Electrolysis/Meetings/2016-05-05

« previous week | index | next week »

billm

• Working through some tryserver failures for bug 1262671.
• Still looking at stability in general. Filed bug 1269961 to figure out whether ShutDownKill crashes could be leading to our horrible stability numbers in telemetry. Filed bug 1270308 to figure out how we can improve the ShutDownKill issues.

mconley

• M9 bug 1263887 - [e10s] Select dropdowns in e10s mode do not notice changes to the <select> after it has opened.
  • Uplifted to Aurora
• bug 1261842 - Make initial browser remote sooner if we're defaulting to using remote tabs
  • I've written a bunch of tests. WIP patches in bug 1267720.
  • In the midst of the nsWindowWatcher refactor.
• bug 1166351 - XBL parent bindings not properly attached - tabs suddenly do not switch or close properly
  • Got r+ from bholley on two of the patches. Last one is waiting. My try push is frustratingly unhappy though, so I'll be looking at this.
• Bunch of reviews and needinfos.

bsmedberg

• b1 crash rates: https://sql.telemetry.mozilla.org/queries/271#445
  • non-e10s: 9.2
  • e10s: 27.2
  • a full 33% of e10s content crashes in b1 were caused by bug 1255823 which is fixed for b2, so we expect the crash rate for b2 to stabilize around 20
  • theory going around that shutdown crashes are inflating this beyond reason (and not showing up in crash-stats): see billm
• OOM is better but still a concern

chutten

• Beta 47 E10s Experiment Notebook is up: https://github.com/vitillo/e10s_analyses/blob/master/beta47-ex1/e10s_experiment.ipynb

kanru

• bug 1259192 REOPENED:: crash in mozalloc_abort | NS_DebugBreak | mozilla::ipc::FatalError | mozilla::net::PHttpChannelChild::FatalError | mozilla::net::PHttpChannelChild::OnMessageReceived
  • Bug reopened because of new ongoing crashes
  • Some initial analysis
• bug 1264642 NEW:: Reduce the size of PBrowser::Msg_AsyncMessage IPC messages
  • passed all js tests and all test_postMessages.html
  • publish WIP patches and summarize current status

thinker

• bug 1253241 NEW:: [e10s] "Scroll of Lorem Ipsum" testcase causes graphic glitches when hide menupopup after open Inspector
  • under review.

cervantes

• bug 1259480 [e10s] FatalError crash in PContentChild::Read from PContentChild::SendReadPrefsArray
  • Running talos to investigate the impact of checksum in IPC messages
• bug 1258317 REOPENED:: crash in mozilla::ipc::TransferHandleToProcess (PBackground related)
  • Looking at the crash reports and dumps
  • Traceing sandbox code
  • Working on the patch as suggested by bobowen

tchou

• bug 1245068 NEW:: [e10s] Ctrl+mouse wheel to change zoom level doesn't work if web page is not focused
  • tracing GetContentViewer whether it is called from the EventStateManager associated with the window where the wheel event occurred
  • submitted wip and asked for feedback
• bug 1005844
  • Persuading to land the patches as it might reduce our OOM rate

wcpan

• bug 1259641 RESOLVED::FIXED [e10s] MediaFeatureValuesChangedAllDocuments takes tons of time when resizing e10s window
  • Request uplift

ddurst

• confirmed that crash-stats data directly integrated with re:dash will not be soon (but is being worked on)
• creating ipynb for use as cron for 1265015, needs review after testing

cpeterson

• test sign-off: elan and I are still following up with module owners to sign-off module test coverage.
• SLOW_SCRIPT_PAGE_COUNT/hour in Beta 47 is 0.34 in non-e10s and .42 in e10s.

felipe

• bug 1268921 - Allow non-integer values to be specified for the rollout sampling
• Tests spreadsheets
• Marionette on Beta

bobowen

• bug 1035125 - On Windows, plugin-container.exe is linked against the sandbox_s library twice - patches up for review, but there could be a problem with VS2015 (see bug 1265615)

aklotz

• bug 1268544 - Refactor a11y IPC to allow for remoting COM objects
• Trying to have 0.1 build to test by mid next week

mccr8

• bug 1268938 - Set name for telemetry of IPC reply messages.
• bug 1270247 - Crash in ParamTraits<mozilla::net::NetAddr>::Write instead of sending a malformed message.
• bug 1269365 - OOMs during deserialization of nsTArray should result in an OOM crash.
• bug 1268130 - Crash more aggressively in serialization of memcpy-able nsTArrays.
• bug 1268721 - Null deref crash in GetSurfaceData() called from PuppetWidget::SetCursor().
• bug 1268662 - Started working on chunking prefs to avoid IPC message OOMs in parent.

mrbkap

• bug 1267844 - dropping a tab into an overflowed tab bar from another window gives an infinite throbber. I've been banging my head against this, still debugging and trying to figure out what's going wrong.
• bug 1183037 - autocomplete dropdown gets "stuck" when the user clicks on the input. I attached a very hacky patch. Now waiting for review.
• Needinfos and reviews.

elan

• Track Back Schedule
• Next milestones Today (May 4) (initial results from experiment), Next is week of May 9 new telemetry dashboards should be available.

ksteuber

• bug 1263774 - Send memory reports. Try run is failing on a test that seems unrelated to my work. I am going to try rebasing and see if they pass.
• bug 1269185 - Crash happening in function that I wrote (nsZipHandle::findDataStart). I think I can get this back to the rate that it was crashing at before by catching the SEH exception in Windows. Just requested review on this.