Bug 219565

Summary:

REGRESSSION(270435) [GStreamer] Many media source tests failing and timing out

Product:

WebKit

Reporter:

Lauro Moura <lmoura>

Component:

Media

Assignee:

Lauro Moura <lmoura>

Status:

RESOLVED FIXED

Severity:

Normal

CC:

bugs-noreply, calvaris, cgarcia, dbates, dpino, ews-watchlist, gustavo, menard, peng.liu6, pnormand, vjaquez, webkit-bug-importer

Priority:

Keywords:

InRadar

Version:

WebKit Nightly Build

Hardware:

Unspecified

OS:

Unspecified

See Also:

https://bugs.webkit.org/show_bug.cgi?id=219332
https://bugs.webkit.org/show_bug.cgi?id=219822

Attachments:

Description	Flags
Patch	none
Updated patch after r270612	none
Patch with updated expectations	none
Patch	none

Lauro Moura

Reported 2020-12-04 20:47:58 PST

In media/media-source: Expected: 71 passes, 2 failures, 4 flakies After r270435: 59 passes, 15 timeouts media/media-source/media-source-abort-resets-parser.html media/media-source/media-source-canplaythrough-event.html media/media-source/media-source-error-crash.html media/media-source/media-source-init-segment-duration.html media/media-source/media-source-paint-to-canvas.html media/media-source/media-source-remove-crash.html media/media-source/media-source-resize.html media/media-source/media-source-samples-resolution-change.html media/media-source/media-source-seek-detach-crash.html media/media-source/media-source-seek-redundant-append.html media/media-source/media-source-seek-twice.html media/media-source/media-source-stalled-holds-sleep-assertion.html media/media-source/media-source-trackid-change.html media/media-source/media-source-unnecessary-seek-seeked.html media/media-source/media-source-webm-init-inside-segment.html Other tests are failing too. Collecting their data in the following comments.

Attachments
Patch (2.06 KB, patch) 2020-12-09 10:31 PST, Peng Liu	no flags	Details Formatted Diff Diff
Updated patch after r270612 (23.11 KB, patch) 2020-12-10 21:55 PST, Lauro Moura	no flags	Details Formatted Diff Diff
Patch with updated expectations (28.89 KB, patch) 2020-12-11 22:05 PST, Lauro Moura	no flags	Details Formatted Diff Diff
Patch (28.78 KB, patch) 2020-12-14 01:42 PST, Philippe Normand	no flags	Details Formatted Diff Diff
Show Obsolete (3) View All Add attachment proposed patch, testcase, etc.

Lauro Moura

Comment 1 2020-12-04 20:54:28 PST

Checking media/media-source/media-source-error-crash.html, it seems it's getting stuck somewhere waiting for events after an appendBuffer command. With GST_DEBUG=7 (trace, generated a 20MB text file), the last mentions of webkitmse are the following: 0:00:00.964280978 106 0x5636a23af610 DEBUG webkitmse MediaPlayerPrivateGStreamerMSE.cpp:712:trackDetected: track ID: V1, caps: video/x-h264, stream-format=(string)avc, alignment=(string)au, level=(string)3, profile=(string)main, codec_data=(buffer)014d401eff010013274d401ea9181407b600d4180418c2b5ef7c0401000428de09c8, width=(int)640, height=(int)480, pixel-aspect-ratio=(fraction)1/1 0:00:00.964318730 106 0x5636a23af610 DEBUG webkitmse PlaybackPipeline.cpp:178:attachTrack:<source> Configured track V1: appsrc=appsrc1, padId=0, mediaType=video/x-h264 0:00:00.982215860 106 0x5636a27692a0 DEBUG webkitmse AppendPipeline.cpp:684:connectDemuxerSrcPadToAppsinkFromStreamingThread: connecting to appsink 0:00:00.982220189 106 0x5636a27692a0 WARN webkitmse AppendPipeline.cpp:687:connectDemuxerSrcPadToAppsinkFromStreamingThread: Only one stream per SourceBuffer is allowed! Ignoring stream 2 by adding a black hole probe. 0:00:00.982249044 106 0x5636a27692a0 DEBUG webkitmse AppendPipeline.cpp:208:operator(): Posting no-more-pads task to main thread 0:00:00.982269273 106 0x5636a23af610 DEBUG webkitmse AppendPipeline.cpp:508:didReceiveInitializationSegment: Notifying SourceBuffer for track V1 0:00:00.982533950 106 0x5636a27692a0 DEBUG webkitmse AppendPipeline.cpp:897:appendPipelineAppsinkPadEventProbe: Handling event stream-start on append pipeline appsinkPad 0:00:00.984564192 106 0x5636a27692a0 DEBUG webkitmse AppendPipeline.cpp:897:appendPipelineAppsinkPadEventProbe: Handling event caps on append pipeline appsinkPad Without the revision, the test has the same messages as above, but followed by the normal flow (after the last common message, about handling event caps): 0:00:00.970354434 106 0x559cb7897e10 TRACE webkitmse AppendPipeline.cpp:560:consumeAppsinkAvailableSamples:<append-pipeline-video-mp4-0> batchedSampleCount = 0 0:00:00.970533065 106 0x559cb7897e10 DEBUG webkitmse MediaPlayerPrivateGStreamerMSE.cpp:712:trackDetected: track ID: V1, caps: video/x-h264, stream-format=(string)avc, alignment=(string)au, level=(string)3, profile=(string)main, codec_data=(buffer)014d401effe10013274d401ea9181407b600d4180418c2b5ef7c0401000428de09c8, width=(int)640, height=(int)480, pixel-aspect-ratio=(fraction)1/1, framerate=(fraction)0/1, chroma-format=(string)4:2:0, bit-depth-luma=(uint)8, bit-depth-chroma=(uint)8, colorimetry=(string)2:4:5:4, parsed=(boolean)true 0:00:00.970589302 106 0x559cb7897e10 DEBUG webkitmse PlaybackPipeline.cpp:220:reattachTrack: Re-attaching track 0:00:00.972209093 106 0x559cb7c52cc0 WARN webkitmse AppendPipeline.cpp:303:handleErrorSyncMessage:<append-pipeline-video-mp4-0> Demuxing error: error message: 0x559cb7ff4620, time 99:99:99.999999999, seq-num 165, element 'qtdemux0', GstMessageError, gerror=(GError)NULL, debug=(string)"../gst/isomp4/qtdemux.c\(7009\):\ gst_qtdemux_process_adapter\ \(\):\ /GstPipeline:append-pipeline-video-mp4-0/GstQTDemux:qtdemux0:\012atom\ ....\ has\ bogus\ size\ 18446744073709551615";

Lauro Moura

Comment 2 2020-12-04 20:56:49 PST

In imported/w3c/web-platform-tests/encrypted-media/: Expected: 34 passes, 3 failures, and 5 flakies Actual: 20 passes, 18 timeouts imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-clear-encrypted.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-encrypted-clear-sources.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-encrypted-clear.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-events.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-multikey-sequential-readyState.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-multikey-sequential.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-multikey.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-multisession.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-setMediaKeys-after-src.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-setMediaKeys-after-update.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-setMediaKeys-onencrypted.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-two-videos.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-waitingforkey.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-setmediakeys-again-after-playback.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-setmediakeys-again-after-resetting-src.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-setmediakeys-multiple-times-with-different-mediakeys.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-waiting-for-a-key.https.html

Lauro Moura

Comment 3 2020-12-04 20:57:50 PST

* Properly formatted list imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-clear-encrypted.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-encrypted-clear-sources.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-encrypted-clear.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-events.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-multikey-sequential-readyState.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-multikey-sequential.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-multikey.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-multisession.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-setMediaKeys-after-src.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-setMediaKeys-after-update.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-setMediaKeys-onencrypted.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-two-videos.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-waitingforkey.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-setmediakeys-again-after-playback.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-setmediakeys-again-after-resetting-src.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-setmediakeys-multiple-times-with-different-mediakeys.https.html imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-waiting-for-a-key.https.html

