Title: [231283] trunk
- Revision
- 231283
- Author
- fpi...@apple.com
- Date
- 2018-05-02 17:37:30 -0700 (Wed, 02 May 2018)
Log Message
JSC should know how to cache custom getter accesses on the prototype chain
https://bugs.webkit.org/show_bug.cgi?id=185213
Reviewed by Keith Miller.
JSTests:
* microbenchmarks/get-custom-getter.js: Added.
(test):
Source/_javascript_Core:
This was a simple fix after the work I did for bug 185174. >4x speed-up on the new get-custom-getter.js test.
* jit/Repatch.cpp:
(JSC::tryCacheGetByID):
Modified Paths
Added Paths
Diff
Modified: trunk/JSTests/ChangeLog (231282 => 231283)
--- trunk/JSTests/ChangeLog 2018-05-03 00:31:50 UTC (rev 231282)
+++ trunk/JSTests/ChangeLog 2018-05-03 00:37:30 UTC (rev 231283)
@@ -1,3 +1,13 @@
+2018-05-02 Filip Pizlo <fpi...@apple.com>
+
+ JSC should know how to cache custom getter accesses on the prototype chain
+ https://bugs.webkit.org/show_bug.cgi?id=185213
+
+ Reviewed by Keith Miller.
+
+ * microbenchmarks/get-custom-getter.js: Added.
+ (test):
+
2018-05-02 Robin Morisset <rmoris...@apple.com>
emitCodeToGetArgumentsArrayLength should not crash on PhantomNewArrayWithSpread
Added: trunk/JSTests/microbenchmarks/get-custom-getter.js (0 => 231283)
--- trunk/JSTests/microbenchmarks/get-custom-getter.js (rev 0)
+++ trunk/JSTests/microbenchmarks/get-custom-getter.js 2018-05-03 00:37:30 UTC (rev 231283)
@@ -0,0 +1,21 @@
+// RegExp.input is a handy getter
+
+var o = RegExp;
+o.input = "foo";
+
+function test(o) {
+ var result = null;
+ for (var i = 0; i < 30000; i++)
+ result = o.input;
+
+ return result;
+}
+
+for (var k = 0; k < 9; k++) {
+ var newResult = test(o)
+ if (newResult != "foo")
+ throw "Failed at " + k + " with " +newResult;
+ result = newResult;
+ o = {__proto__ : o }
+}
+
Modified: trunk/Source/_javascript_Core/ChangeLog (231282 => 231283)
--- trunk/Source/_javascript_Core/ChangeLog 2018-05-03 00:31:50 UTC (rev 231282)
+++ trunk/Source/_javascript_Core/ChangeLog 2018-05-03 00:37:30 UTC (rev 231283)
@@ -1,3 +1,15 @@
+2018-05-02 Filip Pizlo <fpi...@apple.com>
+
+ JSC should know how to cache custom getter accesses on the prototype chain
+ https://bugs.webkit.org/show_bug.cgi?id=185213
+
+ Reviewed by Keith Miller.
+
+ This was a simple fix after the work I did for bug 185174. >4x speed-up on the new get-custom-getter.js test.
+
+ * jit/Repatch.cpp:
+ (JSC::tryCacheGetByID):
+
2018-05-01 Filip Pizlo <fpi...@apple.com>
JSC should be able to cache custom setter calls on the prototype chain
Modified: trunk/Source/_javascript_Core/jit/Repatch.cpp (231282 => 231283)
--- trunk/Source/_javascript_Core/jit/Repatch.cpp 2018-05-03 00:31:50 UTC (rev 231282)
+++ trunk/Source/_javascript_Core/jit/Repatch.cpp 2018-05-03 00:37:30 UTC (rev 231283)
@@ -301,13 +301,19 @@
// We use ObjectPropertyConditionSet instead for faster accesses.
prototypeAccessChain = nullptr;
+ // FIXME: Maybe this `if` should be inside generateConditionsForPropertyBlah.
+ // https://bugs.webkit.org/show_bug.cgi?id=185215
if (slot.isUnset()) {
conditionSet = generateConditionsForPropertyMiss(
vm, codeBlock, exec, structure, propertyName.impl());
- } else {
+ } else if (!slot.isCacheableCustom()) {
conditionSet = generateConditionsForPrototypePropertyHit(
vm, codeBlock, exec, structure, slot.slotBase(),
propertyName.impl());
+ } else {
+ conditionSet = generateConditionsForPrototypePropertyHitCustom(
+ vm, codeBlock, exec, structure, slot.slotBase(),
+ propertyName.impl());
}
if (!conditionSet.isValid())
_______________________________________________
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes
