Mercurial > hg > truffle
graph
-
modified CustomCanonicalizer to deal with Nodes (instead of ValueNodes) to avoid extra type testing during canonicalization (GRAAL-506)Wed, 02 Oct 2013 21:43:19 +0200, by Doug Simon
-
Merge.Wed, 02 Oct 2013 21:40:29 +0200, by Doug Simon
-
replace instanceof type tests for Canonicalizable and Simplifiable with extra boolean properties in NodeClass (GRAAL-506)Wed, 02 Oct 2013 20:57:11 +0200, by Doug Simon
-
moved Canonicalizable and Simplifiable to the com.oracle.graal.graph project (GRAAL-506)Tue, 01 Oct 2013 20:38:25 +0200, by Doug Simon
-
broadened types in Canonicalizable interface so that it can be moved to the com.oracle.graal.graph project (GRAAL-506)Tue, 01 Oct 2013 20:32:47 +0200, by Doug Simon
-
broadened types in SimplifierTool interface so that it can be moved to the com.oracle.graal.graph project (GRAAL-506)Tue, 01 Oct 2013 20:30:04 +0200, by Doug Simon
-
Inlining: remove killing begin node in non-exceptional path of InvokeWithException after inliningWed, 02 Oct 2013 21:54:12 +0200, by Bernhard Urban
-
fix signature of reflective invocation of ReplacementsImpl.makeGraphWed, 02 Oct 2013 18:40:48 +0200, by Lukas Stadler
-
MergeWed, 02 Oct 2013 17:34:27 +0200, by Lukas Stadler
-
don't tail duplicate object allocationsWed, 21 Aug 2013 15:29:55 +0200, by Lukas Stadler
-
some more javadoc for benchmark countersWed, 02 Oct 2013 15:17:16 +0200, by Lukas Stadler
-
in-depth profiling of allocations and monitorsWed, 02 Oct 2013 15:05:04 +0200, by Lukas Stadler
-
remove unused optionWed, 02 Oct 2013 15:04:04 +0200, by Lukas Stadler
-
fix, simplify and enable ObjectClone intrinsificationWed, 02 Oct 2013 14:38:07 +0200, by Lukas Stadler
-
new parameter on Snippet annotation to remove all frame statesWed, 02 Oct 2013 14:16:04 +0200, by Lukas Stadler
-
Merge.Wed, 02 Oct 2013 15:37:06 +0100, by Chris Seaton
-
Merge.Wed, 02 Oct 2013 14:46:30 +0100, by Chris Seaton
-
mx sl command.Wed, 02 Oct 2013 14:19:28 +0100, by Chris Seaton
-
CommitAllocation: use INIT_LOCATION for initializing objects and arraysWed, 02 Oct 2013 15:43:39 +0200, by Bernhard Urban
-
SchedulePhase: bail out with SchedulingError if scheduled block is not dominated by earliestWed, 02 Oct 2013 11:16:21 +0200, by Bernhard Urban
-
NewMemoryAwareScheudling: process loop a second time in order to compute the loop exists correctly with the new merged stateWed, 02 Oct 2013 11:16:21 +0200, by Bernhard Urban
-
NewMemoryAwareScheduling: handle MemoryPhis properlyWed, 02 Oct 2013 13:32:30 +0200, by Bernhard Urban
-
MemoryScheduling: update testcasesWed, 02 Oct 2013 11:16:21 +0200, by Bernhard Urban
-
FloatingReadNode: add verify(), make sure that lastLocationAccess is always a MemoryCheckpointWed, 25 Sep 2013 19:29:14 +0200, by Bernhard Urban
-
InvokeWithException: make next successor a special begin node which has the same locationidentity as the invokeWed, 02 Oct 2013 11:16:21 +0200, by Bernhard Urban
-
Fix bug in constant folding.Wed, 02 Oct 2013 14:57:37 +0200, by Roland Schatz
-
MergeWed, 02 Oct 2013 13:28:17 +0200, by Lukas Stadler
-
