Platform/2013-03-12

From MozillaWiki
Jump to: navigation, search


« previous week | index | next week »

Platform Meeting Details

  • Tuesday 2013-03-12 - 11:00 am Pacific
  • Dial-in: Audio-only conference# 98411
    • People with Mozilla phones or softphones please dial x4000 Conf# 98411
    • US/Toll-free: +1 800 707 2533, (pin 4000) Conf# 98411
    • US/California/Mountain View: +1 650 903 0800, x4000 Conf# 98411
    • US/California/San Francisco: +1 415 762 5700, x4000 Conf# 98411
    • US/Oregon/Portland: +1 971 544 8000, x4000 Conf# 98411
    • CA/British Columbia/Vancouver: +1 778 785 1540, x4000 Conf# 98411
    • CA/Ontario/Toronto: +1 416 848 3114, x4000 Conf# 98411
    • UK/London: +44 (0)207 855 3000, x4000 Conf# 98411
    • FR/Paris: +33 1 84 88 37 37, x4000 Conf# 98411
    • Gmail Chat (requires Flash and the Google Talk plugin): paste +1 650 903 0800 into the Gmail Chat box that doesn't look like it accepts phone numbers
    • SkypeOut is free if you use the 800 number
  • Engineering Vidyo Room / Warp Core / SFO-Boardroom / Tor Commons Finch
  • join irc.mozilla.org #planning for back channel

Actions

Notices/Schedule

  • 20.0b5 (our fourth week of beta) going to build today
  • Another round of CTP blocking: Blocking Flash 11.0.*-11.2.* = 6.9% users affected
    • Will be enabling/testing this week
25 bugs (26 bugs last week)
66 bugs (87 bugs last week)
Unresolved Aurora 54 Trackers (non-security, not tracked for Beta) Unresolved Beta 126 Trackers (non-security)

Key Issues

Products/Projects

Firefox OS

Firefox Desktop

Firefox Mobile

Blog shout out (repeat signal 'cause it's valuable info)
http://gbrownmozilla.wordpress.com/2013/03/02/firefox-for-android-performance-measures-february-check-up/
Usability and Responsiveness
  • Landed preff'd off in Fx22 and will ride the trains: bug 716403 - Request to hide the navigation bar when scrolling down content. We're busy landing follow up fixes including bug 849539, bug 849254, bug 850154
    • Help wanted: take nightly for a spin and set browser.chrome.dynamictoolbar = true via about:config
  • We're landing final support for ARMv6 QVGA support shipping in Fx20 on April 2nd; QA is testing inbound build for a fix for bug 840593 - In content UI cut off on small screens
  • bug 846774 - Make synchronous events single events to reduce hangs
  • Keep eyes open for more outcomes from last week's work week!

Stability Report [Weekly]

  • Bugs that need attention:
    • bug 744836 (... nsThread::Dispatch | nsTimerImpl::PostTimerEvent) - jumped into top 10 on Aurora 21, needs assignee
    • bug 798274 (gfxContext::PushGroupAndCopyBackground) - came back to be top 10 in Nightly, assigned to Bas but no action for months
    • bug 812881 (libsomxcore.so) - #2 on Android release, assigned to cdouble but no action in >6 weeks
    • bug 846465 (libstagefright.so, HTC devices, JB) - #2 topcrash on Android Beta, needs assignee
    • bug 782223 (libGLESv2_tegra.so, Nexus 7, JB) - #3 topcrash on Android Beta, needs assignee
  • Note:
    • bug 752828 - android.database.sqlite.SQLiteDatabaseLockedException: database is locked at android.database.sqlite.SQLiteStatement.native_executeSql(Native Method) on ICS
      • STR in bug, workaround did not work ( bug 843029 )
  • Mobile :
    • if anyone has run into the following bugs, please contact QA/Dev for Firefox for Android Team:
      • bug 782223 - crash in libGLESv2_tegra.so@0x4ae.. on Asus Nexus 7 (NVIDIA Tegra 3) with JB
      • bug 834243 - crash in mozilla::layers::LayerManagerOGL::Initialize with abort message: "We need a context on Android"
      • bug 795675 - crash in JSC::Yarr::interpret mainly on JB

Snappy [Bi-Weekly]

  • Telemetry dashboards have skewed data for the "idle-daily" dashboard view -- the telemetry submission from multi-day sessions are given too much weight. Metrics has made the "saved-session" view the default until the data is cleaned up. You may need to re-evaluate any past analyses,
    • Comparison: oi49.tinypic.com/1zhp1w.jpg
Startup & Shutdown
  • bug 845907: Generic readahead functions. This gives us a consistent way to warm IO caches
    • bug 810151: Readahead of of omni.ja file - 60% drop in omni.ja startup read time on Windows test. Will land soon
Tab Strip
Content

