| 1 | //===- InstrProfWriter.h - Instrumented profiling writer --------*- C++ -*-===// |
| 2 | // |
| 3 | // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. |
| 4 | // See https://llvm.org/LICENSE.txt for license information. |
| 5 | // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception |
| 6 | // |
| 7 | //===----------------------------------------------------------------------===// |
| 8 | // |
| 9 | // This file contains support for writing profiling data for instrumentation |
| 10 | // based PGO and coverage. |
| 11 | // |
| 12 | //===----------------------------------------------------------------------===// |
| 13 | |
| 14 | #ifndef LLVM_PROFILEDATA_INSTRPROFWRITER_H |
| 15 | #define LLVM_PROFILEDATA_INSTRPROFWRITER_H |
| 16 | |
| 17 | #include "llvm/ADT/DenseMap.h" |
| 18 | #include "llvm/ADT/MapVector.h" |
| 19 | #include "llvm/ADT/StringMap.h" |
| 20 | #include "llvm/IR/GlobalValue.h" |
| 21 | #include "llvm/Object/BuildID.h" |
| 22 | #include "llvm/ProfileData/DataAccessProf.h" |
| 23 | #include "llvm/ProfileData/IndexedMemProfData.h" |
| 24 | #include "llvm/ProfileData/InstrProf.h" |
| 25 | #include "llvm/ProfileData/MemProfSummaryBuilder.h" |
| 26 | #include "llvm/Support/Compiler.h" |
| 27 | #include "llvm/Support/Error.h" |
| 28 | #include <cstdint> |
| 29 | #include <memory> |
| 30 | #include <random> |
| 31 | |
| 32 | namespace llvm { |
| 33 | |
| 34 | /// Writer for instrumentation based profile data. |
| 35 | class InstrProfRecordWriterTrait; |
| 36 | class ProfOStream; |
| 37 | class MemoryBuffer; |
| 38 | class raw_fd_ostream; |
| 39 | |
| 40 | class InstrProfWriter { |
| 41 | public: |
| 42 | using ProfilingData = SmallDenseMap<uint64_t, InstrProfRecord>; |
| 43 | |
| 44 | private: |
| 45 | bool Sparse; |
| 46 | StringMap<ProfilingData> FunctionData; |
| 47 | /// The maximum length of a single temporal profile trace. |
| 48 | uint64_t MaxTemporalProfTraceLength; |
| 49 | /// The maximum number of stored temporal profile traces. |
| 50 | uint64_t TemporalProfTraceReservoirSize; |
| 51 | /// The total number of temporal profile traces seen. |
| 52 | uint64_t TemporalProfTraceStreamSize = 0; |
| 53 | /// The list of temporal profile traces. |
| 54 | SmallVector<TemporalProfTraceTy> TemporalProfTraces; |
| 55 | std::mt19937 RNG; |
| 56 | |
| 57 | // The MemProf data. |
| 58 | memprof::IndexedMemProfData MemProfData; |
| 59 | |
| 60 | // List of binary ids. |
| 61 | std::vector<llvm::object::BuildID> BinaryIds; |
| 62 | |
| 63 | // Read the vtable names from raw instr profile reader. |
| 64 | StringSet<> VTableNames; |
| 65 | |
| 66 | // An enum describing the attributes of the profile. |
| 67 | InstrProfKind ProfileKind = InstrProfKind::Unknown; |
| 68 | // Use raw pointer here for the incomplete type object. |
| 69 | InstrProfRecordWriterTrait *InfoObj; |
| 70 | |
| 71 | // Temporary support for writing the previous version of the format, to enable |
| 72 | // some forward compatibility. Currently this suppresses the writing of the |
| 73 | // new vtable names section and header fields. |
| 74 | // TODO: Consider enabling this with future version changes as well, to ease |
| 75 | // deployment of newer versions of llvm-profdata. |
| 76 | bool WritePrevVersion = false; |
| 77 | |
| 78 | // The MemProf version we should write. |
| 79 | memprof::IndexedVersion MemProfVersionRequested; |
| 80 | |
| 81 | // Whether to serialize the full schema. |
| 82 | bool MemProfFullSchema; |
| 83 | |
| 84 | // Whether to generated random memprof hotness for testing. |
| 85 | bool MemprofGenerateRandomHotness; |
| 86 | |
| 87 | std::unique_ptr<memprof::DataAccessProfData> DataAccessProfileData; |
| 88 | |
| 89 | // MemProf summary builder to which records are added as MemProf data is added |
| 90 | // to the writer. |
| 91 | memprof::MemProfSummaryBuilder MemProfSumBuilder; |
| 92 | |
| 93 | public: |
| 94 | // For memprof testing, random hotness can be assigned to the contexts if |
