Closed
Bug 1399137
Opened 7 years ago
Closed 7 years ago
H264 with payload type 97 broken in mochitests
Categories
(Core :: WebRTC: Audio/Video, defect, P3)
Tracking
()
RESOLVED
FIXED
mozilla58
Tracking | Status | |
---|---|---|
firefox-esr52 | --- | unaffected |
firefox56 | --- | wontfix |
firefox57 | --- | wontfix |
firefox58 | --- | fixed |
People
(Reporter: pehrsons, Assigned: pehrsons)
References
Details
(Keywords: regression)
Attachments
(2 files, 1 obsolete file)
5.49 KB,
patch
|
Details | Diff | Splinter Review | |
2.98 KB,
patch
|
pehrsons
:
review+
|
Details | Diff | Splinter Review |
While writing a mochitest for video codecs in peer connection (bug 1395853) I hit a debug assert on using payload type 97 in a sender's offer sdp: > GECKO(8702) | # Fatal error in /home/pehrsons/Dev/mozilla-central/media/webrtc/trunk/webrtc/modules/rtp_rtcp/source/rtp_format_h264.cc, line 250 > GECKO(8702) | # last system error: 0 > GECKO(8702) | # Check failed: payload_size_left >= fragment->length (1176 vs. 10000) > GECKO(8702) | # Payload size left 1176, fragment length 10000, packetization mode SingleNalUnit To reproduce: Add the attached testcase patch to m-c and run > ./mach mochitest dom/media/tests/mochitest/test_peerConnection_videoCodecs.html Some test failures are expected (video flow check is not implemented), but the crash is not.
Comment 1•7 years ago
|
||
Odd. H264 Mode 0 (pt 97) works fine in https://mozilla.github.io/webrtc-landing/pc_test.html (Require H264, Prefer Mode 0).
Updated•7 years ago
|
Rank: 15
Priority: -- → P1
Comment 2•7 years ago
|
||
Attachment #8907215 -
Flags: review?(apehrson)
Updated•7 years ago
|
Assignee: nobody → rjesup
Status: NEW → ASSIGNED
Comment 3•7 years ago
|
||
silly webrtc.org code subtracts overhead from one side and not from the other.... how could that go wrong?
Attachment #8907288 -
Flags: review?(apehrson)
Updated•7 years ago
|
Attachment #8907215 -
Attachment is obsolete: true
Attachment #8907215 -
Flags: review?(apehrson)
Assignee | ||
Comment 4•7 years ago
|
||
Comment on attachment 8907288 [details] [diff] [review] make the fake openh264 encoder support mode 0 h264 Review of attachment 8907288 [details] [diff] [review]: ----------------------------------------------------------------- Thanks Randell!
Attachment #8907288 -
Flags: review?(apehrson) → review+
Assignee | ||
Updated•7 years ago
|
Rank: 15 → 25
Priority: P1 → P2
Summary: H264 with payload type 97 broken → H264 with payload type 97 broken in mochitests
Updated•7 years ago
|
Assignee: rjesup → apehrson
Comment 5•7 years ago
|
||
Mass change P2->P3 to align with new Mozilla triage process.
Priority: P2 → P3
Assignee | ||
Comment 7•7 years ago
|
||
Thanks. It's fine to let it ride on 58 only.
Pushed by pehrsons@gmail.com: https://hg.mozilla.org/integration/mozilla-inbound/rev/94d38bae703e make the fake openh264 encoder support mode 0 h264 r=pehrsons
Comment 9•7 years ago
|
||
bugherder |
https://hg.mozilla.org/mozilla-central/rev/94d38bae703e
Status: ASSIGNED → RESOLVED
Closed: 7 years ago
Resolution: --- → FIXED
Target Milestone: --- → mozilla58
Updated•7 years ago
|
status-firefox56:
--- → wontfix
status-firefox-esr52:
--- → unaffected
You need to log in
before you can comment on or make changes to this bug.
Description
•