[Webkit-unassigned] [Bug 186129] REGRESSION (r232186): Hardware-accelerated CSS animations using steps() timing function no longer work

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Tue Jul 3 03:16:46 PDT 2018


https://bugs.webkit.org/show_bug.cgi?id=186129

Frédéric Wang (:fredw) <fred.wang at free.fr> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |ASSIGNED
           Assignee|webkit-unassigned at lists.web |fred.wang at free.fr
                   |kit.org                     |

--- Comment #12 from Frédéric Wang (:fredw) <fred.wang at free.fr> ---
(In reply to Frédéric Wang (:fredw) from comment #11)
> Note that in the former case, the AnimationBase becomes non-accelerated when
> startAnimation returns false (e.g. when animationCanBeAccelerated returns
> false) but in the latter case, the WebAnimation just ignores the boolean
> returned by startAnimation. It looks like it should somewhat fallback to
> non-accelerated mode too.

OK, just checked again this morning and indeed this seems to be enough:

--- a/Source/WebCore/animation/KeyframeEffectReadOnly.cpp
+++ b/Source/WebCore/animation/KeyframeEffectReadOnly.cpp
@@ -1256,3 +1256,6 @@ void KeyframeEffectReadOnly::applyPendingAcceleratedActions()
         case AcceleratedAction::Play:
-            compositedRenderer->startAnimation(timeOffset, backingAnimationForCompositedRenderer().ptr(), m_blendingKeyframes);
+            if (!compositedRenderer->startAnimation(timeOffset, backingAnimationForCompositedRenderer().ptr(), m_blendingKeyframes)) {
+                m_shouldRunAccelerated = false;
+                m_lastRecordedAcceleratedAction = AcceleratedAction::Stop;
+            }
             break;

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20180703/5b4f2c79/attachment.html>


More information about the webkit-unassigned mailing list