980957 - Autophone - s1s2 - Regression in throbber time/throbber stop on 2014-03-04

Status: RESOLVED WONTFIX

(Firefox for Android Graveyard :: General, defect)

Description

[Bob Clary [:bc] (inactive)]

Reviewing http://phonedash.mozilla.org/#/org.mozilla.fennec/throbberstop/local-blank/norejected/2014-03-01/2014-03-07/notcached/errorbars/standarderror you can see that there was an improvement on throbber start times and a coincident regression in throbber stop times on around 2014-03-04 leading to an even larger regression in total throbber time.

This is most noticeable on the slower phones and local test pages but is apparent elsewhere as well. This appeared first on fx-team, then mozilla-central, then mozilla-inbound. Sorry for the difficult ui.

before: 2014-03-04 23:33:23 http://hg.mozilla.org/integration/mozilla-inbound/rev/ca402c27bf50

after : 2014-03-05 05:02:24 http://hg.mozilla.org/integration/mozilla-inbound/rev/2fcb315f6461

http://hg.mozilla.org/integration/fx-team/pushloghtml?fromchange=7a629e8da204&tochange=f87a3c18196b

Wes Johnston — Bug 967254 - Delay Webapp Uninstall code till delayed startup. r=rnewman
Wes Johnston — Bug 977592 - Don't autofocus the password field in auth dialogs. r=margaret
Wes Johnston — Bug 974289 - Fixup test for FilePicker results. r=lucasr
Phillip Sackl — Bug 979326 - [Australis] Tweak the timing of the customize mode transition. r=mconley.

Comment 1

[Bob Clary [:bc] (inactive)]

crap, pasted the wrong set of links. Here are the fx-team links:

before: 2014-03-04 08:16:58 http://hg.mozilla.org/integration/fx-team/rev/7a629e8da204
after : 2014-03-04 08:36:56 http://hg.mozilla.org/integration/fx-team/rev/f87a3c18196b

Comment 2

[Joel Maher ( :jmaher ) (UTC -8)]

1.5 years with no action here, is this of interest?

Comment 3

[:Margaret Leibovic]

(In reply to Joel Maher (:jmaher) from comment #2)
> 1.5 years with no action here, is this of interest?

At this point, I think it would be hard to see if fixing these issues would actually fix the regression.

I suppose we could keep this open to investigate startup time related to webapps, since that seems like the most suspicious bug listed in comment 0.

However, things have changed sine this bug was filed, and I know snorp's team has been focusing on startup performance, so maybe we should just close this out.