diff options
author | vitalyisaev <vitalyisaev@yandex-team.com> | 2023-06-29 10:00:50 +0300 |
---|---|---|
committer | vitalyisaev <vitalyisaev@yandex-team.com> | 2023-06-29 10:00:50 +0300 |
commit | 6ffe9e53658409f212834330e13564e4952558f6 (patch) | |
tree | 85b1e00183517648b228aafa7c8fb07f5276f419 /contrib/libs/llvm14/lib/Bitcode/Reader | |
parent | 726057070f9c5a91fc10fde0d5024913d10f1ab9 (diff) | |
download | ydb-6ffe9e53658409f212834330e13564e4952558f6.tar.gz |
YQ Connector: support managed ClickHouse
Со стороны dqrun можно обратиться к инстансу коннектора, который работает на streaming стенде, и извлечь данные из облачного CH.
Diffstat (limited to 'contrib/libs/llvm14/lib/Bitcode/Reader')
-rw-r--r-- | contrib/libs/llvm14/lib/Bitcode/Reader/BitReader.cpp | 133 | ||||
-rw-r--r-- | contrib/libs/llvm14/lib/Bitcode/Reader/BitcodeAnalyzer.cpp | 985 | ||||
-rw-r--r-- | contrib/libs/llvm14/lib/Bitcode/Reader/BitcodeReader.cpp | 7144 | ||||
-rw-r--r-- | contrib/libs/llvm14/lib/Bitcode/Reader/MetadataLoader.cpp | 2379 | ||||
-rw-r--r-- | contrib/libs/llvm14/lib/Bitcode/Reader/MetadataLoader.h | 83 | ||||
-rw-r--r-- | contrib/libs/llvm14/lib/Bitcode/Reader/ValueList.cpp | 216 | ||||
-rw-r--r-- | contrib/libs/llvm14/lib/Bitcode/Reader/ValueList.h | 96 | ||||
-rw-r--r-- | contrib/libs/llvm14/lib/Bitcode/Reader/ya.make | 33 |
8 files changed, 11069 insertions, 0 deletions
diff --git a/contrib/libs/llvm14/lib/Bitcode/Reader/BitReader.cpp b/contrib/libs/llvm14/lib/Bitcode/Reader/BitReader.cpp new file mode 100644 index 0000000000..5ac893aef1 --- /dev/null +++ b/contrib/libs/llvm14/lib/Bitcode/Reader/BitReader.cpp @@ -0,0 +1,133 @@ +//===-- BitReader.cpp -----------------------------------------------------===// +// +// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. +// See https://llvm.org/LICENSE.txt for license information. +// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception +// +//===----------------------------------------------------------------------===// + +#include "llvm-c/BitReader.h" +#include "llvm-c/Core.h" +#include "llvm/Bitcode/BitcodeReader.h" +#include "llvm/IR/LLVMContext.h" +#include "llvm/IR/Module.h" +#include "llvm/Support/MemoryBuffer.h" +#include "llvm/Support/raw_ostream.h" +#include <cstring> +#include <string> + +using namespace llvm; + +/* Builds a module from the bitcode in the specified memory buffer, returning a + reference to the module via the OutModule parameter. Returns 0 on success. + Optionally returns a human-readable error message via OutMessage. */ +LLVMBool LLVMParseBitcode(LLVMMemoryBufferRef MemBuf, LLVMModuleRef *OutModule, + char **OutMessage) { + return LLVMParseBitcodeInContext(LLVMGetGlobalContext(), MemBuf, OutModule, + OutMessage); +} + +LLVMBool LLVMParseBitcode2(LLVMMemoryBufferRef MemBuf, + LLVMModuleRef *OutModule) { + return LLVMParseBitcodeInContext2(LLVMGetGlobalContext(), MemBuf, OutModule); +} + +LLVMBool LLVMParseBitcodeInContext(LLVMContextRef ContextRef, + LLVMMemoryBufferRef MemBuf, + LLVMModuleRef *OutModule, + char **OutMessage) { + MemoryBufferRef Buf = unwrap(MemBuf)->getMemBufferRef(); + LLVMContext &Ctx = *unwrap(ContextRef); + + Expected<std::unique_ptr<Module>> ModuleOrErr = parseBitcodeFile(Buf, Ctx); + if (Error Err = ModuleOrErr.takeError()) { + std::string Message; + handleAllErrors(std::move(Err), [&](ErrorInfoBase &EIB) { + Message = EIB.message(); + }); + if (OutMessage) + *OutMessage = strdup(Message.c_str()); + *OutModule = wrap((Module *)nullptr); + return 1; + } + + *OutModule = wrap(ModuleOrErr.get().release()); + return 0; +} + +LLVMBool LLVMParseBitcodeInContext2(LLVMContextRef ContextRef, + LLVMMemoryBufferRef MemBuf, + LLVMModuleRef *OutModule) { + MemoryBufferRef Buf = unwrap(MemBuf)->getMemBufferRef(); + LLVMContext &Ctx = *unwrap(ContextRef); + + ErrorOr<std::unique_ptr<Module>> ModuleOrErr = + expectedToErrorOrAndEmitErrors(Ctx, parseBitcodeFile(Buf, Ctx)); + if (ModuleOrErr.getError()) { + *OutModule = wrap((Module *)nullptr); + return 1; + } + + *OutModule = wrap(ModuleOrErr.get().release()); + return 0; +} + +/* Reads a module from the specified path, returning via the OutModule parameter + a module provider which performs lazy deserialization. Returns 0 on success. + Optionally returns a human-readable error message via OutMessage. */ +LLVMBool LLVMGetBitcodeModuleInContext(LLVMContextRef ContextRef, + LLVMMemoryBufferRef MemBuf, + LLVMModuleRef *OutM, char **OutMessage) { + LLVMContext &Ctx = *unwrap(ContextRef); + std::unique_ptr<MemoryBuffer> Owner(unwrap(MemBuf)); + Expected<std::unique_ptr<Module>> ModuleOrErr = + getOwningLazyBitcodeModule(std::move(Owner), Ctx); + // Release the buffer if we didn't take ownership of it since we never owned + // it anyway. + (void)Owner.release(); + + if (Error Err = ModuleOrErr.takeError()) { + std::string Message; + handleAllErrors(std::move(Err), [&](ErrorInfoBase &EIB) { + Message = EIB.message(); + }); + if (OutMessage) + *OutMessage = strdup(Message.c_str()); + *OutM = wrap((Module *)nullptr); + return 1; + } + + *OutM = wrap(ModuleOrErr.get().release()); + + return 0; +} + +LLVMBool LLVMGetBitcodeModuleInContext2(LLVMContextRef ContextRef, + LLVMMemoryBufferRef MemBuf, + LLVMModuleRef *OutM) { + LLVMContext &Ctx = *unwrap(ContextRef); + std::unique_ptr<MemoryBuffer> Owner(unwrap(MemBuf)); + + ErrorOr<std::unique_ptr<Module>> ModuleOrErr = expectedToErrorOrAndEmitErrors( + Ctx, getOwningLazyBitcodeModule(std::move(Owner), Ctx)); + Owner.release(); + + if (ModuleOrErr.getError()) { + *OutM = wrap((Module *)nullptr); + return 1; + } + + *OutM = wrap(ModuleOrErr.get().release()); + return 0; +} + +LLVMBool LLVMGetBitcodeModule(LLVMMemoryBufferRef MemBuf, LLVMModuleRef *OutM, + char **OutMessage) { + return LLVMGetBitcodeModuleInContext(LLVMGetGlobalContext(), MemBuf, OutM, + OutMessage); +} + +LLVMBool LLVMGetBitcodeModule2(LLVMMemoryBufferRef MemBuf, + LLVMModuleRef *OutM) { + return LLVMGetBitcodeModuleInContext2(LLVMGetGlobalContext(), MemBuf, OutM); +} diff --git a/contrib/libs/llvm14/lib/Bitcode/Reader/BitcodeAnalyzer.cpp b/contrib/libs/llvm14/lib/Bitcode/Reader/BitcodeAnalyzer.cpp new file mode 100644 index 0000000000..ffef352999 --- /dev/null +++ b/contrib/libs/llvm14/lib/Bitcode/Reader/BitcodeAnalyzer.cpp @@ -0,0 +1,985 @@ +//===- BitcodeAnalyzer.cpp - Internal BitcodeAnalyzer implementation ------===// +// +// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. +// See https://llvm.org/LICENSE.txt for license information. +// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception +// +//===----------------------------------------------------------------------===// + +#include "llvm/Bitcode/BitcodeAnalyzer.h" +#include "llvm/Bitcode/BitcodeReader.h" +#include "llvm/Bitcode/LLVMBitCodes.h" +#include "llvm/Bitstream/BitCodes.h" +#include "llvm/Bitstream/BitstreamReader.h" +#include "llvm/Support/Format.h" +#include "llvm/Support/SHA1.h" + +using namespace llvm; + +static Error reportError(StringRef Message) { + return createStringError(std::errc::illegal_byte_sequence, Message.data()); +} + +/// Return a symbolic block name if known, otherwise return null. +static Optional<const char *> GetBlockName(unsigned BlockID, + const BitstreamBlockInfo &BlockInfo, + CurStreamTypeType CurStreamType) { + // Standard blocks for all bitcode files. + if (BlockID < bitc::FIRST_APPLICATION_BLOCKID) { + if (BlockID == bitc::BLOCKINFO_BLOCK_ID) + return "BLOCKINFO_BLOCK"; + return None; + } + + // Check to see if we have a blockinfo record for this block, with a name. + if (const BitstreamBlockInfo::BlockInfo *Info = + BlockInfo.getBlockInfo(BlockID)) { + if (!Info->Name.empty()) + return Info->Name.c_str(); + } + + if (CurStreamType != LLVMIRBitstream) + return None; + + switch (BlockID) { + default: + return None; + case bitc::OPERAND_BUNDLE_TAGS_BLOCK_ID: + return "OPERAND_BUNDLE_TAGS_BLOCK"; + case bitc::MODULE_BLOCK_ID: + return "MODULE_BLOCK"; + case bitc::PARAMATTR_BLOCK_ID: + return "PARAMATTR_BLOCK"; + case bitc::PARAMATTR_GROUP_BLOCK_ID: + return "PARAMATTR_GROUP_BLOCK_ID"; + case bitc::TYPE_BLOCK_ID_NEW: + return "TYPE_BLOCK_ID"; + case bitc::CONSTANTS_BLOCK_ID: + return "CONSTANTS_BLOCK"; + case bitc::FUNCTION_BLOCK_ID: + return "FUNCTION_BLOCK"; + case bitc::IDENTIFICATION_BLOCK_ID: + return "IDENTIFICATION_BLOCK_ID"; + case bitc::VALUE_SYMTAB_BLOCK_ID: + return "VALUE_SYMTAB"; + case bitc::METADATA_BLOCK_ID: + return "METADATA_BLOCK"; + case bitc::METADATA_KIND_BLOCK_ID: + return "METADATA_KIND_BLOCK"; + case bitc::METADATA_ATTACHMENT_ID: + return "METADATA_ATTACHMENT_BLOCK"; + case bitc::USELIST_BLOCK_ID: + return "USELIST_BLOCK_ID"; + case bitc::GLOBALVAL_SUMMARY_BLOCK_ID: + return "GLOBALVAL_SUMMARY_BLOCK"; + case bitc::FULL_LTO_GLOBALVAL_SUMMARY_BLOCK_ID: + return "FULL_LTO_GLOBALVAL_SUMMARY_BLOCK"; + case bitc::MODULE_STRTAB_BLOCK_ID: + return "MODULE_STRTAB_BLOCK"; + case bitc::STRTAB_BLOCK_ID: + return "STRTAB_BLOCK"; + case bitc::SYMTAB_BLOCK_ID: + return "SYMTAB_BLOCK"; + } +} + +/// Return a symbolic code name if known, otherwise return null. +static Optional<const char *> GetCodeName(unsigned CodeID, unsigned BlockID, + const BitstreamBlockInfo &BlockInfo, + CurStreamTypeType CurStreamType) { + // Standard blocks for all bitcode files. + if (BlockID < bitc::FIRST_APPLICATION_BLOCKID) { + if (BlockID == bitc::BLOCKINFO_BLOCK_ID) { + switch (CodeID) { + default: + return None; + case bitc::BLOCKINFO_CODE_SETBID: + return "SETBID"; + case bitc::BLOCKINFO_CODE_BLOCKNAME: + return "BLOCKNAME"; + case bitc::BLOCKINFO_CODE_SETRECORDNAME: + return "SETRECORDNAME"; + } + } + return None; + } + + // Check to see if we have a blockinfo record for this record, with a name. + if (const BitstreamBlockInfo::BlockInfo *Info = + BlockInfo.getBlockInfo(BlockID)) { + for (const std::pair<unsigned, std::string> &RN : Info->RecordNames) + if (RN.first == CodeID) + return RN.second.c_str(); + } + + if (CurStreamType != LLVMIRBitstream) + return None; + +#define STRINGIFY_CODE(PREFIX, CODE) \ + case bitc::PREFIX##_##CODE: \ + return #CODE; + switch (BlockID) { + default: + return None; + case bitc::MODULE_BLOCK_ID: + switch (CodeID) { + default: + return None; + STRINGIFY_CODE(MODULE_CODE, VERSION) + STRINGIFY_CODE(MODULE_CODE, TRIPLE) + STRINGIFY_CODE(MODULE_CODE, DATALAYOUT) + STRINGIFY_CODE(MODULE_CODE, ASM) + STRINGIFY_CODE(MODULE_CODE, SECTIONNAME) + STRINGIFY_CODE(MODULE_CODE, DEPLIB) // Deprecated, present in old bitcode + STRINGIFY_CODE(MODULE_CODE, GLOBALVAR) + STRINGIFY_CODE(MODULE_CODE, FUNCTION) + STRINGIFY_CODE(MODULE_CODE, ALIAS) + STRINGIFY_CODE(MODULE_CODE, GCNAME) + STRINGIFY_CODE(MODULE_CODE, COMDAT) + STRINGIFY_CODE(MODULE_CODE, VSTOFFSET) + STRINGIFY_CODE(MODULE_CODE, METADATA_VALUES_UNUSED) + STRINGIFY_CODE(MODULE_CODE, SOURCE_FILENAME) + STRINGIFY_CODE(MODULE_CODE, HASH) + } + case bitc::IDENTIFICATION_BLOCK_ID: + switch (CodeID) { + default: + return None; + STRINGIFY_CODE(IDENTIFICATION_CODE, STRING) + STRINGIFY_CODE(IDENTIFICATION_CODE, EPOCH) + } + case bitc::PARAMATTR_BLOCK_ID: + switch (CodeID) { + default: + return None; + // FIXME: Should these be different? + case bitc::PARAMATTR_CODE_ENTRY_OLD: + return "ENTRY"; + case bitc::PARAMATTR_CODE_ENTRY: + return "ENTRY"; + } + case bitc::PARAMATTR_GROUP_BLOCK_ID: + switch (CodeID) { + default: + return None; + case bitc::PARAMATTR_GRP_CODE_ENTRY: + return "ENTRY"; + } + case bitc::TYPE_BLOCK_ID_NEW: + switch (CodeID) { + default: + return None; + STRINGIFY_CODE(TYPE_CODE, NUMENTRY) + STRINGIFY_CODE(TYPE_CODE, VOID) + STRINGIFY_CODE(TYPE_CODE, FLOAT) + STRINGIFY_CODE(TYPE_CODE, DOUBLE) + STRINGIFY_CODE(TYPE_CODE, LABEL) + STRINGIFY_CODE(TYPE_CODE, OPAQUE) + STRINGIFY_CODE(TYPE_CODE, INTEGER) + STRINGIFY_CODE(TYPE_CODE, POINTER) + STRINGIFY_CODE(TYPE_CODE, HALF) + STRINGIFY_CODE(TYPE_CODE, ARRAY) + STRINGIFY_CODE(TYPE_CODE, VECTOR) + STRINGIFY_CODE(TYPE_CODE, X86_FP80) + STRINGIFY_CODE(TYPE_CODE, FP128) + STRINGIFY_CODE(TYPE_CODE, PPC_FP128) + STRINGIFY_CODE(TYPE_CODE, METADATA) + STRINGIFY_CODE(TYPE_CODE, X86_MMX) + STRINGIFY_CODE(TYPE_CODE, STRUCT_ANON) + STRINGIFY_CODE(TYPE_CODE, STRUCT_NAME) + STRINGIFY_CODE(TYPE_CODE, STRUCT_NAMED) + STRINGIFY_CODE(TYPE_CODE, FUNCTION) + STRINGIFY_CODE(TYPE_CODE, TOKEN) + STRINGIFY_CODE(TYPE_CODE, BFLOAT) + } + + case bitc::CONSTANTS_BLOCK_ID: + switch (CodeID) { + default: + return None; + STRINGIFY_CODE(CST_CODE, SETTYPE) + STRINGIFY_CODE(CST_CODE, NULL) + STRINGIFY_CODE(CST_CODE, UNDEF) + STRINGIFY_CODE(CST_CODE, INTEGER) + STRINGIFY_CODE(CST_CODE, WIDE_INTEGER) + STRINGIFY_CODE(CST_CODE, FLOAT) + STRINGIFY_CODE(CST_CODE, AGGREGATE) + STRINGIFY_CODE(CST_CODE, STRING) + STRINGIFY_CODE(CST_CODE, CSTRING) + STRINGIFY_CODE(CST_CODE, CE_BINOP) + STRINGIFY_CODE(CST_CODE, CE_CAST) + STRINGIFY_CODE(CST_CODE, CE_GEP) + STRINGIFY_CODE(CST_CODE, CE_INBOUNDS_GEP) + STRINGIFY_CODE(CST_CODE, CE_SELECT) + STRINGIFY_CODE(CST_CODE, CE_EXTRACTELT) + STRINGIFY_CODE(CST_CODE, CE_INSERTELT) + STRINGIFY_CODE(CST_CODE, CE_SHUFFLEVEC) + STRINGIFY_CODE(CST_CODE, CE_CMP) + STRINGIFY_CODE(CST_CODE, INLINEASM) + STRINGIFY_CODE(CST_CODE, CE_SHUFVEC_EX) + STRINGIFY_CODE(CST_CODE, CE_UNOP) + STRINGIFY_CODE(CST_CODE, DSO_LOCAL_EQUIVALENT) + STRINGIFY_CODE(CST_CODE, NO_CFI_VALUE) + case bitc::CST_CODE_BLOCKADDRESS: + return "CST_CODE_BLOCKADDRESS"; + STRINGIFY_CODE(CST_CODE, DATA) + } + case bitc::FUNCTION_BLOCK_ID: + switch (CodeID) { + default: + return None; + STRINGIFY_CODE(FUNC_CODE, DECLAREBLOCKS) + STRINGIFY_CODE(FUNC_CODE, INST_BINOP) + STRINGIFY_CODE(FUNC_CODE, INST_CAST) + STRINGIFY_CODE(FUNC_CODE, INST_GEP_OLD) + STRINGIFY_CODE(FUNC_CODE, INST_INBOUNDS_GEP_OLD) + STRINGIFY_CODE(FUNC_CODE, INST_SELECT) + STRINGIFY_CODE(FUNC_CODE, INST_EXTRACTELT) + STRINGIFY_CODE(FUNC_CODE, INST_INSERTELT) + STRINGIFY_CODE(FUNC_CODE, INST_SHUFFLEVEC) + STRINGIFY_CODE(FUNC_CODE, INST_CMP) + STRINGIFY_CODE(FUNC_CODE, INST_RET) + STRINGIFY_CODE(FUNC_CODE, INST_BR) + STRINGIFY_CODE(FUNC_CODE, INST_SWITCH) + STRINGIFY_CODE(FUNC_CODE, INST_INVOKE) + STRINGIFY_CODE(FUNC_CODE, INST_UNOP) + STRINGIFY_CODE(FUNC_CODE, INST_UNREACHABLE) + STRINGIFY_CODE(FUNC_CODE, INST_CLEANUPRET) + STRINGIFY_CODE(FUNC_CODE, INST_CATCHRET) + STRINGIFY_CODE(FUNC_CODE, INST_CATCHPAD) + STRINGIFY_CODE(FUNC_CODE, INST_PHI) + STRINGIFY_CODE(FUNC_CODE, INST_ALLOCA) + STRINGIFY_CODE(FUNC_CODE, INST_LOAD) + STRINGIFY_CODE(FUNC_CODE, INST_VAARG) + STRINGIFY_CODE(FUNC_CODE, INST_STORE) + STRINGIFY_CODE(FUNC_CODE, INST_EXTRACTVAL) + STRINGIFY_CODE(FUNC_CODE, INST_INSERTVAL) + STRINGIFY_CODE(FUNC_CODE, INST_CMP2) + STRINGIFY_CODE(FUNC_CODE, INST_VSELECT) + STRINGIFY_CODE(FUNC_CODE, DEBUG_LOC_AGAIN) + STRINGIFY_CODE(FUNC_CODE, INST_CALL) + STRINGIFY_CODE(FUNC_CODE, DEBUG_LOC) + STRINGIFY_CODE(FUNC_CODE, INST_GEP) + STRINGIFY_CODE(FUNC_CODE, OPERAND_BUNDLE) + STRINGIFY_CODE(FUNC_CODE, INST_FENCE) + STRINGIFY_CODE(FUNC_CODE, INST_ATOMICRMW) + STRINGIFY_CODE(FUNC_CODE, INST_LOADATOMIC) + STRINGIFY_CODE(FUNC_CODE, INST_STOREATOMIC) + STRINGIFY_CODE(FUNC_CODE, INST_CMPXCHG) + STRINGIFY_CODE(FUNC_CODE, INST_CALLBR) + } + case bitc::VALUE_SYMTAB_BLOCK_ID: + switch (CodeID) { + default: + return None; + STRINGIFY_CODE(VST_CODE, ENTRY) + STRINGIFY_CODE(VST_CODE, BBENTRY) + STRINGIFY_CODE(VST_CODE, FNENTRY) + STRINGIFY_CODE(VST_CODE, COMBINED_ENTRY) + } + case bitc::MODULE_STRTAB_BLOCK_ID: + switch (CodeID) { + default: + return None; + STRINGIFY_CODE(MST_CODE, ENTRY) + STRINGIFY_CODE(MST_CODE, HASH) + } + case bitc::GLOBALVAL_SUMMARY_BLOCK_ID: + case bitc::FULL_LTO_GLOBALVAL_SUMMARY_BLOCK_ID: + switch (CodeID) { + default: + return None; + STRINGIFY_CODE(FS, PERMODULE) + STRINGIFY_CODE(FS, PERMODULE_PROFILE) + STRINGIFY_CODE(FS, PERMODULE_RELBF) + STRINGIFY_CODE(FS, PERMODULE_GLOBALVAR_INIT_REFS) + STRINGIFY_CODE(FS, PERMODULE_VTABLE_GLOBALVAR_INIT_REFS) + STRINGIFY_CODE(FS, COMBINED) + STRINGIFY_CODE(FS, COMBINED_PROFILE) + STRINGIFY_CODE(FS, COMBINED_GLOBALVAR_INIT_REFS) + STRINGIFY_CODE(FS, ALIAS) + STRINGIFY_CODE(FS, COMBINED_ALIAS) + STRINGIFY_CODE(FS, COMBINED_ORIGINAL_NAME) + STRINGIFY_CODE(FS, VERSION) + STRINGIFY_CODE(FS, FLAGS) + STRINGIFY_CODE(FS, TYPE_TESTS) + STRINGIFY_CODE(FS, TYPE_TEST_ASSUME_VCALLS) + STRINGIFY_CODE(FS, TYPE_CHECKED_LOAD_VCALLS) + STRINGIFY_CODE(FS, TYPE_TEST_ASSUME_CONST_VCALL) + STRINGIFY_CODE(FS, TYPE_CHECKED_LOAD_CONST_VCALL) + STRINGIFY_CODE(FS, VALUE_GUID) + STRINGIFY_CODE(FS, CFI_FUNCTION_DEFS) + STRINGIFY_CODE(FS, CFI_FUNCTION_DECLS) + STRINGIFY_CODE(FS, TYPE_ID) + STRINGIFY_CODE(FS, TYPE_ID_METADATA) + STRINGIFY_CODE(FS, BLOCK_COUNT) + STRINGIFY_CODE(FS, PARAM_ACCESS) + } + case bitc::METADATA_ATTACHMENT_ID: + switch (CodeID) { + default: + return None; + STRINGIFY_CODE(METADATA, ATTACHMENT) + } + case bitc::METADATA_BLOCK_ID: + switch (CodeID) { + default: + return None; + STRINGIFY_CODE(METADATA, STRING_OLD) + STRINGIFY_CODE(METADATA, VALUE) + STRINGIFY_CODE(METADATA, NODE) + STRINGIFY_CODE(METADATA, NAME) + STRINGIFY_CODE(METADATA, DISTINCT_NODE) + STRINGIFY_CODE(METADATA, KIND) // Older bitcode has it in a MODULE_BLOCK + STRINGIFY_CODE(METADATA, LOCATION) + STRINGIFY_CODE(METADATA, OLD_NODE) + STRINGIFY_CODE(METADATA, OLD_FN_NODE) + STRINGIFY_CODE(METADATA, NAMED_NODE) + STRINGIFY_CODE(METADATA, GENERIC_DEBUG) + STRINGIFY_CODE(METADATA, SUBRANGE) + STRINGIFY_CODE(METADATA, ENUMERATOR) + STRINGIFY_CODE(METADATA, BASIC_TYPE) + STRINGIFY_CODE(METADATA, FILE) + STRINGIFY_CODE(METADATA, DERIVED_TYPE) + STRINGIFY_CODE(METADATA, COMPOSITE_TYPE) + STRINGIFY_CODE(METADATA, SUBROUTINE_TYPE) + STRINGIFY_CODE(METADATA, COMPILE_UNIT) + STRINGIFY_CODE(METADATA, SUBPROGRAM) + STRINGIFY_CODE(METADATA, LEXICAL_BLOCK) + STRINGIFY_CODE(METADATA, LEXICAL_BLOCK_FILE) + STRINGIFY_CODE(METADATA, NAMESPACE) + STRINGIFY_CODE(METADATA, TEMPLATE_TYPE) + STRINGIFY_CODE(METADATA, TEMPLATE_VALUE) + STRINGIFY_CODE(METADATA, GLOBAL_VAR) + STRINGIFY_CODE(METADATA, LOCAL_VAR) + STRINGIFY_CODE(METADATA, EXPRESSION) + STRINGIFY_CODE(METADATA, OBJC_PROPERTY) + STRINGIFY_CODE(METADATA, IMPORTED_ENTITY) + STRINGIFY_CODE(METADATA, MODULE) + STRINGIFY_CODE(METADATA, MACRO) + STRINGIFY_CODE(METADATA, MACRO_FILE) + STRINGIFY_CODE(METADATA, STRINGS) + STRINGIFY_CODE(METADATA, GLOBAL_DECL_ATTACHMENT) + STRINGIFY_CODE(METADATA, GLOBAL_VAR_EXPR) + STRINGIFY_CODE(METADATA, INDEX_OFFSET) + STRINGIFY_CODE(METADATA, INDEX) + STRINGIFY_CODE(METADATA, ARG_LIST) + } + case bitc::METADATA_KIND_BLOCK_ID: + switch (CodeID) { + default: + return None; + STRINGIFY_CODE(METADATA, KIND) + } + case bitc::USELIST_BLOCK_ID: + switch (CodeID) { + default: + return None; + case bitc::USELIST_CODE_DEFAULT: + return "USELIST_CODE_DEFAULT"; + case bitc::USELIST_CODE_BB: + return "USELIST_CODE_BB"; + } + + case bitc::OPERAND_BUNDLE_TAGS_BLOCK_ID: + switch (CodeID) { + default: + return None; + case bitc::OPERAND_BUNDLE_TAG: + return "OPERAND_BUNDLE_TAG"; + } + case bitc::STRTAB_BLOCK_ID: + switch (CodeID) { + default: + return None; + case bitc::STRTAB_BLOB: + return "BLOB"; + } + case bitc::SYMTAB_BLOCK_ID: + switch (CodeID) { + default: + return None; + case bitc::SYMTAB_BLOB: + return "BLOB"; + } + } +#undef STRINGIFY_CODE +} + +static void printSize(raw_ostream &OS, double Bits) { + OS << format("%.2f/%.2fB/%luW", Bits, Bits / 8, (unsigned long)(Bits / 32)); +} +static void printSize(raw_ostream &OS, uint64_t Bits) { + OS << format("%lub/%.2fB/%luW", (unsigned long)Bits, (double)Bits / 8, + (unsigned long)(Bits / 32)); +} + +static Expected<CurStreamTypeType> ReadSignature(BitstreamCursor &Stream) { + auto tryRead = [&Stream](char &Dest, size_t size) -> Error { + if (Expected<SimpleBitstreamCursor::word_t> MaybeWord = Stream.Read(size)) + Dest = MaybeWord.get(); + else + return MaybeWord.takeError(); + return Error::success(); + }; + + char Signature[6]; + if (Error Err = tryRead(Signature[0], 8)) + return std::move(Err); + if (Error Err = tryRead(Signature[1], 8)) + return std::move(Err); + + // Autodetect the file contents, if it is one we know. + if (Signature[0] == 'C' && Signature[1] == 'P') { + if (Error Err = tryRead(Signature[2], 8)) + return std::move(Err); + if (Error Err = tryRead(Signature[3], 8)) + return std::move(Err); + if (Signature[2] == 'C' && Signature[3] == 'H') + return ClangSerializedASTBitstream; + } else if (Signature[0] == 'D' && Signature[1] == 'I') { + if (Error Err = tryRead(Signature[2], 8)) + return std::move(Err); + if (Error Err = tryRead(Signature[3], 8)) + return std::move(Err); + if (Signature[2] == 'A' && Signature[3] == 'G') + return ClangSerializedDiagnosticsBitstream; + } else if (Signature[0] == 'R' && Signature[1] == 'M') { + if (Error Err = tryRead(Signature[2], 8)) + return std::move(Err); + if (Error Err = tryRead(Signature[3], 8)) + return std::move(Err); + if (Signature[2] == 'R' && Signature[3] == 'K') + return LLVMBitstreamRemarks; + } else { + if (Error Err = tryRead(Signature[2], 4)) + return std::move(Err); + if (Error Err = tryRead(Signature[3], 4)) + return std::move(Err); + if (Error Err = tryRead(Signature[4], 4)) + return std::move(Err); + if (Error Err = tryRead(Signature[5], 4)) + return std::move(Err); + if (Signature[0] == 'B' && Signature[1] == 'C' && Signature[2] == 0x0 && + Signature[3] == 0xC && Signature[4] == 0xE && Signature[5] == 0xD) + return LLVMIRBitstream; + } + return UnknownBitstream; +} + +static Expected<CurStreamTypeType> analyzeHeader(Optional<BCDumpOptions> O, + BitstreamCursor &Stream) { + ArrayRef<uint8_t> Bytes = Stream.getBitcodeBytes(); + const unsigned char *BufPtr = (const unsigned char *)Bytes.data(); + const unsigned char *EndBufPtr = BufPtr + Bytes.size(); + + // If we have a wrapper header, parse it and ignore the non-bc file + // contents. The magic number is 0x0B17C0DE stored in little endian. + if (isBitcodeWrapper(BufPtr, EndBufPtr)) { + if (Bytes.size() < BWH_HeaderSize) + return reportError("Invalid bitcode wrapper header"); + + if (O) { + unsigned Magic = support::endian::read32le(&BufPtr[BWH_MagicField]); + unsigned Version = support::endian::read32le(&BufPtr[BWH_VersionField]); + unsigned Offset = support::endian::read32le(&BufPtr[BWH_OffsetField]); + unsigned Size = support::endian::read32le(&BufPtr[BWH_SizeField]); + unsigned CPUType = support::endian::read32le(&BufPtr[BWH_CPUTypeField]); + + O->OS << "<BITCODE_WRAPPER_HEADER" + << " Magic=" << format_hex(Magic, 10) + << " Version=" << format_hex(Version, 10) + << " Offset=" << format_hex(Offset, 10) + << " Size=" << format_hex(Size, 10) + << " CPUType=" << format_hex(CPUType, 10) << "/>\n"; + } + + if (SkipBitcodeWrapperHeader(BufPtr, EndBufPtr, true)) + return reportError("Invalid bitcode wrapper header"); + } + + // Use the cursor modified by skipping the wrapper header. + Stream = BitstreamCursor(ArrayRef<uint8_t>(BufPtr, EndBufPtr)); + + return ReadSignature(Stream); +} + +static bool canDecodeBlob(unsigned Code, unsigned BlockID) { + return BlockID == bitc::METADATA_BLOCK_ID && Code == bitc::METADATA_STRINGS; +} + +Error BitcodeAnalyzer::decodeMetadataStringsBlob(StringRef Indent, + ArrayRef<uint64_t> Record, + StringRef Blob, + raw_ostream &OS) { + if (Blob.empty()) + return reportError("Cannot decode empty blob."); + + if (Record.size() != 2) + return reportError( + "Decoding metadata strings blob needs two record entries."); + + unsigned NumStrings = Record[0]; + unsigned StringsOffset = Record[1]; + OS << " num-strings = " << NumStrings << " {\n"; + + StringRef Lengths = Blob.slice(0, StringsOffset); + SimpleBitstreamCursor R(Lengths); + StringRef Strings = Blob.drop_front(StringsOffset); + do { + if (R.AtEndOfStream()) + return reportError("bad length"); + + uint32_t Size; + if (Error E = R.ReadVBR(6).moveInto(Size)) + return E; + if (Strings.size() < Size) + return reportError("truncated chars"); + + OS << Indent << " '"; + OS.write_escaped(Strings.slice(0, Size), /*hex=*/true); + OS << "'\n"; + Strings = Strings.drop_front(Size); + } while (--NumStrings); + + OS << Indent << " }"; + return Error::success(); +} + +BitcodeAnalyzer::BitcodeAnalyzer(StringRef Buffer, + Optional<StringRef> BlockInfoBuffer) + : Stream(Buffer) { + if (BlockInfoBuffer) + BlockInfoStream.emplace(*BlockInfoBuffer); +} + +Error BitcodeAnalyzer::analyze(Optional<BCDumpOptions> O, + Optional<StringRef> CheckHash) { + if (Error E = analyzeHeader(O, Stream).moveInto(CurStreamType)) + return E; + + Stream.setBlockInfo(&BlockInfo); + + // Read block info from BlockInfoStream, if specified. + // The block info must be a top-level block. + if (BlockInfoStream) { + BitstreamCursor BlockInfoCursor(*BlockInfoStream); + if (Error E = analyzeHeader(O, BlockInfoCursor).takeError()) + return E; + + while (!BlockInfoCursor.AtEndOfStream()) { + Expected<unsigned> MaybeCode = BlockInfoCursor.ReadCode(); + if (!MaybeCode) + return MaybeCode.takeError(); + if (MaybeCode.get() != bitc::ENTER_SUBBLOCK) + return reportError("Invalid record at top-level in block info file"); + + Expected<unsigned> MaybeBlockID = BlockInfoCursor.ReadSubBlockID(); + if (!MaybeBlockID) + return MaybeBlockID.takeError(); + if (MaybeBlockID.get() == bitc::BLOCKINFO_BLOCK_ID) { + Optional<BitstreamBlockInfo> NewBlockInfo; + if (Error E = + BlockInfoCursor.ReadBlockInfoBlock(/*ReadBlockInfoNames=*/true) + .moveInto(NewBlockInfo)) + return E; + if (!NewBlockInfo) + return reportError("Malformed BlockInfoBlock in block info file"); + BlockInfo = std::move(*NewBlockInfo); + break; + } + + if (Error Err = BlockInfoCursor.SkipBlock()) + return Err; + } + } + + // Parse the top-level structure. We only allow blocks at the top-level. + while (!Stream.AtEndOfStream()) { + Expected<unsigned> MaybeCode = Stream.ReadCode(); + if (!MaybeCode) + return MaybeCode.takeError(); + if (MaybeCode.get() != bitc::ENTER_SUBBLOCK) + return reportError("Invalid record at top-level"); + + Expected<unsigned> MaybeBlockID = Stream.ReadSubBlockID(); + if (!MaybeBlockID) + return MaybeBlockID.takeError(); + + if (Error E = parseBlock(MaybeBlockID.get(), 0, O, CheckHash)) + return E; + ++NumTopBlocks; + } + + return Error::success(); +} + +void BitcodeAnalyzer::printStats(BCDumpOptions O, + Optional<StringRef> Filename) { + uint64_t BufferSizeBits = Stream.getBitcodeBytes().size() * CHAR_BIT; + // Print a summary of the read file. + O.OS << "Summary "; + if (Filename) + O.OS << "of " << Filename->data() << ":\n"; + O.OS << " Total size: "; + printSize(O.OS, BufferSizeBits); + O.OS << "\n"; + O.OS << " Stream type: "; + switch (CurStreamType) { + case UnknownBitstream: + O.OS << "unknown\n"; + break; + case LLVMIRBitstream: + O.OS << "LLVM IR\n"; + break; + case ClangSerializedASTBitstream: + O.OS << "Clang Serialized AST\n"; + break; + case ClangSerializedDiagnosticsBitstream: + O.OS << "Clang Serialized Diagnostics\n"; + break; + case LLVMBitstreamRemarks: + O.OS << "LLVM Remarks\n"; + break; + } + O.OS << " # Toplevel Blocks: " << NumTopBlocks << "\n"; + O.OS << "\n"; + + // Emit per-block stats. + O.OS << "Per-block Summary:\n"; + for (const auto &Stat : BlockIDStats) { + O.OS << " Block ID #" << Stat.first; + if (Optional<const char *> BlockName = + GetBlockName(Stat.first, BlockInfo, CurStreamType)) + O.OS << " (" << *BlockName << ")"; + O.OS << ":\n"; + + const PerBlockIDStats &Stats = Stat.second; + O.OS << " Num Instances: " << Stats.NumInstances << "\n"; + O.OS << " Total Size: "; + printSize(O.OS, Stats.NumBits); + O.OS << "\n"; + double pct = (Stats.NumBits * 100.0) / BufferSizeBits; + O.OS << " Percent of file: " << format("%2.4f%%", pct) << "\n"; + if (Stats.NumInstances > 1) { + O.OS << " Average Size: "; + printSize(O.OS, Stats.NumBits / (double)Stats.NumInstances); + O.OS << "\n"; + O.OS << " Tot/Avg SubBlocks: " << Stats.NumSubBlocks << "/" + << Stats.NumSubBlocks / (double)Stats.NumInstances << "\n"; + O.OS << " Tot/Avg Abbrevs: " << Stats.NumAbbrevs << "/" + << Stats.NumAbbrevs / (double)Stats.NumInstances << "\n"; + O.OS << " Tot/Avg Records: " << Stats.NumRecords << "/" + << Stats.NumRecords / (double)Stats.NumInstances << "\n"; + } else { + O.OS << " Num SubBlocks: " << Stats.NumSubBlocks << "\n"; + O.OS << " Num Abbrevs: " << Stats.NumAbbrevs << "\n"; + O.OS << " Num Records: " << Stats.NumRecords << "\n"; + } + if (Stats.NumRecords) { + double pct = (Stats.NumAbbreviatedRecords * 100.0) / Stats.NumRecords; + O.OS << " Percent Abbrevs: " << format("%2.4f%%", pct) << "\n"; + } + O.OS << "\n"; + + // Print a histogram of the codes we see. + if (O.Histogram && !Stats.CodeFreq.empty()) { + std::vector<std::pair<unsigned, unsigned>> FreqPairs; // <freq,code> + for (unsigned i = 0, e = Stats.CodeFreq.size(); i != e; ++i) + if (unsigned Freq = Stats.CodeFreq[i].NumInstances) + FreqPairs.push_back(std::make_pair(Freq, i)); + llvm::stable_sort(FreqPairs); + std::reverse(FreqPairs.begin(), FreqPairs.end()); + + O.OS << "\tRecord Histogram:\n"; + O.OS << "\t\t Count # Bits b/Rec % Abv Record Kind\n"; + for (const auto &FreqPair : FreqPairs) { + const PerRecordStats &RecStats = Stats.CodeFreq[FreqPair.second]; + + O.OS << format("\t\t%7d %9lu", RecStats.NumInstances, + (unsigned long)RecStats.TotalBits); + + if (RecStats.NumInstances > 1) + O.OS << format(" %9.1f", + (double)RecStats.TotalBits / RecStats.NumInstances); + else + O.OS << " "; + + if (RecStats.NumAbbrev) + O.OS << format(" %7.2f", (double)RecStats.NumAbbrev / + RecStats.NumInstances * 100); + else + O.OS << " "; + + O.OS << " "; + if (Optional<const char *> CodeName = GetCodeName( + FreqPair.second, Stat.first, BlockInfo, CurStreamType)) + O.OS << *CodeName << "\n"; + else + O.OS << "UnknownCode" << FreqPair.second << "\n"; + } + O.OS << "\n"; + } + } +} + +Error BitcodeAnalyzer::parseBlock(unsigned BlockID, unsigned IndentLevel, + Optional<BCDumpOptions> O, + Optional<StringRef> CheckHash) { + std::string Indent(IndentLevel * 2, ' '); + uint64_t BlockBitStart = Stream.GetCurrentBitNo(); + + // Get the statistics for this BlockID. + PerBlockIDStats &BlockStats = BlockIDStats[BlockID]; + + BlockStats.NumInstances++; + + // BLOCKINFO is a special part of the stream. + bool DumpRecords = O.hasValue(); + if (BlockID == bitc::BLOCKINFO_BLOCK_ID) { + if (O && !O->DumpBlockinfo) + O->OS << Indent << "<BLOCKINFO_BLOCK/>\n"; + Optional<BitstreamBlockInfo> NewBlockInfo; + if (Error E = Stream.ReadBlockInfoBlock(/*ReadBlockInfoNames=*/true) + .moveInto(NewBlockInfo)) + return E; + if (!NewBlockInfo) + return reportError("Malformed BlockInfoBlock"); + BlockInfo = std::move(*NewBlockInfo); + if (Error Err = Stream.JumpToBit(BlockBitStart)) + return Err; + // It's not really interesting to dump the contents of the blockinfo + // block, so only do it if the user explicitly requests it. + DumpRecords = O && O->DumpBlockinfo; + } + + unsigned NumWords = 0; + if (Error Err = Stream.EnterSubBlock(BlockID, &NumWords)) + return Err; + + // Keep it for later, when we see a MODULE_HASH record + uint64_t BlockEntryPos = Stream.getCurrentByteNo(); + + Optional<const char *> BlockName = None; + if (DumpRecords) { + O->OS << Indent << "<"; + if ((BlockName = GetBlockName(BlockID, BlockInfo, CurStreamType))) + O->OS << *BlockName; + else + O->OS << "UnknownBlock" << BlockID; + + if (!O->Symbolic && BlockName) + O->OS << " BlockID=" << BlockID; + + O->OS << " NumWords=" << NumWords + << " BlockCodeSize=" << Stream.getAbbrevIDWidth() << ">\n"; + } + + SmallVector<uint64_t, 64> Record; + + // Keep the offset to the metadata index if seen. + uint64_t MetadataIndexOffset = 0; + + // Read all the records for this block. + while (true) { + if (Stream.AtEndOfStream()) + return reportError("Premature end of bitstream"); + + uint64_t RecordStartBit = Stream.GetCurrentBitNo(); + + BitstreamEntry Entry; + if (Error E = Stream.advance(BitstreamCursor::AF_DontAutoprocessAbbrevs) + .moveInto(Entry)) + return E; + + switch (Entry.Kind) { + case BitstreamEntry::Error: + return reportError("malformed bitcode file"); + case BitstreamEntry::EndBlock: { + uint64_t BlockBitEnd = Stream.GetCurrentBitNo(); + BlockStats.NumBits += BlockBitEnd - BlockBitStart; + if (DumpRecords) { + O->OS << Indent << "</"; + if (BlockName) + O->OS << *BlockName << ">\n"; + else + O->OS << "UnknownBlock" << BlockID << ">\n"; + } + return Error::success(); + } + + case BitstreamEntry::SubBlock: { + uint64_t SubBlockBitStart = Stream.GetCurrentBitNo(); + if (Error E = parseBlock(Entry.ID, IndentLevel + 1, O, CheckHash)) + return E; + ++BlockStats.NumSubBlocks; + uint64_t SubBlockBitEnd = Stream.GetCurrentBitNo(); + + // Don't include subblock sizes in the size of this block. + BlockBitStart += SubBlockBitEnd - SubBlockBitStart; + continue; + } + case BitstreamEntry::Record: + // The interesting case. + break; + } + + if (Entry.ID == bitc::DEFINE_ABBREV) { + if (Error Err = Stream.ReadAbbrevRecord()) + return Err; + ++BlockStats.NumAbbrevs; + continue; + } + + Record.clear(); + + ++BlockStats.NumRecords; + + StringRef Blob; + uint64_t CurrentRecordPos = Stream.GetCurrentBitNo(); + unsigned Code; + if (Error E = Stream.readRecord(Entry.ID, Record, &Blob).moveInto(Code)) + return E; + + // Increment the # occurrences of this code. + if (BlockStats.CodeFreq.size() <= Code) + BlockStats.CodeFreq.resize(Code + 1); + BlockStats.CodeFreq[Code].NumInstances++; + BlockStats.CodeFreq[Code].TotalBits += + Stream.GetCurrentBitNo() - RecordStartBit; + if (Entry.ID != bitc::UNABBREV_RECORD) { + BlockStats.CodeFreq[Code].NumAbbrev++; + ++BlockStats.NumAbbreviatedRecords; + } + + if (DumpRecords) { + O->OS << Indent << " <"; + Optional<const char *> CodeName = + GetCodeName(Code, BlockID, BlockInfo, CurStreamType); + if (CodeName) + O->OS << *CodeName; + else + O->OS << "UnknownCode" << Code; + if (!O->Symbolic && CodeName) + O->OS << " codeid=" << Code; + const BitCodeAbbrev *Abbv = nullptr; + if (Entry.ID != bitc::UNABBREV_RECORD) { + Abbv = Stream.getAbbrev(Entry.ID); + O->OS << " abbrevid=" << Entry.ID; + } + + for (unsigned i = 0, e = Record.size(); i != e; ++i) + O->OS << " op" << i << "=" << (int64_t)Record[i]; + + // If we found a metadata index, let's verify that we had an offset + // before and validate its forward reference offset was correct! + if (BlockID == bitc::METADATA_BLOCK_ID) { + if (Code == bitc::METADATA_INDEX_OFFSET) { + if (Record.size() != 2) + O->OS << "(Invalid record)"; + else { + auto Offset = Record[0] + (Record[1] << 32); + MetadataIndexOffset = Stream.GetCurrentBitNo() + Offset; + } + } + if (Code == bitc::METADATA_INDEX) { + O->OS << " (offset "; + if (MetadataIndexOffset == RecordStartBit) + O->OS << "match)"; + else + O->OS << "mismatch: " << MetadataIndexOffset << " vs " + << RecordStartBit << ")"; + } + } + + // If we found a module hash, let's verify that it matches! + if (BlockID == bitc::MODULE_BLOCK_ID && Code == bitc::MODULE_CODE_HASH && + CheckHash.hasValue()) { + if (Record.size() != 5) + O->OS << " (invalid)"; + else { + // Recompute the hash and compare it to the one in the bitcode + SHA1 Hasher; + StringRef Hash; + Hasher.update(*CheckHash); + { + int BlockSize = (CurrentRecordPos / 8) - BlockEntryPos; + auto Ptr = Stream.getPointerToByte(BlockEntryPos, BlockSize); + Hasher.update(ArrayRef<uint8_t>(Ptr, BlockSize)); + Hash = Hasher.result(); + } + std::array<char, 20> RecordedHash; + int Pos = 0; + for (auto &Val : Record) { + assert(!(Val >> 32) && "Unexpected high bits set"); + support::endian::write32be(&RecordedHash[Pos], Val); + Pos += 4; + } + if (Hash == StringRef(RecordedHash.data(), RecordedHash.size())) + O->OS << " (match)"; + else + O->OS << " (!mismatch!)"; + } + } + + O->OS << "/>"; + + if (Abbv) { + for (unsigned i = 1, e = Abbv->getNumOperandInfos(); i != e; ++i) { + const BitCodeAbbrevOp &Op = Abbv->getOperandInfo(i); + if (!Op.isEncoding() || Op.getEncoding() != BitCodeAbbrevOp::Array) + continue; + assert(i + 2 == e && "Array op not second to last"); + std::string Str; + bool ArrayIsPrintable = true; + for (unsigned j = i - 1, je = Record.size(); j != je; ++j) { + if (!isPrint(static_cast<unsigned char>(Record[j]))) { + ArrayIsPrintable = false; + break; + } + Str += (char)Record[j]; + } + if (ArrayIsPrintable) + O->OS << " record string = '" << Str << "'"; + break; + } + } + + if (Blob.data()) { + if (canDecodeBlob(Code, BlockID)) { + if (Error E = decodeMetadataStringsBlob(Indent, Record, Blob, O->OS)) + return E; + } else { + O->OS << " blob data = "; + if (O->ShowBinaryBlobs) { + O->OS << "'"; + O->OS.write_escaped(Blob, /*hex=*/true) << "'"; + } else { + bool BlobIsPrintable = true; + for (char C : Blob) + if (!isPrint(static_cast<unsigned char>(C))) { + BlobIsPrintable = false; + break; + } + + if (BlobIsPrintable) + O->OS << "'" << Blob << "'"; + else + O->OS << "unprintable, " << Blob.size() << " bytes."; + } + } + } + + O->OS << "\n"; + } + + // Make sure that we can skip the current record. + if (Error Err = Stream.JumpToBit(CurrentRecordPos)) + return Err; + if (Expected<unsigned> Skipped = Stream.skipRecord(Entry.ID)) + ; // Do nothing. + else + return Skipped.takeError(); + } +} + diff --git a/contrib/libs/llvm14/lib/Bitcode/Reader/BitcodeReader.cpp b/contrib/libs/llvm14/lib/Bitcode/Reader/BitcodeReader.cpp new file mode 100644 index 0000000000..720ab560f9 --- /dev/null +++ b/contrib/libs/llvm14/lib/Bitcode/Reader/BitcodeReader.cpp @@ -0,0 +1,7144 @@ +//===- BitcodeReader.cpp - Internal BitcodeReader implementation ----------===// +// +// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. +// See https://llvm.org/LICENSE.txt for license information. +// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception +// +//===----------------------------------------------------------------------===// + +#include "llvm/Bitcode/BitcodeReader.h" +#include "MetadataLoader.h" +#include "ValueList.h" +#include "llvm/ADT/APFloat.h" +#include "llvm/ADT/APInt.h" +#include "llvm/ADT/ArrayRef.h" +#include "llvm/ADT/DenseMap.h" +#include "llvm/ADT/Optional.h" +#include "llvm/ADT/STLExtras.h" +#include "llvm/ADT/SmallString.h" +#include "llvm/ADT/SmallVector.h" +#include "llvm/ADT/StringRef.h" +#include "llvm/ADT/Triple.h" +#include "llvm/ADT/Twine.h" +#include "llvm/Bitcode/BitcodeCommon.h" +#include "llvm/Bitcode/LLVMBitCodes.h" +#include "llvm/Bitstream/BitstreamReader.h" +#include "llvm/Config/llvm-config.h" +#include "llvm/IR/Argument.h" +#include "llvm/IR/Attributes.h" +#include "llvm/IR/AutoUpgrade.h" +#include "llvm/IR/BasicBlock.h" +#include "llvm/IR/CallingConv.h" +#include "llvm/IR/Comdat.h" +#include "llvm/IR/Constant.h" +#include "llvm/IR/Constants.h" +#include "llvm/IR/DataLayout.h" +#include "llvm/IR/DebugInfo.h" +#include "llvm/IR/DebugInfoMetadata.h" +#include "llvm/IR/DebugLoc.h" +#include "llvm/IR/DerivedTypes.h" +#include "llvm/IR/Function.h" +#include "llvm/IR/GVMaterializer.h" +#include "llvm/IR/GlobalAlias.h" +#include "llvm/IR/GlobalIFunc.h" +#include "llvm/IR/GlobalObject.h" +#include "llvm/IR/GlobalValue.h" +#include "llvm/IR/GlobalVariable.h" +#include "llvm/IR/InlineAsm.h" +#include "llvm/IR/InstIterator.h" +#include "llvm/IR/InstrTypes.h" +#include "llvm/IR/Instruction.h" +#include "llvm/IR/Instructions.h" +#include "llvm/IR/Intrinsics.h" +#include "llvm/IR/LLVMContext.h" +#include "llvm/IR/Metadata.h" +#include "llvm/IR/Module.h" +#include "llvm/IR/ModuleSummaryIndex.h" +#include "llvm/IR/Operator.h" +#include "llvm/IR/Type.h" +#include "llvm/IR/Value.h" +#include "llvm/IR/Verifier.h" +#include "llvm/Support/AtomicOrdering.h" +#include "llvm/Support/Casting.h" +#include "llvm/Support/CommandLine.h" +#include "llvm/Support/Compiler.h" +#include "llvm/Support/Debug.h" +#include "llvm/Support/Error.h" +#include "llvm/Support/ErrorHandling.h" +#include "llvm/Support/ErrorOr.h" +#include "llvm/Support/ManagedStatic.h" +#include "llvm/Support/MathExtras.h" +#include "llvm/Support/MemoryBuffer.h" +#include "llvm/Support/raw_ostream.h" +#include <algorithm> +#include <cassert> +#include <cstddef> +#include <cstdint> +#include <deque> +#include <map> +#include <memory> +#include <set> +#include <string> +#include <system_error> +#include <tuple> +#include <utility> +#include <vector> + +using namespace llvm; + +static cl::opt<bool> PrintSummaryGUIDs( + "print-summary-global-ids", cl::init(false), cl::Hidden, + cl::desc( + "Print the global id for each value when reading the module summary")); + +namespace { + +enum { + SWITCH_INST_MAGIC = 0x4B5 // May 2012 => 1205 => Hex +}; + +} // end anonymous namespace + +static Error error(const Twine &Message) { + return make_error<StringError>( + Message, make_error_code(BitcodeError::CorruptedBitcode)); +} + +static Error hasInvalidBitcodeHeader(BitstreamCursor &Stream) { + if (!Stream.canSkipToPos(4)) + return createStringError(std::errc::illegal_byte_sequence, + "file too small to contain bitcode header"); + for (unsigned C : {'B', 'C'}) + if (Expected<SimpleBitstreamCursor::word_t> Res = Stream.Read(8)) { + if (Res.get() != C) + return createStringError(std::errc::illegal_byte_sequence, + "file doesn't start with bitcode header"); + } else + return Res.takeError(); + for (unsigned C : {0x0, 0xC, 0xE, 0xD}) + if (Expected<SimpleBitstreamCursor::word_t> Res = Stream.Read(4)) { + if (Res.get() != C) + return createStringError(std::errc::illegal_byte_sequence, + "file doesn't start with bitcode header"); + } else + return Res.takeError(); + return Error::success(); +} + +static Expected<BitstreamCursor> initStream(MemoryBufferRef Buffer) { + const unsigned char *BufPtr = (const unsigned char *)Buffer.getBufferStart(); + const unsigned char *BufEnd = BufPtr + Buffer.getBufferSize(); + + if (Buffer.getBufferSize() & 3) + return error("Invalid bitcode signature"); + + // If we have a wrapper header, parse it and ignore the non-bc file contents. + // The magic number is 0x0B17C0DE stored in little endian. + if (isBitcodeWrapper(BufPtr, BufEnd)) + if (SkipBitcodeWrapperHeader(BufPtr, BufEnd, true)) + return error("Invalid bitcode wrapper header"); + + BitstreamCursor Stream(ArrayRef<uint8_t>(BufPtr, BufEnd)); + if (Error Err = hasInvalidBitcodeHeader(Stream)) + return std::move(Err); + + return std::move(Stream); +} + +/// Convert a string from a record into an std::string, return true on failure. +template <typename StrTy> +static bool convertToString(ArrayRef<uint64_t> Record, unsigned Idx, + StrTy &Result) { + if (Idx > Record.size()) + return true; + + Result.append(Record.begin() + Idx, Record.end()); + return false; +} + +// Strip all the TBAA attachment for the module. +static void stripTBAA(Module *M) { + for (auto &F : *M) { + if (F.isMaterializable()) + continue; + for (auto &I : instructions(F)) + I.setMetadata(LLVMContext::MD_tbaa, nullptr); + } +} + +/// Read the "IDENTIFICATION_BLOCK_ID" block, do some basic enforcement on the +/// "epoch" encoded in the bitcode, and return the producer name if any. +static Expected<std::string> readIdentificationBlock(BitstreamCursor &Stream) { + if (Error Err = Stream.EnterSubBlock(bitc::IDENTIFICATION_BLOCK_ID)) + return std::move(Err); + + // Read all the records. + SmallVector<uint64_t, 64> Record; + + std::string ProducerIdentification; + + while (true) { + BitstreamEntry Entry; + if (Error E = Stream.advance().moveInto(Entry)) + return std::move(E); + + switch (Entry.Kind) { + default: + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + return ProducerIdentification; + case BitstreamEntry::Record: + // The interesting case. + break; + } + + // Read a record. + Record.clear(); + Expected<unsigned> MaybeBitCode = Stream.readRecord(Entry.ID, Record); + if (!MaybeBitCode) + return MaybeBitCode.takeError(); + switch (MaybeBitCode.get()) { + default: // Default behavior: reject + return error("Invalid value"); + case bitc::IDENTIFICATION_CODE_STRING: // IDENTIFICATION: [strchr x N] + convertToString(Record, 0, ProducerIdentification); + break; + case bitc::IDENTIFICATION_CODE_EPOCH: { // EPOCH: [epoch#] + unsigned epoch = (unsigned)Record[0]; + if (epoch != bitc::BITCODE_CURRENT_EPOCH) { + return error( + Twine("Incompatible epoch: Bitcode '") + Twine(epoch) + + "' vs current: '" + Twine(bitc::BITCODE_CURRENT_EPOCH) + "'"); + } + } + } + } +} + +static Expected<std::string> readIdentificationCode(BitstreamCursor &Stream) { + // We expect a number of well-defined blocks, though we don't necessarily + // need to understand them all. + while (true) { + if (Stream.AtEndOfStream()) + return ""; + + BitstreamEntry Entry; + if (Error E = Stream.advance().moveInto(Entry)) + return std::move(E); + + switch (Entry.Kind) { + case BitstreamEntry::EndBlock: + case BitstreamEntry::Error: + return error("Malformed block"); + + case BitstreamEntry::SubBlock: + if (Entry.ID == bitc::IDENTIFICATION_BLOCK_ID) + return readIdentificationBlock(Stream); + + // Ignore other sub-blocks. + if (Error Err = Stream.SkipBlock()) + return std::move(Err); + continue; + case BitstreamEntry::Record: + if (Error E = Stream.skipRecord(Entry.ID).takeError()) + return std::move(E); + continue; + } + } +} + +static Expected<bool> hasObjCCategoryInModule(BitstreamCursor &Stream) { + if (Error Err = Stream.EnterSubBlock(bitc::MODULE_BLOCK_ID)) + return std::move(Err); + + SmallVector<uint64_t, 64> Record; + // Read all the records for this module. + + while (true) { + Expected<BitstreamEntry> MaybeEntry = Stream.advanceSkippingSubblocks(); + if (!MaybeEntry) + return MaybeEntry.takeError(); + BitstreamEntry Entry = MaybeEntry.get(); + + switch (Entry.Kind) { + case BitstreamEntry::SubBlock: // Handled for us already. + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + return false; + case BitstreamEntry::Record: + // The interesting case. + break; + } + + // Read a record. + Expected<unsigned> MaybeRecord = Stream.readRecord(Entry.ID, Record); + if (!MaybeRecord) + return MaybeRecord.takeError(); + switch (MaybeRecord.get()) { + default: + break; // Default behavior, ignore unknown content. + case bitc::MODULE_CODE_SECTIONNAME: { // SECTIONNAME: [strchr x N] + std::string S; + if (convertToString(Record, 0, S)) + return error("Invalid record"); + // Check for the i386 and other (x86_64, ARM) conventions + if (S.find("__DATA,__objc_catlist") != std::string::npos || + S.find("__OBJC,__category") != std::string::npos) + return true; + break; + } + } + Record.clear(); + } + llvm_unreachable("Exit infinite loop"); +} + +static Expected<bool> hasObjCCategory(BitstreamCursor &Stream) { + // We expect a number of well-defined blocks, though we don't necessarily + // need to understand them all. + while (true) { + BitstreamEntry Entry; + if (Error E = Stream.advance().moveInto(Entry)) + return std::move(E); + + switch (Entry.Kind) { + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + return false; + + case BitstreamEntry::SubBlock: + if (Entry.ID == bitc::MODULE_BLOCK_ID) + return hasObjCCategoryInModule(Stream); + + // Ignore other sub-blocks. + if (Error Err = Stream.SkipBlock()) + return std::move(Err); + continue; + + case BitstreamEntry::Record: + if (Error E = Stream.skipRecord(Entry.ID).takeError()) + return std::move(E); + continue; + } + } +} + +static Expected<std::string> readModuleTriple(BitstreamCursor &Stream) { + if (Error Err = Stream.EnterSubBlock(bitc::MODULE_BLOCK_ID)) + return std::move(Err); + + SmallVector<uint64_t, 64> Record; + + std::string Triple; + + // Read all the records for this module. + while (true) { + Expected<BitstreamEntry> MaybeEntry = Stream.advanceSkippingSubblocks(); + if (!MaybeEntry) + return MaybeEntry.takeError(); + BitstreamEntry Entry = MaybeEntry.get(); + + switch (Entry.Kind) { + case BitstreamEntry::SubBlock: // Handled for us already. + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + return Triple; + case BitstreamEntry::Record: + // The interesting case. + break; + } + + // Read a record. + Expected<unsigned> MaybeRecord = Stream.readRecord(Entry.ID, Record); + if (!MaybeRecord) + return MaybeRecord.takeError(); + switch (MaybeRecord.get()) { + default: break; // Default behavior, ignore unknown content. + case bitc::MODULE_CODE_TRIPLE: { // TRIPLE: [strchr x N] + std::string S; + if (convertToString(Record, 0, S)) + return error("Invalid record"); + Triple = S; + break; + } + } + Record.clear(); + } + llvm_unreachable("Exit infinite loop"); +} + +static Expected<std::string> readTriple(BitstreamCursor &Stream) { + // We expect a number of well-defined blocks, though we don't necessarily + // need to understand them all. + while (true) { + Expected<BitstreamEntry> MaybeEntry = Stream.advance(); + if (!MaybeEntry) + return MaybeEntry.takeError(); + BitstreamEntry Entry = MaybeEntry.get(); + + switch (Entry.Kind) { + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + return ""; + + case BitstreamEntry::SubBlock: + if (Entry.ID == bitc::MODULE_BLOCK_ID) + return readModuleTriple(Stream); + + // Ignore other sub-blocks. + if (Error Err = Stream.SkipBlock()) + return std::move(Err); + continue; + + case BitstreamEntry::Record: + if (llvm::Expected<unsigned> Skipped = Stream.skipRecord(Entry.ID)) + continue; + else + return Skipped.takeError(); + } + } +} + +namespace { + +class BitcodeReaderBase { +protected: + BitcodeReaderBase(BitstreamCursor Stream, StringRef Strtab) + : Stream(std::move(Stream)), Strtab(Strtab) { + this->Stream.setBlockInfo(&BlockInfo); + } + + BitstreamBlockInfo BlockInfo; + BitstreamCursor Stream; + StringRef Strtab; + + /// In version 2 of the bitcode we store names of global values and comdats in + /// a string table rather than in the VST. + bool UseStrtab = false; + + Expected<unsigned> parseVersionRecord(ArrayRef<uint64_t> Record); + + /// If this module uses a string table, pop the reference to the string table + /// and return the referenced string and the rest of the record. Otherwise + /// just return the record itself. + std::pair<StringRef, ArrayRef<uint64_t>> + readNameFromStrtab(ArrayRef<uint64_t> Record); + + bool readBlockInfo(); + + // Contains an arbitrary and optional string identifying the bitcode producer + std::string ProducerIdentification; + + Error error(const Twine &Message); +}; + +} // end anonymous namespace + +Error BitcodeReaderBase::error(const Twine &Message) { + std::string FullMsg = Message.str(); + if (!ProducerIdentification.empty()) + FullMsg += " (Producer: '" + ProducerIdentification + "' Reader: 'LLVM " + + LLVM_VERSION_STRING "')"; + return ::error(FullMsg); +} + +Expected<unsigned> +BitcodeReaderBase::parseVersionRecord(ArrayRef<uint64_t> Record) { + if (Record.empty()) + return error("Invalid record"); + unsigned ModuleVersion = Record[0]; + if (ModuleVersion > 2) + return error("Invalid value"); + UseStrtab = ModuleVersion >= 2; + return ModuleVersion; +} + +std::pair<StringRef, ArrayRef<uint64_t>> +BitcodeReaderBase::readNameFromStrtab(ArrayRef<uint64_t> Record) { + if (!UseStrtab) + return {"", Record}; + // Invalid reference. Let the caller complain about the record being empty. + if (Record[0] + Record[1] > Strtab.size()) + return {"", {}}; + return {StringRef(Strtab.data() + Record[0], Record[1]), Record.slice(2)}; +} + +namespace { + +class BitcodeReader : public BitcodeReaderBase, public GVMaterializer { + LLVMContext &Context; + Module *TheModule = nullptr; + // Next offset to start scanning for lazy parsing of function bodies. + uint64_t NextUnreadBit = 0; + // Last function offset found in the VST. + uint64_t LastFunctionBlockBit = 0; + bool SeenValueSymbolTable = false; + uint64_t VSTOffset = 0; + + std::vector<std::string> SectionTable; + std::vector<std::string> GCTable; + + std::vector<Type*> TypeList; + DenseMap<Function *, FunctionType *> FunctionTypes; + BitcodeReaderValueList ValueList; + Optional<MetadataLoader> MDLoader; + std::vector<Comdat *> ComdatList; + DenseSet<GlobalObject *> ImplicitComdatObjects; + SmallVector<Instruction *, 64> InstructionList; + + std::vector<std::pair<GlobalVariable *, unsigned>> GlobalInits; + std::vector<std::pair<GlobalValue *, unsigned>> IndirectSymbolInits; + + struct FunctionOperandInfo { + Function *F; + unsigned PersonalityFn; + unsigned Prefix; + unsigned Prologue; + }; + std::vector<FunctionOperandInfo> FunctionOperands; + + /// The set of attributes by index. Index zero in the file is for null, and + /// is thus not represented here. As such all indices are off by one. + std::vector<AttributeList> MAttributes; + + /// The set of attribute groups. + std::map<unsigned, AttributeList> MAttributeGroups; + + /// While parsing a function body, this is a list of the basic blocks for the + /// function. + std::vector<BasicBlock*> FunctionBBs; + + // When reading the module header, this list is populated with functions that + // have bodies later in the file. + std::vector<Function*> FunctionsWithBodies; + + // When intrinsic functions are encountered which require upgrading they are + // stored here with their replacement function. + using UpdatedIntrinsicMap = DenseMap<Function *, Function *>; + UpdatedIntrinsicMap UpgradedIntrinsics; + // Intrinsics which were remangled because of types rename + UpdatedIntrinsicMap RemangledIntrinsics; + + // Several operations happen after the module header has been read, but + // before function bodies are processed. This keeps track of whether + // we've done this yet. + bool SeenFirstFunctionBody = false; + + /// When function bodies are initially scanned, this map contains info about + /// where to find deferred function body in the stream. + DenseMap<Function*, uint64_t> DeferredFunctionInfo; + + /// When Metadata block is initially scanned when parsing the module, we may + /// choose to defer parsing of the metadata. This vector contains info about + /// which Metadata blocks are deferred. + std::vector<uint64_t> DeferredMetadataInfo; + + /// These are basic blocks forward-referenced by block addresses. They are + /// inserted lazily into functions when they're loaded. The basic block ID is + /// its index into the vector. + DenseMap<Function *, std::vector<BasicBlock *>> BasicBlockFwdRefs; + std::deque<Function *> BasicBlockFwdRefQueue; + + /// Indicates that we are using a new encoding for instruction operands where + /// most operands in the current FUNCTION_BLOCK are encoded relative to the + /// instruction number, for a more compact encoding. Some instruction + /// operands are not relative to the instruction ID: basic block numbers, and + /// types. Once the old style function blocks have been phased out, we would + /// not need this flag. + bool UseRelativeIDs = false; + + /// True if all functions will be materialized, negating the need to process + /// (e.g.) blockaddress forward references. + bool WillMaterializeAllForwardRefs = false; + + bool StripDebugInfo = false; + TBAAVerifier TBAAVerifyHelper; + + std::vector<std::string> BundleTags; + SmallVector<SyncScope::ID, 8> SSIDs; + +public: + BitcodeReader(BitstreamCursor Stream, StringRef Strtab, + StringRef ProducerIdentification, LLVMContext &Context); + + Error materializeForwardReferencedFunctions(); + + Error materialize(GlobalValue *GV) override; + Error materializeModule() override; + std::vector<StructType *> getIdentifiedStructTypes() const override; + + /// Main interface to parsing a bitcode buffer. + /// \returns true if an error occurred. + Error parseBitcodeInto( + Module *M, bool ShouldLazyLoadMetadata = false, bool IsImporting = false, + DataLayoutCallbackTy DataLayoutCallback = [](StringRef) { return None; }); + + static uint64_t decodeSignRotatedValue(uint64_t V); + + /// Materialize any deferred Metadata block. + Error materializeMetadata() override; + + void setStripDebugInfo() override; + +private: + std::vector<StructType *> IdentifiedStructTypes; + StructType *createIdentifiedStructType(LLVMContext &Context, StringRef Name); + StructType *createIdentifiedStructType(LLVMContext &Context); + + Type *getTypeByID(unsigned ID); + + Value *getFnValueByID(unsigned ID, Type *Ty) { + if (Ty && Ty->isMetadataTy()) + return MetadataAsValue::get(Ty->getContext(), getFnMetadataByID(ID)); + return ValueList.getValueFwdRef(ID, Ty); + } + + Metadata *getFnMetadataByID(unsigned ID) { + return MDLoader->getMetadataFwdRefOrLoad(ID); + } + + BasicBlock *getBasicBlock(unsigned ID) const { + if (ID >= FunctionBBs.size()) return nullptr; // Invalid ID + return FunctionBBs[ID]; + } + + AttributeList getAttributes(unsigned i) const { + if (i-1 < MAttributes.size()) + return MAttributes[i-1]; + return AttributeList(); + } + + /// Read a value/type pair out of the specified record from slot 