1851785 - [wpt-sync] Sync PR 41839 - De-flake getFrameStats WPTs.

Status: RESOLVED FIXED

(Testing :: web-platform-tests, task, P4)

Description

[Web Platform Test Sync Bot [:wpt-sync] (Matrix: #interop:mozilla.org)]

Sync web-platform-tests PR 41839 into mozilla-central (this bug is closed when the sync is complete).

PR: https://github.com/web-platform-tests/wpt/pull/41839
Details from upstream follow.

Henrik Boström <[e-mail]> wrote:

De-flake getFrameStats WPTs.

The discardedFrames was incorrectly assuming that no frames would be
discarded initially, but the test only uses ideal so frames can be
discarded, and our implementation may sometimes discard the first frame
so this assertion is removed.

Also timeouts have been observed on the bots. The bot output[1] does
not say which of the tests are timing out but the most likely candidate
is the test that waits for a calculated timeout - which could be large
if the bot is under heavy load. This test is updated to wait for a
fixed number of milliseconds instead.

Lastly, for good measure, since all of these tests wait for media to
flow, the timeout should be "long".

[1] https://ci.chromium.org/ui/p/chromium/builders/ci/Mac12%20Tests/12050/test-results?sortby=&groupby=

Bug: chromium:1472978
Change-Id: If6ddb8f78409a2bec08c2abb8ab28baccdc75454
Reviewed-on: https://chromium-review.googlesource.com/4844860
WPT-Export-Revision: bf2e697c32699b6d71e3eb41564773f7e1bfb715

Comment 1

[Web Platform Test Sync Bot [:wpt-sync] (Matrix: #interop:mozilla.org)]

PR 41839 applied with additional changes from upstream: ee0ad06887da5ab6023092a1a23216b4dfbf324a

Comment 2

[Web Platform Test Sync Bot [:wpt-sync] (Matrix: #interop:mozilla.org)]

Pushed to try (stability) https://treeherder.mozilla.org/#/jobs?repo=try&revision=0eae3511fe4e1f36bf1ee6535e6b71a5164725a4

Comment 3

[Web Platform Test Sync Bot [:wpt-sync] (Matrix: #interop:mozilla.org)]

CI Results

Ran 0 Firefox configurations based on mozilla-central, and Firefox, Chrome, and Safari on GitHub CI

Total 1 tests and 10 subtests

Status Summary

Firefox

OK : 1
FAIL: 10

Chrome

OK : 1
FAIL: 10

Safari

OK : 1
FAIL: 10

Links

GitHub PR Head
GitHub PR Base

Details

New Tests That Don't Pass

• /mediacapture-extensions/MediaStreamTrack-getFrameStats.https.html [wpt.fyi]
  • totalFrames increases over time: FAIL (Chrome: FAIL, Safari: FAIL)
  • deliveredFrames increases, even without sinks: FAIL (Chrome: FAIL, Safari: FAIL)
  • discardedFrames increases when frameRate decimation is happening: FAIL (Chrome: FAIL, Safari: FAIL)
  • totalFrames is the sum of deliveredFrames and discardedFrames: FAIL (Chrome: FAIL, Safari: FAIL)
  • Stats are frozen while disabled: FAIL (Chrome: FAIL, Safari: FAIL)
  • Disabling and re-enabling does not reset the counters: FAIL (Chrome: FAIL, Safari: FAIL)
  • New stats baselines when a track is cloned from a disabled track: FAIL (Chrome: FAIL, Safari: FAIL)
  • New stats baselines when a track is cloned from an enabled track: FAIL (Chrome: FAIL, Safari: FAIL)
  • New stats baselines for the clone of a clone: FAIL (Chrome: FAIL, Safari: FAIL)
  • getFrameStats throws on audio tracks: FAIL (Chrome: FAIL, Safari: FAIL)

Comment 4

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/997baf7032df
[wpt PR 41839] - De-flake getFrameStats WPTs., a=testonly
https://hg.mozilla.org/integration/autoland/rev/b3f9accc2109
[wpt PR 41839] - Update wpt metadata, a=testonly

Comment 5

[Cosmin Sabou [:CosminS]]

https://hg.mozilla.org/mozilla-central/rev/997baf7032df
https://hg.mozilla.org/mozilla-central/rev/b3f9accc2109