Vladimir Kozlov
063286a618
Merge
2017-05-12 13:14:25 -07:00
Zoltan Majo
f3a162f330
8178723: Workaround for failure of CRC32C intrinsic on x86 machines without CLMUL support (JDK-8178720)
...
Disable CRC32C intrinsic on affected machines. Improve tests.
Co-authored-by: Lutz Schmidt <lutz.schmidt@sap.com>
Reviewed-by: kvn, simonis, mdoerr, aph
2017-04-19 14:37:11 +02:00
Jamsheed Mohammed C M
6f2cad0c0c
8173679: Disable ProfileTrap code and UseRTMLocking in emulated client Win32
...
Disabled mdo trap count update on deopt, and made +UseRTMLocking to exit.
Reviewed-by: kvn
2017-02-03 19:26:35 -08:00
Michael Berg
56ed7077a8
8170039: Change UseAVX < 3 on SKX/KNL to not emit evex vector check
...
Reviewed-by: kvn
2016-11-22 11:34:11 -08:00
Vivek Deshpande
7a8af5acc8
8178811: Minimize the AVX <-> SSE transition penalty through generation of vzeroupper instruction on x86
...
Reviewed-by: kvn
2017-05-05 19:28:54 -07:00
Dmitry Chuyko
e70bbbd57f
8067744: XMM/SSE float register values corrupted by JNI_CreateVM call in JRE 8 (Windows)
...
Save/restore registers in generate_get_cpu_info() code which checks interrupts.
Reviewed-by: kvn, mcberg
2016-10-28 12:28:46 -07:00
Smita Kamath
2bc0337093
8165381: Update for x86 SHA512 using AVX2
...
Add intrinsics for x86 AVX2 architecture with no SHA instructions.
Reviewed-by: kvn
2016-10-21 10:16:09 -07:00
Vivek Deshpande
d58e3e0324
8154122: Intrinsify fused mac operations
...
Added FMA intrinsics on x86
Reviewed-by: kvn, aph, darcy
2016-08-26 12:17:50 -07:00
Tobias Hartmann
f92cc0c836
8158214: Crash with "assert(VM_Version::supports_sse4_1()) failed" if UseSSE < 4 is set
...
Do not emit unsupported SSE 4.1 instructions in CRC32 intrinsic.
Reviewed-by: kvn, zmajo
2016-06-02 13:19:05 +02:00
Jan Civlin
f2e9b827d6
8154495: SHA256 AVX2 intrinsic (when no supports_sha() available)
...
Reviewed-by: kvn
2016-04-26 21:54:21 -07:00
Coleen Phillimore
0d3e7977ae
8151939: VM_Version_init() print buffer is too small
...
Increase buffer size, use logging to print out version and os information
Reviewed-by: kvn, rprotacio, mockner
2016-04-07 16:37:35 -04:00
Mikael Vidstedt
d4b4dca194
8151002: Make Assembler methods vextract and vinsert match actual instructions
...
Reviewed-by: kvn, vlivanov, mcberg
2016-03-07 15:03:48 -08:00
Vivek Deshpande
40fe96de0d
8150767: Enables SHA Extensions on x86
...
Add x86 intrinsics for SHA-1 and SHA-256.
Co-authored-by: Shravya Rukmannagari <shravya.rukmannagari@intel.com>
Reviewed-by: kvn, twisti
2016-03-03 22:02:13 -08:00
Zoltan Majo
477c40e4eb
8146478: Node limit exceeded with -XX:AllocateInstancePrefetchLines=1073741823
...
Constrain the set of accepted values for the Allocate{PrefetchLines, InstancePrefetchLines, PrefetchStepSize, PrefetchDistance} flags. Increase macro node expansion budget.
Reviewed-by: kvn
2016-01-28 08:33:45 +01:00
Kishor Kharbas
a8fcb62829
8146581: Minor corrections to the patch submitted for earlier bug id - 8143925
...
Reviewed-by: kvn
2016-01-07 14:29:05 -08:00
Kishor Kharbas
790f5bded4
8143925: Enhancing CounterMode.crypt() for AES
...
