Bug 222505

Summary:

[GPUP] imported/w3c/web-platform-tests/webrtc/RTCRtpSender-replaceTrack.https.html times out with GPU Process enabled

Product:

WebKit

Reporter:

Jon Lee <jonlee>

Component:

WebRTC

Assignee:

youenn fablet <youennf>

Status:

RESOLVED FIXED

Severity:

Normal

CC:

eric.carlson, jer.noble, peng.liu6, simon.fraser, webkit-bug-importer, youennf

Priority:

Keywords:

InRadar

Version:

WebKit Nightly Build

Hardware:

Unspecified

OS:

Unspecified

Attachments:

Description	Flags
Patch	none
Patch	none

Jon Lee

Reported 2021-02-26 17:01:58 PST

Fix WebRTC layout test failures with GPU Process enabled latest from gpu-process/TestExpectations: webrtc/audio-video-element-playing.html [ Failure ] webrtc/captureCanvas-webrtc-software-h264-baseline.html [ Timeout ] webrtc/captureCanvas-webrtc-software-h264-high.html [ Timeout ] webrtc/h264-baseline.html [ Failure ] webrtc/h264-high.html [ Failure ] webrtc/video-rotation-no-cvo.html [ Timeout ] imported/w3c/web-platform-tests/webrtc/RTCRtpSender-replaceTrack.https.html [ Timeout ]

Attachments
Patch (1.54 KB, patch) 2021-03-01 02:16 PST, youenn fablet	no flags	Details Formatted Diff Diff
Patch (1.37 KB, patch) 2021-03-03 08:44 PST, youenn fablet	no flags	Details Formatted Diff Diff
Show Obsolete (1) View All Add attachment proposed patch, testcase, etc.

Radar WebKit Bug Importer

Comment 1 2021-02-26 17:02:30 PST

<rdar://problem/74814573>

youenn fablet

Comment 2 2021-03-01 02:12:56 PST

From results.webkit.org, they now all pass except imported/w3c/web-platform-tests/webrtc/RTCRtpSender-replaceTrack.https.html which does indeed timeout on the bots, but does not reproduce locally.

youenn fablet

Comment 3 2021-03-01 02:16:35 PST

Created attachment 421800 [details] Patch

youenn fablet

Comment 4 2021-03-01 02:19:53 PST

About imported/w3c/web-platform-tests/webrtc/RTCRtpSender-replaceTrack.https.html: 2021-02-28 20:55:01.056 com.apple.WebKit.GPU.Development[78884:107843434] PT: Got usage property: 1 2021-02-28 20:55:01.056 com.apple.WebKit.GPU.Development[78884:107843434] PT: Usage = 1, use AppleGVA 'avc1' codec (RTCVideoEncoderH264.mm:730): Compression session created with hw accl enabled (RTCVideoEncoderH264.mm:497): keyframe requested (RTCVideoEncoderH264.mm:905): Generated keyframe (bitrate_prober.cc:134): Probe delay too high (next_ms:1995566429, now_ms: 1995566434) (bitrate_prober.cc:134): Probe delay too high (next_ms:1995566433, now_ms: 1995566437) (bitrate_prober.cc:134): Probe delay too high (next_ms:1995566437, now_ms: 1995566443) (video_receive_stream2.cc:757): No decodable frame in 3000 ms, requesting keyframe. (rtcp_receiver.cc:315): absl::optional<TimeDelta> webrtc::RTCPReceiver::OnPeriodicRttUpdate(webrtc::Timestamp, bool): Timeout: No increase in RTCP RR extended highest sequence number. (rtcp_receiver.cc:313): absl::optional<TimeDelta> webrtc::RTCPReceiver::OnPeriodicRttUpdate(webrtc::Timestamp, bool): Timeout: No RTCP RR received. (RTCVideoEncoderH264.mm:497): keyframe requested (RTCVideoEncoderH264.mm:905): Generated keyframe (RTCVideoEncoderH264.mm:891): H264 encode dropped frame.

EWS

Comment 5 2021-03-01 02:44:51 PST

Committed r273645: <https://commits.webkit.org/r273645> All reviewed patches have been landed. Closing bug and clearing flags on attachment 421800 [details].

youenn fablet

Comment 6 2021-03-01 02:45:34 PST

Reopening for imported/w3c/web-platform-tests/webrtc/RTCRtpSender-replaceTrack.https.html

youenn fablet

Comment 7 2021-03-03 08:43:34 PST

No longer timing out on bots.

youenn fablet

Comment 8 2021-03-03 08:44:40 PST

Created attachment 422089 [details] Patch

EWS

Comment 9 2021-03-03 09:11:39 PST

Committed r273818: <https://commits.webkit.org/r273818> All reviewed patches have been landed. Closing bug and clearing flags on attachment 422089 [details].

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