#define DEBUG_TYPE "functionattrs"
#include "llvm/Transforms/IPO.h"
+#include "llvm/ADT/SCCIterator.h"
+#include "llvm/ADT/SetVector.h"
+#include "llvm/ADT/SmallSet.h"
+#include "llvm/ADT/Statistic.h"
+#include "llvm/Analysis/AliasAnalysis.h"
+#include "llvm/Analysis/CallGraph.h"
+#include "llvm/Analysis/CaptureTracking.h"
#include "llvm/CallGraphSCCPass.h"
#include "llvm/GlobalVariable.h"
#include "llvm/IntrinsicInst.h"
#include "llvm/LLVMContext.h"
-#include "llvm/Analysis/AliasAnalysis.h"
-#include "llvm/Analysis/CallGraph.h"
-#include "llvm/Analysis/CaptureTracking.h"
-#include "llvm/ADT/SmallSet.h"
-#include "llvm/ADT/Statistic.h"
-#include "llvm/ADT/UniqueVector.h"
#include "llvm/Support/InstIterator.h"
using namespace llvm;
// External node - may write memory. Just give up.
return false;
- if (F->doesNotAccessMemory())
+ AliasAnalysis::ModRefBehavior MRB = AA->getModRefBehavior(F);
+ if (MRB == AliasAnalysis::DoesNotAccessMemory)
// Already perfect!
continue;
// Definitions with weak linkage may be overridden at linktime with
// something that writes memory, so treat them like declarations.
if (F->isDeclaration() || F->mayBeOverridden()) {
- if (!F->onlyReadsMemory())
+ if (!AliasAnalysis::onlyReadsMemory(MRB))
// May write memory. Just give up.
return false;
// Ignore calls to functions in the same SCC.
if (CS.getCalledFunction() && SCCNodes.count(CS.getCalledFunction()))
continue;
- switch (AA->getModRefBehavior(CS)) {
- case AliasAnalysis::DoesNotAccessMemory:
- // Ignore calls that don't access memory.
- continue;
- case AliasAnalysis::OnlyReadsMemory:
- // Handle calls that only read from memory.
- ReadsMemory = true;
- continue;
- case AliasAnalysis::AccessesArguments:
- // Check whether all pointer arguments point to local memory, and
- // ignore calls that only access local memory.
- for (CallSite::arg_iterator CI = CS.arg_begin(), CE = CS.arg_end();
- CI != CE; ++CI) {
- Value *Arg = *CI;
- if (Arg->getType()->isPointerTy()) {
- AliasAnalysis::Location Loc(Arg,
- AliasAnalysis::UnknownSize,
- I->getMetadata(LLVMContext::MD_tbaa));
- if (!AA->pointsToConstantMemory(Loc, /*OrLocal=*/true))
- // Writes memory. Just give up.
- return false;
+ AliasAnalysis::ModRefBehavior MRB = AA->getModRefBehavior(CS);
+ // If the call doesn't access arbitrary memory, we may be able to
+ // figure out something.
+ if (AliasAnalysis::onlyAccessesArgPointees(MRB)) {
+ // If the call does access argument pointees, check each argument.
+ if (AliasAnalysis::doesAccessArgPointees(MRB))
+ // Check whether all pointer arguments point to local memory, and
+ // ignore calls that only access local memory.
+ for (CallSite::arg_iterator CI = CS.arg_begin(), CE = CS.arg_end();
+ CI != CE; ++CI) {
+ Value *Arg = *CI;
+ if (Arg->getType()->isPointerTy()) {
+ AliasAnalysis::Location Loc(Arg,
+ AliasAnalysis::UnknownSize,
+ I->getMetadata(LLVMContext::MD_tbaa));
+ if (!AA->pointsToConstantMemory(Loc, /*OrLocal=*/true)) {
+ if (MRB & AliasAnalysis::Mod)
+ // Writes non-local memory. Give up.
+ return false;
+ if (MRB & AliasAnalysis::Ref)
+ // Ok, it reads non-local memory.
+ ReadsMemory = true;
+ }
+ }
}
- }
- // Only reads and writes local memory.
continue;
- default:
- // Otherwise, be conservative.
- break;
}
+ // The call could access any memory. If that includes writes, give up.
+ if (MRB & AliasAnalysis::Mod)
+ return false;
+ // If it reads, note it.
+ if (MRB & AliasAnalysis::Ref)
+ ReadsMemory = true;
+ continue;
} else if (LoadInst *LI = dyn_cast<LoadInst>(I)) {
- // Ignore non-volatile loads from local memory.
