mnandigama
Current Status - Posted on 2010-03-08
Done:
Looked at the regression ranges [ Regbugfixes/(Genbugfixes+Regbug_fixes) ] by component and by top 10 files [ by total bug fixes ] in each component and by overall top ten files [ by total bug fixes ] to identify any SPC spread exceptions.
Identified a bunch of source files that are part of 1.9.1 and no longer exist in trunk. What would have happened to the corresponding tests. Are they refactored ? retired or running as usual. If they are running as usual, then are they still testing what they are supposed to test ? In order to answer this question, I need to generate test case to source file mapping which would be many to many using modified code coverage test run scripts.
Next:
Design a test routine that executes tests one by one and collects test case to source file mapping.
Older Posts
Member of Projects
Team Posts
Posted on 2010-03-15
cjones
Done:
- Some OOPP topcrashes and beta blockers
- Low-level work on paired plugin/browser "hang dumps"
- Proposed "STL policy"
Next:
- Submit plugin/browser hang dumps
- Push STL through
- Publish layers from OOP (or joe?)
- IPDL semantics
Posted on 2010-03-15
sdwilsh
Done:
A bunch of reviews and work on I/O reduction goal.
Next:
- More reviews
- Start work on an IndexedDB prototype
- Land bug 536978 (async cookie writes) for Q1 goal.
Posted on 2010-03-12
ddahl
Done:
DevTools status.
The current console patches are found on bug 545266 and bug 546708 - also try builds will start once things are actually usable - I hope end of next week. It turns out that trying to attach (and keep attached) an arbitrary object on a nsIDOMWindow is a it of a pain - from outside the window's constructor. sicking told me about his patch for detecting new windows (before content scripts run) and attaching objects to them: bug 549539 - which landed late today. Yay. With this patch on trunk it should be trivial now to detect window construction and attach a console.
Whoops, forgot to mention that the console does log to the output, but once you navigate away, the console object attached to the window is destroyed.
I gave up on the html document approach for now, I think it can be worked out, just was spending too much time on it. For now, it is just a XUL vbox element where the output is appended to.
Next:
Work with the new window detection hooks to implement the observer that keeps window consoles re-appearing. Every load of the browser creates a new nsIDOMWindow - news to me, anyway.
Posted on 2010-03-12
bzbarsky
Done:
- Finally pushed some security fix patches
- Fixed bug 550882, then spent a while debugging why Tsspider regressed as a result.
- Fixed the fact that Tsspider is measuring the wrong thing.
- Reviewed bug 536466
- Reviewed bug 550351
- Various other reviews
- Some JS profiling
- Fixed bug 551118
- Started bug 462076 review
Next:
- Finish bug 462076 review
- Do async redirect review
- Window ids
Posted on 2010-03-12
Mossop
Done:
- Wired up installs and blocked installs from webpages
- Converted the xpinstall test suite to work with the new APIs
- Started preparing the project branch for performance testing
Next:
- Finish support for extensions that don't need restarts
- Personas and plugins support
- Background auto-updates
- Convert the old EM test suite to the new APIs
- Make sure the API docs are in sync with the code
- Make a list of things not yet implemented for QA
- Put together a plan for a trunk landing
Posted on 2010-03-11
jdm
Done:
New patch (hopefully final) up for review for remoting chrome registry (bug 542907) Static analysis for bug 551286 (NS_FINAL attribute for functions) Various investigations into multiprocess Jetpack design
Next:
More Jetpack investigation - properties/restrictions of evalIntoSandbox and scope, CPOWs
Posted on 2010-03-11
mfinkle
Done:
- Create browser-tests for new Form Autocomplete
- Create WIP patch for Open in New Tab via context panel (bug 551711)
- Create patch for auto-updating add-ons (bug 551709)
- Added status flags to Fennec 1.1 planning page (https://wiki.mobile.org/Mobile/Planning)
- Create presentation for Firefox Mobile Add-ons
- Work with Caitlin on the new Firefox Mobile add-on challenge
Next:
- Continue WIP patches
- Create a WIP for start page redesign
- Work with Romaxa on portrait and volume keys bugs
- Get Vivien, Felipe, Romaxa and Robert together to get a solid approach to link-tapping
Posted on 2010-03-11
fredinfinite23
Done:
- Remoted link-visited info from content to chrome, bug 516728 patch posted (review pending)
- Remoted link-visitedness fetches from content to chrome, bug 551181 patch posted (review pending)
- Created an xpcom wrapper to test new IHistory async API, bug 546938 patches posted (review pending)
- made HTTP POST requests work for e10s necko, bug 536273 patches posted (review pending)
Next:
- Correct IHistory xpcshell test leaks (bug 551020)
- Make xpcshell IHistory tests run on parent process too (bug 551014)
- Write xpcshell unit tests for IPC remote prefs and observers (bug 506269)
- Write tests for libpref
Posted on 2010-03-10
bsmedberg
Done:
- Async stream delivery, bug 551049 patches posted (review pending)
- Got record and replay working again, woo
- Debugged topcrash @ _filbuf, bug 546035, appears to be caused by messages racing with a crashed (or hung->crashed) plugin process
- found a serious/common hang while debugging above, bug 551242
Next:
- Backport to Lorentz branch
- Review of IPC-prefservice
- Review of IPC-chromeregistry
Posted on 2010-03-09
bhsieh
Done:
Got review for bug 514705 (content should inform chrome of navigation) and bug 520309 (replacement for fastload cache). Started new patches for both.
