Mozilla Status Board

Recent Posts (All Users)

lbergstrom@mozilla.com today

Done:

  • NOT TWIS (thanks, manish & jdm!)
  • Manager offsite (Monday -- Wednesday)
  • Tracked down the gaps for using the MSVC build of Servo
  • Fixed servo-linux1
  • Contracting negotiations
  • Oxidation & security discussions

Next:

  • TWIS
  • MSVC builder online
  • Fix servo-mac1
  • Share the Android patches or fix Android
  • Look into Windows font stuff on MSVC
  • Try partition/fragment of WPT test runs on mac
  • Oxidation & security discussions

jolesen@mozilla.com today

Done:

  • Implemented file-level testing framework for Cretonne. Go through all the *.cton files in a directory hierarchy, load the Cretonne IL functions in the file, and run the requested tests on them.
  • Added syntax for specifying tests to run, settings to use, and for selecting target ISAs in test files.
  • Wrote a testing guide which documents all the different kinds of tests in Cretonne.
  • Added a table of ISA-specific instruction encodings to functions and implemented the bare basics of the legaliser pass.
  • Added syntax for instruction encoding annotations when writing out functions as text. No parser support yet.
  • Added a source map to the parser output so verifier error messages can refer to line numbers.

Next:

Working on legalization, defining a syntax for specifying instruction transformations.

josh@joshmatthews.net today

Done:

Gecko:

  • addressed review comments for e10s + large serialized uploads, investigated memory leak
  • addressed review comments for cookie path handling
  • wrote patch for intermittent orange for responsive images
  • wrote patch for more efficient AutoIPCSender
  • wrote patch for e10s temp file crash
  • reviewed secure cookie changes

Servo:

  • addressed Promise review comments and rebased
  • reviewed vec conversion PR
  • discussed debugger design
  • reviewed mozmap PR
  • tried to unit test https
  • wrote up introduction and first steps for UPenn student
  • TWiS
  • reviewed response.text(), fetch(), and OpenEndedDictionary PRs
  • looked for issues for promising contributors
  • updated some old PRs for merging
  • reviewed insert_many PR for smallvec

Next:

  • figure out next step for e10s + SW changes
  • address review comments for e10s + large serialization
  • fix responsive image test failures
  • set up blink development environment

masayuki@d-toybox.com today

Done:

This post includes my work between 2016-09-08 and 2016-09-26 (FYI: On vacation between 8 and 12, PTO because of national holiday of Japan on 19, 20 and 22)

  • Writing patches: Bug 1305355 - [e10s] Hits MOZ_ASSERT() when removing composition string of MS Pinyin on Win10
  • Writing patches: Bug 1303273 - NativeKey::mCommittedCharsAndModifiers should be computed with following WM_CHAR messages as far as possible
  • Wrote a patch: Bug 1304624 - Unexpected newline inserted when typing with Chinese input method
  • FIXED: Bug 1300937 - test_keycodes.xul should check KeyboardEvent.key value (and KeyboardEvent.code too?)
  • FIXED: Bug 1138159 - selecting text with ibus pinyin input method on Xubuntu immediately deletes it
  • FIXED: Bug 1300003 - NativeKey should remove all following WM_CHAR messages at handling WM_KEYDOWN first
  • FIXED: Bug 972110 - Intermittent test_bug674770-1.html | The click operation worked successfully | The click operation shouldn't work in the contenteditable area
  • FIXED: Bug 1297013 - mozilla::NativeKey should log its behavior
  • FIXED: Bug 1302956 - NativeKey shouldn't run recursively
  • Reviewed: Bug 1300143
  • Reviewed: Bug 1300140
  • Reviewed: Bug 1272893 - remove some remaining uses of nsISupportsArray in send/import code of mailnews
  • Reviewed: Bug 1302980 - iBus Korean character composition breaks in the awesome bar & the search bar
  • Reviewed: Bug 1273001 - convert nsIHTMLEditor::getLinkedObjects() to nsIArray
  • Reviewed: Bug 1287983 - Implement transitionstart/transitionrun event
  • Answered ni?: Bug 1301476 - EventListener for CTRL-N doesn't work Firefox 48 as in Firefox 47
  • Answered ni?: Bug 1205573 - Google Spreadsheets adds extra newline to cells when cutting text and tabbing between cells
  • Answered ni?: Bug 1299985 - Japanese email is displayed with different font and it's ugly since 2016-08-27 build
  • Answered ni?: Bug 1299553 - Keyup event dropped on mac if cmd/meta key is pressed.
  • Answered ni?: Bug 1284825 - copy a text phrase of a link does not always work
  • Answered ni?: Bug 1295098 - WidgetMosueEventBase::buttonType should be an enum class in mozilla::

