Mercurial > hg > graal-jvmci-8
graph
-
less canonicalization during InliningPhaseWed, 28 May 2014 17:47:12 +0200, by Lukas Stadler
-
NodeWorkList refactoringWed, 28 May 2014 17:20:35 +0200, by Lukas Stadler
-
NodeBitMap refactoringWed, 28 May 2014 17:19:41 +0200, by Lukas Stadler
-
long values and scale on DebugHistogramWed, 28 May 2014 17:19:41 +0200, by Lukas Stadler
-
cache last receiver stamp in MethodCallTargetNodeWed, 28 May 2014 17:14:24 +0200, by Lukas Stadler
-
ignore transient fields in NodeClassWed, 28 May 2014 17:14:24 +0200, by Lukas Stadler
-
grow NodeMaps exponentiallyWed, 28 May 2014 17:14:24 +0200, by Lukas Stadler
-
Interface to do graph verification after High/Mid/LowTier in unittests.Wed, 28 May 2014 12:17:16 +0200, by Roland Schatz
-
Fix insertion of profile data in unit tests.Wed, 28 May 2014 12:17:09 +0200, by Roland Schatz
-
Ensure values stay finite in block probability computation.Wed, 28 May 2014 12:17:00 +0200, by Roland Schatz
-
Remove unused oop compression code from backends.Wed, 28 May 2014 12:16:44 +0200, by Roland Schatz
-
Remove isCompressible flags from memory access nodes.Wed, 28 May 2014 12:15:50 +0200, by Roland Schatz
-
Merge with 5c73b162eec248fc2d06f59d8f25860871a21be5Tue, 27 May 2014 21:20:01 -0700, by Michael Van De Vanter
-
Truffle: additional methods on ExecutionContextTue, 27 May 2014 21:18:45 -0700, by Michael Van De Vanter
-
Truffle: NPE guard in InstrumentationNodeTue, 27 May 2014 21:18:16 -0700, by Michael Van De Vanter
-
reduced execution time of ReplacementsImple.registerSubstitutions() by deferring parsing of substitution classes until the first request for a substitute method is receivedWed, 28 May 2014 00:50:11 +0200, by Doug Simon
-
added comments to explain the origin of generated sourcesTue, 27 May 2014 22:00:41 +0200, by Doug Simon
-
renamed project (and package) com.oracle.graal.hotspot.codegen to com.oracle.graal.hotspot.sourcegenTue, 27 May 2014 22:00:22 +0200, by Doug Simon
-
added ${workspace}/com.oracle.graal.hotspot/src_gen/hotspot to include pathsTue, 27 May 2014 21:20:36 +0200, by Doug Simon
-
LSRA optimization: activate by default.Tue, 27 May 2014 16:27:03 +0200, by Josef Eisl
-
LSRA optimization: assign location to intervals.Tue, 27 May 2014 15:25:50 +0200, by Josef Eisl
-
LSRA optimization: split intervals at block boundaries.Mon, 26 May 2014 19:21:55 +0200, by Josef Eisl
-
LSRA optimization: check if optimization is feasible.Mon, 26 May 2014 16:57:16 +0200, by Josef Eisl
-
LSRA: skip handled intervals in IntervalWalker.updateUnhandledStackIntervals(int).Mon, 26 May 2014 16:35:59 +0200, by Josef Eisl
-
LSRA: add Interval.getIntervalCoveringOpId(int).Mon, 26 May 2014 15:44:57 +0200, by Josef Eisl
-
LSRA optimization: add spilled intervals to unhandled list.Mon, 26 May 2014 15:12:14 +0200, by Josef Eisl
-
LSRA: move stack intervals to active list.Mon, 26 May 2014 15:11:25 +0200, by Josef Eisl
-
LSRA optimization: add LinearScanWalker.handleSpillSlot().Mon, 26 May 2014 12:06:05 +0200, by Josef Eisl
-
LSRA Optimization: add support for stack intervals.Mon, 26 May 2014 11:47:45 +0200, by Josef Eisl
-