HELP WANTED:

  • bug 715376: Nathan Froyd's implementation of per-page event queues has been block on DOM tests for a while. He needs help passing/fixing the tests
Hangs
  • bug 566746: Neil Deakin's refactor of form history is almost complete. This will eliminate a common source of main-thread I/O jank
  • bug 669034: David Teller is working on a Session Storage refactor. He is cutting the state collection/serialization into chunks and moving I/O off the main thread
Add-ons
Tools
  • Thunderbird profiling support

Metro [1st Tuesday of each month]

Games [1st Tuesday of Month]

Mobile Web Compat [2nd Tuesday of Month]

Outreach
  • A couple of weeks ago I blogged Help Wanted: Firefox OS UA Detection
    • Fixes for Dailymotion, Dropbox, and Stack Overflow
    • New introductions for Abril, BBC, Bing, eHow, ESPN, Globo, Google (Accounts, GMail, Maps, News), Orange, TechTudo, ThinkGeek, Yahoo, Yelp
    • The best way to get sites fixed is with technical contacts, please share!
  • UA detection frameworks WURFL and DeviceAtlas recognize Firefox OS as mobile
Gecko
  • Two Gecko Firefox OS compat issues
    • bug 832923 Implement <input type='file'> on B2G (blocks Facebook and Twitter)
    • bug 850209 - Support h.264 in <video> element (blocks Web video but does have workaround - use Web activity)
  • Firefox for Android Tablet UA to remain unchanged
    • bug 829596 Remove "Tablet" from UA string of Firefox on Android tablets
Misc

Critsmash [3rd Tuesday of Month]

Memshrink [4th Tuesday of Month]

Roundtable

OrangeFactor Update

  • Last week, we lost all previous OrangeFactor data due to a datacenter migration.
  • Last week's OrangeFactor: 4.03 (Previous Week: lost)
  • Top 10 open trunk intermittent failure bugs:
ID Summary Count
833769 Intermittent leak of 156, 168, 180, 192, 204, 264, 288, 312, 336, 384, 412 bytes (1 Mutex, 1 ReentrantMonitor, 1|2|3|5|9|10 nsRunnable, 1 nsTArray_base, 1 nsThread) 249
761987 Frequent Android Shutdown,tcheckerboard,tcheck2,trobopan,tsvg,tprovider PROCESS-CRASH [@ nssCertificate_Destroy] top frame can also be [@ PORT_FreeArena_Util][@ nssCertificateStore_Lock][@ PR_Lock][@ 0xffff0fc4][@ FreeArenaList][@ nssArena_Destroy] 112
822321 Intermittent Panda "Could not connect; sleeping for 5 seconds. reconnecting socket" in runtestsremote.py, then "Remote Device Error: unable to connect socket: [Errno 111] Connection refused" during reboot device step 109
663657 Intermittent Android "command timed out: 2400 seconds without output, attempting to kill" 94
845735 Intermittent test_bug582412-1.html, 640272.html | This test left crash dumps behind (ABORT: __delete__()d actor: file PPluginScriptableObject.cpp, line 28) [@ mozilla::plugins::PPluginScriptableObject::Transition] 90
788999 Intermittent test_fileapi_slice.html | application timed out after 330 seconds with no output 62
848943 Frequent components/search/test/browser_contextmenu.js | Check context menu label - Got Search Google for "open-suse.ru)", expected Search Foo for "test search" | Checking context menu search URL - Got http://mochi.test:8888/browser/browser/comp... 48
846606 Intermittent browser_dbg_bug723069_editor-breakpoints.js | Test timed out | correct number of editor breakpoint changes - Got 3, expected 4 46
681138 Intermittent failure in test_reftests_with_caret.html | Reftest http://mochi.test:8888/tests/layout/base/tests/bug106855-2.html FAILED 45
603147 Intermittent "child process nnnn still alive after shutdown" during mochitest-chrome or mochitest shutdown 42
  • Landed a patch for bug 848943 on inbound this morning (yay!), though there appears to be a related failure from it now (boo!).
  • Currently without an owner: bug 822321, bug 663657, bug 788999, and bug 603147.
  • Progress on bug 833769 in the last week. No progress on bug 761987 since mid-February.
  • 25 intermittent failures fixed in the last week (thanks!) - List

Designing a New HTTP Disk Cache

The networking team has started the process of designing a new HTTP disk cache. We have lost confidence in our ability to improve the existing disk cache as necessary. We just started, but you can monitor plans as they evolve via this page. Feel free to join the discussion if you have ideas, contact Josh Aas.