Next:

  • Bug 1304624 - Unexpected newline inserted when typing with Chinese input method
  • Bug 1305355 - [e10s] Hits MOZ_ASSERT() when removing composition string of MS Pinyin on Win10
  • Bug 1303273 - NativeKey::mCommittedCharsAndModifiers should be computed with following WM_CHAR messages as far as possible
  • Bug 1257617 - keypress event with ctrl modifier fires twice
  • Bug 1217700 - Create automated tests of NOTIFY_IME_OF_SELECTION_CHANGE and NOTIFY_IME_OF_TEXT_CHANGE

lhansen@mozilla.com today

Done:

  • Memory model further work
  • A bunch of reviews

Next:

  • Something non-memory-modelish for a little while

bbouvier@mozilla.com today

Done:

  • Finally closed bug 1248555: uncanonicalized NaNs in WebAssembly.
  • Try to repro bug 1296640.
  • Investigate wasm failing tests on the ARM simulator under MacOS X in bug 1300380.
  • Report wasm memory reserved for optimization tricks (as it's huge) under about:memory in bug 1303013.
  • Enhance wasm binary-to-text error reporting (instead of "plain crash") in bug 1305097.
  • Investigate TeaVM bugs in bug 1303122.
  • Think about testing + fuzzing strategies for wasm.
  • Try doesn't run some JS test suites: opened bug 1304775.

Next:

  • Needinfos after the wasm 0xc landing + uplifts + empty the (mips) reviews list.
  • ParisWeb (Thursday + Friday), where I'll attend conferences and give one talk about WebAssembly.

mkato@mozilla-japan.org today

Done:

2 weeks activities (9/12-9/23)

  • Bug 1138159 - selecting text with ibus pinyin input method on Xubuntu immediately deletes it
    • Writing patch for 3rd party code and review for our workaround
  • Bug 1279842 - Firefox is reading the ' character as "and number 39" on Reader Mode
    • Review
  • Bug 1290022 - Investigating we can use Rust based xi-unicode as Linebreaker
    • Writing patch to improve Unicode support and PR
    • Investigating our line breaker rules to fit xi-unicode.
  • Bug 1297013 - mozilla::NativeKey should log its behavior
    • Review
  • Bug 1299897 - Widgets (scrollbar etc) don't respect HiDPI setting with GTK 3.21+
  • Bug 1300003 - NativeKey should remove all following WM_CHAR messages at handling WM_KEYDOWN first
    • Review
  • Bug 1301317 - Don't build telephony API without MOZ_B2G_RIL
    • Ni?
  • Bug 1300140
    • Landed
  • Bug 1300143
    • Landed
  • Bug 1301486 - Flash - The candidate window displayed at wrong position when inputting some CCJK characters
    • Investigating and discuss with Adobe
  • Bug 1302980 - iBus Korean character composition breaks in the awesome bar & the search bar
    • Landed
  • Bug 1303982 - CarrySet and CarryClear are missing on non-JIT platform
    • Landed
  • 1 Day PTO
  • 2 Days National Holiday

Next:

  • Bug 1290022 - Investigating we can use Rust based xi-unicode as Linebreaker
  • Bug 1215247 - Enable ECMAScript Internationalization API for Firefox on Android

nnethercote@mozilla.com today

Done:

Landed:

Other:

  • Helped Marco Castellucio set up crash-stats development.
  • Triaged crashes in Friday's Nightly build.

gwatson@mozilla.com yesterday

Done:

  • Work on WR render task / primitive caching support.
  • Opened osmesa / wpt / reftest PRs for review.
  • Moved osmesa-src into the servo repo.
  • Various osmesa-src fixes for build machines.
  • Various small WR optimizations and fixes.
  • Various reftest fixes under WR.

