TPEPlatform/2014-07-03: Difference between revisions

Jump to navigation Jump to search
 
(25 intermediate revisions by 13 users not shown)
Line 56: Line 56:
==Media==
==Media==
====Randy Lin====
====Randy Lin====
* {{Bug|951008}} - Media Recorder - fire onstart event when encoder start to generate encoded data
** Landed
* {{Bug|1023783}} - [Flame][V1.4][Music]The songs will play by loudspeaker when there is a headset insert.
**  OOAD
** Should be regression, audio channel services isn't fire the pause event to speaker manager.
* Profile WebRTC h264 performance.
** b2g process cpu usage become higher than 15 days ago.
* PTO 1 day
* TOEIC test 0.5 day
====Alfredo Yang====
====Alfredo Yang====
*{{Bug|984274}} - Intermittent test_sandbox_permission.html | Test timed out.
** r+ and under try run.
*{{Bug|916643}} - ImageCapture - Implement WebIDL and takePhoto()
** The first draft will be ready at this week.
====John Lin====
====John Lin====
* {{Bug|997593}} Cannot playback a recorded mp4 video on Firefox OS
** Investigate the problem of "only one MP4 video can be played in a page" - Impl. of video playback doesn't release HW decoder. Ask rlin to update landing test page to workaround it.
** Prepare patch for review.
* {{Bug|1009420}} [B2G][WebRTC] ACodec decode thread block on waiting consumed graphic buffer from GonkNativeWindow
** Compare the average latencies of decoding frames before and after applying Sotaro's patch - no significant difference. Suspect that QCOM's decoder contributes most of the delay.
** Talk with bwu about the result.
* Help ctai with {{Bug|969312}}
* Explore ringtone trimmer solution.
** Discuss with slee & jwwang.
** Check media stream & recording W3C spec.
* 1 day PTO
====JW Wang====
====JW Wang====
*{{Bug|902686}} - Change manifest.js to use SpecialPowers.pushPrefEnv
** review patch
*{{Bug|1033121}} - Race in nsTimerEvent destructor
** debug
** race condition in nsTimerImpl that causes timer cease to fire and timeout in media tests
*{{Bug|1033172}} - Should mCurrentSeekTarget be reset at exit of MediaDecoderStateMachine::SeekCompleted?
** debug
** report bug
*{{Bug|1032078}} - TEST-UNEXPECTED-FAIL | /tests/content/media/test/test_fastSeek.html | uncaught exception - TypeError: document.body is null
** mentor bug
====Benjamin Chen====
====Benjamin Chen====
*{{Bug|991776}} - Intermittent test_mediarecorder_record_audiocontext.html | Test timed out.
** found root cause: Hook the MediaEncoder to TrackUnionStream is not synchronuos + AudioBufferSource run out of the buffer.
** r?
*{{Bug|1015856}} - [MediaEncoder::GTest] Testcase for bug 970774, WebM aspect ratio
** r+, last nits.
==Platform==
==Platform==
====Steven Lee====
====Steven Lee====
* {{Bug|1016184}} - [Touch Caret] Enable touch caret on B2G
** one shutdown timeout problem existing. Trying to figure out the problem
** Can reproduce the problem on my emulator
* {{Bug|1020725}} - [Text Selection] Enable text selection on B2G
** Pending
* Help Andre on SpeechRecognition
* ringtone cropping discussion
* take over media topics
====Chiahung Tai====
====Chiahung Tai====
*PTO
*Health Exam/PTO
**1 day
** 0.5/0.5 day
*{{Bug|969312}} - [MediaEncoder] Support data from GonkCameraSource in HW video recording path.
**WIP
**ACodec.cpp(libstagefright) has a bug. It only set the value of store-meta-in-buffer when the value is 1. So I think we can just re-configure the OMXEncoder. Try to implement a version which use |release|. The whole release version takes around 90-150ms to restart and wroks fine.


