blob: 55418d9d0f9cb60a05c4907c72216a4ab0f5e7f0 [file] [log] [blame]
//===- SampleProf.h - Sampling profiling format support ---------*- C++ -*-===//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
// This file contains common definitions used in the reading and writing of
// sample profile data.
#include "llvm/ADT/DenseSet.h"
#include "llvm/ADT/SmallVector.h"
#include "llvm/ADT/StringMap.h"
#include "llvm/ADT/StringRef.h"
#include "llvm/ADT/StringSet.h"
#include "llvm/IR/Function.h"
#include "llvm/IR/GlobalValue.h"
#include "llvm/IR/Module.h"
#include "llvm/Support/Debug.h"
#include "llvm/Support/ErrorOr.h"
#include "llvm/Support/MathExtras.h"
#include "llvm/Support/raw_ostream.h"
#include <algorithm>
#include <cstdint>
#include <map>
#include <set>
#include <string>
#include <system_error>
#include <utility>
namespace llvm {
class raw_ostream;
const std::error_category &sampleprof_category();
enum class sampleprof_error {
success = 0,
inline std::error_code make_error_code(sampleprof_error E) {
return std::error_code(static_cast<int>(E), sampleprof_category());
inline sampleprof_error MergeResult(sampleprof_error &Accumulator,
sampleprof_error Result) {
// Prefer first error encountered as later errors may be secondary effects of
// the initial problem.
if (Accumulator == sampleprof_error::success &&
Result != sampleprof_error::success)
Accumulator = Result;
return Accumulator;
} // end namespace llvm
namespace std {
template <>
struct is_error_code_enum<llvm::sampleprof_error> : std::true_type {};
} // end namespace std
namespace llvm {
namespace sampleprof {
enum SampleProfileFormat {
SPF_None = 0,
SPF_Text = 0x1,
SPF_Compact_Binary = 0x2,
SPF_GCC = 0x3,
SPF_Ext_Binary = 0x4,
SPF_Binary = 0xff
static inline uint64_t SPMagic(SampleProfileFormat Format = SPF_Binary) {
return uint64_t('S') << (64 - 8) | uint64_t('P') << (64 - 16) |
uint64_t('R') << (64 - 24) | uint64_t('O') << (64 - 32) |
uint64_t('F') << (64 - 40) | uint64_t('4') << (64 - 48) |
uint64_t('2') << (64 - 56) | uint64_t(Format);
// Get the proper representation of a string in the input Format.
static inline StringRef getRepInFormat(StringRef Name,
SampleProfileFormat Format,
std::string &GUIDBuf) {
if (Name.empty())
return Name;
GUIDBuf = std::to_string(Function::getGUID(Name));
return (Format == SPF_Compact_Binary) ? StringRef(GUIDBuf) : Name;
static inline uint64_t SPVersion() { return 103; }
// Section Type used by SampleProfileExtBinaryBaseReader and
// SampleProfileExtBinaryBaseWriter. Never change the existing
// value of enum. Only append new ones.
enum SecType {
SecInValid = 0,
SecProfSummary = 1,
SecNameTable = 2,
SecProfileSymbolList = 3,
SecFuncOffsetTable = 4,
// marker for the first type of profile.
SecFuncProfileFirst = 32,
SecLBRProfile = SecFuncProfileFirst
static inline std::string getSecName(SecType Type) {
switch (Type) {
case SecInValid:
return "InvalidSection";
case SecProfSummary:
return "ProfileSummarySection";
case SecNameTable:
return "NameTableSection";
case SecProfileSymbolList:
return "ProfileSymbolListSection";
case SecFuncOffsetTable:
return "FuncOffsetTableSection";
case SecLBRProfile:
return "LBRProfileSection";
llvm_unreachable("A SecType has no name for output");
// Entry type of section header table used by SampleProfileExtBinaryBaseReader
// and SampleProfileExtBinaryBaseWriter.
