//
//===----------------------------------------------------------------------===//
-#define DEBUG_TYPE "sample-profile"
-
#include "llvm/Transforms/Scalar.h"
#include "llvm/ADT/DenseMap.h"
#include "llvm/ADT/SmallPtrSet.h"
#include "llvm/ADT/SmallSet.h"
-#include "llvm/ADT/StringMap.h"
#include "llvm/ADT/StringRef.h"
#include "llvm/Analysis/LoopInfo.h"
#include "llvm/Analysis/PostDominators.h"
#include "llvm/IR/Metadata.h"
#include "llvm/IR/Module.h"
#include "llvm/Pass.h"
+#include "llvm/ProfileData/SampleProfReader.h"
#include "llvm/Support/CommandLine.h"
#include "llvm/Support/Debug.h"
-#include "llvm/Support/LineIterator.h"
-#include "llvm/Support/MemoryBuffer.h"
-#include "llvm/Support/Regex.h"
#include "llvm/Support/raw_ostream.h"
#include <cctype>
using namespace llvm;
+using namespace sampleprof;
+
+#define DEBUG_TYPE "sample-profile"
// Command line option to specify the file to read samples from. This is
// mainly used for debugging.
"sample block/edge weights through the CFG."));
namespace {
-/// \brief Represents the relative location of an instruction.
-///
-/// Instruction locations are specified by the line offset from the
-/// beginning of the function (marked by the line where the function
-/// header is) and the discriminator value within that line.
-///
-/// The discriminator value is useful to distinguish instructions
-/// that are on the same line but belong to different basic blocks
-/// (e.g., the two post-increment instructions in "if (p) x++; else y++;").
-struct InstructionLocation {
- InstructionLocation(int L, unsigned D) : LineOffset(L), Discriminator(D) {}
- int LineOffset;
- unsigned Discriminator;
-};
-}
-
-namespace llvm {
-template <> struct DenseMapInfo<InstructionLocation> {
- typedef DenseMapInfo<int> OffsetInfo;
- typedef DenseMapInfo<unsigned> DiscriminatorInfo;
- static inline InstructionLocation getEmptyKey() {
- return InstructionLocation(OffsetInfo::getEmptyKey(),
- DiscriminatorInfo::getEmptyKey());
- }
- static inline InstructionLocation getTombstoneKey() {
- return InstructionLocation(OffsetInfo::getTombstoneKey(),
- DiscriminatorInfo::getTombstoneKey());
- }
- static inline unsigned getHashValue(InstructionLocation Val) {
- return DenseMapInfo<std::pair<int, unsigned>>::getHashValue(
- std::pair<int, unsigned>(Val.LineOffset, Val.Discriminator));
- }
- static inline bool isEqual(InstructionLocation LHS, InstructionLocation RHS) {
- return LHS.LineOffset == RHS.LineOffset &&
- LHS.Discriminator == RHS.Discriminator;
- }
-};
-}
-
-namespace {
-typedef DenseMap<InstructionLocation, unsigned> BodySampleMap;
typedef DenseMap<BasicBlock *, unsigned> BlockWeightMap;
typedef DenseMap<BasicBlock *, BasicBlock *> EquivalenceClassMap;
typedef std::pair<BasicBlock *, BasicBlock *> Edge;
typedef DenseMap<Edge, unsigned> EdgeWeightMap;
typedef DenseMap<BasicBlock *, SmallVector<BasicBlock *, 8>> BlockEdgeMap;
-/// \brief Representation of the runtime profile for a function.
+/// \brief Sample profile pass.
///
-/// This data structure contains the runtime profile for a given
-/// function. It contains the total number of samples collected
-/// in the function and a map of samples collected in every statement.
-class SampleFunctionProfile {
+/// This pass reads profile data from the file specified by
+/// -sample-profile-file and annotates every affected function with the
+/// profile information found in that file.
