Autophone shows a throbber stop regression on 2016-12-23 on mozilla-inbound: https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=f79edc3888e5ce7647fd259cbceb6e806df2a715&tochange=fb7b4a449847ab2e0149a4a3b1d9f7b511414cca http://phonedash.mozilla.org/#/2016-12-23/2016-12-26/binning=repo-phonetype-phoneid-test_name-cached_label-metric&rejected=norejected&errorbars=noerrorbars&errorbartype=standarderror&valuetype=median&remote-blank=on&remote-nytimes=on&remote-twitter=on&throbberstart=on&throbberstop=on&first=on&mozilla-central=on&mozilla-inbound=on&nexus-4=on&nexus-4-5=on&nexus-4-6=on that is merged onto mozilla-central on 2016-12-25. This is most apparent on the slower Nexus 4 devices but does appear to affect the other devices to a lesser extent. either Bug 1321644 or Bug 1321644 or both together appear to be the cause(s).

That should have read: Bug 1319850

and seen on perfherder: == Change summary for alert #4608 (as of December 23 2016 17:09 UTC) == Regressions: 6% remote-twitter summary android-4-2-armv7-api15 opt 2580.55 -> 2737.59 4% remote-blank summary android-6-0-armv8-api15 opt 839.64 -> 874.67 3% remote-twitter summary android-6-0-armv8-api15 opt 967.22 -> 993.66 2% remote-blank summary android-4-4-armv7-api15 opt 1276.51 -> 1304.9 For up to date results, see: https://treeherder.mozilla.org/perf.html#/alerts?id=4608

There was a synchronous creation even added in Bug 1319850. That's the only thing I can think of that might be causing this. I've created a patch that tries to eliminate or at least decrease it's impact. I'll run on try to see if it addresses the issue.

Here is the try build: https://treeherder.mozilla.org/#/jobs?repo=try&revision=8275a6d66106fc9e2ba38ade78b42d928637ace6

Not sure how to test if this patch addresses the issue. The try build posted in Comment 4 has finished.

new try run with autophone results: https://treeherder.mozilla.org/#/jobs?repo=try&revision=2d76f65047a16105faba6d8e92dcbdfa3ed12e93

Randall, you should post the patch and get this in

(In reply to James Willcox (:snorp) (jwillcox@mozilla.com) from comment #7) > Randall, you should post the patch and get this in Since I couldn't figure out if this patch helped with regression I created Bug 1328747. I just pushed it to inbound. If things improve I guess we can close this. Otherwise I'm not sure what to look at to fix this issue.

bug 1328747 did improve things. -> fixed