struct SecHdrTableEntry {
SecType Type;
uint64_t Flags;
uint64_t Offset;
uint64_t Size;
enum SecFlags { SecFlagInValid = 0, SecFlagCompress = (1 << 0) };
static inline void addSecFlags(SecHdrTableEntry &Entry, uint64_t Flags) {
Entry.Flags |= Flags;
static inline void removeSecFlags(SecHdrTableEntry &Entry, uint64_t Flags) {
Entry.Flags &= ~Flags;
static inline bool hasSecFlag(SecHdrTableEntry &Entry, SecFlags Flag) {
return Entry.Flags & Flag;
/// 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 LineLocation {
LineLocation(uint32_t L, uint32_t D) : LineOffset(L), Discriminator(D) {}
void print(raw_ostream &OS) const;
void dump() const;
bool operator<(const LineLocation &O) const {
return LineOffset < O.LineOffset ||
(LineOffset == O.LineOffset && Discriminator < O.Discriminator);
uint32_t LineOffset;
uint32_t Discriminator;
raw_ostream &operator<<(raw_ostream &OS, const LineLocation &Loc);
/// Representation of a single sample record.
/// A sample record is represented by a positive integer value, which
/// indicates how frequently was the associated line location executed.
/// Additionally, if the associated location contains a function call,
/// the record will hold a list of all the possible called targets. For
/// direct calls, this will be the exact function being invoked. For
/// indirect calls (function pointers, virtual table dispatch), this
/// will be a list of one or more functions.
class SampleRecord {
using CallTarget = std::pair<StringRef, uint64_t>;
struct CallTargetComparator {
bool operator()(const CallTarget &LHS, const CallTarget &RHS) const {
if (LHS.second != RHS.second)
return LHS.second > RHS.second;
return LHS.first < RHS.first;
using SortedCallTargetSet = std::set<CallTarget, CallTargetComparator>;
using CallTargetMap = StringMap<uint64_t>;
SampleRecord() = default;
/// Increment the number of samples for this record by \p S.
/// Optionally scale sample count \p S by \p Weight.
/// Sample counts accumulate using saturating arithmetic, to avoid wrapping
/// around unsigned integers.
sampleprof_error addSamples(uint64_t S, uint64_t Weight = 1) {
bool Overflowed;
NumSamples = SaturatingMultiplyAdd(S, Weight, NumSamples, &Overflowed);
return Overflowed ? sampleprof_error::counter_overflow
: sampleprof_error::success;
/// Add called function \p F with samples \p S.
/// Optionally scale sample count \p S by \p Weight.
/// Sample counts accumulate using saturating arithmetic, to avoid wrapping
/// around unsigned integers.
sampleprof_error addCalledTarget(StringRef F, uint64_t S,
uint64_t Weight = 1) {
uint64_t &TargetSamples = CallTargets[F];
bool Overflowed;
TargetSamples =
SaturatingMultiplyAdd(S, Weight, TargetSamples, &Overflowed);
return Overflowed ? sampleprof_error::counter_overflow
: sampleprof_error::success;
/// Return true if this sample record contains function calls.
bool hasCalls() const { return !CallTargets.empty(); }
uint64_t getSamples() const { return NumSamples; }
const CallTargetMap &getCallTargets() const { return CallTargets; }
const SortedCallTargetSet getSortedCallTargets() const {
return SortCallTargets(CallTargets);
/// Sort call targets in descending order of call frequency.
static const SortedCallTargetSet SortCallTargets(const CallTargetMap &Targets) {
SortedCallTargetSet SortedTargets;
for (const auto &I : Targets) {
SortedTargets.emplace(I.first(), I.second);
return SortedTargets;
/// Merge the samples in \p Other into this record.
/// Optionally scale sample counts by \p Weight.
sampleprof_error merge(const SampleRecord &Other, uint64_t Weight = 1) {
sampleprof_error Result = addSamples(Other.getSamples(), Weight);
for (const auto &I : Other.getCallTargets()) {
MergeResult(Result, addCalledTarget(I.first(), I.second, Weight));
return Result;
void print(raw_ostream &OS, unsigned Indent) const;
void dump() const;
uint64_t NumSamples = 0;
CallTargetMap CallTargets;
raw_ostream &operator<<(raw_ostream &OS, const SampleRecord &Sample);
class FunctionSamples;
using BodySampleMap = std::map<LineLocation, SampleRecord>;
// NOTE: Using a StringMap here makes parsed profiles consume around 17% more
// memory, which is *very* significant for large profiles.
using FunctionSamplesMap = std::map<std::string, FunctionSamples, std::less<>>;
using CallsiteSampleMap = std::map<LineLocation, FunctionSamplesMap>;
/// Representation of the samples collected for a function.
