The non-virtual test is also slow. And also on other platforms, apparently, since the non-virtual test even has an entry in LayoutTests/SlowTests

This test was added while fixing  bug 445194 , which seems to have exchanged something that used to crash with something that's just slow.

Is it fine to keep slow tests around like this? Can the test be removed or improved? Can the code be improved?

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/cce7fee0c26ee7229faa7501ab2d371cb8d8ad9e

commit cce7fee0c26ee7229faa7501ab2d371cb8d8ad9e
Author: Morten Stenshorne <mstensho@chromium.org>
Date: Mon Jan 15 14:33:04 2018

Mark a virtual focus-controller-recursion-crash.html test as slow.

There may be other virtual versions out there as well, but this one
triggered a timeout on a bot:
https://ci.chromium.org/buildbot/chromium.webkit/WebKit%20Mac10.11%20(dbg)/12058

TBR=kochi@chromium.org

Bug:  445194 ,  802029 
Change-Id: Ia5f0ccd589b167d3d8f2589fc14b1a74e67412bb
Reviewed-on: https://chromium-review.googlesource.com/866774
Reviewed-by: Morten Stenshorne <mstensho@chromium.org>
Commit-Queue: Morten Stenshorne <mstensho@chromium.org>
Cr-Commit-Position: refs/heads/master@{#529274}
[modify] https://crrev.com/cce7fee0c26ee7229faa7501ab2d371cb8d8ad9e/third_party/WebKit/LayoutTests/SlowTests

Is this observed on Mac only?
It may be due to recently added code for Android...
Adding Ajith.

FYI issue 787759 is the tracking bug for the perf.

I only saw the Timeout on Mac, but it's super-slow everywhere, so it might happen anywhere, I suppose.

Let me close this bug as a duplicate to 787759.