+class SampleProfileLoader : public FunctionPass {
public:
- SampleFunctionProfile()
- : TotalSamples(0), TotalHeadSamples(0), HeaderLineno(0), DT(0), PDT(0),
- LI(0), Ctx(0) {}
+ // Class identification, replacement for typeinfo
+ static char ID;
+
+ SampleProfileLoader(StringRef Name = SampleProfileFile)
+ : FunctionPass(ID), DT(nullptr), PDT(nullptr), LI(nullptr), Ctx(nullptr),
+ Reader(), Samples(nullptr), Filename(Name), ProfileIsValid(false) {
+ initializeSampleProfileLoaderPass(*PassRegistry::getPassRegistry());
+ }
+
+ bool doInitialization(Module &M) override;
+
+ void dump() { Reader->dump(); }
+
+ const char *getPassName() const override { return "Sample profile pass"; }
+
+ bool runOnFunction(Function &F) override;
+
+ void getAnalysisUsage(AnalysisUsage &AU) const override {
+ AU.setPreservesCFG();
+ AU.addRequired<LoopInfo>();
+ AU.addRequired<DominatorTreeWrapperPass>();
+ AU.addRequired<PostDominatorTree>();
+ }
+protected:
unsigned getFunctionLoc(Function &F);
- bool emitAnnotations(Function &F, DominatorTree *DomTree,
- PostDominatorTree *PostDomTree, LoopInfo *Loops);
+ bool emitAnnotations(Function &F);
unsigned getInstWeight(Instruction &I);
- unsigned getBlockWeight(BasicBlock *B);
- void addTotalSamples(unsigned Num) { TotalSamples += Num; }
- void addHeadSamples(unsigned Num) { TotalHeadSamples += Num; }
- void addBodySamples(int LineOffset, unsigned Discriminator, unsigned Num) {
- assert(LineOffset >= 0);
- BodySamples[InstructionLocation(LineOffset, Discriminator)] += Num;
- }
- void print(raw_ostream &OS);
+ unsigned getBlockWeight(BasicBlock *BB);
void printEdgeWeight(raw_ostream &OS, Edge E);
void printBlockWeight(raw_ostream &OS, BasicBlock *BB);
void printBlockEquivalence(raw_ostream &OS, BasicBlock *BB);
unsigned visitEdge(Edge E, unsigned *NumUnknownEdges, Edge *UnknownEdge);
void buildEdges(Function &F);
bool propagateThroughEdges(Function &F);
- bool empty() { return BodySamples.empty(); }
-protected:
- /// \brief Total number of samples collected inside this function.
- ///
- /// Samples are cumulative, they include all the samples collected
- /// inside this function and all its inlined callees.
- unsigned TotalSamples;
-
- /// \brief Total number of samples collected at the head of the function.
- /// FIXME: Use head samples to estimate a cold/hot attribute for the function.
- unsigned TotalHeadSamples;
-
- /// \brief Line number for the function header. Used to compute relative
- /// line numbers from the absolute line LOCs found in instruction locations.
- /// The relative line numbers are needed to address the samples from the
- /// profile file.
+ /// \brief Line number for the function header. Used to compute absolute
+ /// line numbers from the relative line numbers found in the profile.
unsigned HeaderLineno;
- /// \brief Map line offsets to collected samples.
- ///
- /// Each entry in this map contains the number of samples
- /// collected at the corresponding line offset. All line locations
- /// are an offset from the start of the function.
- BodySampleMap BodySamples;
-
/// \brief Map basic blocks to their computed weights.
///
/// The weight of a basic block is defined to be the maximum
/// \brief LLVM context holding the debug data we need.
LLVMContext *Ctx;
-};
-
-/// \brief Sample-based profile reader.
-///
-/// Each profile contains sample counts for all the functions
-/// executed. Inside each function, statements are annotated with the
-/// collected samples on all the instructions associated with that
-/// statement.
-///
-/// For this to produce meaningful data, the program needs to be
-/// compiled with some debug information (at minimum, line numbers:
-/// -gline-tables-only). Otherwise, it will be impossible to match IR
-/// instructions to the line numbers collected by the profiler.
-///
-/// From the profile file, we are interested in collecting the
-/// following information:
-///
-/// * A list of functions included in the profile (mangled names).
-///
-/// * For each function F:
-/// 1. The total number of samples collected in F.
-///
-/// 2. The samples collected at each line in F. To provide some
-/// protection against source code shuffling, line numbers should
-/// be relative to the start of the function.
-class SampleModuleProfile {
-public:
- SampleModuleProfile(const Module &M, StringRef F)
- : Profiles(0), Filename(F), M(M) {}
-
- void dump();
- bool loadText();
- void loadNative() { llvm_unreachable("not implemented"); }
- void printFunctionProfile(raw_ostream &OS, StringRef FName);
- void dumpFunctionProfile(StringRef FName);
- SampleFunctionProfile &getProfile(const Function &F) {
- return Profiles[F.getName()];
- }
-
- /// \brief Report a parse error message and stop compilation.