simplify PEA block states by turning scalarAliases and objectAliases into one global NodeMapWed, 02 Oct 2013 13:18:25 +0200, by Lukas Stadler
-
remove unused method from VirtualizerToolWed, 02 Oct 2013 10:19:43 +0200, by Lukas Stadler
-
small javadoc fixWed, 02 Oct 2013 10:18:50 +0200, by Lukas Stadler
-
remove unused EscapeState.ThreadLocalWed, 02 Oct 2013 13:26:32 +0200, by Lukas Stadler
-
MergeMon, 30 Sep 2013 18:32:47 +0200, by Lukas Stadler
-
move benchmark counters into separate class and make them correct for multithreaded applicationsMon, 30 Sep 2013 18:32:26 +0200, by Lukas Stadler
-
Simplify constant folding.Wed, 02 Oct 2013 12:09:51 +0200, by Roland Schatz
-
Truffle: minor legibility improvement in NodeUtil.printSourceAttributionTree()Tue, 01 Oct 2013 17:26:56 -0700, by Michael Van De Vanter
-
Merge with f5faf49a86dd4bfd8e5ff6b9d6beedd280ea3f69Tue, 01 Oct 2013 07:23:33 -0700, by Michael Van De Vanter
-
Truffle: add NodeUtil.printSourceAttributionTree() for debuggingMon, 30 Sep 2013 20:47:41 -0700, by Michael Van De Vanter
-
remove MemoryState, make MemoryMap a node, add MemoryMap input to ReturnNodeMon, 30 Sep 2013 22:05:40 +0200, by Bernhard Urban
-
MemoryMap: move to graal.nodes packageMon, 30 Sep 2013 16:36:24 +0200, by Bernhard Urban
-
NewObjectSnippets: don't use a seperate location for initializing the array length fieldMon, 30 Sep 2013 16:30:18 +0200, by Bernhard Urban
-
SnippetTemplate: remove canonicalizer. remove loop for return node. copy Set on user sideMon, 30 Sep 2013 16:28:42 +0200, by Bernhard Urban
-
gtx_ptx.cpp: fix switch default label and operator precedence in ncores().Mon, 30 Sep 2013 21:46:41 +0200, by Andreas Woess
-
PTX warp limiter to available GPU processorsMon, 30 Sep 2013 13:03:47 -0400, by Morris Meyer
-
Fix wrong local variable name in LoopPolicies.shouldFullUnrollMon, 30 Sep 2013 17:03:14 +0200, by Gilles Duboscq
-
Keep the correct Deoptimization reason and action while grouping DeoptimizationsMon, 30 Sep 2013 16:10:32 +0200, by Gilles Duboscq
-
Temporarily move encodeDeoptActionAndReason to MetaAccessProviderMon, 30 Sep 2013 16:09:40 +0200, by Gilles Duboscq
-
MergeMon, 30 Sep 2013 18:05:25 +0200, by Gilles Duboscq
-
Use ControlSplitNode and ControlSinkNode in GraphBuilderPhase.isBlockEndThu, 26 Sep 2013 17:33:04 +0200, by Gilles Duboscq
-
Add a DynamicDeoptimizeNode where the action and reason is a input node.Thu, 26 Sep 2013 17:30:19 +0200, by Gilles Duboscq
-
Fix copyright date in NodeWithStateThu, 26 Sep 2013 16:26:56 +0200, by Gilles Duboscq
-
Remove DeoptimizingNode.getDeoptimizationReasonThu, 26 Sep 2013 16:10:02 +0200, by Gilles Duboscq
-
SpeculationLog gives back the necessary Constant when calling maySpeculate. Remove the deoptimizationReasonfrom the DebugInfoThu, 26 Sep 2013 16:03:28 +0200, by Gilles Duboscq
-
CodeCacheProvider.encodeDeoptActionAndReason now returns a ConstantThu, 26 Sep 2013 14:41:12 +0200, by Gilles Duboscq
-
Merge.Mon, 30 Sep 2013 16:49:08 +0200, by Doug Simon
-
enhanced test for CipherBlockChaining substitutions to cover path where the substitutions call the original (i.e. substituted) methods on slow pathsMon, 30 Sep 2013 16:48:24 +0200, by Doug Simon
