rev 56639 : loosen a couple more counter checks due to races observed in testing; simplify om_release() extraction of mid since list head or cur_mid_in_use is marked; simplify deflate_monitor_list() extraction of mid since there are no parallel deleters due to the safepoint; simplify deflate_monitor_list_using_JT() extraction of mid since list head or cur_mid_in_use is marked; prepend_block_to_lists() - simplify based on David H's comments; does not need load_acquire() or release_store() because of the cmpxchg(); prepend_to_common() - simplify to use mark_next_loop() for m and use mark_list_head() and release_store() for the non-empty list case; add more debugging for "Non-balanced monitor enter/exit" failure mode; fix race in inflate() in the "CASE: neutral" code path; install_displaced_markword_in_object() does not need to clear the header field since that is handled when the ObjectMonitor is moved from the global free list; LSuccess should clear boxReg to set ICC.ZF=1 to avoid depending on existing boxReg contents; update fast_unlock() to detect when object no longer refers to the same ObjectMonitor and take fast path exit instead; clarify fast_lock() code where we detect when object no longer refers to the same ObjectMonitor; add/update comments for movptr() calls where we move a literal into an Address; remove set_owner(); refactor setting of owner field into set_owner_from(2 versions), set_owner_from_BasicLock(), and try_set_owner_from(); the new functions include monitorinflation+owner logging; extract debug code from v2.06 and v2.07 and move to v2.07.debug; change 'jccb' -> 'jcc' and 'jmpb' -> 'jmp' as needed; checkpoint initial version of MacroAssembler::inc_om_ref_count(); update LP64 MacroAssembler::fast_lock() and fast_unlock() to use inc_om_ref_count(); fast_lock() return flag setting logic can use 'testptr(tmpReg, tmpReg)' instead of 'cmpptr(tmpReg, 0)' since that's more efficient; fast_unlock() LSuccess return flag setting logic can use 'testl (boxReg, 0)' instead of 'xorptr(boxReg, boxReg)' since that's more efficient; cleanup "fast-path" vs "fast path" and "slow-path" vs "slow path"; update MacroAssembler::rtm_inflated_locking() to use inc_om_ref_count(); update MacroAssembler::fast_lock() to preserve the flags before decrementing ref_count and restore the flags afterwards; this is more clean than depending on the contents of rax/tmpReg; coleenp CR - refactor async monitor deflation work from ServiceThread::service_thread_entry() to ObjectSynchronizer::deflate_idle_monitors_using_JT(); rehn,eosterlund CR - add support for HandshakeAfterDeflateIdleMonitors for platforms that don't have ObjectMonitor ref_count support implemented in C2 fast_lock() and fast_unlock().

   1 /*
   2  * Copyright (c) 2002, 2019, Oracle and/or its affiliates. All rights reserved.
   3  * Copyright (c) 2012, 2018 SAP SE. All rights reserved.
   4  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
   5  *
   6  * This code is free software; you can redistribute it and/or modify it
   7  * under the terms of the GNU General Public License version 2 only, as
   8  * published by the Free Software Foundation.
   9  *
  10  * This code is distributed in the hope that it will be useful, but WITHOUT
  11  * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
  12  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
  13  * version 2 for more details (a copy is included in the LICENSE file that
  14  * accompanied this code).
  15  *
  16  * You should have received a copy of the GNU General Public License version
  17  * 2 along with this work; if not, write to the Free Software Foundation,
  18  * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
  19  *
  20  * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
  21  * or visit www.oracle.com if you need additional information or have any
  22  * questions.
  23  *
  24  */
  25 
  26 #ifndef CPU_PPC_GLOBALS_PPC_HPP
  27 #define CPU_PPC_GLOBALS_PPC_HPP
  28 
  29 #include "utilities/globalDefinitions.hpp"
  30 #include "utilities/macros.hpp"
  31 
  32 // Sets the default values for platform dependent flags used by the runtime system.
  33 // (see globals.hpp)
  34 
  35 define_pd_global(bool, ImplicitNullChecks,    true);  // Generate code for implicit null checks.
  36 define_pd_global(bool, TrapBasedNullChecks,   true);
  37 define_pd_global(bool, UncommonNullCast,      true);  // Uncommon-trap NULLs passed to check cast.
  38 
  39 #define DEFAULT_STACK_YELLOW_PAGES (2)
  40 #define DEFAULT_STACK_RED_PAGES (1)
  41 // Java_java_net_SocketOutputStream_socketWrite0() uses a 64k buffer on the
  42 // stack if compiled for unix and LP64. To pass stack overflow tests we need
  43 // 20 shadow pages.
