1903047 - Perma Late Beta REFTEST TEST-UNEXPECTED-FAIL | dom/media/test/reftest/uneven_frame_duration_video.html == dom/media/test/reftest/uneven_frame_duration_video-ref.html | when Gecko 129 switches from Early Beta to Late Beta on 2024-07-19

Status: VERIFIED FIXED

(Core :: Audio/Video: Playback, defect)

Description

[Sandor Molnar[:smolnar]]

Central as Late Beta simulation

How to run these simulations

Failure log

OAD | file:///D:/task_171862308063301/build/tests/reftest/tests/dom/media/test/reftest/uneven_frame_duration_video-ref.html | 8 / 12 (66%)
[task 2024-06-17T12:10:31.810Z] 12:10:31     INFO - REFTEST INFO | REFTEST fuzzy test (0, 0) <= (174, 767322) <= (84, 774213)
[task 2024-06-17T12:10:34.186Z] 12:10:34     INFO - REFTEST TEST-UNEXPECTED-FAIL | dom/media/test/reftest/uneven_frame_duration_video.html == dom/media/test/reftest/uneven_frame_duration_video-ref.html | image comparison, max difference: 174, number of differing pixels: 767322
[task 2024-06-17T12:10:34.189Z] 12:10:34     INFO - REFTEST   IMAGE 1 (TEST): data
INFO - REFTEST   IMAGE 2 (REFERENCE)
INFO - REFTEST TEST-END | dom/media/test/reftest/uneven_frame_duration_video.html == dom/media/test/reftest/uneven_frame_duration_video-ref.html
[task 2024-06-17T12:10:34.192Z] 12:10:34     INFO - REFTEST TEST-START | dom/media/test/reftest/frame_order_mp4.html == dom/media/test/reftest/frame_order_mp4-ref.html

Comment 1

[C.M.Chang[:chunmin]]

Hi Sandor, How do you find out if this failure is caused by patches in bug 1902351? I am not able to locate which patch causes the issue by mozregression since the test works expectedly in Nightly. In addition, the LATE BETA simulation on top of Bug 1902157's patches is green.

Comment 2

[Cristian Tuns]

I will include the backout of bug 1902351 in the today's simulation to see if that is the culprit.
(Edit) I tried to backout bug 1902351 but I can't do that because it has some dependencies in the other bugs in that push: https://hg.mozilla.org/integration/autoland/pushloghtml?changeset=916a5b8267fccc4b0de2f6ad2426aefdb0fcc1f1) (I can't confirm which specific bug is the culprit for this if I can't back it out)

Comment 3

[Jim Mathies [:jimm]]

bug 1902351 landed comment changes. No way that regressed anything in the product.

Comment 4

[Jim Mathies [:jimm]]

I think we'll wait for more current test data (closer to the merge).

Comment 5

[C.M.Chang[:chunmin]]

See bug 1903058 comment 5

Comment 6

[BugBot [:suhaib / :marco/ :calixte]]

:sotaro, since you are the author of the regressor, bug 1889275, could you take a look? Also, could you set the severity field?

For more information, please visit BugBot documentation.

Comment 7

[Sotaro Ikeda [:sotaro]]

I take the bug.

Comment 8

[Norisz Fay [:noriszfay]]

This failure might also be related

Comment 9

[Sebastian Hengst [:aryx] (needinfo me if it's about an intermittent or backout)]

Likely unrelated: The reftest failure in this is cause by the content slightly shifted horizontally while the log in comment 8 is a green frame vs. the video rendered.

Comment 10

[Donal Meehan [:dmeehan]]

:sotaro, the Fx129 soft code freeze starts on Thursday 2024-07-04, ahead of Fx129 merging to beta on Monday 2024-07-08.
Do you plan on landing a fix this week?

Comment 11

[Sotaro Ikeda [:sotaro]]

I am going to create a patch and asking for review today.

Comment 12

[Sotaro Ikeda [:sotaro]]

It seems necessary to wait ID3D11Query of before color conversion if the ID3D11Query exists in DXGITextureHostD3D11::GetAsSurfaceWithDevice() like Bug 1817617. Since the color conversion seems to use VideoProcessor internally.

Comment 13

[Sotaro Ikeda [:sotaro]]

Attachment: Bug 1903047 - Wait ID3D11Query of D3D11Texture2D copy complete before color conversion if ID3D11Query of DXGITextureHostD3D11 exists

Comment 14

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/d138ac8c7ec9
Wait ID3D11Query of D3D11Texture2D copy complete before color conversion if ID3D11Query of DXGITextureHostD3D11 exists r=gfx-reviewers,lsalzman

Comment 15

[pstanciu]

https://hg.mozilla.org/mozilla-central/rev/d138ac8c7ec9

Comment 16

[Norisz Fay [:noriszfay]]

Verified fixed in today's central as beta sim

Comment 17

[BugBot [:suhaib / :marco/ :calixte]]

Set release status flags based on info from the regressing bug 1889275