This is usually indicative of a Chrome crash. You didn't manage to get a trace, did you?

I sync'd back to ec1790662dbd3cb93a057757fb5e6dd52d8cbb6f and can't repro either. Was this on a specific website?

Often when we get crashes like this entering VR, it's because of a DCHECK being hit in resizing code somewhere in blink, as we resize the page when we enter VR.

Assigning back to estark@ for more information, can't do much about this without a trace or repro.

I did just see a crash on google.com on that revision:
[FATAL:PaintLayer.cpp(2300)] Check failed: !result.InnerNode() || (result.GetHitTestRequest().ListBased() && result.ListBasedTestResult().size()).

This seems like a pretty likely culprit, so if this is the crash you saw Emily, I think we can close this.

I'm traveling this week and don't have the Pixel with me. I can ask a coworker to look if chrome://crashes is enabled and if it has a trace from last week. I'm pretty sure I saw it happen on multiple websites, but I'm not sure.

Re comment #3: if that was the culprit, is there another bug filed for it?

There wasn't a bug for comment #3's issue that we know of.

Comment #3 is speculation, I was unable to reproduce the issue you saw. I'm also unable to find a bug for the crash - it was likely transient ToT badness as I can't repro on a more recent ToT.

Don't think there's any actionable information here, and the issue is likely gone. Going to close this for now, please re-open if you encounter this again.

Here is the adb log when the screen went black. I don't see any Chrome related crashes, so let me know if this is useful.

Deleted: vr_adb.txt 74.8 KB

vr_adb.txt 74.8 KB View Download

Thanks meacer, can you please clarify when the screen went black for you? Did it stay black? Did this happen all of the time, or was it occasional?

If it was occasional, and happened when entering WebVR presentation, it's likely that you're running into  issue 738443 .

Looks like what I see is different than what estark saw.

In my case, the screen goes black for about a second after the "You can recenter the screen..." screen whereas in estark's case it's before this message. It happens consistently but is also a short delay and not a few seconds, so perhaps I mistakenly assumed it was the bug that we are hunting?

Going black for about a second is fully expected, as long as you correctly entered VR after that black period. Is that what happened?