1634258 - 2.89 - 4.18% ts_paint / ts_paint_webext (linux64-shippable-qr) regression on push 2760b8a14f17ac39b2b6cc1ac2d63f2f1d05bb77 (Mon April 27 2020)

Status: RESOLVED WONTFIX

(Core :: Graphics: WebRender, defect, P3)

Description

[Alexandru Ionescu (needinfo me) [:alexandrui]]

Perfherder has detected a talos performance regression from push 2760b8a14f17ac39b2b6cc1ac2d63f2f1d05bb77. As author of one of the patches included in that push, we need your help to address this regression.

Regressions:

4% ts_paint linux64-shippable-qr opt e10s stylo 257.00 -> 267.75
4% ts_paint_webext linux64-shippable-qr opt e10s stylo 256.83 -> 267.42
3% ts_paint_webext linux64-shippable-qr opt e10s stylo 259.67 -> 267.17

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) will be backed out in accordance with our regression policy.

For more information on performance sheriffing please see our FAQ.

Comment 1

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

Set release status flags based on info from the regressing bug 1632795

Comment 2

[Jamie Nicol [:jnicol]]

Bug 1632795 made it so that we actually do a render when the OS/window manager requests it. Previously we were ignoring that request, which was incorrect.

The specific effect in these tests is an additional render when the window is first created. I have verified there are not extra ones after that. This render doesn't actually draw anything other than clearing the screen, but that still takes a bit of time as there is bookkeeping to do.

I think we just need to live with this regression, the correctness is worth the hit. So please do not back out the patches.