--- /home/slave/webkitgtk/gtk-linux-64-release-tests/build/layout-test-results/animations/transition-and-animation-1-expected.txt +++ /home/slave/webkitgtk/gtk-linux-64-release-tests/build/layout-test-results/animations/transition-and-animation-1-actual.txt @@ -1,3 +1,3 @@ This test has a transition and animation on the same property (-webkit-transform). But the transition is never triggered, so nothing should be moving when the animation finishes. -PASS - "webkitTransform" property for "box" element at 0.55s saw something close to: none +FAIL - "webkitTransform" property for "box" element at 0.55s expected: none but saw: matrix(1, 0, 0, 1, 200.1999969482422, 0) The stderr is more interesting, failed to create drawable This was caused by r233164 from bug 186981, diff --git a/Source/WebCore/animation/KeyframeEffectReadOnly.cpp b/Source/WebCore/animation/KeyframeEffectReadOnly.cpp index 96396956a9d..caf57bcd202 100644 --- a/Source/WebCore/animation/KeyframeEffectReadOnly.cpp +++ b/Source/WebCore/animation/KeyframeEffectReadOnly.cpp @@ -1164,6 +1164,9 @@ void KeyframeEffectReadOnly::updateAcceleratedAnimationState() if (!m_shouldRunAccelerated) return; + if (!renderer()) + return; + auto localTime = animation()->currentTime(); // If we don't have a localTime or localTime < 0, we either don't have a start time or we're before the startTime
Also the layout test added in that commit fails on GTK, webanimations/opacity-animation-yields-compositing-span.html --- /home/slave/webkitgtk/gtk-linux-64-release-tests/build/layout-test-results/webanimations/opacity-animation-yields-compositing-span-expected.txt +++ /home/slave/webkitgtk/gtk-linux-64-release-tests/build/layout-test-results/webanimations/opacity-animation-yields-compositing-span-actual.txt @@ -8,7 +8,7 @@ (children 1 (GraphicsLayer (position 8.00 8.00) - (bounds 25.00 18.00) + (bounds 25.00 17.00) (opacity 0.50) (drawsContent 1) )
This test is now timing out, see webkit.org/b/187264
animations/transition-and-animation-1.html is now passing since r233512