1579391 - 2.95 - 3.05% build times (linux32-shippable) regression on push 43e96d7429198eecbbbb6241c0fc0cddc8f41442 (Wed August 28 2019)

Status: RESOLVED DUPLICATE of bug 1577747

(Core :: DOM: Core & HTML, defect)

Description

[Marian Raiciof [:marauder]]

We have detected a build metrics regression from push:

https://hg.mozilla.org/integration/autoland/pushloghtml?changeset=43e96d7429198eecbbbb6241c0fc0cddc8f41442

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

Regressions:

3% build times linux32-shippable opt nightly taskcluster-c4.4xlarge 3,897.42 -> 4,016.12
3% build times linux32-shippable opt nightly taskcluster-c4.4xlarge 3,898.07 -> 4,013.07

You can find links to graphs and comparison views for each of the above tests at: https://treeherder.mozilla.org/perf.html#/alerts?id=22976

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 jobs in a pushlog format.

To learn more about the regressing test(s), please see: https://developer.mozilla.org/en-US/docs/Mozilla/Performance/Automated_Performance_Testing_and_Sheriffing/Build_Metrics

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

Comment 1

[Emilio Cobos Álvarez (:emilio)]

I guess I can look into removing useless includes or what not, but it's unclear to me otherwise how the regressing patch could affect build times. How sure are we of it being the regressing bug?

But this is only linux32 build times so maybe not a huge deal?

Mike, should I spend time looking into this other than more generally removing redundant includes? My patch moved some code from a virtual interface implementation to some inline functions in nsIContentInlines.h (which is not a terribly included file) which only get called in two translation units.

Comment 2

[Mike Hommey [:glandium]]

I don't know why there weren't alerts on the other linux build types, but this is the same regression as bug 1577747: cranelift being reenabled.