| 95 | // MemprofGenerateRandomHotness is enabled. The random seed can be either |
| 96 | // provided by MemprofGenerateRandomHotnessSeed, or if that is 0, one will be |
| 97 | // generated in the writer using the current time. |
| 98 | LLVM_ABI InstrProfWriter(bool Sparse = false, |
| 99 | uint64_t TemporalProfTraceReservoirSize = 0, |
| 100 | uint64_t MaxTemporalProfTraceLength = 0, |
| 101 | bool WritePrevVersion = false, |
| 102 | memprof::IndexedVersion MemProfVersionRequested = |
| 103 | static_cast<memprof::IndexedVersion>( |
| 104 | memprof::MinimumSupportedVersion), |
| 105 | bool MemProfFullSchema = false, |
| 106 | bool MemprofGenerateRandomHotness = false, |
| 107 | unsigned MemprofGenerateRandomHotnessSeed = 0); |
| 108 | LLVM_ABI ~InstrProfWriter(); |
| 109 | |
| 110 | StringMap<ProfilingData> &getProfileData() { return FunctionData; } |
| 111 | |
| 112 | /// Add function counts for the given function. If there are already counts |
| 113 | /// for this function and the hash and number of counts match, each counter is |
| 114 | /// summed. Optionally scale counts by \p Weight. |
| 115 | LLVM_ABI void addRecord(NamedInstrProfRecord &&I, uint64_t Weight, |
| 116 | function_ref<void(Error)> Warn); |
| 117 | void addRecord(NamedInstrProfRecord &&I, function_ref<void(Error)> Warn) { |
| 118 | addRecord(I: std::move(I), Weight: 1, Warn); |
| 119 | } |
| 120 | void addVTableName(StringRef VTableName) { VTableNames.insert(key: VTableName); } |
| 121 | |
| 122 | /// Add \p SrcTraces using reservoir sampling where \p SrcStreamSize is the |
| 123 | /// total number of temporal profiling traces the source has seen. |
| 124 | LLVM_ABI void |
| 125 | addTemporalProfileTraces(SmallVectorImpl<TemporalProfTraceTy> &SrcTraces, |
| 126 | uint64_t SrcStreamSize); |
| 127 | |
| 128 | /// Add the entire MemProfData \p Incoming to the writer context. |
| 129 | LLVM_ABI bool addMemProfData(memprof::IndexedMemProfData Incoming, |
| 130 | function_ref<void(Error)> Warn); |
| 131 | |
| 132 | // Add a binary id to the binary ids list. |
| 133 | LLVM_ABI void addBinaryIds(ArrayRef<llvm::object::BuildID> BIs); |
| 134 | |
| 135 | LLVM_ABI void addDataAccessProfData( |
| 136 | std::unique_ptr<memprof::DataAccessProfData> DataAccessProfile); |
| 137 | |
| 138 | /// Merge existing function counts from the given writer. |
| 139 | LLVM_ABI void mergeRecordsFromWriter(InstrProfWriter &&IPW, |
| 140 | function_ref<void(Error)> Warn); |
| 141 | |
| 142 | /// Write the profile to \c OS |
| 143 | LLVM_ABI Error write(raw_fd_ostream &OS); |
| 144 | |
| 145 | /// Write the profile to a string output stream \c OS |
| 146 | LLVM_ABI Error write(raw_string_ostream &OS); |
| 147 | |
| 148 | /// Write the profile in text format to \c OS |
| 149 | LLVM_ABI Error writeText(raw_fd_ostream &OS); |
| 150 | |
| 151 | /// Write temporal profile trace data to the header in text format to \c OS |
| 152 | LLVM_ABI void writeTextTemporalProfTraceData(raw_fd_ostream &OS, |
| 153 | InstrProfSymtab &Symtab); |
| 154 | |
| 155 | LLVM_ABI Error validateRecord(const InstrProfRecord &Func); |
| 156 | |
| 157 | /// Write \c Record in text format to \c OS |
| 158 | LLVM_ABI static void writeRecordInText(StringRef Name, uint64_t Hash, |
| 159 | const InstrProfRecord &Counters, |
| 160 | InstrProfSymtab &Symtab, |
| 161 | raw_fd_ostream &OS); |
| 162 | |
| 163 | /// Write the profile, returning the raw data. For testing. |
| 164 | LLVM_ABI std::unique_ptr<MemoryBuffer> writeBuffer(); |
| 165 | |
| 166 | /// Update the attributes of the current profile from the attributes |
| 167 | /// specified. An error is returned if IR and FE profiles are mixed. |
| 168 | Error mergeProfileKind(const InstrProfKind Other) { |
| 169 | // If the kind is unset, this is the first profile we are merging so just |
| 170 | // set it to the given type. |
| 171 | if (ProfileKind == InstrProfKind::Unknown) { |
| 172 | ProfileKind = Other; |
| 173 | return Error::success(); |
| 174 | } |
| 175 | |
| 176 | // Returns true if merging is should fail assuming A and B are incompatible. |
