From MozillaWiki
Jump to: navigation, search


  • continued work updating arewee10syet with incoming reports
  • browser chrome test case fixes are in try. I'll get reviews and request check-in by tomorrow.
    • bug 1194475
    • bug 1100664
  • note: stability on Nightly and DevEd has come back down into reasonable crashes per ADI range.


  • bug 1196539 (m8, plugin painting bug) - fixes plus tests - waiting on a try push, has reviews.
  • bug 1137944 (m7, plugin positioning bug) - wip of async approach posted. generally when composition deferment is turned on, rendering performance sucks. But hopefully I can avoid turning it on for extended periods of time. to accomplish this I currently have plugins hiding for apz wheel scroll, looking at adding that for certain types of dom scroll (page down/up, scrollbar drag) next. non-apz smooth scroll is pretty hosed, need to figure out if we'll be releasing without apz. if not, or if users can turn apz off, i'll need to figure something out here too. remaining dom scroll fixes I need to figure out might solve this.


  • bug 1200336 - Sandboxing doesn't work on a local build built with VS2015 - on inbound.
  • bug 1156742 - print to xps with low integrity sandbox - feedback from roc over using Moz2D recording and sending a page at a time over IPC. Working on initial patch for page at a time. Just had feedback from Bas to explore using Moz2D recording instead of EMF further.


  • bug 1196973 - securityUI not upated on remoteness transitions - checkin-needed
  • bug 1109146 - browser.altClickSave doesn't work - patch in progress
  • bug 1203184 - disable on Aurora - done
  • bug 1190297 - mistyped keyword search clears urlbar if keyword.enabled = false, next


Clear decision not to do any e10s testing in 42 beta, correct? Yes. e10s will be disabled in Aurora as of today in preparation for beta 1 Plan for beta 43 testing? Need decisions on sample size and UI well before 2-Nov. Note: funnelcake build for Fx43 Beta


  • More work on m8 bugs.
  • My review/needinfo queue has (as of yesterday) gotten long. Will spend some time knocking it back into submission.
  • Will be taking a few days over the next 6 as PTO.


My job right now is to help solidify the plan of record Scope perf tooling for v1 with Vladan Procure actionable optimization work items during M8 Talos tests - get with Joel


  • Un-breaking popular add-ons:
 - FlashBlock: developed dirty hacks to keep it working,
   author is reviewing and merging
 - Mega: developers released new version which appears to work,
   invited them to port their Chrome version to WebExt
 - Google Translator: no feedback from author, starting a WebExt
   porting myself :)
  • What could AMO do if we fork a popular add-on with unresponsive
 developer (e.g. Google Translator)? 
 Maybe Firefox could prompt users to switch?
  • Likely the most Firefox-friendly Slashdot thread in ages,
 where I get quoted a lot in comments rejecting WebExt FUD :) 


  • Bug 1189034 - Haven't managed to reproduce this yet. I want to try and find a machine with the same gfx card and so on as the reporter.
  • Bug 1183296/1109714 - The <select> stuff appears to be mostly fleshed out and sort of working (I think?) for bug 1109714, but we need to fix bug 1183296 to make it work properly. I've been looking into this and basically have a semi fleshed out implementation of

nsIWebContentConverterService to make this work.

  • Bug 1111892 - Flash plugin hang on Linux. Can reproduce now, have it broken in the debugger and asking jimm for help in finding out the root cause. I have a few leads on where to investigate further.


  • no actual work accomplished this week