LSRA Optimization: walk basic block boundaries.Mon, 26 May 2014 09:32:51 +0200, by Josef Eisl
-
LSRA: add OptimizingLinearScanWalker.Mon, 26 May 2014 09:29:51 +0200, by Josef Eisl
-
reduced time to initialize ForeignCallProviders by avoiding triggering class initialization of Node subclasses as well as making annotation parsing lazy in SnippetInfoTue, 27 May 2014 15:43:36 +0200, by Doug Simon
-
mxtool: remove python downloaderTue, 27 May 2014 13:16:12 +0200, by Bernhard Urban
-
BciBlockMapping: allocate smaller array if possibleTue, 27 May 2014 13:04:29 +0200, by Bernhard Urban
-
computeBlockOrder: no need to check if block is active, since it's anywayTue, 27 May 2014 12:15:31 +0200, by Bernhard Urban
-
mx: add verbose mode to download helperTue, 27 May 2014 12:14:54 +0200, by Bernhard Urban
-
Support for compressed constants in HSAIL backend.Tue, 27 May 2014 12:08:11 +0200, by Roland Schatz
-
backout 88a6017687c9Tue, 27 May 2014 10:22:28 +0200, by Josef Eisl
-
Merge.Mon, 26 May 2014 20:27:28 +0200, by Doug Simon
-
fixed code generation error for debug buildsMon, 26 May 2014 19:54:15 +0200, by Doug Simon
-
Merge.Mon, 26 May 2014 18:55:21 +0200, by Doug Simon
-
removed use of ServiceLoader in runtime initializationMon, 26 May 2014 18:55:06 +0200, by Doug Simon
-
added more runtime initialization timersMon, 26 May 2014 17:44:19 +0200, by Doug Simon
-
LSRA: fix getMaterializedValue() (respect MustHaveRegister priorities).Wed, 21 May 2014 18:40:48 +0200, by Josef Eisl
-
Use new jacocoreport versionMon, 26 May 2014 18:18:19 +0200, by Gilles Duboscq
-
Optimize compare compressed pattern.Mon, 26 May 2014 17:43:45 +0200, by Roland Schatz
-
Use correct stamp when creating ConstantNode.Mon, 26 May 2014 16:16:47 +0200, by Roland Schatz
-
Support direct comparison of compressed pointers.Mon, 26 May 2014 16:13:58 +0200, by Roland Schatz
-
Support for compressed constants.Mon, 26 May 2014 16:09:53 +0200, by Roland Schatz
-
Allow signed values in Buffer.emit(Byte|Short).Fri, 23 May 2014 17:21:37 +0200, by Roland Schatz
-
mx: option to force a GC after each unit testFri, 23 May 2014 11:44:18 +0200, by Roland Schatz
-
hotspot eclipse project: add generated folders for client and server buildsMon, 26 May 2014 12:03:04 +0200, by Gilles Duboscq
-
mx: use python downloader if stderr is not a tty to avoid spamming logsMon, 26 May 2014 12:01:30 +0200, by Gilles Duboscq
-
LSRA: remove IntervalWalker.currentInterval and change the behavior of nextInterval and walkTo(int).Thu, 22 May 2014 20:36:31 +0200, by Josef Eisl
-
LSRA: make IntervalWalker.currentInterval private.Thu, 22 May 2014 19:42:02 +0200, by Josef Eisl
-
LSRA: restrict access to IntervalWalker members.Thu, 22 May 2014 23:02:46 +0200, by Josef Eisl
-
tests for NodeMapMon, 26 May 2014 13:37:21 +0200, by Lukas Stadler
-
explicit getAndGrow and setAndGrow functionality on NodeMapMon, 26 May 2014 13:37:11 +0200, by Lukas Stadler
-
mx: create Eclipse projects for distributionsMon, 26 May 2014 13:02:29 +0200, by Doug Simon
-
mx: prevent CTRL-C from being blocked while subprocess is running (re-applied without unrelated changes)Mon, 26 May 2014 12:55:17 +0200, by Doug Simon