Mercurial > hg > graal-jvmci-8
graph
-
Only report snippet cache overflow onceThu, 19 Mar 2015 17:22:24 -0700, by Tom Rodriguez
-
added null checking for the receiver when inlining or applying an InvocationPlugin during graph parsingThu, 19 Mar 2015 22:31:42 +0100, by Doug Simon
-
TraceDeoptimization: print compiler and compile_idThu, 19 Mar 2015 19:27:25 +0100, by Gilles Duboscq
-
Small fix for memory schedule verification. The begin node is not always the first node in the block if it is at the same time also the end node.Thu, 19 Mar 2015 15:38:33 +0100, by Thomas Wuerthinger
-
Merge.Thu, 19 Mar 2015 12:47:06 +0100, by Doug Simon
-
Guard constant folding with !ImmutableCode.Thu, 19 Mar 2015 11:53:36 +0100, by Thomas Wuerthinger
-
Merge.Thu, 19 Mar 2015 11:22:37 +0100, by Thomas Wuerthinger
-
Merge.Wed, 18 Mar 2015 23:26:47 +0100, by Thomas Wuerthinger
-
Fix an issue when the result of canonicalization is appended in the graph builder. Introduce GraphBuilderPhase#recursiveAppend.Wed, 18 Mar 2015 23:26:23 +0100, by Thomas Wuerthinger
-
Merge.Wed, 18 Mar 2015 21:36:35 +0100, by Thomas Wuerthinger
-
Fix a bug in the schedule phase and tighten asserts.Wed, 18 Mar 2015 21:36:12 +0100, by Thomas Wuerthinger
-
Canonicalizer: tryCanonicalize: return early if the custom canonicalizer has simplified the node awayThu, 19 Mar 2015 10:28:28 +0100, by Gilles Duboscq
-
removed logic for estimating invocation plugin countWed, 18 Mar 2015 21:01:56 +0100, by Doug Simon
-
Dom-based ConditionalElimination: fix fliped check in processConditionAnchorWed, 18 Mar 2015 20:17:41 +0100, by Gilles Duboscq
-
Properly handle Guard usages during on stack replacementWed, 18 Mar 2015 11:38:38 -0700, by Tom Rodriguez
-
MergeWed, 18 Mar 2015 10:07:47 -0700, by Tom Rodriguez
-
Ignore -Xbatch when the thread is in the middle of class loadingWed, 18 Mar 2015 10:01:25 -0700, by Tom Rodriguez
-
Relax asserts in ArrayEqualsNodesTue, 17 Mar 2015 19:24:47 -0700, by Tom Rodriguez
-
Experimentally disallow Graal compiling itself in tieredTue, 17 Mar 2015 19:21:57 -0700, by Tom Rodriguez
-
Add option to print CompileQueue occupancy from LogCompilation outputTue, 17 Mar 2015 19:21:54 -0700, by Tom Rodriguez
-
Log dequeue of methods from CompileQueueTue, 17 Mar 2015 19:21:22 -0700, by Tom Rodriguez
-
MergeWed, 18 Mar 2015 16:39:06 +0100, by Stefan Anzinger
-
[SPARC] (Re)move SPARCScratchRegister into the SPARCMacroAssembler to force usage only during assembly.Wed, 18 Mar 2015 15:23:39 +0100, by Stefan Anzinger
-
MergeWed, 18 Mar 2015 13:53:23 +0100, by Stefan Anzinger
-
[SPARC] Set lower plugin count estimate for SPARC backendTue, 17 Mar 2015 18:57:47 +0100, by Stefan Anzinger
-
FloatStamp.Abs.foldStamp support known-NaN inputWed, 18 Mar 2015 15:58:55 +0100, by Gilles Duboscq
-
CanonicalizerPhase: can not use Debug.handle without a corresponding scope/sandboxWed, 18 Mar 2015 15:51:15 +0100, by Gilles Duboscq
-
Add assert in FloatStampWed, 18 Mar 2015 15:14:27 +0100, by Gilles Duboscq
-
Implement foldStamp for Float to Integer opsWed, 08 Oct 2014 16:20:38 +0200, by Gilles Duboscq
-
Implement foldStamp for Integer to Float conversionsWed, 08 Oct 2014 15:27:10 +0200, by Gilles Duboscq
-
Remove TailDuplicationPhaseWed, 18 Mar 2015 14:36:13 +0100, by Gilles Duboscq
