| Tue, 13 May 2014 19:19:27 +0200 |
Thomas Wuerthinger |
Merge. |
| Tue, 13 May 2014 12:15:40 +0200 |
Thomas Wuerthinger |
Merge. |
| Tue, 13 May 2014 12:15:32 +0200 |
Thomas Wuerthinger |
Fix exponential explosion when propagating zero probabilities. |
| Tue, 13 May 2014 03:57:06 +0200 |
Thomas Wuerthinger |
Stop propagating probability above loop header. |
| Tue, 13 May 2014 02:31:41 +0200 |
Thomas Wuerthinger |
Merge. |
| Tue, 13 May 2014 02:31:20 +0200 |
Thomas Wuerthinger |
Propagate 0.0 probabilities when simplifying IfNode. |
| Sat, 24 May 2014 00:54:20 +0200 |
Doug Simon |
added clarifying comment |
| Sat, 24 May 2014 00:46:33 +0200 |
Doug Simon |
HotSpotOptions.inline.hpp generator writes to System.out to make generator errors more visible (they will show up when compiling the generated source) |
| Sat, 24 May 2014 00:25:29 +0200 |
Doug Simon |
HotSpotOptions error messages should go to System.err |
| Sat, 24 May 2014 00:23:52 +0200 |
Doug Simon |
mx: drain all subprocess output to callables before returning from mx.run |
| Fri, 23 May 2014 19:36:34 +0200 |
Doug Simon |
made HotSpotOptions processing faster by removing use of service loader in VM startup and only doing work for options specified on the command line |
| Fri, 23 May 2014 17:47:44 +0200 |
Lukas Stadler |
small cleanup in ReadElimination |
| Fri, 23 May 2014 17:47:15 +0200 |
Lukas Stadler |
consume less memory in ReentrantBlockIterator and ReentrantNodeIterator |
| Fri, 23 May 2014 17:43:07 +0200 |
Lukas Stadler |
read elimination without schedule |
| Fri, 23 May 2014 14:51:59 +0200 |
Gilles Duboscq |
CanonicalizerPhase: canonicalize usages when stamp changes |
| Fri, 23 May 2014 13:21:49 +0200 |
Gilles Duboscq |
Simplify removeOrMaterializeIf and make it handle merges with more predecessors. |
| Fri, 23 May 2014 13:22:54 +0200 |
Bernhard Urban |
midtier: remove ReadEliminationPhase (superseded by EarlyReadEliminationPhase) |
| Fri, 23 May 2014 11:50:47 +0200 |
Bernhard Urban |
mx: add pack200 to javaconfig |
| Fri, 23 May 2014 11:25:01 +0200 |
Bernhard Urban |
mx: update vm choice helptext |
| Fri, 23 May 2014 13:10:14 +0200 |
Doug Simon |
mx: annotation processor paths must include libraries that are also Eclipse containers |
| Thu, 22 May 2014 14:15:01 +0200 |
Miguel Garcia |
[inlining] check maxMethodPerInlining after discarding methods below threshold |
| Thu, 22 May 2014 11:18:11 +0200 |
Miguel Garcia |
[inlining] more precise type in createDispatchOnTypeBeforeInvoke() |
| Wed, 21 May 2014 21:47:19 +0200 |
Miguel Garcia |
[inlining] isEmpty() favored over size() == 0 |
| Wed, 21 May 2014 20:51:25 +0200 |
Miguel Garcia |
[inlining] forgotten assertion, counterpart to the one in pushGraph() |
| Wed, 21 May 2014 20:49:07 +0200 |
Miguel Garcia |
[inlining] operation that pushes invocation goes ahead and pushes graphs too |
| Wed, 21 May 2014 20:04:33 +0200 |
Miguel Garcia |
[inlining] readability in CallsiteHolder constructor, part 2 |
| Wed, 21 May 2014 19:49:48 +0200 |
Miguel Garcia |
[inlining] readability in CallsiteHolder constructor, part 1 |
| Thu, 22 May 2014 16:51:32 +0200 |
Lukas Stadler |
only canonicalize CustomizedUnsafeLoadFinalNode if the condition is constant |
| Thu, 22 May 2014 16:51:09 +0200 |
Lukas Stadler |
put LoopSafepointEliminationPhase into an IncrementalCanonicalizerPhase |
| Thu, 22 May 2014 14:04:55 +0200 |
Lukas Stadler |
preserve context in Debug.forceLog |