+ // Ignore non-volatile loads from local memory. (Atomic is okay here.)
if (!LI->isVolatile()) {
- AliasAnalysis::Location Loc(LI->getPointerOperand(),
- AA->getTypeStoreSize(LI->getType()),
- LI->getMetadata(LLVMContext::MD_tbaa));
+ AliasAnalysis::Location Loc = AA->getLocation(LI);
if (AA->pointsToConstantMemory(Loc, /*OrLocal=*/true))
continue;
}
} else if (StoreInst *SI = dyn_cast<StoreInst>(I)) {
- // Ignore non-volatile stores to local memory.
+ // Ignore non-volatile stores to local memory. (Atomic is okay here.)
if (!SI->isVolatile()) {
- const Type *StoredType = SI->getValueOperand()->getType();
- AliasAnalysis::Location Loc(SI->getPointerOperand(),
- AA->getTypeStoreSize(StoredType),
- SI->getMetadata(LLVMContext::MD_tbaa));
+ AliasAnalysis::Location Loc = AA->getLocation(SI);
if (AA->pointsToConstantMemory(Loc, /*OrLocal=*/true))
continue;
}
+ } else if (VAArgInst *VI = dyn_cast<VAArgInst>(I)) {
+ // Ignore vaargs on local memory.
+ AliasAnalysis::Location Loc = AA->getLocation(VI);
+ if (AA->pointsToConstantMemory(Loc, /*OrLocal=*/true))
+ continue;
}
// Any remaining instructions need to be taken seriously! Check if they
MadeChange = true;
// Clear out any existing attributes.
- F->removeAttribute(~0, Attribute::ReadOnly | Attribute::ReadNone);
+ AttrBuilder B;
+ B.addAttribute(Attribute::ReadOnly)
+ .addAttribute(Attribute::ReadNone);
+ F->removeAttribute(AttributeSet::FunctionIndex,
+ Attribute::get(F->getContext(), B));
// Add in the new attribute.
- F->addAttribute(~0, ReadsMemory? Attribute::ReadOnly : Attribute::ReadNone);
+ B.clear();
+ B.addAttribute(ReadsMemory ? Attribute::ReadOnly : Attribute::ReadNone);
+ F->addAttribute(AttributeSet::FunctionIndex,
+ Attribute::get(F->getContext(), B));
if (ReadsMemory)
++NumReadOnly;
return MadeChange;
}
+namespace {
+ // For a given pointer Argument, this retains a list of Arguments of functions
+ // in the same SCC that the pointer data flows into. We use this to build an
+ // SCC of the arguments.
+ struct ArgumentGraphNode {
+ Argument *Definition;
+ SmallVector<ArgumentGraphNode*, 4> Uses;
+ };
+
+ class ArgumentGraph {
+ // We store pointers to ArgumentGraphNode objects, so it's important that
+ // that they not move around upon insert.
+ typedef std::map<Argument*, ArgumentGraphNode> ArgumentMapTy;
+
+ ArgumentMapTy ArgumentMap;
+
+ // There is no root node for the argument graph, in fact:
+ // void f(int *x, int *y) { if (...) f(x, y); }
+ // is an example where the graph is disconnected. The SCCIterator requires a
+ // single entry point, so we maintain a fake ("synthetic") root node that
+ // uses every node. Because the graph is directed and nothing points into
+ // the root, it will not participate in any SCCs (except for its own).
+ ArgumentGraphNode SyntheticRoot;
+
+ public:
+ ArgumentGraph() { SyntheticRoot.Definition = 0; }
+
+ typedef SmallVectorImpl<ArgumentGraphNode*>::iterator iterator;
+
+ iterator begin() { return SyntheticRoot.Uses.begin(); }
+ iterator end() { return SyntheticRoot.Uses.end(); }
+ ArgumentGraphNode *getEntryNode() { return &SyntheticRoot; }
+
+ ArgumentGraphNode *operator[](Argument *A) {
+ ArgumentGraphNode &Node = ArgumentMap[A];
+ Node.Definition = A;
+ SyntheticRoot.Uses.push_back(&Node);
+ return &Node;
+ }
+ };
+
+ // This tracker checks whether callees are in the SCC, and if so it does not
+ // consider that a capture, instead adding it to the "Uses" list and
+ // continuing with the analysis.
