Summary: | Rewire the same-origin checks for the JavaScriptCore bindings through BindingSecurity | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
Product: | WebKit | Reporter: | Adam Barth <abarth> | ||||||||
Component: | WebCore JavaScript | Assignee: | Adam Barth <abarth> | ||||||||
Status: | RESOLVED FIXED | ||||||||||
Severity: | Normal | CC: | darin, eric, gyuyoung.kim, ossy, rakuco, sam, webkit.review.bot | ||||||||
Priority: | P2 | ||||||||||
Version: | 528+ (Nightly build) | ||||||||||
Hardware: | Unspecified | ||||||||||
OS: | Unspecified | ||||||||||
Bug Depends on: | 93654 | ||||||||||
Bug Blocks: | 75793, 93407 | ||||||||||
Attachments: |
|
Description
Adam Barth
2012-08-07 12:17:21 PDT
Created attachment 156972 [details]
Patch
Created attachment 156973 [details]
Patch
Comment on attachment 156973 [details] Patch Attachment 156973 [details] did not pass efl-ews (efl): Output: http://queues.webkit.org/results/13445851 Created attachment 156989 [details]
hopefully builds on EFL now
Comment on attachment 156989 [details]
hopefully builds on EFL now
I wonder what, if any, perf affect this has on DOM benchmarks.
> I wonder what, if any, perf affect this has on DOM benchmarks.
I'll run the perf tests before landing, but I don't expect any trouble.
The work to rewrite allowsAccessFrom through BindingSecurity is in Bug 93407. This seems OK to me, but I would like to see the extra indirect go away. A follow up patch would be fine for that. Thanks. Will do. Committed r125126: <http://trac.webkit.org/changeset/125126> (In reply to comment #10) > Committed r125126: <http://trac.webkit.org/changeset/125126> It caused an assertion on Qt and on GTK. Could you check it, please? https://bugs.webkit.org/show_bug.cgi?id=93654 > It caused an assertion on Qt and on GTK. Could you check it, please? https://bugs.webkit.org/show_bug.cgi?id=93654
Certainly.
|