mnunez [Mon, 05 May 2014 14:17:20 +0200] rev 24750
8037572: Add more test cases to check static types
Reviewed-by: attila, lagergren
lagergren [Fri, 02 May 2014 18:22:29 +0200] rev 24749
8041625: AccessorProperty currentType must only by Object.class when non-primitive, and scoping followup problem for lazily generated with bodies
Reviewed-by: jlaskey, attila
mnunez [Tue, 29 Apr 2014 16:00:53 +0200] rev 24748
8038398: OptimisticRecompilationTest fails on staging repo nashorn/jdk9/nashorn due to test framework
Reviewed-by: attila, lagergren
lagergren [Mon, 28 Apr 2014 16:37:36 +0200] rev 24747
8041995: Problems when loading tree expressions with several optimistic program points when optimistically initializing ObjectNodes
Reviewed-by: jlaskey, attila
lagergren [Fri, 25 Apr 2014 14:26:07 +0200] rev 24746
8041905: Fix apply2call bug that prevented avatar.js unit tests from running correctly
Reviewed-by: attila, hannesw
lagergren [Wed, 23 Apr 2014 17:37:41 +0200] rev 24745
8041434: Add synchronization to the common global constants structure
Reviewed-by: attila, hannesw
lagergren [Wed, 23 Apr 2014 16:13:47 +0200] rev 24744
8038426: Move all loggers from process wide scope into Global scope
Reviewed-by: attila, hannesw
lagergren [Tue, 22 Apr 2014 14:09:46 +0200] rev 24743
8033105: Make sure Nashorn test harness can run zlib benchmark
Reviewed-by: attila, hannesw
lagergren [Fri, 18 Apr 2014 21:24:34 +0200] rev 24742
8040655: When processing a RewriteException debug object, the return value has already been reset to null. We need to catch this value before that.
Reviewed-by: attila, lagergren
Contributed-by: matherey.nunez@oracle.com
lagergren [Fri, 18 Apr 2014 20:12:21 +0200] rev 24741
8040102: Remove all references to Unsafe and definition of anonymous clases from the code
Summary: As the catch combinator optimization is now part of java.lang.invoke we don't need to put our own in the boot class path in any configuration anymore. Furthermore, with the completion of the array performance subtask of optimistic typing, we can remove the experimental (commented out) Unsafe accessors in the ArrayData classes
Reviewed-by: attila, jlaskey
lagergren [Thu, 17 Apr 2014 20:01:19 +0200] rev 24740
8040089: Apply to call transform was incomplete. Now passes all tests and performance is back
Reviewed-by: hannesw, attila, sundar, jlaskey
lagergren [Fri, 11 Apr 2014 16:52:56 +0200] rev 24739
Merge
lagergren [Fri, 11 Apr 2014 16:52:14 +0200] rev 24738
8039746: Transform applies to calls wherever possible, for ScriptFunctions and JSObjects.
Reviewed-by: hannesw, attila, sundar, jlaskey
attila [Fri, 11 Apr 2014 16:40:06 +0200] rev 24737
8040024: BranchOptimizer produces bad code for NaN FP comparison
Reviewed-by: jlaskey, lagergren
lagergren [Wed, 02 Apr 2014 10:52:39 +0200] rev 24736
8039044: Expand undefined intrinsics for all commutative combinators of scrict undefined checks
Reviewed-by: jlaskey, hannesw
lagergren [Tue, 01 Apr 2014 16:12:38 +0200] rev 24735
8038945: Simplify strict undefined checks
Reviewed-by: jlaskey, hannesw
lagergren [Tue, 01 Apr 2014 11:19:32 +0200] rev 24734
8038799: Guard and unbox boxed primitives types on setting them in Properties to avoid megamorphisism
Reviewed-by: attila, jlaskey
lagergren [Mon, 31 Mar 2014 14:13:34 +0200] rev 24733
8027043: Turn global accesses into MethodHandle.constant, with one chance of reassignment, e.g. x = value occuring once in the global scope is ok, twice is not.