- void reportParseError(int64_t LineNumber, Twine Msg) const {
- DiagnosticInfoSampleProfile Diag(Filename.data(), LineNumber, Msg);
- M.getContext().diagnose(Diag);
- }
-
-protected:
- /// \brief Map every function to its associated profile.
- ///
- /// The profile of every function executed at runtime is collected
- /// in the structure SampleFunctionProfile. This maps function objects
- /// to their corresponding profiles.
- StringMap<SampleFunctionProfile> Profiles;
-
- /// \brief Path name to the file holding the profile data.
- ///
- /// The format of this file is defined by each profiler
- /// independently. If possible, the profiler should have a text
- /// version of the profile format to be used in constructing test
- /// cases and debugging.
- StringRef Filename;
-
- /// \brief Module being compiled. Used mainly to access the current
- /// LLVM context for diagnostics.
- const Module &M;
-};
-
-/// \brief Sample profile pass.
-///
-/// This pass reads profile data from the file specified by
-/// -sample-profile-file and annotates every affected function with the
-/// profile information found in that file.
-class SampleProfileLoader : public FunctionPass {
-public:
- // Class identification, replacement for typeinfo
- static char ID;
-
- SampleProfileLoader(StringRef Name = SampleProfileFile)
- : FunctionPass(ID), Profiler(), Filename(Name), ProfileIsValid(false) {
- initializeSampleProfileLoaderPass(*PassRegistry::getPassRegistry());
- }
-
- bool doInitialization(Module &M) override;
- void dump() { Profiler->dump(); }
-
- const char *getPassName() const override { return "Sample profile pass"; }
-
- bool runOnFunction(Function &F) override;
-
- void getAnalysisUsage(AnalysisUsage &AU) const override {
- AU.setPreservesCFG();
- AU.addRequired<LoopInfo>();
- AU.addRequired<DominatorTreeWrapperPass>();
- AU.addRequired<PostDominatorTree>();
- }
-
-protected:
/// \brief Profile reader object.
- std::unique_ptr<SampleModuleProfile> Profiler;
+ std::unique_ptr<SampleProfileReader> Reader;
+
+ /// \brief Samples collected for the body of this function.
+ FunctionSamples *Samples;
/// \brief Name of the profile file to load.
StringRef Filename;
- /// \brief Flag indicating whether the profile input loaded succesfully.
+ /// \brief Flag indicating whether the profile input loaded successfully.
bool ProfileIsValid;
};
}
-/// \brief Print this function profile on stream \p OS.
-///
-/// \param OS Stream to emit the output to.
-void SampleFunctionProfile::print(raw_ostream &OS) {
- OS << TotalSamples << ", " << TotalHeadSamples << ", " << BodySamples.size()
- << " sampled lines\n";
- for (BodySampleMap::const_iterator SI = BodySamples.begin(),
- SE = BodySamples.end();
- SI != SE; ++SI)
- OS << "\tline offset: " << SI->first.LineOffset
- << ", discriminator: " << SI->first.Discriminator
- << ", number of samples: " << SI->second << "\n";
- OS << "\n";
-}
-
/// \brief Print the weight of edge \p E on stream \p OS.
///
/// \param OS Stream to emit the output to.
/// \param E Edge to print.
-void SampleFunctionProfile::printEdgeWeight(raw_ostream &OS, Edge E) {
+void SampleProfileLoader::printEdgeWeight(raw_ostream &OS, Edge E) {
OS << "weight[" << E.first->getName() << "->" << E.second->getName()
<< "]: " << EdgeWeights[E] << "\n";
}
///
/// \param OS Stream to emit the output to.
/// \param BB Block to print.
-void SampleFunctionProfile::printBlockEquivalence(raw_ostream &OS,
- BasicBlock *BB) {
+void SampleProfileLoader::printBlockEquivalence(raw_ostream &OS,
+ BasicBlock *BB) {
BasicBlock *Equiv = EquivalenceClass[BB];
OS << "equivalence[" << BB->getName()
<< "]: " << ((Equiv) ? EquivalenceClass[BB]->getName() : "NONE") << "\n";
///
/// \param OS Stream to emit the output to.
