The description in comment8 was not quite right. As we would do the time adjustment in ReaderProxy [1], the audio sample's `mTime` should be adjusted to be larger than 0, unless the first sample's start time is less than the start time we got from metadata. So usually the start and end time of audio sample are both larger than zero. If this crash were caused by negative start time (which might cause overflow if we substract it from a super large positve end time). We should force that all samle's start time in MediaQueue should be larger than zero. [1] https://searchfox.org/mozilla-central/rev/d302c3058330a57f238be4062fddea629311ce66/dom/media/ReaderProxy.cpp#46
Bug 1540748 Comment 9 Edit History
Note: The actual edited comment in the bug view page will always show the original commenter’s name and original timestamp.
The description in comment8 was not quite right. As we would do the time adjustment in ReaderProxy [1], the audio sample's `mTime` should be adjusted to be larger than 0, unless the first sample's start time is less than the start time we got from metadata. So usually the start and end time of audio sample are both larger than zero. If this crash were caused by negative start time (which might cause overflow if we substract it from a super large positive end time). We should force that all sample's start time in MediaQueue should be larger than zero. [1] https://searchfox.org/mozilla-central/rev/d302c3058330a57f238be4062fddea629311ce66/dom/media/ReaderProxy.cpp#46