Made patch for bug 548810 (crash reporter bug in child process shutdown).
Next:
Get new patches, reviews, and land 514705 and 548810.
Get API approval for 520309, get implementation review.
Posted on 2010-03-09
bent
Done:
- Structured clone, worker PostMessage up for review.
- Tried unsuccessfully to get VMWare to record bug 541707.
Next:
- Add some binary hooks to test builds to get VMWare recording automated from within the guest.
- Add some hooks to runtests.py to start and stop recording around mochitest runs.
- Cached identifiers for silverlight, Lorentz beta blocker.
Posted on 2010-03-09
pettay
Done:
- Few changes to TabChildGlobal API (based on the feedback). All the comments very welcome https://wiki.mozilla.org/ContentProcessEvent_Handlers
- reviews (editor, e10s, event handling)
- Some more dialog handling in E10s
- meeting with hybi wg chair (websocket protocol)
Next:
- dialogs for e10s
- CPOW for content process events listeners (depends on CPOW)
Posted on 2010-03-09
jduell
Done:
- reviewed bug 546756 (xpcshell syntax errors), bug 534764 (NSPRLOGFILE for e10s), bug 543726 (change to HttpChannel cache IDL), bug 536279 (request/reponse HTTP headers for e10s). Started review o f bug 546581 (base class for Http channels).
- made some progress on e10s HTTP notifications
Next:
- more e10s reviews: bug 546581 (base class), bug 536273 (HTTP POSTs),
- pile of non-e10s reviews to do
- HTTP notifications
- testXUL bootstrap
Posted on 2010-03-08
ctalbert
Done:
- Got heartbeat daemon working.
- Presented Automation Roadmap to QA Team
- Hammered away on mobile automation bugs
Next:
- Complete mobile automation bugs
- Clean out review queue
- Help Joel land patches for mobile test refactoring
- Meet with Henrik and Heather on Mozmill plans going forward
Posted on 2010-03-08
mnandigama
Done:
Looked at the regression ranges [ Regbugfixes/(Genbugfixes+Regbug_fixes) ] by component and by top 10 files [ by total bug fixes ] in each component and by overall top ten files [ by total bug fixes ] to identify any SPC spread exceptions.
Identified a bunch of source files that are part of 1.9.1 and no longer exist in trunk. What would have happened to the corresponding tests. Are they refactored ? retired or running as usual. If they are running as usual, then are they still testing what they are supposed to test ? In order to answer this question, I need to generate test case to source file mapping which would be many to many using modified code coverage test run scripts.
Next:
Design a test routine that executes tests one by one and collects test case to source file mapping.
