Allow proper highlighting on universal overflow scroll.

LayoutTests/TestExpectations

Index: LayoutTests/TestExpectations
diff --git a/LayoutTests/TestExpectations b/LayoutTests/TestExpectations
index d30fd1c4a42155a2b91c67b0062aa2def83029da..fee8522e145aa62073c8594882ed97b89608701f 100644
--- a/LayoutTests/TestExpectations
+++ b/LayoutTests/TestExpectations
@@ -181,6 +181,10 @@ crbug.com/318978 [ Linux ] permissionclient/plugin-permission.html [ Skip ]
 crbug.com/318978 [ Linux ] plugins/ [ Skip ]
 crbug.com/318978 [ Linux ] virtual/softwarecompositing/plugins/invalidate_rect.html [ Skip ]
 
+Bug(hartmanng) compositing/overflow/selection-gaps.html [ NeedsRebaseline ]
+Bug(hartmanng) virtual/softwarecompositing/overflow/selection-gaps.html [ NeedsRebaseline ]
+Bug(hartmanng) virtual/gpu/compositedscrolling/overflow/selection-gaps.html [ NeedsRebaseline ]

[Ian Vollick, 2014/05/31 01:17:36]

I wonder if we could make this a ref test, and in the -expected.html, we could
make the background opaque white rather than turning off accelerated overflow
scrolling. That way we should have the same stuff being composited and we
shouldn't (hopefully) run into the same AA-related pixel diffs.

[hartmanng, 2014/06/02 20:23:58]

I like it. Done.

+
 # Hits an assert on Snow-leopard Debug but not on any other platforms.
 crbug.com/347186 [ Snowleopard Debug ] inspector/sources/debugger/async-callstack-mutation-observer.html [ Skip ]
 crbug.com/347186 [ Snowleopard Debug ] inspector/sources/debugger/async-callstack.html [ Skip ]