Add intrinsic for CounterMode.crypt() to leverage the parallel nature of AES in Counter(CTR) Mode.
Reviewed-by: kvn, ascarpino
2015-12-28 23:11:01 -08:00
Andrew Haley
cee2a179e6
8143072: [JVMCI] Port JVMCI to AArch64
...
Reviewed-by: gdub, rschatz, twisti, kvn
2015-12-23 20:19:42 -10:00
Vladimir Kozlov
223668aab4
Merge
2015-12-14 10:27:33 -08:00
Jamsheed Mohammed
1f2a9c1407
6808665: Use486InstrsOnly aborts 32-bit VM
...
The code supporting -XX:+/-Use486InstrsOnly was removed.
Reviewed-by: dholmes, thartmann, vlivanov
2015-12-09 11:06:39 +01:00
Vivek R Deshpande
2d9a6cfd3f
8143355: Update for addition of vectorizedMismatch intrinsic for x86
...
Co-authored-by: Liqi Yi <liqi.yi@intel.com>
Reviewed-by: kvn
2015-12-07 16:35:07 -08:00
Zoltan Majo
d0382697c4
8143208: compiler/c1/6855215/Test6855215.java supports_sse2() failed
...
Check that SSE is enabled before using SSE instructions in string-related intrinsics. Make the UseSSE42Intrinsics dependent on UseSSE >= 4.
Reviewed-by: kvn
2015-11-23 08:11:30 +01:00
Konstantin Shefov
6de50f10f5
8131778: java disables UseAES flag when using VIS=2 on sparc
...
Reviewed-by: iignatyev, kvn
2015-11-13 18:14:41 +03:00
Michael C Berg
e86e38619e
8140779: Code generation fixes for avx512
...
Assembler refactoring. Abstract away instruction attributes.
Reviewed-by: kvn, roland, iveresov
2015-11-09 11:26:41 -08:00
Zoltan Majo
fc2a5e9d53
8078554: Compiler: implement ranges (optionally constraints) for those flags that have them missing
...
Add range check or constraint where necessary.
Reviewed-by: roland, thartmann
2015-10-09 14:21:26 +02:00
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
Tomasz Wojtowicz
61b77b8590
8134553: CRC32C implementations for x86/x64 targets
...
Reviewed-by: kvn
2015-09-16 15:54:32 -07:00
Michael Berg
d67924dc8e
8132160: support for AVX 512 call frames and stack management
...
Simplify save/restore frame on x86 systems which support EVEX.
Reviewed-by: kvn, iveresov
2015-09-11 17:02:44 -07:00
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
Zoltan Majo
547a40e75a
8130120: Handling of SHA intrinsics inconsistent across platforms
...
Introduce common warning message and common processing of SHA intrinsic-related arguments.
Reviewed-by: kvn, mcberg
2015-07-03 09:33:04 +02:00
Andrew Haley
917f70414f
Merge
2015-07-02 11:12:59 +01:00
Andrew Haley
52b991b411
8130150: Implement BigInteger.montgomeryMultiply intrinsic
...
Add montgomeryMultiply intrinsics
Reviewed-by: kvn
2015-06-16 17:31:53 +01:00
James Cheng
e2533553f6
8073583: C2 support for CRC32C on SPARC
...
Reviewed-by: jrose, kvn
2015-06-29 00:10:01 -07:00
Anthony Scarpino
c4c528df14
8073108: Use x86 and SPARC CPU instructions for GHASH acceleration
...
Reviewed-by: kvn, jrose
2015-06-17 17:48:25 -07:00
Vladimir Kozlov
93d1573c5e
8081778: Use Intel x64 CPU instructions for RSA acceleration
...
Add intrinsics for BigInteger squareToLen and mulAdd methods.
Reviewed-by: kvn, jrose
2015-06-03 15:02:10 -07:00
Michael C Berg
4fca8dbb1f
8076276: Add support for AVX512
...
Reviewed-by: kvn, roland
2015-05-08 11:49:20 -07:00
Vladimir Kozlov
b9c00b1904
Merge
2015-04-16 14:05:48 -07:00
Thomas Stuefe
704c02a180
8074860: Structured Exception Catcher missing around CreateJavaVM on Windows
...