'Slot'. + /// Increment Slot past the number of slots used in the record. Return true on + /// failure. + bool getValueTypePair(const SmallVectorImpl<uint64_t> &Record, unsigned &Slot, + unsigned InstNum, Value *&ResVal) { + if (Slot == Record.size()) return true; + unsigned ValNo = (unsigned)Record[Slot++]; + // Adjust the ValNo, if it was encoded relative to the InstNum. + if (UseRelativeIDs) + ValNo = InstNum - ValNo; + if (ValNo < InstNum) { + // If this is not a forward reference, just return the value we already + // have. + ResVal = getFnValueByID(ValNo, nullptr); + return ResVal == nullptr; + } + if (Slot == Record.size()) + return true; + + unsigned TypeNo = (unsigned)Record[Slot++]; + ResVal = getFnValueByID(ValNo, getTypeByID(TypeNo)); + return ResVal == nullptr; + } + + /// Read a value out of the specified record from slot 'Slot'. Increment Slot + /// past the number of slots used by the value in the record. Return true if + /// there is an error. + bool popValue(const SmallVectorImpl<uint64_t> &Record, unsigned &Slot, + unsigned InstNum, Type *Ty, Value *&ResVal) { + if (getValue(Record, Slot, InstNum, Ty, ResVal)) + return true; + // All values currently take a single record slot. + ++Slot; + return false; + } + + /// Like popValue, but does not increment the Slot number. + bool getValue(const SmallVectorImpl<uint64_t> &Record, unsigned Slot, + unsigned InstNum, Type *Ty, Value *&ResVal) { + ResVal = getValue(Record, Slot, InstNum, Ty); + return ResVal == nullptr; + } + + /// Version of getValue that returns ResVal directly, or 0 if there is an + /// error. + Value *getValue(const SmallVectorImpl<uint64_t> &Record, unsigned Slot, + unsigned InstNum, Type *Ty) { + if (Slot == Record.size()) return nullptr; + unsigned ValNo = (unsigned)Record[Slot]; + // Adjust the ValNo, if it was encoded relative to the InstNum. + if (UseRelativeIDs) + ValNo = InstNum - ValNo; + return getFnValueByID(ValNo, Ty); + } + + /// Like getValue, but decodes signed VBRs. + Value *getValueSigned(const SmallVectorImpl<uint64_t> &Record, unsigned Slot, + unsigned InstNum, Type *Ty) { + if (Slot == Record.size()) return nullptr; + unsigned ValNo = (unsigned)decodeSignRotatedValue(Record[Slot]); + // Adjust the ValNo, if it was encoded relative to the InstNum. + if (UseRelativeIDs) + ValNo = InstNum - ValNo; + return getFnValueByID(ValNo, Ty); + } + + /// Upgrades old-style typeless byval/sret/inalloca attributes by adding the + /// corresponding argument's pointee type. Also upgrades intrinsics that now + /// require an elementtype attribute. + void propagateAttributeTypes(CallBase *CB, ArrayRef<Type *> ArgsTys); + + /// Converts alignment exponent (i.e. power of two (or zero)) to the + /// corresponding alignment to use. If alignment is too large, returns + /// a corresponding error code. + Error parseAlignmentValue(uint64_t Exponent, MaybeAlign &Alignment); + Error parseAttrKind(uint64_t Code, Attribute::AttrKind *Kind); + Error parseModule( + uint64_t ResumeBit, bool ShouldLazyLoadMetadata = false, + DataLayoutCallbackTy DataLayoutCallback = [](StringRef) { return None; }); + + Error parseComdatRecord(ArrayRef<uint64_t> Record); + Error parseGlobalVarRecord(ArrayRef<uint64_t> Record); + Error parseFunctionRecord(ArrayRef<uint64_t> Record); + Error parseGlobalIndirectSymbolRecord(unsigned BitCode, + ArrayRef<uint64_t> Record); + + Error parseAttributeBlock(); + Error parseAttributeGroupBlock(); + Error parseTypeTable(); + Error parseTypeTableBody(); + Error parseOperandBundleTags(); + Error parseSyncScopeNames(); + + Expected<Value *> recordValue(SmallVectorImpl<uint64_t> &Record, + unsigned NameIndex, Triple &TT); + void setDeferredFunctionInfo(unsigned FuncBitcodeOffsetDelta, Function *F, + ArrayRef<uint64_t> Record); + Error parseValueSymbolTable(uint64_t Offset = 0); + Error parseGlobalValueSymbolTable(); + Error parseConstants(); + Error rememberAndSkipFunctionBodies(); + Error rememberAndSkipFunctionBody(); + /// Save the positions of the Metadata blocks and skip parsing the blocks. + Error rememberAndSkipMetadata(); + Error typeCheckLoadStoreInst(Type *ValType, Type *PtrType); + Error parseFunctionBody(Function *F); + Error globalCleanup(); + Error resolveGlobalAndIndirectSymbolInits(); + Error parseUseLists(); + Error findFunctionInStream( + Function *F, + DenseMap<Function *, uint64_t>::iterator DeferredFunctionInfoIterator); + + SyncScope::ID getDecodedSyncScopeID(unsigned Val); +}; + +/// Class to manage reading and parsing function summary index bitcode +/// files/sections. +class ModuleSummaryIndexBitcodeReader : public BitcodeReaderBase { + /// The module index built during parsing. + ModuleSummaryIndex &TheIndex; + + /// Indicates whether we have encountered a global value summary section + /// yet during parsing. + bool SeenGlobalValSummary = false; + + /// Indicates whether we have already parsed the VST, used for error checking. + bool SeenValueSymbolTable = false; + + /// Set to the offset of the VST recorded in the MODULE_CODE_VSTOFFSET record. + /// Used to enable on-demand parsing of the VST. + uint64_t VSTOffset = 0; + + // Map to save ValueId to ValueInfo association that was recorded in the + // ValueSymbolTable. It is used after the VST is parsed to convert + // call graph edges read from the function summary from referencing + // callees by their ValueId to using the ValueInfo instead, which is how + // they are recorded in the summary index being built. + // We save a GUID which refers to the same global as the ValueInfo, but + // ignoring the linkage, i.e. for values other than local linkage they are + // identical. + DenseMap<unsigned, std::pair<ValueInfo, GlobalValue::GUID>> + ValueIdToValueInfoMap; + + /// Map populated during module path string table parsing, from the + /// module ID to a string reference owned by the index's module + /// path string table, used to correlate with combined index + /// summary records. + DenseMap<uint64_t, StringRef> ModuleIdMap; + + /// Original source file name recorded in a bitcode record. + std::string SourceFileName; + + /// The string identifier given to this module by the client, normally the + /// path to the bitcode file. + StringRef ModulePath; + + /// For per-module summary indexes, the unique numerical identifier given to + /// this module by the client. + unsigned ModuleId; + +public: + ModuleSummaryIndexBitcodeReader(BitstreamCursor Stream, StringRef Strtab, + ModuleSummaryIndex &TheIndex, + StringRef ModulePath, unsigned ModuleId); + + Error parseModule(); + +private: + void setValueGUID(uint64_t ValueID, StringRef ValueName, + GlobalValue::LinkageTypes Linkage, + StringRef SourceFileName); + Error parseValueSymbolTable( + uint64_t Offset, + DenseMap<unsigned, GlobalValue::LinkageTypes> &ValueIdToLinkageMap); + std::vector<ValueInfo> makeRefList(ArrayRef<uint64_t> Record); + std::vector<FunctionSummary::EdgeTy> makeCallList(ArrayRef<uint64_t> Record, + bool IsOldProfileFormat, + bool HasProfile, + bool HasRelBF); + Error parseEntireSummary(unsigned ID); + Error parseModuleStringTable(); + void parseTypeIdCompatibleVtableSummaryRecord(ArrayRef<uint64_t> Record); + void parseTypeIdCompatibleVtableInfo(ArrayRef<uint64_t> Record, size_t &Slot, + TypeIdCompatibleVtableInfo &TypeId); + std::vector<FunctionSummary::ParamAccess> + parseParamAccesses(ArrayRef<uint64_t> Record); + + std::pair<ValueInfo, GlobalValue::GUID> + getValueInfoFromValueId(unsigned ValueId); + + void addThisModule(); + ModuleSummaryIndex::ModuleInfo *getThisModule(); +}; + +} // end anonymous namespace + +std::error_code llvm::errorToErrorCodeAndEmitErrors(LLVMContext &Ctx, + Error Err) { + if (Err) { + std::error_code EC; + handleAllErrors(std::move(Err), [&](ErrorInfoBase &EIB) { + EC = EIB.convertToErrorCode(); + Ctx.emitError(EIB.message()); + }); + return EC; + } + return std::error_code(); +} + +BitcodeReader::BitcodeReader(BitstreamCursor Stream, StringRef Strtab, + StringRef ProducerIdentification, + LLVMContext &Context) + : BitcodeReaderBase(std::move(Stream), Strtab), Context(Context), + ValueList(Context, Stream.SizeInBytes()) { + this->ProducerIdentification = std::string(ProducerIdentification); +} + +Error BitcodeReader::materializeForwardReferencedFunctions() { + if (WillMaterializeAllForwardRefs) + return Error::success(); + + // Prevent recursion. + WillMaterializeAllForwardRefs = true; + + while (!BasicBlockFwdRefQueue.empty()) { + Function *F = BasicBlockFwdRefQueue.front(); + BasicBlockFwdRefQueue.pop_front(); + assert(F && "Expected valid function"); + if (!BasicBlockFwdRefs.count(F)) + // Already materialized. + continue; + + // Check for a function that isn't materializable to prevent an infinite + // loop. When parsing a blockaddress stored in a global variable, there + // isn't a trivial way to check if a function will have a body without a + // linear search through FunctionsWithBodies, so just check it here. + if (!F->isMaterializable()) + return error("Never resolved function from blockaddress"); + + // Try to materialize F. + if (Error Err = materialize(F)) + return Err; + } + assert(BasicBlockFwdRefs.empty() && "Function missing from queue"); + + // Reset state. + WillMaterializeAllForwardRefs = false; + return Error::success(); +} + +//===----------------------------------------------------------------------===// +// Helper functions to implement forward reference resolution, etc. +//===----------------------------------------------------------------------===// + +static bool hasImplicitComdat(size_t Val) { + switch (Val) { + default: + return false; + case 1: // Old WeakAnyLinkage + case 4: // Old LinkOnceAnyLinkage + case 10: // Old WeakODRLinkage + case 11: // Old LinkOnceODRLinkage + return true; + } +} + +static GlobalValue::LinkageTypes getDecodedLinkage(unsigned Val) { + switch (Val) { + default: // Map unknown/new linkages to external + case 0: + return GlobalValue::ExternalLinkage; + case 2: + return GlobalValue::AppendingLinkage; + case 3: + return GlobalValue::InternalLinkage; + case 5: + return GlobalValue::ExternalLinkage; // Obsolete DLLImportLinkage + case 6: + return GlobalValue::ExternalLinkage; // Obsolete DLLExportLinkage + case 7: + return GlobalValue::ExternalWeakLinkage; + case 8: + return GlobalValue::CommonLinkage; + case 9: + return GlobalValue::PrivateLinkage; + case 12: + return GlobalValue::AvailableExternallyLinkage; + case 13: + return GlobalValue::PrivateLinkage; // Obsolete LinkerPrivateLinkage + case 14: + return GlobalValue::PrivateLinkage; // Obsolete LinkerPrivateWeakLinkage + case 15: + return GlobalValue::ExternalLinkage; // Obsolete LinkOnceODRAutoHideLinkage + case 1: // Old value with implicit comdat. + case 16: + return GlobalValue::WeakAnyLinkage; + case 10: // Old value with implicit comdat. + case 17: + return GlobalValue::WeakODRLinkage; + case 4: // Old value with implicit comdat. + case 18: + return GlobalValue::LinkOnceAnyLinkage; + case 11: // Old value with implicit comdat. + case 19: + return GlobalValue::LinkOnceODRLinkage; + } +} + +static FunctionSummary::FFlags getDecodedFFlags(uint64_t RawFlags) { + FunctionSummary::FFlags Flags; + Flags.ReadNone = RawFlags & 0x1; + Flags.ReadOnly = (RawFlags >> 1) & 0x1; + Flags.NoRecurse = (RawFlags >> 2) & 0x1; + Flags.ReturnDoesNotAlias = (RawFlags >> 3) & 0x1; + Flags.NoInline = (RawFlags >> 4) & 0x1; + Flags.AlwaysInline = (RawFlags >> 5) & 0x1; + Flags.NoUnwind = (RawFlags >> 6) & 0x1; + Flags.MayThrow = (RawFlags >> 7) & 0x1; + Flags.HasUnknownCall = (RawFlags >> 8) & 0x1; + Flags.MustBeUnreachable = (RawFlags >> 9) & 0x1; + return Flags; +} + +// Decode the flags for GlobalValue in the summary. The bits for each attribute: +// +// linkage: [0,4), notEligibleToImport: 4, live: 5, local: 6, canAutoHide: 7, +// visibility: [8, 10). +static GlobalValueSummary::GVFlags getDecodedGVSummaryFlags(uint64_t RawFlags, + uint64_t Version) { + // Summary were not emitted before LLVM 3.9, we don't need to upgrade Linkage + // like getDecodedLinkage() above. Any future change to the linkage enum and + // to getDecodedLinkage() will need to be taken into account here as above. + auto Linkage = GlobalValue::LinkageTypes(RawFlags & 0xF); // 4 bits + auto Visibility = GlobalValue::VisibilityTypes((RawFlags >> 8) & 3); // 2 bits + RawFlags = RawFlags >> 4; + bool NotEligibleToImport = (RawFlags & 0x1) || Version < 3; + // The Live flag wasn't introduced until version 3. For dead stripping + // to work correctly on earlier versions, we must conservatively treat all + // values as live. + bool Live = (RawFlags & 0x2) || Version < 3; + bool Local = (RawFlags & 0x4); + bool AutoHide = (RawFlags & 0x8); + + return GlobalValueSummary::GVFlags(Linkage, Visibility, NotEligibleToImport, + Live, Local, AutoHide); +} + +// Decode the flags for GlobalVariable in the summary +static GlobalVarSummary::GVarFlags getDecodedGVarFlags(uint64_t RawFlags) { + return GlobalVarSummary::GVarFlags( + (RawFlags & 0x1) ? true : false, (RawFlags & 0x2) ? true : false, + (RawFlags & 0x4) ? true : false, + (GlobalObject::VCallVisibility)(RawFlags >> 3)); +} + +static GlobalValue::VisibilityTypes getDecodedVisibility(unsigned Val) { + switch (Val) { + default: // Map unknown visibilities to default. + case 0: return GlobalValue::DefaultVisibility; + case 1: return GlobalValue::HiddenVisibility; + case 2: return GlobalValue::ProtectedVisibility; + } +} + +static GlobalValue::DLLStorageClassTypes +getDecodedDLLStorageClass(unsigned Val) { + switch (Val) { + default: // Map unknown values to default. + case 0: return GlobalValue::DefaultStorageClass; + case 1: return GlobalValue::DLLImportStorageClass; + case 2: return GlobalValue::DLLExportStorageClass; + } +} + +static bool getDecodedDSOLocal(unsigned Val) { + switch(Val) { + default: // Map unknown values to preemptable. + case 0: return false; + case 1: return true; + } +} + +static GlobalVariable::ThreadLocalMode getDecodedThreadLocalMode(unsigned Val) { + switch (Val) { + case 0: return GlobalVariable::NotThreadLocal; + default: // Map unknown non-zero value to general dynamic. + case 1: return GlobalVariable::GeneralDynamicTLSModel; + case 2: return GlobalVariable::LocalDynamicTLSModel; + case 3: return GlobalVariable::InitialExecTLSModel; + case 4: return GlobalVariable::LocalExecTLSModel; + } +} + +static GlobalVariable::UnnamedAddr getDecodedUnnamedAddrType(unsigned Val) { + switch (Val) { + default: // Map unknown to UnnamedAddr::None. + case 0: return GlobalVariable::UnnamedAddr::None; + case 1: return GlobalVariable::UnnamedAddr::Global; + case 2: return GlobalVariable::UnnamedAddr::Local; + } +} + +static int getDecodedCastOpcode(unsigned Val) { + switch (Val) { + default: return -1; + case bitc::CAST_TRUNC : return Instruction::Trunc; + case bitc::CAST_ZEXT : return Instruction::ZExt; + case bitc::CAST_SEXT : return Instruction::SExt; + case bitc::CAST_FPTOUI : return Instruction::FPToUI; + case bitc::CAST_FPTOSI : return Instruction::FPToSI; + case bitc::CAST_UITOFP : return Instruction::UIToFP; + case bitc::CAST_SITOFP : return Instruction::SIToFP; + case bitc::CAST_FPTRUNC : return Instruction::FPTrunc; + case bitc::CAST_FPEXT : return Instruction::FPExt; + case bitc::CAST_PTRTOINT: return Instruction::PtrToInt; + case bitc::CAST_INTTOPTR: return Instruction::IntToPtr; + case bitc::CAST_BITCAST : return Instruction::BitCast; + case bitc::CAST_ADDRSPACECAST: return Instruction::AddrSpaceCast; + } +} + +static int getDecodedUnaryOpcode(unsigned Val, Type *Ty) { + bool IsFP = Ty->isFPOrFPVectorTy(); + // UnOps are only valid for int/fp or vector of int/fp types + if (!IsFP && !Ty->isIntOrIntVectorTy()) + return -1; + + switch (Val) { + default: + return -1; + case bitc::UNOP_FNEG: + return IsFP ? Instruction::FNeg : -1; + } +} + +static int getDecodedBinaryOpcode(unsigned Val, Type *Ty) { + bool IsFP = Ty->isFPOrFPVectorTy(); + // BinOps are only valid for int/fp or vector of int/fp types + if (!IsFP && !Ty->isIntOrIntVectorTy()) + return -1; + + switch (Val) { + default: + return -1; + case bitc::BINOP_ADD: + return IsFP ? Instruction::FAdd : Instruction::Add; + case bitc::BINOP_SUB: + return IsFP ? Instruction::FSub : Instruction::Sub; + case bitc::BINOP_MUL: + return IsFP ? Instruction::FMul : Instruction::Mul; + case bitc::BINOP_UDIV: + return IsFP ? -1 : Instruction::UDiv; + case bitc::BINOP_SDIV: + return IsFP ? Instruction::FDiv : Instruction::SDiv; + case bitc::BINOP_UREM: + return IsFP ? -1 : Instruction::URem; + case bitc::BINOP_SREM: + return IsFP ? Instruction::FRem : Instruction::SRem; + case bitc::BINOP_SHL: + return IsFP ? -1 : Instruction::Shl; + case bitc::BINOP_LSHR: + return IsFP ? -1 : Instruction::LShr; + case bitc::BINOP_ASHR: + return IsFP ? -1 : Instruction::AShr; + case bitc::BINOP_AND: + return IsFP ? -1 : Instruction::And; + case bitc::BINOP_OR: + return IsFP ? -1 : Instruction::Or; + case bitc::BINOP_XOR: + return IsFP ? -1 : Instruction::Xor; + } +} + +static AtomicRMWInst::BinOp getDecodedRMWOperation(unsigned Val) { + switch (Val) { + default: return AtomicRMWInst::BAD_BINOP; + case bitc::RMW_XCHG: return AtomicRMWInst::Xchg; + case bitc::RMW_ADD: return AtomicRMWInst::Add; + case bitc::RMW_SUB: return AtomicRMWInst::Sub; + case bitc::RMW_AND: return AtomicRMWInst::And; + case bitc::RMW_NAND: return AtomicRMWInst::Nand; + case bitc::RMW_OR: return AtomicRMWInst::Or; + case bitc::RMW_XOR: return AtomicRMWInst::Xor; + case bitc::RMW_MAX: return AtomicRMWInst::Max; + case bitc::RMW_MIN: return AtomicRMWInst::Min; + case bitc::RMW_UMAX: return AtomicRMWInst::UMax; + case bitc::RMW_UMIN: return AtomicRMWInst::UMin; + case bitc::RMW_FADD: return AtomicRMWInst::FAdd; + case bitc::RMW_FSUB: return AtomicRMWInst::FSub; + } +} + +static AtomicOrdering getDecodedOrdering(unsigned Val) { + switch (Val) { + case bitc::ORDERING_NOTATOMIC: return AtomicOrdering::NotAtomic; + case bitc::ORDERING_UNORDERED: return AtomicOrdering::Unordered; + case bitc::ORDERING_MONOTONIC: return AtomicOrdering::Monotonic; + case bitc::ORDERING_ACQUIRE: return AtomicOrdering::Acquire; + case bitc::ORDERING_RELEASE: return AtomicOrdering::Release; + case bitc::ORDERING_ACQREL: return AtomicOrdering::AcquireRelease; + default: // Map unknown orderings to sequentially-consistent. + case bitc::ORDERING_SEQCST: return AtomicOrdering::SequentiallyConsistent; + } +} + +static Comdat::SelectionKind getDecodedComdatSelectionKind(unsigned Val) { + switch (Val) { + default: // Map unknown selection kinds to any. + case bitc::COMDAT_SELECTION_KIND_ANY: + return Comdat::Any; + case bitc::COMDAT_SELECTION_KIND_EXACT_MATCH: + return Comdat::ExactMatch; + case bitc::COMDAT_SELECTION_KIND_LARGEST: + return Comdat::Largest; + case bitc::COMDAT_SELECTION_KIND_NO_DUPLICATES: + return Comdat::NoDeduplicate; + case bitc::COMDAT_SELECTION_KIND_SAME_SIZE: + return Comdat::SameSize; + } +} + +static FastMathFlags getDecodedFastMathFlags(unsigned Val) { + FastMathFlags FMF; + if (0 != (Val & bitc::UnsafeAlgebra)) + FMF.setFast(); + if (0 != (Val & bitc::AllowReassoc)) + FMF.setAllowReassoc(); + if (0 != (Val & bitc::NoNaNs)) + FMF.setNoNaNs(); + if (0 != (Val & bitc::NoInfs)) + FMF.setNoInfs(); + if (0 != (Val & bitc::NoSignedZeros)) + FMF.setNoSignedZeros(); + if (0 != (Val & bitc::AllowReciprocal)) + FMF.setAllowReciprocal(); + if (0 != (Val & bitc::AllowContract)) + FMF.setAllowContract(true); + if (0 != (Val & bitc::ApproxFunc)) + FMF.setApproxFunc(); + return FMF; +} + +static void upgradeDLLImportExportLinkage(GlobalValue *GV, unsigned Val) { + switch (Val) { + case 5: GV->setDLLStorageClass(GlobalValue::DLLImportStorageClass); break; + case 6: GV->setDLLStorageClass(GlobalValue::DLLExportStorageClass); break; + } +} + +Type *BitcodeReader::getTypeByID(unsigned ID) { + // The type table size is always specified correctly. + if (ID >= TypeList.size()) + return nullptr; + + if (Type *Ty = TypeList[ID]) + return Ty; + + // If we have a forward reference, the only possible case is when it is to a + // named struct. Just create a placeholder for now. + return TypeList[ID] = createIdentifiedStructType(Context); +} + +StructType *BitcodeReader::createIdentifiedStructType(LLVMContext &Context, + StringRef Name) { + auto *Ret = StructType::create(Context, Name); + IdentifiedStructTypes.push_back(Ret); + return Ret; +} + +StructType *BitcodeReader::createIdentifiedStructType(LLVMContext &Context) { + auto *Ret = StructType::create(Context); + IdentifiedStructTypes.push_back(Ret); + return Ret; +} + +//===----------------------------------------------------------------------===// +// Functions for parsing blocks from the bitcode file +//===----------------------------------------------------------------------===// + +static uint64_t getRawAttributeMask(Attribute::AttrKind Val) { + switch (Val) { + case Attribute::EndAttrKinds: + case Attribute::EmptyKey: + case Attribute::TombstoneKey: + llvm_unreachable("Synthetic enumerators which should never get here"); + + case Attribute::None: return 0; + case Attribute::ZExt: return 1 << 0; + case Attribute::SExt: return 1 << 1; + case Attribute::NoReturn: return 1 << 2; + case Attribute::InReg: return 1 << 3; + case Attribute::StructRet: return 1 << 4; + case Attribute::NoUnwind: return 1 << 5; + case Attribute::NoAlias: return 1 << 6; + case Attribute::ByVal: return 1 << 7; + case Attribute::Nest: return 1 << 8; + case Attribute::ReadNone: return 1 << 9; + case Attribute::ReadOnly: return 1 << 10; + case Attribute::NoInline: return 1 << 11; + case Attribute::AlwaysInline: return 1 << 12; + case Attribute::OptimizeForSize: return 1 << 13; + case Attribute::StackProtect: return 1 << 14; + case Attribute::StackProtectReq: return 1 << 15; + case Attribute::Alignment: return 31 << 16; + case Attribute::NoCapture: return 1 << 21; + case Attribute::NoRedZone: return 1 << 22; + case Attribute::NoImplicitFloat: return 1 << 23; + case Attribute::Naked: return 1 << 24; + case Attribute::InlineHint: return 1 << 25; + case Attribute::StackAlignment: return 7 << 26; + case Attribute::ReturnsTwice: return 1 << 29; + case Attribute::UWTable: return 1 << 30; + case Attribute::NonLazyBind: return 1U << 31; + case Attribute::SanitizeAddress: return 1ULL << 32; + case Attribute::MinSize: return 1ULL << 33; + case Attribute::NoDuplicate: return 1ULL << 34; + case Attribute::StackProtectStrong: return 1ULL << 35; + case Attribute::SanitizeThread: return 1ULL << 36; + case Attribute::SanitizeMemory: return 1ULL << 37; + case Attribute::NoBuiltin: return 1ULL << 38; + case Attribute::Returned: return 1ULL << 39; + case Attribute::Cold: return 1ULL << 40; + case Attribute::Builtin: return 1ULL << 41; + case Attribute::OptimizeNone: return 1ULL << 42; + case Attribute::InAlloca: return 1ULL << 43; + case Attribute::NonNull: return 1ULL << 44; + case Attribute::JumpTable: return 1ULL << 45; + case Attribute::Convergent: return 1ULL << 46; + case Attribute::SafeStack: return 1ULL << 47; + case Attribute::NoRecurse: return 1ULL << 48; + case Attribute::InaccessibleMemOnly: return 1ULL << 49; + case Attribute::InaccessibleMemOrArgMemOnly: return 1ULL << 50; + case Attribute::SwiftSelf: return 1ULL << 51; + case Attribute::SwiftError: return 1ULL << 52; + case Attribute::WriteOnly: return 1ULL << 53; + case Attribute::Speculatable: return 1ULL << 54; + case Attribute::StrictFP: return 1ULL << 55; + case Attribute::SanitizeHWAddress: return 1ULL << 56; + case Attribute::NoCfCheck: return 1ULL << 57; + case Attribute::OptForFuzzing: return 1ULL << 58; + case Attribute::ShadowCallStack: return 1ULL << 59; + case Attribute::SpeculativeLoadHardening: + return 1ULL << 60; + case Attribute::ImmArg: + return 1ULL << 61; + case Attribute::WillReturn: + return 1ULL << 62; + case Attribute::NoFree: + return 1ULL << 63; + default: + // Other attributes are not supported in the raw format, + // as we ran out of space. + return 0; + } + llvm_unreachable("Unsupported attribute type"); +} + +static void addRawAttributeValue(AttrBuilder &B, uint64_t Val) { + if (!Val) return; + + for (Attribute::AttrKind I = Attribute::None; I != Attribute::EndAttrKinds; + I = Attribute::AttrKind(I + 1)) { + if (uint64_t A = (Val & getRawAttributeMask(I))) { + if (I == Attribute::Alignment) + B.addAlignmentAttr(1ULL << ((A >> 16) - 1)); + else if (I == Attribute::StackAlignment) + B.addStackAlignmentAttr(1ULL << ((A >> 26)-1)); + else if (Attribute::isTypeAttrKind(I)) + B.addTypeAttr(I, nullptr); // Type will be auto-upgraded. + else + B.addAttribute(I); + } + } +} + +/// This fills an AttrBuilder object with the LLVM attributes that have +/// been decoded from the given integer. This function must stay in sync with +/// 'encodeLLVMAttributesForBitcode'. +static void decodeLLVMAttributesForBitcode(AttrBuilder &B, + uint64_t EncodedAttrs) { + // The alignment is stored as a 16-bit raw value from bits 31--16. We shift + // the bits above 31 down by 11 bits. + unsigned Alignment = (EncodedAttrs & (0xffffULL << 16)) >> 16; + assert((!Alignment || isPowerOf2_32(Alignment)) && + "Alignment must be a power of two."); + + if (Alignment) + B.addAlignmentAttr(Alignment); + addRawAttributeValue(B, ((EncodedAttrs & (0xfffffULL << 32)) >> 11) | + (EncodedAttrs & 0xffff)); +} + +Error BitcodeReader::parseAttributeBlock() { + if (Error Err = Stream.EnterSubBlock(bitc::PARAMATTR_BLOCK_ID)) + return Err; + + if (!MAttributes.empty()) + return error("Invalid multiple blocks"); + + SmallVector<uint64_t, 64> Record; + + SmallVector<AttributeList, 8> Attrs; + + // Read all the records. + while (true) { + Expected<BitstreamEntry> MaybeEntry = Stream.advanceSkippingSubblocks(); + if (!MaybeEntry) + return MaybeEntry.takeError(); + BitstreamEntry Entry = MaybeEntry.get(); + + switch (Entry.Kind) { + case BitstreamEntry::SubBlock: // Handled for us already. + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + return Error::success(); + case BitstreamEntry::Record: + // The interesting case. + break; + } + + // Read a record. + Record.clear(); + Expected<unsigned> MaybeRecord = Stream.readRecord(Entry.ID, Record); + if (!MaybeRecord) + return MaybeRecord.takeError(); + switch (MaybeRecord.get()) { + default: // Default behavior: ignore. + break; + case bitc::PARAMATTR_CODE_ENTRY_OLD: // ENTRY: [paramidx0, attr0, ...] + // Deprecated, but still needed to read old bitcode files. + if (Record.size() & 1) + return error("Invalid record"); + + for (unsigned i = 0, e = Record.size(); i != e; i += 2) { + AttrBuilder B(Context); + decodeLLVMAttributesForBitcode(B, Record[i+1]); + Attrs.push_back(AttributeList::get(Context, Record[i], B)); + } + + MAttributes.push_back(AttributeList::get(Context, Attrs)); + Attrs.clear(); + break; + case bitc::PARAMATTR_CODE_ENTRY: // ENTRY: [attrgrp0, attrgrp1, ...] + for (unsigned i = 0, e = Record.size(); i != e; ++i) + Attrs.push_back(MAttributeGroups[Record[i]]); + + MAttributes.push_back(AttributeList::get(Context, Attrs)); + Attrs.clear(); + break; + } + } +} + +// Returns Attribute::None on unrecognized codes. +static Attribute::AttrKind getAttrFromCode(uint64_t Code) { + switch (Code) { + default: + return Attribute::None; + case bitc::ATTR_KIND_ALIGNMENT: + return Attribute::Alignment; + case bitc::ATTR_KIND_ALWAYS_INLINE: + return Attribute::AlwaysInline; + case bitc::ATTR_KIND_ARGMEMONLY: + return Attribute::ArgMemOnly; + case bitc::ATTR_KIND_BUILTIN: + return Attribute::Builtin; + case bitc::ATTR_KIND_BY_VAL: + return Attribute::ByVal; + case bitc::ATTR_KIND_IN_ALLOCA: + return Attribute::InAlloca; + case bitc::ATTR_KIND_COLD: + return Attribute::Cold; + case bitc::ATTR_KIND_CONVERGENT: + return Attribute::Convergent; + case bitc::ATTR_KIND_DISABLE_SANITIZER_INSTRUMENTATION: + return Attribute::DisableSanitizerInstrumentation; + case bitc::ATTR_KIND_ELEMENTTYPE: + return Attribute::ElementType; + case bitc::ATTR_KIND_INACCESSIBLEMEM_ONLY: + return Attribute::InaccessibleMemOnly; + case bitc::ATTR_KIND_INACCESSIBLEMEM_OR_ARGMEMONLY: + return Attribute::InaccessibleMemOrArgMemOnly; + case bitc::ATTR_KIND_INLINE_HINT: + return Attribute::InlineHint; + case bitc::ATTR_KIND_IN_REG: + return Attribute::InReg; + case