/// \param BB Block to print.
-void SampleFunctionProfile::printBlockWeight(raw_ostream &OS, BasicBlock *BB) {
+void SampleProfileLoader::printBlockWeight(raw_ostream &OS, BasicBlock *BB) {
OS << "weight[" << BB->getName() << "]: " << BlockWeights[BB] << "\n";
}
-/// \brief Print the function profile for \p FName on stream \p OS.
-///
-/// \param OS Stream to emit the output to.
-/// \param FName Name of the function to print.
-void SampleModuleProfile::printFunctionProfile(raw_ostream &OS,
- StringRef FName) {
- OS << "Function: " << FName << ":\n";
- Profiles[FName].print(OS);
-}
-
-/// \brief Dump the function profile for \p FName.
-///
-/// \param FName Name of the function to print.
-void SampleModuleProfile::dumpFunctionProfile(StringRef FName) {
- printFunctionProfile(dbgs(), FName);
-}
-
-/// \brief Dump all the function profiles found.
-void SampleModuleProfile::dump() {
- for (StringMap<SampleFunctionProfile>::const_iterator I = Profiles.begin(),
- E = Profiles.end();
- I != E; ++I)
- dumpFunctionProfile(I->getKey());
-}
-
-/// \brief Load samples from a text file.
-///
-/// The file contains a list of samples for every function executed at
-/// runtime. Each function profile has the following format:
-///
-/// function1:total_samples:total_head_samples
-/// offset1[.discriminator]: number_of_samples [fn1:num fn2:num ... ]
-/// offset2[.discriminator]: number_of_samples [fn3:num fn4:num ... ]
-/// ...
-/// offsetN[.discriminator]: number_of_samples [fn5:num fn6:num ... ]
-///
-/// Function names must be mangled in order for the profile loader to
-/// match them in the current translation unit. The two numbers in the
-/// function header specify how many total samples were accumulated in
-/// the function (first number), and the total number of samples accumulated
-/// at the prologue of the function (second number). This head sample
-/// count provides an indicator of how frequent is the function invoked.
-///
-/// Each sampled line may contain several items. Some are optional
-/// (marked below):
-///
-/// a- Source line offset. This number represents the line number
-/// in the function where the sample was collected. The line number
-/// is always relative to the line where symbol of the function
-/// is defined. So, if the function has its header at line 280,
-/// the offset 13 is at line 293 in the file.
-///
-/// b- [OPTIONAL] Discriminator. This is used if the sampled program
-/// was compiled with DWARF discriminator support
-/// (http://wiki.dwarfstd.org/index.php?title=Path_Discriminators)
-///
-/// c- Number of samples. This is the number of samples collected by
-/// the profiler at this source location.
-///
-/// d- [OPTIONAL] Potential call targets and samples. If present, this
-/// line contains a call instruction. This models both direct and
-/// indirect calls. Each called target is listed together with the
-/// number of samples. For example,
-///
-/// 130: 7 foo:3 bar:2 baz:7
-///
-/// The above means that at relative line offset 130 there is a
-/// call instruction that calls one of foo(), bar() and baz(). With
-/// baz() being the relatively more frequent call target.
-///
-/// FIXME: This is currently unhandled, but it has a lot of
-/// potential for aiding the inliner.
-///
-///
-/// Since this is a flat profile, a function that shows up more than
-/// once gets all its samples aggregated across all its instances.
-///
-/// FIXME: flat profiles are too imprecise to provide good optimization
-/// opportunities. Convert them to context-sensitive profile.
-///
-/// This textual representation is useful to generate unit tests and
-/// for debugging purposes, but it should not be used to generate
-/// profiles for large programs, as the representation is extremely
-/// inefficient.
-///
-/// \returns true if the file was loaded successfully, false otherwise.
-bool SampleModuleProfile::loadText() {
- std::unique_ptr<MemoryBuffer> Buffer;
- error_code EC = MemoryBuffer::getFile(Filename, Buffer);
- if (EC) {
- std::string Msg(EC.message());
- M.getContext().diagnose(DiagnosticInfoSampleProfile(Filename.data(), Msg));
- return false;
- }
- line_iterator LineIt(*Buffer, '#');
-
- // Read the profile of each function. Since each function may be
- // mentioned more than once, and we are collecting flat profiles,
- // accumulate samples as we parse them.
