// Generate code for a single tree root. private void generateTree(OPT_BURS_TreeNode k, OPT_BURS_STATE burs) { OPT_BURS_TreeNode child1 = k.child1; OPT_BURS_TreeNode child2 = k.child2; if (child1 != null) { if (child2 != null) { // k has two children; use register labeling to // determine order that minimizes register pressure if (k.isSuperNodeRoot()) { byte act = burs.action[k.rule(k.getNonTerminal())]; if ((act & burs.RIGHT_CHILD_FIRST) != 0) { // rule selected forces order of evaluation generateTree(child2, burs); generateTree(child1, burs); } else { generateTree(child1, burs); generateTree(child2, burs); } } else { generateTree(child1, burs); generateTree(child2, burs); } } else { generateTree(child1, burs); } } else if (child2 != null) { generateTree(child2, burs); } if (k.isSuperNodeRoot()) { int nonterminal = k.getNonTerminal(); int rule = k.rule(nonterminal); burs.code(k, nonterminal, rule); if (DEBUG) VM.sysWrite(k + " " + OPT_BURS_Debug.string[rule] + "\n"); } }
/** * Logic to select the appropriate guarding mechanism for the edge from caller to callee according * to the controlling {@link OPT_Options}. If we are using IG_CODE_PATCH, then this method also * records the required dependency. Precondition: lock on {@link VM_Class#OptCLDepManager} is * held. * * @param caller The caller method * @param callee The callee method * @param codePatchSupported Can we use code patching at this call site? */ private byte chooseGuard( VM_Method caller, VM_Method singleImpl, VM_Method callee, OPT_CompilationState state, boolean codePatchSupported) { byte guard = state.getOptions().INLINING_GUARD; if (codePatchSupported) { if (VM.VerifyAssertions && VM.runningVM) { VM._assert(VM_Lock.owns(VM_Class.OptCLDepManager)); } if (guard == OPT_Options.IG_CODE_PATCH) { if (OPT_ClassLoadingDependencyManager.TRACE || OPT_ClassLoadingDependencyManager.DEBUG) { VM_Class.OptCLDepManager.report( "CODE PATCH: Inlined " + singleImpl + " into " + caller + "\n"); } VM_Class.OptCLDepManager.addNotOverriddenDependency(callee, state.getCompiledMethod()); } } else if (guard == OPT_Options.IG_CODE_PATCH) { guard = OPT_Options.IG_METHOD_TEST; } if (guard == OPT_Options.IG_METHOD_TEST && singleImpl.getDeclaringClass().isFinal()) { // class test is more efficient and just as effective guard = OPT_Options.IG_CLASS_TEST; } return guard; }
/** * Compute node congruence over the value number graph. * * <p>Algorithm: Muchnick pp. 348-355 * * <p>Note: the Muchnick algorithm is buggy. In particular, it does not put all needed congruence * classes on the worklist. * * <p>Two nodes in the value graph are congruent if one of the following holds: * * <ul> * <li>they are the same node * <li>their labels are identical constants * <li>they have the same operators and their operands are congruent * </ul> * * <p>Optimistic algorithm: * * <ul> * <li>Initially assume all nodes with the same label are congruent * <li>start a work list with all congruence classes that have multiple operands * <li>choose a congruence class from the worklist. partition its elements into new congruence * classes if we can discover that they are not congruent. * <li>Add any newly created congruence classes to the work list. * <li>(Here's the step Muchnick omits:) For each class C which has a dependence on any of the * newly created congruence classes, add C to the work list * <li>repeat until work list is empty * </ul> * * <p>The following method breaks Muchnick's algorithm, which will assign m and n the same value * number. Muchnick's problem is that it does not put the congruence class for 'int_mul' back on * the worklist when we discover, for example, that i is not congruent to k * * <pre> * public int foo(int a, int b, int c, int d, int e, int f, int g, int h) { * int i = a+b; * int j = c+d; * int k = e+f; * int l = g+h; * int m = i * j; * int n = k * l; * int o = m/n; * return o; * } * </pre> */ public void globalValueNumber() { valueGraph = new OPT_ValueGraph(ir); if (DEBUG) VM.sysWrite(valueGraph.toString()); // initialize the congurence classes initialize(); // initialize the work list initializeWorkList(); // drain the work list while (!workList.empty()) { OPT_GVCongruenceClass partition = (OPT_GVCongruenceClass) workList.pop(); partitionClass(partition); } // all done if (DEBUG) printValueNumbers(); }
