Rev3 WINNT 6.1 mozilla-inbound debug test mochitest-1 on 2012-11-25 15:20:21 PST for push 4b816b309e52 slave: talos-r3-w7-088 https://tbpl.mozilla.org/php/getParsedLog.php?id=17335446&tree=Mozilla-Inbound { ... ... ... [Parent 3480] ###!!! ASSERTION: Should have positive clock time.: 'clock_time >= mStartTime', file e:/builds/moz2_slave/m-in-w32-dbg/build/content/media/MediaDecoderStateMachine.cpp, line 2277 130379 INFO TEST-PASS | /tests/content/media/test/test_playback_rate.html | playbackRate should be initially 1.0 130380 INFO TEST-PASS | /tests/content/media/test/test_playback_rate.html | defaultPlaybackRate should be initially 1.0 130381 INFO TEST-PASS | /tests/content/media/test/test_playback_rate.html | PlaybackRate should be clamped to 0.25. 130382 INFO TEST-PASS | /tests/content/media/test/test_playback_rate.html | playbackRate should be reset to 1.0 on play() call xul!JSD_AttemptUCScriptInStackFrame+0x0000000000AC3EBA xul!JSD_AttemptUCScriptInStackFrame+0x0000000000AC4112 xul!JSD_AttemptUCScriptInStackFrame+0x000000000136FABF xul!JSD_AttemptUCScriptInStackFrame+0x000000000136FCF1 xul!JSD_AttemptUCScriptInStackFrame+0x000000000136A8F8 xul!JSD_AttemptUCScriptInStackFrame+0x0000000001317FF0 xul!JSD_AttemptUCScriptInStackFrame+0x0000000001369F00 nspr4!PRP_TryLock+0x000000000000062B MSVCR100D!beginthreadex+0x00000000000001D4 kernel32!BaseThreadInitThunk+0x0000000000000012 ntdll!RtlInitializeExceptionChain+0x0000000000000063 ntdll!RtlInitializeExceptionChain+0x0000000000000036 [Parent 3480] ###!!! ASSERTION: Should have positive clock time.: 'clock_time >= mStartTime', file e:/builds/moz2_slave/m-in-w32-dbg/build/content/media/MediaDecoderStateMachine.cpp, line 2277 TEST-UNEXPECTED-FAIL | /tests/content/media/test/test_playback_rate.html | application timed out after 330 seconds with no output }

Paul, are you looking into this? I get a horrendous amount of assertion failure spam when I run this test on Windows, I haven't tried other platforms. It looks like AudioClock::GetPosition() is returning a bogus value because AudioClock::mBasePosition is bogus.

OK, I looked into this because it was causing tests to fail for me with my WMF backend... The problem is that AudioClock::GetPosition() uses mOldOutRate to calculate the "position" variable [1], that gets stored in mPreviousPosition, which then later is stored in mBasePosition [2]. mOldOutRate is initialized by the constructor to 0, and it's set to mOutRate in one but not all code paths in AudioClock::SetPlaybackRate(). So I guess we're taking the path in SetPlaybackRate() that doesn't set mOldOutRate, and so mOldOutRate is then 0 when we come to all AudioClock::GetPosition(), which poisons all future calculations. We don't appear to be crashing, despite the fact that we're dividing by 0. I'd guess the fix is to initialize mOldOutRate to mOutRate in AudioClock::Init(). [1] http://mxr.mozilla.org/mozilla-central/source/content/media/AudioStream.cpp#1088 [2] http://mxr.mozilla.org/mozilla-central/source/content/media/AudioStream.cpp#1096

Initialize mOldOutRate to mOutRate, and initialize mOldBasePosition to 0. This seems like the right thing to do, and it stops the assertion spam and the current playback position now advances for me when playing MP4 videos using the Windows Media Foundation backend I'm working on.

Comment on attachment 690157 [details] [diff] [review] Patch v1 Review of attachment 690157 [details] [diff] [review]: ----------------------------------------------------------------- Funnily enough, I've came to the same conclusion and was about to upload the same patch. Thanks!