- Regex HeadRE("^([^:]+):([0-9]+):([0-9]+)$");
- Regex LineSample("^([0-9]+)\\.?([0-9]+)?: ([0-9]+)(.*)$");
- while (!LineIt.is_at_eof()) {
- // Read the header of each function. The function header should
- // have this format:
- //
- // function_name:total_samples:total_head_samples
- //
- // See above for an explanation of each field.
- SmallVector<StringRef, 3> Matches;
- if (!HeadRE.match(*LineIt, &Matches)) {
- reportParseError(LineIt.line_number(),
- "Expected 'mangled_name:NUM:NUM', found " + *LineIt);
- return false;
- }
- assert(Matches.size() == 4);
- StringRef FName = Matches[1];
- unsigned NumSamples, NumHeadSamples;
- Matches[2].getAsInteger(10, NumSamples);
- Matches[3].getAsInteger(10, NumHeadSamples);
- Profiles[FName] = SampleFunctionProfile();
- SampleFunctionProfile &FProfile = Profiles[FName];
- FProfile.addTotalSamples(NumSamples);
- FProfile.addHeadSamples(NumHeadSamples);
- ++LineIt;
-
- // Now read the body. The body of the function ends when we reach
- // EOF or when we see the start of the next function.
- while (!LineIt.is_at_eof() && isdigit((*LineIt)[0])) {
- if (!LineSample.match(*LineIt, &Matches)) {
- reportParseError(
- LineIt.line_number(),
- "Expected 'NUM[.NUM]: NUM[ mangled_name:NUM]*', found " + *LineIt);
- return false;
- }
- assert(Matches.size() == 5);
- unsigned LineOffset, NumSamples, Discriminator = 0;
- Matches[1].getAsInteger(10, LineOffset);
- if (Matches[2] != "")
- Matches[2].getAsInteger(10, Discriminator);
- Matches[3].getAsInteger(10, NumSamples);
-
- // FIXME: Handle called targets (in Matches[4]).
-
- // When dealing with instruction weights, we use the value
- // zero to indicate the absence of a sample. If we read an
- // actual zero from the profile file, return it as 1 to
- // avoid the confusion later on.
- if (NumSamples == 0)
- NumSamples = 1;
- FProfile.addBodySamples(LineOffset, Discriminator, NumSamples);
- ++LineIt;
- }
- }
-
- return true;
-}
-
/// \brief Get the weight for an instruction.
///
/// The "weight" of an instruction \p Inst is the number of samples
/// \param Inst Instruction to query.
///
/// \returns The profiled weight of I.
-unsigned SampleFunctionProfile::getInstWeight(Instruction &Inst) {
+unsigned SampleProfileLoader::getInstWeight(Instruction &Inst) {
DebugLoc DLoc = Inst.getDebugLoc();
unsigned Lineno = DLoc.getLine();
if (Lineno < HeaderLineno)
DILocation DIL(DLoc.getAsMDNode(*Ctx));
int LOffset = Lineno - HeaderLineno;
unsigned Discriminator = DIL.getDiscriminator();
- unsigned Weight =
- BodySamples.lookup(InstructionLocation(LOffset, Discriminator));
+ unsigned Weight = Samples->samplesAt(LOffset, Discriminator);
DEBUG(dbgs() << " " << Lineno << "." << Discriminator << ":" << Inst
<< " (line offset: " << LOffset << "." << Discriminator
<< " - weight: " << Weight << ")\n");
/// \brief Compute the weight of a basic block.
///
-/// The weight of basic block \p B is the maximum weight of all the
-/// instructions in B. The weight of \p B is computed and cached in
+/// The weight of basic block \p BB is the maximum weight of all the
+/// instructions in BB. The weight of \p BB is computed and cached in
/// the BlockWeights map.
///
-/// \param B The basic block to query.
+/// \param BB The basic block to query.
///
-/// \returns The computed weight of B.
-unsigned SampleFunctionProfile::getBlockWeight(BasicBlock *B) {
- // If we've computed B's weight before, return it.
+/// \returns The computed weight of BB.
+unsigned SampleProfileLoader::getBlockWeight(BasicBlock *BB) {
+ // If we've computed BB's weight before, return it.
std::pair<BlockWeightMap::iterator, bool> Entry =
- BlockWeights.insert(std::make_pair(B, 0));
+ BlockWeights.insert(std::make_pair(BB, 0));
if (!Entry.second)
return Entry.first->second;
- // Otherwise, compute and cache B's weight.
