Unreviewed, add a FIXME comment referencing https://bugs.webkit.org/show_bug.cgi...
authorfpizlo@apple.com <fpizlo@apple.com@268f45cc-cd09-0410-ab3c-d52691b4dbfc>
Thu, 30 Apr 2015 17:44:23 +0000 (17:44 +0000)
committerfpizlo@apple.com <fpizlo@apple.com@268f45cc-cd09-0410-ab3c-d52691b4dbfc>
Thu, 30 Apr 2015 17:44:23 +0000 (17:44 +0000)
* jit/JITOperations.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@183631 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Source/JavaScriptCore/ChangeLog
Source/JavaScriptCore/jit/JITOperations.cpp

index a828cfa..c698c3b 100644 (file)
@@ -1,5 +1,11 @@
 2015-04-30  Filip Pizlo  <fpizlo@apple.com>
 
+        Unreviewed, add a FIXME comment referencing https://bugs.webkit.org/show_bug.cgi?id=144458.
+
+        * jit/JITOperations.cpp:
+
+2015-04-30  Filip Pizlo  <fpizlo@apple.com>
+
         Add a comment clarifying the behavior and semantics of getCallData/getConstructData, in
         particular that they cannot change their minds and may be called from compiler threads.
 
index 7214973..98d8773 100644 (file)
@@ -721,8 +721,12 @@ inline char* linkFor(
     
     JSValue calleeAsValue = execCallee->calleeAsValue();
     JSCell* calleeAsFunctionCell = getJSFunction(calleeAsValue);
-    if (!calleeAsFunctionCell)
+    if (!calleeAsFunctionCell) {
+        // FIXME: We should cache these kinds of calls. They can be common and currently they are
+        // expensive.
+        // https://bugs.webkit.org/show_bug.cgi?id=144458
         return reinterpret_cast<char*>(handleHostCall(execCallee, calleeAsValue, kind));
+    }
 
     JSFunction* callee = jsCast<JSFunction*>(calleeAsFunctionCell);
     JSScope* scope = callee->scopeUnchecked();