Posted on 2010-03-08
vivien
Done:
- Hack around xresponse for getting some scrolling performances data between different fennec
- Analysing data for the scrollbars indicators in fennec
- Working on a touch hit patch for fennec
Next:
- Finalizing the hit patch for fennec and ask review for some layout problem i have
- try mfinkle's approach to avoid perfs regression on kinetic scrolling
- Correct the 400ms double click delay patch which sounds make fennec feel more responsive
Posted on 2010-03-03
karlt
Done:
Next:
Posted on 2010-03-02
kmacinnis
Done:
Puppetization doc / bz handed to ops Nagios checks bz -> ops Firefighting: disk, logs, et al
Next:
Metrics via JMX
Posted on 2010-03-02
deinspanjer
Done:
Pushed updateType feature to production for AMO ADU Pushed Socorro HBase storage to production Finished first draft of CDA Kettle data access layer
Posted on 2010-03-02
michal
Done:
bug 537164 - e10s HTTP: caching bug 549241 - Channel classifier
Next:
bug 537164 - e10s HTTP: caching bug 549241 - Channel classifier bug 536317 - e10s HTTP: implement Cancel bug 486199 - SCTP
Posted on 2010-03-01
honzab
Done:
Made some progress on sec UI for e10s, but still in the design phase Bug 543726 HTML5: test_foreign.html fails Some e10s discussions
Next:
Bug 534667 app cache for e10s sec UI
Posted on 2010-02-16
jmaher
Done:
Next:
- think more about hardware for winmo automation
- get a few winmo automation patches reviewed and checked in
Posted on 2010-02-08
Tristan
Done:
Refactored processor code into library.
Added random port selection to stackwalk-net.
Began jsonz loader.
Next:
Finish up jsonz loader.
Continue helping ozten.
Posted on 2010-02-02
benjamn
Done:
- Progress on window-closing related bugs (543528, 523885) unrelated to electrolysis.
- CPOW patches continuing to get simpler as supporting patches land.
Next:
- Exploring high latency, low bandwidth talos infrastructure to support regression testing of HTML5 speculative parsing.
Posted on 2010-01-26
kkovash
Done:
Analyzing responses from recent Major Update survey (answering the question, "why don't people upgrade?").
Next:
Publish resulting insights and recommendations sometime this week.
Posted on 2009-12-22
harth
Done:
Initial triage and workers view up for crash automation.
Next:
Integrating crash automation interface with the Bugzilla workflow using REST API.
Posted on 2009-12-22
Unfocused
Done:
- Got first review of Tab matches in Awesomebar.
- Started work on new Extension Manager UI.
Next:
- Tab matches in Awesomebar: review, rinse, repeat.
- Building Extension Manager UI prototype and hooking it up to the new API
Posted on 2009-12-02
bcutler
Done:
Analyzed: First Run A/B Test Running: What's New A/B Test; First Run Support Test
Next:
SiteSpect: Survey test, Speed test, 2.* First Run tests Web Analytics: Migrate AMO to Omniture
Posted on 2009-11-23
bcombee
Next:
Bug 499904 - after exploring adding DBus handling directly into Fennec as part of the phone support component, I'm instead writing a small proxy utility that will intercept Maemo browser requests and use them to launch Fennec. This relies on bug 529576 to land to have proper "open in new tab" behavior
Posted on 2009-11-20
antoine
Done:
Work on bug 516749, finalized the parent/chrome side.
Next:
Fix a remaining bug with casting, and check all prompts that should use the new IPDL protocol.
Posted on 2009-11-12
blassey
Done:
password echoing follow ups
Next:
taking a look at xptcall patch from jacob
Posted on 2009-11-09
jimm
Done:
win32 windowless plugin drawing patch is working and passing tests. (bug 522299)
Next:
create a stand alone shared dib class based on the above work. integrate with shmem if possible.
Posted on 2009-10-30
dietrich
Done:
Status update: http://autonome.wordpress.com/2009/10/31/firefox-startup-performance-weekly-summary-8/
Posted on 2009-10-19
mconnor
Done:
- Short week due to Thanksgiving/travel to Bay Area.
- Labs offsite!
- Some work on Weave Firefox UI
- Dealing with perceived sec implications of Personas integration in Firefox
Next:
- Finish and land new Weave UI for Firefox for pre2
- Fixing followups
- Fennec UI evolution
Posted on 2009-10-13
jrmuizel
Done:
Very basic event forwarding on windows
Next:
Get more complex events working
Posted on 2009-10-07
crowder
Done:
- Just finished helping to track down and kill what is hopefully the last ARM bug merged from Tracemonkey (bug 518227).
Next:
- Bug 506926 - fixing broken Fennec reftests on WinCE
- Faster fast-startup component (need to do remote start checking/comm before XUL setup is done, if possible)
- Need to be able to inhibit pop-up windows for FastStartup to work cleanly. In other words, if the user isn't interacting with the browser, it shouldn't be trying to interact with them. Best current example of this is bug 516794 (the part where we complain at startup)
- jsfunfuzz on WinCE