Platform/2009-08-04
From MozillaWiki
< Platform
« previous week | index | next week »
Contents
Notices / Schedule
- shipped!
- shipped!
- haven't pushed Major Update yet, need to determine when we'll do that
- code freeze tonight!
- is it possible?
- currently 18 blockers that need patches
- targeting release in sync with 3.0.14
- code freeze is in one week
Firefox 3.6a1
- 7 blockers remaining (see blocker report)
- do we need to hold Alpha on these blockers?
- what does this mean for branching?
Blocker Report
Gecko 1.9.2 / Namoroka
- Alpha 1 Blockers (flag: blocking1.9.2, priority: P1)
- Release Blockers (flag: blocking1.9.2 or blocking-firefox3.6)
- 93 open
- 166 nominations (most in blocking-firefox3?)
Browser / Front End
Landed
- Asynchronous location bar has landed
- form awesomecomplete has fully landed
Ongoing work
- updater is now working in WinCE, but there are some polish issues bug 507338
- fixed a bunch of Ts low hanging fruit bugs
- vlad's got a patch to get probes working so we can produce a timeline
- ddahl will be adding probes, putting out a call for others to do the same
- looking at combining XPCOM and JS components to reduce file I/O cost
- streaming video UI mockups available for comment
- theme explorations now starting to consider how to support Add-ons
- doorhanger notficiation work underway
- changing mouse scrolling model to improve perception of performance bug 462809
- check out our list of active projects at any time, and get involved!
- Polish update: Firefox is 59% shiny (0% change)
- Thanks go to: tyler downer, dao
- 46 remaining easy polish bugs (whiteboard [polish-easy])
- 38 remaining hard polish bugs(whiteboard [polish-hard])
GFX Update
- Decode-on-draw (bug 435296) chugging along, working through various test failures. Of particular annoyance was bug 507902, but that should be all sorted out now. Only major feature left to add is an image locking mechanism for layout. Won't make the branch. For the full scoop see the bug or the GFX unmeeting notes.
Layout Update
- background-size (waldo) and CSS gradients (ventnor) landed
- Web-tech posts soon (need to fix Mac gradient bug)
- asynchronous containing block reconstruction landed (tnikkel)
- removed widgets from decks
- bz's finegrained reflow hints for dynamic abs-pos landed
- video/audio tests landing
- frame poisoning looking good (zwol)
- Still to do/land for 1.9.2:
- frame poisoning (zwol)
- scroll repaint optimizations (roc)
- pointer-events:none for HTML (jwatt,
bug 380573bug 508179)
- Post-1.9.2 work
- sharing CSS interpolation code (dholbert, bug 504652)
- Brian Birtles started contracting, working on SVG animation improvements
- www-font wars
Content Update
No updates (BlackHat happened last week, vacations, sick leave).
peterv's got a fix to speed up xpconnectish stuff and quickstubs some more.
Platform-specific Support Update
Work continues on getting infrastructure updated to win7 RC SDK.
JS
Tracing getters is happening!
Security
Electrolysis
Tree Management
- Turning off tests in our current build+tests unittest job (the 'U' on the tbpl), and relying on the packaged tests ('M', 'E').
- Talos TP4 now enabled on all branches except 1.9.0
- going to disable TP3, fast talos.
- Increasing time between idle builds on mozilla-central, mozilla-1.9.1 from 2 hours to 9.
Roundtable
- when to land infallible malloc (beltzner)
- infallible malloc should land after we branch for 1.9.2, definitely
- if it's targetted at 1.9.3, we need to commit to stabilizing for Novemberish for the planned 1.9.3 branch
- what other pieces are we aiming at for 1.9.3? we should get that list
- shipping jQuery (beltzner)
- jetpack ships jQuery, and mozilla-central is considering it
- pressure is to leverage the existing group of web developers who are used to it, and who might want to write add-ons for mozilla platform products
- unsure whether shipping it, or shims is the right thing to do
- unsure what the use cases are
- unsure what to do about other libraries (moo.fx, prototype, YUI)
- will start a dev-platform thread soon to gather requirements
- what's the firebug plan?
- What's the extent of our commitment to firebug working well?
- What platform hooks does firebug need to work well? How do we go about making sure we provide them in an expeditious manner?
- How do we make sure that Firebug developers come to us with issues instead of just hacking around them?
- Do we want to have someone competent at UI design take a look at Firebug UI?
- What's the multi-thread/process story? Web workers, electrolysis, etc.
- Quick update on Code Coverage (timr)
- Latest c/c++ results
- latest JS results
- Content discussion Wednesday Aug 5th 3pm; JS discussion Thursday Aug 6th, 1pm. Others to be scheduled. Feel free to contact Murali or myself if you want to dig into this with us.