| Thu, 22 May 2014 14:04:55 +0200 |
Lukas Stadler |
small cleanup in LinearScan |
| Thu, 22 May 2014 14:04:55 +0200 |
Lukas Stadler |
skip PiNode in AMD64HotSpotNodeLIRBuilder.filterCompression |
| Thu, 22 May 2014 14:04:55 +0200 |
Lukas Stadler |
fix bug in ConditionalEliminationPhase that loses the connection from guard to checkcast PiNode |
| Thu, 22 May 2014 14:04:55 +0200 |
Lukas Stadler |
make FixedNodeProbabilityCache behave better in the presence of dead code |
| Thu, 22 May 2014 13:19:47 +0200 |
twisti |
HSAIL: fix for -UseHSAILDeoptimization |
| Thu, 22 May 2014 12:41:53 +0200 |
twisti |
mx: fixed incorrect test for subprocess being alive |
| Wed, 21 May 2014 22:44:46 -0700 |
Tom Rodriguez |
minor cleanups |
| Wed, 21 May 2014 22:22:06 -0700 |
Tom Rodriguez |
rename initializeBytecode to getBytecode and eliminate extra copy |
| Wed, 21 May 2014 21:07:15 -0700 |
Michael Van De Vanter |
Truffle: revise instrumentation support APIs in ExecutionContext |
| Wed, 21 May 2014 10:25:54 -0700 |
Christian Wimmer |
Add test for getStaticFields() |
| Wed, 21 May 2014 10:08:39 -0700 |
Christian Wimmer |
Merge |
| Tue, 20 May 2014 19:06:41 -0700 |
Christian Wimmer |
Add method ResolvedJavaType.getStaticFields |
| Tue, 20 May 2014 19:02:33 -0700 |
Christian Wimmer |
Pass the compiled method to LIR factory |
| Tue, 20 May 2014 19:01:08 -0700 |
Christian Wimmer |
Rename methods to have consistent names, allow subclasses of bytecode parsers |
| Tue, 20 May 2014 18:59:45 -0700 |
Christian Wimmer |
Introduce class BarrieredAccess for low-level object access with read and write barriers |
| Tue, 20 May 2014 18:58:42 -0700 |
Christian Wimmer |
Remove overly restrictive assertion |
| Tue, 20 May 2014 18:58:03 -0700 |
Christian Wimmer |
Introduce method to customize type size |
| Tue, 20 May 2014 18:57:27 -0700 |
Christian Wimmer |
Avoid NullPointerException when only some assertions are enabled |
| Tue, 20 May 2014 18:56:57 -0700 |
Christian Wimmer |
Remove overly strict assertion; avoid NullPointerException when canonicalizing invokes without a state |
| Tue, 20 May 2014 18:55:59 -0700 |
Christian Wimmer |
Make classes extensible |
| Tue, 20 May 2014 18:55:35 -0700 |
Christian Wimmer |
Bugfix: as long as snippets are preprocessed, PiNode must not be canonicalized |
| Tue, 20 May 2014 18:54:48 -0700 |
Christian Wimmer |
Add accessor method |
| Tue, 20 May 2014 18:54:09 -0700 |
Christian Wimmer |
Cache result of toJava and toJavaConstructor, since it is an expensive operation |
| Tue, 20 May 2014 18:53:31 -0700 |
Christian Wimmer |
Make NodeClass more flexible |
| Tue, 20 May 2014 18:52:54 -0700 |
Christian Wimmer |
Factor out rule creation in its own method |
| Tue, 20 May 2014 18:51:54 -0700 |
Christian Wimmer |
Recompute probability only when number of types in profile changed |
| Tue, 20 May 2014 18:51:16 -0700 |
Christian Wimmer |
Remove unused method |
| Wed, 21 May 2014 17:57:41 +0200 |
Doug Simon |
mx: fixed more spurious "error while killing subprocess" messages (GRAAL-350) |
| Wed, 21 May 2014 16:41:57 +0200 |
Bernhard Urban |
mx archive: fix log message |
| Wed, 21 May 2014 15:34:35 +0200 |
Bernhard Urban |
mx archive: avoid multiple directory entries too |
| Wed, 21 May 2014 15:29:38 +0200 |
Bernhard Urban |
mx: remove unused packagejar command |
| Wed, 21 May 2014 15:11:14 +0200 |
Bernhard Urban |
mx: add support for setting a main class in distributions |