/// This data structure contains all the collected samples for the body
/// of a function. Each sample corresponds to a LineLocation instance
/// within the body of the function.
class FunctionSamples {
FunctionSamples() = default;
void print(raw_ostream &OS = dbgs(), unsigned Indent = 0) const;
void dump() const;
sampleprof_error addTotalSamples(uint64_t Num, uint64_t Weight = 1) {
bool Overflowed;
TotalSamples =
SaturatingMultiplyAdd(Num, Weight, TotalSamples, &Overflowed);
return Overflowed ? sampleprof_error::counter_overflow
: sampleprof_error::success;
sampleprof_error addHeadSamples(uint64_t Num, uint64_t Weight = 1) {
bool Overflowed;
TotalHeadSamples =
SaturatingMultiplyAdd(Num, Weight, TotalHeadSamples, &Overflowed);
return Overflowed ? sampleprof_error::counter_overflow
: sampleprof_error::success;
sampleprof_error addBodySamples(uint32_t LineOffset, uint32_t Discriminator,
uint64_t Num, uint64_t Weight = 1) {
return BodySamples[LineLocation(LineOffset, Discriminator)].addSamples(
Num, Weight);
sampleprof_error addCalledTargetSamples(uint32_t LineOffset,
uint32_t Discriminator,
StringRef FName, uint64_t Num,
uint64_t Weight = 1) {
return BodySamples[LineLocation(LineOffset, Discriminator)].addCalledTarget(
FName, Num, Weight);
/// Return the number of samples collected at the given location.
/// Each location is specified by \p LineOffset and \p Discriminator.
/// If the location is not found in profile, return error.
ErrorOr<uint64_t> findSamplesAt(uint32_t LineOffset,
uint32_t Discriminator) const {
const auto &ret = BodySamples.find(LineLocation(LineOffset, Discriminator));
if (ret == BodySamples.end())
return std::error_code();
return ret->second.getSamples();
/// Returns the call target map collected at a given location.
/// Each location is specified by \p LineOffset and \p Discriminator.
/// If the location is not found in profile, return error.
findCallTargetMapAt(uint32_t LineOffset, uint32_t Discriminator) const {
const auto &ret = BodySamples.find(LineLocation(LineOffset, Discriminator));
if (ret == BodySamples.end())
return std::error_code();
return ret->second.getCallTargets();
/// Return the function samples at the given callsite location.
FunctionSamplesMap &functionSamplesAt(const LineLocation &Loc) {
return CallsiteSamples[Loc];
/// Returns the FunctionSamplesMap at the given \p Loc.
const FunctionSamplesMap *
findFunctionSamplesMapAt(const LineLocation &Loc) const {
auto iter = CallsiteSamples.find(Loc);
if (iter == CallsiteSamples.end())
return nullptr;
return &iter->second;
/// Returns a pointer to FunctionSamples at the given callsite location \p Loc
/// with callee \p CalleeName. If no callsite can be found, relax the
/// restriction to return the FunctionSamples at callsite location \p Loc
/// with the maximum total sample count.
const FunctionSamples *findFunctionSamplesAt(const LineLocation &Loc,
StringRef CalleeName) const {
std::string CalleeGUID;
CalleeName = getRepInFormat(CalleeName, Format, CalleeGUID);
auto iter = CallsiteSamples.find(Loc);
if (iter == CallsiteSamples.end())
return nullptr;
auto FS = iter->second.find(CalleeName);
if (FS != iter->second.end())
return &FS->second;
// If we cannot find exact match of the callee name, return the FS with
// the max total count.
uint64_t MaxTotalSamples = 0;
const FunctionSamples *R = nullptr;
for (const auto &NameFS : iter->second)
if (NameFS.second.getTotalSamples() >= MaxTotalSamples) {
MaxTotalSamples = NameFS.second.getTotalSamples();
R = &NameFS.second;
return R;
bool empty() const { return TotalSamples == 0; }
/// Return the total number of samples collected inside the function.
uint64_t getTotalSamples() const { return TotalSamples; }
/// Return the total number of branch samples that have the function as the
/// branch target. This should be equivalent to the sample of the first
/// instruction of the symbol. But as we directly get this info for raw
/// profile without referring to potentially inaccurate debug info, this
/// gives more accurate profile data and is preferred for standalone symbols.
uint64_t getHeadSamples() const { return TotalHeadSamples; }
/// Return the sample count of the first instruction of the function.
