Closed Bug 1620096 Opened 5 years ago Closed 6 months ago

Intermittent gfx/tests/reftest/709477-1.html == gfx/tests/reftest/709477-1-ref.html | load failed: timed out waiting for reftest-wait to be removed

Categories

(Core :: Graphics: WebRender, defect)

Product:
Core
Component:
Graphics: WebRender
Type:
defect

Tracking

()

Status:
RESOLVED WORKSFORME

People

(Reporter: intermittent-bug-filer, Unassigned)

Details

(Keywords: intermittent-failure, Whiteboard: [stockwell disabled][stockwell unknown])

Attachments

(2 files)

Bug 1620096 - Disabled 709477-1.html on Linux and Android. r?#intermittent-reviewers
47 bytes, text/x-phabricator-request
Details | Review
Bug 1620096 - disable 709477-1.html on Debug_builds for frequent failures. r=#intermittent-reviewers
48 bytes, text/x-phabricator-request
Details | Review

Filed by: cbrindusan [at] mozilla.com
Parsed log: https://treeherder.mozilla.org/logviewer.html#?job_id=291705191&repo=autoland
Full log: https://firefox-ci-tc.services.mozilla.com/api/queue/v1/task/PF3ZbKmcRLWHSoTDKib6WA/runs/0/artifacts/public/logs/live_backing.log
Reftest URL: https://hg.mozilla.org/mozilla-central/raw-file/tip/layout/tools/reftest/reftest-analyzer.xhtml#logurl=https://firefox-ci-tc.services.mozilla.com/api/queue/v1/task/PF3ZbKmcRLWHSoTDKib6WA/runs/0/artifacts/public/logs/live_backing.log&only_show_unexpected=1

[task 2020-03-04T21:49:27.262Z] 21:49:27 INFO - REFTEST TEST-START | gfx/tests/reftest/709477-1.html == gfx/tests/reftest/709477-1-ref.html
[task 2020-03-04T21:49:27.262Z] 21:49:27 INFO - REFTEST TEST-LOAD | http://10.0.2.2:8854/tests/gfx/tests/reftest/709477-1.html | 797 / 7027 (11%)
[task 2020-03-04T21:54:33.053Z] 21:54:33 WARNING - REFTEST TEST-UNEXPECTED-FAIL | gfx/tests/reftest/709477-1.html == gfx/tests/reftest/709477-1-ref.html | load failed: timed out waiting for reftest-wait to be removed

Kris can you please assign someone to take a look?

Flags: needinfo?(ktaeleman)

@Lee: Looks like this test is related to Glyph clipping. There is some state machine logic in the test, maybe that can be removed?

Flags: needinfo?(ktaeleman) → needinfo?(lsalzman)

Karl appears to be test author...

Flags: needinfo?(lsalzman) → needinfo?(karlt)

The two macosx1014-64, one linux1804-64, and one windows10-64 attributions in comment 25 are all errors. So all failures in that week occur with Quantum Render.

START file:///builds/worker/workspace/build/tests/reftest/tests/gfx/tests/reftest/709477-1.html
[CONTENT] OnDocumentLoad triggering WaitForTestEnd
[CONTENT] WaitForTestEnd: Adding listeners
Initializing canvas snapshot
DoDrawWindow 0,0,800,1000
[CONTENT] MakeProgress
[CONTENT] AfterPaintListener in file:///builds/worker/workspace/build/tests/reftest/tests/gfx/tests/reftest/709477-1.html
[CONTENT] AfterPaintListener in file:///builds/worker/workspace/build/tests/reftest/tests/gfx/tests/reftest/709477-1.html
[CONTENT] MakeProgress: STATE_WAITING_TO_FIRE_INVALIDATE_EVENT
[CONTENT] MakeProgress: waiting for updateCanvasPending
[CONTENT] HandlePendingTasksAfterMakeProgress updating canvas
[CONTENT] Webrender enabled, sending update whole canvas for invalidation
Updating entire canvas for invalidation
DoDrawWindow 0,0,800,1000
[CONTENT] MakeProgress
[CONTENT] MakeProgress: STATE_WAITING_TO_FIRE_INVALIDATE_EVENT
[CONTENT] MakeProgress: dispatching MozReftestInvalidate
[CONTENT] MakeProgress
[CONTENT] AfterPaintListener in file:///builds/worker/workspace/build/tests/reftest/tests/gfx/tests/reftest/709477-1.html
[CONTENT] MakeProgress: STATE_WAITING_FOR_REFTEST_WAIT_REMOVAL
[CONTENT] MakeProgress: waiting for reftest-wait to be removed
[CONTENT] HandlePendingTasksAfterMakeProgress updating canvas
[CONTENT] Webrender enabled, sending update whole canvas for invalidation
Updating entire canvas for invalidation
DoDrawWindow 0,0,800,1000
[CONTENT] MakeProgress
[CONTENT] MakeProgress: STATE_WAITING_FOR_REFTEST_WAIT_REMOVAL
[CONTENT] MakeProgress: waiting for reftest-wait to be removed