| Wed, 21 May 2014 15:17:21 +0200 |
Bernhard Urban |
mxtool: fix archive subcommand such that it will return a successful returncode |
| Tue, 20 May 2014 17:26:27 +0200 |
Roland Schatz |
Introduce AbstractObjectStamp, make ObjectStamp and NarrowOopStamp incompatible. |
| Tue, 20 May 2014 16:47:19 +0200 |
Roland Schatz |
Allow using StampFactory.forConstant(Constant, MetaAccessProvider) for primitive constants. |
| Tue, 20 May 2014 15:46:02 +0200 |
Miguel Garcia |
[inlining] another renaming to avoid misleading type suggestion (2 of 2) |
| Tue, 20 May 2014 15:45:07 +0200 |
Miguel Garcia |
[inlining] another renaming to avoid misleading type suggestion (1 of 2) |
| Tue, 20 May 2014 15:42:10 +0200 |
Miguel Garcia |
[inlining] renaming to convey underlying types (2 of 2) |
| Tue, 20 May 2014 15:39:59 +0200 |
Miguel Garcia |
[inlining] renaming to convey underlying types (1 of 2) |
| Tue, 20 May 2014 15:01:20 +0200 |
Miguel Garcia |
[inline-info] step 3, InlineInfo leaves populateInlineInfo fully initialized |
| Tue, 20 May 2014 14:53:29 +0200 |
Miguel Garcia |
[inline-info] step 2, simpler inter-procedural communication |
| Tue, 20 May 2014 14:48:46 +0200 |
Miguel Garcia |
[inline-info] step 1 of de-aliasing MethodInvocation assumptions |
| Tue, 20 May 2014 14:39:42 +0200 |
Miguel Garcia |
[inline-info] towards initializing InlineInfo in one place |
| Tue, 20 May 2014 21:40:13 -0700 |
Tom Rodriguez |
handle expected phis when converting to trapping null checks |
| Tue, 20 May 2014 21:35:32 -0700 |
Tom Rodriguez |
don't use JNI natives to interact with VM metadata |
| Tue, 20 May 2014 13:46:34 -0700 |
Tom Rodriguez |
initialize HotSpotVMConfig fields efficiently from C++ |
| Tue, 20 May 2014 21:38:31 +0200 |
Miguel Garcia |
Merge |
| Tue, 20 May 2014 13:05:07 +0200 |
Miguel Garcia |
[inlining-7] end of refactoring trail, helper methods now closer to users |
| Tue, 20 May 2014 13:00:23 +0200 |
Miguel Garcia |
[inlining-7] InlineableGraph takes care of setup chores during construction |
| Tue, 20 May 2014 12:15:55 +0200 |
Miguel Garcia |
[inlining-7] moved three utilities methods to where they belong |
| Tue, 20 May 2014 12:08:09 +0200 |
Miguel Garcia |
[inlining-6] InlineableGraph now in package for inlineable elements |
| Tue, 20 May 2014 12:06:07 +0200 |
Miguel Garcia |
[inlining-6] InlineableMacroNode now in package for inlineable elements |
| Tue, 20 May 2014 12:04:10 +0200 |
Miguel Garcia |
[inlining-6] moved Inlineable to dedicated package for inlineable elements |
| Mon, 19 May 2014 21:39:49 +0200 |
Miguel Garcia |
[inlining-5] separate check code (fewer args, pure, concise) from logging code |
| Mon, 19 May 2014 21:30:07 +0200 |
Miguel Garcia |
[inlining-5] "where does optimisticOpts come from?" answered |
| Mon, 19 May 2014 21:26:46 +0200 |
Miguel Garcia |
[inlining-5] "where does replacements come from?" answered |
| Mon, 19 May 2014 21:23:53 +0200 |
Miguel Garcia |
[inlining-5] checkTargetConditions() about to lose some of its formal params |
| Tue, 20 May 2014 20:32:05 +0200 |
Gilles Duboscq |
Ignore synthetic methods in TestResolvedJavaField |
| Tue, 20 May 2014 16:36:37 +0200 |
Gilles Duboscq |
Update JaCoCo libs |
| Tue, 20 May 2014 16:17:16 +0200 |
Lukas Stadler |
better stamps for IntegerRemNode |
| Tue, 20 May 2014 15:28:53 +0200 |
Lukas Stadler |
@Ignore (and not expect GraalInternalError) long-running EscapingNewStringConcatTest |