| 177 | auto testIncompatible = [&](InstrProfKind A, InstrProfKind B) { |
| 178 | return (static_cast<bool>(ProfileKind & A) && |
| 179 | static_cast<bool>(Other & B)) || |
| 180 | (static_cast<bool>(ProfileKind & B) && |
| 181 | static_cast<bool>(Other & A)); |
| 182 | }; |
| 183 | |
| 184 | // Check if the profiles are in-compatible. Clang frontend profiles can't be |
| 185 | // merged with other profile types. |
| 186 | if (static_cast<bool>( |
| 187 | (ProfileKind & InstrProfKind::FrontendInstrumentation) ^ |
| 188 | (Other & InstrProfKind::FrontendInstrumentation))) { |
| 189 | return make_error<InstrProfError>(Args: instrprof_error::unsupported_version); |
| 190 | } |
| 191 | if (testIncompatible(InstrProfKind::FunctionEntryOnly, |
| 192 | InstrProfKind::FunctionEntryInstrumentation) || |
| 193 | testIncompatible(InstrProfKind::FunctionEntryOnly, |
| 194 | InstrProfKind::LoopEntriesInstrumentation)) { |
| 195 | return make_error<InstrProfError>( |
| 196 | Args: instrprof_error::unsupported_version, |
| 197 | Args: "cannot merge FunctionEntryOnly profiles and BB profiles together" ); |
| 198 | } |
| 199 | |
| 200 | // Now we update the profile type with the bits that are set. |
| 201 | ProfileKind |= Other; |
| 202 | return Error::success(); |
| 203 | } |
| 204 | |
| 205 | InstrProfKind getProfileKind() const { return ProfileKind; } |
| 206 | |
| 207 | bool hasSingleByteCoverage() const { |
| 208 | return static_cast<bool>(ProfileKind & InstrProfKind::SingleByteCoverage); |
| 209 | } |
| 210 | |
| 211 | // Internal interfaces for testing purpose only. |
| 212 | LLVM_ABI void setValueProfDataEndianness(llvm::endianness Endianness); |
| 213 | LLVM_ABI void setOutputSparse(bool Sparse); |
| 214 | void setMemProfVersionRequested(memprof::IndexedVersion Version) { |
| 215 | MemProfVersionRequested = Version; |
| 216 | } |
| 217 | void setMemProfFullSchema(bool Full) { MemProfFullSchema = Full; } |
| 218 | // Compute the overlap b/w this object and Other. Program level result is |
| 219 | // stored in Overlap and function level result is stored in FuncLevelOverlap. |
| 220 | LLVM_ABI void overlapRecord(NamedInstrProfRecord &&Other, |
| 221 | OverlapStats &Overlap, |
| 222 | OverlapStats &FuncLevelOverlap, |
| 223 | const OverlapFuncFilters &FuncFilter); |
| 224 | |
| 225 | private: |
| 226 | void addRecord(StringRef Name, uint64_t Hash, InstrProfRecord &&I, |
| 227 | uint64_t Weight, function_ref<void(Error)> Warn); |
| 228 | bool shouldEncodeData(const ProfilingData &PD); |
| 229 | |
| 230 | /// Add a memprof record for a function identified by its \p Id. |
| 231 | void addMemProfRecord(const GlobalValue::GUID Id, |
| 232 | const memprof::IndexedMemProfRecord &Record); |
| 233 | |
| 234 | /// Add a memprof frame identified by the hash of the contents of the frame in |
| 235 | /// \p FrameId. |
| 236 | bool addMemProfFrame(const memprof::FrameId, const memprof::Frame &F, |
| 237 | function_ref<void(Error)> Warn); |
| 238 | |
| 239 | /// Add a call stack identified by the hash of the contents of the call stack |
| 240 | /// in \p CallStack. |
| 241 | bool addMemProfCallStack(const memprof::CallStackId CSId, |
| 242 | const llvm::SmallVector<memprof::FrameId> &CallStack, |
| 243 | function_ref<void(Error)> Warn); |
| 244 | |
| 245 | Error writeImpl(ProfOStream &OS); |
| 246 | |
| 247 | // Writes known header fields and reserves space for fields whose value are |
| 248 | // known only after payloads are written. Returns the start byte offset for |
| 249 | // back patching. |
| 250 | uint64_t (const IndexedInstrProf::Header &, |
| 251 | const bool WritePrevVersion, ProfOStream &OS); |
| 252 | |
| 253 | // Writes binary IDs. |
| 254 | Error writeBinaryIds(ProfOStream &OS); |
| 255 | |
| 256 | // Writes compressed vtable names to profiles. |
| 257 | Error writeVTableNames(ProfOStream &OS); |
| 258 | }; |
| 259 | |
| 260 | } // end namespace llvm |
| 261 | |
| 262 | #endif // LLVM_PROFILEDATA_INSTRPROFWRITER_H |
| 263 | |