-
omit canonicalization in IncrementalCanonicalizerPhase if none of the contained phases changed the graphMon, 30 Sep 2013 10:03:05 +0200, by Doug Simon
-
new PrintAfterCompilation option, simplify PrintCompilation outputMon, 30 Sep 2013 14:02:07 +0200, by Lukas Stadler
-
graal code installer: minor style cleanupMon, 30 Sep 2013 11:36:52 +0200, by Bernhard Urban
-
AMD64HotSpot: use conditional jump for IC_MISS_HANDLERMon, 30 Sep 2013 11:16:36 +0200, by Bernhard Urban
-
AMD64HotSpot: emit jump to IC miss handler directly instead of emitting it inside verfied_entryMon, 30 Sep 2013 09:32:18 +0200, by Bernhard Urban
-
improved javadoc for ResolvedJavaField.readConstantValue and .readValueMon, 30 Sep 2013 09:47:41 +0200, by Doug Simon
-
improved documentation and testing for ResolvedJavaType.resolveMethod() (GRAAL-489)Mon, 30 Sep 2013 09:47:32 +0200, by Doug Simon
-
PTX: one-dimensional parallel warp invocation, ParallelOver annotationSun, 29 Sep 2013 14:47:12 -0400, by Morris Meyer
-
PTX single-threaded array store, Warp annotationSat, 28 Sep 2013 21:06:12 -0400, by Morris Meyer
-
MergeFri, 27 Sep 2013 19:51:01 +0200, by Christos Kotselidis
-
Readjust branch probabilities of G1 BarriersFri, 27 Sep 2013 19:50:35 +0200, by Christos Kotselidis
-
put Node.getEncapsulatedSourceSection() on slow pathFri, 27 Sep 2013 16:39:54 +0200, by Michael Haupt
-
Augment Write Barrier countersFri, 27 Sep 2013 12:36:29 +0200, by Christos Kotselidis
-
Provide object read/write methods similar to Pointer.readXxxx/writeXxx that do not require casts to Word first, and use them in relevant placesThu, 26 Sep 2013 15:53:21 -0700, by Christian Wimmer
-
attempt to canonicalize array length access when lowering array load|store operationsThu, 26 Sep 2013 22:45:25 +0200, by Doug Simon
-
MergeThu, 26 Sep 2013 16:46:27 +0200, by Andreas Woess
-
Truffle: force slow path on StringBuilder methods.Thu, 26 Sep 2013 16:44:54 +0200, by Andreas Woess
-
Truffle: approximate source location for "illegal recursive call".Thu, 26 Sep 2013 16:38:35 +0200, by Andreas Woess
-
MergeThu, 26 Sep 2013 13:48:19 +0200, by Andreas Woess
-
Truffle: add not-in-frame assumption feature.Thu, 26 Sep 2013 13:47:12 +0200, by Andreas Woess
-
improve NodeUtil.cloneNode.Wed, 25 Sep 2013 17:34:01 +0200, by Andreas Woess
-
include isValid in Assumption.toString(); minor javadoc fix.Wed, 25 Sep 2013 17:31:55 +0200, by Andreas Woess
-
fix endless recursion in _find_eclipse_wsroot on windows when eclipse workspace is not in repository path.Wed, 25 Sep 2013 16:16:54 +0200, by Andreas Woess
-
Use 32 byte code alignment for Graal on x86.Thu, 26 Sep 2013 13:17:48 +0200, by Roland Schatz
-
Make AbstractBeginNode a IterableNodeType and use this in ControlFlowGraphThu, 26 Sep 2013 11:15:45 +0200, by Gilles Duboscq
-
LoweringPhase: add commentThu, 26 Sep 2013 08:50:55 +0200, by Bernhard Urban
-
HSAIL, SPARC: fix some occurrences of object identity, using equals() insteadWed, 25 Sep 2013 17:07:33 +0200, by Bernhard Urban
-
SnippetTemplate: avoid two getNodes() in assertions. comment fix and renamesWed, 25 Sep 2013 16:55:03 +0200, by Bernhard Urban
