The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/b650e659c291030324aebd4cf50c003ad652bd98

commit b650e659c291030324aebd4cf50c003ad652bd98
Author: tkent <tkent@chromium.org>
Date: Wed May 24 00:01:04 2017

Update TestExpectatinos for  crbug.com/725016 

BUG= 725016 ,725542,725545,725548
TBR=cfroussios@chromium.org
NOTRY=true

Review-Url: https://codereview.chromium.org/2899193002
Cr-Commit-Position: refs/heads/master@{#474112}

[modify] https://crrev.com/b650e659c291030324aebd4cf50c003ad652bd98/third_party/WebKit/LayoutTests/TestExpectations

Any idea whats up?

I put together some builds on my Z840 desktop and ran the test using the command below, with different trees.
time third_party/WebKit/Tools/Scripts/run-webkit-tests -t Default storage/indexeddb/objectstore-keycursor.html

Resulting times are below. 
~2.4s release
~2.7s release + DCHECKs
 ~10s debug

Somewhat surprisingly, I also got about 10 seconds from a debug build on a MacBookPro13,3 (15-inch 2016).

The failure at https://build.chromium.org/p/chromium.webkit/builders/WebKit%20Win7%20%28dbg%29 (WebKit Win7 Dbg) referenced in  Issue 725016  shows an 18 second timeout. Given the 10-second time on a quiet computer, exceeding 18 seconds on test bots doesn't seem crazy to me.

The time on a release build with DCHECKs enabled is close to the time on a vanilla release build, which makes me think it wouldn't be fruitful to mess with DCHECKs and hope to shave off some time. Seems like storage/indexeddb/objectstore-keycursor.html is just a slow test.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/fab65e15673fe3f3825b883238d7be20c32e4962

commit fab65e15673fe3f3825b883238d7be20c32e4962
Author: Victor Costan <pwnall@chromium.org>
Date: Fri Jul 07 02:11:58 2017

Mark storage/indexeddb/objectstore-keycursor.html as slow.

The IndexedDB layout test was previously maked as timing out. However,
the test is slowed down a lot on Debug builds, as shown by the times
below, which were obtained on a HP Z840 running Ubuntu 14.04.

Z840 Time  Build Type
    ~2.4s  Debug
    ~2.7s  Release with DCHECKs enabled
     ~10s  Release

BUG=725548

Change-Id: I4ca2cdf7c728dd0f256e97259cb9e8fb4537a020
Reviewed-on: https://chromium-review.googlesource.com/560546
Commit-Queue: Victor Costan <pwnall@chromium.org>
Reviewed-by: Joshua Bell <jsbell@chromium.org>
Cr-Commit-Position: refs/heads/master@{#484794}
[modify] https://crrev.com/fab65e15673fe3f3825b883238d7be20c32e4962/third_party/WebKit/LayoutTests/SlowTests
[modify] https://crrev.com/fab65e15673fe3f3825b883238d7be20c32e4962/third_party/WebKit/LayoutTests/TestExpectations

Not closing this bug, because the test is now listed in SlowTests. At the same time, I don't see any immediate solution, so I'm going to mark this as Available.

I marked the bug as OS-independent, because the large difference between debug and optimized builds is visible on Linux and Mac, so it doesn't seem to be Windows-specific.

Thanks for investigating this Victor :)

going to mark as test disabled even though it's just slow