bitc::ATTR_KIND_JUMP_TABLE: + return Attribute::JumpTable; + case bitc::ATTR_KIND_MIN_SIZE: + return Attribute::MinSize; + case bitc::ATTR_KIND_NAKED: + return Attribute::Naked; + case bitc::ATTR_KIND_NEST: + return Attribute::Nest; + case bitc::ATTR_KIND_NO_ALIAS: + return Attribute::NoAlias; + case bitc::ATTR_KIND_NO_BUILTIN: + return Attribute::NoBuiltin; + case bitc::ATTR_KIND_NO_CALLBACK: + return Attribute::NoCallback; + case bitc::ATTR_KIND_NO_CAPTURE: + return Attribute::NoCapture; + case bitc::ATTR_KIND_NO_DUPLICATE: + return Attribute::NoDuplicate; + case bitc::ATTR_KIND_NOFREE: + return Attribute::NoFree; + case bitc::ATTR_KIND_NO_IMPLICIT_FLOAT: + return Attribute::NoImplicitFloat; + case bitc::ATTR_KIND_NO_INLINE: + return Attribute::NoInline; + case bitc::ATTR_KIND_NO_RECURSE: + return Attribute::NoRecurse; + case bitc::ATTR_KIND_NO_MERGE: + return Attribute::NoMerge; + case bitc::ATTR_KIND_NON_LAZY_BIND: + return Attribute::NonLazyBind; + case bitc::ATTR_KIND_NON_NULL: + return Attribute::NonNull; + case bitc::ATTR_KIND_DEREFERENCEABLE: + return Attribute::Dereferenceable; + case bitc::ATTR_KIND_DEREFERENCEABLE_OR_NULL: + return Attribute::DereferenceableOrNull; + case bitc::ATTR_KIND_ALLOC_SIZE: + return Attribute::AllocSize; + case bitc::ATTR_KIND_NO_RED_ZONE: + return Attribute::NoRedZone; + case bitc::ATTR_KIND_NO_RETURN: + return Attribute::NoReturn; + case bitc::ATTR_KIND_NOSYNC: + return Attribute::NoSync; + case bitc::ATTR_KIND_NOCF_CHECK: + return Attribute::NoCfCheck; + case bitc::ATTR_KIND_NO_PROFILE: + return Attribute::NoProfile; + case bitc::ATTR_KIND_NO_UNWIND: + return Attribute::NoUnwind; + case bitc::ATTR_KIND_NO_SANITIZE_COVERAGE: + return Attribute::NoSanitizeCoverage; + case bitc::ATTR_KIND_NULL_POINTER_IS_VALID: + return Attribute::NullPointerIsValid; + case bitc::ATTR_KIND_OPT_FOR_FUZZING: + return Attribute::OptForFuzzing; + case bitc::ATTR_KIND_OPTIMIZE_FOR_SIZE: + return Attribute::OptimizeForSize; + case bitc::ATTR_KIND_OPTIMIZE_NONE: + return Attribute::OptimizeNone; + case bitc::ATTR_KIND_READ_NONE: + return Attribute::ReadNone; + case bitc::ATTR_KIND_READ_ONLY: + return Attribute::ReadOnly; + case bitc::ATTR_KIND_RETURNED: + return Attribute::Returned; + case bitc::ATTR_KIND_RETURNS_TWICE: + return Attribute::ReturnsTwice; + case bitc::ATTR_KIND_S_EXT: + return Attribute::SExt; + case bitc::ATTR_KIND_SPECULATABLE: + return Attribute::Speculatable; + case bitc::ATTR_KIND_STACK_ALIGNMENT: + return Attribute::StackAlignment; + case bitc::ATTR_KIND_STACK_PROTECT: + return Attribute::StackProtect; + case bitc::ATTR_KIND_STACK_PROTECT_REQ: + return Attribute::StackProtectReq; + case bitc::ATTR_KIND_STACK_PROTECT_STRONG: + return Attribute::StackProtectStrong; + case bitc::ATTR_KIND_SAFESTACK: + return Attribute::SafeStack; + case bitc::ATTR_KIND_SHADOWCALLSTACK: + return Attribute::ShadowCallStack; + case bitc::ATTR_KIND_STRICT_FP: + return Attribute::StrictFP; + case bitc::ATTR_KIND_STRUCT_RET: + return Attribute::StructRet; + case bitc::ATTR_KIND_SANITIZE_ADDRESS: + return Attribute::SanitizeAddress; + case bitc::ATTR_KIND_SANITIZE_HWADDRESS: + return Attribute::SanitizeHWAddress; + case bitc::ATTR_KIND_SANITIZE_THREAD: + return Attribute::SanitizeThread; + case bitc::ATTR_KIND_SANITIZE_MEMORY: + return Attribute::SanitizeMemory; + case bitc::ATTR_KIND_SPECULATIVE_LOAD_HARDENING: + return Attribute::SpeculativeLoadHardening; + case bitc::ATTR_KIND_SWIFT_ERROR: + return Attribute::SwiftError; + case bitc::ATTR_KIND_SWIFT_SELF: + return Attribute::SwiftSelf; + case bitc::ATTR_KIND_SWIFT_ASYNC: + return Attribute::SwiftAsync; + case bitc::ATTR_KIND_UW_TABLE: + return Attribute::UWTable; + case bitc::ATTR_KIND_VSCALE_RANGE: + return Attribute::VScaleRange; + case bitc::ATTR_KIND_WILLRETURN: + return Attribute::WillReturn; + case bitc::ATTR_KIND_WRITEONLY: + return Attribute::WriteOnly; + case bitc::ATTR_KIND_Z_EXT: + return Attribute::ZExt; + case bitc::ATTR_KIND_IMMARG: + return Attribute::ImmArg; + case bitc::ATTR_KIND_SANITIZE_MEMTAG: + return Attribute::SanitizeMemTag; + case bitc::ATTR_KIND_PREALLOCATED: + return Attribute::Preallocated; + case bitc::ATTR_KIND_NOUNDEF: + return Attribute::NoUndef; + case bitc::ATTR_KIND_BYREF: + return Attribute::ByRef; + case bitc::ATTR_KIND_MUSTPROGRESS: + return Attribute::MustProgress; + case bitc::ATTR_KIND_HOT: + return Attribute::Hot; + } +} + +Error BitcodeReader::parseAlignmentValue(uint64_t Exponent, + MaybeAlign &Alignment) { + // Note: Alignment in bitcode files is incremented by 1, so that zero + // can be used for default alignment. + if (Exponent > Value::MaxAlignmentExponent + 1) + return error("Invalid alignment value"); + Alignment = decodeMaybeAlign(Exponent); + return Error::success(); +} + +Error BitcodeReader::parseAttrKind(uint64_t Code, Attribute::AttrKind *Kind) { + *Kind = getAttrFromCode(Code); + if (*Kind == Attribute::None) + return error("Unknown attribute kind (" + Twine(Code) + ")"); + return Error::success(); +} + +Error BitcodeReader::parseAttributeGroupBlock() { + if (Error Err = Stream.EnterSubBlock(bitc::PARAMATTR_GROUP_BLOCK_ID)) + return Err; + + if (!MAttributeGroups.empty()) + return error("Invalid multiple blocks"); + + SmallVector<uint64_t, 64> Record; + + // Read all the records. + while (true) { + Expected<BitstreamEntry> MaybeEntry = Stream.advanceSkippingSubblocks(); + if (!MaybeEntry) + return MaybeEntry.takeError(); + BitstreamEntry Entry = MaybeEntry.get(); + + switch (Entry.Kind) { + case BitstreamEntry::SubBlock: // Handled for us already. + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + return Error::success(); + case BitstreamEntry::Record: + // The interesting case. + break; + } + + // Read a record. + Record.clear(); + Expected<unsigned> MaybeRecord = Stream.readRecord(Entry.ID, Record); + if (!MaybeRecord) + return MaybeRecord.takeError(); + switch (MaybeRecord.get()) { + default: // Default behavior: ignore. + break; + case bitc::PARAMATTR_GRP_CODE_ENTRY: { // ENTRY: [grpid, idx, a0, a1, ...] + if (Record.size() < 3) + return error("Invalid record"); + + uint64_t GrpID = Record[0]; + uint64_t Idx = Record[1]; // Index of the object this attribute refers to. + + AttrBuilder B(Context); + for (unsigned i = 2, e = Record.size(); i != e; ++i) { + if (Record[i] == 0) { // Enum attribute + Attribute::AttrKind Kind; + if (Error Err = parseAttrKind(Record[++i], &Kind)) + return Err; + + // Upgrade old-style byval attribute to one with a type, even if it's + // nullptr. We will have to insert the real type when we associate + // this AttributeList with a function. + if (Kind == Attribute::ByVal) + B.addByValAttr(nullptr); + else if (Kind == Attribute::StructRet) + B.addStructRetAttr(nullptr); + else if (Kind == Attribute::InAlloca) + B.addInAllocaAttr(nullptr); + else if (Attribute::isEnumAttrKind(Kind)) + B.addAttribute(Kind); + else + return error("Not an enum attribute"); + } else if (Record[i] == 1) { // Integer attribute + Attribute::AttrKind Kind; + if (Error Err = parseAttrKind(Record[++i], &Kind)) + return Err; + if (!Attribute::isIntAttrKind(Kind)) + return error("Not an int attribute"); + if (Kind == Attribute::Alignment) + B.addAlignmentAttr(Record[++i]); + else if (Kind == Attribute::StackAlignment) + B.addStackAlignmentAttr(Record[++i]); + else if (Kind == Attribute::Dereferenceable) + B.addDereferenceableAttr(Record[++i]); + else if (Kind == Attribute::DereferenceableOrNull) + B.addDereferenceableOrNullAttr(Record[++i]); + else if (Kind == Attribute::AllocSize) + B.addAllocSizeAttrFromRawRepr(Record[++i]); + else if (Kind == Attribute::VScaleRange) + B.addVScaleRangeAttrFromRawRepr(Record[++i]); + } else if (Record[i] == 3 || Record[i] == 4) { // String attribute + bool HasValue = (Record[i++] == 4); + SmallString<64> KindStr; + SmallString<64> ValStr; + + while (Record[i] != 0 && i != e) + KindStr += Record[i++]; + assert(Record[i] == 0 && "Kind string not null terminated"); + + if (HasValue) { + // Has a value associated with it. + ++i; // Skip the '0' that terminates the "kind" string. + while (Record[i] != 0 && i != e) + ValStr += Record[i++]; + assert(Record[i] == 0 && "Value string not null terminated"); + } + + B.addAttribute(KindStr.str(), ValStr.str()); + } else { + assert((Record[i] == 5 || Record[i] == 6) && + "Invalid attribute group entry"); + bool HasType = Record[i] == 6; + Attribute::AttrKind Kind; + if (Error Err = parseAttrKind(Record[++i], &Kind)) + return Err; + if (!Attribute::isTypeAttrKind(Kind)) + return error("Not a type attribute"); + + B.addTypeAttr(Kind, HasType ? getTypeByID(Record[++i]) : nullptr); + } + } + + UpgradeAttributes(B); + MAttributeGroups[GrpID] = AttributeList::get(Context, Idx, B); + break; + } + } + } +} + +Error BitcodeReader::parseTypeTable() { + if (Error Err = Stream.EnterSubBlock(bitc::TYPE_BLOCK_ID_NEW)) + return Err; + + return parseTypeTableBody(); +} + +Error BitcodeReader::parseTypeTableBody() { + if (!TypeList.empty()) + return error("Invalid multiple blocks"); + + SmallVector<uint64_t, 64> Record; + unsigned NumRecords = 0; + + SmallString<64> TypeName; + + // Read all the records for this type table. + while (true) { + Expected<BitstreamEntry> MaybeEntry = Stream.advanceSkippingSubblocks(); + if (!MaybeEntry) + return MaybeEntry.takeError(); + BitstreamEntry Entry = MaybeEntry.get(); + + switch (Entry.Kind) { + case BitstreamEntry::SubBlock: // Handled for us already. + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + if (NumRecords != TypeList.size()) + return error("Malformed block"); + return Error::success(); + case BitstreamEntry::Record: + // The interesting case. + break; + } + + // Read a record. + Record.clear(); + Type *ResultTy = nullptr; + Expected<unsigned> MaybeRecord = Stream.readRecord(Entry.ID, Record); + if (!MaybeRecord) + return MaybeRecord.takeError(); + switch (MaybeRecord.get()) { + default: + return error("Invalid value"); + case bitc::TYPE_CODE_NUMENTRY: // TYPE_CODE_NUMENTRY: [numentries] + // TYPE_CODE_NUMENTRY contains a count of the number of types in the + // type list. This allows us to reserve space. + if (Record.empty()) + return error("Invalid record"); + TypeList.resize(Record[0]); + continue; + case bitc::TYPE_CODE_VOID: // VOID + ResultTy = Type::getVoidTy(Context); + break; + case bitc::TYPE_CODE_HALF: // HALF + ResultTy = Type::getHalfTy(Context); + break; + case bitc::TYPE_CODE_BFLOAT: // BFLOAT + ResultTy = Type::getBFloatTy(Context); + break; + case bitc::TYPE_CODE_FLOAT: // FLOAT + ResultTy = Type::getFloatTy(Context); + break; + case bitc::TYPE_CODE_DOUBLE: // DOUBLE + ResultTy = Type::getDoubleTy(Context); + break; + case bitc::TYPE_CODE_X86_FP80: // X86_FP80 + ResultTy = Type::getX86_FP80Ty(Context); + break; + case bitc::TYPE_CODE_FP128: // FP128 + ResultTy = Type::getFP128Ty(Context); + break; + case bitc::TYPE_CODE_PPC_FP128: // PPC_FP128 + ResultTy = Type::getPPC_FP128Ty(Context); + break; + case bitc::TYPE_CODE_LABEL: // LABEL + ResultTy = Type::getLabelTy(Context); + break; + case bitc::TYPE_CODE_METADATA: // METADATA + ResultTy = Type::getMetadataTy(Context); + break; + case bitc::TYPE_CODE_X86_MMX: // X86_MMX + ResultTy = Type::getX86_MMXTy(Context); + break; + case bitc::TYPE_CODE_X86_AMX: // X86_AMX + ResultTy = Type::getX86_AMXTy(Context); + break; + case bitc::TYPE_CODE_TOKEN: // TOKEN + ResultTy = Type::getTokenTy(Context); + break; + case bitc::TYPE_CODE_INTEGER: { // INTEGER: [width] + if (Record.empty()) + return error("Invalid record"); + + uint64_t NumBits = Record[0]; + if (NumBits < IntegerType::MIN_INT_BITS || + NumBits > IntegerType::MAX_INT_BITS) + return error("Bitwidth for integer type out of range"); + ResultTy = IntegerType::get(Context, NumBits); + break; + } + case bitc::TYPE_CODE_POINTER: { // POINTER: [pointee type] or + // [pointee type, address space] + if (Record.empty()) + return error("Invalid record"); + unsigned AddressSpace = 0; + if (Record.size() == 2) + AddressSpace = Record[1]; + ResultTy = getTypeByID(Record[0]); + if (!ResultTy || + !PointerType::isValidElementType(ResultTy)) + return error("Invalid type"); + ResultTy = PointerType::get(ResultTy, AddressSpace); + break; + } + case bitc::TYPE_CODE_OPAQUE_POINTER: { // OPAQUE_POINTER: [addrspace] + if (Record.size() != 1) + return error("Invalid record"); + if (Context.supportsTypedPointers()) + return error( + "Opaque pointers are only supported in -opaque-pointers mode"); + unsigned AddressSpace = Record[0]; + ResultTy = PointerType::get(Context, AddressSpace); + break; + } + case bitc::TYPE_CODE_FUNCTION_OLD: { + // Deprecated, but still needed to read old bitcode files. + // FUNCTION: [vararg, attrid, retty, paramty x N] + if (Record.size() < 3) + return error("Invalid record"); + SmallVector<Type*, 8> ArgTys; + for (unsigned i = 3, e = Record.size(); i != e; ++i) { + if (Type *T = getTypeByID(Record[i])) + ArgTys.push_back(T); + else + break; + } + + ResultTy = getTypeByID(Record[2]); + if (!ResultTy || ArgTys.size() < Record.size()-3) + return error("Invalid type"); + + ResultTy = FunctionType::get(ResultTy, ArgTys, Record[0]); + break; + } + case bitc::TYPE_CODE_FUNCTION: { + // FUNCTION: [vararg, retty, paramty x N] + if (Record.size() < 2) + return error("Invalid record"); + SmallVector<Type*, 8> ArgTys; + for (unsigned i = 2, e = Record.size(); i != e; ++i) { + if (Type *T = getTypeByID(Record[i])) { + if (!FunctionType::isValidArgumentType(T)) + return error("Invalid function argument type"); + ArgTys.push_back(T); + } + else + break; + } + + ResultTy = getTypeByID(Record[1]); + if (!ResultTy || ArgTys.size() < Record.size()-2) + return error("Invalid type"); + + ResultTy = FunctionType::get(ResultTy, ArgTys, Record[0]); + break; + } + case bitc::TYPE_CODE_STRUCT_ANON: { // STRUCT: [ispacked, eltty x N] + if (Record.empty()) + return error("Invalid record"); + SmallVector<Type*, 8> EltTys; + for (unsigned i = 1, e = Record.size(); i != e; ++i) { + if (Type *T = getTypeByID(Record[i])) + EltTys.push_back(T); + else + break; + } + if (EltTys.size() != Record.size()-1) + return error("Invalid type"); + ResultTy = StructType::get(Context, EltTys, Record[0]); + break; + } + case bitc::TYPE_CODE_STRUCT_NAME: // STRUCT_NAME: [strchr x N] + if (convertToString(Record, 0, TypeName)) + return error("Invalid record"); + continue; + + case bitc::TYPE_CODE_STRUCT_NAMED: { // STRUCT: [ispacked, eltty x N] + if (Record.empty()) + return error("Invalid record"); + + if (NumRecords >= TypeList.size()) + return error("Invalid TYPE table"); + + // Check to see if this was forward referenced, if so fill in the temp. + StructType *Res = cast_or_null<StructType>(TypeList[NumRecords]); + if (Res) { + Res->setName(TypeName); + TypeList[NumRecords] = nullptr; + } else // Otherwise, create a new struct. + Res = createIdentifiedStructType(Context, TypeName); + TypeName.clear(); + + SmallVector<Type*, 8> EltTys; + for (unsigned i = 1, e = Record.size(); i != e; ++i) { + if (Type *T = getTypeByID(Record[i])) + EltTys.push_back(T); + else + break; + } + if (EltTys.size() != Record.size()-1) + return error("Invalid record"); + Res->setBody(EltTys, Record[0]); + ResultTy = Res; + break; + } + case bitc::TYPE_CODE_OPAQUE: { // OPAQUE: [] + if (Record.size() != 1) + return error("Invalid record"); + + if (NumRecords >= TypeList.size()) + return error("Invalid TYPE table"); + + // Check to see if this was forward referenced, if so fill in the temp. + StructType *Res = cast_or_null<StructType>(TypeList[NumRecords]); + if (Res) { + Res->setName(TypeName); + TypeList[NumRecords] = nullptr; + } else // Otherwise, create a new struct with no body. + Res = createIdentifiedStructType(Context, TypeName); + TypeName.clear(); + ResultTy = Res; + break; + } + case bitc::TYPE_CODE_ARRAY: // ARRAY: [numelts, eltty] + if (Record.size() < 2) + return error("Invalid record"); + ResultTy = getTypeByID(Record[1]); + if (!ResultTy || !ArrayType::isValidElementType(ResultTy)) + return error("Invalid type"); + ResultTy = ArrayType::get(ResultTy, Record[0]); + break; + case bitc::TYPE_CODE_VECTOR: // VECTOR: [numelts, eltty] or + // [numelts, eltty, scalable] + if (Record.size() < 2) + return error("Invalid record"); + if (Record[0] == 0) + return error("Invalid vector length"); + ResultTy = getTypeByID(Record[1]); + if (!ResultTy || !VectorType::isValidElementType(ResultTy)) + return error("Invalid type"); + bool Scalable = Record.size() > 2 ? Record[2] : false; + ResultTy = VectorType::get(ResultTy, Record[0], Scalable); + break; + } + + if (NumRecords >= TypeList.size()) + return error("Invalid TYPE table"); + if (TypeList[NumRecords]) + return error( + "Invalid TYPE table: Only named structs can be forward referenced"); + assert(ResultTy && "Didn't read a type?"); + TypeList[NumRecords++] = ResultTy; + } +} + +Error BitcodeReader::parseOperandBundleTags() { + if (Error Err = Stream.EnterSubBlock(bitc::OPERAND_BUNDLE_TAGS_BLOCK_ID)) + return Err; + + if (!BundleTags.empty()) + return error("Invalid multiple blocks"); + + SmallVector<uint64_t, 64> Record; + + while (true) { + Expected<BitstreamEntry> MaybeEntry = Stream.advanceSkippingSubblocks(); + if (!MaybeEntry) + return MaybeEntry.takeError(); + BitstreamEntry Entry = MaybeEntry.get(); + + switch (Entry.Kind) { + case BitstreamEntry::SubBlock: // Handled for us already. + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + return Error::success(); + case BitstreamEntry::Record: + // The interesting case. + break; + } + + // Tags are implicitly mapped to integers by their order. + + Expected<unsigned> MaybeRecord = Stream.readRecord(Entry.ID, Record); + if (!MaybeRecord) + return MaybeRecord.takeError(); + if (MaybeRecord.get() != bitc::OPERAND_BUNDLE_TAG) + return error("Invalid record"); + + // OPERAND_BUNDLE_TAG: [strchr x N] + BundleTags.emplace_back(); + if (convertToString(Record, 0, BundleTags.back())) + return error("Invalid record"); + Record.clear(); + } +} + +Error BitcodeReader::parseSyncScopeNames() { + if (Error Err = Stream.EnterSubBlock(bitc::SYNC_SCOPE_NAMES_BLOCK_ID)) + return Err; + + if (!SSIDs.empty()) + return error("Invalid multiple synchronization scope names blocks"); + + SmallVector<uint64_t, 64> Record; + while (true) { + Expected<BitstreamEntry> MaybeEntry = Stream.advanceSkippingSubblocks(); + if (!MaybeEntry) + return MaybeEntry.takeError(); + BitstreamEntry Entry = MaybeEntry.get(); + + switch (Entry.Kind) { + case BitstreamEntry::SubBlock: // Handled for us already. + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + if (SSIDs.empty()) + return error("Invalid empty synchronization scope names block"); + return Error::success(); + case BitstreamEntry::Record: + // The interesting case. + break; + } + + // Synchronization scope names are implicitly mapped to synchronization + // scope IDs by their order. + + Expected<unsigned> MaybeRecord = Stream.readRecord(Entry.ID, Record); + if (!MaybeRecord) + return MaybeRecord.takeError(); + if (MaybeRecord.get() != bitc::SYNC_SCOPE_NAME) + return error("Invalid record"); + + SmallString<16> SSN; + if (convertToString(Record, 0, SSN)) + return error("Invalid record"); + + SSIDs.push_back(Context.getOrInsertSyncScopeID(SSN)); + Record.clear(); + } +} + +/// Associate a value with its name from the given index in the provided record. +Expected<Value *> BitcodeReader::recordValue(SmallVectorImpl<uint64_t> &Record, + unsigned NameIndex, Triple &TT) { + SmallString<128> ValueName; + if (convertToString(Record, NameIndex, ValueName)) + return error("Invalid record"); + unsigned ValueID = Record[0]; + if (ValueID >= ValueList.size() || !ValueList[ValueID]) + return error("Invalid record"); + Value *V = ValueList[ValueID]; + + StringRef NameStr(ValueName.data(), ValueName.size()); + if (NameStr.find_first_of(0) != StringRef::npos) + return error("Invalid value name"); + V->setName(NameStr); + auto *GO = dyn_cast<GlobalObject>(V); + if (GO && ImplicitComdatObjects.contains(GO) && TT.supportsCOMDAT()) + GO->setComdat(TheModule->getOrInsertComdat(V->getName())); + return V; +} + +/// Helper to note and return the current location, and jump to the given +/// offset. +static Expected<uint64_t> jumpToValueSymbolTable(uint64_t Offset, + BitstreamCursor &Stream) { + // Save the current parsing location so we can jump back at the end + // of the VST read. + uint64_t CurrentBit = Stream.GetCurrentBitNo(); + if (Error JumpFailed = Stream.JumpToBit(Offset * 32)) + return std::move(JumpFailed); + Expected<BitstreamEntry> MaybeEntry = Stream.advance(); + if (!MaybeEntry) + return MaybeEntry.takeError(); + assert(MaybeEntry.get().Kind == BitstreamEntry::SubBlock); + assert(MaybeEntry.get().ID == bitc::VALUE_SYMTAB_BLOCK_ID); + return CurrentBit; +} + +void BitcodeReader::setDeferredFunctionInfo(unsigned FuncBitcodeOffsetDelta, + Function *F, + ArrayRef<uint64_t> Record) { + // Note that we subtract 1 here because the offset is relative to one word + // before the start of the identification or module block, which was + // historically always the start of the regular bitcode header. + uint64_t FuncWordOffset = Record[1] - 1; + uint64_t FuncBitOffset = FuncWordOffset * 32; + DeferredFunctionInfo[F] = FuncBitOffset + FuncBitcodeOffsetDelta; + // Set the LastFunctionBlockBit to point to the last function block. + // Later when parsing is resumed after function materialization, + // we can simply skip that last function block. + if (FuncBitOffset > LastFunctionBlockBit) + LastFunctionBlockBit = FuncBitOffset; +} + +/// Read a new-style GlobalValue symbol table. +Error BitcodeReader::parseGlobalValueSymbolTable() { + unsigned FuncBitcodeOffsetDelta = + Stream.getAbbrevIDWidth() + bitc::BlockIDWidth; + + if (Error Err = Stream.EnterSubBlock(bitc::VALUE_SYMTAB_BLOCK_ID)) + return Err; + + SmallVector<uint64_t, 64> Record; + while (true) { + Expected<BitstreamEntry> MaybeEntry = Stream.advanceSkippingSubblocks(); + if (!MaybeEntry) + return MaybeEntry.takeError(); + BitstreamEntry Entry = MaybeEntry.get(); + + switch (Entry.Kind) { + case BitstreamEntry::SubBlock: + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + return Error::success(); + case BitstreamEntry::Record: + break; + } + + Record.clear(); + Expected<unsigned> MaybeRecord = Stream.readRecord(Entry.ID, Record); + if (!MaybeRecord) + return MaybeRecord.takeError(); + switch (MaybeRecord.get()) { + case bitc::VST_CODE_FNENTRY: // [valueid, offset] + setDeferredFunctionInfo(FuncBitcodeOffsetDelta, + cast<Function>(ValueList[Record[0]]), Record); + break; + } + } +} + +/// Parse the value symbol table at either the current parsing location or +/// at the given bit offset if provided. +Error BitcodeReader::parseValueSymbolTable(uint64_t Offset) { + uint64_t CurrentBit; + // Pass in the Offset to distinguish between calling for the module-level + // VST (where we want to jump to the VST offset) and the function-level + // VST (where we don't). + if (Offset > 0) { + Expected<uint64_t> MaybeCurrentBit = jumpToValueSymbolTable(Offset, Stream); + if (!MaybeCurrentBit) + return MaybeCurrentBit.takeError(); + CurrentBit = MaybeCurrentBit.get(); + // If this module uses a string table, read this as a module-level VST. + if (UseStrtab) { + if (Error Err = parseGlobalValueSymbolTable()) + return Err; + if (Error JumpFailed = Stream.JumpToBit(CurrentBit)) + return JumpFailed; + return Error::success(); + } + // Otherwise, the VST will be in a similar format to a function-level VST, + // and will contain symbol names. + } + + // Compute the delta between the bitcode indices in the VST (the word offset + // to the word-aligned ENTER_SUBBLOCK for the function block, and that + // expected by the lazy reader. The reader's EnterSubBlock expects to have + // already read the ENTER_SUBBLOCK code (size getAbbrevIDWidth) and BlockID + // (size BlockIDWidth). Note that we access the stream's AbbrevID width here + // just before entering the VST subblock because: 1) the EnterSubBlock + // changes the AbbrevID width; 2) the VST block is nested within the same + // outer MODULE_BLOCK as the FUNCTION_BLOCKs and therefore have the same + // AbbrevID width before calling EnterSubBlock; and 3) when we want to + // jump to the FUNCTION_BLOCK using this offset later, we don't want + // to rely on the stream's AbbrevID width being that of the MODULE_BLOCK. + unsigned FuncBitcodeOffsetDelta = + Stream.getAbbrevIDWidth() + bitc::BlockIDWidth; + + if (Error Err = Stream.EnterSubBlock(bitc::VALUE_SYMTAB_BLOCK_ID)) + return Err; + + SmallVector<uint64_t, 64> Record; + + Triple TT(TheModule->getTargetTriple()); + + // Read all the records for this value table. + SmallString<128> ValueName; + + while (true) { + Expected<BitstreamEntry> MaybeEntry = Stream.advanceSkippingSubblocks(); + if (!MaybeEntry) + return MaybeEntry.takeError(); + BitstreamEntry Entry = MaybeEntry.get(); + + switch (Entry.Kind) { + case BitstreamEntry::SubBlock: // Handled for us already. + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + if (Offset > 0) + if (Error JumpFailed = Stream.JumpToBit(CurrentBit)) + return JumpFailed; + return Error::success(); + case BitstreamEntry::Record: + // The interesting case. + break; + } + + // Read a record. + Record.clear(); + Expected<unsigned> MaybeRecord = Stream.readRecord(Entry.ID, Record); + if (!MaybeRecord) + return MaybeRecord.takeError(); + switch (MaybeRecord.get()) { + default: // Default behavior: unknown type. + break; + case bitc::VST_CODE_ENTRY: { // VST_CODE_ENTRY: [valueid, namechar x N] + Expected<Value *> ValOrErr = recordValue(Record, 1, TT); + if (Error Err = ValOrErr.takeError()) + return Err; + ValOrErr.get(); + break; + } + case bitc::VST_CODE_FNENTRY: { + // VST_CODE_FNENTRY: [valueid, offset, namechar x N] + Expected<Value *> ValOrErr = recordValue(Record, 2, TT); + if (Error Err = ValOrErr.takeError()) + return Err; + Value *V = ValOrErr.get(); + + // Ignore function offsets emitted for aliases of functions in older + // versions of LLVM. + if (auto *F = dyn_cast<Function>(V)) + setDeferredFunctionInfo(FuncBitcodeOffsetDelta, F, Record); + break; + } + case bitc::VST_CODE_BBENTRY: { + if (convertToString(Record, 1, ValueName)) + return error("Invalid record"); + BasicBlock *BB = getBasicBlock(Record[0]); + if (!BB) + return error("Invalid record"); + + BB->setName(StringRef(ValueName.data(), ValueName.size())); + ValueName.clear(); + break; + } + } + } +} + +/// Decode a signed value stored with the sign bit in the LSB for dense VBR +/// encoding. +uint64_t BitcodeReader::decodeSignRotatedValue(uint64_t V) { + if ((V & 1) == 0) + return V >> 1; + if (V != 1) + return -(V >> 1); + // There is no such thing as -0 with integers. "-0" really means MININT. + return 1ULL << 63; +} + +/// Resolve all of the initializers for global values and aliases that we can. +Error BitcodeReader::resolveGlobalAndIndirectSymbolInits() { + std::vector<std::pair<GlobalVariable *, unsigned>> GlobalInitWorklist; + std::vector<std::pair<GlobalValue *, unsigned>> IndirectSymbolInitWorklist; + std::vector<FunctionOperandInfo> FunctionOperandWorklist; + + GlobalInitWorklist.swap(GlobalInits); + IndirectSymbolInitWorklist.swap(IndirectSymbolInits); + FunctionOperandWorklist.swap(FunctionOperands); + + while (!GlobalInitWorklist.empty()) { + unsigned ValID = GlobalInitWorklist.back().second; + if (ValID >= ValueList.size()) { + // Not ready to resolve this yet, it requires something later in the file. + GlobalInits.push_back(GlobalInitWorklist.back()); + } else { + if (Constant *C = dyn_cast_or_null<Constant>(ValueList[ValID])) + GlobalInitWorklist.back().first->setInitializer(C); + else + return error("Expected a constant"); + } + GlobalInitWorklist.pop_back(); + } + + while (!IndirectSymbolInitWorklist.empty()) { + unsigned ValID = IndirectSymbolInitWorklist.back().second; + if (ValID >= ValueList.size()) { + IndirectSymbolInits.push_back(IndirectSymbolInitWorklist.back()); + } else { + Constant *C = dyn_cast_or_null<Constant>(ValueList[ValID]); + if (!C) + return error("Expected a constant"); + GlobalValue *GV = IndirectSymbolInitWorklist.back().first; + if (auto *GA = dyn_cast<GlobalAlias>(GV)) { + if (C->getType() != GV->getType()) + return error("Alias and aliasee types don't match"); + GA->setAliasee(C); + } else if (auto *GI = dyn_cast<GlobalIFunc>(GV)) { + Type *ResolverFTy = + GlobalIFunc::getResolverFunctionType(GI->getValueType()); + // Transparently fix up the type for compatiblity with older bitcode + GI->setResolver( + ConstantExpr::getBitCast(C, ResolverFTy->getPointerTo())); + } else { + return error("Expected an alias or an ifunc"); + } + } + IndirectSymbolInitWorklist.pop_back(); + } + + while (!FunctionOperandWorklist.empty()) { + FunctionOperandInfo &Info = FunctionOperandWorklist.back(); + if (Info.PersonalityFn) { + unsigned ValID = Info.PersonalityFn - 1; + if (ValID < ValueList.size()) { + if (Constant *C = dyn_cast_or_null<Constant>(ValueList[ValID])) + Info.F->setPersonalityFn(C); + else + return error("Expected a constant"); + Info.PersonalityFn = 0; + } + } + if (Info.Prefix) { + unsigned ValID = Info.Prefix - 1; + if (ValID < ValueList.size()) { + if (Constant *C = dyn_cast_or_null<Constant>(ValueList[ValID])) + Info.F->setPrefixData(C); + else + return error("Expected a constant"); + Info.Prefix = 0; + } + } + if (Info.Prologue) { + unsigned ValID = Info.Prologue - 1; + if (ValID < ValueList.size()) { + if (Constant *C = dyn_cast_or_null<Constant>(ValueList[ValID])) + Info.F->setPrologueData(C); + else + return error("Expected a constant"); + Info.Prologue = 0; + } + } + if (Info.PersonalityFn || Info.Prefix || Info.Prologue) + FunctionOperands.push_back(Info); + FunctionOperandWorklist.pop_back(); + } + + return Error::success(); +} + +APInt llvm::readWideAPInt(ArrayRef<uint64_t> Vals, unsigned TypeBits) { + SmallVector<uint64_t, 8> Words(Vals.size()); + transform(Vals, Words.begin(), + BitcodeReader::decodeSignRotatedValue); + + return APInt(TypeBits, Words); +} + +Error BitcodeReader::parseConstants() { + if (Error Err = Stream.EnterSubBlock(bitc::CONSTANTS_BLOCK_ID)) + return Err; + + SmallVector<uint64_t, 64> Record; + + // Read all the records for this value table. + Type *CurTy = Type::getInt32Ty(Context); + unsigned NextCstNo = ValueList.size(); + + struct DelayedShufTy { + VectorType *OpTy; + VectorType *RTy; + uint64_t Op0Idx; + uint64_t Op1Idx; + uint64_t Op2Idx; + unsigned CstNo; + }; + std::vector<DelayedShufTy> DelayedShuffles; + struct DelayedSelTy { + Type *OpTy; + uint64_t Op0Idx; + uint64_t Op1Idx; + uint64_t Op2Idx; + unsigned CstNo; + }; + std::vector<DelayedSelTy> DelayedSelectors; + + while (true) { + Expected<BitstreamEntry> MaybeEntry = Stream.advanceSkippingSubblocks(); + if (!MaybeEntry) + return MaybeEntry.takeError(); + BitstreamEntry Entry = MaybeEntry.get(); + + switch (Entry.Kind) { + case BitstreamEntry::SubBlock: // Handled for us already. + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + // Once all the constants have been read, go through and resolve forward + // references. + // + // We have to treat shuffles specially because they don't have three + // operands anymore. We need to convert the shuffle mask into an array, + // and we can't convert a forward reference. + for (auto &DelayedShuffle : DelayedShuffles) { + VectorType *OpTy = DelayedShuffle.OpTy; + VectorType *RTy = DelayedShuffle.RTy; + uint64_t Op0Idx = DelayedShuffle.Op0Idx; + uint64_t Op1Idx = DelayedShuffle.Op1Idx; + uint64_t Op2Idx = DelayedShuffle.Op2Idx; + uint64_t CstNo = DelayedShuffle.CstNo; + Constant *Op0 = ValueList.getConstantFwdRef(Op0Idx, OpTy); + Constant *Op1 = ValueList.getConstantFwdRef(Op1Idx, OpTy); + Type *ShufTy = + VectorType::get(Type::getInt32Ty(Context), RTy->getElementCount()); + Constant *Op2 = ValueList.getConstantFwdRef(Op2Idx, ShufTy); + if (!ShuffleVectorInst::isValidOperands(Op0, Op1, Op2)) + return error("Invalid shufflevector operands"); + SmallVector<int, 16> Mask; + ShuffleVectorInst::getShuffleMask(Op2, Mask); + Value *V = ConstantExpr::getShuffleVector(Op0, Op1, Mask); + ValueList.assignValue(V, CstNo); + } + for (auto &DelayedSelector : DelayedSelectors) { + Type *OpTy = DelayedSelector.OpTy; + Type *SelectorTy = Type::getInt1Ty(Context); + uint64_t Op0Idx = DelayedSelector.Op0Idx; + uint64_t Op1Idx = DelayedSelector.Op1Idx; + uint64_t Op2Idx = DelayedSelector.Op2Idx; + uint64_t CstNo = DelayedSelector.CstNo; + Constant *Op1 = ValueList.getConstantFwdRef(Op1Idx, OpTy); + Constant *Op2 = ValueList.getConstantFwdRef(Op2Idx, OpTy); + // The selector might be an i1 or an <n x i1> + // Get the type from the ValueList before getting a forward ref. + if (VectorType *VTy = dyn_cast<VectorType>(OpTy)) { + Value *V = ValueList[Op0Idx]; + assert(V); + if (SelectorTy != V->getType()) + SelectorTy = VectorType::get(SelectorTy, VTy->getElementCount()); + } + Constant *Op0 = ValueList.getConstantFwdRef(Op0Idx, SelectorTy); + Value *V = ConstantExpr::getSelect(Op0, Op1, Op2); + ValueList.assignValue(V, CstNo); + } + + if (NextCstNo != ValueList.size()) + return error("Invalid constant reference"); + + ValueList.resolveConstantForwardRefs(); + return Error::success(); + case BitstreamEntry::Record: + // The interesting case. + break; + } + + // Read a record. + Record.clear(); + Type *VoidType = Type::getVoidTy(Context); + Value *V = nullptr; + Expected<unsigned> MaybeBitCode = Stream.readRecord(Entry.ID, Record); + if (!MaybeBitCode) + return MaybeBitCode.takeError(); + switch (unsigned BitCode = MaybeBitCode.get()) { + default: // Default behavior: unknown constant + case bitc::CST_CODE_UNDEF: // UNDEF + V = UndefValue::get(CurTy); + break; + case bitc::CST_CODE_POISON: // POISON + V = PoisonValue::get(CurTy); + break; + case bitc::CST_CODE_SETTYPE: // SETTYPE: [typeid] + if (Record.empty()) + return error("Invalid record"); + if (Record[0] >= TypeList.size() || !TypeList[Record[0]]) + return error("Invalid record"); + if (TypeList[Record[0]] == VoidType) + return error("Invalid constant type"); + CurTy = TypeList[Record[0]]; + continue; // Skip the ValueList manipulation. + case bitc::CST_CODE_NULL: // NULL + if (CurTy->isVoidTy() || CurTy->isFunctionTy() || CurTy->isLabelTy()) + return error("Invalid type for a constant null value"); + V = Constant::getNullValue(CurTy); + break; + case bitc::CST_CODE_INTEGER: // INTEGER: [intval] + if (!CurTy->isIntegerTy() || Record.empty()) + return error("Invalid record"); + V = ConstantInt::get(CurTy, decodeSignRotatedValue(Record[0])); + break; + case bitc::CST_CODE_WIDE_INTEGER: {// WIDE_INTEGER: [n x intval] + if (!CurTy->isIntegerTy() || Record.empty()) + return error("Invalid record"); + + APInt VInt = + readWideAPInt(Record, cast<IntegerType>(CurTy)->getBitWidth()); + V = ConstantInt::get(Context, VInt); + + break; + } + case bitc::CST_CODE_FLOAT: { // FLOAT: [fpval] + if (Record.empty()) + return error("Invalid record"); + if (CurTy->isHalfTy()) + V = ConstantFP::get(Context, APFloat(APFloat::IEEEhalf(), + APInt(16, (uint16_t)Record[0]))); + else if (CurTy->isBFloatTy()) + V = ConstantFP::get(Context, APFloat(APFloat::BFloat(), + APInt(16, (uint32_t)Record[0]))); + else if (CurTy->isFloatTy()) + V = ConstantFP::get(Context, APFloat(APFloat::IEEEsingle(), + APInt(32, (uint32_t)Record[0]))); + else if (CurTy->isDoubleTy()) + V = ConstantFP::get(Context, APFloat(APFloat::IEEEdouble(), + APInt(64, Record[0]))); + else if (CurTy->isX86_FP80Ty()) { + // Bits are not stored the same way as a normal i80 APInt, compensate. + uint64_t Rearrange[2]; + Rearrange[0] = (Record[1] & 0xffffLL) | (Record[0] << 16); + Rearrange[1] = Record[0] >> 48; + V = ConstantFP::get(Context, APFloat(APFloat::x87DoubleExtended(), + APInt(80, Rearrange))); + } else if (CurTy->isFP128Ty()) + V = ConstantFP::get(Context, APFloat(APFloat::IEEEquad(), + APInt(128, Record))); + else if (CurTy->isPPC_FP128Ty()) + V = ConstantFP::get(Context, APFloat(APFloat::PPCDoubleDouble(), + APInt(128, Record))); + else + V = UndefValue::get(CurTy); + break; + } + + case bitc::CST_CODE_AGGREGATE: {// AGGREGATE: [n x value number] + if (Record.empty()) + return error("Invalid record"); + + unsigned Size = Record.size(); + SmallVector<Constant*, 16> Elts; + + if (StructType *STy = dyn_cast<StructType>(CurTy)) { + for (unsigned i = 0; i != Size; ++i) + Elts.push_back(ValueList.getConstantFwdRef(Record[i], + STy->getElementType(i))); + V = ConstantStruct::get(STy, Elts); + } else if (ArrayType *ATy = dyn_cast<ArrayType>(CurTy)) { + Type *EltTy = ATy->getElementType(); + for (unsigned i = 0; i != Size; ++i) + Elts.push_back(ValueList.getConstantFwdRef(Record[i], EltTy)); + V = ConstantArray::get(ATy, Elts); + } else if (VectorType *VTy = dyn_cast<VectorType>(CurTy)) { + Type *EltTy = VTy->getElementType(); + for (unsigned i = 0; i != Size; ++i) + Elts.push_back(ValueList.getConstantFwdRef(Record[i], EltTy)); + V = ConstantVector::get(Elts); + } else { + V = UndefValue::get(CurTy); + } + break; + } + case bitc::CST_CODE_STRING: // STRING: [values] + case bitc::CST_CODE_CSTRING: { // CSTRING: [values] + if (Record.empty()) + return error("Invalid record"); + + SmallString<16> Elts(Record.begin(), Record.end()); + V = ConstantDataArray::getString(Context, Elts, + BitCode == bitc::CST_CODE_CSTRING); + break; + } + case bitc::CST_CODE_DATA: {// DATA: [n x value] + if (Record.empty()) + return error("Invalid record"); + + Type *EltTy; + if (auto *Array = dyn_cast<ArrayType>(CurTy)) + EltTy = Array->getElementType(); + else + EltTy = cast<VectorType>(CurTy)->getElementType(); + if (EltTy->isIntegerTy(8)) { + SmallVector<uint8_t, 16> Elts(Record.begin(), Record.end()); + if (isa<VectorType>(CurTy)) + V = ConstantDataVector::get(Context, Elts); + else + V = ConstantDataArray::get(Context, Elts); + } else if (EltTy->isIntegerTy(16)) { + SmallVector<uint16_t, 16> Elts(Record.begin(), Record.end()); + if (isa<VectorType>(CurTy)) + V = ConstantDataVector::get(Context, Elts); + else + V = ConstantDataArray::get(Context, Elts); + } else if (EltTy->isIntegerTy(32)) { + SmallVector<uint32_t, 16> Elts(Record.begin(), Record.end()); + if (isa<VectorType>(CurTy)) + V = ConstantDataVector::get(Context, Elts); + else + V = ConstantDataArray::get(Context, Elts); + } else if (EltTy->isIntegerTy(64)) { + SmallVector<uint64_t, 16> Elts(Record.begin(), Record.end()); + if (isa<VectorType>(CurTy)) + V = ConstantDataVector::get(Context, Elts); + else + V = ConstantDataArray::get(Context, Elts); + } else if (EltTy->isHalfTy()) { + SmallVector<uint16_t, 16> Elts(Record.begin(), Record.end()); + if (isa<VectorType>(CurTy)) + V = ConstantDataVector::getFP(EltTy, Elts); + else + V = ConstantDataArray::getFP(EltTy, Elts); + } else if (EltTy->isBFloatTy()) { + SmallVector<uint16_t, 16> Elts(Record.begin(), Record.end()); + if (isa<VectorType>(CurTy)) + V = ConstantDataVector::getFP(EltTy, Elts); + else + V = ConstantDataArray::getFP(EltTy, Elts); + } else if (EltTy->isFloatTy()) { + SmallVector<uint32_t, 16> Elts(Record.begin(), Record.end()); + if (isa<VectorType>(CurTy)) + V = ConstantDataVector::getFP(EltTy, Elts); + else + V = ConstantDataArray::getFP(EltTy, Elts); + } else if (EltTy->isDoubleTy()) { + SmallVector<uint64_t, 16> Elts(Record.begin(), Record.end()); + if (isa<VectorType>(CurTy)) + V = ConstantDataVector::getFP(EltTy, Elts); + else + V = ConstantDataArray::getFP(EltTy, Elts); + } else { + return error("Invalid type for value"); + } + break; + } + case bitc::CST_CODE_CE_UNOP: { // CE_UNOP: [opcode, opval] + if (Record.size() < 2) + return error("Invalid record"); + int Opc = getDecodedUnaryOpcode(Record[0], CurTy); + if (Opc < 0) { + V = UndefValue::get(CurTy); // Unknown unop. + } else { + Constant *LHS = ValueList.getConstantFwdRef(Record[1], CurTy); + unsigned Flags = 0; + V = ConstantExpr::get(Opc, LHS, Flags); + } + break; + } + case bitc::CST_CODE_CE_BINOP: { // CE_BINOP: [opcode, opval, opval] + if (Record.size() < 3) + return error("Invalid record"); + int Opc = getDecodedBinaryOpcode(Record[0], CurTy); + if (Opc < 0) { + V = UndefValue::get(CurTy); // Unknown binop. + } else { + Constant *LHS = ValueList.getConstantFwdRef(Record[1], CurTy); + Constant *RHS = ValueList.getConstantFwdRef(Record[2], CurTy); + unsigned Flags = 0; + if (Record.size() >= 4) { + if (Opc == Instruction::Add || + Opc == Instruction::Sub || + Opc == Instruction::Mul || + Opc == Instruction::Shl) { + if (Record[3] & (1 << bitc::OBO_NO_SIGNED_WRAP)) + Flags |= OverflowingBinaryOperator::NoSignedWrap; + if (Record[3] & (1 << bitc::OBO_NO_UNSIGNED_WRAP)) + Flags |= OverflowingBinaryOperator::NoUnsignedWrap; + } else if (Opc == Instruction::SDiv || + Opc == Instruction::UDiv || + Opc == Instruction::LShr || + Opc == Instruction::AShr) { + if (Record[3] & (1 << bitc::PEO_EXACT)) + Flags |= SDivOperator::IsExact; + } + } + V = ConstantExpr::get(Opc, LHS, RHS, Flags); + } + break; + } + case bitc::CST_CODE_CE_CAST: { // CE_CAST: [opcode, opty, opval] + if (Record.size() < 3) + return error("Invalid record"); + int Opc = getDecodedCastOpcode(Record[0]); + if (Opc < 0) { + V = UndefValue::get(CurTy); // Unknown cast. + } else { + Type *OpTy = getTypeByID(Record[1]); + if (!OpTy) + return error("Invalid record"); + Constant *Op = ValueList.getConstantFwdRef(Record[2], OpTy); + V = UpgradeBitCastExpr(Opc, Op, CurTy); + if (!V) V = ConstantExpr::getCast(Opc, Op, CurTy); + } + break; + } + case bitc::CST_CODE_CE_INBOUNDS_GEP: // [ty, n x operands] + case bitc::CST_CODE_CE_GEP: // [ty, n x operands] + case bitc::CST_CODE_CE_GEP_WITH_INRANGE_INDEX: { // [ty, flags, n x + // operands] + unsigned OpNum = 0; + Type *PointeeType = nullptr; + if (BitCode == bitc::CST_CODE_CE_GEP_WITH_INRANGE_INDEX || + Record.size() % 2) + PointeeType = getTypeByID(Record[OpNum++]); + + bool InBounds = false; + Optional<unsigned> InRangeIndex; + if (BitCode == bitc::CST_CODE_CE_GEP_WITH_INRANGE_INDEX) { + uint64_t Op = Record[OpNum++]; + InBounds = Op & 1; + InRangeIndex = Op >> 1; + } else if (BitCode == bitc::CST_CODE_CE_INBOUNDS_GEP) + InBounds = true; + + SmallVector<Constant*, 16> Elts; + Type *Elt0FullTy = nullptr; + while (OpNum != Record.size()) { + if (!Elt0FullTy) + Elt0FullTy = getTypeByID(Record[OpNum]); + Type *ElTy = getTypeByID(Record[OpNum++]); + if (!ElTy) + return error("Invalid record"); + Elts.push_back(ValueList.getConstantFwdRef(Record[OpNum++], ElTy)); + } + + if (Elts.size() < 1) + return error("Invalid gep with no operands"); + + PointerType *OrigPtrTy = cast<PointerType>(Elt0FullTy->getScalarType()); + if (!PointeeType) + PointeeType = OrigPtrTy->getPointerElementType(); + else if (!OrigPtrTy->isOpaqueOrPointeeTypeMatches(PointeeType)) + return error("Explicit gep operator type does not match pointee type " + "of pointer operand"); + + ArrayRef<Constant *> Indices(Elts.begin() + 1, Elts.end()); + V = ConstantExpr::getGetElementPtr(PointeeType, Elts[0], Indices, + InBounds, InRangeIndex); + break; + } + case bitc::CST_CODE_CE_SELECT: { // CE_SELECT: [opval#, opval#, opval#] + if (Record.size() < 3) + return error("Invalid record"); + + DelayedSelectors.push_back( + {CurTy, Record[0], Record[1], Record[2], NextCstNo}); + (void)ValueList.getConstantFwdRef(NextCstNo, CurTy); + ++NextCstNo; + continue; + } + case bitc::CST_CODE_CE_EXTRACTELT + : { // CE_EXTRACTELT: [opty, opval, opty, opval] + if (Record.size() < 3) + return error("Invalid record"); + VectorType *OpTy = + dyn_cast_or_null<VectorType>(getTypeByID(Record[0])); + if (!OpTy) + return error("Invalid record"); + Constant *Op0 = ValueList.getConstantFwdRef(Record[1], OpTy); + Constant *Op1 = nullptr; + if (Record.size() == 4) { + Type *IdxTy = getTypeByID(Record[2]); + if (!IdxTy) + return error("Invalid record"); + Op1 = ValueList.getConstantFwdRef(Record[3], IdxTy); + } else { + // Deprecated, but still needed to read old bitcode files. + Op1 = ValueList.getConstantFwdRef(Record[2], Type::getInt32Ty(Context)); + } + if (!Op1) + return error("Invalid record"); + V = ConstantExpr::getExtractElement(Op0, Op1); + break; + } + case bitc::CST_CODE_CE_INSERTELT + : { // CE_INSERTELT: [opval, opval, opty, opval] + VectorType *OpTy = dyn_cast<VectorType>(CurTy); + if (Record.size() < 3 || !OpTy) + return error("Invalid record"); + Constant *Op0 = ValueList.getConstantFwdRef(Record[0], OpTy); + Constant *Op1 = ValueList.getConstantFwdRef(Record[1], + OpTy->getElementType()); + Constant *Op2 = nullptr; + if (Record.size() == 4) { + Type *IdxTy = getTypeByID(Record[2]); + if (!IdxTy) + return error("Invalid record"); + Op2 = ValueList.getConstantFwdRef(Record[3], IdxTy); + } else { + // Deprecated, but still needed to read old bitcode files. + Op2 = ValueList.getConstantFwdRef(Record[2], Type::getInt32Ty(Context)); + } + if (!Op2) + return error("Invalid record"); + V = ConstantExpr::getInsertElement(Op0, Op1, Op2); + break; + } + case bitc::CST_CODE_CE_SHUFFLEVEC: { // CE_SHUFFLEVEC: [opval, opval, opval] + VectorType *OpTy = dyn_cast<VectorType>(CurTy); + if (Record.size() < 3 || !OpTy) + return error("Invalid record"); + DelayedShuffles.push_back( + {OpTy, OpTy, Record[0], Record[1], Record[2], NextCstNo}); + ++NextCstNo; + continue; + } + case bitc::CST_CODE_CE_SHUFVEC_EX: { // [opty, opval, opval, opval] + VectorType *RTy = dyn_cast<VectorType>(CurTy); + VectorType *OpTy = + dyn_cast_or_null<VectorType>(getTypeByID(Record[0])); + if (Record.size() < 4 || !RTy || !OpTy) + return error("Invalid record"); + DelayedShuffles.push_back( + {OpTy, RTy, Record[1], Record[2], Record[3], NextCstNo}); + ++NextCstNo; + continue; + } + case bitc::CST_CODE_CE_CMP: { // CE_CMP: [opty, opval, opval, pred] + if (Record.size() < 4) + return error("Invalid record"); + Type *OpTy = getTypeByID(Record[0]); + if (!OpTy) + return error("Invalid record"); + Constant *Op0 = ValueList.getConstantFwdRef(Record[1], OpTy); + Constant *Op1 = ValueList.getConstantFwdRef(Record[2], OpTy); + + if (OpTy->isFPOrFPVectorTy()) + V = ConstantExpr::getFCmp(Record[3], Op0, Op1); + else + V = ConstantExpr::getICmp(Record[3], Op0, Op1); + break; + } + // This maintains backward compatibility, pre-asm dialect keywords. + // Deprecated, but still needed to read old bitcode files. + case bitc::CST_CODE_INLINEASM_OLD: { + if (Record.size() < 2) + return error("Invalid record"); + std::string AsmStr, ConstrStr; + bool HasSideEffects = Record[0] & 1; + bool IsAlignStack = Record[0] >> 1; + unsigned AsmStrSize = Record[1]; + if (2+AsmStrSize >= Record.size()) + return error("Invalid record"); + unsigned ConstStrSize = Record[2+AsmStrSize]; + if (3+AsmStrSize+ConstStrSize > Record.size()) + return error("Invalid record"); + + for (unsigned i = 0; i != AsmStrSize; ++i) + AsmStr += (char)Record[2+i]; + for (unsigned i = 0; i != ConstStrSize; ++i) + ConstrStr += (char)Record[3+AsmStrSize+i]; + UpgradeInlineAsmString(&AsmStr); + // FIXME: support upgrading in opaque pointers mode. + V = InlineAsm::get(cast<FunctionType>(CurTy->getPointerElementType()), + AsmStr, ConstrStr, HasSideEffects, IsAlignStack); + break; + } + // This version adds support for the asm dialect keywords (e.g., + // inteldialect). + case bitc::CST_CODE_INLINEASM_OLD2: { + if (Record.size() < 2) + return error("Invalid record"); + std::string AsmStr, ConstrStr; + bool HasSideEffects = Record[0] & 1; + bool IsAlignStack = (Record[0] >> 1) & 1; + unsigned AsmDialect = Record[0] >> 2; + unsigned AsmStrSize = Record[1]; + if (2+AsmStrSize >= Record.size()) + return error("Invalid record"); + unsigned ConstStrSize = Record[2+AsmStrSize]; + if (3+AsmStrSize+ConstStrSize > Record.size()) + return error("Invalid record"); + + for (unsigned i = 0; i != AsmStrSize; ++i) + AsmStr += (char)Record[2+i]; + for (unsigned i = 0; i != ConstStrSize; ++i) + ConstrStr += (char)Record[3+AsmStrSize+i]; + UpgradeInlineAsmString(&AsmStr); + // FIXME: support upgrading in opaque pointers mode. + V = InlineAsm::get(cast<FunctionType>(CurTy->getPointerElementType()), + AsmStr, ConstrStr, HasSideEffects, IsAlignStack, + InlineAsm::AsmDialect(AsmDialect)); + break; + } + // This version adds support for the unwind keyword. + case bitc::CST_CODE_INLINEASM_OLD3: { + if (Record.size() < 2) + return error("Invalid record"); + unsigned OpNum = 0; + std::string AsmStr, ConstrStr; + bool HasSideEffects = Record[OpNum] & 1; + bool IsAlignStack = (Record[OpNum] >> 1) & 1; + unsigned AsmDialect = (Record[OpNum] >> 2) & 1; + bool CanThrow = (Record[OpNum] >> 3) & 1; + ++OpNum; + unsigned AsmStrSize = Record[OpNum]; + ++OpNum; + if (OpNum + AsmStrSize >= Record.size()) + return error("Invalid record"); + unsigned ConstStrSize = Record[OpNum + AsmStrSize]; + if (OpNum + 1 + AsmStrSize + ConstStrSize > Record.size()) + return error("Invalid record"); + + for (unsigned i = 0; i != AsmStrSize; ++i) + AsmStr += (char)Record[OpNum + i]; + ++OpNum; + for (unsigned i = 0; i != ConstStrSize; ++i) + ConstrStr += (char)Record[OpNum + AsmStrSize + i]; + UpgradeInlineAsmString(&AsmStr); + // FIXME: support upgrading in opaque pointers mode. + V = InlineAsm::get(cast<FunctionType>(CurTy->getPointerElementType()), + AsmStr, ConstrStr, HasSideEffects, IsAlignStack, + InlineAsm::AsmDialect(AsmDialect), CanThrow); + break; + } + // This version adds explicit function type. + case bitc::CST_CODE_INLINEASM: { + if (Record.size() < 3) + return error("Invalid record"); + unsigned OpNum = 0; + auto *FnTy = dyn_cast_or_null<FunctionType>(getTypeByID(Record[OpNum])); + ++OpNum; + if (!FnTy) + return error("Invalid record"); + std::string AsmStr, ConstrStr; + bool HasSideEffects = Record[OpNum] & 1; + bool IsAlignStack = (Record[OpNum] >> 1) & 1; + unsigned AsmDialect = (Record[OpNum] >> 2) & 1; + bool CanThrow = (Record[OpNum] >> 3) & 1; + ++OpNum; + unsigned AsmStrSize = Record[OpNum]; + ++OpNum; + if (OpNum + AsmStrSize >= Record.size()) + return error("Invalid record"); + unsigned ConstStrSize = Record[OpNum + AsmStrSize]; + if (OpNum + 1 + AsmStrSize + ConstStrSize > Record.size()) + return error("Invalid record"); + + for (unsigned i = 0; i != AsmStrSize; ++i) + AsmStr += (char)Record[OpNum + i]; + ++OpNum; + for (unsigned i = 0; i != ConstStrSize; ++i) + ConstrStr += (char)Record[OpNum + AsmStrSize + i]; + UpgradeInlineAsmString(&AsmStr); + V = InlineAsm::get(FnTy, AsmStr, ConstrStr, HasSideEffects, IsAlignStack, + InlineAsm::AsmDialect(AsmDialect), CanThrow); + break; + } + case bitc::CST_CODE_BLOCKADDRESS:{ + if (Record.size() < 3) + return error("Invalid record"); + Type *FnTy = getTypeByID(Record[0]); + if (!FnTy) + return error("Invalid record"); + Function *Fn = + dyn_cast_or_null<Function>(ValueList.getConstantFwdRef(Record[1],FnTy)); + if (!Fn) + return error("Invalid record"); + + // If the function is already parsed we can insert the block address right + // away. + BasicBlock *BB; + unsigned BBID = Record[2]; + if (!BBID) + // Invalid reference to entry block. + return error("Invalid ID"); + if (!Fn->empty()) { + Function::iterator BBI = Fn->begin(), BBE = Fn->end(); + for (size_t I = 0, E = BBID; I != E; ++I) { + if (BBI == BBE) + return error("Invalid ID"); + ++BBI; + } + BB = &*BBI; + } else { + // Otherwise insert a placeholder and remember it so it can be inserted + // when the function is parsed. + auto &FwdBBs = BasicBlockFwdRefs[Fn]; + if (FwdBBs.empty()) + BasicBlockFwdRefQueue.push_back(Fn); + if (FwdBBs.size() < BBID + 1) + FwdBBs.resize(BBID + 1); + if (!FwdBBs[BBID]) + FwdBBs[BBID] = BasicBlock::Create(Context); + BB = FwdBBs[BBID]; + } + V = BlockAddress::get(Fn, BB); + break; + } + case bitc::CST_CODE_DSO_LOCAL_EQUIVALENT: { + if (Record.size() < 2) + return error("Invalid record"); + Type *GVTy = getTypeByID(Record[0]); + if (!GVTy) + return error("Invalid record"); + GlobalValue *GV = dyn_cast_or_null<GlobalValue>( + ValueList.getConstantFwdRef(Record[1], GVTy)); + if (!GV) + return error("Invalid record"); + + V = DSOLocalEquivalent::get(GV); + break; + } + case bitc::CST_CODE_NO_CFI_VALUE: { + if (Record.size() < 2) + return error("Invalid record"); + Type *GVTy = getTypeByID(Record[0]); + if (!GVTy) + return error("Invalid record"); + GlobalValue *GV = dyn_cast_or_null<GlobalValue>( + ValueList.getConstantFwdRef(Record[1], GVTy)); + if (!GV) + return error("Invalid record"); + V = NoCFIValue::get(GV); + break; + } + } + + ValueList.assignValue(V, NextCstNo); + ++NextCstNo; + } +} + +Error BitcodeReader::parseUseLists() { + if (Error Err = Stream.EnterSubBlock(bitc::USELIST_BLOCK_ID)) + return Err; + + // Read all the records. + SmallVector<uint64_t, 64> Record; + + while (true) { + Expected<BitstreamEntry> MaybeEntry = Stream.advanceSkippingSubblocks(); + if (!MaybeEntry) + return MaybeEntry.takeError(); + BitstreamEntry Entry = MaybeEntry.get(); + + switch (Entry.Kind) { + case BitstreamEntry::SubBlock: // Handled for us already. + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + return Error::success(); + case BitstreamEntry::Record: + // The interesting case. + break; + } + + // Read a use list record. + Record.clear(); + bool IsBB = false; + Expected<unsigned> MaybeRecord = Stream.readRecord(Entry.ID, Record); + if (!MaybeRecord) + return MaybeRecord.takeError(); + switch (MaybeRecord.get()) { + default: // Default behavior: unknown type. + break; + case bitc::USELIST_CODE_BB: + IsBB = true; + LLVM_FALLTHROUGH; + case bitc::USELIST_CODE_DEFAULT: { + unsigned RecordLength = Record.size(); + if (RecordLength < 3) + // Records should have at least an ID and two indexes. + return error("Invalid record"); + unsigned ID = Record.pop_back_val(); + + Value *V; + if (IsBB) { + assert(ID < FunctionBBs.size() && "Basic block not found"); + V = FunctionBBs[ID]; + } else + V = ValueList[ID]; + unsigned NumUses = 0; + SmallDenseMap<const Use *, unsigned, 16> Order; + for (const Use &U : V->materialized_uses()) { + if (++NumUses > Record.size()) + break; + Order[&U] = Record[NumUses - 1]; + } + if (Order.size() != Record.size() || NumUses > Record.size()) + // Mismatches can happen if the functions are being materialized lazily + // (out-of-order), or a value has been upgraded. + break; + + V->sortUseList([&](const Use &L, const Use &R) { + return Order.lookup(&L) < Order.lookup(&R); + }); + break; + } + } + } +} + +/// When we see the block for metadata, remember where it is and then skip it. +/// This lets us lazily deserialize the metadata. +Error BitcodeReader::rememberAndSkipMetadata() { + // Save the current stream state. + uint64_t CurBit = Stream.GetCurrentBitNo(); + DeferredMetadataInfo.push_back(CurBit); + + // Skip over the block for now. + if (Error Err = Stream.SkipBlock()) + return Err; + return Error::success(); +} + +Error BitcodeReader::materializeMetadata() { + for (uint64_t BitPos : DeferredMetadataInfo) { + // Move the bit stream to the saved position. + if (Error JumpFailed = Stream.JumpToBit(BitPos)) + return JumpFailed; + if (Error Err = MDLoader->parseModuleMetadata()) + return Err; + } + + // Upgrade "Linker Options" module flag to "llvm.linker.options" module-level + // metadata. Only upgrade if the new option doesn't exist to avoid upgrade + // multiple times. + if (!TheModule->getNamedMetadata("llvm.linker.options")) { + if (Metadata *Val = TheModule->getModuleFlag("Linker Options")) { + NamedMDNode *LinkerOpts = + TheModule->getOrInsertNamedMetadata("llvm.linker.options"); + for (const MDOperand &MDOptions : cast<MDNode>(Val)->operands()) + LinkerOpts->addOperand(cast<MDNode>(MDOptions)); + } + } + + DeferredMetadataInfo.clear(); + return Error::success(); +} + +void BitcodeReader::setStripDebugInfo() { StripDebugInfo = true; } + +/// When we see the block for a function body, remember where it is and then +/// skip it. This lets us lazily deserialize the functions. +Error BitcodeReader::rememberAndSkipFunctionBody() { + // Get the function we are talking about. + if (FunctionsWithBodies.empty()) + return error("Insufficient function protos"); + + Function *Fn = FunctionsWithBodies.back(); + FunctionsWithBodies.pop_back(); + + // Save the current stream state. + uint64_t CurBit = Stream.GetCurrentBitNo(); + assert( + (DeferredFunctionInfo[Fn] == 0 || DeferredFunctionInfo[Fn] == CurBit) && + "Mismatch between VST and scanned function offsets"); + DeferredFunctionInfo[Fn] = CurBit; + + // Skip over the function block for now. + if (Error Err = Stream.SkipBlock()) + return Err; + return