-
TruffleCache: do not cut off ControlFlowException constructors.Thu, 26 Sep 2013 03:04:38 +0200, by Andreas Woess
-
Improvements to recent unsafe cast node changesWed, 25 Sep 2013 17:40:58 -0700, by Christian Wimmer
-
SPARC: added CodeInstaller::pd_relocate_pollWed, 25 Sep 2013 13:26:02 -0700, by twisti
-
made lowering recursive instead of iterativeWed, 25 Sep 2013 21:49:39 +0200, by Doug Simon
-
be verbose when the JDK specified by --installed-jdks is missingWed, 25 Sep 2013 21:48:38 +0200, by Doug Simon
-
Merge.Wed, 25 Sep 2013 21:25:06 +0200, by Doug Simon
-
Merge.Wed, 25 Sep 2013 12:22:07 +0200, by Doug Simon
-
Merge.Tue, 24 Sep 2013 15:35:59 +0200, by Doug Simon
-
removed creation of intermediate LoadHubNode when lowering StoreIndexedNodeFri, 20 Sep 2013 11:25:53 +0200, by Doug Simon
-
removed creation of intermediate ArrayLengthNode when lowering [Load|Store]IndexedNodesThu, 19 Sep 2013 23:51:37 +0200, by Doug Simon
-
Remove unused AbstractCallNodeWed, 25 Sep 2013 13:11:10 +0200, by Gilles Duboscq
-
StubForeignCallNode is not a DeoptimizingNodeWed, 25 Sep 2013 13:05:16 +0200, by Gilles Duboscq
-
Nodes that extend AbstractStateSplit do not need to implement StateSplit againWed, 25 Sep 2013 12:12:45 +0200, by Gilles Duboscq
-
Make InfopointNode a NodeWithState, it is not a StateSplit anymoreWed, 25 Sep 2013 12:08:28 +0200, by Gilles Duboscq
-
Introduce NodeWithState for nodes that hold some VirtualState. Use this interface in the required special cases (Scheduling and PEA)Wed, 25 Sep 2013 12:08:04 +0200, by Gilles Duboscq
-
AMD64HotSpotLIRGenerator: ignore InfopointNodes with AFTER_BCIWed, 25 Sep 2013 11:16:17 +0200, by Gilles Duboscq
-
Fix SnippetFrameStateCleanupPhase's handling of Merges: if there is no statesplit after a merge, the last statesplit inside the merges's incomming branches should get an AFTER_BCITue, 24 Sep 2013 16:21:59 +0200, by Gilles Duboscq
-
add "sun4u" as a valid identifier for the SPARC architectureWed, 25 Sep 2013 12:17:55 +0200, by Doug Simon
-
make GuardingPiNode canonicalize to a deopt when the guard is guaranteed to failWed, 25 Sep 2013 12:17:21 +0200, by Doug Simon
-
add support for a node to canonicalize itself to a ControlSinkNodeWed, 25 Sep 2013 12:16:26 +0200, by Doug Simon
-
Rewrite write barrier addition phaseWed, 25 Sep 2013 10:20:11 +0200, by Christos Kotselidis
-
Refactor write barriersWed, 25 Sep 2013 10:08:14 +0200, by Christos Kotselidis
-
Move barriers in hotspot specific packageWed, 25 Sep 2013 09:58:05 +0200, by Christos Kotselidis
-
MergeTue, 24 Sep 2013 21:37:53 -0700, by Mick Jordan
-
move pylint to mxtoolTue, 24 Sep 2013 21:36:46 -0700, by Mick Jordan
-
WriteBarrierSnipppets: make locations visible to other packagesTue, 24 Sep 2013 21:58:04 +0200, by Bernhard Urban
-
Finished PTX assembler and Register -> Variable conversionTue, 24 Sep 2013 14:24:47 -0400, by Morris Meyer
-
G1 Barriers: don't use ANY_LOCATIONTue, 24 Sep 2013 16:08:07 +0200, by Bernhard Urban
-
NewMultiArrayNode: use INIT_LOCATION instead of ANY_LOCATIONTue, 24 Sep 2013 14:11:38 +0200, by Bernhard Urban