1447076 - Intermittent GECKO(4006) | thread 'WRWorker#1' panicked at 'Moz2D replay problem', gfx/webrender_bindings/src/moz2d_renderer.rs:187:21

Status: RESOLVED FIXED

(Core :: Graphics: WebRender, defect, P5)

Description

[Treeherder Bug Filer]

Filed by: kgupta [at] mozilla.com

https://treeherder.mozilla.org/logviewer.html#?job_id=168902686&repo=try

https://queue.taskcluster.net/v1/task/c9DW4Y2xSVGJqUVgj6tBzA/runs/0/artifacts/public/logs/live_backing.log

Comment 1

[Kartikaya Gupta (email:[e-mail])]

This "Moz2D replay problem" error is showing up in bunch of intermittent failures and real-world crashes (see dep bugs). Do we need more logging to understand why it's happening?

Comment 2

[Kartikaya Gupta (email:[e-mail])]

Oh, I guess this is just the non-debug variant of bug 1436861. Instead of asserting it returns false and then panics.

Comment 3

[Jeff Muizelaar [:jrmuizel]]

Attachment: Bug 1447076. Don't accept blob images that don't have any area.

Review commit: https://reviewboard.mozilla.org/r/231312/diff/#index_header
See other reviews: https://reviewboard.mozilla.org/r/231312/

Comment 4

[Kartikaya Gupta (email:[e-mail])]

Comment on attachment 8962479 [details]
Bug 1447076. Don't accept blob images that don't have any area.

https://reviewboard.mozilla.org/r/231312/#review236742

Comment 5

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/mozilla-inbound/rev/829f884d8d4e
Don't accept blob images that don't have any area. r=kats

Comment 6

[Andreea Pavel [:apavel]]

:kats hi. there already was https://bugzilla.mozilla.org/show_bug.cgi?id=1446371 for this. could you update progress there since that is what we used so far for classifying?

Comment 7

[Darkspirit]

I see your new assert:

(Jan Andre Ikenmeyer [:darkspirit] from bug 1446286 comment 4)
> 1. open https://www.qualcomm.com/products/server-processors
> 2. click on Search in the left navi
> 3. click on X in the left navi to close Search
> 4. bp-e5773a76-8f18-499e-ab96-4a9090180324

mozregression --repo mozilla-inbound --launch 829f884d8d4e --pref gfx.webrender.all:true startup.homepage_welcome_url:"https://www.qualcomm.com/products/server-processors" -B debug
> 0:44.05 INFO: [Child 29655, Main Thread] WARNING: '!IsUpdating()', file /builds/worker/workspace/build/src/dom/events/IMEContentObserver.cpp, line 2314
> 0:45.47 INFO: Assertion failure: aDescriptor.width > 0 && aDescriptor.height > 0, at /builds/worker/workspace/build/src/gfx/layers/wr/IpcResourceUpdateQueue.cpp:267

tab crash.

Comment 8

[Josh Matthews [:jdm]]

(In reply to Jan Andre Ikenmeyer [:darkspirit] from comment #7)
> I see your new assert:
> 
> (Jan Andre Ikenmeyer [:darkspirit] from bug 1446286 comment 4)
> > 1. open https://www.qualcomm.com/products/server-processors
> > 2. click on Search in the left navi
> > 3. click on X in the left navi to close Search
> > 4. bp-e5773a76-8f18-499e-ab96-4a9090180324
> 
> mozregression --repo mozilla-inbound --launch 829f884d8d4e --pref
> gfx.webrender.all:true
> startup.homepage_welcome_url:"https://www.qualcomm.com/products/server-
> processors" -B debug
> > 0:44.05 INFO: [Child 29655, Main Thread] WARNING: '!IsUpdating()', file /builds/worker/workspace/build/src/dom/events/IMEContentObserver.cpp, line 2314
> > 0:45.47 INFO: Assertion failure: aDescriptor.width > 0 && aDescriptor.height > 0, at /builds/worker/workspace/build/src/gfx/layers/wr/IpcResourceUpdateQueue.cpp:267
> 
> tab crash.

Do you have a link to the crash report?

Comment 9

[Darkspirit]

(In reply to Josh Matthews [:jdm] from comment #8)
> Do you have a link to the crash report?

No. I clicked on the blue button to restore the tab and to send a report. Checkboxes were ticked. There was no crash report with and without gpu process. No fallback, no failure. (Do inbound builds send crash reports at all?)

Comment 10

[Darkspirit]

Ha! I can't reproduce when gfx.webrender.blob.invalidation is true.
Something is just spamming this:
> 1:01.66 INFO: app unit -454761244 60

Comment 11

[Kartikaya Gupta (email:[e-mail])]

(In reply to Andreea Pavel [:apavel] from comment #6)
> :kats hi. there already was
> https://bugzilla.mozilla.org/show_bug.cgi?id=1446371 for this. could you
> update progress there since that is what we used so far for classifying?

There are a lot of bugs filed for this issue, see the dependency list. We are trying to figure out the root cause still, the assertions being added should narrow it down.

Comment 12

[Stefan Hindli [:stefan_hindli]]

https://hg.mozilla.org/mozilla-central/rev/829f884d8d4e

Comment 13

[Darkspirit]

(In reply to Jan Andre Ikenmeyer [:darkspirit] from comment #7)
> tab crash.

bp-bb706cb8-0d2e-44b3-8249-8a8490180330

Comment 27

[Intermittent Failures Robot]

1 failures in 2808 pushes (0.0 failures/push) were associated with this bug in the last 7 days.

Repository breakdown:
* autoland: 1

Platform breakdown:
* osx-10-10: 1

For more details, see:
https://treeherder.mozilla.org/intermittent-failures.html#/bugdetails?bug=1447076&startday=2018-11-26&endday=2018-12-02&tree=all