Error::success(); +} + +Error BitcodeReader::globalCleanup() { + // Patch the initializers for globals and aliases up. + if (Error Err = resolveGlobalAndIndirectSymbolInits()) + return Err; + if (!GlobalInits.empty() || !IndirectSymbolInits.empty()) + return error("Malformed global initializer set"); + + // Look for intrinsic functions which need to be upgraded at some point + // and functions that need to have their function attributes upgraded. + for (Function &F : *TheModule) { + MDLoader->upgradeDebugIntrinsics(F); + Function *NewFn; + if (UpgradeIntrinsicFunction(&F, NewFn)) + UpgradedIntrinsics[&F] = NewFn; + else if (auto Remangled = Intrinsic::remangleIntrinsicFunction(&F)) + // Some types could be renamed during loading if several modules are + // loaded in the same LLVMContext (LTO scenario). In this case we should + // remangle intrinsics names as well. + RemangledIntrinsics[&F] = Remangled.getValue(); + // Look for functions that rely on old function attribute behavior. + UpgradeFunctionAttributes(F); + } + + // Look for global variables which need to be renamed. + std::vector<std::pair<GlobalVariable *, GlobalVariable *>> UpgradedVariables; + for (GlobalVariable &GV : TheModule->globals()) + if (GlobalVariable *Upgraded = UpgradeGlobalVariable(&GV)) + UpgradedVariables.emplace_back(&GV, Upgraded); + for (auto &Pair : UpgradedVariables) { + Pair.first->eraseFromParent(); + TheModule->getGlobalList().push_back(Pair.second); + } + + // Force deallocation of memory for these vectors to favor the client that + // want lazy deserialization. + std::vector<std::pair<GlobalVariable *, unsigned>>().swap(GlobalInits); + std::vector<std::pair<GlobalValue *, unsigned>>().swap(IndirectSymbolInits); + return Error::success(); +} + +/// Support for lazy parsing of function bodies. This is required if we +/// either have an old bitcode file without a VST forward declaration record, +/// or if we have an anonymous function being materialized, since anonymous +/// functions do not have a name and are therefore not in the VST. +Error BitcodeReader::rememberAndSkipFunctionBodies() { + if (Error JumpFailed = Stream.JumpToBit(NextUnreadBit)) + return JumpFailed; + + if (Stream.AtEndOfStream()) + return error("Could not find function in stream"); + + if (!SeenFirstFunctionBody) + return error("Trying to materialize functions before seeing function blocks"); + + // An old bitcode file with the symbol table at the end would have + // finished the parse greedily. + assert(SeenValueSymbolTable); + + SmallVector<uint64_t, 64> Record; + + while (true) { + Expected<llvm::BitstreamEntry> MaybeEntry = Stream.advance(); + if (!MaybeEntry) + return MaybeEntry.takeError(); + llvm::BitstreamEntry Entry = MaybeEntry.get(); + + switch (Entry.Kind) { + default: + return error("Expect SubBlock"); + case BitstreamEntry::SubBlock: + switch (Entry.ID) { + default: + return error("Expect function block"); + case bitc::FUNCTION_BLOCK_ID: + if (Error Err = rememberAndSkipFunctionBody()) + return Err; + NextUnreadBit = Stream.GetCurrentBitNo(); + return Error::success(); + } + } + } +} + +bool BitcodeReaderBase::readBlockInfo() { + Expected<Optional<BitstreamBlockInfo>> MaybeNewBlockInfo = + Stream.ReadBlockInfoBlock(); + if (!MaybeNewBlockInfo) + return true; // FIXME Handle the error. + Optional<BitstreamBlockInfo> NewBlockInfo = + std::move(MaybeNewBlockInfo.get()); + if (!NewBlockInfo) + return true; + BlockInfo = std::move(*NewBlockInfo); + return false; +} + +Error BitcodeReader::parseComdatRecord(ArrayRef<uint64_t> Record) { + // v1: [selection_kind, name] + // v2: [strtab_offset, strtab_size, selection_kind] + StringRef Name; + std::tie(Name, Record) = readNameFromStrtab(Record); + + if (Record.empty()) + return error("Invalid record"); + Comdat::SelectionKind SK = getDecodedComdatSelectionKind(Record[0]); + std::string OldFormatName; + if (!UseStrtab) { + if (Record.size() < 2) + return error("Invalid record"); + unsigned ComdatNameSize = Record[1]; + OldFormatName.reserve(ComdatNameSize); + for (unsigned i = 0; i != ComdatNameSize; ++i) + OldFormatName += (char)Record[2 + i]; + Name = OldFormatName; + } + Comdat *C = TheModule->getOrInsertComdat(Name); + C->setSelectionKind(SK); + ComdatList.push_back(C); + return Error::success(); +} + +static void inferDSOLocal(GlobalValue *GV) { + // infer dso_local from linkage and visibility if it is not encoded. + if (GV->hasLocalLinkage() || + (!GV->hasDefaultVisibility() && !GV->hasExternalWeakLinkage())) + GV->setDSOLocal(true); +} + +Error BitcodeReader::parseGlobalVarRecord(ArrayRef<uint64_t> Record) { + // v1: [pointer type, isconst, initid, linkage, alignment, section, + // visibility, threadlocal, unnamed_addr, externally_initialized, + // dllstorageclass, comdat, attributes, preemption specifier, + // partition strtab offset, partition strtab size] (name in VST) + // v2: [strtab_offset, strtab_size, v1] + StringRef Name; + std::tie(Name, Record) = readNameFromStrtab(Record); + + if (Record.size() < 6) + return error("Invalid record"); + Type *Ty = getTypeByID(Record[0]); + if (!Ty) + return error("Invalid record"); + bool isConstant = Record[1] & 1; + bool explicitType = Record[1] & 2; + unsigned AddressSpace; + if (explicitType) { + AddressSpace = Record[1] >> 2; + } else { + if (!Ty->isPointerTy()) + return error("Invalid type for value"); + AddressSpace = cast<PointerType>(Ty)->getAddressSpace(); + Ty = Ty->getPointerElementType(); + } + + uint64_t RawLinkage = Record[3]; + GlobalValue::LinkageTypes Linkage = getDecodedLinkage(RawLinkage); + MaybeAlign Alignment; + if (Error Err = parseAlignmentValue(Record[4], Alignment)) + return Err; + std::string Section; + if (Record[5]) { + if (Record[5] - 1 >= SectionTable.size()) + return error("Invalid ID"); + Section = SectionTable[Record[5] - 1]; + } + GlobalValue::VisibilityTypes Visibility = GlobalValue::DefaultVisibility; + // Local linkage must have default visibility. + // auto-upgrade `hidden` and `protected` for old bitcode. + if (Record.size() > 6 && !GlobalValue::isLocalLinkage(Linkage)) + Visibility = getDecodedVisibility(Record[6]); + + GlobalVariable::ThreadLocalMode TLM = GlobalVariable::NotThreadLocal; + if (Record.size() > 7) + TLM = getDecodedThreadLocalMode(Record[7]); + + GlobalValue::UnnamedAddr UnnamedAddr = GlobalValue::UnnamedAddr::None; + if (Record.size() > 8) + UnnamedAddr = getDecodedUnnamedAddrType(Record[8]); + + bool ExternallyInitialized = false; + if (Record.size() > 9) + ExternallyInitialized = Record[9]; + + GlobalVariable *NewGV = + new GlobalVariable(*TheModule, Ty, isConstant, Linkage, nullptr, Name, + nullptr, TLM, AddressSpace, ExternallyInitialized); + NewGV->setAlignment(Alignment); + if (!Section.empty()) + NewGV->setSection(Section); + NewGV->setVisibility(Visibility); + NewGV->setUnnamedAddr(UnnamedAddr); + + if (Record.size() > 10) + NewGV->setDLLStorageClass(getDecodedDLLStorageClass(Record[10])); + else + upgradeDLLImportExportLinkage(NewGV, RawLinkage); + + ValueList.push_back(NewGV); + + // Remember which value to use for the global initializer. + if (unsigned InitID = Record[2]) + GlobalInits.push_back(std::make_pair(NewGV, InitID - 1)); + + if (Record.size() > 11) { + if (unsigned ComdatID = Record[11]) { + if (ComdatID > ComdatList.size()) + return error("Invalid global variable comdat ID"); + NewGV->setComdat(ComdatList[ComdatID - 1]); + } + } else if (hasImplicitComdat(RawLinkage)) { + ImplicitComdatObjects.insert(NewGV); + } + + if (Record.size() > 12) { + auto AS = getAttributes(Record[12]).getFnAttrs(); + NewGV->setAttributes(AS); + } + + if (Record.size() > 13) { + NewGV->setDSOLocal(getDecodedDSOLocal(Record[13])); + } + inferDSOLocal(NewGV); + + // Check whether we have enough values to read a partition name. + if (Record.size() > 15) + NewGV->setPartition(StringRef(Strtab.data() + Record[14], Record[15])); + + return Error::success(); +} + +Error BitcodeReader::parseFunctionRecord(ArrayRef<uint64_t> Record) { + // v1: [type, callingconv, isproto, linkage, paramattr, alignment, section, + // visibility, gc, unnamed_addr, prologuedata, dllstorageclass, comdat, + // prefixdata, personalityfn, preemption specifier, addrspace] (name in VST) + // v2: [strtab_offset, strtab_size, v1] + StringRef Name; + std::tie(Name, Record) = readNameFromStrtab(Record); + + if (Record.size() < 8) + return error("Invalid record"); + Type *FTy = getTypeByID(Record[0]); + if (!FTy) + return error("Invalid record"); + if (auto *PTy = dyn_cast<PointerType>(FTy)) + FTy = PTy->getPointerElementType(); + + if (!isa<FunctionType>(FTy)) + return error("Invalid type for value"); + auto CC = static_cast<CallingConv::ID>(Record[1]); + if (CC & ~CallingConv::MaxID) + return error("Invalid calling convention ID"); + + unsigned AddrSpace = TheModule->getDataLayout().getProgramAddressSpace(); + if (Record.size() > 16) + AddrSpace = Record[16]; + + Function *Func = + Function::Create(cast<FunctionType>(FTy), GlobalValue::ExternalLinkage, + AddrSpace, Name, TheModule); + + assert(Func->getFunctionType() == FTy && + "Incorrect fully specified type provided for function"); + FunctionTypes[Func] = cast<FunctionType>(FTy); + + Func->setCallingConv(CC); + bool isProto = Record[2]; + uint64_t RawLinkage = Record[3]; + Func->setLinkage(getDecodedLinkage(RawLinkage)); + Func->setAttributes(getAttributes(Record[4])); + + // Upgrade any old-style byval or sret without a type by propagating the + // argument's pointee type. There should be no opaque pointers where the byval + // type is implicit. + for (unsigned i = 0; i != Func->arg_size(); ++i) { + for (Attribute::AttrKind Kind : {Attribute::ByVal, Attribute::StructRet, + Attribute::InAlloca}) { + if (!Func->hasParamAttribute(i, Kind)) + continue; + + if (Func->getParamAttribute(i, Kind).getValueAsType()) + continue; + + Func->removeParamAttr(i, Kind); + + Type *PTy = cast<FunctionType>(FTy)->getParamType(i); + Type *PtrEltTy = PTy->getPointerElementType(); + Attribute NewAttr; + switch (Kind) { + case Attribute::ByVal: + NewAttr = Attribute::getWithByValType(Context, PtrEltTy); + break; + case Attribute::StructRet: + NewAttr = Attribute::getWithStructRetType(Context, PtrEltTy); + break; + case Attribute::InAlloca: + NewAttr = Attribute::getWithInAllocaType(Context, PtrEltTy); + break; + default: + llvm_unreachable("not an upgraded type attribute"); + } + + Func->addParamAttr(i, NewAttr); + } + } + + MaybeAlign Alignment; + if (Error Err = parseAlignmentValue(Record[5], Alignment)) + return Err; + Func->setAlignment(Alignment); + if (Record[6]) { + if (Record[6] - 1 >= SectionTable.size()) + return error("Invalid ID"); + Func->setSection(SectionTable[Record[6] - 1]); + } + // Local linkage must have default visibility. + // auto-upgrade `hidden` and `protected` for old bitcode. + if (!Func->hasLocalLinkage()) + Func->setVisibility(getDecodedVisibility(Record[7])); + if (Record.size() > 8 && Record[8]) { + if (Record[8] - 1 >= GCTable.size()) + return error("Invalid ID"); + Func->setGC(GCTable[Record[8] - 1]); + } + GlobalValue::UnnamedAddr UnnamedAddr = GlobalValue::UnnamedAddr::None; + if (Record.size() > 9) + UnnamedAddr = getDecodedUnnamedAddrType(Record[9]); + Func->setUnnamedAddr(UnnamedAddr); + + FunctionOperandInfo OperandInfo = {Func, 0, 0, 0}; + if (Record.size() > 10) + OperandInfo.Prologue = Record[10]; + + if (Record.size() > 11) + Func->setDLLStorageClass(getDecodedDLLStorageClass(Record[11])); + else + upgradeDLLImportExportLinkage(Func, RawLinkage); + + if (Record.size() > 12) { + if (unsigned ComdatID = Record[12]) { + if (ComdatID > ComdatList.size()) + return error("Invalid function comdat ID"); + Func->setComdat(ComdatList[ComdatID - 1]); + } + } else if (hasImplicitComdat(RawLinkage)) { + ImplicitComdatObjects.insert(Func); + } + + if (Record.size() > 13) + OperandInfo.Prefix = Record[13]; + + if (Record.size() > 14) + OperandInfo.PersonalityFn = Record[14]; + + if (Record.size() > 15) { + Func->setDSOLocal(getDecodedDSOLocal(Record[15])); + } + inferDSOLocal(Func); + + // Record[16] is the address space number. + + // Check whether we have enough values to read a partition name. Also make + // sure Strtab has enough values. + if (Record.size() > 18 && Strtab.data() && + Record[17] + Record[18] <= Strtab.size()) { + Func->setPartition(StringRef(Strtab.data() + Record[17], Record[18])); + } + + ValueList.push_back(Func); + + if (OperandInfo.PersonalityFn || OperandInfo.Prefix || OperandInfo.Prologue) + FunctionOperands.push_back(OperandInfo); + + // If this is a function with a body, remember the prototype we are + // creating now, so that we can match up the body with them later. + if (!isProto) { + Func->setIsMaterializable(true); + FunctionsWithBodies.push_back(Func); + DeferredFunctionInfo[Func] = 0; + } + return Error::success(); +} + +Error BitcodeReader::parseGlobalIndirectSymbolRecord( + unsigned BitCode, ArrayRef<uint64_t> Record) { + // v1 ALIAS_OLD: [alias type, aliasee val#, linkage] (name in VST) + // v1 ALIAS: [alias type, addrspace, aliasee val#, linkage, visibility, + // dllstorageclass, threadlocal, unnamed_addr, + // preemption specifier] (name in VST) + // v1 IFUNC: [alias type, addrspace, aliasee val#, linkage, + // visibility, dllstorageclass, threadlocal, unnamed_addr, + // preemption specifier] (name in VST) + // v2: [strtab_offset, strtab_size, v1] + StringRef Name; + std::tie(Name, Record) = readNameFromStrtab(Record); + + bool NewRecord = BitCode != bitc::MODULE_CODE_ALIAS_OLD; + if (Record.size() < (3 + (unsigned)NewRecord)) + return error("Invalid record"); + unsigned OpNum = 0; + Type *Ty = getTypeByID(Record[OpNum++]); + if (!Ty) + return error("Invalid record"); + + unsigned AddrSpace; + if (!NewRecord) { + auto *PTy = dyn_cast<PointerType>(Ty); + if (!PTy) + return error("Invalid type for value"); + Ty = PTy->getPointerElementType(); + AddrSpace = PTy->getAddressSpace(); + } else { + AddrSpace = Record[OpNum++]; + } + + auto Val = Record[OpNum++]; + auto Linkage = Record[OpNum++]; + GlobalValue *NewGA; + if (BitCode == bitc::MODULE_CODE_ALIAS || + BitCode == bitc::MODULE_CODE_ALIAS_OLD) + NewGA = GlobalAlias::create(Ty, AddrSpace, getDecodedLinkage(Linkage), Name, + TheModule); + else + NewGA = GlobalIFunc::create(Ty, AddrSpace, getDecodedLinkage(Linkage), Name, + nullptr, TheModule); + + // Local linkage must have default visibility. + // auto-upgrade `hidden` and `protected` for old bitcode. + if (OpNum != Record.size()) { + auto VisInd = OpNum++; + if (!NewGA->hasLocalLinkage()) + NewGA->setVisibility(getDecodedVisibility(Record[VisInd])); + } + if (BitCode == bitc::MODULE_CODE_ALIAS || + BitCode == bitc::MODULE_CODE_ALIAS_OLD) { + if (OpNum != Record.size()) + NewGA->setDLLStorageClass(getDecodedDLLStorageClass(Record[OpNum++])); + else + upgradeDLLImportExportLinkage(NewGA, Linkage); + if (OpNum != Record.size()) + NewGA->setThreadLocalMode(getDecodedThreadLocalMode(Record[OpNum++])); + if (OpNum != Record.size()) + NewGA->setUnnamedAddr(getDecodedUnnamedAddrType(Record[OpNum++])); + } + if (OpNum != Record.size()) + NewGA->setDSOLocal(getDecodedDSOLocal(Record[OpNum++])); + inferDSOLocal(NewGA); + + // Check whether we have enough values to read a partition name. + if (OpNum + 1 < Record.size()) { + NewGA->setPartition( + StringRef(Strtab.data() + Record[OpNum], Record[OpNum + 1])); + OpNum += 2; + } + + ValueList.push_back(NewGA); + IndirectSymbolInits.push_back(std::make_pair(NewGA, Val)); + return Error::success(); +} + +Error BitcodeReader::parseModule(uint64_t ResumeBit, + bool ShouldLazyLoadMetadata, + DataLayoutCallbackTy DataLayoutCallback) { + if (ResumeBit) { + if (Error JumpFailed = Stream.JumpToBit(ResumeBit)) + return JumpFailed; + } else if (Error Err = Stream.EnterSubBlock(bitc::MODULE_BLOCK_ID)) + return Err; + + SmallVector<uint64_t, 64> Record; + + // Parts of bitcode parsing depend on the datalayout. Make sure we + // finalize the datalayout before we run any of that code. + bool ResolvedDataLayout = false; + auto ResolveDataLayout = [&] { + if (ResolvedDataLayout) + return; + + // datalayout and triple can't be parsed after this point. + ResolvedDataLayout = true; + + // Upgrade data layout string. + std::string DL = llvm::UpgradeDataLayoutString( + TheModule->getDataLayoutStr(), TheModule->getTargetTriple()); + TheModule->setDataLayout(DL); + + if (auto LayoutOverride = + DataLayoutCallback(TheModule->getTargetTriple())) + TheModule->setDataLayout(*LayoutOverride); + }; + + // Read all the records for this module. + while (true) { + Expected<llvm::BitstreamEntry> MaybeEntry = Stream.advance(); + if (!MaybeEntry) + return MaybeEntry.takeError(); + llvm::BitstreamEntry Entry = MaybeEntry.get(); + + switch (Entry.Kind) { + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + ResolveDataLayout(); + return globalCleanup(); + + case BitstreamEntry::SubBlock: + switch (Entry.ID) { + default: // Skip unknown content. + if (Error Err = Stream.SkipBlock()) + return Err; + break; + case bitc::BLOCKINFO_BLOCK_ID: + if (readBlockInfo()) + return error("Malformed block"); + break; + case bitc::PARAMATTR_BLOCK_ID: + if (Error Err = parseAttributeBlock()) + return Err; + break; + case bitc::PARAMATTR_GROUP_BLOCK_ID: + if (Error Err = parseAttributeGroupBlock()) + return Err; + break; + case bitc::TYPE_BLOCK_ID_NEW: + if (Error Err = parseTypeTable()) + return Err; + break; + case bitc::VALUE_SYMTAB_BLOCK_ID: + if (!SeenValueSymbolTable) { + // Either this is an old form VST without function index and an + // associated VST forward declaration record (which would have caused + // the VST to be jumped to and parsed before it was encountered + // normally in the stream), or there were no function blocks to + // trigger an earlier parsing of the VST. + assert(VSTOffset == 0 || FunctionsWithBodies.empty()); + if (Error Err = parseValueSymbolTable()) + return Err; + SeenValueSymbolTable = true; + } else { + // We must have had a VST forward declaration record, which caused + // the parser to jump to and parse the VST earlier. + assert(VSTOffset > 0); + if (Error Err = Stream.SkipBlock()) + return Err; + } + break; + case bitc::CONSTANTS_BLOCK_ID: + if (Error Err = parseConstants()) + return Err; + if (Error Err = resolveGlobalAndIndirectSymbolInits()) + return Err; + break; + case bitc::METADATA_BLOCK_ID: + if (ShouldLazyLoadMetadata) { + if (Error Err = rememberAndSkipMetadata()) + return Err; + break; + } + assert(DeferredMetadataInfo.empty() && "Unexpected deferred metadata"); + if (Error Err = MDLoader->parseModuleMetadata()) + return Err; + break; + case bitc::METADATA_KIND_BLOCK_ID: + if (Error Err = MDLoader->parseMetadataKinds()) + return Err; + break; + case bitc::FUNCTION_BLOCK_ID: + ResolveDataLayout(); + + // If this is the first function body we've seen, reverse the + // FunctionsWithBodies list. + if (!SeenFirstFunctionBody) { + std::reverse(FunctionsWithBodies.begin(), FunctionsWithBodies.end()); + if (Error Err = globalCleanup()) + return Err; + SeenFirstFunctionBody = true; + } + + if (VSTOffset > 0) { + // If we have a VST forward declaration record, make sure we + // parse the VST now if we haven't already. It is needed to + // set up the DeferredFunctionInfo vector for lazy reading. + if (!SeenValueSymbolTable) { + if (Error Err = BitcodeReader::parseValueSymbolTable(VSTOffset)) + return Err; + SeenValueSymbolTable = true; + // Fall through so that we record the NextUnreadBit below. + // This is necessary in case we have an anonymous function that + // is later materialized. Since it will not have a VST entry we + // need to fall back to the lazy parse to find its offset. + } else { + // If we have a VST forward declaration record, but have already + // parsed the VST (just above, when the first function body was + // encountered here), then we are resuming the parse after + // materializing functions. The ResumeBit points to the + // start of the last function block recorded in the + // DeferredFunctionInfo map. Skip it. + if (Error Err = Stream.SkipBlock()) + return Err; + continue; + } + } + + // Support older bitcode files that did not have the function + // index in the VST, nor a VST forward declaration record, as + // well as anonymous functions that do not have VST entries. + // Build the DeferredFunctionInfo vector on the fly. + if (Error Err = rememberAndSkipFunctionBody()) + return Err; + + // Suspend parsing when we reach the function bodies. Subsequent + // materialization calls will resume it when necessary. If the bitcode + // file is old, the symbol table will be at the end instead and will not + // have been seen yet. In this case, just finish the parse now. + if (SeenValueSymbolTable) { + NextUnreadBit = Stream.GetCurrentBitNo(); + // After the VST has been parsed, we need to make sure intrinsic name + // are auto-upgraded. + return globalCleanup(); + } + break; + case bitc::USELIST_BLOCK_ID: + if (Error Err = parseUseLists()) + return Err; + break; + case bitc::OPERAND_BUNDLE_TAGS_BLOCK_ID: + if (Error Err = parseOperandBundleTags()) + return Err; + break; + case bitc::SYNC_SCOPE_NAMES_BLOCK_ID: + if (Error Err = parseSyncScopeNames()) + return Err; + break; + } + continue; + + case BitstreamEntry::Record: + // The interesting case. + break; + } + + // Read a record. + Expected<unsigned> MaybeBitCode = Stream.readRecord(Entry.ID, Record); + if (!MaybeBitCode) + return MaybeBitCode.takeError(); + switch (unsigned BitCode = MaybeBitCode.get()) { + default: break; // Default behavior, ignore unknown content. + case bitc::MODULE_CODE_VERSION: { + Expected<unsigned> VersionOrErr = parseVersionRecord(Record); + if (!VersionOrErr) + return VersionOrErr.takeError(); + UseRelativeIDs = *VersionOrErr >= 1; + break; + } + case bitc::MODULE_CODE_TRIPLE: { // TRIPLE: [strchr x N] + if (ResolvedDataLayout) + return error("target triple too late in module"); + std::string S; + if (convertToString(Record, 0, S)) + return error("Invalid record"); + TheModule->setTargetTriple(S); + break; + } + case bitc::MODULE_CODE_DATALAYOUT: { // DATALAYOUT: [strchr x N] + if (ResolvedDataLayout) + return error("datalayout too late in module"); + std::string S; + if (convertToString(Record, 0, S)) + return error("Invalid record"); + TheModule->setDataLayout(S); + break; + } + case bitc::MODULE_CODE_ASM: { // ASM: [strchr x N] + std::string S; + if (convertToString(Record, 0, S)) + return error("Invalid record"); + TheModule->setModuleInlineAsm(S); + break; + } + case bitc::MODULE_CODE_DEPLIB: { // DEPLIB: [strchr x N] + // Deprecated, but still needed to read old bitcode files. + std::string S; + if (convertToString(Record, 0, S)) + return error("Invalid record"); + // Ignore value. + break; + } + case bitc::MODULE_CODE_SECTIONNAME: { // SECTIONNAME: [strchr x N] + std::string S; + if (convertToString(Record, 0, S)) + return error("Invalid record"); + SectionTable.push_back(S); + break; + } + case bitc::MODULE_CODE_GCNAME: { // SECTIONNAME: [strchr x N] + std::string S; + if (convertToString(Record, 0, S)) + return error("Invalid record"); + GCTable.push_back(S); + break; + } + case bitc::MODULE_CODE_COMDAT: + if (Error Err = parseComdatRecord(Record)) + return Err; + break; + // FIXME: BitcodeReader should handle {GLOBALVAR, FUNCTION, ALIAS, IFUNC} + // written by ThinLinkBitcodeWriter. See + // `ThinLinkBitcodeWriter::writeSimplifiedModuleInfo` for the format of each + // record + // (https://github.com/llvm/llvm-project/blob/b6a93967d9c11e79802b5e75cec1584d6c8aa472/llvm/lib/Bitcode/Writer/BitcodeWriter.cpp#L4714) + case bitc::MODULE_CODE_GLOBALVAR: + if (Error Err = parseGlobalVarRecord(Record)) + return Err; + break; + case bitc::MODULE_CODE_FUNCTION: + ResolveDataLayout(); + if (Error Err = parseFunctionRecord(Record)) + return Err; + break; + case bitc::MODULE_CODE_IFUNC: + case bitc::MODULE_CODE_ALIAS: + case bitc::MODULE_CODE_ALIAS_OLD: + if (Error Err = parseGlobalIndirectSymbolRecord(BitCode, Record)) + return Err; + break; + /// MODULE_CODE_VSTOFFSET: [offset] + case bitc::MODULE_CODE_VSTOFFSET: + if (Record.empty()) + return error("Invalid record"); + // Note that we subtract 1 here because the offset is relative to one word + // before the start of the identification or module block, which was + // historically always the start of the regular bitcode header. + VSTOffset = Record[0] - 1; + break; + /// MODULE_CODE_SOURCE_FILENAME: [namechar x N] + case bitc::MODULE_CODE_SOURCE_FILENAME: + SmallString<128> ValueName; + if (convertToString(Record, 0, ValueName)) + return error("Invalid record"); + TheModule->setSourceFileName(ValueName); + break; + } + Record.clear(); + } +} + +Error BitcodeReader::parseBitcodeInto(Module *M, bool ShouldLazyLoadMetadata, + bool IsImporting, + DataLayoutCallbackTy DataLayoutCallback) { + TheModule = M; + MDLoader = MetadataLoader(Stream, *M, ValueList, IsImporting, + [&](unsigned ID) { return getTypeByID(ID); }); + return parseModule(0, ShouldLazyLoadMetadata, DataLayoutCallback); +} + +Error BitcodeReader::typeCheckLoadStoreInst(Type *ValType, Type *PtrType) { + if (!isa<PointerType>(PtrType)) + return error("Load/Store operand is not a pointer type"); + + if (!cast<PointerType>(PtrType)->isOpaqueOrPointeeTypeMatches(ValType)) + return error("Explicit load/store type does not match pointee " + "type of pointer operand"); + if (!PointerType::isLoadableOrStorableType(ValType)) + return error("Cannot load/store from pointer"); + return Error::success(); +} + +void BitcodeReader::propagateAttributeTypes(CallBase *CB, + ArrayRef<Type *> ArgsTys) { + for (unsigned i = 0; i != CB->arg_size(); ++i) { + for (Attribute::AttrKind Kind : {Attribute::ByVal, Attribute::StructRet, + Attribute::InAlloca}) { + if (!CB->paramHasAttr(i, Kind) || + CB->getParamAttr(i, Kind).getValueAsType()) + continue; + + CB->removeParamAttr(i, Kind); + + Type *PtrEltTy = ArgsTys[i]->getPointerElementType(); + Attribute NewAttr; + switch (Kind) { + case Attribute::ByVal: + NewAttr = Attribute::getWithByValType(Context, PtrEltTy); + break; + case Attribute::StructRet: + NewAttr = Attribute::getWithStructRetType(Context, PtrEltTy); + break; + case Attribute::InAlloca: + NewAttr = Attribute::getWithInAllocaType(Context, PtrEltTy); + break; + default: + llvm_unreachable("not an upgraded type attribute"); + } + + CB->addParamAttr(i, NewAttr); + } + } + + if (CB->isInlineAsm()) { + const InlineAsm *IA = cast<InlineAsm>(CB->getCalledOperand()); + unsigned ArgNo = 0; + for (const InlineAsm::ConstraintInfo &CI : IA->ParseConstraints()) { + if (!CI.hasArg()) + continue; + + if (CI.isIndirect && !CB->getAttributes().getParamElementType(ArgNo)) { + Type *ElemTy = ArgsTys[ArgNo]->getPointerElementType(); + CB->addParamAttr( + ArgNo, Attribute::get(Context, Attribute::ElementType, ElemTy)); + } + + ArgNo++; + } + } + + switch (CB->getIntrinsicID()) { + case Intrinsic::preserve_array_access_index: + case Intrinsic::preserve_struct_access_index: + if (!CB->getAttributes().getParamElementType(0)) { + Type *ElTy = ArgsTys[0]->getPointerElementType(); + Attribute NewAttr = Attribute::get(Context, Attribute::ElementType, ElTy); + CB->addParamAttr(0, NewAttr); + } + break; + default: + break; + } +} + +/// Lazily parse the specified function body block. +Error BitcodeReader::parseFunctionBody(Function *F) { + if (Error Err = Stream.EnterSubBlock(bitc::FUNCTION_BLOCK_ID)) + return Err; + + // Unexpected unresolved metadata when parsing function. + if (MDLoader->hasFwdRefs()) + return error("Invalid function metadata: incoming forward references"); + + InstructionList.clear(); + unsigned ModuleValueListSize = ValueList.size(); + unsigned ModuleMDLoaderSize = MDLoader->size(); + + // Add all the function arguments to the value table. +#ifndef NDEBUG + unsigned ArgNo = 0; + FunctionType *FTy = FunctionTypes[F]; +#endif + for (Argument &I : F->args()) { + assert(I.getType() == FTy->getParamType(ArgNo++) && + "Incorrect fully specified type for Function Argument"); + ValueList.push_back(&I); + } + unsigned NextValueNo = ValueList.size(); + BasicBlock *CurBB = nullptr; + unsigned CurBBNo = 0; + + DebugLoc LastLoc; + auto getLastInstruction = [&]() -> Instruction * { + if (CurBB && !CurBB->empty()) + return &CurBB->back(); + else if (CurBBNo && FunctionBBs[CurBBNo - 1] && + !FunctionBBs[CurBBNo - 1]->empty()) + return &FunctionBBs[CurBBNo - 1]->back(); + return nullptr; + }; + + std::vector<OperandBundleDef> OperandBundles; + + // Read all the records. + SmallVector<uint64_t, 64> Record; + + while (true) { + Expected<llvm::BitstreamEntry> MaybeEntry = Stream.advance(); + if (!MaybeEntry) + return MaybeEntry.takeError(); + llvm::BitstreamEntry Entry = MaybeEntry.get(); + + switch (Entry.Kind) { + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + goto OutOfRecordLoop; + + case BitstreamEntry::SubBlock: + switch (Entry.ID) { + default: // Skip unknown content. + if (Error Err = Stream.SkipBlock()) + return Err; + break; + case bitc::CONSTANTS_BLOCK_ID: + if (Error Err = parseConstants()) + return Err; + NextValueNo = ValueList.size(); + break; + case bitc::VALUE_SYMTAB_BLOCK_ID: + if (Error Err = parseValueSymbolTable()) + return Err; + break; + case bitc::METADATA_ATTACHMENT_ID: + if (Error Err = MDLoader->parseMetadataAttachment(*F, InstructionList)) + return Err; + break; + case bitc::METADATA_BLOCK_ID: + assert(DeferredMetadataInfo.empty() && + "Must read all module-level metadata before function-level"); + if (Error Err = MDLoader->parseFunctionMetadata()) + return Err; + break; + case bitc::USELIST_BLOCK_ID: + if (Error Err = parseUseLists()) + return Err; + break; + } + continue; + + case BitstreamEntry::Record: + // The interesting case. + break; + } + + // Read a record. + Record.clear(); + Instruction *I = nullptr; + Expected<unsigned> MaybeBitCode = Stream.readRecord(Entry.ID, Record); + if (!MaybeBitCode) + return MaybeBitCode.takeError(); + switch (unsigned BitCode = MaybeBitCode.get()) { + default: // Default behavior: reject + return error("Invalid value"); + case bitc::FUNC_CODE_DECLAREBLOCKS: { // DECLAREBLOCKS: [nblocks] + if (Record.empty() || Record[0] == 0) + return error("Invalid record"); + // Create all the basic blocks for the function. + FunctionBBs.resize(Record[0]); + + // See if anything took the address of blocks in this function. + auto BBFRI = BasicBlockFwdRefs.find(F); + if (BBFRI == BasicBlockFwdRefs.end()) { + for (BasicBlock *&BB : FunctionBBs) + BB = BasicBlock::Create(Context, "", F); + } else { + auto &BBRefs = BBFRI->second; + // Check for invalid basic block references. + if (BBRefs.size() > FunctionBBs.size()) + return error("Invalid ID"); + assert(!BBRefs.empty() && "Unexpected empty array"); + assert(!BBRefs.front() && "Invalid reference to entry block"); + for (unsigned I = 0, E = FunctionBBs.size(), RE = BBRefs.size(); I != E; + ++I) + if (I < RE && BBRefs[I]) { + BBRefs[I]->insertInto(F); + FunctionBBs[I] = BBRefs[I]; + } else { + FunctionBBs[I] = BasicBlock::Create(Context, "", F); + } + + // Erase from the table. + BasicBlockFwdRefs.erase(BBFRI); + } + + CurBB = FunctionBBs[0]; + continue; + } + + case bitc::FUNC_CODE_DEBUG_LOC_AGAIN: // DEBUG_LOC_AGAIN + // This record indicates that the last instruction is at the same + // location as the previous instruction with a location. + I = getLastInstruction(); + + if (!I) + return error("Invalid record"); + I->setDebugLoc(LastLoc); + I = nullptr; + continue; + + case bitc::FUNC_CODE_DEBUG_LOC: { // DEBUG_LOC: [line, col, scope, ia] + I = getLastInstruction(); + if (!I || Record.size() < 4) + return error("Invalid record"); + + unsigned Line = Record[0], Col = Record[1]; + unsigned ScopeID = Record[2], IAID = Record[3]; + bool isImplicitCode = Record.size() == 5 && Record[4]; + + MDNode *Scope = nullptr, *IA = nullptr; + if (ScopeID) { + Scope = dyn_cast_or_null<MDNode>( + MDLoader->getMetadataFwdRefOrLoad(ScopeID - 1)); + if (!Scope) + return error("Invalid record"); + } + if (IAID) { + IA = dyn_cast_or_null<MDNode>( + MDLoader->getMetadataFwdRefOrLoad(IAID - 1)); + if (!IA) + return error("Invalid record"); + } + LastLoc = DILocation::get(Scope->getContext(), Line, Col, Scope, IA, + isImplicitCode); + I->setDebugLoc(LastLoc); + I = nullptr; + continue; + } + case bitc::FUNC_CODE_INST_UNOP: { // UNOP: [opval, ty, opcode] + unsigned OpNum = 0; + Value *LHS; + if (getValueTypePair(Record, OpNum, NextValueNo, LHS) || + OpNum+1 > Record.size()) + return error("Invalid record"); + + int Opc = getDecodedUnaryOpcode(Record[OpNum++], LHS->getType()); + if (Opc == -1) + return error("Invalid record"); + I = UnaryOperator::Create((Instruction::UnaryOps)Opc, LHS); + InstructionList.push_back(I); + if (OpNum < Record.size()) { + if (isa<FPMathOperator>(I)) { + FastMathFlags FMF = getDecodedFastMathFlags(Record[OpNum]); + if (FMF.any()) + I->setFastMathFlags(FMF); + } + } + break; + } + case bitc::FUNC_CODE_INST_BINOP: { // BINOP: [opval, ty, opval, opcode] + unsigned OpNum = 0; + Value *LHS, *RHS; + if (getValueTypePair(Record, OpNum, NextValueNo, LHS) || + popValue(Record, OpNum, NextValueNo, LHS->getType(), RHS) || + OpNum+1 > Record.size()) + return error("Invalid record"); + + int Opc = getDecodedBinaryOpcode(Record[OpNum++], LHS->getType()); + if (Opc == -1) + return error("Invalid record"); + I = BinaryOperator::Create((Instruction::BinaryOps)Opc, LHS, RHS); + InstructionList.push_back(I); + if (OpNum < Record.size()) { + if (Opc == Instruction::Add || + Opc == Instruction::Sub || + Opc == Instruction::Mul || + Opc == Instruction::Shl) { + if (Record[OpNum] & (1 << bitc::OBO_NO_SIGNED_WRAP)) + cast<BinaryOperator>(I)->setHasNoSignedWrap(true); + if (Record[OpNum] & (1 << bitc::OBO_NO_UNSIGNED_WRAP)) + cast<BinaryOperator>(I)->setHasNoUnsignedWrap(true); + } else if (Opc == Instruction::SDiv || + Opc == Instruction::UDiv || + Opc == Instruction::LShr || + Opc == Instruction::AShr) { + if (Record[OpNum] & (1 << bitc::PEO_EXACT)) + cast<BinaryOperator>(I)->setIsExact(true); + } else if (isa<FPMathOperator>(I)) { + FastMathFlags FMF = getDecodedFastMathFlags(Record[OpNum]); + if (FMF.any()) + I->setFastMathFlags(FMF); + } + + } + break; + } + case bitc::FUNC_CODE_INST_CAST: { // CAST: [opval, opty, destty, castopc] + unsigned OpNum = 0; + Value *Op; + if (getValueTypePair(Record, OpNum, NextValueNo, Op) || + OpNum+2 != Record.size()) + return error("Invalid record"); + + Type *ResTy = getTypeByID(Record[OpNum]); + int Opc = getDecodedCastOpcode(Record[OpNum + 1]); + if (Opc == -1 || !ResTy) + return error("Invalid record"); + Instruction *Temp = nullptr; + if ((I = UpgradeBitCastInst(Opc, Op, ResTy, Temp))) { + if (Temp) { + InstructionList.push_back(Temp); + assert(CurBB && "No current BB?"); + CurBB->getInstList().push_back(Temp); + } + } else { + auto CastOp = (Instruction::CastOps)Opc; + if (!CastInst::castIsValid(CastOp, Op, ResTy)) + return error("Invalid cast"); + I = CastInst::Create(CastOp, Op, ResTy); + } + InstructionList.push_back(I); + break; + } + case bitc::FUNC_CODE_INST_INBOUNDS_GEP_OLD: + case bitc::FUNC_CODE_INST_GEP_OLD: + case bitc::FUNC_CODE_INST_GEP: { // GEP: type, [n x operands] + unsigned OpNum = 0; + + Type *Ty; + bool InBounds; + + if (BitCode == bitc::FUNC_CODE_INST_GEP) { + InBounds = Record[OpNum++]; + Ty = getTypeByID(Record[OpNum++]); + } else { + InBounds = BitCode == bitc::FUNC_CODE_INST_INBOUNDS_GEP_OLD; + Ty = nullptr; + } + + Value *BasePtr; + if (getValueTypePair(Record, OpNum, NextValueNo, BasePtr)) + return error("Invalid record"); + + if (!Ty) { + Ty = BasePtr->getType()->getScalarType()->getPointerElementType(); + } else if (!cast<PointerType>(BasePtr->getType()->getScalarType()) + ->isOpaqueOrPointeeTypeMatches(Ty)) { + return error( + "Explicit gep type does not match pointee type of pointer operand"); + } + + SmallVector<Value*, 16> GEPIdx; + while (OpNum != Record.size()) { + Value *Op; + if (getValueTypePair(Record, OpNum, NextValueNo, Op)) + return error("Invalid record"); + GEPIdx.push_back(Op); + } + + I = GetElementPtrInst::Create(Ty, BasePtr, GEPIdx); + + InstructionList.push_back(I); + if (InBounds) + cast<GetElementPtrInst>(I)->setIsInBounds(true); + break; + } + + case bitc::FUNC_CODE_INST_EXTRACTVAL: { + // EXTRACTVAL: [opty, opval, n x indices] + unsigned OpNum = 0; + Value *Agg; + if (getValueTypePair(Record, OpNum, NextValueNo, Agg)) + return error("Invalid record"); + Type *Ty = Agg->getType(); + + unsigned RecSize = Record.size(); + if (OpNum == RecSize) + return error("EXTRACTVAL: Invalid instruction with 0 indices"); + + SmallVector<unsigned, 4> EXTRACTVALIdx; + for (; OpNum != RecSize; ++OpNum) { + bool IsArray = Ty->isArrayTy(); + bool IsStruct = Ty->isStructTy(); + uint64_t Index = Record[OpNum]; + + if (!IsStruct && !IsArray) + return error("EXTRACTVAL: Invalid type"); + if ((unsigned)Index != Index) + return error("Invalid value"); + if (IsStruct && Index >= Ty->getStructNumElements()) + return error("EXTRACTVAL: Invalid struct index"); + if (IsArray && Index >= Ty->getArrayNumElements()) + return error("EXTRACTVAL: Invalid array index"); + EXTRACTVALIdx.push_back((unsigned)Index); + + if (IsStruct) + Ty = Ty->getStructElementType(Index); + else + Ty = Ty->getArrayElementType(); + } + + I = ExtractValueInst::Create(Agg, EXTRACTVALIdx); + InstructionList.push_back(I); + break; + } + + case bitc::FUNC_CODE_INST_INSERTVAL: { + // INSERTVAL: [opty, opval, opty, opval, n x indices] + unsigned OpNum = 0; + Value *Agg; + if (getValueTypePair(Record, OpNum, NextValueNo, Agg)) + return error("Invalid record"); + Value *Val; + if (getValueTypePair(Record, OpNum, NextValueNo, Val)) + return error("Invalid record"); + + unsigned RecSize = Record.size(); + if (OpNum == RecSize) + return error("INSERTVAL: Invalid instruction with 0 indices"); + + SmallVector<unsigned, 4> INSERTVALIdx; + Type *CurTy = Agg->getType(); + for (; OpNum != RecSize; ++OpNum) { + bool IsArray = CurTy->isArrayTy(); + bool IsStruct = CurTy->isStructTy(); + uint64_t Index = Record[OpNum]; + + if (!IsStruct && !IsArray) + return error("INSERTVAL: Invalid type"); + if ((unsigned)Index != Index) + return error("Invalid value"); + if (IsStruct && Index >= CurTy->getStructNumElements()) + return error("INSERTVAL: Invalid struct index"); + if (IsArray && Index >= CurTy->getArrayNumElements()) + return error("INSERTVAL: Invalid array index"); + + INSERTVALIdx.push_back((unsigned)Index); + if (IsStruct) + CurTy = CurTy->getStructElementType(Index); + else + CurTy = CurTy->getArrayElementType(); + } + + if (CurTy != Val->getType()) + return error("Inserted value type doesn't match aggregate type"); + + I = InsertValueInst::Create(Agg, Val, INSERTVALIdx); + InstructionList.push_back(I); + break; + } + + case bitc::FUNC_CODE_INST_SELECT: { // SELECT: [opval, ty, opval, opval] + // obsolete form of select + // handles select i1 ... in old bitcode + unsigned OpNum = 0; + Value *TrueVal, *FalseVal, *Cond; + if (getValueTypePair(Record, OpNum, NextValueNo, TrueVal) || + popValue(Record, OpNum, NextValueNo, TrueVal->getType(), FalseVal) || + popValue(Record, OpNum, NextValueNo, Type::getInt1Ty(Context), Cond)) + return error("Invalid record"); + + I = SelectInst::Create(Cond, TrueVal, FalseVal); + InstructionList.push_back(I); + break; + } + + case bitc::FUNC_CODE_INST_VSELECT: {// VSELECT: [ty,opval,opval,predty,pred] + // new form of select + // handles select i1 or select [N x i1] + unsigned OpNum = 0; + Value *TrueVal, *FalseVal, *Cond; + if (getValueTypePair(Record, OpNum, NextValueNo, TrueVal) || + popValue(Record, OpNum, NextValueNo, TrueVal->getType(), FalseVal) || + getValueTypePair(Record, OpNum, NextValueNo, Cond)) + return error("Invalid record"); + + // select condition can be either i1 or [N x i1] + if (VectorType* vector_type = + dyn_cast<VectorType>(Cond->getType())) { + // expect <n x i1> + if (vector_type->getElementType() != Type::getInt1Ty(Context)) + return error("Invalid type for value"); + } else { + // expect i1 + if (Cond->getType() != Type::getInt1Ty(Context)) + return error("Invalid type for value"); + } + + I = SelectInst::Create(Cond, TrueVal, FalseVal); + InstructionList.push_back(I); + if (OpNum < Record.size() && isa<FPMathOperator>(I)) { + FastMathFlags FMF = getDecodedFastMathFlags(Record[OpNum]); + if (FMF.any()) + I->setFastMathFlags(FMF); + } + break; + } + + case bitc::FUNC_CODE_INST_EXTRACTELT: { // EXTRACTELT: [opty, opval, opval] + unsigned OpNum = 0; + Value *Vec, *Idx; + if (getValueTypePair(Record, OpNum, NextValueNo, Vec) || + getValueTypePair(Record, OpNum, NextValueNo, Idx)) + return error("Invalid record"); + if (!Vec->getType()->isVectorTy()) + return error("Invalid type for value"); + I = ExtractElementInst::Create(Vec, Idx); + InstructionList.push_back(I); + break; + } + + case bitc::FUNC_CODE_INST_INSERTELT: { // INSERTELT: [ty, opval,opval,opval] + unsigned OpNum = 0; + Value *Vec, *Elt, *Idx; + if (getValueTypePair(Record, OpNum, NextValueNo, Vec)) + return error("Invalid record"); + if (!Vec->getType()->isVectorTy()) + return error("Invalid type for value"); + if (popValue(Record, OpNum, NextValueNo, + cast<VectorType>(Vec->getType())->getElementType(), Elt) || + getValueTypePair(Record, OpNum, NextValueNo, Idx)) + return error("Invalid record"); + I = InsertElementInst::Create(Vec, Elt, Idx); + InstructionList.push_back(I); + break; + } + + case bitc::FUNC_CODE_INST_SHUFFLEVEC: {// SHUFFLEVEC: [opval,ty,opval,opval] + unsigned OpNum = 0; + Value *Vec1, *Vec2, *Mask; + if (getValueTypePair(Record, OpNum, NextValueNo, Vec1) || + popValue(Record, OpNum, NextValueNo, Vec1->getType(), Vec2)) + return error("Invalid record"); + + if (getValueTypePair(Record, OpNum, NextValueNo, Mask)) + return error("Invalid record"); + if (!Vec1->getType()->isVectorTy() || !Vec2->getType()->isVectorTy()) + return error("Invalid type for value"); + + I = new ShuffleVectorInst(Vec1, Vec2, Mask); + InstructionList.push_back(I); + break; + } + + case bitc::FUNC_CODE_INST_CMP: // CMP: [opty, opval, opval, pred] + // Old form of ICmp/FCmp returning bool + // Existed to differentiate between icmp/fcmp and vicmp/vfcmp which were + // both legal on vectors but had different behaviour. + case bitc::FUNC_CODE_INST_CMP2: { // CMP2: [opty, opval, opval, pred] + // FCmp/ICmp returning bool or vector of bool + + unsigned OpNum = 0; + Value *LHS, *RHS; + if (getValueTypePair(Record, OpNum, NextValueNo, LHS) || + popValue(Record, OpNum, NextValueNo, LHS->getType(), RHS)) + return error("Invalid record"); + + if (OpNum >= Record.size()) + return error( + "Invalid record: operand number exceeded available operands"); + + unsigned PredVal = Record[OpNum]; + bool IsFP = LHS->getType()->isFPOrFPVectorTy(); + FastMathFlags FMF; + if (IsFP && Record.size() > OpNum+1) + FMF = getDecodedFastMathFlags(Record[++OpNum]); + + if (OpNum+1 != Record.size()) + return error("Invalid record"); + + if (LHS->getType()->isFPOrFPVectorTy()) + I = new FCmpInst((FCmpInst::Predicate)PredVal, LHS, RHS); + else + I = new ICmpInst((ICmpInst::Predicate)PredVal, LHS, RHS); + + if (FMF.any()) + I->setFastMathFlags(FMF); + InstructionList.push_back(I); + break; + } + + case bitc::FUNC_CODE_INST_RET: // RET: [opty,opval<optional>] + { + unsigned Size = Record.size(); + if (Size == 0) { + I = ReturnInst::Create(Context); + InstructionList.push_back(I); + break; + } + + unsigned OpNum = 0; + Value *Op = nullptr; + if (getValueTypePair(Record, OpNum, NextValueNo, Op)) + return error("Invalid record"); + if (OpNum != Record.size()) + return error("Invalid record"); + + I = ReturnInst::Create(Context, Op); + InstructionList.push_back(I); + break; + } + case bitc::FUNC_CODE_INST_BR: { // BR: [bb#, bb#, opval] or [bb#] + if (Record.size() != 1 && Record.size() != 3) + return error("Invalid record"); + BasicBlock *TrueDest = getBasicBlock(Record[0]); + if (!TrueDest) + return error("Invalid record"); + + if (Record.size() == 1) { + I = BranchInst::Create(TrueDest); + InstructionList.push_back(I); + } + else { + BasicBlock *FalseDest = getBasicBlock(Record[1]); + Value *Cond = getValue(Record, 2, NextValueNo, + Type::getInt1Ty(Context)); + if (!FalseDest || !Cond) + return error("Invalid record"); + I = BranchInst::Create(TrueDest, FalseDest, Cond); + InstructionList.push_back(I); + } + break; + } + case bitc::FUNC_CODE_INST_CLEANUPRET: { // CLEANUPRET: [val] or [val,bb#] + if (Record.size() != 1 && Record.size() != 2) + return error("Invalid record"); + unsigned Idx = 0; + Value *CleanupPad = + getValue(Record, Idx++, NextValueNo, Type::getTokenTy(Context)); + if (!CleanupPad) + return error("Invalid record"); + BasicBlock *UnwindDest = nullptr; + if (Record.size() == 2) { + UnwindDest = getBasicBlock(Record[Idx++]); + if (!UnwindDest) + return error("Invalid record"); + } + + I = CleanupReturnInst::Create(CleanupPad, UnwindDest); + InstructionList.push_back(I); + break; + } + case bitc::FUNC_CODE_INST_CATCHRET: { // CATCHRET: [val,bb#] + if (Record.size() != 2) + return error("Invalid record"); + unsigned Idx = 0; + Value *CatchPad = + getValue(Record, Idx++, NextValueNo, Type::getTokenTy(Context)); + if (!CatchPad) + return error("Invalid record"); + BasicBlock *BB = getBasicBlock(Record[Idx++]); + if (!BB) + return error("Invalid record"); + + I = CatchReturnInst::Create(CatchPad, BB); + InstructionList.push_back(I); + break; + } + case bitc::FUNC_CODE_INST_CATCHSWITCH: { // CATCHSWITCH: [tok,num,(bb)*,bb?] + // We must have, at minimum, the outer scope and the number of arguments. + if (Record.size() < 2) + return error("Invalid record"); + + unsigned Idx = 0; + + Value *ParentPad = + getValue(Record, Idx++, NextValueNo, Type::getTokenTy(Context)); + + unsigned NumHandlers = Record[Idx++]; + + SmallVector<BasicBlock *, 2> Handlers; + for (unsigned Op = 0; Op != NumHandlers; ++Op) { + BasicBlock *BB = getBasicBlock(Record[Idx++]); + if (!BB) + return error("Invalid record"); + Handlers.push_back(BB); + } + + BasicBlock *UnwindDest = nullptr; + if (Idx + 1 == Record.size()) { + UnwindDest = getBasicBlock(Record[Idx++]); + if (!UnwindDest) + return error("Invalid record"); + } + + if (Record.size() != Idx) + return error("Invalid record"); + + auto *CatchSwitch = + CatchSwitchInst::Create(ParentPad, UnwindDest, NumHandlers); + for (BasicBlock *Handler : Handlers) + CatchSwitch->addHandler(Handler); + I = CatchSwitch; + InstructionList.push_back(I); + break; + } + case bitc::FUNC_CODE_INST_CATCHPAD: + case bitc::FUNC_CODE_INST_CLEANUPPAD: { // [tok,num,(ty,val)*] + // We must have, at minimum, the outer scope and the number of arguments. + if (Record.size() < 2) + return error("Invalid record"); + + unsigned Idx = 0; + + Value *ParentPad = + getValue(Record, Idx++, NextValueNo, Type::getTokenTy(Context)); + + unsigned NumArgOperands = Record[Idx++]; + + SmallVector<Value *, 2> Args; + for (unsigned Op = 0; Op != NumArgOperands; ++Op) { + Value *Val; + if (getValueTypePair(Record, Idx, NextValueNo, Val)) + return error("Invalid record"); + Args.push_back(Val); + } + + if (Record.size() != Idx) + return error("Invalid record"); + + if (BitCode == bitc::FUNC_CODE_INST_CLEANUPPAD) + I = CleanupPadInst::Create(ParentPad, Args); + else + I = CatchPadInst::Create(ParentPad, Args); + InstructionList.push_back(I); + break; + } + case bitc::FUNC_CODE_INST_SWITCH: { // SWITCH: [opty, op0, op1, ...] + // Check magic + if ((Record[0] >> 16) == SWITCH_INST_MAGIC) { + // "New" SwitchInst format with case ranges. The changes to write this + // format were reverted but we still recognize bitcode that uses it. + // Hopefully someday we will have support for case ranges and can use + // this format again. + + Type *OpTy = getTypeByID(Record[1]); + unsigned ValueBitWidth = cast<IntegerType>(OpTy)->getBitWidth(); + + Value *Cond = getValue(Record, 2, NextValueNo, OpTy); + BasicBlock *Default = getBasicBlock(Record[3]); + if (!OpTy || !Cond || !Default) + return error("Invalid record"); + + unsigned NumCases = Record[4]; + + SwitchInst *SI = SwitchInst::Create(Cond, Default, NumCases); + InstructionList.push_back(SI); + + unsigned CurIdx = 5; + for (unsigned i = 0; i != NumCases; ++i) { + SmallVector<ConstantInt*, 1> CaseVals; + unsigned NumItems = Record[CurIdx++]; + for (unsigned ci = 0; ci != NumItems; ++ci) { + bool isSingleNumber = Record[CurIdx++]; + + APInt Low; + unsigned ActiveWords = 1; + if (ValueBitWidth > 64) + ActiveWords = Record[CurIdx++]; + Low = readWideAPInt(makeArrayRef(&Record[CurIdx], ActiveWords), + ValueBitWidth); + CurIdx += ActiveWords; + + if (!isSingleNumber) { + ActiveWords = 1; + if (ValueBitWidth > 64) + ActiveWords = Record[CurIdx++]; + APInt High = readWideAPInt( + makeArrayRef(&Record[CurIdx], ActiveWords), ValueBitWidth); + CurIdx += ActiveWords; + + // FIXME: It is not clear whether values in the range should be + // compared as signed or unsigned values. The partially + // implemented changes that used this format in the past used + // unsigned comparisons. + for ( ; Low.ule(High); ++Low) + CaseVals.push_back(ConstantInt::get(Context, Low)); + } else + CaseVals.push_back(ConstantInt::get(Context, Low)); + } + BasicBlock *DestBB = getBasicBlock(Record[CurIdx++]); + for (ConstantInt *Cst : CaseVals) + SI->addCase(Cst, DestBB); + } + I = SI; + break; + } + + // Old SwitchInst format without case ranges. + + if (Record.size() < 3 || (Record.size() & 1) == 0) + return error("Invalid record"); + Type *OpTy = getTypeByID(Record[0]); + Value *Cond = getValue(Record, 1, NextValueNo, OpTy); + BasicBlock *Default = getBasicBlock(Record[2]); + if (!OpTy || !Cond || !Default) + return error("Invalid record"); + unsigned NumCases = (Record.size()-3)/2; + SwitchInst *SI = SwitchInst::Create(Cond, Default, NumCases); + InstructionList.push_back(SI); + for (unsigned i = 0, e = NumCases; i != e; ++i) { + ConstantInt *CaseVal = + dyn_cast_or_null<ConstantInt>(getFnValueByID(Record[3+i*2], OpTy)); + BasicBlock *DestBB = getBasicBlock(Record[1+3+i*2]); + if (!CaseVal || !DestBB) { + delete SI; + return error("Invalid record"); + } + SI->addCase(CaseVal, DestBB); + } + I = SI; + break; + } + case bitc::FUNC_CODE_INST_INDIRECTBR: { // INDIRECTBR: [opty, op0, op1, ...] + if (Record.size() < 2) + return error("Invalid record"); + Type *OpTy = getTypeByID(Record[0]); + Value *Address = getValue(Record, 1, NextValueNo, OpTy); + if (!OpTy || !Address) + return error("Invalid record"); + unsigned NumDests = Record.size()-2; + IndirectBrInst *IBI = IndirectBrInst::Create(Address, NumDests); + InstructionList.push_back(IBI); + for (unsigned i = 0, e = NumDests; i != e; ++i) { + if (BasicBlock *DestBB = getBasicBlock(Record[2+i])) { + IBI->addDestination(DestBB); + } else { + delete IBI; + return error("Invalid record"); + } + } + I = IBI; + break; + } + + case bitc::FUNC_CODE_INST_INVOKE: { + // INVOKE: [attrs, cc, normBB, unwindBB, fnty, op0,op1,op2, ...] + if (Record.size() < 4) + return error("Invalid record"); + unsigned OpNum = 0; + AttributeList PAL = getAttributes(Record[OpNum++]); + unsigned CCInfo = Record[OpNum++]; + BasicBlock *NormalBB = getBasicBlock(Record[OpNum++]); + BasicBlock *UnwindBB = getBasicBlock(Record[OpNum++]); + + FunctionType *FTy = nullptr; + if ((CCInfo >> 13) & 1) { + FTy = dyn_cast<FunctionType>(getTypeByID(Record[OpNum++])); + if (!FTy) + return error("Explicit invoke type is not a function type"); + } + + Value *Callee; + if (getValueTypePair(Record, OpNum, NextValueNo, Callee)) + return error("Invalid record"); + + PointerType *CalleeTy = dyn_cast<PointerType>(Callee->getType()); + if (!CalleeTy) + return error("Callee is not a pointer"); + if (!FTy) { + FTy = + dyn_cast<FunctionType>(Callee->getType()->getPointerElementType()); + if (!FTy) + return error("Callee is not of pointer to function type"); + } else if (!CalleeTy->isOpaqueOrPointeeTypeMatches(FTy)) + return error("Explicit invoke type does not match pointee type of " + "callee operand"); + if (Record.size() < FTy->getNumParams() + OpNum) + return error("Insufficient operands to call"); + + SmallVector<Value*, 16> Ops; + SmallVector<Type *, 16> ArgsTys; + for (unsigned i = 0, e = FTy->getNumParams(); i != e; ++i, ++OpNum) { + Ops.push_back(getValue(Record, OpNum, NextValueNo, + FTy->getParamType(i))); + ArgsTys.push_back(FTy->getParamType(i)); + if (!Ops.back()) + return error("Invalid record"); + } + + if (!FTy->isVarArg()) { + if (Record.size() != OpNum) + return error("Invalid record"); + } else { + // Read type/value pairs for varargs params. + while (OpNum != Record.size()) { + Value *Op; + if (getValueTypePair(Record, OpNum, NextValueNo, Op)) + return error("Invalid record"); + Ops.push_back(Op); + ArgsTys.push_back(Op->getType()); + } + } + + I = InvokeInst::Create(FTy, Callee, NormalBB, UnwindBB, Ops, + OperandBundles); + OperandBundles.clear(); + InstructionList.push_back(I); + cast<InvokeInst>(I)->setCallingConv( + static_cast<CallingConv::ID>(CallingConv::MaxID & CCInfo)); + cast<InvokeInst>(I)->setAttributes(PAL); + propagateAttributeTypes(cast<CallBase>(I), ArgsTys); + + break; + } + case bitc::FUNC_CODE_INST_RESUME: { // RESUME: [opval] + unsigned Idx = 0; + Value *Val = nullptr; + if (getValueTypePair(Record, Idx, NextValueNo, Val)) + return error("Invalid record"); + I = ResumeInst::Create(Val); + InstructionList.push_back(I); + break; + } + case bitc::FUNC_CODE_INST_CALLBR: { + // CALLBR: [attr, cc, norm, transfs, fty, fnid, args] + unsigned OpNum = 0; + AttributeList PAL = getAttributes(Record[OpNum++]); + unsigned CCInfo = Record[OpNum++]; + + BasicBlock *DefaultDest = getBasicBlock(Record[OpNum++]); + unsigned NumIndirectDests = Record[OpNum++]; + SmallVector<BasicBlock *, 16> IndirectDests; + for (unsigned i = 0, e = NumIndirectDests; i != e; ++i) + IndirectDests.push_back(getBasicBlock(Record[OpNum++])); + + FunctionType *FTy = nullptr; + if ((CCInfo >> bitc::CALL_EXPLICIT_TYPE) & 1) { + FTy = dyn_cast<FunctionType>(getTypeByID(Record[OpNum++])); + if (!FTy) + return error("Explicit call type is not a function type"); + } + + Value *Callee; + if (getValueTypePair(Record, OpNum, NextValueNo, Callee)) + return error("Invalid record"); + + PointerType *OpTy = dyn_cast<PointerType>(Callee->getType()); + if (!OpTy) + return error("Callee is not a pointer type"); + if (!FTy) { + FTy = + dyn_cast<FunctionType>(Callee->getType()->getPointerElementType()); + if (!FTy) + return error("Callee is not of pointer to function type"); + } else if (!OpTy->isOpaqueOrPointeeTypeMatches(FTy)) + return error("Explicit call type does not match pointee type of " + "callee operand"); + if (Record.size() < FTy->getNumParams() + OpNum) + return error("Insufficient operands to call"); + + SmallVector<Value*, 16> Args; + SmallVector<Type *, 16> ArgsTys; + // Read the fixed params. + for (unsigned i = 0, e = FTy->getNumParams(); i != e; ++i, ++OpNum) { + Value *Arg; + if (FTy->getParamType(i)->isLabelTy()) + Arg = getBasicBlock(Record[OpNum]); + else + Arg = getValue(Record, OpNum, NextValueNo, FTy->getParamType(i)); + if (!Arg) + return error("Invalid record"); + Args.push_back(Arg); + ArgsTys.push_back(Arg->getType()); + } + + // Read type/value pairs for varargs params. + if (!FTy->isVarArg()) { + if (OpNum != Record.size()) + return error("Invalid record"); + } else { + while (OpNum != Record.size()) { + Value *Op; + if (getValueTypePair(Record, OpNum, NextValueNo, Op)) + return error("Invalid record"); + Args.push_back(Op); + ArgsTys.push_back(Op->getType()); + } + } + + I = CallBrInst::Create(FTy, Callee, DefaultDest, IndirectDests, Args, + OperandBundles); + OperandBundles.clear(); + InstructionList.push_back(I); + cast<CallBrInst>(I)->setCallingConv( + static_cast<CallingConv::ID>((0x7ff & CCInfo) >> bitc::CALL_CCONV)); + cast<CallBrInst>(I)->setAttributes(PAL); + propagateAttributeTypes(cast<CallBase>(I), ArgsTys); + break; + } + case bitc::FUNC_CODE_INST_UNREACHABLE: // UNREACHABLE + I = new UnreachableInst(Context); + InstructionList.push_back(I); + break; + case bitc::FUNC_CODE_INST_PHI: { // PHI: [ty, val0,bb0, ...] + if (Record.empty()) + return error("Invalid record"); + // The first record specifies the type. + Type *Ty = getTypeByID(Record[0]); + if (!Ty) + return error("Invalid record"); + + // Phi arguments are pairs of records of [value, basic block]. + // There is an optional final record for fast-math-flags if this phi has a + // floating-point type. + size_t NumArgs = (Record.size() - 1) / 2; + PHINode *PN = PHINode::Create(Ty, NumArgs); + if ((Record.size() - 1) % 2 == 1 && !isa<FPMathOperator>(PN)) + return error("Invalid record"); + InstructionList.push_back(PN); + + for (unsigned i = 0; i != NumArgs; i++) { + Value *V; + // With the new function encoding, it is possible that operands have + // negative IDs (for forward references). Use a signed VBR + // representation to keep the encoding small. + if (UseRelativeIDs) + V = getValueSigned(Record, i * 2 + 1, NextValueNo, Ty); + else + V = getValue(Record, i * 2 + 1, NextValueNo, Ty); + BasicBlock *BB = getBasicBlock(Record[i * 2 + 2]); + if (!V || !BB) + return error("Invalid record"); + PN->addIncoming(V, BB); + } + I = PN; + + // If there are an even number of records, the final record must be FMF. + if (Record.size() % 2 == 0) { + assert(isa<FPMathOperator>(I) && "Unexpected phi type"); + FastMathFlags FMF = getDecodedFastMathFlags(Record[Record.size() - 1]); + if (FMF.any()) + I->setFastMathFlags(FMF); + } + + break; + } + + case bitc::FUNC_CODE_INST_LANDINGPAD: + case bitc::FUNC_CODE_INST_LANDINGPAD_OLD: { + // LANDINGPAD: [ty, val, val, num, (id0,val0 ...)