1799488 - 18.66% microsoft loadtime (OSX) regression on Sat October 22 2022

Status: RESOLVED INVALID

(Core :: DOM: Forms, defect)

Description

[Alex Finder]

Perfherder has detected a browsertime performance regression from push 8baa19d12931bdf9b5d61ea4f06a9d6b8e214d81. As author of one of the patches included in that push, we need your help to address this regression.

Regressions:

Ratio	Test	Platform	Options	Absolute values (old vs new)
19%	microsoft loadtime	macosx1015-64-shippable-qr	cold fission webrender	256.96 -> 304.92

Details of the alert can be found in the alert summary, including links to graphs and comparisons for each of the affected tests. Please follow our guide to handling regression bugs and let us know your plans within 3 business days, or the offending patch(es) may be backed out in accordance with our regression policy.

If you need the profiling jobs you can trigger them yourself from treeherder job view or ask a sheriff to do that for you.

For more information on performance sheriffing please see our FAQ.

Comment 1

[BugBot [:suhaib / :marco]]

Set release status flags based on info from the regressing bug 1795887

Comment 2

[Jim Mathies [:jimm]]

Attachment: image.png

Hard to say that this landing caused this. It's a very noisy test that seems to have gotten noisier around the landing, but I wouldn't say blame is definitive. Also the 19% regression is questionable.

Comment 3

[Adam Vandolder [:avandolder]]

I'm really unsure of how this regression could've been caused by this patch; the change should be pretty much trivial.

Comment 4

[Sean Feng [:sefeng211]]

Hi Alex,

Adam's patch isn't in the regression range you posted in the description, so we are a bit confused here. Also, as Jim noted, the graph was noisy and looks like something was improved around Oct 29.

Can you clarify those for us? Thanks!

Comment 5

[Alex Finder]

There is a visible difference between revisions 8baa19d12931b and a374e40ad4e2f, making bug 1795887 as the actual culprit. Regression seems to have been resolved around d1982cee06ca5. Backfilled range d1982cee06ca5 - caddc49839d30 to see if there is still a regression visible after that, or it's just noise.

Comment 6

[Alex Finder]

Closing bug and alert as invalid due to noise. Sorry for the confusion!