1008965 - 5.4% tresize regression for windows 7/8 on May 12 (rev 02eb695c8640) on inbound (fx32)

Status: RESOLVED WONTFIX

(Core :: Graphics, defect)

Description

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

Talos alerts are doing their job, here is a graph to show what is happening:
http://graphs.mozilla.org/graph.html#tests=[[254,131,31],[254,131,25]]&sel=1399299427035,1399904227035&displayrange=7&datatype=running

here is the pushlog that shows the problem:
http://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?changeset=02eb695c8640

I did some retriggers, there is not any issue with noise in the test:
https://tbpl.mozilla.org/?tree=Mozilla-Inbound&jobname=WINNT%206.2%20mozilla-inbound%20talos%20chromez

Comment 1

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

Matt- can you comment here on what patch is suspect and what options we have (new feature vs perf, investigate some better ways to do this, back it out, wait for bug XYZ, etc.)

Comment 2

[Matt Woodrow (:mattwoodrow)]

http://hg.mozilla.org/integration/mozilla-inbound/rev/d1f0df851145 seems like the only one that could possibly have affected windows.

I'll do some try pushes to try confirm. https://tbpl.mozilla.org/?tree=Try&rev=b6f9a4e64972

Comment 3

[Matt Woodrow (:mattwoodrow)]

Confirmed that this is the first part of d1f0df851145.

This is code that was accidentally disabled during the conversion to Moz2D. I suspect we got a corresponding win at some point and didn't attribute it correctly.

Not much we can do here.

Comment 4

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

it looks like we say a win on May 8th from http://hg.mozilla.org/integration/mozilla-inbound/rev/d1155f994a3a (bug 938395).  This removes some of the win.  Looking back for a full week, we have an overall win which is a good thing :)

I vote for wontfix

Comment 5

[Avi Halachmi (:avih)]

(In reply to Matt Woodrow (:mattwoodrow) from comment #3)
> Not much we can do here.

Thanks.