====Tingyu Lin====
====Tingyu Lin====
* {{Bug|982560}} - Mach command fails to find tests if executed from a sub directory
** r?
* {{Bug|1024471}} - Make run/debug commands use a scratch profile by default
** land
** Just help rebase those patches since I think this feature is useful.
* {{Bug|1016184}} - [Touch Caret] Enable touch caret on B2G test cases and fix the existing test case failures
** debug
** Debug the application timeout when running mochitest.
====Boris Chiou====
====Boris Chiou====
* {{Bug|959118}} - [LayerScope] Combine layer dump and layer scope
** OOAD | WIP
** Implement a data structure according to original the packet format and unpack these packets into JSON format in the viewer.
** Design Protocol Buffer message and implement it
* {{Bug|975346}} - [LayerScope] LayerScope is not work while HWC is chosen
** r+
* {{Bug|1029956}} - [LayerScope] mLayerRef of color layers are always not correct (0xa5a5a5a5) on the viewer
** r+ | land
* {{Bug|991227}} - clang 3.5 build warning: gfx/layers/LayerScope.cpp:460:35: warning: 'auto_ptr' is deprecated [-Wdeprecated-declarations]
** WIP


==Graphics==
==Graphics==
====Peter Chang====
====Peter Chang====
====Morris Tseng====
====Morris Tseng====
* {{Bug|987040}} - Implement mozbrowserSelection
** r+
** Still have some test failures. Figure out how to resolve this failure.
* {{Bug|1029982}} - Don't change mNeedLayoutFlush and mNeedStyleFlush when nsPresShell::FlushPendingNotification failed
** L
* {{Bug|864595}} - "ASSERTION: point.x should not be outside of rect r" with getBoundingClientRect on an nsRange that falls within trailing whitespace of a text node
** L
====Jerry Shih====
====Jerry Shih====
* {{Bug|987529}} - Implement Vsync dispatch pipeline
** OOAD | WIP
** disucss with Masson about the optimized content process tick() path.
* {{Bug|987527}} - Register Vsync monitor to HWComposer and notify to gecko/hal
** OOAD | WIP
** Try to disable hw vsync event when b2g suspends.
====Chiajung Hung====
====Chiajung Hung====
* ARM open speech
====Solomon Chiou====
====Solomon Chiou====
* PTO 1 day
* {{Bug|937133}}
** r+
** back porting to v1.4 for Dolphin/Tarako
* {{Bug|1002823}}
** OOAD
** Change the design accroding to ROC's comment, which is creating a new kind of Image for overlaying TV screen.
** Write the prototype HW composer for low level overlaying operation(which the vendor calls in ImageLayer directly).


==Manager==
==Manager==
====CJ Ku====
====CJ Ku====
* {{Bug|1028092}} - Dump testing log when testing pass but we expect testing fail.
* {{bug|1023928}} - [System][UX] The waiting icon now comes with a dirty dot
** file|r?
** debug
* PTO
** Root cause address. It's a bug in Gralloc Image path.
** 1 day
* {{bug|976923}} - reftest fails - font-inflation disable-fontinfl-on-mobile-4.html
* {{Bug|1028128}} - Duplicate reftest log
** land
** remove a out of date test case
* {{bug|972697}} - [reftest::B2G] All reftest cases in layout/reftests/font-inflation fail after enable OOP
** debug | r?
** We still have many disable reftest case, especially on B2G. By enable font-inflation test case, we get around 100 test case work on B2G.
* {{bug|1031677}} - Enable reftest 413928-1.xht/413928-2.xht on B2G
* {{bug|1031676}} - Enable reftest box-sizing-replaced-002.xht/box-sizing-replaced-003.xht on B2G
* {{bug|1031674}} - Enable reftest 381497-n.html on B2G
* {{bug|1031673}} - Enable reftest 960822.html on B2G
* {{bug|1031672}} - Enable reftest 728983-1.html on B2G
** file
** file
* {{Bug|1000722}} - Many B2G reftests fail after enabling OOP, because partial drawWindow calls are drawing too small an area
** r+
** rollback change in bug 981477... damn.. it's a hard work...
Confirmed users
1,643

edits

Navigation menu