Commit Graph

650 Commits

Author SHA1 Message Date
Dmitrij Pochepko
df5a5904ce 8141331: Test hotspot/compiler/oracle/MethodMatcher.java fails with NPE
Reviewed-by: iignatyev
2015-11-04 00:59:44 +03:00
Chris Plummer
d78d0c4fa4 8140189: [TESTBUG] Get rid of "@library /../../test/lib" in jtreg tests
Use new external.lib.roots property in TEST.ROOT so /../../test/lib is not needed.

Reviewed-by: mseledtsov, sla, iklam
2015-10-29 12:04:04 -07:00
Gerard Ziemski
684ae3fa6e 8140539: compiler/membars/DekkerTest.java fails with -XX:CICompilerCount=1
Add back "-XX:+IgnoreUnrecognizedVMOptions" that was removed in a prototype.

Reviewed-by: dcubed, ddmitriev
2015-10-28 09:09:55 -05:00
Chris Hegarty
ca4787d77b 8139891: Prepare Unsafe for true encapsulation
Reviewed-by: alanb, dholmes, jrose, psandoz, twisti
2015-10-27 14:18:56 +00:00
Gerard Ziemski
f84561609b 8129855: "-XX:+IgnoreUnrecognizedVMOptions" hides out of range VM options
Implement strict spec rgarding how IgnoreUnrecognizedVMOptions works, supplied with the corresponding test case.

Reviewed-by: dcubed, ddmitriev
2015-10-23 11:17:23 -05:00
Christian Tornqvist
de06de1d7b 8140243: [TESTBUG] Exclude compiler/jvmci/compilerToVM/GetConstantPoolTest.java
Reviewed-by: gtriantafill, kvn
2015-10-21 09:47:23 -07:00
Alejandro Murillo
cfbb592424 Merge 2015-10-19 12:30:17 -07:00
Igor Ignatyev
c1408bb0c8 8139707: [TESTBUG] Quarantine unstable compiler/jvmci tests
Reviewed-by: twisti
2015-10-16 02:05:54 +03:00
Konstantin Shefov
a4cb6b3eb4 8072369: [TESTBUG] Remove potentially insecure class cast in some hotspot tests
Reviewed-by: twisti, kvn, iignatyev, tpivovarova
2015-10-15 18:00:00 +03:00
Dmitrij Pochepko
72de147944 8139438: [TESTBUG] JVMCI test fails with RuntimeException: Has no virtual object before materialization
Reviewed-by: iignatyev, twisti
2015-10-15 02:46:51 +03:00
Tatiana Pivovarova
d3ad64d9d6 8139375: [TESTBUG] compiler/jvmci/SecurityRestrictionsTest checks are too tight
Reviewed-by: twisti, iignatyev
2015-10-15 01:58:28 +03:00
Igor Ignatyev
815be268ee 8139376: [TESTBUG] ExecuteInstalledCodeTest should be run only on amd64 and sparcv9
Reviewed-by: twisti, kvn
2015-10-13 16:21:25 +03:00
Chris Hegarty
1a83e59745 8138978: Examine usages of sun.misc.IOUtils
Reviewed-by: alanb, mullan, psandoz, rriggs, weijun
2015-10-09 14:21:33 +01:00
Christian Thalinger
16526e000e 8136421: JEP 243: Java-Level JVM Compiler Interface
Reviewed-by: ihse, alanb, roland, coleenp, iveresov, kvn, kbarrett
2015-10-08 12:49:30 -10:00
Tobias Hartmann
b9bf98abb2 8139048: Quarantine compiler/startup/SmallCodeCacheStartup.java
Quarantine the test because it fails on JPRT for the CPU, CompactStrings and JVMCI repositories.

