Findit identified the culprit r617828 as introducing flaky test(s)
summarized in https://findit-for-me.appspot.com/waterfall/flake/flake-culprit?key=ag9zfmZpbmRpdC1mb3ItbWVyQwsSDEZsYWtlQ3VscHJpdCIxY2hyb21pdW0vMmZkZGJiZjQxMjQ5MTIzMzc3YTNjMWJjYWQ4NTRlMzUyNjNmNTlkMQw

Please revert the culprit or disable the test(s) asap. If you are the owner,
please fix!

If the culprit above is wrong, please file a bug using this link:
https://bugs.chromium.org/p/chromium/issues/entry?status=Unconfirmed&labels=Pri-1,Test-Findit-Wrong&components=Tools%3ETest%3EFindit%3EFlakiness&summary=%5BFindit%5D%20Flake%20Analyzer%20-%20Wrong%20culprit%20r617828&comment=Link%20to%20Culprit%3A%20https://findit-for-me.appspot.com/waterfall/flake/flake-culprit?key=ag9zfmZpbmRpdC1mb3ItbWVyQwsSDEZsYWtlQ3VscHJpdCIxY2hyb21pdW0vMmZkZGJiZjQxMjQ5MTIzMzc3YTNjMWJjYWQ4NTRlMzUyNjNmNTlkMQw

Automatically posted by the findit-for-me app (https://goo.gl/Ot9f7N).

CL is pretty old. Kyle, could you please take a look?

Findit identified the culprit r617828 as introducing flaky test(s)
summarized in https://findit-for-me.appspot.com/waterfall/flake/flake-culprit?key=ag9zfmZpbmRpdC1mb3ItbWVyQwsSDEZsYWtlQ3VscHJpdCIxY2hyb21pdW0vMmZkZGJiZjQxMjQ5MTIzMzc3YTNjMWJjYWQ4NTRlMzUyNjNmNTlkMQw

Please revert the culprit or disable the test(s) asap. If you are the owner,
please fix!

If the culprit above is wrong, please file a bug using this link:
https://bugs.chromium.org/p/chromium/issues/entry?status=Unconfirmed&labels=Pri-1,Test-Findit-Wrong&components=Tools%3ETest%3EFindit%3EFlakiness&summary=%5BFindit%5D%20Flake%20Analyzer%20-%20Wrong%20culprit%20r617828&comment=Link%20to%20Culprit%3A%20https://findit-for-me.appspot.com/waterfall/flake/flake-culprit?key=ag9zfmZpbmRpdC1mb3ItbWVyQwsSDEZsYWtlQ3VscHJpdCIxY2hyb21pdW0vMmZkZGJiZjQxMjQ5MTIzMzc3YTNjMWJjYWQ4NTRlMzUyNjNmNTlkMQw

Automatically posted by the findit-for-me app (https://goo.gl/Ot9f7N).

The flaky failures have the test output image going from a green square to nothing, so the offscreen canvas surface probably isn't included in the display frame. I'll take a look at how the test is (or isn't) ensuring things have drawn before grabbing the image output.

It looks like this test was disabled before and only got enabled in https://crrev.com/c/1342801 3 days ago. 

The test finishes as soon as the offscreen canvas has green pixels. Blinks has the right state at that point but it doesn't necessarily mean the display compositor has drawn the green pixels? The offscreen canvas still to ensure it's submit a CompositorFrame, that the display compositor has run Display::DrawAndSwap() which picked up the offscreen CompositorFrame and the output was drawn to the window.

It looks like it's pretty common for the other web tests call requestAnimationFrame() twice. I wonder if adding changing "resolve();" to "requestAnimationFrame(resolve);" at [1] would fix this flake? aaronhk@ does that sound right/wrong to you?

The failure doesn't reproduce locally on my macbook so I can't test the theory easily.

[1] https://cs.chromium.org/chromium/src/third_party/blink/web_tests/printing/offscreencanvas-2d-printing.html?l=20&rcl=dca578643994678e047bdcdfe1f63fd3548afc04

I ran https://crrev.com/c/1405540 through the mac bots multiple times and didn't see anymore flake from this test. That is promising although the fix is super speculative. Will try to land and see if that impacts flake.

https://test-results.appspot.com/dashboards/flakiness_dashboard.html#testType=webkit_layout_tests&tests=printing%2Foffscreencanvas-2d-printing.html

printing/offscreencanvas-2d-printing.html was globally disabled to land mojo changes in https://crrev.com/c/1145692, see https://crbug.com/921719. The speculative fix I had isn't enough to ensure the pixels are actually drawn after mojo changes so I'm abandoning it. Enabling mac OOP-D just exposed the existing race and the test is disabled so I'm going to close this.