Bug 55100 - Make WeakGCMap use new handle infrastructure
Summary: Make WeakGCMap use new handle infrastructure
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: New Bugs (show other bugs)
Version: 528+ (Nightly build)
Hardware: Other OS X 10.5
: P2 Normal
Assignee: Oliver Hunt
URL:
Keywords:
Depends on:
Blocks: 56383
  Show dependency treegraph
 
Reported: 2011-02-23 16:28 PST by Oliver Hunt
Modified: 2011-03-15 09:43 PDT (History)
1 user (show)

See Also:


Attachments
Patch (69.19 KB, patch)
2011-02-23 16:42 PST, Oliver Hunt
no flags Details | Formatted Diff | Diff
Patch (54.58 KB, patch)
2011-02-24 12:37 PST, Oliver Hunt
ggaren: review+
Details | Formatted Diff | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Oliver Hunt 2011-02-23 16:28:06 PST
Make WeakGCMap use new handle infrastructure
Comment 1 Oliver Hunt 2011-02-23 16:42:54 PST
Created attachment 83576 [details]
Patch
Comment 2 Oliver Hunt 2011-02-23 16:43:31 PST
Comment on attachment 83576 [details]
Patch

I've realised that i can separate out a few bits of this patch completely
Comment 3 Oliver Hunt 2011-02-24 12:37:03 PST
Created attachment 83702 [details]
Patch
Comment 4 Geoffrey Garen 2011-02-24 12:51:00 PST
Comment on attachment 83702 [details]
Patch

r=me

I think you should verify performance using the Droameo and DOM Mandlebrot tests.
Comment 5 Oliver Hunt 2011-02-24 14:00:34 PST
Committed r79616: <http://trac.webkit.org/changeset/79616>
Comment 6 Eric Seidel (no email) 2011-02-25 00:06:06 PST
Worker tests are crashing since around this time:

Thread 8 Crashed:  WebCore: Worker
0   com.apple.JavaScriptCore      	0x00000001002b3907 JSC::MarkedBlock::destroy(JSC::MarkedBlock*) + 39 (MarkedBlock.cpp:44)
1   com.apple.JavaScriptCore      	0x000000010022ce48 JSC::MarkedSpace::freeBlocks(WTF::DoublyLinkedList<JSC::MarkedBlock>&) + 392 (MarkedSpace.cpp:66)
2   com.apple.JavaScriptCore      	0x000000010022d593 JSC::MarkedSpace::shrink() + 163 (MarkedSpace.cpp:103)
3   com.apple.JavaScriptCore      	0x0000000100167ced JSC::Heap::destroy() + 109 (Heap.cpp:84)
4   com.apple.WebCore             	0x00000001013c43fc WebCore::WorkerScriptController::~WorkerScriptController() + 76 (WorkerScriptController.cpp:65)
5   com.apple.WebCore             	0x00000001013c7802 WebCore::WorkerThreadShutdownFinishTask::performTask(WebCore::ScriptExecutionContext*) + 50 (WorkerScriptController.h:48)
6   com.apple.WebCore             	0x00000001013c3841 WebCore::WorkerRunLoop::runInMode(WebCore::WorkerContext*, WebCore::ModePredicate const&) + 385 (WorkerRunLoop.cpp:163)
7   com.apple.WebCore             	0x00000001013c3a1e WebCore::WorkerRunLoop::run(WebCore::WorkerContext*) + 238 (WorkerRunLoop.cpp:133)
8   com.apple.WebCore             	0x00000001013c71c0 WebCore::WorkerThread::workerThread() + 992 (WorkerThread.cpp:143)
9   libSystem.B.dylib             	0x00007fff859e3456 _pthread_start + 331
10  libSystem.B.dylib             	0x00007fff859e3309 thread_start + 13

Either this, or bug 55159 is to blame.