[Webkit-unassigned] [Bug 266195] New: [GStreamer] media/media-source/media-managedmse-multipletracks-bufferedchange.html is a constant failure
bugzilla-daemon at webkit.org
bugzilla-daemon at webkit.org
Sun Dec 10 13:38:52 PST 2023
https://bugs.webkit.org/show_bug.cgi?id=266195
Bug ID: 266195
Summary: [GStreamer]
media/media-source/media-managedmse-multipletracks-buf
feredchange.html is a constant failure
Product: WebKit
Version: WebKit Nightly Build
Hardware: Unspecified
OS: Unspecified
Status: NEW
Severity: Normal
Priority: P2
Component: Media
Assignee: webkit-unassigned at lists.webkit.org
Reporter: jean-yves.avenard at apple.com
in test introduced in bug 266169
it is always failing:
```
--- /home/ews/worker/WPE-WK2-Tests-EWS/build/layout-test-results/media/media-source/media-managedmse-multipletracks-bufferedchange-expected.txt
+++ /home/ews/worker/WPE-WK2-Tests-EWS/build/layout-test-results/media/media-source/media-managedmse-multipletracks-bufferedchange-actual.txt
@@ -4,19 +4,19 @@
RUN(sourceBuffer = source.addSourceBuffer('video/webm; codecs="vp9,opus"'))
RUN(sourceBuffer.appendBuffer(buffer))
onbufferedchange called.
-e.addedRanges = [0, 9.999)
+e.addedRanges = [0, 3.416666),[5.333, 9.999666)
e.removedRanges = []
EVENT(bufferedchange)
EVENT(update)
-EXPECTED (sourceBuffer.buffered.length == '1') OK
-video.buffered = [0, 9.999)
-sourceBuffer.buffered = [0, 9.999)
+EXPECTED (sourceBuffer.buffered.length == '1'), OBSERVED '2' FAIL
+video.buffered = [0, 3.416666),[5.333, 9.999666)
+sourceBuffer.buffered = [0, 3.416666),[5.333, 9.999666)
RUN(source.endOfStream())
-video.buffered = [0, 10.0555)
-sourceBuffer.buffered = [0, 10.0555)
+video.buffered = [0, 3.416666),[5.333, 10.0555)
+sourceBuffer.buffered = [0, 3.416666),[5.333, 10.0555)
EVENT(sourceended)
onbufferedchange called.
-e.addedRanges = [9.999, 10.0555)
+e.addedRanges = [9.999666, 10.0555)
e.removedRanges = []
EVENT(bufferedchange)
END OF TEST
```
GStreamer doesn't appear to demux the whole content and is showing as having gap. The last vp9 frame finishes exactly at 9.999, not 9.999666
Inspection with ffprobe shows that there are no discontinuity in this file at 3.416666:
`ffprobe -show_entries packet=pts_time,duration_time,stream_index LayoutTests/media/media-source/content/test-fragmented.webm | less`
```
[PACKET]
stream_index=1
pts_time=3.395000
duration_time=0.020000
[/PACKET]
[PACKET]
stream_index=1
pts_time=3.415000
duration_time=0.020000
[/PACKET]
[PACKET]
stream_index=0
pts_time=3.417000
duration_time=0.041000
[/PACKET]
[PACKET]
stream_index=1
pts_time=3.435000
duration_time=0.020000
[/PACKET]
[PACKET]
stream_index=1
pts_time=3.455000
duration_time=0.020000
[/PACKET]
[PACKET]
stream_index=0
pts_time=3.458000
duration_time=0.041000
[/PACKET]
[PACKET]
stream_index=1
pts_time=3.475000
duration_time=0.020000
```
The timestamps are also weirdly rounded they are all in milliseconds.
output from mkvinfo -a:
```
| + Simple block: key, track number 2, 1 frame(s), timestamp 00:00:03.382000000
| + Frame with size 284
| + Simple block: key, track number 2, 1 frame(s), timestamp 00:00:03.402000000
| + Frame with size 286
| + Simple block: key, track number 2, 1 frame(s), timestamp 00:00:03.422000000
| + Frame with size 274
| + Simple block: track number 1, 1 frame(s), timestamp 00:00:03.417000000
| + Frame with size 608
| + Simple block: key, track number 2, 1 frame(s), timestamp 00:00:03.442000000
| + Frame with size 255
| + Simple block: key, track number 2, 1 frame(s), timestamp 00:00:03.462000000
| + Frame with size 260
| + Simple block: track number 1, 1 frame(s), timestamp 00:00:03.458000000
| + Frame with size 398
| + Simple block: key, track number 2, 1 frame(s), timestamp 00:00:03.482000000
| + Frame with size 474
| + Simple block: key, track number 2, 1 frame(s), timestamp 00:00:03.502000000
| + Frame with size 233
| + Simple block: track number 1, 1 frame(s), timestamp 00:00:03.500000000
| + Frame with size 458
```
This webm file is made of 2 clusters, the 2nd one starting at 00:00:04.982000000
--
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20231210/f76f465c/attachment.htm>
More information about the webkit-unassigned
mailing list