Lauro Moura

Comment 4 2020-12-04 21:36:46 PST

In imported/w3c/web-platform-tests/media-source/: Expected: 45 passes, 18 failures, and 8 flakies Actual: 15 passes and 49 timeouts imported/w3c/web-platform-tests/media-source/mediasource-activesourcebuffers.html imported/w3c/web-platform-tests/media-source/mediasource-append-buffer.html imported/w3c/web-platform-tests/media-source/mediasource-appendbuffer-quota-exceeded.html imported/w3c/web-platform-tests/media-source/mediasource-appendwindow.html imported/w3c/web-platform-tests/media-source/mediasource-avtracks.html imported/w3c/web-platform-tests/media-source/mediasource-buffered-seek.html imported/w3c/web-platform-tests/media-source/mediasource-buffered.html imported/w3c/web-platform-tests/media-source/mediasource-changetype-play-implicit.html imported/w3c/web-platform-tests/media-source/mediasource-changetype-play-negative.html imported/w3c/web-platform-tests/media-source/mediasource-changetype-play-without-codecs-parameter.html imported/w3c/web-platform-tests/media-source/mediasource-changetype-play.html imported/w3c/web-platform-tests/media-source/mediasource-changetype.html imported/w3c/web-platform-tests/media-source/mediasource-config-change-mp4-a-bitrate.html imported/w3c/web-platform-tests/media-source/mediasource-config-change-mp4-av-audio-bitrate.html imported/w3c/web-platform-tests/media-source/mediasource-config-change-mp4-av-framesize.html imported/w3c/web-platform-tests/media-source/mediasource-config-change-mp4-av-video-bitrate.html imported/w3c/web-platform-tests/media-source/mediasource-config-change-mp4-v-bitrate.html imported/w3c/web-platform-tests/media-source/mediasource-config-change-mp4-v-framerate.html imported/w3c/web-platform-tests/media-source/mediasource-config-change-mp4-v-framesize.html imported/w3c/web-platform-tests/media-source/mediasource-config-change-webm-a-bitrate.html imported/w3c/web-platform-tests/media-source/mediasource-config-change-webm-av-audio-bitrate.html imported/w3c/web-platform-tests/media-source/mediasource-config-change-webm-av-framesize.html imported/w3c/web-platform-tests/media-source/mediasource-config-change-webm-av-video-bitrate.html imported/w3c/web-platform-tests/media-source/mediasource-config-change-webm-v-bitrate.html imported/w3c/web-platform-tests/media-source/mediasource-config-change-webm-v-framerate.html imported/w3c/web-platform-tests/media-source/mediasource-config-change-webm-v-framesize.html imported/w3c/web-platform-tests/media-source/mediasource-correct-frames-after-reappend.html imported/w3c/web-platform-tests/media-source/mediasource-correct-frames.html imported/w3c/web-platform-tests/media-source/mediasource-detach.html imported/w3c/web-platform-tests/media-source/mediasource-duration-boundaryconditions.html imported/w3c/web-platform-tests/media-source/mediasource-duration.html imported/w3c/web-platform-tests/media-source/mediasource-endofstream.html imported/w3c/web-platform-tests/media-source/mediasource-errors.html imported/w3c/web-platform-tests/media-source/mediasource-getvideoplaybackquality.html imported/w3c/web-platform-tests/media-source/mediasource-h264-play-starved.html imported/w3c/web-platform-tests/media-source/mediasource-liveseekable.html imported/w3c/web-platform-tests/media-source/mediasource-play-then-seek-back.html imported/w3c/web-platform-tests/media-source/mediasource-play.html imported/w3c/web-platform-tests/media-source/mediasource-redundant-seek.html imported/w3c/web-platform-tests/media-source/mediasource-remove.html imported/w3c/web-platform-tests/media-source/mediasource-removesourcebuffer.html imported/w3c/web-platform-tests/media-source/mediasource-replay.html imported/w3c/web-platform-tests/media-source/mediasource-seek-beyond-duration.html imported/w3c/web-platform-tests/media-source/mediasource-seek-during-pending-seek.html imported/w3c/web-platform-tests/media-source/mediasource-seekable.html imported/w3c/web-platform-tests/media-source/mediasource-sequencemode-append-buffer.html imported/w3c/web-platform-tests/media-source/mediasource-sourcebuffer-mode.html imported/w3c/web-platform-tests/media-source/mediasource-timestamp-offset.html imported/w3c/web-platform-tests/media-source/mediasource-video-is-visible.html

Lauro Moura

Comment 5 2020-12-04 21:50:04 PST

Tentative gardening in r270469

Diego Pino

Comment 6 2020-12-06 23:36:16 PST

(In reply to Lauro Moura from comment #5) > Tentative gardening in r270469 Also r270471 and r270495.

Philippe Normand

Comment 7 2020-12-09 08:36:35 PST

Well, clang is helpful about this... ../../Source/WebCore/platform/graphics/gstreamer/mse/SourceBufferPrivateGStreamer.cpp:185:1: warning: all paths through this function will call itself [-Winfinite-recursion] { ^ ../../Source/WebCore/platform/graphics/gstreamer/mse/SourceBufferPrivateGStreamer.cpp:190:1: warning: all paths through this function will call itself [-Winfinite-recursion] { ^ 2 warnings generated. I suspect this is the culprit.