Reviewed-by: roland
2015-10-08 08:54:57 +02:00
Felix Yang
12cba200a3 8138583: aarch64: add support for vectorizing fabs/fneg
Reviewed-by: aph, roland
2015-09-30 04:35:39 -04:00
Ed Nevill
cb8cff7e6e 8135231: aarch64: add support for vectorizing double precision sqrt
Reviewed-by: roland, aph
2015-09-23 12:39:30 -04:00
Pavel Punegov
4af470b866 8066157: JEP-JDK-8046155: Test task: method matcher
MethodMatcher test and framework

Reviewed-by: iignatyev, neliasso
2015-09-24 20:13:04 +03:00
Roland Westrelin
92af9b015c 8136926: phi == NULL assert in PhaseIdealLoop::try_move_store_after_loop
Multiple phis on same slice are possible in a loop

Reviewed-by: kvn, mcberg
2015-09-22 15:25:34 +02:00
Zoltan Majo
a61a0cec2e 8136914: compiler/loopopts/superword/SumRedSqrt_Double.java times out
Change test to execute only on relevant (x86-based) platforms.

Reviewed-by: kvn, dlong
2015-09-22 13:42:09 +02:00
Tatiana Pivovarova
89e62a6274 8136693: add package statement 'compiler.testlibrary' to CompilerUtils
Reviewed-by: iignatyev
2015-09-19 12:03:36 +03:00
Nils Eliasson
5ca8983920 8135068: Extract method matchers from CompilerOracle
Ecapsulate code to enable reuse

Reviewed-by: roland, kvn
2015-09-18 10:11:11 +02:00
Roland Westrelin
c55d212196 8136461: PhaseIdealLoop::try_move_store_before_loop() may bypass early loop exit
PhaseIdealLoop::try_move_store_before_loop() needs to check for early loop exit before candidate Stores

Reviewed-by: kvn
2015-09-15 13:08:43 +02:00
Roland Westrelin
404fc5caa8 8134974: 8130847 broken with loop predicates
Pinned eliminated arraycopy loads in uncommon trap path for loop predicates may need to be moved

Reviewed-by: kvn
2015-09-08 19:19:08 +02:00
Roland Westrelin
8dd169a71f 8135069: C2 replaces range checks by unsigned comparison with -1
I < 0 || i > -1 wrongly folded as i >u -1

Reviewed-by: kvn
2015-09-11 16:56:56 +02:00
Igor Veresov
e88940fae6 Merge 2015-09-10 17:56:43 -07:00
Michael Berg
e75f5a5cde 8135028: support for vectorizing double precision sqrt
Reviewed-by: kvn, twisti
2015-09-09 10:34:17 -07:00
Jesper Wilhelmsson
1736e104a1 Merge 2015-09-08 16:10:37 +02:00
Ahmed Khawaja
d7b8032741 8132081: C2 support for Adler32 on SPARC
Add C2 instrinsic support for Adler32 checksum on SPARC.

Reviewed-by: kvn
2015-09-03 15:03:12 -07:00
Dmitry Samersoff
6ed285af0a Merge 2015-08-31 21:46:33 +03:00
Roland Westrelin
ed9e8b9ff2 8134031: Incorrect JIT compilation of complex code with inlining and escape analysis
Bad rewiring of memory edges when we split unique types during EA

Reviewed-by: kvn
2015-08-31 17:20:08 +02:00
Roland Westrelin
7ef7c96fde 8134468: Lucene test failures with 32 bit JDK 9b78, Server compiler
Test that checks whether an array load falls into the range of an arraycopy is incorrect on 32bits

Reviewed-by: iveresov, kvn
2015-09-02 10:00:27 +02:00
Roland Westrelin
9dcbe322e0 8134288: compiler/runtime/6859338/Test6859338.java crashes in PhaseIdealLoop::try_move_store_after_loop
Store nodes may not have a control if used to update profiling

Reviewed-by: kvn
2015-08-26 12:43:15 +02:00
Roland Westrelin
8942ef410f 8134321: tools/pack200/Pack200Test.java crashes in the VM PIT jdk9 b79
Code that capture field values of eliminated allocation at a safepoint when there's an arraycopy behind a Phi is broken