Next:

  • Fix WR + CI patches review feedback.
  • Hopefully land the WR + CI patches.
  • Land the initial WR prim caching support, and convert box-shadow to use it.

mantaroh@mozilla-japan.org 3 days ago

Done:

  • Write patch : Bug 1287983 - Implement transitionstart and transitionrun event
  • Write patch : Bug 1302038 - Add DocumentTimelineOptions
  • Write patch : Bug 1302007 - [Test] Add css-transitions tests of changing to the display property.
  • Write patch : Bug 1293945 - The animation will finish when setting small negative playbackRate.
  • Write patch :Bug 1202333 - AnimationEvent elapsedTime should reflect playbackRate
  • Bug 1298554 - [Test] Remove the some animation.id tests from mochitest.
  • Bug 1302007 - [Test] Add css-transitions tests of changing to the display property.
  • File : Bug 1302644 - [Test] Add tests of canceling the CSS-Animation and CSS-Transition.
  • File : Bug 1302648 - Implement animationcancel event.

Scroll-Driven Animations:

  • Wrote the scroll driven animation spec.
  • Create the demonstration app for TPAC.

Next:

hiikezoe@mozilla-japan.org 3 days ago

Done:

Fixed
  • Bug 1294651 - iterationComposite feature should be behind dom.animations-api.core.enabled pref
  • Bug 1303793 - Beta mochitest(1) failing with dom/animation/test/mozilla/test_set-easing.html | Setting easing should not throw in sandbox
WIP
  • Bug 1304922 - Replace mWinsInCascade with another flag that EffectSet has
  • Bug 1303233 - Reverse composite order of animation effects
Filed
  • Bug 1304565 - Some tests fail in test_restyle.html on MacOS 10.10
  • Bug 1304937 - Removing important rule preventing an animation from running while the animation is active triggers transition
Reviewed
  • Bug 1304805 - test_animation_observers.html | [set_spacing:subtree] records after animation is changed - number of records - got +0, expected 1
Landed test case
Answered to ni?
  • Bug 1235286 - Need automation tests for animation optimizations or detection of regressions about optimizations

Next:

  • Bug 1304922 - Replace mWinsInCascade with another flag that EffectSet has
  • Bug 1303233 - Reverse composite order of animation effects

ajeffrey@mozilla.com 4 days ago

Done:

  • Transactional memory: got enough of the STM API implemented to run a tree-editing benchmark; preliminary results show up to 10x speed-up compared to locking when lock contention is high (https://github.com/asajeffrey/tra-mut-loc).
  • JS shared memory: discussions with shu, and with James Riely of DePaul U, about the memory model.
  • Debugger: discussions with ejpbruel.
  • Reviewed https://github.com/servo/servo/pull/13167.

Next:

  • Propose a JS shared memory model.
  • Servo script thread sharing.
  • Servo script/layout concurrency.

mgoregaokar@mozilla.com 5 days ago

Done:

Stylo: - Make all reftests green (https://bugzilla.mozilla.org/show_bug.cgi?id=1302993 ). - (ongoing) Start work on uncompute (https://github.com/servo/servo/pull/13228 ) - Handle predefined types and shorthands in css-properties-list

Servo: - Mentor/debug https://github.com/servo/servo/pull/13188 (lots of spec bugs to stamp out) - (ongoing) Mentor https://github.com/servo/servo/issues/13235

Review: https://bugzilla.mozilla.org/show_bug.cgi?id=1302949#c0

Next:

  • Continue work on uncompute
  • Review emilio's rule tree
  • Merge geckolib crates

slyu@mozilla.com 5 days ago

Done:

  • ViewSource Conf
  • RustFest.eu

Next:

