Platform/2013-04-23

« previous week | index | next week »

Platform Meeting Details

• Tuesday 2013-04-23 - 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
• join irc.mozilla.org #planning for back channel

Actions

• Clint to follow-up with Ted about review for bug 597064
• Jet to find owners for OrangeFactor bug 761987, bug 806046, bug 861158, and bug 858948

Notices/Schedule

33 bugs (33 bugs last week)	33 bugs (48 bugs last week)
Unresolved Aurora 54 Trackers (non-security, not tracked for Beta)	Unresolved Beta 128 Trackers (non-security)

Key Issues

Threads

• Dev Kit - feedback needed (morgamic)
• Using a pre-processing flag to auto-disable features in later Beta versions (lsblakk)
• Revamping touch input on Windows (jmathies)
• W3C Proposed Recommendation: Web Storage (dbaron)
• Rethinking the amount of system JS we use in Gecko on B2G (jlebar)

Products/Projects

Stability

• bug 856670 (js::RemapWrapper) is our #2 crash on Aurora 22, something made the signature go away on Nightly but we're not sure why. Benjamin Smedberg thinks something might have reduced sizes of thread pools (indexedDB? DOM workers?) or reduced stack frame size, or similar (see comment 28), does anyone know what happened there so we have a candidate for uplift?
• Release crash rates are becoming quite low (1.63 yesterday - we consider everything below 2.0 to be "good"), multiple reasons - Flash 11.7 seems to be slightly better in hangs than 11.6, and then a number of signatures are down significantly in volume for unknown reasons, e.g. bug 812683 and bug 819028 (D3D11-6.2/Win7).

OrangeFactor

• Past week's OrangeFactor: 8.28 (Previous Week: 5.84) - 7.31 if bug 761987 is left out.
• bug 846156 (ASSERTION: mArguments wasn't cleaned up properly!: '!mArguments') has many clones and hits across a variety of tests.
• Still seeing many crashes [@ PL_DHashTableEnumerate] with "ASSERTION: (*(uint32_t*)(table->entryStore + ((uint32_t)1 << (32 - (table)->hashShift)) * table->entrySize)) > 0" spread across many tests (i.e. bug 855185).
• 15 intermittent failures fixed in the last week - List - Thanks!.

Performance [Bi-Weekly]

• bug 859558: Project to eliminate font jank
• bug 830492: gps changed FHR DB's behavior to require fewer fsyncs
• bug 844331 and bug 862599: Bug caused Telemetry submission rates to drop significantly, fixed now.
• Tool for comparing Telemetry simple measures. Make sure to disable mixed-content protection on the page

Blog posts:

• David Teller's project "Async & Responsive"
• Avi's Tabstrip Animation Project and recent progress
• Irving's refactor of Add-On manager storage
• Nathan Froyd took a look at Talos pageload benchmark data
• Julian Seward's update on native stack unwinding progress

Memshrink [4th Tuesday of Month]

about:memory now has the ability to load and save memory reports as gzipped JSON. An improved UI and the ability to diff two profiles will be here soon.

Memory consumption on B2G is an ongoing concern. The amount of memory available to Gecko is a little over 100 MiB, so any app that misbehaves memory-wise can chew that up quickly.

MemShrink reports: week 93--94, week 95--96.

Roundtable

• The birch project branch will remain open for B2G development going forward. Can we turn off triplicate OSX/Windows tests (bug 864517)?