Disabled the test in crbug.com/883463.

Dave,

As owner of the test, could you please take a look?

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/96b9bc1d1e1357608c2f4f57c82ea23d586a8a55

commit 96b9bc1d1e1357608c2f4f57c82ea23d586a8a55
Author: Ramin Halavati <rhalavati@chromium.org>
Date: Wed Jan 24 15:36:10 2018

Disable flaky test: MouseMove.

MainThreadEventQueueBrowserTest.MouseMove on Win7 sometimes results in
the following error:
ERROR:delegated_frame_host.cc(230)] NOT IMPLEMENTED

It is excluded from vis unit tests as viz does not support
SetNeedsBeginFrames.

Bug: 805419
Change-Id: Ifb490ef07785a91b9602c4cd8f0b3c5e97a42ba0
TBR: dtapuska
Reviewed-on: https://chromium-review.googlesource.com/883463
Commit-Queue: Ramin Halavati <rhalavati@chromium.org>
Reviewed-by: Dave Tapuska <dtapuska@chromium.org>
Cr-Commit-Position: refs/heads/master@{#531537}
[modify] https://crrev.com/96b9bc1d1e1357608c2f4f57c82ea23d586a8a55/testing/buildbot/filters/viz.content_browsertests.filter

 Issue 803872  has been merged into this issue.

jonross@ are you working on viz's unit tests?

Is there a range for these flakes appearing?

The delegated_frame_host message is expected and not a sign of an error.

The actual error in the test is that expected values aren't correct:

../../content/browser/renderer_host/input/main_thread_event_queue_browsertest.cc(139): error: Expected equality of these values:
  20
  last_mouse_x
    Which is: 116
../../content/browser/renderer_host/input/main_thread_event_queue_browsertest.cc(140): error: Expected equality of these values:
  25
  last_mouse_y
    Which is: 61

Since the test is not timing out, and since it is receiving its data, nothing initially stands out as the viz test suite not delivering information correctly.

Could the timing changes introduced by Viz be exposing a flake within this test? It has been disabled in the past for flaking in non-viz environments.

Sending to riajiang@

This is another test reliant on InputEventAcks, where the result is not what is expected in Viz. It will likely be cleared by the hit test work that you are already working on.

Removing myself.

This test is currently disabled on Windows generally not just in Viz, due to flakiness see  issue 805666 .

I've ran it locally on Windows with both normal and Viz modes and couldn't reproduce, so it may be fixed by now.

Either way this is no longer blocking OOP-D