| Tue, 20 May 2014 15:17:43 +0200 |
Lukas Stadler |
a bit of javadoc in TruffleRuntime |
| Tue, 20 May 2014 15:16:33 +0200 |
Lukas Stadler |
fix wrong assert in ObjectStampMeetTest |
| Tue, 20 May 2014 15:16:04 +0200 |
Lukas Stadler |
simplify getInterfaces jtt test |
| Tue, 20 May 2014 15:35:35 +0200 |
Doug Simon |
mx: propagate failure from forked Java compilation task back up to parent (GRAAL-350) |
| Tue, 20 May 2014 11:56:19 +0200 |
Josef Eisl |
Add JRE library support to mx projectgraph. |
| Tue, 20 May 2014 13:13:28 +0200 |
Gilles Duboscq |
ResolvedJavaType.resolveMethod: fix javadoc, add assert in native code. update changelog |
| Tue, 20 May 2014 12:40:07 +0200 |
Erik Eckstein |
Merge |
| Tue, 20 May 2014 12:38:57 +0200 |
Erik Eckstein |
fix wrong handling of memory anti-dependencies in scheduler |
| Tue, 20 May 2014 11:50:30 +0200 |
Doug Simon |
mx: fixed spurious "error while killing subprocess" messages (GRAAL-350) |
| Mon, 19 May 2014 17:21:30 -0700 |
Michael Van De Vanter |
Merge with 9ae1d2f3bda60f9d91243c883c5aa7812e2ab256 |
| Mon, 19 May 2014 17:14:36 -0700 |
Michael Van De Vanter |
Truffle/Instrumentation: significant reorganization of the instrumentation framework's implementation and connection to the runtime ExecutionContext, with some new features, including a Tag-based "trap" mechanisms. |
| Mon, 19 May 2014 14:14:12 -0700 |
Tom Rodriguez |
convert asserts into if tests and check for phis at merge |
| Mon, 19 May 2014 10:45:01 -0700 |
twisti |
enable some HSAIL virtual call inlining |
| Mon, 19 May 2014 18:59:04 +0200 |
Christian Wirth |
SL: adding NodeInfo.descriptions to SL statements |
| Mon, 19 May 2014 18:52:39 +0200 |
Christian Wirth |
Add description and language to the NodeInfo annotation |
| Mon, 19 May 2014 17:21:35 +0200 |
Miguel Garcia |
Merge |
| Mon, 19 May 2014 16:52:10 +0200 |
Miguel Garcia |
[inlining-4] privatizing methods that can be made private |
| Mon, 19 May 2014 16:45:11 +0200 |
Miguel Garcia |
[inlining-4] one less alias in getTypeCheckedInlineInfo() |
| Mon, 19 May 2014 16:34:38 +0200 |
Miguel Garcia |
[inlining-4] one less alias in getAssumptionInlineInfo() |
| Mon, 19 May 2014 16:32:02 +0200 |
Miguel Garcia |
[inlining-4] one less alias in getExactInlineInfo() |
| Mon, 19 May 2014 16:27:24 +0200 |
Miguel Garcia |
[inlining-4] removed alias for InliningData.maxMethodPerInlining |
| Mon, 19 May 2014 16:24:31 +0200 |
Miguel Garcia |
[inlining-4] getTypeCheckedInlineInfo() can get context.getReplacements() itself |
| Mon, 19 May 2014 16:17:43 +0200 |
Miguel Garcia |
[inlining-4] getAssumptionInlineInfo() can get context.getReplacements() itself |
| Mon, 19 May 2014 16:14:56 +0200 |
Miguel Garcia |
[inlining-4] no need to pass context.getReplacements() to getExactInlineInfo() |
| Mon, 19 May 2014 16:10:24 +0200 |
Miguel Garcia |
[inlining-4] parameter aliasing context.getOptimisticOptimizations() goes away |
| Mon, 19 May 2014 16:08:06 +0200 |
Miguel Garcia |
[inlining-4] parameter aliasing context.getReplacements() goes away |
| Mon, 19 May 2014 16:05:18 +0200 |
Miguel Garcia |
[inlining-4] the method param that aliased maxMethodPerInlining goes away |
| Mon, 19 May 2014 15:58:48 +0200 |
Miguel Garcia |
[inlining-4] getTypeInlineInfo() becomes instance method of InliningData |
| Mon, 19 May 2014 15:39:43 +0200 |
Miguel Garcia |
[inlining-4] getTypeCheckedInlineInfo() becomes instance method of InliningData |