  • Prepare ViewSource + RustFest sharing
  • Prepare Rust beginner tutorial for Taipei and Hong Kong community
  • Prepare for Web Engine Hackfest

Coordination:

9/24-30 Web Engine Hackfest in Spain

bholley@mozilla.com 5 days ago

Done:

  • PTO
  • stylo meetup TPE
  • landed: bug 1292278 - Stylo Assertion failure: aNameSpaceID > 0 && (int64_t) aNameSpaceID <= (int64_t) mURIArray.Length() at dom/base/nsNameSpaceManager.h
  • investigated style situation with various types of anonymous content
  • landed: bug 1295370 - stylo: Make nsTextControlFrame::AppendAnonymousContentTo safe to call from parallel style traversal
  • landed: bug 1292279 - Stylo crash in parser/htmlparser/tests/reftest/bug608373-1.html
  • landed: bug 1296509 - Make GetFlattenedTreeParent common-case fast
  • landed: bug 1297572 - AllChildrenIterator misses native anonymous content from the root scroll frame
  • landed: bug 1292662 - stylo: Handle style for Anonymous XBL and NAC
  • landed: bug 1299276 - Add a separate --enable-stylo option to control MOZ_STYLO separate from the linkage effects of --with-servo
  • landed: bug 1299348 - stylo: Crash in layout/reftests/bugs/212563-2.html trying to eagerly style scroll frame anonymous content before styling the root element
  • landed: bug 1299629 - stylo: Assertion failure: nsContentUtils::IsSafeToRunScript() in nsPresContext::MediaFeatureValuesChanged
  • repo wrangling
  • bug 1297687 - With "never remember history" enabled, postMessage() is rejected despite origins matching
  • landed: bug 1300287 - stylo: move stylo shutdown hook to the same level as the init hook

mh+mozilla@glandium.org 6 days ago

Done:

Reviewed: - bug 1300695, bug 1299864, bug 1299694, bug 1256155, bug 1301076, bug 1298569, bug 1300654, bug 1301665, bug 1298472, bug 1283203, bug 1290932, bug 1302189, bug 1295486, bug 921063, bug 1301442, bug 1272629, bug 1303085, bug 1302909, bug 1261255, bug 1303227, bug 1303212.

Landed bugs:

  • bug 1300948: Add thread identifier to LogAlloc output
  • bug 1300974: Work around race condition leading to dead lock on fork when enabling LogAlloc
  • bug 1299919: Debug linux static analysis builds are finding the wrong preprocessor
  • bug 1289946: Building with mingw-gcc fails with EnumStringComparisonError
  • bug 1300900: Add gdb helper to get information about memory allocations
  • bug 1295542
  • bug 1303210: Show compressed size in find_dupes.py report, and sort its output.
  • bug 1303232: Use per-process reduced thread-ids when munging logalloc logs.

Other:

  • Looked into bug 1303013
  • Looked into bug 1285531
  • On the side: Rewriting pulsebot in rust.
  • bug 1303926: Beta Mac opt build failing with "mozbuild.base.ObjdirMismatchException: Objdir mismatch" after 50 merge
  • Still working on python configure pseudo-linter
  • Restarted sorting out standalone spidermonkey mess (bug 1176787, bug 1275639)

Next:

  • More python configure
  • Probably more poking at bug 1285531

Coordination:

Waiting for review: - bug 1287671: Simplify jscpucfg.h

Waiting for other bug to be fixed: - bug 1256730: Fail configure when api keyfiles don't exist

alessio.placitelli@gmail.com 6 days ago

Done:

  • Bug 1276200 - Validate engagement measurements
  • Bug 1293954: Make HW Survey data available on games.mozilla.org
  • Bug 1277806 - Implement keyed scalar measurements in Telemetry (in review)
  • Bug 1302812 - Total URI Count shows different behavior when error page visited in different order

Next:

  • Bug 1293955: Add new fields to the HW survey report
  • Bug 1303333 - Implement the probe to count URI loads triggered by search
  • Addon Recommender stuff (Workweek in London)

amarchesini@mozilla.com 6 days ago

Done:

