Bug 201521 – Refactor the Gigacage code to require less pointer casting.

• Status: RESOLVED
• Resolution: FIXED
• Component: JavaScriptCore

[Mark Lam]

Also clean up the code a bit.

[Mark Lam]

Created attachment 378129 [details]
proposed patch.

[Radar WebKit Bug Importer]

<rdar://problem/55087505>

[Radar WebKit Bug Importer]

<rdar://problem/55087515>

[Saam Barati]

Comment on attachment 378129 [details]
proposed patch.

View in context: https://bugs.webkit.org/attachment.cgi?id=378129&action=review

r=me

> Source/bmalloc/ChangeLog:17
> +        5. Change LLInt's loadCagedJSValue() to skip the caging if Gigacage is not enabled
> +           in the build.  This allows us to remove the unneeded stubs in WTF Gigacage.h.

nit: some part of this belongs in the JSC changelog.

> Source/bmalloc/bmalloc/Gigacage.h:50
> +    NumberOfKinds // This should always be at the end of this list.

this comment is implied by the name of the enum value

> Source/bmalloc/bmalloc/Gigacage.h:61
> +        break; // Invalid: go to BCRASH();

no need for this comment

[Mark Lam]

Thanks for the review.

(In reply to Saam Barati from comment #4)
> > Source/bmalloc/ChangeLog:17
> > +        5. Change LLInt's loadCagedJSValue() to skip the caging if Gigacage is not enabled
> > +           in the build.  This allows us to remove the unneeded stubs in WTF Gigacage.h.
> 
> nit: some part of this belongs in the JSC changelog.

Fixed.

> 
> > Source/bmalloc/bmalloc/Gigacage.h:50
> > +    NumberOfKinds // This should always be at the end of this list.
> 
> this comment is implied by the name of the enum value

Fixed.

> > Source/bmalloc/bmalloc/Gigacage.h:61
> > +        break; // Invalid: go to BCRASH();
> 
> no need for this comment

Fixed.

[Mark Lam]

Created attachment 378133 [details]
patch for landing.

[Mark Lam]

The jsc EWS is red purportedly due to a build failure.  However, that failure can only manifest if the bot is picking up an outdated Gigacage.h.  Obviously, the JSC lib built correctly for all the other bots.  So the jsc EWS failure is bogus.

[Mark Lam]

Landed in r249556: <http://trac.webkit.org/r249556>.