/// The function can be either a standalone symbol or an inlined function.
uint64_t getEntrySamples() const {
// Use either BodySamples or CallsiteSamples which ever has the smaller
// lineno.
if (!BodySamples.empty() &&
(CallsiteSamples.empty() ||
BodySamples.begin()->first < CallsiteSamples.begin()->first))
return BodySamples.begin()->second.getSamples();
if (!CallsiteSamples.empty()) {
uint64_t T = 0;
// An indirect callsite may be promoted to several inlined direct calls.
// We need to get the sum of them.
for (const auto &N_FS : CallsiteSamples.begin()->second)
T += N_FS.second.getEntrySamples();
return T;
return 0;
/// Return all the samples collected in the body of the function.
const BodySampleMap &getBodySamples() const { return BodySamples; }
/// Return all the callsite samples collected in the body of the function.
const CallsiteSampleMap &getCallsiteSamples() const {
return CallsiteSamples;
/// Merge the samples in \p Other into this one.
/// Optionally scale samples by \p Weight.
sampleprof_error merge(const FunctionSamples &Other, uint64_t Weight = 1) {
sampleprof_error Result = sampleprof_error::success;
Name = Other.getName();
MergeResult(Result, addTotalSamples(Other.getTotalSamples(), Weight));
MergeResult(Result, addHeadSamples(Other.getHeadSamples(), Weight));
for (const auto &I : Other.getBodySamples()) {
const LineLocation &Loc = I.first;
const SampleRecord &Rec = I.second;
MergeResult(Result, BodySamples[Loc].merge(Rec, Weight));
for (const auto &I : Other.getCallsiteSamples()) {
const LineLocation &Loc = I.first;
FunctionSamplesMap &FSMap = functionSamplesAt(Loc);
for (const auto &Rec : I.second)
MergeResult(Result, FSMap[Rec.first].merge(Rec.second, Weight));
return Result;
/// Recursively traverses all children, if the total sample count of the
/// corresponding function is no less than \p Threshold, add its corresponding
/// GUID to \p S. Also traverse the BodySamples to add hot CallTarget's GUID
/// to \p S.
void findInlinedFunctions(DenseSet<GlobalValue::GUID> &S, const Module *M,
uint64_t Threshold) const {
if (TotalSamples <= Threshold)
// Import hot CallTargets, which may not be available in IR because full
// profile annotation cannot be done until backend compilation in ThinLTO.
for (const auto &BS : BodySamples)
for (const auto &TS : BS.second.getCallTargets())
if (TS.getValue() > Threshold) {
const Function *Callee =
M->getFunction(getNameInModule(TS.getKey(), M));
if (!Callee || !Callee->getSubprogram())
for (const auto &CS : CallsiteSamples)
for (const auto &NameFS : CS.second)
NameFS.second.findInlinedFunctions(S, M, Threshold);
/// Set the name of the function.
void setName(StringRef FunctionName) { Name = FunctionName; }
/// Return the function name.
StringRef getName() const { return Name; }
/// Return the original function name if it exists in Module \p M.
StringRef getFuncNameInModule(const Module *M) const {
return getNameInModule(Name, M);
/// Return the canonical name for a function, taking into account
/// suffix elision policy attributes.
static StringRef getCanonicalFnName(const Function &F) {
static const char *knownSuffixes[] = { ".llvm.", ".part." };
auto AttrName = "sample-profile-suffix-elision-policy";
auto Attr = F.getFnAttribute(AttrName).getValueAsString();
if (Attr == "" || Attr == "all") {
return F.getName().split('.').first;
} else if (Attr == "selected") {
StringRef Cand(F.getName());
for (const auto &Suf : knownSuffixes) {
StringRef Suffix(Suf);
auto It = Cand.rfind(Suffix);
if (It == StringRef::npos)
return Cand;
auto Dit = Cand.rfind('.');
if (Dit == It + Suffix.size() - 1)
Cand = Cand.substr(0, It);
return Cand;
} else if (Attr == "none") {
return F.getName();
} else {
assert(false && "internal error: unknown suffix elision policy");
return F.getName();
/// Translate \p Name into its original name in Module.
/// When the Format is not SPF_Compact_Binary, \p Name needs no translation.
/// When the Format is SPF_Compact_Binary, \p Name in current FunctionSamples
/// is actually GUID of the original function name. getNameInModule will
/// translate \p Name in current FunctionSamples into its original name.
