[webkit-reviews] review granted: [Bug 72106] Consolidate the logic that creates run loop observers for flushing layer tree changes to CoreAnimation : [Attachment 115105] Patch

[bugzilla-daemon at webkit.org]

Anders Carlsson <andersca at apple.com> has granted Andy Estes
<aestes at apple.com>'s request for review:
Bug 72106: Consolidate the logic that creates run loop observers for flushing
layer tree changes to CoreAnimation
https://bugs.webkit.org/show_bug.cgi?id=72106

Attachment 115105: Patch
https://bugs.webkit.org/attachment.cgi?id=115105&action=review

------- Additional Comments from Anders Carlsson <andersca at apple.com>
View in context: https://bugs.webkit.org/attachment.cgi?id=115105&action=review


>> Source/WebCore/platform/graphics/ca/LayerFlushScheduler.cpp:40
>> +	ASSERT(m_suspendCount);
> 
> This should be m_suspendCount > 0

I don't think this assertion buys us anything unless someone calls suspend more
than 2^31 times :)

> Source/WebCore/platform/graphics/ca/LayerFlushScheduler.cpp:46
> +    --m_suspendCount;
> +    ASSERT(m_suspendCount > -1);

I think it's cleaner to put the ASSERT before the decrement here - then you can
just assert that m_suspendCount is not 0.

> Source/WebCore/platform/graphics/ca/LayerFlushScheduler.h:51
> +    int m_suspendCount;

I'd make this an unsigned integer instead.