1407984 - 1.87 - 4.77% tp5o_webext (linux64, windows10-64, windows7-32) regression on push 7a4a9b65e2e773ea74d5485e6ddbf97ab3160252 (Thu Oct 12 2017)

Status: RESOLVED WONTFIX

(WebExtensions :: Request Handling, defect)

Description

[Ionuț Goldan [:igoldan]]

Talos has detected a Firefox performance regression from push:

https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?changeset=7a4a9b65e2e773ea74d5485e6ddbf97ab3160252

As author of one of the patches included in that push, we need your help to address this regression.

Regressions:

  5%  tp5o_webext summary windows10-64 pgo e10s     282.10 -> 295.54
  4%  tp5o_webext summary windows7-32 pgo e10s      278.97 -> 291.17
  2%  tp5o_webext summary linux64 pgo e10s          349.09 -> 356.45
  2%  tp5o_webext summary windows7-32 opt e10s      377.99 -> 385.07


You can find links to graphs and comparison views for each of the above tests at: https://treeherder.mozilla.org/perf.html#/alerts?id=9975

On the page above you can see an alert for each affected platform as well as a link to a graph showing the history of scores for this test. There is also a link to a treeherder page showing the Talos jobs in a pushlog format.

To learn more about the regressing test(s), please see: https://wiki.mozilla.org/Buildbot/Talos/Tests

For information on reproducing and debugging the regression, either on try or locally, see: https://wiki.mozilla.org/Buildbot/Talos/Running

*** Please let us know your plans within 3 business days, or the offending patch(es) will be backed out! ***

Our wiki page outlines the common responses and expectations: https://wiki.mozilla.org/Buildbot/Talos/RegressionBugsHandling

Comment 1

[Ionuț Goldan [:igoldan]]

:kmag I see you own bug 1404652. Can we fix these performance regressions? Or should we consider backout/accepting them?

Comment 2

[Kris Maglione [:kmag]]

This is somewhat expected. We're essentially trading better responsiveness for somewhat higher request latency. The latency shows up in the tp5o numbers, but is generally not visible to users. The difference in responsiveness, however, can be significant.

So I suggest we accept these regressions.