1568459 - 9.36% perf_reftest_singletons style-attr-1.html (windows7-32-shippable) regression on push d45068799783935d7a110307ca9430159d6909d4

Status: RESOLVED WONTFIX

(Firefox Build System :: General, defect)

Description

[Ionuț Goldan [:igoldan]]

Talos has detected a Firefox performance regression from push:

https://hg.mozilla.org/integration/autoland/pushloghtml?fromchange=7f45ff373dff62e961476c4998601c9e04bf7237&tochange=d45068799783935d7a110307ca9430159d6909d4

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

Regressions:

9% perf_reftest_singletons style-attr-1.html windows7-32-shippable opt e10s stylo 2.98 -> 3.26

You can find links to graphs and comparison views for each of the above tests at: https://treeherder.mozilla.org/perf.html#/alerts?id=22038

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/Performance_sheriffing/Talos/Tests

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

Comment 1

[Ionuț Goldan [:igoldan]]

I'm pretty sure this a broken PGO issue, as I only see this on Windows 7. Other platforms aren't at all affected.
:glandium I filed this only FYI. But from my experience, we can close this as WONTFIX. I expect this to get fixed by itself, in the following days/weeks.

Comment 2

[Mike Hommey [:glandium]]

I think it's a plausible regression, due to the reduction in number of registers available, and maybe the addition of code to handle the frame pointer, that has some impact in a tight loop or something. I'm afraid there's not much we can do here. It's worth noting that this is kind of equivalent to what happened when we enable frame pointers on win32 for C++ in bug 1322735. It's also worth noting that this change improves crash and hang reports where rust code is involved somehow. So... wontfix? Eric what do you think?

Comment 3

[Eric Rahm [:erahm]]

I'm okay with a 0.28ms regression on a style stress test on Win32 in order to get frame pointers enabled.