+ struct ArgumentUsesTracker : public CaptureTracker {
+ ArgumentUsesTracker(const SmallPtrSet<Function*, 8> &SCCNodes)
+ : Captured(false), SCCNodes(SCCNodes) {}
+
+ void tooManyUses() { Captured = true; }
+
+ bool captured(Use *U) {
+ CallSite CS(U->getUser());
+ if (!CS.getInstruction()) { Captured = true; return true; }
+
+ Function *F = CS.getCalledFunction();
+ if (!F || !SCCNodes.count(F)) { Captured = true; return true; }
+
+ Function::arg_iterator AI = F->arg_begin(), AE = F->arg_end();
+ for (CallSite::arg_iterator PI = CS.arg_begin(), PE = CS.arg_end();
+ PI != PE; ++PI, ++AI) {
+ if (AI == AE) {
+ assert(F->isVarArg() && "More params than args in non-varargs call");
+ Captured = true;
+ return true;
+ }
+ if (PI == U) {
+ Uses.push_back(AI);
+ break;
+ }
+ }
+ assert(!Uses.empty() && "Capturing call-site captured nothing?");
+ return false;
+ }
+
+ bool Captured; // True only if certainly captured (used outside our SCC).
+ SmallVector<Argument*, 4> Uses; // Uses within our SCC.
+
+ const SmallPtrSet<Function*, 8> &SCCNodes;
+ };
+}
+
+namespace llvm {
+ template<> struct GraphTraits<ArgumentGraphNode*> {
+ typedef ArgumentGraphNode NodeType;
+ typedef SmallVectorImpl<ArgumentGraphNode*>::iterator ChildIteratorType;
+
+ static inline NodeType *getEntryNode(NodeType *A) { return A; }
+ static inline ChildIteratorType child_begin(NodeType *N) {
+ return N->Uses.begin();
+ }
+ static inline ChildIteratorType child_end(NodeType *N) {
+ return N->Uses.end();
+ }
+ };
+ template<> struct GraphTraits<ArgumentGraph*>
+ : public GraphTraits<ArgumentGraphNode*> {
+ static NodeType *getEntryNode(ArgumentGraph *AG) {
+ return AG->getEntryNode();
+ }
+ static ChildIteratorType nodes_begin(ArgumentGraph *AG) {
+ return AG->begin();
+ }
+ static ChildIteratorType nodes_end(ArgumentGraph *AG) {
+ return AG->end();
+ }
+ };
+}
+
/// AddNoCaptureAttrs - Deduce nocapture attributes for the SCC.
bool FunctionAttrs::AddNoCaptureAttrs(const CallGraphSCC &SCC) {
bool Changed = false;
+ SmallPtrSet<Function*, 8> SCCNodes;
+
+ // Fill SCCNodes with the elements of the SCC. Used for quickly
+ // looking up whether a given CallGraphNode is in this SCC.
+ for (CallGraphSCC::iterator I = SCC.begin(), E = SCC.end(); I != E; ++I) {
+ Function *F = (*I)->getFunction();
+ if (F && !F->isDeclaration() && !F->mayBeOverridden())
+ SCCNodes.insert(F);
+ }
+
+ ArgumentGraph AG;
+
+ AttrBuilder B;
+ B.addAttribute(Attribute::NoCapture);
+
// Check each function in turn, determining which pointer arguments are not
// captured.
for (CallGraphSCC::iterator I = SCC.begin(), E = SCC.end(); I != E; ++I) {
Function *F = (*I)->getFunction();
if (F == 0)
- // External node - skip it;
+ // External node - only a problem for arguments that we pass to it.
continue;
// Definitions with weak linkage may be overridden at linktime with
- // something that writes memory, so treat them like declarations.
+ // something that captures pointers, so treat them like declarations.
if (F->isDeclaration() || F->mayBeOverridden())
continue;
+ // Functions that are readonly (or readnone) and nounwind and don't return
+ // a value can't capture arguments. Don't analyze them.
+ if (F->onlyReadsMemory() && F->doesNotThrow() &&
+ F->getReturnType()->isVoidTy()) {
+ for (Function::arg_iterator A = F->arg_begin(), E = F->arg_end();
+ A != E; ++A) {
+ if (A->getType()->isPointerTy() && !A->hasNoCaptureAttr()) {
+ A->addAttr(Attribute::get(F->getContext(), B));
+ ++NumNoCapture;
+ Changed = true;
+ }
+ }
+ continue;
+ }
+
for (Function::arg_iterator A = F->arg_begin(), E = F->arg_end(); A!=E; ++A)
- if (A->getType()->isPointerTy() && !A->hasNoCaptureAttr() &&
- !PointerMayBeCaptured(A, true, /*StoreCaptures=*/false)) {
- A->addAttr(Attribute::NoCapture);
+ if (A->getType()->isPointerTy() && !A->hasNoCaptureAttr()) {
+ ArgumentUsesTracker Tracker(SCCNodes);
+ PointerMayBeCaptured(A, &Tracker);
+ if (!Tracker.Captured) {
+ if (Tracker.Uses.empty()) {
+ // If it's trivially not captured, mark it nocapture now.