+ // Otherwise, compute and cache BB's weight.
unsigned Weight = 0;
- for (BasicBlock::iterator I = B->begin(), E = B->end(); I != E; ++I) {
- unsigned InstWeight = getInstWeight(*I);
+ for (auto &I : BB->getInstList()) {
+ unsigned InstWeight = getInstWeight(I);
if (InstWeight > Weight)
Weight = InstWeight;
}
/// the weights of every basic block in the CFG.
///
/// \param F The function to query.
-bool SampleFunctionProfile::computeBlockWeights(Function &F) {
+bool SampleProfileLoader::computeBlockWeights(Function &F) {
bool Changed = false;
DEBUG(dbgs() << "Block weights\n");
- for (Function::iterator B = F.begin(), E = F.end(); B != E; ++B) {
- unsigned Weight = getBlockWeight(B);
+ for (auto &BB : F) {
+ unsigned Weight = getBlockWeight(&BB);
Changed |= (Weight > 0);
- DEBUG(printBlockWeight(dbgs(), B));
+ DEBUG(printBlockWeight(dbgs(), &BB));
}
return Changed;
/// \param DomTree Opposite dominator tree. If \p Descendants is filled
/// with blocks from \p BB1's dominator tree, then
/// this is the post-dominator tree, and vice versa.
-void SampleFunctionProfile::findEquivalencesFor(
+void SampleProfileLoader::findEquivalencesFor(
BasicBlock *BB1, SmallVector<BasicBlock *, 8> Descendants,
DominatorTreeBase<BasicBlock> *DomTree) {
- for (SmallVectorImpl<BasicBlock *>::iterator I = Descendants.begin(),
- E = Descendants.end();
- I != E; ++I) {
- BasicBlock *BB2 = *I;
+ for (auto *BB2 : Descendants) {
bool IsDomParent = DomTree->dominates(BB2, BB1);
bool IsInSameLoop = LI->getLoopFor(BB1) == LI->getLoopFor(BB2);
if (BB1 != BB2 && VisitedBlocks.insert(BB2) && IsDomParent &&
/// dominates B2, B2 post-dominates B1 and both are in the same loop.
///
/// \param F The function to query.
-void SampleFunctionProfile::findEquivalenceClasses(Function &F) {
+void SampleProfileLoader::findEquivalenceClasses(Function &F) {
SmallVector<BasicBlock *, 8> DominatedBBs;
DEBUG(dbgs() << "\nBlock equivalence classes\n");
// Find equivalence sets based on dominance and post-dominance information.
- for (Function::iterator B = F.begin(), E = F.end(); B != E; ++B) {
- BasicBlock *BB1 = B;
+ for (auto &BB : F) {
+ BasicBlock *BB1 = &BB;
// Compute BB1's equivalence class once.
if (EquivalenceClass.count(BB1)) {
// each equivalence class has the largest weight, assign that weight
// to all the blocks in that equivalence class.
DEBUG(dbgs() << "\nAssign the same weight to all blocks in the same class\n");
- for (Function::iterator B = F.begin(), E = F.end(); B != E; ++B) {
- BasicBlock *BB = B;
+ for (auto &BI : F) {
+ BasicBlock *BB = &BI;
BasicBlock *EquivBB = EquivalenceClass[BB];
if (BB != EquivBB)
BlockWeights[BB] = BlockWeights[EquivBB];
/// \param UnknownEdge Set if E has not been visited before.
///
/// \returns E's weight, if known. Otherwise, return 0.
-unsigned SampleFunctionProfile::visitEdge(Edge E, unsigned *NumUnknownEdges,
- Edge *UnknownEdge) {
+unsigned SampleProfileLoader::visitEdge(Edge E, unsigned *NumUnknownEdges,
+ Edge *UnknownEdge) {
if (!VisitedEdges.count(E)) {
(*NumUnknownEdges)++;
*UnknownEdge = E;
/// \param F Function to process.
///
/// \returns True if new weights were assigned to edges or blocks.