-
GuardsStage: add areDeoptsFixed helper methodTue, 17 Mar 2015 16:13:54 +0100, by Gilles Duboscq
-
BenchmarkCounters: add option to turn of number grouping separators.Wed, 18 Mar 2015 11:30:45 +0100, by Josef Eisl
-
removed overly strict assertions in InstanceOfDynamicNode constructorThu, 19 Mar 2015 12:46:38 +0100, by Doug Simon
-
added support for checking nodes added to the graph by an Invocation plugin and used this to check that only legal constants are added under ImmutableCodeThu, 19 Mar 2015 12:46:06 +0100, by Doug Simon
-
use cheaper test for whether or not a method has a substitutionWed, 18 Mar 2015 12:39:43 +0100, by Doug Simon
-
factored out assertion checks around application of an InvocationPlugin and added a check that all StateSplit nodes added by a plugin have a non-null frame stateWed, 18 Mar 2015 11:46:06 +0100, by Doug Simon
-
Truffle: implemented new -G:+PrintTruffleExpansionHistogram tool.Wed, 18 Mar 2015 02:15:37 +0100, by Christian Humer
-
Truffle: truffle compilation might result in a CancellationException.Wed, 18 Mar 2015 02:17:34 +0100, by Christian Humer
-
Merge.Wed, 18 Mar 2015 01:46:01 +0100, by Thomas Wuerthinger
-
Add memory proxy support in the new MemoryScheduleVerification.Wed, 18 Mar 2015 01:45:49 +0100, by Thomas Wuerthinger
-
MergeTue, 17 Mar 2015 11:23:06 -0700, by Tom Rodriguez
-
Fix arg limit on MacTue, 17 Mar 2015 11:22:56 -0700, by Tom Rodriguez
-
Make the native initialization path for Graal more robustMon, 16 Mar 2015 17:37:32 -0700, by Tom Rodriguez
-
Don't walk out of loop when estimating unswitch benefitsMon, 16 Mar 2015 16:38:17 -0700, by Tom Rodriguez
-
More checks that InductionVariable.direction is validMon, 16 Mar 2015 15:51:43 -0700, by Tom Rodriguez
-
Don't force resolution for unloaded field typesMon, 16 Mar 2015 15:46:28 -0700, by Tom Rodriguez
-
Do not include MoveProfiling phases in stub compilation.Mon, 16 Mar 2015 18:19:12 +0100, by Josef Eisl
-
LIRSuites: add copy constructor.Mon, 16 Mar 2015 16:38:01 +0100, by Josef Eisl
-
LIRPhaseSuite: add copy().Mon, 16 Mar 2015 16:37:21 +0100, by Josef Eisl
-
Add MoveProfiling.Mon, 16 Mar 2015 17:23:22 +0100, by Josef Eisl
-
Add BenchmarkCounterFactory to PostAllocationOptimizationContext.Mon, 16 Mar 2015 17:15:52 +0100, by Josef Eisl
-
Make DynamicCounterNode LIRLowerable.Mon, 16 Mar 2015 17:36:45 +0100, by Josef Eisl
-
{AMD64,SPARC}HotSpotLIRGenerator: implement createBenchmarkCounter().Tue, 17 Mar 2015 17:19:04 +0100, by Josef Eisl
-
Add AMD64HotSpotCounterOp and SPARCHotSpotCounterOp.Tue, 17 Mar 2015 17:18:47 +0100, by Josef Eisl
-
AMD64Assembler: introduce addq(AMD64Address, Register).Mon, 16 Mar 2015 16:35:20 +0100, by Josef Eisl
-
Make LIRGeneratorTool a BenchmarkCounterFactory.Tue, 17 Mar 2015 17:19:39 +0100, by Josef Eisl
-
Introduce BenchmarkCounterFactory.Tue, 17 Mar 2015 17:19:18 +0100, by Josef Eisl
-
BenchmarkCounters: make thread-safe and refactoring.Tue, 17 Mar 2015 17:29:49 +0100, by Josef Eisl
-
NewObjectSnippets: work around problems with late binding of snippet parameters in GraphBuilderPlugins.Tue, 17 Mar 2015 13:07:12 +0100, by Josef Eisl
-
Merge.Tue, 17 Mar 2015 17:16:04 +0100, by Thomas Wuerthinger
-