/// If the original name doesn't exist in \p M, return empty StringRef.
StringRef getNameInModule(StringRef Name, const Module *M) const {
if (Format != SPF_Compact_Binary)
return Name;
assert(GUIDToFuncNameMap && "GUIDToFuncNameMap needs to be popluated first");
auto iter = GUIDToFuncNameMap->find(std::stoull(;
if (iter == GUIDToFuncNameMap->end())
return StringRef();
return iter->second;
/// Returns the line offset to the start line of the subprogram.
/// We assume that a single function will not exceed 65535 LOC.
static unsigned getOffset(const DILocation *DIL);
/// Get the FunctionSamples of the inline instance where DIL originates
/// from.
/// The FunctionSamples of the instruction (Machine or IR) associated to
/// \p DIL is the inlined instance in which that instruction is coming from.
/// We traverse the inline stack of that instruction, and match it with the
/// tree nodes in the profile.
/// \returns the FunctionSamples pointer to the inlined instance.
const FunctionSamples *findFunctionSamples(const DILocation *DIL) const;
static SampleProfileFormat Format;
/// GUIDToFuncNameMap saves the mapping from GUID to the symbol name, for
/// all the function symbols defined or declared in current module.
DenseMap<uint64_t, StringRef> *GUIDToFuncNameMap = nullptr;
// Assume the input \p Name is a name coming from FunctionSamples itself.
// If the format is SPF_Compact_Binary, the name is already a GUID and we
// don't want to return the GUID of GUID.
static uint64_t getGUID(StringRef Name) {
return (Format == SPF_Compact_Binary) ? std::stoull(
: Function::getGUID(Name);
/// Mangled name of the function.
StringRef Name;
/// 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.
uint64_t TotalSamples = 0;
/// Total number of samples collected at the head of the function.
/// This is an approximation of the number of calls made to this function
/// at runtime.
uint64_t TotalHeadSamples = 0;
/// Map instruction locations 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;
/// Map call sites to collected samples for the called function.
/// Each entry in this map corresponds to all the samples
/// collected for the inlined function call at the given
/// location. For example, given:
/// void foo() {
/// 1 bar();
/// ...
/// 8 baz();
/// }
/// If the bar() and baz() calls were inlined inside foo(), this
/// map will contain two entries. One for all the samples collected
/// in the call to bar() at line offset 1, the other for all the samples
/// collected in the call to baz() at line offset 8.
CallsiteSampleMap CallsiteSamples;
raw_ostream &operator<<(raw_ostream &OS, const FunctionSamples &FS);
/// Sort a LocationT->SampleT map by LocationT.
/// It produces a sorted list of <LocationT, SampleT> records by ascending
/// order of LocationT.
template <class LocationT, class SampleT> class SampleSorter {
using SamplesWithLoc = std::pair<const LocationT, SampleT>;
using SamplesWithLocList = SmallVector<const SamplesWithLoc *, 20>;
SampleSorter(const std::map<LocationT, SampleT> &Samples) {
for (const auto &I : Samples)
llvm::stable_sort(V, [](const SamplesWithLoc *A, const SamplesWithLoc *B) {
return A->first < B->first;
const SamplesWithLocList &get() const { return V; }
SamplesWithLocList V;
/// ProfileSymbolList records the list of function symbols shown up
/// in the binary used to generate the profile. It is useful to
/// to discriminate a function being so cold as not to shown up
/// in the profile and a function newly added.
class ProfileSymbolList {
/// copy indicates whether we need to copy the underlying memory
/// for the input Name.
void add(StringRef Name, bool copy = false) {
if (!copy) {
bool contains(StringRef Name) { return Syms.count(Name); }
void merge(const ProfileSymbolList &List) {
for (auto Sym : List.Syms)
add(Sym, true);
unsigned size() { return Syms.size(); }
void setToCompress(bool TC) { ToCompress = TC; }
bool toCompress() { return ToCompress; }
std::error_code read(const uint8_t *Data, uint64_t ListSize);
std::error_code write(raw_ostream &OS);
void dump(raw_ostream &OS = dbgs()) const;
// Determine whether or not to compress the symbol list when
// writing it into profile. The variable is unused when the symbol
// list is read from an existing profile.
bool ToCompress = false;
DenseSet<StringRef> Syms;
BumpPtrAllocator Allocator;
} // end namespace sampleprof
} // end namespace llvm