DOM:

  • Bug 1303517 - Are the ChromeFilePropertyBag versions of the File constructor supposed to be usable in workers?
  • Bug 1303029 - nsLocation doesn't have username/password
  • Bug 1302987 - Rename nsLocation to mozilla::dom::Location
  • Bug 1301254 - nsINode::GetBaseURI() should be fallible
  • Bug 1301251 - Handle GetSpec() failure in nsContentUtils::GetWrapperSafeScriptFilename()
  • Bug 1301249 - nsIDocument::GetDocumentURI() should be fallible
  • Bug 1287747 - Intermittent dom/filesystem/compat/tests/test_formSubmission.html,test_worker_basic.html | application crashed [@ mozilla::dom::ContentParent::Release] or [@ mozilla::dom::Blob::Release]
  • Bug 1288997 - MOZ_CRASH(IPC message size is too large)

XHR:

  • Bug 1303095 - Crash in OOM | large | NS_ABORT_OOM | nsAString_internal::Assign | mozilla::dom::XMLHttpRequestStringBuffer::GetAsString

nsIPrincipal:

  • Bug 1297393 - Make passing of subject principals to webidl entry points explicit

sawang@mozilla.com 6 days ago

Done:

Next:

  • Bug 1276553 - Make nsISHistory support "partial" mode
    • Keep working on the GroupedSHistory and swapping things.
    • :mystor is interested in another use case. May align the implementation.

shuang@mozilla.com 6 days ago

Done:

  • Working on worker version: Bug 1286717 - Expose persist() method to StorageManager
  • Having meeting with Anne and Janv

Next:

  • Bug 1286717 - Expose persist() method to StorageManager

Coordination:

Bug 1269052 - Implement WorkerGlobalScope.isSecureContext is not finished. So we cannot access navigator.storage from worker.

ttung@mozilla.com 6 days ago

Done:

Next:

  • Bug 1298329
  • Bug 1301966
  • Bug 1302145 - Intermittent dom/workers/test/serviceworkers/test_fetch_integrity.html | monitorConsole | number of messages [{errorMessage:"The script element has a malformed hash in its integrity attribute: \u201Cabc\u201D. The correct format is...

bhsu@mozilla.com 6 days ago

Done:

Pointer Event

  • Bug 1299202 - [Pointer Events] Make sure only mousemove, mousedown, mouseup and mousecancel used in testcases (RESOLVED DUPLICATE)

  • Bug 1299215 - [Pointer Events] Decouple button and buttons when synthesizing Widget Events. (RESOLVED DUPLICATE)

  • Study APZ test automation

    • The API
    • The mochitest wrapper

Next:

Pointer Event

  • Bug 1299209 - [Pointer Events] Make pointerId a mandatory parameter for test automation APIs (WIP)

  • Bug 1299215 - [Pointer Events] Decouple button and buttons when synthesizing Widget Events.

Bug Fixing

htsai@mozilla.com 6 days ago

Done:

  • Q4 plan

Next:

  • Q4 plan
  • Bug nomination
  • Prepare Q3 report with EPM

jjong@mozilla.com 6 days ago

Done:

  • Bug 1288591 - Implement the layout for <input type=time> (2/3 r+)
  • Bug 1278186 - Implement valueAsNumber and valueAsDate for <input type=week> (ongoing)

Next:

  • Bug 1278186 - Implement valueAsNumber and valueAsDate for <input type=week>
  • Bug 1295402 - Implement the min and max attribute for <input type=week>

Coordination:

  • HTML Forms and Input sharing (Sep 23)

jdai@mozilla.com 6 days ago

Done:

  • Bug 1275838 - Implement CustomElementsRegistry get function(r+)
  • Bug 1275832 - Implement custom element name validation for custom element(r+)

Next:

  • Bug 1275839 - Implement CustomElementsRegistry whenDefined function(Ongoing)
  • Bug 1281375 - Navigating to a javascript URL does not override the document's URL

btseng@mozilla.com 6 days ago

Done:

