Bug 215529 – Scrolling sync changes in r261985 regressed CPU usage by ~2 ms/s

• Status: RESOLVED
• Resolution: FIXED
• Component: New Bugs

[Simon Fraser (smfr)]

Scrolling sync changes in r261985 regressed CPU usage by ~2 ms/s

[Simon Fraser (smfr)]

Created attachment 406641 [details]
Patch

[Simon Fraser (smfr)]

<rdar://problem/66866163>

[Geoffrey Garen]

Comment on attachment 406641 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=406641&action=review

r=me

> Source/WebCore/page/scrolling/ScrollingTree.h:212
> +    bool isRecentlyActive();
> +    WEBCORE_EXPORT void setRecentlyActive();

Should we make these names more specific to wheel events?

isProcessingWheelEvent()
willProcessWheelEvent()

> Source/WebCore/page/scrolling/ScrollingTree.h:278
> +    Lock m_lastEventTimeMutex;
> +    MonotonicTime m_lastEventTime;

Should we make these names more specific to wheel events?

m_lastWheelEventTimeMutex
m_lastWheelEventTime

[Simon Fraser (smfr)]

(In reply to Geoffrey Garen from comment #3)
> Comment on attachment 406641 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=406641&action=review
> 
> r=me
> 
> > Source/WebCore/page/scrolling/ScrollingTree.h:212
> > +    bool isRecentlyActive();
> > +    WEBCORE_EXPORT void setRecentlyActive();
> 
> Should we make these names more specific to wheel events?
> 
> isProcessingWheelEvent()
> willProcessWheelEvent()
> 
> > Source/WebCore/page/scrolling/ScrollingTree.h:278
> > +    Lock m_lastEventTimeMutex;
> > +    MonotonicTime m_lastEventTime;
> 
> Should we make these names more specific to wheel events?
> 
> m_lastWheelEventTimeMutex
> m_lastWheelEventTime

Perhaps, though the ScrollingThread exists only to process wheel events, so I don't feel the need to qualify further.

[Simon Fraser (smfr)]

https://trac.webkit.org/changeset/265743/webkit