Philippe Normand

Comment 8 2020-12-09 08:40:34 PST

Just remove the recursing calls and leave implementation of these methods empty. rs=me ;)

Philippe Normand

Comment 9 2020-12-09 09:05:19 PST

(In reply to Philippe Normand from comment #8) > Just remove the recursing calls and leave implementation of these methods > empty. rs=me ;) Ah, I mean, chain to parent class method implementation

Peng Liu

Comment 10 2020-12-09 10:13:58 PST

(In reply to Philippe Normand from comment #7) > Well, clang is helpful about this... > > ../../Source/WebCore/platform/graphics/gstreamer/mse/ > SourceBufferPrivateGStreamer.cpp:185:1: warning: all paths through this > function will call itself [-Winfinite-recursion] > { > ^ > ../../Source/WebCore/platform/graphics/gstreamer/mse/ > SourceBufferPrivateGStreamer.cpp:190:1: warning: all paths through this > function will call itself [-Winfinite-recursion] > { > ^ > 2 warnings generated. > > > I suspect this is the culprit. Oops! Sorry for the mistake. :-(

Peng Liu

Comment 11 2020-12-09 10:31:44 PST

Created attachment 415773 [details] Patch

Lauro Moura

Comment 12 2020-12-10 21:55:50 PST

Created attachment 415967 [details] Updated patch after r270612 Still some failures related to r270612, though.

Lauro Moura

Comment 13 2020-12-10 22:02:23 PST

After this, there are still a few failures, likely related to timeouts after r270612. For example, with media/media-source/media-source-seek-twice.html, the test times out while waiting for the first seek event. GST webkitmse trace output: 0:00:01.224777512 106 0x55905b9e5610 INFO webkitmse MediaPlayerPrivateGStreamerMSE.cpp:167:seek:<media-player-0> [Seek] seek attempt to {-1/1 = nan} secs 0:00:01.224833488 106 0x55905b9e5610 DEBUG webkitmse MediaPlayerPrivateGStreamerMSE.cpp:186:seek:<media-player-0> Seeking from {0/1 = 0.000000} to {-1/1 = nan} seconds 0:00:01.224872773 106 0x55905b9e5610 DEBUG webkitmse MediaPlayerPrivateGStreamerMSE.cpp:261:doSeek:<media-player-0> [Seek] cannot seek, current state change is FAILURE 0:00:01.224882762 106 0x55905b9e5610 WARN webkitmse MediaPlayerPrivateGStreamerMSE.cpp:193:seek:<media-player-0> Seeking to {-1/1 = nan} failed 0:00:15.715545364 106 0x55905b9e5610 DEBUG webkitmse PlaybackPipeline.cpp:150:removeSourceBuffer:<source> Element removed from MediaSource 0:00:15.716031378 106 0x55905b9e5610 TRACE webkitmse MediaPlayerPrivateGStreamerMSE.cpp:115:~MediaPlayerPrivateGStreamerMSE: destroying the player (0x7f9cfccc9700) 0:00:15.716046066 106 0x55905b9e5610 DEBUG webkitmse PlaybackPipeline.cpp:95:setWebKitMediaSrc: webKitMediaSrc=(nil) Looking into MediaSource.cpp, in comparison to the behavior before r270612, one difference I could find was the buffered() call now returning an empty range list, causing hasCurrentTime to return false and the ReadyState always being kept in the HaveMetadata state (in MediaSource::monitorSourceBuffers()).

Lauro Moura

Comment 14 2020-12-10 22:20:27 PST

This investigation was done on top of r270630. Tomorrow I'll check it against r270639 (seeking support).

Lauro Moura

Comment 15 2020-12-11 05:50:17 PST

(In reply to Lauro Moura from comment #14) > This investigation was done on top of r270630. Tomorrow I'll check it > against r270639 (seeking support). Indeed the issue is still present as of this patch on top of r270667.

Peng Liu

Comment 16 2020-12-11 09:46:48 PST

(In reply to Lauro Moura from comment #15) > (In reply to Lauro Moura from comment #14) > > This investigation was done on top of r270630. Tomorrow I'll check it > > against r270639 (seeking support). > > > Indeed the issue is still present as of this patch on top of r270667. I guess we can land the current patch first and file a new bug to track the seeking issue. What do you think?

Lauro Moura

Comment 17 2020-12-11 22:05:18 PST

Created attachment 416090 [details] Patch with updated expectations Updated the expectations and opened bug219822 for the new failures.

EWS

Comment 18 2020-12-13 02:26:42 PST

Tools/Scripts/svn-apply failed to apply attachment 416090 [details] to trunk. Please resolve the conflicts and upload a new patch.

Philippe Normand

Comment 19 2020-12-14 01:42:33 PST

Created attachment 416141 [details] Patch

EWS

Comment 20 2020-12-14 02:35:32 PST

Committed r270765: <https://trac.webkit.org/changeset/270765> All reviewed patches have been landed. Closing bug and clearing flags on attachment 416141 [details].

Radar WebKit Bug Importer

Comment 21 2020-12-14 02:36:20 PST

<rdar://problem/72290021>

Note You need to log in before you can comment on or make changes to this bug.