With the update of webrtc.org code to update 43 (bug 1198458), the first Android 4.3 opt test in webrtc in the first sent RTCP has an invalid timestamp of 2085978496000. Very likely this is due to a default initialization of the timestamp/last-rtp-time fields in the RTCP structure, and the RTCP is being sent before RTP (or some such). The value reported in the RTCP stats to JS is always the exact same value. In other to land bug 1198458, the test for a valid RTCP timestamp checks for the "uninitialized" value, and ignores it (throws an info()). This is to resolve the problem, or decide this is correct behavior and formally whitelist the specific default value for timestamp.

Comment on attachment 8688831 [details] [diff] [review] Whitelist specific bad RTCP timestamp value r=drno on IRC

Mass change P2->P3 to align with new Mozilla triage process.

Android no longer seems to generate the mysterious 2085978496000 timestamps, so we can run these tests on Android again. Depends on D9355

Pushed by cpeterson@mozilla.com: https://hg.mozilla.org/integration/mozilla-inbound/rev/69d881a78667 Restore RTCP timestamp tests on Android. r=jib

Android no longer seems to generate the mysterious 2085978496000 timestamps, so we can run these tests on Android again. My patch in comment 6 just re-enables the test on Android. It didn't fix the timestamp bug. We don't need to uplift this test fix to 64 Beta. The patch won't apply to cleanly 64 Beta because patches in bug 979649 made other changes to the same test in 65 Nightly.