Platform/Planning/Video2010Q4: Difference between revisions
Jump to navigation
Jump to search
No edit summary |
No edit summary |
||
| (5 intermediate revisions by the same user not shown) | |||
| Line 2: | Line 2: | ||
* Reduce orange, reduce assertion failures in mochitests. | * Reduce orange, reduce assertion failures in mochitests. | ||
** VideoData leak (patch in hand) | |||
** test_seek, test_seek_out_of_range, test_playback, test_play_events, test_buffered | |||
** Upgrade PulseAudio to eliminate that source of errors | |||
** Use record and replay | |||
** P1 | |||
* Implement playbackRate {{bug|495040}} | * Implement playbackRate {{bug|495040}} | ||
** Post-FF4 | |||
* Implement 'seekable' | * Implement 'seekable' | ||
** Post-FF4 | |||
* HTMLMediaElement startTime attribute {{bug|498253}} | |||
** Post-FF4 | |||
* libsydneyaudio replacement and audio refactoring | * libsydneyaudio replacement and audio refactoring | ||
** P2 | |||
* Rewrite nsWaveDecoder to use nsBuiltinDecoder. | * Rewrite nsWaveDecoder to use nsBuiltinDecoder. | ||
** Post-FF4 | |||
* Thread pool nsBuiltinDecoderStateMachine. {{bug|592833}} | * Thread pool nsBuiltinDecoderStateMachine. {{bug|592833}} | ||
** P2 | |||
* Refactoring to allow multiple decoders per element. Required by: | * Refactoring to allow multiple decoders per element. Required by: | ||
** Seamless looping {{bug|449157}} | ** Seamless looping {{bug|449157}} | ||
** Chained Ogg support {{bug|455165}} | ** Chained Ogg support {{bug|455165}} | ||
** Adaptive bandwidth | ** Adaptive bandwidth | ||
** Post-FF4 | |||
* Subtitle support... | * Subtitle support... | ||
* | ** Post-FF4 | ||
* Setup environment to measure performance and compare vs Flash. | |||
** | ** P1 | ||
* Perf work for Fennec? (Tremor {{bug|511348}}? VP8-ARM?) | |||
** | ** P2 | ||
* Adaptive bandwidth / RTP (Tim & Blizzard?) | * Adaptive bandwidth / RTP (Tim & Blizzard?) | ||
** Post-FF4 | |||
* Review/Land GStreamer patch (Should this be refactored on top of nsBuiltinDecoder before landing?) | * Review/Land GStreamer patch (Should this be refactored on top of nsBuiltinDecoder before landing?) | ||
** Not our priority; we'll review whatever someone comes up with, IF someone steps up to commit to addressing review comments; they can demonstrate commitment by addressing the currently outstanding review comments | |||
* Pushing video directly from decoder thread to compositing process (can develop on Fennec) | * Pushing video directly from decoder thread to compositing process (can develop on Fennec) | ||
** Need to start on this ASAP (P1) | |||
* Fullscreen API, backport to FF4 perhaps. | * Fullscreen API, backport to FF4 perhaps. | ||
* | ** Not really a video API; will be done when it's done, we can consider backporting | ||
Latest revision as of 21:51, 14 September 2010
Candidate work items:
- Reduce orange, reduce assertion failures in mochitests.
- VideoData leak (patch in hand)
- test_seek, test_seek_out_of_range, test_playback, test_play_events, test_buffered
- Upgrade PulseAudio to eliminate that source of errors
- Use record and replay
- P1
- Implement playbackRate bug 495040
- Post-FF4
- Implement 'seekable'
- Post-FF4
- HTMLMediaElement startTime attribute bug 498253
- Post-FF4
- libsydneyaudio replacement and audio refactoring
- P2
- Rewrite nsWaveDecoder to use nsBuiltinDecoder.
- Post-FF4
- Thread pool nsBuiltinDecoderStateMachine. bug 592833
- P2
- Refactoring to allow multiple decoders per element. Required by:
- Seamless looping bug 449157
- Chained Ogg support bug 455165
- Adaptive bandwidth
- Post-FF4
- Subtitle support...
- Post-FF4
- Setup environment to measure performance and compare vs Flash.
- P1
- Perf work for Fennec? (Tremor bug 511348? VP8-ARM?)
- P2
- Adaptive bandwidth / RTP (Tim & Blizzard?)
- Post-FF4
- Review/Land GStreamer patch (Should this be refactored on top of nsBuiltinDecoder before landing?)
- Not our priority; we'll review whatever someone comes up with, IF someone steps up to commit to addressing review comments; they can demonstrate commitment by addressing the currently outstanding review comments
- Pushing video directly from decoder thread to compositing process (can develop on Fennec)
- Need to start on this ASAP (P1)
- Fullscreen API, backport to FF4 perhaps.
- Not really a video API; will be done when it's done, we can consider backporting