/** * Should we inline a particular call site? * * @param state information needed to make the inlining decision * @return an OPT_InlineDecision with the result */ public OPT_InlineDecision shouldInline(final OPT_CompilationState state) { final OPT_Options opts = state.getOptions(); final boolean verbose = opts.PRINT_DETAILED_INLINE_REPORT; if (!opts.INLINE) { return OPT_InlineDecision.NO("inlining not enabled"); } final VM_Method staticCallee = state.obtainTarget(); final VM_NormalMethod rootMethod = state.getRootMethod(); final VM_Method caller = state.getMethod(); final int bcIndex = state.getBytecodeIndex(); if (verbose) VM.sysWriteln( "Begin inline decision for " + "<" + caller + "," + bcIndex + "," + staticCallee + ">"); // Stage 1: At all optimization levels we should attempt to inline // trivial methods. Even if the inline code is never executed, // inlining a trivial method is a no cost operation as the impact // on code size should be negligible and compile time usually is // reduced since we expect to eliminate the call instruction (or // at worse replace one call instruction with another one). if (!state.isInvokeInterface()) { if (staticCallee.isNative()) { if (verbose) VM.sysWriteln("\tNO: native method\n"); return OPT_InlineDecision.NO("native method"); } if (hasNoInlinePragma(staticCallee, state)) { if (verbose) VM.sysWriteln("\tNO: pragmaNoInline\n"); return OPT_InlineDecision.NO("pragmaNoInline"); } if (!staticCallee.isAbstract()) { int inlinedSizeEstimate = inlinedSizeEstimate((VM_NormalMethod) staticCallee, state); boolean guardless = state.getHasPreciseTarget() || !needsGuard(staticCallee); if (inlinedSizeEstimate < opts.IC_MAX_ALWAYS_INLINE_TARGET_SIZE && guardless && !state.getSequence().containsMethod(staticCallee)) { if (verbose) VM.sysWriteln("\tYES: trivial guardless inline\n"); return OPT_InlineDecision.YES(staticCallee, "trivial inline"); } } } if (opts.getOptLevel() == 0) { // at opt level 0, trivial unguarded inlines are the only kind we consider if (verbose) VM.sysWriteln("\tNO: only do trivial inlines at O0\n"); return OPT_InlineDecision.NO("Only do trivial inlines at O0"); } if (rootMethod.inlinedSizeEstimate() > opts.IC_MASSIVE_METHOD_SIZE) { // In massive methods, we do not do any additional inlining to // avoid completely blowing out compile time by making a bad situation worse if (verbose) VM.sysWriteln("\tNO: only do trivial inlines into massive methods\n"); return OPT_InlineDecision.NO("Root method is massive; no non-trivial inlines"); } // Stage 2: Determine based on profile data and static information // what are the possible targets of this call. // VM_WeightedCallTargets targets = null; boolean purelyStatic = true; if (VM_Controller.dcg != null && VM_Controller.options.ADAPTIVE_INLINING) { targets = VM_Controller.dcg.getCallTargets(caller, bcIndex); if (targets != null) { if (verbose) VM.sysWriteln("\tFound profile data"); purelyStatic = false; if (state.getHasPreciseTarget()) { // static analysis tells us that there is only one possible target. // Filter the profile information accordingly. targets = targets.filter(staticCallee); if (verbose) VM.sysWriteln("\tFiltered to match precise target"); if (targets == null) { if (verbose) VM.sysWriteln("\tNow no profile data..."); // After filtering, no matching profile data, fall back to // static information to avoid degradations targets = VM_WeightedCallTargets.create(staticCallee, 0); purelyStatic = true; } } } } // Critical section: must prevent class hierarchy from changing while // we are inspecting it to determine how/whether to do the inline guard. synchronized (VM_Class.OptCLDepManager) { boolean guardOverrideOnStaticCallee = false; if (targets == null) { if (verbose) VM.sysWriteln("\tNo profile data"); // No profile information. // Fake up "profile data" based on static information to // be able to share all the decision making logic. if (state.isInvokeInterface()) { if (opts.GUARDED_INLINE_INTERFACE) { VM_Method singleImpl = OPT_InterfaceHierarchy.getUniqueImplementation(staticCallee); if (singleImpl != null && hasBody(singleImpl)) { if (verbose) VM.sysWriteln( "\tFound a single