Title: [185042] trunk/Source/_javascript_Core
- Revision
- 185042
- Author
- [email protected]
- Date
- 2015-05-31 11:03:52 -0700 (Sun, 31 May 2015)
Log Message
Unreviewed, add a FIXME referencing https://bugs.webkit.org/show_bug.cgi?id=145503.
* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::inliningCost):
Modified Paths
Diff
Modified: trunk/Source/_javascript_Core/ChangeLog (185041 => 185042)
--- trunk/Source/_javascript_Core/ChangeLog 2015-05-31 15:10:09 UTC (rev 185041)
+++ trunk/Source/_javascript_Core/ChangeLog 2015-05-31 18:03:52 UTC (rev 185042)
@@ -1,3 +1,10 @@
+2015-05-31 Filip Pizlo <[email protected]>
+
+ Unreviewed, add a FIXME referencing https://bugs.webkit.org/show_bug.cgi?id=145503.
+
+ * dfg/DFGByteCodeParser.cpp:
+ (JSC::DFG::ByteCodeParser::inliningCost):
+
2015-05-31 Yusuke Suzuki <[email protected]>
[ES6] Drop WeakMap#clear
Modified: trunk/Source/_javascript_Core/dfg/DFGByteCodeParser.cpp (185041 => 185042)
--- trunk/Source/_javascript_Core/dfg/DFGByteCodeParser.cpp 2015-05-31 15:10:09 UTC (rev 185041)
+++ trunk/Source/_javascript_Core/dfg/DFGByteCodeParser.cpp 2015-05-31 18:03:52 UTC (rev 185042)
@@ -1253,6 +1253,11 @@
// this function.
// https://bugs.webkit.org/show_bug.cgi?id=127627
+ // FIXME: We currently inline functions that have run in LLInt but not in Baseline. These
+ // functions have very low fidelity profiling, and presumably they weren't very hot if they
+ // haven't gotten to Baseline yet. Consider not inlining these functions.
+ // https://bugs.webkit.org/show_bug.cgi?id=145503
+
// Have we exceeded inline stack depth, or are we trying to inline a recursive call to
// too many levels? If either of these are detected, then don't inline. We adjust our
// heuristics if we are dealing with a function that cannot otherwise be compiled.
_______________________________________________
webkit-changes mailing list
[email protected]
https://lists.webkit.org/mailman/listinfo/webkit-changes
