Bug 186596 - [GTK] MediaStream related crashes and issues on tests on platform GTK since r232589
Summary: [GTK] MediaStream related crashes and issues on tests on platform GTK since r...
Status: NEW
Alias: None
Product: WebKit
Classification: Unclassified
Component: WebKitGTK (show other bugs)
Version: WebKit Nightly Build
Hardware: Unspecified Unspecified
: P2 Normal
Assignee: Nobody
URL:
Keywords:
Depends on:
Blocks: 185787
  Show dependency treegraph
 
Reported: 2018-06-13 10:11 PDT by Carlos Alberto Lopez Perez
Modified: 2021-03-25 01:56 PDT (History)
4 users (show)

See Also:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Carlos Alberto Lopez Perez 2018-06-13 10:11:22 PDT
The following mediastream tests crash on platform GTK since r232589 <https://trac.webkit.org/changeset/232589>

  fast/mediastream/MediaStream-video-element-displays-buffer.html [ Crash ]
  fast/mediastream/MediaStream-video-element-track-stop.html [ Crash ]
  fast/mediastream/MediaStream-video-element-video-tracks-disabled-then-enabled.html [ Crash ]
  fast/mediastream/MediaStream-video-element.html [ Crash ]
  fast/mediastream/apply-constraints-audio.html [ Crash ]
  fast/mediastream/change-tracks-media-stream-being-played.html [ Crash ]
  fast/mediastream/media-stream-track-source-failure.html [ Crash ]


Crash logs https://build.webkit.org/results/GTK%20Linux%2064-bit%20Release%20(Tests)/r232791%20(6991)/results.html
Comment 1 Carlos Alberto Lopez Perez 2018-06-13 10:17:35 PDT
but fast/mediastream/media-stream-track-source-failure.html is flaky

sometimes crashes and sometimes fails with this diff: 

--- a/gtk-linux-64-release-tests/build/layout-test-results/fast/mediastream/media-stream-track-source-failure-expected.txt
+++ b/gtk-linux-64-release-tests/build/layout-test-results/fast/mediastream/media-stream-track-source-failure-actual.txt
@@ -1,6 +1,10 @@
+CONSOLE MESSAGE: line 2451: Unhandled Promise Rejection: Error: assert_equals: Video track should have no size expected 0 but got 1120
+CONSOLE MESSAGE: line 2451: Unhandled Promise Rejection: Error: assert_equals: Video track should have no size expected 0 but got 1120
 
 
+Harness Error (FAIL), message = assert_equals: Video track should have no size expected 0 but got 1120
+
 PASS Create stream 
-PASS End video track only 
-PASS End audio track only 
+FAIL End video track only promise_test: Unhandled rejection with value: "Device state did not change in 5 second"
+FAIL End audio track only promise_test: Unhandled rejection with value: "Device state did not change in 5 second"
Comment 2 Carlos Alberto Lopez Perez 2018-06-13 10:24:43 PDT
And another one:

 - The test imported/w3c/web-platform-tests/mediacapture-streams/MediaStream-default-feature-policy.https.html is timing out
Comment 3 Thibault Saunier 2018-06-13 13:21:53 PDT
(In reply to Carlos Alberto Lopez Perez from comment #2)
> And another one:
> 
>  - The test
> imported/w3c/web-platform-tests/mediacapture-streams/MediaStream-default-
> feature-policy.https.html is timing out

I think this one is unrelated to the commit (it was not activated in there).

(In reply to Carlos Alberto Lopez Perez from comment #0)
> The following mediastream tests crash on platform GTK since r232589


Remaining after r232796

+ fast/mediastream/change-tracks-media-stream-being-played.htm
+ fast/mediastream/MediaStream-video-element.html

Those seem racy to me (can't reproduce locally yet), they do not happen with Debug builds also it seems. I am investigating it anyway!

(In reply to Carlos Alberto Lopez Perez from comment #1)
> but fast/mediastream/media-stream-track-source-failure.html is flaky
> 
> sometimes crashes and sometimes fails with this diff: 
> 
> ---
> a/gtk-linux-64-release-tests/build/layout-test-results/fast/mediastream/
> media-stream-track-source-failure-expected.txt
> +++
> b/gtk-linux-64-release-tests/build/layout-test-results/fast/mediastream/
> media-stream-track-source-failure-actual.txt
> @@ -1,6 +1,10 @@
> +CONSOLE MESSAGE: line 2451: Unhandled Promise Rejection: Error:
> assert_equals: Video track should have no size expected 0 but got 1120
> +CONSOLE MESSAGE: line 2451: Unhandled Promise Rejection: Error:
> assert_equals: Video track should have no size expected 0 but got 1120
>  
>  
> +Harness Error (FAIL), message = assert_equals: Video track should have no
> size expected 0 but got 1120
> +
>  PASS Create stream 
> -PASS End video track only 
> -PASS End audio track only 
> +FAIL End video track only promise_test: Unhandled rejection with value:
> "Device state did not change in 5 second"
> +FAIL End audio track only promise_test: Unhandled rejection with value:
> "Device state did not change in 5 second"

Will investigate.
Comment 4 Thibault Saunier 2018-06-14 11:50:17 PDT
All looks green with latest build:


https://build.webkit.org/results/GTK%20Linux%2064-bit%20Release%20(Tests)/r232842%20(7008)/results.html

I have not been able to reproduce issues that happened in r232796 (namely  fast/mediastream/change-tracks-media-stream-being-played.html) and will follow test results see if it flakes again.
Comment 5 Xabier Rodríguez Calvar 2021-03-25 01:56:09 PDT
fast/mediastream/apply-constraints-audio.html seems to be passing.
fast/mediastream/change-tracks-media-stream-being-played.html times out flakily.
fast/mediastream/media-stream-track-source-failure.html times out flakily.

The rest seem to pass to me.

Applies to bots and locally.
Comment 6 Xabier Rodríguez Calvar 2021-03-25 01:56:41 PDT
What I mean is that I see no crash.