Electrolysis/Meetings/2015-07-02
From MozillaWiki
< Electrolysis | Meetings
ursula
- Bug 1021654 Convert about:newtab to content - worked on getting tiles to show up (it happened!) now working on newtab page interactions (pinning, blocking, dragging etc) to work well with e10s enabled.
- That includes fixing Bug 1176061 to stop writing to prefs directly in content and instead use messaging framework
mconley
- Recovering from work week + Canada Day
- M6 - bug 863514 - Electrolysis: Make gesture support work
- APZC for OS X is getting enabled sooner than anticipated. The content-process kludge that we wrote will not be necessary. mstange is taking this work, since he knows what to do.
- M7 - bug 1166351 - Unable to access new tabs and unable to close any tab
- This was renamed to: "XBL parent bindings not properly attached - tabs suddenly do not switch or close properly"
- Reproduced on non-e10s, so removed from milestone tracking. This is a pretty disturbing bug, and Enn has been assigned.
- M8 - bug 1173219 - Make TabParent::RecvCreateWindow less crash prone
- Patch reviewed by billm, doing fixes, will land today.
- M8 - bug 1114299 - [e10s] link with target="_blank" opens window with no tabs/toolbars/menus
- I understand the problem, have a tentative patch, need to fix failing tests
- Reviewed blassey's tab switch talos test - looking forward to tps reports!
- Latest talos numbers - now with OS X! (we should be getting these weekly now): https://bugzilla.mozilla.org/show_bug.cgi?id=1144120#c5
- This is only showing regressions, and not our epic wins elsewhere
jimicy
- Bug 653065 Make the lightweight theme web installer ready for e10s
* Lightweight web themes now work in e10s. * Changed tests to pass in e10s. https://treeherder.mozilla.org/#/jobs?repo=try&revision=1ce1a261383f * Cleanup and patch should be ready for review
- There's a current bug non e10s related where the lightweight themes cause pocket and hello views to blank https://bugzilla.mozilla.org/show_bug.cgi?id=1177046
gabor
- Bug 1170484 - facebook spellchecking issue
The problem is that when nsFrameMessageManager receives a message, it executes script without entering in a micro task. I'm discussing it with smaug what is the correct way to fix it. I'm leaning toward addin an nsAutoMicroTask, but also afraid that it does too much.
- Bug 1173451 - RemoteAddonsChild init
Filed a patch that sends a message from child to parent to check if RemoteAddonsParent is inited or not, and just cancel the init if not. Got r- because of the sync message.
- some reviews, and just have started to read Bill's new add-on API
jimm
- bug 1175677 (m7, mouse click hang with test plugin) - resolved wfm
- bug 1080687 (m7, tabs crash on system shutdown) - waiting on review
- bug 1152300 (m7, Flash steals keyboard input) - wip
- bug 1178998 (identify which hang detector detects a hang) - waiting on review
mrbkap
- Landed patches for the rest of my m7 bugs with the exception of bug 1169268 (crash dealing with clipboard stuff). I have a plan and hope to have to patch written by the end of today.
- Up next: test stuff and m8 bugs.
blassey
- bug 1158561 - Hang with ClearSiteData. Landed and bounced twice yesterday because of problems with WinXP.
- bug 1166132 - Talos test for tab switching, ave review from jmaher, waiting for mconley
felipe
- bug 1129957 - postdata headers - stuck on likely unrelated intermittent orange to land this bug
- bug 903016 - content area click - will finish up before pto, only one patch remaining
- Will be on PTO starting next Wednesday - ask for review requests until Monday
gw280
- bug 1157941 (m7) - Spent a lot of time working on the navigation "page flashing" bug. Finally have a solution that is not that invasive, and fixes it. Waiting on a review, but it currently crashes a few mochitests on try. They should be reasonably easy to fix, I think.
- No more m7s, so going to start working on m8s as soon as that's done.
Giorgio
- Helping add-on developers:
- bug 1177309 "Add-ons get shims despite multiprocessCompatible flag on Fx DE" - bug 1178692 "webProgressListener in framescript gets no event" (combo of about:newTab being in the main process and tab API weirdness, really)
- ABE/NoScript porting: more dependency resolution, legacy cruft removal
and SDK-fication (e.g. AddressMatcher and ChannelReplacement)
- Regrets: I'll likely be unable to join the meetings on 7/9 and 7/10 :(