Defer a couple of svg layout tests.

webkit/tools/layout_tests/test_lists/tests_fixable.txt

 // These tests are expected to fail until we get around to fixing
 // them.
 
 // Bug 1124548: Copying with no selection is sometimes supposed to work
 // Also skipped by Apple on Windows (rdar://problem/5015941)
 DEFER : LayoutTests/editing/execCommand/copy-without-selection.html = FAIL
 
 // onload race condition due to poorly designed test.
 // Works fine when run stand-alone.  Not needed for Beta.
 // Also skipped by Apple on Windows, due to intermittent failure
@@ skipping 388 matching lines @@
 
 // SVG text fills not working. Note that linux baseline is wrong.
 LINUX : LayoutTests/svg/custom/pointer-events-text.svg = FAIL
 
 // Marking this again as failed. This test failed even after rebaseline due to
 // the fact that the font is not specified in the test and it seems that the
 // test falls back on a different font locally than on the buildbot so the
 // baseline is invalid. This might indicate a failure in our font fallback code
 // or might be expected and that we should either go with the buildbot output or
 // change the test upstream to specify explicitly what font to use.
-LINUX WIN : LayoutTests/fast/backgrounds/svg-as-background-6.html = FAIL
+// http://crbug.com/3244 : The mask part of this test is totally wrong.
+DEFER LINUX WIN : LayoutTests/fast/backgrounds/svg-as-background-6.html = FAIL
 
 // BUG: 992321
 // The failures listed below have never passed in Chromium and need to be
 // investigated, categorized, and (one hopes) fixed.
 
 // The following tests fail because SVG animation is not yet implemented
 DEFER LINUX WIN : LayoutTests/svg/W3C-SVG-1.1/animate-elem-78-t.svg = FAIL
 
 // These test fail full text diff (but not simplified diff) most likely due
 // to differing implementations of SVG fonts. They may or may not represent real
@@ skipping 145 matching lines @@
 LINUX : LayoutTests/svg/dynamic-updates/SVGEllipseElement-svgdom-cx-prop.html = PASS CRASH
 
 // WebKit Merge 38950:39000 failures:
 LINUX : LayoutTests/svg/custom/js-late-gradient-and-object-creation.svg = FAIL
 
 // Linux pixeltest failure: slightly different shade of composited
 // alpha in bottom group.
 LINUX : LayoutTests/svg/W3C-SVG-1.1/masking-opacity-01-b.svg = FAIL
 
 // Fixed upstream, see https://bugs.webkit.org/show_bug.cgi?id=21910.
-// However, we don't seem to draw the "eyeballs" at the correct location.
-LINUX WIN : LayoutTests/fast/backgrounds/svg-as-background-3.html = FAIL
+// However, we need to support focal point for radial gradients (see
+// http://crbug.com/8696)
+DEFER LINUX WIN : LayoutTests/fast/backgrounds/svg-as-background-3.html = FAIL
 
 //
 // -----------------------------------------------------------------
 // End of SVG tests
 // -----------------------------------------------------------------
 
 // This fails because of differences between how CG and Skia tests for
 // a point in a path. See
 // http://code.google.com/p/chromium/issues/detail?id=7465 for details.
 DEFER LINUX WIN : LayoutTests/fast/canvas/pointInPath.html = FAIL
@@ skipping 1463 matching lines @@
 LINUX : LayoutTests/tables/mozilla/bugs/bug9123-1.html = FAIL
 LINUX : LayoutTests/tables/mozilla/collapsing_borders/bug41262-3.html = FAIL
 LINUX : LayoutTests/tables/mozilla/marvin/backgr_index.html = FAIL
 LINUX : LayoutTests/tables/mozilla/marvin/backgr_layers-opacity.html = FAIL
 LINUX : LayoutTests/tables/mozilla_expected_failures/bugs/bug1010.html = FAIL
 LINUX : LayoutTests/tables/mozilla_expected_failures/bugs/bug23847.html = FAIL
 LINUX : LayoutTests/tables/mozilla_expected_failures/bugs/bug32205-4.html = FAIL
 LINUX : LayoutTests/tables/mozilla_expected_failures/marvin/backgr_fixed-bg.html = FAIL
 LINUX : LayoutTests/tables/mozilla_expected_failures/marvin/backgr_layers-show.html = FAIL