  • Bug fixing
    • Bug 1298599 - Intermittent TEST-UNEXPECTED-TIMEOUT | /IndexedDB/idbcursor-continuePrimaryKey-exception-order.htm. (landed)
    • Bug 1302261 - Incorrect state in the IDBIndex handle after the deletion of the IDBIndex is aborted. (r?)
    • Bug 1300454 - [meta] Intermittent mochitest timeout after tests in worker is finished.
      • Not reproducible locally but more frequent in Linux32 debug build in treeherder.
      • Logs added to narrow down the root cause.
  • Peer Review:
    • Bug 1293174 - [Pointer Event] Implement implicit pointer capture for touch
  • DOM team meeting in MTV

Next:

  • Bug 1300454 - [meta] Intermittent mochitest timeout after tests in worker is finished.
    • Figure out why the DirectoryLock was not released by previous test.
  • Bug 1282862 - Intermittent failure, PR_ASSERT during xpcshell shutdown on Mac
    • dead in PR_ASSERT in _pt_thread_death().

sshih@mozilla.com 6 days ago

Done:

  • Bug 1297635 - Add a helper function for input to check whether input events with modifier should change the value (landed)
  • Bug 1293174 - [Pointer Event] Implement implicit pointer capture for touch (f?)
  • Bug 1292063 - Support Event.composed (WIP)
  • Bug 1301327 - Enabling w3c_touch_events on desktop Firefox is causing major problems with ExtJS6 (found root cause)

Next:

  • Pointer events on Linux

Coordination:

  • PTO from 9/26~9/29

daisuke@mozilla-japan.org 6 days ago

Done:

  • wip: Bug 1210795 - Display animations' timing-functions in the animation-inspector
  • wip: Bug 1210796 - Display keyframes' timing-functions in the animation-inspector
  • Animation tool meeting with an animator.

Next:

  • Bug 1210795 - Display animations' timing-functions in the animation-inspector
  • Bug 1210796 - Display keyframes' timing-functions in the animation-inspector

mbrubeck@mozilla.com last week

Done:

  • Working on OpenType shaping code
  • Talked to author of opentype and font crates about status of OpenType implementations

Next:

  • Continue work on OpenType positioning
  • Work on Stylo features

nfroyd@mozilla.com last week

Done:

One-offs:

  • Registered for Mozloha.
  • 1:1 with Nick.
  • Reviews...felt like a lot this week.
  • Worked on trying to rationalize string iterators for safety.
  • Looked through bhackett's monster web replay patches.
  • Battled cough/sore throat/sniffles for the latter half of the week.

Next:

  • More of the same.
  • Really try to get mozilla-central to build unmodified with clang-cl.
  • Want to get clang-cl jobs running in taskcluster.

kvijayan@mozilla.com last week

Done:

Next:

  • Finish and land test suites (bug 1301454).
  • Work on more documentation material for flyweb.
  • More BizDev follow-up work.
  • Work on phone-to-phone demo.

Ms2ger@gmail.com last week

Done:

Servo DOM

Servo constellation

Servo other

Testing

Administration

  • 2016-09-12: expenses

SpiderMonkey

Reviews

Next:

  • mozmap

Coordination:

  • may end up taking some time off this week or next

sphink@gmail.com last week

Done:

  • writing up some documentation and getting familiarized with more GC stuff in preparation for terrence's departure :-(
  • thinking through Structured Clone variants, which is unfortunately delaying my review on bug 1302036
  • minor patch to avoid log spam for leaks - bug 1302125
  • Waldo done learned me about perfect forwarding - bug 1302276
  • Minor fix to SM(pkg) job (but it fixes the cause of a bustage mergearound) - bug 1301122
  • Minor fix for MOZ_CCTIMER=filename, bug 1303123
  • worked on a tool to partition profiler samplers into understandable buckets (scripting, GC, layout, painting, etc.) and provide an at-a-glance summary of what a particular page (well, profile) is spending its time on
    • one main goal is to help with first-line triage when finding a page or app that works better on another browser, or otherwise seems slower than it ought to be. No point in digging into layout behavior if it's spending all of its time collecting garbage.
    • mstange pointed me to two other people who have already done something similar. \o/
  • retested some of the gdocs slowdowns with the new nursery CCWrappers. Sadly, none of them were fully "fixed" relative to Chrome
    • but the Taipei team is rerunning their test suite with the patch now, so hopefully we'll get some more solid numbers than my stopwatch-based local runs

