Mercurial > hg > graal-compiler
view src/share/vm/c1/c1_globals.hpp @ 2149:7e37af9d69ef
7011379: G1: overly long concurrent marking cycles
Summary: This changeset introduces filtering of SATB buffers at the point when they are about to be enqueued. If this filtering clears enough entries on each buffer, the buffer can then be re-used and not enqueued. This cuts down the number of SATB buffers that need to be processed by the concurrent marking threads.
Reviewed-by: johnc, ysr
author | tonyp |
---|---|
date | Wed, 19 Jan 2011 09:35:17 -0500 |
parents | f95d63e2154a |
children | 06f017f7daa7 b92c45f2bc75 |
line wrap: on
line source
/* * Copyright (c) 2000, 2010, Oracle and/or its affiliates. All rights reserved. * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. * * This code is free software; you can redistribute it and/or modify it * under the terms of the GNU General Public License version 2 only, as * published by the Free Software Foundation. * * This code is distributed in the hope that it will be useful, but WITHOUT * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License * version 2 for more details (a copy is included in the LICENSE file that * accompanied this code). * * You should have received a copy of the GNU General Public License version * 2 along with this work; if not, write to the Free Software Foundation, * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. * * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA * or visit www.oracle.com if you need additional information or have any * questions. * */ #ifndef SHARE_VM_C1_C1_GLOBALS_HPP #define SHARE_VM_C1_C1_GLOBALS_HPP #include "runtime/globals.hpp" #ifdef TARGET_ARCH_x86 # include "c1_globals_x86.hpp" #endif #ifdef TARGET_ARCH_sparc # include "c1_globals_sparc.hpp" #endif #ifdef TARGET_OS_FAMILY_linux # include "c1_globals_linux.hpp" #endif #ifdef TARGET_OS_FAMILY_solaris # include "c1_globals_solaris.hpp" #endif #ifdef TARGET_OS_FAMILY_windows # include "c1_globals_windows.hpp" #endif // // Defines all global flags used by the client compiler. // #define C1_FLAGS(develop, develop_pd, product, product_pd, notproduct) \ \ /* Printing */ \ notproduct(bool, PrintC1Statistics, false, \ "Print Compiler1 statistics" ) \ \ notproduct(bool, PrintInitialBlockList, false, \ "Print block list of BlockListBuilder") \ \ notproduct(bool, PrintCFG, false, \ "Print control flow graph after each change") \ \ notproduct(bool, PrintCFG0, false, \ "Print control flow graph after construction") \ \ notproduct(bool, PrintCFG1, false, \ "Print control flow graph after optimizations") \ \ notproduct(bool, PrintCFG2, false, \ "Print control flow graph before code generation") \ \ notproduct(bool, PrintIRDuringConstruction, false, \ "Print IR as it's being constructed (helpful for debugging frontend)")\ \ notproduct(bool, PrintPhiFunctions, false, \ "Print phi functions when they are created and simplified") \ \ notproduct(bool, PrintIR, false, \ "Print full intermediate representation after each change") \ \ notproduct(bool, PrintIR0, false, \ "Print full intermediate representation after construction") \ \ notproduct(bool, PrintIR1, false, \ "Print full intermediate representation after optimizations") \ \ notproduct(bool, PrintIR2, false, \ "Print full intermediate representation before code generation") \ \ notproduct(bool, PrintSimpleStubs, false, \ "Print SimpleStubs") \ \ /* C1 optimizations */ \ \ develop(bool, UseC1Optimizations, true, \ "Turn on C1 optimizations") \ \ develop(bool, SelectivePhiFunctions, true, \ "create phi functions at loop headers only when