Bug 33863 - Occasional crash when interacting with page with reflected WebGL
Summary: Occasional crash when interacting with page with reflected WebGL
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: Layout and Rendering (show other bugs)
Version: 528+ (Nightly build)
Hardware: Mac OS X 10.5
: P2 Normal
Assignee: Simon Fraser (smfr)
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2010-01-19 13:42 PST by Simon Fraser (smfr)
Modified: 2010-01-19 16:45 PST (History)
0 users

See Also:

Attachments
Patch (5.89 KB, patch)
2010-01-19 13:50 PST, Simon Fraser (smfr) 		mitz: review+
Details | Formatted Diff | Diff
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Simon Fraser (smfr) 2010-01-19 13:42:19 PST 
There's an issue where compositing layers for a reflection can be torn down, without clearing the original layer's replicaLayer pointer.
Comment 1 Simon Fraser (smfr) 2010-01-19 13:50:09 PST 
Created attachment 46949 [details]
Patch
Comment 2 mitz 2010-01-19 14:39:20 PST 
Comment on attachment 46949 [details]
Patch

>          if (layer->backing()) {
> +            // If we're removing backing on a reflection, clear the source GraphicsLayer's pointer to
> +            // its replica GraphicsLayer. In practise this should never happen

So can we add an assertion?

> +            if (layer->isReflection()) {
> +                RenderLayer* sourceLayer = toRenderBoxModelObject(layer->renderer()->parent())->layer();
> +                if (RenderLayerBacking* backing = sourceLayer->backing())
> +                    backing->graphicsLayer()->setReplicatedByLayer(0);

And also assert right before here that the graphicsLayer currently thinks that it’s replicated by our layer?

r=me
Comment 3 Simon Fraser (smfr) 2010-01-19 16:37:59 PST 
http://trac.webkit.org/changeset/53507
Comment 4 Simon Fraser (smfr) 2010-01-19 16:45:37 PST 
http://trac.webkit.org/changeset/53507