Add another memory schedule test.Tue, 17 Mar 2015 17:15:40 +0100, by Thomas Wuerthinger
-
Fix another bug in the new scheduler found by the new verification phase.Tue, 17 Mar 2015 17:13:25 +0100, by Thomas Wuerthinger
-
Assign result maps in schedule phase earlier to have better context in case of exceptions.Tue, 17 Mar 2015 17:12:47 +0100, by Thomas Wuerthinger
-
Add memory schedule verification pass. Fix a bug in the new schedule implementation.Tue, 17 Mar 2015 16:17:30 +0100, by Thomas Wuerthinger
-
Merge.Tue, 17 Mar 2015 13:14:13 +0100, by Thomas Wuerthinger
-
Fix for earliest possible schedule when the last node in a block is an invoke. Added a new scheduling test.Tue, 17 Mar 2015 13:13:57 +0100, by Thomas Wuerthinger
-
Merge.Mon, 16 Mar 2015 21:50:15 +0100, by Thomas Wuerthinger
-
Add GraalHostThreads flag to overwrite the number of threads spawned for the baseline compiler.Mon, 16 Mar 2015 21:49:59 +0100, by Thomas Wuerthinger
-
remove unused code in CurrentJavaThreadNodeTue, 17 Mar 2015 15:52:41 +0100, by Andreas Woess
-
Truffle: repair TraceTruffleTransferToInterpreter optionTue, 17 Mar 2015 15:49:33 +0100, by Andreas Woess
-
Truffle: try to inline method handle invocations in FastPETue, 17 Mar 2015 15:39:46 +0100, by Andreas Woess
-
MethodHandleNode: record unique concrete method assumptionsTue, 17 Mar 2015 15:36:09 +0100, by Andreas Woess
-
Truffle: respect truffle inlining option in FastPE and replace Stack with ArrayDequeTue, 17 Mar 2015 14:40:35 +0100, by Andreas Woess
-
Merge.Tue, 17 Mar 2015 13:57:33 +0100, by Doug Simon
-
handle proxies with null inputs in SchedulePhaseTue, 17 Mar 2015 13:27:50 +0100, by Lukas Stadler
-
add null check in Materialize graphbuilder pluginTue, 17 Mar 2015 11:18:55 +0100, by Lukas Stadler
-
fix faulty check in HotSpotTruffleRuntimeTue, 17 Mar 2015 11:11:03 +0100, by Lukas Stadler
-
remove obsolete assertion from MonitorExitNodeTue, 17 Mar 2015 11:10:32 +0100, by Lukas Stadler
-
removed placeholder codeTue, 17 Mar 2015 13:45:05 +0100, by Doug Simon
-
Merge.Tue, 17 Mar 2015 12:49:45 +0100, by Doug Simon
-
MergeTue, 17 Mar 2015 12:05:51 +0100, by Stefan Anzinger
-
[SPARC] remove debug code for mmap()Tue, 17 Mar 2015 12:05:11 +0100, by Stefan Anzinger
-
MergeFri, 13 Mar 2015 11:28:01 +0100, by Stefan Anzinger
-
[SPARC] Add com.oracle.nfi.test.NativeFunctionInterfaceTest to blacklist for SPARCThu, 12 Mar 2015 17:08:19 +0100, by Stefan Anzinger
-
Merge.Tue, 17 Mar 2015 12:13:24 +0100, by Doug Simon
-
Merge with f803f49c9ec423cb1c87a510c29fd61afd6a8c57Mon, 16 Mar 2015 16:54:10 -0700, by Michael Van De Vanter
-
Truffle/Instrumentation: remove support for "probe-lite", an optimization for a particular use case that no longer seems worth the complexity.Mon, 16 Mar 2015 15:59:57 -0700, by Michael Van De Vanter
-
Truffle: fixed TruffleCompilationExceptionsAreThrown was ignored for non permanent bailouts.Mon, 16 Mar 2015 13:47:45 +0100, by Christian Humer
-
Refactoring of AMD64 code generation.Mon, 16 Mar 2015 15:12:22 +0100, by Roland Schatz
-
Make LIRGeneratorTool.append(op) return the op.Mon, 16 Mar 2015 14:17:00 +0100, by Roland Schatz
-
Allow arbitrary constants in CompilationResultBuilder.recordInlineDataInCode.Fri, 13 Mar 2015 15:23:44 +0100, by Roland Schatz
-