+ A->addAttr(Attribute::get(F->getContext(), B));
+ ++NumNoCapture;
+ Changed = true;
+ } else {
+ // If it's not trivially captured and not trivially not captured,
+ // then it must be calling into another function in our SCC. Save
+ // its particulars for Argument-SCC analysis later.
+ ArgumentGraphNode *Node = AG[A];
+ for (SmallVectorImpl<Argument*>::iterator UI = Tracker.Uses.begin(),
+ UE = Tracker.Uses.end(); UI != UE; ++UI)
+ Node->Uses.push_back(AG[*UI]);
+ }
+ }
+ // Otherwise, it's captured. Don't bother doing SCC analysis on it.
+ }
+ }
+
+ // The graph we've collected is partial because we stopped scanning for
+ // argument uses once we solved the argument trivially. These partial nodes
+ // show up as ArgumentGraphNode objects with an empty Uses list, and for
+ // these nodes the final decision about whether they capture has already been
+ // made. If the definition doesn't have a 'nocapture' attribute by now, it
+ // captures.
+
+ for (scc_iterator<ArgumentGraph*> I = scc_begin(&AG), E = scc_end(&AG);
+ I != E; ++I) {
+ std::vector<ArgumentGraphNode*> &ArgumentSCC = *I;
+ if (ArgumentSCC.size() == 1) {
+ if (!ArgumentSCC[0]->Definition) continue; // synthetic root node
+
+ // eg. "void f(int* x) { if (...) f(x); }"
+ if (ArgumentSCC[0]->Uses.size() == 1 &&
+ ArgumentSCC[0]->Uses[0] == ArgumentSCC[0]) {
+ ArgumentSCC[0]->
+ Definition->
+ addAttr(Attribute::get(ArgumentSCC[0]->Definition->getContext(), B));
++NumNoCapture;
Changed = true;
}
+ continue;
+ }
+
+ bool SCCCaptured = false;
+ for (std::vector<ArgumentGraphNode*>::iterator I = ArgumentSCC.begin(),
+ E = ArgumentSCC.end(); I != E && !SCCCaptured; ++I) {
+ ArgumentGraphNode *Node = *I;
+ if (Node->Uses.empty()) {
+ if (!Node->Definition->hasNoCaptureAttr())
+ SCCCaptured = true;
+ }
+ }
+ if (SCCCaptured) continue;
+
+ SmallPtrSet<Argument*, 8> ArgumentSCCNodes;
+ // Fill ArgumentSCCNodes with the elements of the ArgumentSCC. Used for
+ // quickly looking up whether a given Argument is in this ArgumentSCC.
+ for (std::vector<ArgumentGraphNode*>::iterator I = ArgumentSCC.begin(),
+ E = ArgumentSCC.end(); I != E; ++I) {
+ ArgumentSCCNodes.insert((*I)->Definition);
+ }
+
+ for (std::vector<ArgumentGraphNode*>::iterator I = ArgumentSCC.begin(),
+ E = ArgumentSCC.end(); I != E && !SCCCaptured; ++I) {
+ ArgumentGraphNode *N = *I;
+ for (SmallVectorImpl<ArgumentGraphNode*>::iterator UI = N->Uses.begin(),
+ UE = N->Uses.end(); UI != UE; ++UI) {
+ Argument *A = (*UI)->Definition;
+ if (A->hasNoCaptureAttr() || ArgumentSCCNodes.count(A))
+ continue;
+ SCCCaptured = true;
+ break;
+ }
+ }
+ if (SCCCaptured) continue;
+
+ for (unsigned i = 0, e = ArgumentSCC.size(); i != e; ++i) {
+ Argument *A = ArgumentSCC[i]->Definition;
+ A->addAttr(Attribute::get(A->getContext(), B));
+ ++NumNoCapture;
+ Changed = true;
+ }
}
return Changed;
/// or a pointer that doesn't alias any other pointer visible to the caller.
bool FunctionAttrs::IsFunctionMallocLike(Function *F,
SmallPtrSet<Function*, 8> &SCCNodes) const {
- UniqueVector<Value *> FlowsToReturn;
+ SmallSetVector<Value *, 8> FlowsToReturn;
for (Function::iterator I = F->begin(), E = F->end(); I != E; ++I)
if (ReturnInst *Ret = dyn_cast<ReturnInst>(I->getTerminator()))
FlowsToReturn.insert(Ret->getReturnValue());
for (unsigned i = 0; i != FlowsToReturn.size(); ++i) {
- Value *RetVal = FlowsToReturn[i+1]; // UniqueVector[0] is reserved.
+ Value *RetVal = FlowsToReturn[i];
if (Constant *C = dyn_cast<Constant>(RetVal)) {
if (!C->isNullValue() && !isa<UndefValue>(C))
projects / oota-llvm.git / blobdiff