tablet mode to the laptop mode. What is the expected behavior? What went wrong? A few users are affected by this issue but we are not even sure we can do something about it: can you confirm whether this is a bug in the OS level or, in alternative, what should we look at in our app? Thanks! Did this work before? N/A Chrome version: <Copy from: 'about:version'> Channel: stable OS Version: 67.0.4496.99 Flash Version:
,
Aug 3
Also, I am attaching crashlytics stats about the affected devices.
,
Aug 10
,
Aug 12
This might not be the right channel to report issues with Android on Chrome OS, as I normally see those in our internal tracker. However, if you could tell us what application that is and how you reproduce the crash, someone from the graphics team may be able to check it. marcheu@, chadversary@, what do you think? Note that this issue tracker is public, so if you think some of the information is sensitive, I'd recommend going with the internal path. Unfortunately I don't know how to do that, but let me add few more folks who may know.
,
Aug 13
The app is Bandlab, you can just download the latest version from Play Store: https://play.google.com/store/apps/details?id=com.bandlab.bandlab We have crash reports for versions of the app up to 7.7.2, which is currently the latest but will be replaced in a few days. The app has some native code, only for audio i/o and business logic, no native graphics rendering. I hope the breakpad microdump attached to this issue can somehow help, if you need we can also provide a debug build of our app.
,
Dec 3
Can we have any updates about this issue? We still have a lot of such crashes. We would happy to properly support Chromebooks, we believe our app is pretty useful on Chromebooks (this is musical editor) but looks that we have a lot of crashes on Chromebooks with ARC 7.1. Maybe someone could forward this bug report to a proper issue tracker, we would happy to provide any required debug information that could help investigate this issue
,
Dec 3
+posciak for ARC++ platform +Intel folks If this is dying in the Intel driver we may need to bring in Intel, I am not sure how much we can get out of that form of crash dump.
,
Dec 4
We would happy to provide any required information, including debug builds of the app. The problem that we don't have clear steps how to reproduce this, but it's possible even on debug builds
,
Dec 4
,
Dec 4
@bhthompson Can you add the Intel at Chromium entry to the CC list so we can have additional folks vectored in as well?
,
Dec 4
@athos and other Bandlab folks on this thread. We tried (recording an audio track and changing the chrome book mode from tablet to Laptop and vice versa) on 64-bit devices (Pixelbook and X2) 32-bit devices (Samsung Pro and Acer R11) and the older version of the app where issue is reported (ver 7.7.2) and latest version of the app (ver 7.15.2), but couldn’t reproduce the issue. We did run into an ANR with the Acer R11 on R72 Dev that is not seen on an R11 on R70 Stable Channel. But the ANR was not related to i965_dri.so. Bandlab folks: Above you mention you were able to duplicate a crash. Do you have any info on how to reproduce that crash? Thanks
,
Dec 4
,
Jan 7
,
Jan 8
Hi everyone - this bug has gone dark. Can the submitter please verify the issue is still present? If so, please provide repro instructions OR data on failure rate? We're more than happy to look into it.
,
Jan 11
This issue has an owner, a component and a priority, but is still listed as untriaged or unconfirmed. By definition, this bug is triaged. Changing status to "assigned". Please reach out to me if you disagree with how I've done this. |
||||||||
►
Sign in to add a comment |
||||||||
Comment 1 by athos.ba...@bandlab.com
, Aug 3