Changeset 225270 in webkit for trunk/Source/JavaScriptCore/ChangeLog

Timestamp:

Nov 29, 2017, 9:31:54 AM (8 years ago)

Author:

[email protected]

Message:

The recursive tail call optimisation is wrong on closures
​https://bugs.webkit.org/show_bug.cgi?id=179835

Reviewed by Saam Barati.

JSTests:

• stress/closure-recursive-tail-call.js: Added.

(makeClosure):

PerformanceTests:

This new benchmark is a very close variant of the merge-sort benchmark, that writes mergeSorted in a kinda CPS style,
to stress the use of closures, and of polymorphic calls.

• TailBench9000/merge-sort-cps.js: Added.

(createRNG):
(mergeSorted):
(checkSorted.check):
(add):
(build):
(compare):
(checkSpectrum):
(buildArray):
(test):

Source/JavaScriptCore:

The problem is that we only check the executable of the callee, not whatever variables might have been captured.
As a stopgap measure this patch just does not do the optimisation for closures.

• dfg/DFGByteCodeParser.cpp:

(JSC::DFG::ByteCodeParser::handleRecursiveTailCall):

Tools:

This just includes merge-sort-cps.js to the list of benchmarks ran by run-jsc-benchmarks --tail-bench

• Scripts/run-jsc-benchmarks:

File:

• trunk/Source/JavaScriptCore/ChangeLog (modified) (1 diff)

trunk/Source/JavaScriptCore/ChangeLog

@@ +1 @@
+2017-11-29  Robin Morisset  <[email protected]>
+
+        The recursive tail call optimisation is wrong on closures
+        https://bugs.webkit.org/show_bug.cgi?id=179835
+
+        Reviewed by Saam Barati.
+
+        The problem is that we only check the executable of the callee, not whatever variables might have been captured.
+        As a stopgap measure this patch just does not do the optimisation for closures.
+
+        * dfg/DFGByteCodeParser.cpp:
+        (JSC::DFG::ByteCodeParser::handleRecursiveTailCall):
+
 2017-11-28  Joseph Pecoraro  <[email protected]>