"AfterPaintListener in" indicates that the "MozAfterPaint" event has fired only once after "dispatching MozReftestInvalidate". Is there a reason why scrollTo(0,1) would not generate a "MozAfterPaint" event with Quantum Render? If so, is there a way that the reftest harness and this test could know that the scroll had changed the rendering?

(In reply to Kris Taeleman (:ktaeleman) from comment #26)

There is some state machine logic in the test, maybe that can be removed?

I'm not clear what change to the test is being proposed, nor how the document modification and paint stages would be maintained with that change.

Flags: needinfo?(karlt)

This bug failed 49 times in the last 7 days. Fails on linux1804-64-qr and linux1804-64-asan-qr platforms on debug and opt build types.

Recent log:
https://treeherder.mozilla.org/logviewer.html#/jobs?job_id=316150562&repo=mozilla-central&lineNumber=11890

Hi Karl, are you still working on this?

Flags: needinfo?(karlt)

No. I suspect a difference in the dispatch of MozAfterPaint with Quantum Render.
I'm hoping someone familiar with those concepts can comment on their interaction, and perhaps answer the questions in comment 28.

Flagging Jim, as triage owner, in the hope he can identify an appropriate person.

Flags: needinfo?(karlt)
Summary: Intermittent gfx/tests/reftest/709477-1.html == gfx/tests/reftest/709477-1-ref.html | load failed: timed out waiting for reftest-wait to be removed → Intermittent QR gfx/tests/reftest/709477-1.html == gfx/tests/reftest/709477-1-ref.html | load failed: timed out waiting for reftest-wait to be removed
Flags: needinfo?(jmathies)
Component: Graphics → Graphics: WebRender
Blocks: gfx-triage
Flags: needinfo?(jmathies)
No longer blocks: gfx-triage
Severity: normal → N/A
Priority: -- → P5

In the last 7 days there were 39 failures of this bug. Occurs on android-hw-p2-8-0-android-aarch64-qr debug, linux1804-64-asan-qr opt, linux1804-64-qr debug.

Recent failure log: https://treeherder.mozilla.org/logviewer.html#/jobs?job_id=318278403&repo=autoland&lineNumber=12498

Jim, could you assign someone to look into this?

Flags: needinfo?(jmathies)
Blocks: gfx-triage
Flags: needinfo?(jmathies)

Feel free to disable for affected platforms.

No longer blocks: gfx-triage
Priority: -- → P5
Assignee: nobody → rmaries
Status: NEW → ASSIGNED
Pushed by smolnar@mozilla.com: https://hg.mozilla.org/integration/autoland/rev/73bf88ce09ce Disabled 709477-1.html on Linux and Android. r=kats DONTBUILD
Assignee: rmaries → nobody
Status: ASSIGNED → NEW
Keywords: leave-open
Whiteboard: [stockwell needswork:owner] → [stockwell disabled]
Assignee: nobody → rmaries
Status: NEW → ASSIGNED

There have been 30 total failures in the last 7 days.
Recent failure log.

Affected platforms are:

  • macosx1015-64-qr debug
  • windows10-64-ccov-qr
  • windows10-64-qr debug

Jim, could you please redirect this to someone who can take a look?

Flags: needinfo?(jmathies)
Assignee: rmaries → nobody
Status: ASSIGNED → NEW
Whiteboard: [stockwell disabled] → [stockwell disabled][stockwell needswork:owner]
Pushed by abutkovits@mozilla.com: https://hg.mozilla.org/integration/autoland/rev/fb9e855fae12 disable 709477-1.html on Debug_builds for frequent failures. r=intermittent-reviewers,ahal DONTBUILD

Maybe we should add this to the list of tests someone needs to look at.

Flags: needinfo?(jmathies) → needinfo?(aosmond)
Summary: Intermittent QR gfx/tests/reftest/709477-1.html == gfx/tests/reftest/709477-1-ref.html | load failed: timed out waiting for reftest-wait to be removed → Intermittent gfx/tests/reftest/709477-1.html == gfx/tests/reftest/709477-1-ref.html | load failed: timed out waiting for reftest-wait to be removed
Flags: needinfo?(aosmond)
Status: NEW → RESOLVED
Closed: 6 months ago
Resolution: --- → WORKSFORME
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: