1870869 - 13.66 - 6.8% canvas2dvideo Mean tick time across 100 ticks: / canvas2dvideo Mean tick time across 100 ticks: + 1 more (Windows) regression on Thu December 7 2023

Status: NEW

(Core :: Graphics: Canvas2D, defect, P2)

Description

[Andrej (:aglavic)]

Perfherder has detected a talos performance regression from push 71b6060b60151e157e94110d28fb51d76bb6415a. 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)
14%	canvas2dvideo Mean tick time across 100 ticks:	windows10-64-ref-hw-2017-qr	e10s fission stylo webgl-ipc webrender	47.47 -> 53.95
13%	canvas2dvideo Mean tick time across 100 ticks:	windows10-64-ref-hw-2017-qr	e10s fission stylo webgl-ipc webrender	47.82 -> 53.95
7%	canvas2dvideo Mean tick time across 100 ticks:	windows10-64-shippable-qr	e10s fission stylo webgl-ipc webrender	63.45 -> 67.77

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 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

[Andrej (:aglavic)]

I am sorry for the rather late filing of this bug, we were dealing with a few days worth of windows reference hardware timing out on us. We are aware it is holiday time, the 3 business day deadline can be extended to January 8th (monday of the second week of January) as a time to let us know your plans

Comment 2

[Andrew Osmond [:aosmond] (he/him)]

As the recent author and maintainer of the test, let's not back out over this. I have patches in the works that will dramatically improve the performance of this test, anticipated to be far more than the regression. Bob's patches solve many stability problems associated with D2D canvas.

Comment 3

[Andrew Osmond [:aosmond] (he/him)]

I would also say that because we haven't had performance tests measuring D2D canvas in this scenario until recently, we ended up with a design that was truly unsuitable. The performance is already abysmal. With a proper fix, it should running at 1/4 to 1/3 of the current benchmark.

Comment 4

[BugBot [:suhaib / :marco/ :calixte]]

Set release status flags based on info from the regressing bug 1863914

:bobowen, since you are the author of the regressor, bug 1863914, could you take a look? Also, could you set the severity field?

For more information, please visit BugBot documentation.

Comment 5

[Brad Werth [:bradwerth] | Out until March 3]

Tracking Bug, so S2.

Comment 6

[Brad Werth [:bradwerth] | Out until March 3]

And assigning to Andrew so our S2s have an owner.

Comment 8

[Donal Meehan [:dmeehan]]

Removing tracking and setting to fix-optional.
Got info that this is a new talos canvas test that was just recently landed.