http://trac.webkit.org/changeset/154120 accidentally changed DFGCapabilities to read...
authorfpizlo@apple.com <fpizlo@apple.com@268f45cc-cd09-0410-ab3c-d52691b4dbfc>
Thu, 15 Aug 2013 21:54:01 +0000 (21:54 +0000)
committerfpizlo@apple.com <fpizlo@apple.com@268f45cc-cd09-0410-ab3c-d52691b4dbfc>
Thu, 15 Aug 2013 21:54:01 +0000 (21:54 +0000)
Rubber stamped by Oliver Hunt.

This was causing some test crashes for me.

* dfg/DFGCapabilities.cpp:
(JSC::DFG::capabilityLevel):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@154139 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Source/JavaScriptCore/ChangeLog
Source/JavaScriptCore/dfg/DFGCapabilities.cpp

index 2db35fe..8f9dc91 100644 (file)
@@ -1,3 +1,14 @@
+2013-08-15  Filip Pizlo  <fpizlo@apple.com>
+
+        http://trac.webkit.org/changeset/154120 accidentally changed DFGCapabilities to read the resolve type from operand 4, not 3; it should be 3.
+
+        Rubber stamped by Oliver Hunt.
+        
+        This was causing some test crashes for me.
+
+        * dfg/DFGCapabilities.cpp:
+        (JSC::DFG::capabilityLevel):
+
 2013-08-15  Brent Fulgham  <bfulgham@apple.com>
 
         [Windows] Clear up improper export declaration.
index 82ee47b..ea88697 100644 (file)
@@ -183,7 +183,7 @@ CapabilityLevel capabilityLevel(OpcodeID opcodeID, CodeBlock* codeBlock, Instruc
 
     case op_resolve_scope: {
         // We don't compile 'catch' or 'with', so there's no point in compiling variable resolution within them.
-        ResolveType resolveType = ResolveModeAndType(pc[4].u.operand).type();
+        ResolveType resolveType = ResolveModeAndType(pc[3].u.operand).type();
         if (resolveType == Dynamic)
             return CannotCompile;
         return CanCompileAndInline;