[webkit-reviews] review granted: [Bug 198005] REGRESSION (r245170): gmail.com inbox table header flickers : [Attachment 370171] Patch

[bugzilla-daemon at webkit.org]

Antti Koivisto <koivisto at iki.fi> has granted Simon Fraser (smfr)
<simon.fraser at apple.com>'s request for review:
Bug 198005: REGRESSION (r245170): gmail.com inbox table header flickers
https://bugs.webkit.org/show_bug.cgi?id=198005

Attachment 370171: Patch

https://bugs.webkit.org/attachment.cgi?id=370171&action=review




--- Comment #7 from Antti Koivisto <koivisto at iki.fi> ---
Comment on attachment 370171
  --> https://bugs.webkit.org/attachment.cgi?id=370171
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=370171&action=review

> Source/WebCore/ChangeLog:16
> +	   Mak some logging changes to help diagnose things like this.

Typo

> Source/WebCore/rendering/RenderLayerCompositor.cpp:935
> +	   if (layerPaintsIntoProvidedBacking) {
> +	       layerPaintsIntoProvidedBacking = false;
> +	       // layerPaintsIntoProvidedBacking was only true for layers that
would otherwise composite because of overlap. If we can
> +	       // no longer share, put this this indirect reason back on the
layer so that requiresOwnBackingStore() sees it.
> +	      
layer.setIndirectCompositingReason(IndirectCompositingReason::Overlap);

It is bit non-obvious that layerPaintsIntoProvidedBacking implies
IndirectCompositingReason::Overlap. Maybe rename it to
layerPaintsIntoProvidedBackingDueToOverlap or similar to avoid future mistakes?