517794 - Txul regression on 9/18

Status: RESOLVED WORKSFORME

(Firefox for Android Graveyard :: General, defect)

Description

[Mark Finkle (:mfinkle) (use needinfo?)]

More than doubled the Txul (window open) time:
http://graphs-old.mozilla.org/#show=3989399,3991126,3985914,3986342,3986630,4007440,3996300,2774336,2774334,2774332&sel=1253189665,1253451238

Based on mobile tinderbox, here's the likely checkin:
http://hg.mozilla.org/mobile-browser/rev/cafbf066fef9

From bug 517600

Interestingly, Ts over the same time period is not affected:
http://graphs-old.mozilla.org/#show=3989585,3991259,3986092,3986478,3986728,4007642,3996440,2774468,2774466,2774464&sel=1253156141,1253430170

Comment 1

[Mark Finkle (:mfinkle) (use needinfo?)]

Looking at the code for the checkin, the problem is probably due to the fact that we are always opening a tab now. Previously, the Txul test probably never caused a tab to be opened. It was only measuring the time needed to open the chrome window, not add the tab.

The patch causes Fennec to always open a tab - which is more realistic. This is also why Ts is not affected. Ts always caused a tab to be opened in the past.

Comment 2

[:Gavin Sharp [email: [e-mail]]]

WORKSFORME, then? I think this is what we want to be measuring.

Comment 3

[Mark Finkle (:mfinkle) (use needinfo?)]

agreed