Add __try/__except around JNI_CreateJavaVM
Reviewed-by: mgronlun, dcubed
2015-04-03 05:55:33 -04:00
Andrew Haley
1dfbc44c1f
8026049: (bf) Intrinsify ByteBuffer.put{Int, Double, Float, ...} methods
...
Use unaligned Unsafe loads and stores for ByteBuffer access on platforms which support unaligned access. Add intrinsics for Unsafe.{get,put}-X-Unaligned methods.
Reviewed-by: dholmes, jrose, psandoz, kvn
2015-03-31 12:31:18 -07:00
Paul Sandoz
c002e71565
8068977: Remove unused sun.misc.Unsafe prefetch intrinsic support
...
Reviewed-by: kvn, vlivanov
2015-02-10 21:32:05 +01:00
Vladimir Kozlov
44d1787a26
8059139: It should be possible to explicitly disable usage of TZCNT instr w/ -XX:-UseBMI1Instructions
...
Reviewed-by: iveresov
2014-10-17 15:35:25 -07:00
Vladimir Kozlov
b9e949183d
8055494: Add C2 x86 intrinsic for BigInteger::multiplyToLen() method
...
Add new C2 intrinsic for BigInteger::multiplyToLen() on x86 in 64-bit VM.
Reviewed-by: roland
2014-09-02 12:48:45 -07:00
Vladimir Kozlov
f813683519
8055069: TSX and RTM should be deprecated more strongly until hardware is corrected
...
Require to specify UnlockExperimentalVMOptions flag together with UseRTMLocking flag on un-patched systems where CPUID allows it but is unsupported otherwise.
Reviewed-by: iveresov, fzhinkin
2014-08-22 12:03:49 -07:00
Zhengyu Gu
8a690a1250
6424123: JVM crashes on failed 'strdup' call
...
Calling os::malloc()/os::strdup() and new os::strdup_check_oom() instead of ::malloc()/::strdup() for native memory tracking purpose
Reviewed-by: coleenp, ctornqvi, kvn
2014-08-11 10:18:09 -07:00
Vladimir Kozlov
7764490363
8052081: Optimize generated by C2 code for Intel's Atom processor
...
Allow to execute vectorization and crc32 optimization on Atom. Enable UseFPUForSpilling by default on x86.
Reviewed-by: roland
2014-08-05 15:02:10 -07:00
Daniel D. Daugherty
a06d36cada
8049717: expose L1_data_cache_line_size for diagnostic/sanity checks
...
Add support for VM_Version::L1_data_cache_line_size().
Reviewed-by: dsimms, kvn, dholmes
2014-07-15 07:33:49 -07:00
Goetz Lindenmaier
f2051ed6fa
8048241: Introduce umbrella header os.inline.hpp and clean up includes
...
Reviewed-by: coleenp, dholmes, lfoltan
2014-06-26 16:05:15 +02:00
James Cheng
395560c428
8035968: Leverage CPU Instructions to Improve SHA Performance on SPARC
...
Add C2 SHA intrinsics on SPARC
Reviewed-by: kvn, roland
2014-06-11 11:05:10 -07:00
David Chase
305ec3bd3f
8037816: Fix for 8036122 breaks build with Xcode5/clang
...
Repaired or selectively disabled offending formats; future-proofed with additional checking
Reviewed-by: kvn, jrose, stefank
2014-05-09 16:50:54 -04:00
Vladimir Kozlov
d9056b18ee
8041957: -XX:UseAVX=0 cause assert(UseAVX) failed
...
Temporary set UseAVX=1 and UseSSE=2 in generate_get_cpu_info()
Reviewed-by: twisti
2014-04-29 12:20:53 -07:00
Vladimir Kozlov
77d38feb95
8038633: crash in VM_Version::get_processor_features() on startup
...
Windows need an exception wrapper around getPsrInfo_stub() call in order to properly handle SEGV for YMM registers test.
Reviewed-by: iveresov, iignatyev
2014-03-31 13:08:03 -07:00