Filed by: nbeleuzu [at] mozilla.com
Parsed log: https://treeherder.mozilla.org/logviewer?job_id=371153367&repo=autoland
Full log: https://firefox-ci-tc.services.mozilla.com/api/queue/v1/task/HDlNnsc7QBqpeOG6NB-HDQ/runs/0/artifacts/public/logs/live_backing.log

[task 2022-03-15T20:05:00.936Z] 20:05:00     INFO - PID 6336 | [Parent 4992, Main Thread] WARNING: 'aOwner->IsDiscarded()', file /builds/worker/workspace/obj-build/dist/include/mozilla/dom/SyncedContextInlines.h:95
[task 2022-03-15T20:05:00.963Z] 20:05:00     INFO - PID 6336 | JavaScript error: resource://gre/actors/BrowserElementParent.jsm, line 24: TypeError: can't access property "ownerGlobal", browser is null
[task 2022-03-15T20:05:01.442Z] 20:05:01     INFO - PID 6336 | [Parent 4992, Main Thread] WARNING: 'aOwner->IsDiscarded()', file /builds/worker/workspace/obj-build/dist/include/mozilla/dom/SyncedContextInlines.h:95
[task 2022-03-15T20:05:01.526Z] 20:05:01     INFO - PID 6336 | [Child 1972, Main Thread] ###!!! ASSERTION: we should have saved a frame property: 'Error', file /builds/worker/checkouts/gecko/layout/painting/nsDisplayList.cpp:4026
[task 2022-03-15T20:05:01.530Z] 20:05:01     INFO - PID 6336 | [Child 1972, Main Thread] ###!!! ASSERTION: we should have saved a frame property: 'Error', file /builds/worker/checkouts/gecko/layout/painting/nsDisplayList.cpp:4026
[task 2022-03-15T20:05:01.574Z] 20:05:01     INFO - Got content assert count 2
[task 2022-03-15T20:05:01.579Z] 20:05:01     INFO - 
[task 2022-03-15T20:05:01.579Z] 20:05:01     INFO - TEST-KNOWN-INTERMITTENT-FAIL | /html/cross-origin-opener-policy/resource-popup.https.html | /common/dummy.xml - parent COOP: ""; child COOP: "same-origin" - assert_equals: expected (object) null but got (string) "FAIL"
[task 2022-03-15T20:05:01.579Z] 20:05:01     INFO - validate@https://web-platform.test:8443/html/cross-origin-opener-policy/resource-popup.https.html:63:20
[task 2022-03-15T20:05:01.579Z] 20:05:01     INFO - coop_resource_test/</bc.onmessage<@https://web-platform.test:8443/html/cross-origin-opener-policy/resource-popup.https.html:26:57
[task 2022-03-15T20:05:01.579Z] 20:05:01     INFO - Test.prototype.step@https://web-platform.test:8443/resources/testharness.js:2534:25
[task 2022-03-15T20:05:01.579Z] 20:05:01     INFO - Test.prototype.step_func_done/<@https://web-platform.test:8443/resources/testharness.js:2609:32
[task 2022-03-15T20:05:01.589Z] 20:05:01     INFO - TEST-UNEXPECTED-FAIL | /html/cross-origin-opener-policy/resource-popup.https.html | assertion count 2 is more than expected 0 assertions
[task 2022-03-15T20:05:01.589Z] 20:05:01     INFO - ...........
[task 2022-03-15T20:05:01.589Z] 20:05:01     INFO - TEST-OK | /html/cross-origin-opener-policy/resource-popup.https.html | took 3443ms

R+B range: https://treeherder.mozilla.org/jobs?repo=autoland&searchStr=Android%2C7.0%2Cx86-64%2CLite%2CWebRender%2Cdebug%2CWeb%2Cplatform%2Ctests%2Cwithout%2Cfission%2Cenabled%2Ctest-android-em-7.0-x86_64-lite-qr%2Fdebug-geckoview-web-platform-tests-nofis-e10s%2Cwpt12&tochange=da82675f86d7db615b7e8881fe6ef9ede05f7bef&fromchange=86858987c84c49bf479f7a52d55832e6f6833bad points out to Bug 1759514 as a culprit.
Miko, could this be a regression from there?

Before this patch we might not update the ink overflow areas when
'outline-style' changes without changing outline-width, if
'outline-width' was zero.

This was triggered by TopLevelVideoDocument which had 'outline-width: 0'
which before bug 1759602 disabled the default outline.

The test reliably triggers the assertion (though I think we should try
to avoid the frame property in the common case the inner rect and frame
rects just match, which would remove the assertion...).

Now that we don't include overflow in outlines (bug 1739894) the
overwhelmingly common case is that innerRect == GetRectRelativeToSelf(),
except on a couple special cases.

Avoid storing the property in those cases. The test introduced in the
patch above still ensures we invalidate rendering correctly even without
the presence of the assert.

Depends on D141320

I took a look since bug 1759602 made this more frequent. I think this was really caused by bug 1759602.

Set release status flags based on info from the regressing bug 1759602

(In reply to Emilio Cobos Álvarez (:emilio) from comment #5)

I took a look since bug 1759602 made this more frequent. I think this was really caused by bug 1759602.

Thank you. I was going to bisect this today but this saved a lot of time. :)

https://hg.mozilla.org/mozilla-central/rev/aec742599e3d
https://hg.mozilla.org/mozilla-central/rev/eebddb9a5984