  44 #define DEFAULT_STACK_SHADOW_PAGES (20 DEBUG_ONLY(+2))
  45 #define DEFAULT_STACK_RESERVED_PAGES (1)
  46 
  47 #define MIN_STACK_YELLOW_PAGES DEFAULT_STACK_YELLOW_PAGES
  48 #define MIN_STACK_RED_PAGES DEFAULT_STACK_RED_PAGES
  49 #define MIN_STACK_SHADOW_PAGES (3 DEBUG_ONLY(+1))
  50 #define MIN_STACK_RESERVED_PAGES (0)
  51 
  52 define_pd_global(intx, StackYellowPages,      DEFAULT_STACK_YELLOW_PAGES);
  53 define_pd_global(intx, StackRedPages,         DEFAULT_STACK_RED_PAGES);
  54 define_pd_global(intx, StackShadowPages,      DEFAULT_STACK_SHADOW_PAGES);
  55 define_pd_global(intx, StackReservedPages,    DEFAULT_STACK_RESERVED_PAGES);
  56 
  57 // Use large code-entry alignment.
  58 define_pd_global(uintx, CodeCacheSegmentSize,  128);
  59 define_pd_global(intx,  CodeEntryAlignment,    128);
  60 define_pd_global(intx,  OptoLoopAlignment,     16);
  61 define_pd_global(intx,  InlineFrequencyCount,  100);
  62 define_pd_global(intx,  InlineSmallCode,       1500);
  63 
  64 // Flags for template interpreter.
  65 define_pd_global(bool, RewriteBytecodes,      true);
  66 define_pd_global(bool, RewriteFrequentPairs,  true);
  67 
  68 define_pd_global(bool, PreserveFramePointer,  false);
  69 
  70 // GC Ergo Flags
  71 define_pd_global(size_t, CMSYoungGenPerWorker, 16*M);  // Default max size of CMS young gen, per GC worker thread.
  72 
  73 define_pd_global(uintx, TypeProfileLevel, 111);
  74 
  75 define_pd_global(bool, CompactStrings, true);
  76 
  77 // 2x unrolled loop is shorter with more than 9 HeapWords.
  78 define_pd_global(intx, InitArrayShortSize, 9*BytesPerLong);
  79 
  80 define_pd_global(bool, ThreadLocalHandshakes, true);



  81 
  82 // Platform dependent flag handling: flags only defined on this platform.
  83 #define ARCH_FLAGS(develop,      \
  84                    product,      \
  85                    diagnostic,   \
  86                    experimental, \
  87                    notproduct,   \
  88                    range,        \
  89                    constraint,   \
  90                    writeable)    \
  91                                                                             \
  92   product(uintx, PowerArchitecturePPC64, 0,                                 \
  93           "CPU Version: x for PowerX. Currently recognizes Power5 to "      \
  94           "Power8. Default is 0. Newer CPUs will be recognized as Power8.") \
  95                                                                             \
  96   product(bool, SuperwordUseVSX, false,                                     \
  97           "Use Power8 VSX instructions for superword optimization.")        \
  98                                                                             \
  99   /* Reoptimize code-sequences of calls at runtime, e.g. replace an */      \
 100   /* indirect call by a direct call.                                */      \
 101   product(bool, ReoptimizeCallSequences, true,                              \
 102           "Reoptimize code-sequences of calls at runtime.")                 \
 103                                                                             \
 104   /* Power 8: Configure Data Stream Control Register. */                    \
 105   product(uint64_t,DSCR_PPC64, (uintx)-1,                                   \
 106           "Power8 or later: Specify encoded value for Data Stream Control " \
 107           "Register")                                                       \
 108   product(uint64_t,DSCR_DPFD_PPC64, 8,                                      \
 109           "Power8 or later: DPFD (default prefetch depth) value of the "    \
 110           "Data Stream Control Register."                                   \
 111           " 0: hardware default, 1: none, 2-7: min-max, 8: don't touch")    \
 112   product(uint64_t,DSCR_URG_PPC64, 8,                                       \
 113           "Power8 or later: URG (depth attainment urgency) value of the "   \
 114           "Data Stream Control Register."                                   \
 115           " 0: hardware default, 1: none, 2-7: min-max, 8: don't touch")    \
 116                                                                             \
 117   product(bool, UseLoadInstructionsForStackBangingPPC64, false,             \
 118           "Use load instructions for stack banging.")                       \
 119                                                                             \
 120   /* special instructions */                                                \
 121                                                                             \
 122   product(bool, UseCountLeadingZerosInstructionsPPC64, true,                \
 123           "Use count leading zeros instructions.")                          \
 124                                                                             \
 125   product(bool, UseCountTrailingZerosInstructionsPPC64, false,              \
 126           "Use count trailing zeros instructions.")                         \
 127                                                                             \
 128   product(bool, UseExtendedLoadAndReserveInstructionsPPC64, false,          \
 129           "Use extended versions of load-and-reserve instructions.")        \
 130                                                                             \
 131   product(bool, UseRotateAndMaskInstructionsPPC64, true,                    \
 132           "Use rotate and mask instructions.")                              \
 133                                                                             \
 134   product(bool, UseStaticBranchPredictionInCompareAndSwapPPC64, true,       \
 135           "Use static branch prediction hints in CAS operations.")          \