-bool SampleFunctionProfile::propagateThroughEdges(Function &F) {
+bool SampleProfileLoader::propagateThroughEdges(Function &F) {
bool Changed = false;
DEBUG(dbgs() << "\nPropagation through edges\n");
- for (Function::iterator BI = F.begin(), EI = F.end(); BI != EI; ++BI) {
- BasicBlock *BB = BI;
+ for (auto &BI : F) {
+ BasicBlock *BB = &BI;
// Visit all the predecessor and successor edges to determine
// which ones have a weight assigned already. Note that it doesn't
if (i == 0) {
// First, visit all predecessor edges.
- for (size_t I = 0; I < Predecessors[BB].size(); I++) {
- Edge E = std::make_pair(Predecessors[BB][I], BB);
+ for (auto *Pred : Predecessors[BB]) {
+ Edge E = std::make_pair(Pred, BB);
TotalWeight += visitEdge(E, &NumUnknownEdges, &UnknownEdge);
if (E.first == E.second)
SelfReferentialEdge = E;
}
} else {
// On the second round, visit all successor edges.
- for (size_t I = 0; I < Successors[BB].size(); I++) {
- Edge E = std::make_pair(BB, Successors[BB][I]);
+ for (auto *Succ : Successors[BB]) {
+ Edge E = std::make_pair(BB, Succ);
TotalWeight += visitEdge(E, &NumUnknownEdges, &UnknownEdge);
}
}
///
/// We are interested in unique edges. If a block B1 has multiple
/// edges to another block B2, we only add a single B1->B2 edge.
-void SampleFunctionProfile::buildEdges(Function &F) {
- for (Function::iterator I = F.begin(), E = F.end(); I != E; ++I) {
- BasicBlock *B1 = I;
+void SampleProfileLoader::buildEdges(Function &F) {
+ for (auto &BI : F) {
+ BasicBlock *B1 = &BI;
// Add predecessors for B1.
SmallPtrSet<BasicBlock *, 16> Visited;
/// \brief Propagate weights into edges
///
-/// The following rules are applied to every block B in the CFG:
+/// The following rules are applied to every block BB in the CFG:
///
-/// - If B has a single predecessor/successor, then the weight
+/// - If BB has a single predecessor/successor, then the weight
/// of that edge is the weight of the block.
///
/// - If all incoming or outgoing edges are known except one, and the
/// weight of the block is already known, the weight of the unknown
/// edge will be the weight of the block minus the sum of all the known
-/// edges. If the sum of all the known edges is larger than B's weight,
+/// edges. If the sum of all the known edges is larger than BB's weight,
/// we set the unknown edge weight to zero.
///
/// - If there is a self-referential edge, and the weight of the block is
/// known, the weight for that edge is set to the weight of the block
/// minus the weight of the other incoming edges to that block (if
/// known).
-void SampleFunctionProfile::propagateWeights(Function &F) {
+void SampleProfileLoader::propagateWeights(Function &F) {
bool Changed = true;
unsigned i = 0;
// edge weights computed during propagation.
DEBUG(dbgs() << "\nPropagation complete. Setting branch weights\n");
MDBuilder MDB(F.getContext());
- for (Function::iterator I = F.begin(), E = F.end(); I != E; ++I) {
- BasicBlock *B = I;
- TerminatorInst *TI = B->getTerminator();
+ for (auto &BI : F) {
+ BasicBlock *BB = &BI;
+ TerminatorInst *TI = BB->getTerminator();
if (TI->getNumSuccessors() == 1)
continue;
if (!isa<BranchInst>(TI) && !isa<SwitchInst>(TI))
bool AllWeightsZero = true;
for (unsigned I = 0; I < TI->getNumSuccessors(); ++I) {
BasicBlock *Succ = TI->getSuccessor(I);
- Edge E = std::make_pair(B, Succ);
+ Edge E = std::make_pair(BB, Succ);
unsigned Weight = EdgeWeights[E];
DEBUG(dbgs() << "\t"; printEdgeWeight(dbgs(), E));
Weights.push_back(Weight);
///
/// \returns the line number where \p F is defined. If it returns 0,
/// it means that there is no debug information available for \p F.