Reviewed-by: attila, sundar, jlaskey
attila [Thu, 27 Mar 2014 14:09:40 +0100] rev 24732
8038416: Access to undefined scoped variables deoptimized too much
Reviewed-by: jlaskey, lagergren
lagergren [Thu, 27 Mar 2014 11:45:54 +0100] rev 24731
8038406: Testability: as a first step of moving loggers away from the process global space, the Debug object now supports logging POJOs from log entries as an event queue, which can be introspected from test scripts. This is way better than screen scraping brittle and subject-to-change log output.
Reviewed-by: attila, hannesw, sundar
attila [Wed, 26 Mar 2014 15:00:32 +0100] rev 24730
8038396: fix for the compiler expression evaluator to be more inquisitive about types
Reviewed-by: hannesw, lagergren
attila [Mon, 24 Mar 2014 18:41:06 +0100] rev 24729
8037534: Use scope types to determine optimistic types
Reviewed-by: hannesw, lagergren
attila [Mon, 24 Mar 2014 18:40:11 +0100] rev 24728
8038223: Symbol trace debug output takes time
Reviewed-by: hannesw, lagergren
attila [Fri, 21 Mar 2014 17:52:03 +0100] rev 24727
Merge
lagergren [Thu, 20 Mar 2014 16:26:54 +0100] rev 24726
8037967: Broke the build, by commiting without saving the last review comment
Reviewed-by: jlaskey, hannesw
lagergren [Thu, 20 Mar 2014 16:16:42 +0100] rev 24725
8033334: Make sure that scope depth information is maintained in the RecompilableScriptFunctionDatas, to avoid unnecessary slow proto linkage when doing on demand compilation
Summary: Compute RecompiledScriptFunctionDatas eagerly, annotate them with scope depth information and use them in recompilations.
Reviewed-by: attila, hannesw, jlaskey
mnunez [Mon, 17 Mar 2014 15:05:18 +0100] rev 24724
8037086: Check that deoptimizing recompilations are correct
Reviewed-by: attila, lagergren
lagergren [Fri, 14 Mar 2014 14:27:32 +0100] rev 24723
8036986: Test should check that correctly type is returned running with optimistic. If optimistic assumption was wrong we should get the right one.
Summary: This uses the inspection framework for generated code to ensure optimisim.
Reviewed-by: attila, lagergren
Contributed-by: matherey.nunez@oracle.com
lagergren [Wed, 12 Mar 2014 10:31:22 +0100] rev 24722
8037177: -Dnashorn.optimistic should be enabled by default, meaning that it has to be explicitly set to false to run with the jdk 8 style conservative types
Summary: This will be the new default setting, so the explicit flag should be removed, and instead reverse checked for explicitly DISABLED optimistic types. In the future, we might remove the flag altogether.
Reviewed-by: attila, hannesw
lagergren [Wed, 05 Mar 2014 09:51:00 +0100] rev 24721
8036127: Prototype filter needs to be applied to getter guard as well, not just getter
Summary: This manifests itself as a bug in optimistic types, as inner functions may access properties of the wrong type, but it is also a bug in tip. Without optimistic types, we have been unable to find a reproducer due to more similar PropertyMaps
Reviewed-by: attila, jlaskey, sundar
lagergren [Mon, 03 Mar 2014 11:24:44 +0100] rev 24720
8035836: Array performance improvements
Summary: Implement typed arrays with native byte buffers and do fast linking for all continuous arrays
Reviewed-by: attila, jlaskey, sundar, hannesw
attila [Wed, 26 Feb 2014 13:17:57 +0100] rev 24719
8035820: Optimistic recompilation
Reviewed-by: hannesw, jlaskey, sundar
Contributed-by: attila.szegedi@oracle.com, marcus.lagergren@oracle.com