 136   product(bool, UseStaticBranchPredictionForUncommonPathsPPC64, false,      \
 137           "Use static branch prediction hints for uncommon paths.")         \
 138                                                                             \
 139   product(bool, UsePower6SchedulerPPC64, false,                             \
 140           "Use Power6 Scheduler.")                                          \
 141                                                                             \
 142   product(bool, InsertEndGroupPPC64, false,                                 \
 143           "Insert EndGroup instructions to optimize for Power6.")           \
 144                                                                             \
 145   /* Trap based checks. */                                                  \
 146   /* Trap based checks use the ppc trap instructions to check certain */    \
 147   /* conditions. This instruction raises a SIGTRAP caught by the      */    \
 148   /* exception handler of the VM.                                     */    \
 149   product(bool, UseSIGTRAP, true,                                           \
 150           "Allow trap instructions that make use of SIGTRAP. Use this to "  \
 151           "switch off all optimizations requiring SIGTRAP.")                \
 152   product(bool, TrapBasedICMissChecks, true,                                \
 153           "Raise and handle SIGTRAP if inline cache miss detected.")        \
 154   product(bool, TrapBasedNotEntrantChecks, true,                            \
 155           "Raise and handle SIGTRAP if calling not entrant or zombie"       \
 156           " method.")                                                       \
 157   product(bool, TraceTraps, false, "Trace all traps the signal handler"     \
 158           "handles.")                                                       \
 159                                                                             \
 160   product(bool, ZapMemory, false, "Write 0x0101... to empty memory."        \
 161           " Use this to ease debugging.")                                   \
 162                                                                             \
 163   /* Use Restricted Transactional Memory for lock elision */                \
 164   product(bool, UseRTMLocking, false,                                       \
 165           "Enable RTM lock eliding for inflated locks in compiled code")    \
 166                                                                             \
 167   experimental(bool, UseRTMForStackLocks, false,                            \
 168           "Enable RTM lock eliding for stack locks in compiled code")       \
 169                                                                             \
 170   product(bool, UseRTMDeopt, false,                                         \
 171           "Perform deopt and recompilation based on RTM abort ratio")       \
 172                                                                             \
 173   product(int, RTMRetryCount, 5,                                            \
 174           "Number of RTM retries on lock abort or busy")                    \
 175           range(0, max_jint)                                                \
 176                                                                             \
 177   experimental(int, RTMSpinLoopCount, 100,                                  \
 178           "Spin count for lock to become free before RTM retry")            \
 179           range(0, 32767) /* immediate operand limit on ppc */              \
 180                                                                             \
 181   experimental(int, RTMAbortThreshold, 1000,                                \
 182           "Calculate abort ratio after this number of aborts")              \
 183           range(0, max_jint)                                                \
 184                                                                             \
 185   experimental(int, RTMLockingThreshold, 10000,                             \
 186           "Lock count at which to do RTM lock eliding without "             \
 187           "abort ratio calculation")                                        \
 188           range(0, max_jint)                                                \
 189                                                                             \
 190   experimental(int, RTMAbortRatio, 50,                                      \
 191           "Lock abort ratio at which to stop use RTM lock eliding")         \
 192           range(0, 100) /* natural range */                                 \
 193                                                                             \
 194   experimental(int, RTMTotalCountIncrRate, 64,                              \
 195           "Increment total RTM attempted lock count once every n times")    \
 196           range(1, 32767) /* immediate operand limit on ppc */              \
 197           constraint(RTMTotalCountIncrRateConstraintFunc,AfterErgo)         \
 198                                                                             \
 199   experimental(intx, RTMLockingCalculationDelay, 0,                         \
 200           "Number of milliseconds to wait before start calculating aborts " \
 201           "for RTM locking")                                                \
 202                                                                             \
 203   experimental(bool, UseRTMXendForLockBusy, true,                           \
 204           "Use RTM Xend instead of Xabort when lock busy")                  \
 205 
 206 #endif // CPU_PPC_GLOBALS_PPC_HPP
--- EOF ---