-unsigned SampleFunctionProfile::getFunctionLoc(Function &F) {
- NamedMDNode *CUNodes = F.getParent()->getNamedMetadata("llvm.dbg.cu");
- if (CUNodes) {
- for (unsigned I = 0, E1 = CUNodes->getNumOperands(); I != E1; ++I) {
- DICompileUnit CU(CUNodes->getOperand(I));
- DIArray Subprograms = CU.getSubprograms();
- for (unsigned J = 0, E2 = Subprograms.getNumElements(); J != E2; ++J) {
- DISubprogram Subprogram(Subprograms.getElement(J));
- if (Subprogram.describes(&F))
- return Subprogram.getLineNumber();
- }
- }
- }
-
- StringRef FnName = F.getName();
- Twine Msg = "No debug information found in function " + FnName;
- DiagnosticInfoSampleProfile Diag(Msg);
- F.getContext().diagnose(Diag);
+unsigned SampleProfileLoader::getFunctionLoc(Function &F) {
+ DISubprogram S = getDISubprogram(&F);
+ if (S.isSubprogram())
+ return S.getLineNumber();
+
+ // If could not find the start of \p F, emit a diagnostic to inform the user
+ // about the missed opportunity.
+ F.getContext().diagnose(DiagnosticInfoSampleProfile(
+ "No debug information found in function " + F.getName() +
+ ": Function profile not used",
+ DS_Warning));
return 0;
}
///
/// 3- Propagation of block weights into edges. This uses a simple
/// propagation heuristic. The following rules are applied to every
-/// block B in the CFG:
+/// block BB in the CFG:
///
-/// - If B has a single predecessor/successor, then the weight
+/// - If BB has a single predecessor/successor, then the weight
/// of that edge is the weight of the block.
///
/// - If all the edges are known except one, and the weight of the
/// block is already known, the weight of the unknown edge will
/// be the weight of the block minus the sum of all the known
-/// edges. If the sum of all the known edges is larger than B's weight,
+/// edges. If the sum of all the known edges is larger than BB's weight,
/// we set the unknown edge weight to zero.
///
/// - If there is a self-referential edge, and the weight of the block is
/// work here.
///
/// Once all the branch weights are computed, we emit the MD_prof
-/// metadata on B using the computed values for each of its branches.
+/// metadata on BB using the computed values for each of its branches.
///
/// \param F The function to query.
///
/// \returns true if \p F was modified. Returns false, otherwise.
-bool SampleFunctionProfile::emitAnnotations(Function &F, DominatorTree *DomTree,
- PostDominatorTree *PostDomTree,
- LoopInfo *Loops) {
+bool SampleProfileLoader::emitAnnotations(Function &F) {
bool Changed = false;
// Initialize invariants used during computation and propagation.
DEBUG(dbgs() << "Line number for the first instruction in " << F.getName()
<< ": " << HeaderLineno << "\n");
- DT = DomTree;
- PDT = PostDomTree;
- LI = Loops;
- Ctx = &F.getParent()->getContext();
// Compute basic block weights.
Changed |= computeBlockWeights(F);
"Sample Profile loader", false, false)
bool SampleProfileLoader::doInitialization(Module &M) {
- Profiler.reset(new SampleModuleProfile(M, Filename));
- ProfileIsValid = Profiler->loadText();
+ if (std::error_code EC =
+ SampleProfileReader::create(Filename, Reader, M.getContext())) {
+ std::string Msg = "Could not open profile: " + EC.message();
+ M.getContext().diagnose(DiagnosticInfoSampleProfile(Filename.data(), Msg));
+ return false;
+ }
+ ProfileIsValid = (Reader->read() == sampleprof_error::success);
return true;
}
bool SampleProfileLoader::runOnFunction(Function &F) {
if (!ProfileIsValid)
return false;
- DominatorTree *DT = &getAnalysis<DominatorTreeWrapperPass>().getDomTree();
- PostDominatorTree *PDT = &getAnalysis<PostDominatorTree>();
- LoopInfo *LI = &getAnalysis<LoopInfo>();
- SampleFunctionProfile &FunctionProfile = Profiler->getProfile(F);
- if (!FunctionProfile.empty())
- return FunctionProfile.emitAnnotations(F, DT, PDT, LI);
+
+ DT = &getAnalysis<DominatorTreeWrapperPass>().getDomTree();
+ PDT = &getAnalysis<PostDominatorTree>();
+ LI = &getAnalysis<LoopInfo>();
+ Ctx = &F.getParent()->getContext();
+ Samples = Reader->getSamplesFor(F);
+ if (!Samples->empty())
+ return emitAnnotations(F);
return false;
}
projects / oota-llvm.git / blobdiff