Bug 41176 - Fix testapi.c to account for movable objects
Summary: Fix testapi.c to account for movable objects
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: JavaScriptCore (show other bugs)
Version: 528+ (Nightly build)
Hardware: All All
: P2 Normal
Assignee: Nathan Lawrence
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2010-06-24 14:22 PDT by Nathan Lawrence
Modified: 2010-06-25 23:26 PDT (History)
5 users (show)

See Also:

Attachments
Patch (1.96 KB, patch)
2010-06-24 14:22 PDT, Nathan Lawrence 		ggaren: review+
ggaren: commit-queue+
Details | Formatted Diff | Diff
Patch (2.14 KB, patch)
2010-06-25 11:45 PDT, Nathan Lawrence 		no flags Details | Formatted Diff | Diff
Show Obsolete (1) 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 Nathan Lawrence 2010-06-24 14:22:02 PDT 
Created attachment 59701 [details]
Patch

We assume in testapi.c that the value aHeapRef refers to will not be moved.  When we have movable objects, this will not be the case.
Comment 1 Geoffrey Garen 2010-06-24 15:53:29 PDT 
Comment on attachment 59701 [details]
Patch

r=me
Comment 2 Oliver Hunt 2010-06-24 23:45:49 PDT 
Comment on attachment 59701 [details]
Patch

aStackRef should be cleared immediately after JSObjectSetPrivateProperty as the private storage should then keep the heap reference live.
Comment 3 Oliver Hunt 2010-06-24 23:48:06 PDT 
Basically because aHeapRef needs to be pinned somehow -- realistically we'll need some kind of layer for API objects to GC allocate handles i guess :-/
Comment 4 Nathan Lawrence 2010-06-25 11:45:12 PDT 
Created attachment 59784 [details]
Patch
Comment 5 Nathan Lawrence 2010-06-25 11:46:09 PDT 
Updated per Oliver's comment.
Comment 6 Geoffrey Garen 2010-06-25 11:54:13 PDT 
(In reply to comment #3)
> Basically because aHeapRef needs to be pinned somehow -- realistically we'll need some kind of layer for API objects to GC allocate handles i guess :-/

The current plan is just not to move objects pointed to by the stack.
Comment 7 Geoffrey Garen 2010-06-25 16:54:26 PDT 
Comment on attachment 59784 [details]
Patch

r=me
Comment 8 WebKit Commit Bot 2010-06-25 23:25:58 PDT 
Comment on attachment 59784 [details]
Patch

Clearing flags on attachment: 59784

Committed r61946: <http://trac.webkit.org/changeset/61946>
Comment 9 WebKit Commit Bot 2010-06-25 23:26:03 PDT 
All reviewed patches have been landed.  Closing bug.