implementation " + singleImpl + " of an interface method " + staticCallee); targets = VM_WeightedCallTargets.create(singleImpl, 0); guardOverrideOnStaticCallee = true; } } } else { // invokestatic, invokevirtual, invokespecial if (staticCallee.isAbstract()) { // look for single non-abstract implementation of the abstract method VM_Class klass = staticCallee.getDeclaringClass(); while (true) { VM_Class[] subClasses = klass.getSubClasses(); if (subClasses.length != 1) break; // multiple subclasses => multiple targets VM_Method singleImpl = subClasses[0].findDeclaredMethod( staticCallee.getName(), staticCallee.getDescriptor()); if (singleImpl != null && !singleImpl.isAbstract()) { // found something if (verbose) VM.sysWriteln("\tsingle impl of abstract method"); targets = VM_WeightedCallTargets.create(singleImpl, 0); guardOverrideOnStaticCallee = true; break; } klass = subClasses[0]; // keep crawling down the hierarchy } } else { targets = VM_WeightedCallTargets.create(staticCallee, 0); } } } // At this point targets is either null, or accurately represents what we // think are the likely target(s) of the call site. // This information may be either derived from profile information or // from static heuristics. To the first approximation, we don't care which. // If there is a precise target, then targets contains exactly that target method. if (targets == null) return OPT_InlineDecision.NO("No potential targets identified"); // Stage 3: We have one or more targets. Determine what if anything should be done with them. final ArrayList methodsToInline = new ArrayList(); final ArrayList methodsNeedGuard = new ArrayList(); final double callSiteWeight = targets.totalWeight(); final boolean goosc = guardOverrideOnStaticCallee; // real closures anyone? final boolean ps = purelyStatic; // real closures anyone? targets.visitTargets( new VM_WeightedCallTargets.Visitor() { public void visit(VM_Method callee, double weight) { if (hasBody(callee)) { if (verbose) { VM.sysWriteln( "\tEvaluating target " + callee + " with " + weight + " samples (" + (100 * VM_AdaptiveInlining.adjustedWeight(weight)) + "%)"); } // Don't inline recursively and respect no inline pragmas OPT_InlineSequence seq = state.getSequence(); if (seq.containsMethod(callee)) { if (verbose) VM.sysWriteln("\t\tReject: recursive"); return; } if (hasNoInlinePragma(callee, state)) { if (verbose) VM.sysWriteln("\t\tReject: noinline pragma"); return; } // more or less figure out the guard situation early -- impacts size estimate. boolean needsGuard = !state.getHasPreciseTarget() && (staticCallee != callee || needsGuard(staticCallee)); if (needsGuard && isForbiddenSpeculation(state.getRootMethod(), callee)) { if (verbose) VM.sysWriteln("\t\tReject: forbidden speculation"); return; } boolean currentlyFinal = (goosc || (staticCallee == callee)) && isCurrentlyFinal(callee, !opts.guardWithClassTest()); boolean preEx = needsGuard && state.getIsExtant() && opts.PREEX_INLINE && currentlyFinal; if (needsGuard && !preEx) { if (!opts.GUARDED_INLINE) { if (verbose) VM.sysWriteln("\t\tReject: guarded inlining disabled"); return; } if (!currentlyFinal && ps) { if (verbose) VM.sysWriteln("\t\tReject: multiple targets and no profile data"); return; } } // Estimate cost of performing this inlining action. // Includes cost of guard & off-branch call if they are going to be generated. boolean decideYes = false; if (hasInlinePragma(callee, state)) { if (verbose) VM.sysWriteln("\t\tSelect: pragma inline"); decideYes = true; } else { // Preserve previous inlining decisions // Not the best thing in the world due to phase shifts, but // it does buy some degree of stability. So, it is probably the lesser // of two evils. VM_CompiledMethod prev = state.getRootMethod().getCurrentCompiledMethod(); if (prev != null && prev.getCompilerType() == VM_CompiledMethod.OPT) { if (((VM_OptCompiledMethod) prev) .getMCMap() .hasInlinedEdge(caller, bcIndex, callee)) { if (verbose) VM.sysWriteln("\t\tSelect: Previously inlined"); decideYes = true; } } if (!decideYes) { int inlinedSizeEstimate = inlinedSizeEstimate((VM_NormalMethod) callee, state); int cost = inliningActionCost(inlinedSizeEstimate, needsGuard, preEx, opts); int maxCost = opts.IC_MAX_TARGET_SIZE; if (callSiteWeight > VM_Controller.options.AI_SEED_MULTIPLIER) { // real profile data with enough samples for us to trust it. // Use weight and shape of call site distrubution to compute // a higher maxCost. double fractionOfSample = weight / callSiteWeight; if (needsGuard && fractionOfSample < opts.AI_MIN_CALLSITE_FRACTION) { // This call accounts for less than AI_MIN_CALLSITE_FRACTION // of the profiled targets at this call site. // It is highly unlikely to be profitable to inline it. if (verbose) VM.sysWriteln( "\t\tReject: less than AI_MIN_CALLSITE_FRACTION of distribution"); maxCost = 0; } else { if (cost > maxCost) { // adjust up based on weight of callsite double adjustedWeight = VM_AdaptiveInlining.adjustedWeight(weight); if (adjustedWeight > VM_Controller.options.AI_CONTROL_POINT) { maxCost = opts.AI_MAX_TARGET_SIZE; } else { int range = opts.AI_MAX_TARGET_SIZE - opts.IC_MAX_TARGET_SIZE; double slope = ((double) range) / VM_Controller.options.AI_CONTROL_POINT; int sizeAdj = (int) (slope * adjustedWeight); maxCost += sizeAdj; } } } } // Somewhat bogus, but if we get really deeply inlined we start backing off. int curDepth = state.getInlineDepth(); if (curDepth > opts.IC_MAX_INLINE_DEPTH) { maxCost /= (curDepth - opts.IC_MAX_INLINE_DEPTH + 1); } decideYes = cost <= maxCost; if (verbose) { if (decideYes) { VM.sysWriteln( "\t\tAccept: cost of " + cost + " was below threshold " + maxCost); } else { VM.sysWriteln( "\t\tReject: cost of " + cost + " was above threshold " + maxCost); } } } } if (decideYes) { // Ok, we're going to inline it. // Record that and also whether or not we think it needs a guard. methodsToInline.add(callee); if (preEx) { if (OPT_ClassLoadingDependencyManager.TRACE || OPT_ClassLoadingDependencyManager.DEBUG) { VM_Class.OptCLDepManager.report( "PREEX_INLINE: Inlined " + callee + " into " + caller + "\n"); } VM_Class.OptCLDepManager.addNotOverriddenDependency( callee, state.getCompiledMethod()); if (goosc) { VM_Class.OptCLDepManager.addNotOverriddenDependency( staticCallee, state.getCompiledMethod()); } methodsNeedGuard.add(Boolean.FALSE); } else { methodsNeedGuard.add(Boolean.valueOf(needsGuard)); } } } } }); // Stage 4: Choose guards and package up the results in an InlineDecision object if (methodsToInline.size() == 0) { OPT_InlineDecision d = OPT_InlineDecision.NO("No desirable targets"); if (verbose) VM.sysWriteln("\tDecide: " + d); return d; } else if (methodsToInline.size() == 1) { VM_Method target = (VM_Method) methodsToInline.get(0); boolean needsGuard = ((Boolean) methodsNeedGuard.get(0)).booleanValue(); if (needsGuard) { if ((guardOverrideOnStaticCallee || target == staticCallee) && isCurrentlyFinal(target, !opts.guardWithClassTest())) { OPT_InlineDecision d = OPT_InlineDecision.guardedYES( target, chooseGuard(caller, target, staticCallee, state, true), "Guarded inline of single static target"); // -#if RVM_WITH_OSR if (opts.OSR_GUARDED_INLINING && OPT_Compiler.getAppStarted() && VM_Controller.options.ENABLE_RECOMPILATION) { // note that we will OSR the failed case. d.setOSRTestFailed(); } // -#endif if (verbose) VM.sysWriteln("\tDecide: " + d); return d; } else { OPT_InlineDecision d = OPT_InlineDecision.guardedYES( target, chooseGuard(caller, target, staticCallee, state, false), "Guarded inlining of one potential target"); if (verbose) VM.sysWriteln("\tDecide: " + d); return d; } } else { OPT_InlineDecision d = OPT_InlineDecision.YES(target, "Unique and desirable target"); if (verbose) VM.sysWriteln("\tDecide: " + d); return d; } } else { VM_Method[] methods = new VM_Method[methodsNeedGuard.size()]; byte[] guards = new byte[methods.length]; int idx = 0; for (Iterator methodIterator = methodsToInline.iterator(), guardIterator = methodsNeedGuard.iterator(); methodIterator.hasNext(); ) { VM_Method target = (VM_Method) methodIterator.next(); boolean needsGuard = ((Boolean) guardIterator.next()).booleanValue(); if (VM.VerifyAssertions) VM._assert(needsGuard); methods[idx] = target; guards[idx] = chooseGuard(caller, target, staticCallee, state, false); idx++; } OPT_InlineDecision d = OPT_InlineDecision.guardedYES(methods, guards, "Inline multiple targets"); if (verbose) VM.sysWriteln("\tDecide: " + d); return d; } } }