Move isCompressed method to HotSpotConstant interface.Fri, 13 Mar 2015 15:20:16 +0100, by Roland Schatz
-
mx: followup fix for SC_ARG_MAX, add more assetionsMon, 16 Mar 2015 12:23:11 +0100, by Gilles Duboscq
-
Support optional last access location in the scheduler phase.Sun, 15 Mar 2015 16:22:37 +0100, by Thomas Wuerthinger
-
Remove fine grained debug scopes in the canonicalizer.Sun, 15 Mar 2015 14:22:50 +0100, by Thomas Wuerthinger
-
Remove metric ConstantNodes. Make LocalLiveness log output only available when assertions are enabled.Sun, 15 Mar 2015 14:16:21 +0100, by Thomas Wuerthinger
-
Add utility Node#pushInputs(NodeStack).Sun, 15 Mar 2015 13:40:28 +0100, by Thomas Wuerthinger
-
Remove dead code in GraphBuilderPhase.Sat, 14 Mar 2015 22:45:52 +0100, by Thomas Wuerthinger
-
Fix javadoc.Sat, 14 Mar 2015 22:38:12 +0100, by Thomas Wuerthinger
-
Consolidation of new allocation stubs to use the standard ForeignCallNode.Sat, 14 Mar 2015 22:29:44 +0100, by Thomas Wuerthinger
-
Turn a check for non-null stateAfter of an invoke into an assertion.Sat, 14 Mar 2015 20:42:09 +0100, by Thomas Wuerthinger
-
Assign the state after when appending the node and not automatically in the main graph builder loop.Sat, 14 Mar 2015 19:15:51 +0100, by Thomas Wuerthinger
-
Small simplification in GraphBuilderPhase.Sat, 14 Mar 2015 17:59:33 +0100, by Thomas Wuerthinger
-
Reduce # of MemoryCheckpoint#Multi.Sat, 14 Mar 2015 14:14:54 +0100, by Thomas Wuerthinger
-
Simplified NamedLocationIdentity #hashCode and #equals.Sat, 14 Mar 2015 14:05:16 +0100, by Thomas Wuerthinger
-
Relax check invariants for usage of ==.Sat, 14 Mar 2015 14:03:59 +0100, by Thomas Wuerthinger
-
Add a unit test for the new ResolvedJavaField API method getLocationIdentity.Sat, 14 Mar 2015 13:19:16 +0100, by Thomas Wuerthinger
-
Fix WriteBarrierAdditionTest.Sat, 14 Mar 2015 12:14:26 +0100, by Thomas Wuerthinger
-
Merge.Sat, 14 Mar 2015 02:04:40 +0100, by Thomas Wuerthinger
-
Rewrite gathering of loop kill information of FloatingReadPhase to reduce set allocations.Sat, 14 Mar 2015 01:57:26 +0100, by Thomas Wuerthinger
-
Create utilities LocationIdentity#isAny, LocationIdentity#isSingle, LocationIdentity#any, LocationIdentity#overlaps.Sat, 14 Mar 2015 01:28:20 +0100, by Thomas Wuerthinger
-
Convert LocationIdentity from an interface into an abstract class.Sat, 14 Mar 2015 01:09:21 +0100, by Thomas Wuerthinger
-
Remove LocationIdentity interface from ResolvedJavaField and add ResolvedJavaField#getLocationIdentity method instead.Sat, 14 Mar 2015 01:02:08 +0100, by Thomas Wuerthinger
-
Merge.Sat, 14 Mar 2015 00:24:40 +0100, by Doug Simon
-
inline intrinsics during bytecode paring only if -G:+Inline or -G:+InlineDuringParsingSat, 14 Mar 2015 00:23:48 +0100, by Doug Simon
-
disabled crypto intrinsics until partial intrinsification (i.e., self recursive intrinsics) bug is fixedFri, 13 Mar 2015 23:46:12 +0100, by Doug Simon
-
don't treat Truffle method inlining substitutions as graph builder replacementsFri, 13 Mar 2015 23:27:53 +0100, by Doug Simon
-
only do eager resolving for replacements that are intrinsicsFri, 13 Mar 2015 23:15:18 +0100, by Doug Simon
-
Small data structure optimizations in SchedulePhase.Fri, 13 Mar 2015 22:59:50 +0100, by Thomas Wuerthinger