Reviewed-by: kvn
2015-08-26 09:49:37 +02:00
Gerard Ziemski
e085f49adb 8134239: compiler/arguments/CheckCICompilerCount.java still fails
Compiler team updated the test by adding a new test case, but due to merging timing it was still using the old constraint error message format, which needed updating

Reviewed-by: ctornqvi, hseigel
2015-08-24 16:17:15 -05:00
Roland Westrelin
a9e232a8ef Merge 2015-08-21 09:12:42 +02:00
Roland Westrelin
33b2496286 Merge 2015-08-20 09:31:28 +02:00
Roland Westrelin
b7b1b2778c 8131969: jit/FloatingPoint/gen_math/Loops05 assert(2 <= size && size <= 16) failed: update low bits table
Assert in register allocation code when vector Phi for a loop is processed because code assumes all inputs already processed

Reviewed-by: kvn
2015-08-18 16:10:34 +02:00
Gerard Ziemski
a00b32c815 8133678: test fails due to 'CICompilerCount=0 must be at least 1' missing from stdout/stderr
We changed the constraint error message formatting for CICompilerCount, but forgot to change the corresponding test.

Reviewed-by: kbarrett, dcubed
2015-08-16 09:33:52 -05:00
Zoltan Majo
6d858b63cb Merge 2015-08-19 10:14:04 +02:00
Zoltan Majo
1e55e60cb1 8076373: In 32-bit VM interpreter and compiled code process NaN values differently
Change interpreter to use XMM registers on x86_32 if they are available. Add stubs for methods transforming from/to int/long float/double.

Reviewed-by: kvn, mcberg
2015-08-19 08:55:18 +02:00
Roland Westrelin
0baf2f7e8a 8130847: Cloned object's fields observed as null after C2 escape analysis
Eliminated instance/array written to by an array copy variant must be correctly initialized when reallocated at a deopt

Reviewed-by: kvn, vlivanov
2015-08-15 02:54:18 +02:00
Roland Westrelin
96b004cd82 8080289: Intermediate writes in a loop not eliminated by optimizer
Move Stores out of loop (after or before) when possible

Reviewed-by: kvn, vlivanov
2015-07-29 17:25:04 +02:00
Roland Westrelin
c030d4e363 8132525: java -client -XX:+TieredCompilation -XX:CICompilerCount=1 -version asserts since 8130858
TieredCompilation not validated when CICompilerCount is processed

Reviewed-by: kvn, gziemski
2015-07-29 12:33:48 +02:00
Zoltan Majo
422cd18f23 8130832: Extend the WhiteBox API to provide information about the availability of compiler intrinsics
Add a new method, sun.hotspot.WhiteBox.isIntrinsicAvailable, that can be used to determine if an intrinsic is available.

Reviewed-by: kvn, jrose
2015-07-28 19:20:42 +02:00
Roland Westrelin
ae7cdb5972 8130858: CICompilerCount=1 when tiered is off is not allowed any more
8122937 broke handling of CICompilerCount

Reviewed-by: kvn, vlivanov, gziemski
2015-07-27 13:44:07 +02:00
Roland Westrelin
51ddedd5c7 Merge 2015-07-23 19:11:28 +02:00
Daniel D. Daugherty
d7f565d9eb 8075171: Contended Locking fast notify bucket
JEP-143/JDK-8073165 Contended Locking fast notify bucket

Co-authored-by: Dave Dice <dave.dice@oracle.com>
Co-authored-by: Karen Kinnear <karen.kinnear@oracle.com>
Reviewed-by: dholmes, acorn, dice, dcubed
2015-07-21 07:28:37 -07:00
Vladimir Ivanov
92d8281194 8078629: VM should constant fold Unsafe.get*() loads from final fields
Reviewed-by: kvn, jrose, psandoz
2015-07-14 06:44:50 -07:00
Anthony Scarpino
963c3852a3 8130341: GHASH 32bit intrinsics has AEADBadTagException
Reviewed-by: kvn, mcberg
2015-07-10 11:31:49 -07:00