Next:

  • Finally do what I should have done long ago -- distribute these slowdowns to other JS team members, since I don't seem to be getting anywhere with them
    • so be prepared for some incoming needinfo's that may end up not being relevant to your area at all
  • Resolve bug 1303461 properly (stack limits on android), since it's holding up a majorly important unmark gray fix (bug 1294747)
  • Resolve my silly talos regression
  • Look more into till's unmark gray bug (I have some better diagnostics for it already; a gray object is getting pulled off of the stack)
  • PTO Friday through the following Wednesday

Coordination:

adw@mozilla.com last week

Done:

  • posted patch, landed: Bug 1297976 - [One-off searches] Screen reader does not read the one-offs searches buttons
  • posted patches: Bug 1299458 - Telemetry data from Search bar is not properly collected when searching in new tab from context menu
  • closed: Bug 1300053 - One-off search engine buttons and the search settings button have duplicate ids
  • mentored, reviewed: Bug 1140286 - Intermittent e10s browser_thumbnails_bg_no_cookies_stored.js | uncaught exception - Error: operation not possible on dead CPOW at :0
  • reviewed: Bug 1302952 - Simplify gear.svg by using the fill filter
  • reviewed: Bug 1283329 - perceptible pause before navigating when pasting a link and quickly hitting "enter"
  • filed: Bug 1302484 - Graphics glitch when browser.urlbar.maxRichResults forces the urlbar popup to scroll
  • talked with Stephen again about one-off search buttons, urlbar, and searchbar behavior (bug 1295458 and others)
  • talked with Jared about urlbar API
  • worked on urlbar API
  • triaged search and location bar bugs

rharter@mozilla.com last week

Done:

Next:

  • XSec Dataset
    • Finalize Aggregation PR
    • Check in 10 new xsec metrics
    • Add Airflow config
    • Fix discrepancy in Null Value handling

Coordination:

I'll be on PTO the last week of September.

bbirtles@mozilla.com last week

Done:

Next:

  • TPAC

jseward@acm.org last week

Done:

Races in Gecko / Orange-race correlation analysis:

  • Completed orange/race correlation analyser. Got results for circa 2700 xpcshell tests and did first analysis against orange-factor results for 3-10 Sept.
  • bug 1274256 (ffmpeg races): made a temp fix that builds on all targets.

UBSan checking of Gecko:

  • Tracked, hacked, tested, the long running bug 1299379 (invalid Side values) -- mucho hassle for a seemingly trivial problem.
  • Decided how to prioritise the remaining, surprisingly few, complaints.

Undef value errors in Gecko:

  • Two complete Mochitest (plain) runs. Slowly working through the backlog of uninitialised value reports that had accumulated.
  • bug 1296716 (Uninits in nsSocketTransportService::AnalyzeConnection): landed
  • bug 1301856 (FPs with gcc-5.3): In progress.

Valgrind:

  • More looking at outstanding bug reports, for a 3.12 release.
  • /**/

Other:

  • Another uptime meeting, very short.
  • Lost quite some time due to repeated hard hangs of my laptop. Display hardware going flaky? Maybe.
  • Spent a day at the Berlin office. Always a nice place to be.

Next:

Coordination:

  • Look at orange/race analysis results.
  • Try to get bug 1274256 landed.
  • More V bug triage/fixing.

amccreight@mozilla.com last week

Done:

  • landed: Bug 1301189 - Use ExposeObjectToActiveJS in nsModuleScript.
  • Bug 1299934 - Run shutdown collection in workers in opt builds.
  • Bug 1300830 - Remove XPCJSRuntime::mDetachedWrappedNativeProtoMap.
  • Bug 1301796 - Fix potential race in RegisterJSRuntime.
  • Work week.