necessary") \ \ develop(bool, OptimizeIfOps, true, \ "Optimize multiple IfOps") \ \ develop(bool, DoCEE, true, \ "Do Conditional Expression Elimination to simplify CFG") \ \ develop(bool, PrintCEE, false, \ "Print Conditional Expression Elimination") \ \ develop(bool, UseLocalValueNumbering, true, \ "Use Local Value Numbering (embedded in GraphBuilder)") \ \ develop(bool, UseGlobalValueNumbering, true, \ "Use Global Value Numbering (separate phase)") \ \ develop(bool, PrintValueNumbering, false, \ "Print Value Numbering") \ \ product(intx, ValueMapInitialSize, 11, \ "Initial size of a value map") \ \ product(intx, ValueMapMaxLoopSize, 8, \ "maximum size of a loop optimized by global value numbering") \ \ develop(bool, EliminateBlocks, true, \ "Eliminate unneccessary basic blocks") \ \ develop(bool, PrintBlockElimination, false, \ "Print basic block elimination") \ \ develop(bool, EliminateNullChecks, true, \ "Eliminate unneccessary null checks") \ \ develop(bool, PrintNullCheckElimination, false, \ "Print null check elimination") \ \ develop(bool, EliminateFieldAccess, true, \ "Optimize field loads and stores") \ \ develop(bool, InlineMethodsWithExceptionHandlers, true, \ "Inline methods containing exception handlers " \ "(NOTE: does not work with current backend)") \ \ develop(bool, InlineSynchronizedMethods, true, \ "Inline synchronized methods") \ \ develop(bool, InlineNIOCheckIndex, true, \ "Intrinsify java.nio.Buffer.checkIndex") \ \ develop(bool, CanonicalizeNodes, true, \ "Canonicalize graph nodes") \ \ develop(bool, CanonicalizeExperimental, false, \ "Canonicalize graph nodes, experimental code") \ \ develop(bool, PrintCanonicalization, false, \ "Print graph node canonicalization") \ \ develop(bool, UseTableRanges, true, \ "Faster versions of lookup table using ranges") \ \ develop(bool, UseFastExceptionHandling, true, \ "Faster handling of exceptions") \ \ develop_pd(bool, RoundFPResults, \ "Indicates whether rounding is needed for floating point results")\ \ develop(intx, NestedInliningSizeRatio, 90, \ "Percentage of prev. allowed inline size in recursive inlining") \ \ notproduct(bool, PrintIRWithLIR, false, \ "Print IR instructions with generated LIR") \ \ notproduct(bool, PrintLIRWithAssembly, false, \ "Show LIR instruction with generated assembly") \ \ develop(bool, CommentedAssembly, trueInDebug, \ "Show extra info in PrintNMethods output") \ \ develop(bool, LIRTracePeephole, false, \ "Trace peephole optimizer") \ \ develop(bool, LIRTraceExecution, false, \ "add LIR code which logs the execution of blocks") \ \ product_pd(bool, LIRFillDelaySlots, \ "fill delays on on SPARC with LIR") \ \ develop_pd(bool, CSEArrayLength, \ "Create separate nodes for length in array accesses") \ \ develop_pd(bool, TwoOperandLIRForm, \ "true if LIR requires src1 and dst to match in binary LIR ops") \ \ develop(intx, TraceLinearScanLevel, 0, \ "Debug levels for the linear scan allocator") \ \ develop(bool, StressLinearScan, false, \ "scramble block order used by LinearScan (stress test)") \ \ product(bool, TimeLinearScan, false, \ "detailed timing of LinearScan phases") \ \ develop(bool, TimeEachLinearScan, false, \ "print detailed timing of each LinearScan run") \ \ develop(bool, CountLinearScan, false, \ "collect statistic counters during LinearScan") \ \ /* C1 variable */ \ \ develop(bool, C1Breakpoint, false, \ "Sets a breakpoint at entry of each compiled method") \ \ develop(bool, ImplicitDiv0Checks, true, \ "Use implicit division by zero checks") \ \ develop(bool, PinAllInstructions, false, \ "All instructions are pinned") \ \ develop(bool, ValueStackPinStackAll, true, \ "Pinning in ValueStack pin everything") \ \ develop(bool, UseFastNewInstance, true, \ "Use fast inlined instance allocation") \ \ develop(bool, UseFastNewTypeArray, true, \ "Use fast inlined type array allocation") \ \ develop(bool, UseFastNewObjectArray, true, \ "Use fast inlined object array allocation") \ \ develop(bool, UseFastLocking, true, \ "Use fast inlined locking code") \ \ develop(bool, UseSlowPath, false, \ "For debugging: test slow cases by always using them") \ \ develop(bool, GenerateArrayStoreCheck, true, \ "Generates code for array store checks") \ \ develop(bool, DeoptC1, true, \ "Use deoptimization in C1") \ \ develop(bool, PrintBailouts, false, \ "Print bailout and its reason") \ \ develop(bool, TracePatching, false, \ "Trace patching of field access on uninitialized classes") \ \ develop(bool, PatchALot, false, \ "Marks all fields as having unloaded classes") \ \ develop(bool, PrintNotLoaded, false, \ "Prints where classes are not loaded during code generation") \ \ notproduct(bool, VerifyOopMaps, false, \ "Adds oopmap verification code to the generated code") \ \ develop(bool, PrintLIR, false, \ "print low-level IR") \ \ develop(bool, BailoutAfterHIR, false, \ "bailout of compilation after building of HIR") \ \ develop(bool, BailoutAfterLIR, false, \ "bailout of compilation after building of LIR") \ \ develop(bool, BailoutOnExceptionHandlers, false, \ "bailout of compilation for methods with exception handlers") \ \ develop(bool, InstallMethods, true, \ "Install methods at the end of successful compilations") \ \ product(intx, CompilationRepeat, 0, \ "Number of times to recompile method before returning result") \ \ develop(intx, NMethodSizeLimit, (32*K)*wordSize, \ "Maximum size of a compiled method.") \ \ develop(bool, TraceFPUStack, false, \ "Trace emulation of the FPU stack (intel only)") \ \ develop(bool, TraceFPURegisterUsage, false, \ "Trace usage of FPU registers at start of blocks (intel only)") \ \ develop(bool, OptimizeUnsafes, true, \ "Optimize raw unsafe ops") \ \ develop(bool, PrintUnsafeOptimization, false, \ "Print optimization of raw unsafe ops") \ \ develop(intx, InstructionCountCutoff, 37000, \ "If GraphBuilder adds this many instructions, bails out") \ \ product_pd(intx, SafepointPollOffset, \ "Offset added to polling address (Intel only)") \ \ develop(bool, ComputeExactFPURegisterUsage, true, \ "Compute additional live set for fpu registers to simplify fpu stack merge (Intel only)") \ \ product(bool, C1ProfileCalls, true, \ "Profile calls when generating code for updating MDOs") \ \ product(bool, C1ProfileVirtualCalls, true, \ "Profile virtual calls when generating code for updating MDOs") \ \ product(bool, C1ProfileInlinedCalls, true, \ "Profile inlined calls when generating code for updating MDOs") \ \ product(bool, C1ProfileBranches, true, \ "Profile branches when generating code for updating MDOs") \ \ product(bool, C1ProfileCheckcasts, true, \ "Profile checkcasts when generating code for updating MDOs") \ \ product(bool, C1OptimizeVirtualCallProfiling, true, \ "Use CHA and exact type results at call sites when updating MDOs")\ \ product(bool, C1UpdateMethodData, trueInTiered, \ "Update methodDataOops in Tier1-generated code") \ \ develop(bool, PrintCFGToFile, false, \ "print control flow graph to a separate file during compilation") \ \ // Read default values for c1 globals C1_FLAGS(DECLARE_DEVELOPER_FLAG, DECLARE_PD_DEVELOPER_FLAG, DECLARE_PRODUCT_FLAG, DECLARE_PD_PRODUCT_FLAG, DECLARE_NOTPRODUCT_FLAG) #endif // SHARE_VM_C1_C1_GLOBALS_HPP