Bug 158144 – Web Inspector: ReportExtraMemoryCost IDL attribute should also be used to generate estimatedSize method

• Status: RESOLVED
• Resolution: FIXED
• Component: Web Inspector

[Joseph Pecoraro]

* SUMMARY
ReportExtraMemoryCost IDL attribute should also be used to generate estimatedSize method.

The pattern for including the JSCell estimatedSize() method is "if a cell reports extra memory to the Heap, it should also include that in its estimated size". We were missing this for WebCore objects (of which there are only a few). The existing ReportExtraMemoryCost IDL attribute already handles reporting extra memory to the Heap using wrapper().memoryCost(). The exact same thing can be used in estimatedSize().

[Radar WebKit Bug Importer]

<rdar://problem/26510790>

[Joseph Pecoraro]

Created attachment 279942 [details]
[PATCH] Proposed Fix

[Darin Adler]

Comment on attachment 279942 [details]
[PATCH] Proposed Fix

View in context: https://bugs.webkit.org/attachment.cgi?id=279942&action=review

> Source/WebCore/bindings/scripts/CodeGeneratorJS.pm:1049
> +    my $needsEstimatedSize = InstanceNeedsEstimatedSize($interface);

Seems better to just call this the one place it’s used instead of putting it into a local variable.

> Source/WebCore/bindings/scripts/CodeGeneratorJS.pm:2012
> +    my $needsEstimatedSize = InstanceNeedsEstimatedSize($interface);

Seems better to just call this the one place it’s used instead of putting it into a local variable.

> Source/WebCore/bindings/scripts/CodeGeneratorJS.pm:3276
> +        push(@implContent, "    ASSERT_GC_OBJECT_INHERITS(thisObject, info());\n");

It seems peculiar to do this assertion after the jsCast. Doesn’t the JSCast do this check already? If it doesn’t then I’d suggest asserting first.

> Source/WebCore/bindings/scripts/CodeGeneratorJS.pm:3278
> +        push(@implContent, "    size_t extraSize = thisObject->wrapped().memoryCost();\n");
> +        push(@implContent, "    return Base::estimatedSize(thisObject) + extraSize;\n");

This would read better without the local variable, I think.

[Joseph Pecoraro]

Created attachment 280199 [details]
[PATCH] For Landing

[WebKit Commit Bot]

Comment on attachment 280199 [details]
[PATCH] For Landing

Clearing flags on attachment: 280199

Committed r201541: <http://trac.webkit.org/changeset/201541>