Make error-handling for feColorMatrix consistent

third_party/WebKit/LayoutTests/TestExpectations

Index: third_party/WebKit/LayoutTests/TestExpectations
diff --git a/third_party/WebKit/LayoutTests/TestExpectations b/third_party/WebKit/LayoutTests/TestExpectations
index 9d28191382c5441ed5d096f613299e8848558616..e6726734c10ded814b7bc36561492c4d397f45dd 100644
--- a/third_party/WebKit/LayoutTests/TestExpectations
+++ b/third_party/WebKit/LayoutTests/TestExpectations
@@ -174,6 +174,14 @@ crbug.com/463358 [ Mac Linux Debug ] fast/backgrounds/transformed-body-backgroun
 crbug.com/463358 [ Mac Linux Debug ] svg/W3C-SVG-1.1/paths-data-02-t.svg [ ImageOnlyFailure ]
 crbug.com/463358 [ Mac Linux Debug ] css3/masking/clip-path-polygon.html [ ImageOnlyFailure ]
 
+crbug.com/534763 css3/filters/effect-reference-hw.html [ NeedsRebaseline ]

[Stephen White, 2015/09/24 16:19:57]

How are these tests affected, BTW? Did they have error conditions in their
feColorMatrix elements? If so, could we fix them instead?

[fs, 2015/09/24 16:30:33]

There's no longer an (explicit) identity matrix in the DRT output, so all of
these are basically:

-[feColorMatrix type=MATRIX values="1.00 0.00 ..."]
+[feColorMatrix type=MATRIX]

Which stems from <feColorMatrix>es without a type (or values).

So no error condition - just different dumping of default values.

+crbug.com/534763 css3/filters/effect-reference-subregion-hw.html [ NeedsRebaseline ]
+crbug.com/534763 css3/filters/effect-reference-subregion.html [ NeedsRebaseline ]
+crbug.com/534763 css3/filters/effect-reference-tile-hw.html [ NeedsRebaseline ]
+crbug.com/534763 css3/filters/effect-reference-tile.html [ NeedsRebaseline ]
+crbug.com/534763 css3/filters/effect-reference.html [ NeedsRebaseline ]
+crbug.com/534763 svg/filters/feColorMatrix-invalid-animation.svg [ NeedsRebaseline ]
+
 crbug.com/267206 [ Mac ] virtual/rootlayerscrolls/fast/scrolling/scrollbar-tickmarks-hittest.html [ Timeout ]
 
 crbug.com/535141 fast/text/international/combining-marks-position.html [ NeedsRebaseline ]