The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/bb2a1b3b1f328c3e64f49e59e30aef7ba30c05b6

commit bb2a1b3b1f328c3e64f49e59e30aef7ba30c05b6
Author: Ria Jiang <riajiang@chromium.org>
Date: Tue Feb 27 03:05:16 2018

Fix SitePerProcessHitTestBrowserTest.SurfaceHitTestPointerEventsNone.

In Viz hit-test, when we set "pointer-events: none;" to an OOPIF, we
fall back to the slow-path targeting because all OOPIFs have
kHitTestAsk flag set, which means we should still get the right target
even though HitTestQuery doesn't know about "pointer-events: none;" at
all.

This CL adds an InputEventAckWaiter in the test to wait for async
targeting results. We didn't have to wait in current targeting because
SurfaceHittest skips OOPIFs with "pointer-events: none;". If there's
a regression (this case should be rare), we can have HitTestQuery
asking RenderWidgetHostInputEventRouter for this information like
what SurfaceHittest is doing right now.

Bug: 812012
Test: SitePerProcessHitTestBrowserTest.SurfaceHitTestPointerEventsNone/1
Change-Id: Ia13979750a87a37fd4a65b98f5951f11c097148e
Reviewed-on: https://chromium-review.googlesource.com/938483
Reviewed-by: Avi Drissman <avi@chromium.org>
Reviewed-by: Sadrul Chowdhury <sadrul@chromium.org>
Commit-Queue: Ria Jiang <riajiang@chromium.org>
Cr-Commit-Position: refs/heads/master@{#539385}
[modify] https://crrev.com/bb2a1b3b1f328c3e64f49e59e30aef7ba30c05b6/content/browser/site_per_process_hit_test_browsertest.cc

The InputEventAckWaiter added in #2 does not work in Viz. It currently times out having never seen the event.

I'm going to try to get Viz hit-test working with OOP-D next week - I can check if that's going to fix this test case then; or re-enable it if disabling this test for Viz right now is more appropriate?

The waiter does work now. There are some known failures being tracked in  issue 848348 .

No longer blocking Viz testing.