?] + unsigned Idx = 0; + if (BitCode == bitc::FUNC_CODE_INST_LANDINGPAD) { + if (Record.size() < 3) + return error("Invalid record"); + } else { + assert(BitCode == bitc::FUNC_CODE_INST_LANDINGPAD_OLD); + if (Record.size() < 4) + return error("Invalid record"); + } + Type *Ty = getTypeByID(Record[Idx++]); + if (!Ty) + return error("Invalid record"); + if (BitCode == bitc::FUNC_CODE_INST_LANDINGPAD_OLD) { + Value *PersFn = nullptr; + if (getValueTypePair(Record, Idx, NextValueNo, PersFn)) + return error("Invalid record"); + + if (!F->hasPersonalityFn()) + F->setPersonalityFn(cast<Constant>(PersFn)); + else if (F->getPersonalityFn() != cast<Constant>(PersFn)) + return error("Personality function mismatch"); + } + + bool IsCleanup = !!Record[Idx++]; + unsigned NumClauses = Record[Idx++]; + LandingPadInst *LP = LandingPadInst::Create(Ty, NumClauses); + LP->setCleanup(IsCleanup); + for (unsigned J = 0; J != NumClauses; ++J) { + LandingPadInst::ClauseType CT = + LandingPadInst::ClauseType(Record[Idx++]); (void)CT; + Value *Val; + + if (getValueTypePair(Record, Idx, NextValueNo, Val)) { + delete LP; + return error("Invalid record"); + } + + assert((CT != LandingPadInst::Catch || + !isa<ArrayType>(Val->getType())) && + "Catch clause has a invalid type!"); + assert((CT != LandingPadInst::Filter || + isa<ArrayType>(Val->getType())) && + "Filter clause has invalid type!"); + LP->addClause(cast<Constant>(Val)); + } + + I = LP; + InstructionList.push_back(I); + break; + } + + case bitc::FUNC_CODE_INST_ALLOCA: { // ALLOCA: [instty, opty, op, align] + if (Record.size() != 4) + return error("Invalid record"); + using APV = AllocaPackedValues; + const uint64_t Rec = Record[3]; + const bool InAlloca = Bitfield::get<APV::UsedWithInAlloca>(Rec); + const bool SwiftError = Bitfield::get<APV::SwiftError>(Rec); + Type *Ty = getTypeByID(Record[0]); + if (!Bitfield::get<APV::ExplicitType>(Rec)) { + auto *PTy = dyn_cast_or_null<PointerType>(Ty); + if (!PTy) + return error("Old-style alloca with a non-pointer type"); + Ty = PTy->getPointerElementType(); + } + Type *OpTy = getTypeByID(Record[1]); + Value *Size = getFnValueByID(Record[2], OpTy); + MaybeAlign Align; + uint64_t AlignExp = + Bitfield::get<APV::AlignLower>(Rec) | + (Bitfield::get<APV::AlignUpper>(Rec) << APV::AlignLower::Bits); + if (Error Err = parseAlignmentValue(AlignExp, Align)) { + return Err; + } + if (!Ty || !Size) + return error("Invalid record"); + + // FIXME: Make this an optional field. + const DataLayout &DL = TheModule->getDataLayout(); + unsigned AS = DL.getAllocaAddrSpace(); + + SmallPtrSet<Type *, 4> Visited; + if (!Align && !Ty->isSized(&Visited)) + return error("alloca of unsized type"); + if (!Align) + Align = DL.getPrefTypeAlign(Ty); + + AllocaInst *AI = new AllocaInst(Ty, AS, Size, *Align); + AI->setUsedWithInAlloca(InAlloca); + AI->setSwiftError(SwiftError); + I = AI; + InstructionList.push_back(I); + break; + } + case bitc::FUNC_CODE_INST_LOAD: { // LOAD: [opty, op, align, vol] + unsigned OpNum = 0; + Value *Op; + if (getValueTypePair(Record, OpNum, NextValueNo, Op) || + (OpNum + 2 != Record.size() && OpNum + 3 != Record.size())) + return error("Invalid record"); + + if (!isa<PointerType>(Op->getType())) + return error("Load operand is not a pointer type"); + + Type *Ty = nullptr; + if (OpNum + 3 == Record.size()) { + Ty = getTypeByID(Record[OpNum++]); + } else { + Ty = Op->getType()->getPointerElementType(); + } + + if (Error Err = typeCheckLoadStoreInst(Ty, Op->getType())) + return Err; + + MaybeAlign Align; + if (Error Err = parseAlignmentValue(Record[OpNum], Align)) + return Err; + SmallPtrSet<Type *, 4> Visited; + if (!Align && !Ty->isSized(&Visited)) + return error("load of unsized type"); + if (!Align) + Align = TheModule->getDataLayout().getABITypeAlign(Ty); + I = new LoadInst(Ty, Op, "", Record[OpNum + 1], *Align); + InstructionList.push_back(I); + break; + } + case bitc::FUNC_CODE_INST_LOADATOMIC: { + // LOADATOMIC: [opty, op, align, vol, ordering, ssid] + unsigned OpNum = 0; + Value *Op; + if (getValueTypePair(Record, OpNum, NextValueNo, Op) || + (OpNum + 4 != Record.size() && OpNum + 5 != Record.size())) + return error("Invalid record"); + + if (!isa<PointerType>(Op->getType())) + return error("Load operand is not a pointer type"); + + Type *Ty = nullptr; + if (OpNum + 5 == Record.size()) { + Ty = getTypeByID(Record[OpNum++]); + } else { + Ty = Op->getType()->getPointerElementType(); + } + + if (Error Err = typeCheckLoadStoreInst(Ty, Op->getType())) + return Err; + + AtomicOrdering Ordering = getDecodedOrdering(Record[OpNum + 2]); + if (Ordering == AtomicOrdering::NotAtomic || + Ordering == AtomicOrdering::Release || + Ordering == AtomicOrdering::AcquireRelease) + return error("Invalid record"); + if (Ordering != AtomicOrdering::NotAtomic && Record[OpNum] == 0) + return error("Invalid record"); + SyncScope::ID SSID = getDecodedSyncScopeID(Record[OpNum + 3]); + + MaybeAlign Align; + if (Error Err = parseAlignmentValue(Record[OpNum], Align)) + return Err; + if (!Align) + return error("Alignment missing from atomic load"); + I = new LoadInst(Ty, Op, "", Record[OpNum + 1], *Align, Ordering, SSID); + InstructionList.push_back(I); + break; + } + case bitc::FUNC_CODE_INST_STORE: + case bitc::FUNC_CODE_INST_STORE_OLD: { // STORE2:[ptrty, ptr, val, align, vol] + unsigned OpNum = 0; + Value *Val, *Ptr; + if (getValueTypePair(Record, OpNum, NextValueNo, Ptr) || + (BitCode == bitc::FUNC_CODE_INST_STORE + ? getValueTypePair(Record, OpNum, NextValueNo, Val) + : popValue(Record, OpNum, NextValueNo, + Ptr->getType()->getPointerElementType(), Val)) || + OpNum + 2 != Record.size()) + return error("Invalid record"); + + if (Error Err = typeCheckLoadStoreInst(Val->getType(), Ptr->getType())) + return Err; + MaybeAlign Align; + if (Error Err = parseAlignmentValue(Record[OpNum], Align)) + return Err; + SmallPtrSet<Type *, 4> Visited; + if (!Align && !Val->getType()->isSized(&Visited)) + return error("store of unsized type"); + if (!Align) + Align = TheModule->getDataLayout().getABITypeAlign(Val->getType()); + I = new StoreInst(Val, Ptr, Record[OpNum + 1], *Align); + InstructionList.push_back(I); + break; + } + case bitc::FUNC_CODE_INST_STOREATOMIC: + case bitc::FUNC_CODE_INST_STOREATOMIC_OLD: { + // STOREATOMIC: [ptrty, ptr, val, align, vol, ordering, ssid] + unsigned OpNum = 0; + Value *Val, *Ptr; + if (getValueTypePair(Record, OpNum, NextValueNo, Ptr) || + !isa<PointerType>(Ptr->getType()) || + (BitCode == bitc::FUNC_CODE_INST_STOREATOMIC + ? getValueTypePair(Record, OpNum, NextValueNo, Val) + : popValue(Record, OpNum, NextValueNo, + Ptr->getType()->getPointerElementType(), Val)) || + OpNum + 4 != Record.size()) + return error("Invalid record"); + + if (Error Err = typeCheckLoadStoreInst(Val->getType(), Ptr->getType())) + return Err; + AtomicOrdering Ordering = getDecodedOrdering(Record[OpNum + 2]); + if (Ordering == AtomicOrdering::NotAtomic || + Ordering == AtomicOrdering::Acquire || + Ordering == AtomicOrdering::AcquireRelease) + return error("Invalid record"); + SyncScope::ID SSID = getDecodedSyncScopeID(Record[OpNum + 3]); + if (Ordering != AtomicOrdering::NotAtomic && Record[OpNum] == 0) + return error("Invalid record"); + + MaybeAlign Align; + if (Error Err = parseAlignmentValue(Record[OpNum], Align)) + return Err; + if (!Align) + return error("Alignment missing from atomic store"); + I = new StoreInst(Val, Ptr, Record[OpNum + 1], *Align, Ordering, SSID); + InstructionList.push_back(I); + break; + } + case bitc::FUNC_CODE_INST_CMPXCHG_OLD: { + // CMPXCHG_OLD: [ptrty, ptr, cmp, val, vol, ordering, synchscope, + // failure_ordering?, weak?] + const size_t NumRecords = Record.size(); + unsigned OpNum = 0; + Value *Ptr = nullptr; + if (getValueTypePair(Record, OpNum, NextValueNo, Ptr)) + return error("Invalid record"); + + if (!isa<PointerType>(Ptr->getType())) + return error("Cmpxchg operand is not a pointer type"); + + Value *Cmp = nullptr; + if (popValue(Record, OpNum, NextValueNo, + cast<PointerType>(Ptr->getType())->getPointerElementType(), + Cmp)) + return error("Invalid record"); + + Value *New = nullptr; + if (popValue(Record, OpNum, NextValueNo, Cmp->getType(), New) || + NumRecords < OpNum + 3 || NumRecords > OpNum + 5) + return error("Invalid record"); + + const AtomicOrdering SuccessOrdering = + getDecodedOrdering(Record[OpNum + 1]); + if (SuccessOrdering == AtomicOrdering::NotAtomic || + SuccessOrdering == AtomicOrdering::Unordered) + return error("Invalid record"); + + const SyncScope::ID SSID = getDecodedSyncScopeID(Record[OpNum + 2]); + + if (Error Err = typeCheckLoadStoreInst(Cmp->getType(), Ptr->getType())) + return Err; + + const AtomicOrdering FailureOrdering = + NumRecords < 7 + ? AtomicCmpXchgInst::getStrongestFailureOrdering(SuccessOrdering) + : getDecodedOrdering(Record[OpNum + 3]); + + if (FailureOrdering == AtomicOrdering::NotAtomic || + FailureOrdering == AtomicOrdering::Unordered) + return error("Invalid record"); + + const Align Alignment( + TheModule->getDataLayout().getTypeStoreSize(Cmp->getType())); + + I = new AtomicCmpXchgInst(Ptr, Cmp, New, Alignment, SuccessOrdering, + FailureOrdering, SSID); + cast<AtomicCmpXchgInst>(I)->setVolatile(Record[OpNum]); + + if (NumRecords < 8) { + // Before weak cmpxchgs existed, the instruction simply returned the + // value loaded from memory, so bitcode files from that era will be + // expecting the first component of a modern cmpxchg. + CurBB->getInstList().push_back(I); + I = ExtractValueInst::Create(I, 0); + } else { + cast<AtomicCmpXchgInst>(I)->setWeak(Record[OpNum + 4]); + } + + InstructionList.push_back(I); + break; + } + case bitc::FUNC_CODE_INST_CMPXCHG: { + // CMPXCHG: [ptrty, ptr, cmp, val, vol, success_ordering, synchscope, + // failure_ordering, weak, align?] + const size_t NumRecords = Record.size(); + unsigned OpNum = 0; + Value *Ptr = nullptr; + if (getValueTypePair(Record, OpNum, NextValueNo, Ptr)) + return error("Invalid record"); + + if (!isa<PointerType>(Ptr->getType())) + return error("Cmpxchg operand is not a pointer type"); + + Value *Cmp = nullptr; + if (getValueTypePair(Record, OpNum, NextValueNo, Cmp)) + return error("Invalid record"); + + Value *Val = nullptr; + if (popValue(Record, OpNum, NextValueNo, Cmp->getType(), Val)) + return error("Invalid record"); + + if (NumRecords < OpNum + 3 || NumRecords > OpNum + 6) + return error("Invalid record"); + + const bool IsVol = Record[OpNum]; + + const AtomicOrdering SuccessOrdering = + getDecodedOrdering(Record[OpNum + 1]); + if (!AtomicCmpXchgInst::isValidSuccessOrdering(SuccessOrdering)) + return error("Invalid cmpxchg success ordering"); + + const SyncScope::ID SSID = getDecodedSyncScopeID(Record[OpNum + 2]); + + if (Error Err = typeCheckLoadStoreInst(Cmp->getType(), Ptr->getType())) + return Err; + + const AtomicOrdering FailureOrdering = + getDecodedOrdering(Record[OpNum + 3]); + if (!AtomicCmpXchgInst::isValidFailureOrdering(FailureOrdering)) + return error("Invalid cmpxchg failure ordering"); + + const bool IsWeak = Record[OpNum + 4]; + + MaybeAlign Alignment; + + if (NumRecords == (OpNum + 6)) { + if (Error Err = parseAlignmentValue(Record[OpNum + 5], Alignment)) + return Err; + } + if (!Alignment) + Alignment = + Align(TheModule->getDataLayout().getTypeStoreSize(Cmp->getType())); + + I = new AtomicCmpXchgInst(Ptr, Cmp, Val, *Alignment, SuccessOrdering, + FailureOrdering, SSID); + cast<AtomicCmpXchgInst>(I)->setVolatile(IsVol); + cast<AtomicCmpXchgInst>(I)->setWeak(IsWeak); + + InstructionList.push_back(I); + break; + } + case bitc::FUNC_CODE_INST_ATOMICRMW_OLD: + case bitc::FUNC_CODE_INST_ATOMICRMW: { + // ATOMICRMW_OLD: [ptrty, ptr, val, op, vol, ordering, ssid, align?] + // ATOMICRMW: [ptrty, ptr, valty, val, op, vol, ordering, ssid, align?] + const size_t NumRecords = Record.size(); + unsigned OpNum = 0; + + Value *Ptr = nullptr; + if (getValueTypePair(Record, OpNum, NextValueNo, Ptr)) + return error("Invalid record"); + + if (!isa<PointerType>(Ptr->getType())) + return error("Invalid record"); + + Value *Val = nullptr; + if (BitCode == bitc::FUNC_CODE_INST_ATOMICRMW_OLD) { + if (popValue(Record, OpNum, NextValueNo, + cast<PointerType>(Ptr->getType())->getPointerElementType(), + Val)) + return error("Invalid record"); + } else { + if (getValueTypePair(Record, OpNum, NextValueNo, Val)) + return error("Invalid record"); + } + + if (!(NumRecords == (OpNum + 4) || NumRecords == (OpNum + 5))) + return error("Invalid record"); + + const AtomicRMWInst::BinOp Operation = + getDecodedRMWOperation(Record[OpNum]); + if (Operation < AtomicRMWInst::FIRST_BINOP || + Operation > AtomicRMWInst::LAST_BINOP) + return error("Invalid record"); + + const bool IsVol = Record[OpNum + 1]; + + const AtomicOrdering Ordering = getDecodedOrdering(Record[OpNum + 2]); + if (Ordering == AtomicOrdering::NotAtomic || + Ordering == AtomicOrdering::Unordered) + return error("Invalid record"); + + const SyncScope::ID SSID = getDecodedSyncScopeID(Record[OpNum + 3]); + + MaybeAlign Alignment; + + if (NumRecords == (OpNum + 5)) { + if (Error Err = parseAlignmentValue(Record[OpNum + 4], Alignment)) + return Err; + } + + if (!Alignment) + Alignment = + Align(TheModule->getDataLayout().getTypeStoreSize(Val->getType())); + + I = new AtomicRMWInst(Operation, Ptr, Val, *Alignment, Ordering, SSID); + cast<AtomicRMWInst>(I)->setVolatile(IsVol); + + InstructionList.push_back(I); + break; + } + case bitc::FUNC_CODE_INST_FENCE: { // FENCE:[ordering, ssid] + if (2 != Record.size()) + return error("Invalid record"); + AtomicOrdering Ordering = getDecodedOrdering(Record[0]); + if (Ordering == AtomicOrdering::NotAtomic || + Ordering == AtomicOrdering::Unordered || + Ordering == AtomicOrdering::Monotonic) + return error("Invalid record"); + SyncScope::ID SSID = getDecodedSyncScopeID(Record[1]); + I = new FenceInst(Context, Ordering, SSID); + InstructionList.push_back(I); + break; + } + case bitc::FUNC_CODE_INST_CALL: { + // CALL: [paramattrs, cc, fmf, fnty, fnid, arg0, arg1...] + if (Record.size() < 3) + return error("Invalid record"); + + unsigned OpNum = 0; + AttributeList PAL = getAttributes(Record[OpNum++]); + unsigned CCInfo = Record[OpNum++]; + + FastMathFlags FMF; + if ((CCInfo >> bitc::CALL_FMF) & 1) { + FMF = getDecodedFastMathFlags(Record[OpNum++]); + if (!FMF.any()) + return error("Fast math flags indicator set for call with no FMF"); + } + + FunctionType *FTy = nullptr; + if ((CCInfo >> bitc::CALL_EXPLICIT_TYPE) & 1) { + FTy = dyn_cast<FunctionType>(getTypeByID(Record[OpNum++])); + if (!FTy) + return error("Explicit call type is not a function type"); + } + + Value *Callee; + if (getValueTypePair(Record, OpNum, NextValueNo, Callee)) + return error("Invalid record"); + + PointerType *OpTy = dyn_cast<PointerType>(Callee->getType()); + if (!OpTy) + return error("Callee is not a pointer type"); + if (!FTy) { + FTy = + dyn_cast<FunctionType>(Callee->getType()->getPointerElementType()); + if (!FTy) + return error("Callee is not of pointer to function type"); + } else if (!OpTy->isOpaqueOrPointeeTypeMatches(FTy)) + return error("Explicit call type does not match pointee type of " + "callee operand"); + if (Record.size() < FTy->getNumParams() + OpNum) + return error("Insufficient operands to call"); + + SmallVector<Value*, 16> Args; + SmallVector<Type *, 16> ArgsTys; + // Read the fixed params. + for (unsigned i = 0, e = FTy->getNumParams(); i != e; ++i, ++OpNum) { + if (FTy->getParamType(i)->isLabelTy()) + Args.push_back(getBasicBlock(Record[OpNum])); + else + Args.push_back(getValue(Record, OpNum, NextValueNo, + FTy->getParamType(i))); + ArgsTys.push_back(FTy->getParamType(i)); + if (!Args.back()) + return error("Invalid record"); + } + + // Read type/value pairs for varargs params. + if (!FTy->isVarArg()) { + if (OpNum != Record.size()) + return error("Invalid record"); + } else { + while (OpNum != Record.size()) { + Value *Op; + if (getValueTypePair(Record, OpNum, NextValueNo, Op)) + return error("Invalid record"); + Args.push_back(Op); + ArgsTys.push_back(Op->getType()); + } + } + + I = CallInst::Create(FTy, Callee, Args, OperandBundles); + OperandBundles.clear(); + InstructionList.push_back(I); + cast<CallInst>(I)->setCallingConv( + static_cast<CallingConv::ID>((0x7ff & CCInfo) >> bitc::CALL_CCONV)); + CallInst::TailCallKind TCK = CallInst::TCK_None; + if (CCInfo & 1 << bitc::CALL_TAIL) + TCK = CallInst::TCK_Tail; + if (CCInfo & (1 << bitc::CALL_MUSTTAIL)) + TCK = CallInst::TCK_MustTail; + if (CCInfo & (1 << bitc::CALL_NOTAIL)) + TCK = CallInst::TCK_NoTail; + cast<CallInst>(I)->setTailCallKind(TCK); + cast<CallInst>(I)->setAttributes(PAL); + propagateAttributeTypes(cast<CallBase>(I), ArgsTys); + if (FMF.any()) { + if (!isa<FPMathOperator>(I)) + return error("Fast-math-flags specified for call without " + "floating-point scalar or vector return type"); + I->setFastMathFlags(FMF); + } + break; + } + case bitc::FUNC_CODE_INST_VAARG: { // VAARG: [valistty, valist, instty] + if (Record.size() < 3) + return error("Invalid record"); + Type *OpTy = getTypeByID(Record[0]); + Value *Op = getValue(Record, 1, NextValueNo, OpTy); + Type *ResTy = getTypeByID(Record[2]); + if (!OpTy || !Op || !ResTy) + return error("Invalid record"); + I = new VAArgInst(Op, ResTy); + InstructionList.push_back(I); + break; + } + + case bitc::FUNC_CODE_OPERAND_BUNDLE: { + // A call or an invoke can be optionally prefixed with some variable + // number of operand bundle blocks. These blocks are read into + // OperandBundles and consumed at the next call or invoke instruction. + + if (Record.empty() || Record[0] >= BundleTags.size()) + return error("Invalid record"); + + std::vector<Value *> Inputs; + + unsigned OpNum = 1; + while (OpNum != Record.size()) { + Value *Op; + if (getValueTypePair(Record, OpNum, NextValueNo, Op)) + return error("Invalid record"); + Inputs.push_back(Op); + } + + OperandBundles.emplace_back(BundleTags[Record[0]], std::move(Inputs)); + continue; + } + + case bitc::FUNC_CODE_INST_FREEZE: { // FREEZE: [opty,opval] + unsigned OpNum = 0; + Value *Op = nullptr; + if (getValueTypePair(Record, OpNum, NextValueNo, Op)) + return error("Invalid record"); + if (OpNum != Record.size()) + return error("Invalid record"); + + I = new FreezeInst(Op); + InstructionList.push_back(I); + break; + } + } + + // Add instruction to end of current BB. If there is no current BB, reject + // this file. + if (!CurBB) { + I->deleteValue(); + return error("Invalid instruction with no BB"); + } + if (!OperandBundles.empty()) { + I->deleteValue(); + return error("Operand bundles found with no consumer"); + } + CurBB->getInstList().push_back(I); + + // If this was a terminator instruction, move to the next block. + if (I->isTerminator()) { + ++CurBBNo; + CurBB = CurBBNo < FunctionBBs.size() ? FunctionBBs[CurBBNo] : nullptr; + } + + // Non-void values get registered in the value table for future use. + if (!I->getType()->isVoidTy()) + ValueList.assignValue(I, NextValueNo++); + } + +OutOfRecordLoop: + + if (!OperandBundles.empty()) + return error("Operand bundles found with no consumer"); + + // Check the function list for unresolved values. + if (Argument *A = dyn_cast<Argument>(ValueList.back())) { + if (!A->getParent()) { + // We found at least one unresolved value. Nuke them all to avoid leaks. + for (unsigned i = ModuleValueListSize, e = ValueList.size(); i != e; ++i){ + if ((A = dyn_cast_or_null<Argument>(ValueList[i])) && !A->getParent()) { + A->replaceAllUsesWith(UndefValue::get(A->getType())); + delete A; + } + } + return error("Never resolved value found in function"); + } + } + + // Unexpected unresolved metadata about to be dropped. + if (MDLoader->hasFwdRefs()) + return error("Invalid function metadata: outgoing forward refs"); + + // Trim the value list down to the size it was before we parsed this function. + ValueList.shrinkTo(ModuleValueListSize); + MDLoader->shrinkTo(ModuleMDLoaderSize); + std::vector<BasicBlock*>().swap(FunctionBBs); + return Error::success(); +} + +/// Find the function body in the bitcode stream +Error BitcodeReader::findFunctionInStream( + Function *F, + DenseMap<Function *, uint64_t>::iterator DeferredFunctionInfoIterator) { + while (DeferredFunctionInfoIterator->second == 0) { + // This is the fallback handling for the old format bitcode that + // didn't contain the function index in the VST, or when we have + // an anonymous function which would not have a VST entry. + // Assert that we have one of those two cases. + assert(VSTOffset == 0 || !F->hasName()); + // Parse the next body in the stream and set its position in the + // DeferredFunctionInfo map. + if (Error Err = rememberAndSkipFunctionBodies()) + return Err; + } + return Error::success(); +} + +SyncScope::ID BitcodeReader::getDecodedSyncScopeID(unsigned Val) { + if (Val == SyncScope::SingleThread || Val == SyncScope::System) + return SyncScope::ID(Val); + if (Val >= SSIDs.size()) + return SyncScope::System; // Map unknown synchronization scopes to system. + return SSIDs[Val]; +} + +//===----------------------------------------------------------------------===// +// GVMaterializer implementation +//===----------------------------------------------------------------------===// + +Error BitcodeReader::materialize(GlobalValue *GV) { + Function *F = dyn_cast<Function>(GV); + // If it's not a function or is already material, ignore the request. + if (!F || !F->isMaterializable()) + return Error::success(); + + DenseMap<Function*, uint64_t>::iterator DFII = DeferredFunctionInfo.find(F); + assert(DFII != DeferredFunctionInfo.end() && "Deferred function not found!"); + // If its position is recorded as 0, its body is somewhere in the stream + // but we haven't seen it yet. + if (DFII->second == 0) + if (Error Err = findFunctionInStream(F, DFII)) + return Err; + + // Materialize metadata before parsing any function bodies. + if (Error Err = materializeMetadata()) + return Err; + + // Move the bit stream to the saved position of the deferred function body. + if (Error JumpFailed = Stream.JumpToBit(DFII->second)) + return JumpFailed; + if (Error Err = parseFunctionBody(F)) + return Err; + F->setIsMaterializable(false); + + if (StripDebugInfo) + stripDebugInfo(*F); + + // Upgrade any old intrinsic calls in the function. + for (auto &I : UpgradedIntrinsics) { + for (User *U : llvm::make_early_inc_range(I.first->materialized_users())) + if (CallInst *CI = dyn_cast<CallInst>(U)) + UpgradeIntrinsicCall(CI, I.second); + } + + // Update calls to the remangled intrinsics + for (auto &I : RemangledIntrinsics) + for (User *U : llvm::make_early_inc_range(I.first->materialized_users())) + // Don't expect any other users than call sites + cast<CallBase>(U)->setCalledFunction(I.second); + + // Finish fn->subprogram upgrade for materialized functions. + if (DISubprogram *SP = MDLoader->lookupSubprogramForFunction(F)) + F->setSubprogram(SP); + + // Check if the TBAA Metadata are valid, otherwise we will need to strip them. + if (!MDLoader->isStrippingTBAA()) { + for (auto &I : instructions(F)) { + MDNode *TBAA = I.getMetadata(LLVMContext::MD_tbaa); + if (!TBAA || TBAAVerifyHelper.visitTBAAMetadata(I, TBAA)) + continue; + MDLoader->setStripTBAA(true); + stripTBAA(F->getParent()); + } + } + + for (auto &I : instructions(F)) { + // "Upgrade" older incorrect branch weights by dropping them. + if (auto *MD = I.getMetadata(LLVMContext::MD_prof)) { + if (MD->getOperand(0) != nullptr && isa<MDString>(MD->getOperand(0))) { + MDString *MDS = cast<MDString>(MD->getOperand(0)); + StringRef ProfName = MDS->getString(); + // Check consistency of !prof branch_weights metadata. + if (!ProfName.equals("branch_weights")) + continue; + unsigned ExpectedNumOperands = 0; + if (BranchInst *BI = dyn_cast<BranchInst>(&I)) + ExpectedNumOperands = BI->getNumSuccessors(); + else if (SwitchInst *SI = dyn_cast<SwitchInst>(&I)) + ExpectedNumOperands = SI->getNumSuccessors(); + else if (isa<CallInst>(&I)) + ExpectedNumOperands = 1; + else if (IndirectBrInst *IBI = dyn_cast<IndirectBrInst>(&I)) + ExpectedNumOperands = IBI->getNumDestinations(); + else if (isa<SelectInst>(&I)) + ExpectedNumOperands = 2; + else + continue; // ignore and continue. + + // If branch weight doesn't match, just strip branch weight. + if (MD->getNumOperands() != 1 + ExpectedNumOperands) + I.setMetadata(LLVMContext::MD_prof, nullptr); + } + } + + // Remove incompatible attributes on function calls. + if (auto *CI = dyn_cast<CallBase>(&I)) { + CI->removeRetAttrs(AttributeFuncs::typeIncompatible( + CI->getFunctionType()->getReturnType())); + + for (unsigned ArgNo = 0; ArgNo < CI->arg_size(); ++ArgNo) + CI->removeParamAttrs(ArgNo, AttributeFuncs::typeIncompatible( + CI->getArgOperand(ArgNo)->getType())); + } + } + + // Look for functions that rely on old function attribute behavior. + UpgradeFunctionAttributes(*F); + + // Bring in any functions that this function forward-referenced via + // blockaddresses. + return materializeForwardReferencedFunctions(); +} + +Error BitcodeReader::materializeModule() { + if (Error Err = materializeMetadata()) + return Err; + + // Promise to materialize all forward references. + WillMaterializeAllForwardRefs = true; + + // Iterate over the module, deserializing any functions that are still on + // disk. + for (Function &F : *TheModule) { + if (Error Err = materialize(&F)) + return Err; + } + // At this point, if there are any function bodies, parse the rest of + // the bits in the module past the last function block we have recorded + // through either lazy scanning or the VST. + if (LastFunctionBlockBit || NextUnreadBit) + if (Error Err = parseModule(LastFunctionBlockBit > NextUnreadBit + ? LastFunctionBlockBit + : NextUnreadBit)) + return Err; + + // Check that all block address forward references got resolved (as we + // promised above). + if (!BasicBlockFwdRefs.empty()) + return error("Never resolved function from blockaddress"); + + // Upgrade any intrinsic calls that slipped through (should not happen!) and + // delete the old functions to clean up. We can't do this unless the entire + // module is materialized because there could always be another function body + // with calls to the old function. + for (auto &I : UpgradedIntrinsics) { + for (auto *U : I.first->users()) { + if (CallInst *CI = dyn_cast<CallInst>(U)) + UpgradeIntrinsicCall(CI, I.second); + } + if (!I.first->use_empty()) + I.first->replaceAllUsesWith(I.second); + I.first->eraseFromParent(); + } + UpgradedIntrinsics.clear(); + // Do the same for remangled intrinsics + for (auto &I : RemangledIntrinsics) { + I.first->replaceAllUsesWith(I.second); + I.first->eraseFromParent(); + } + RemangledIntrinsics.clear(); + + UpgradeDebugInfo(*TheModule); + + UpgradeModuleFlags(*TheModule); + + UpgradeARCRuntime(*TheModule); + + return Error::success(); +} + +std::vector<StructType *> BitcodeReader::getIdentifiedStructTypes() const { + return IdentifiedStructTypes; +} + +ModuleSummaryIndexBitcodeReader::ModuleSummaryIndexBitcodeReader( + BitstreamCursor Cursor, StringRef Strtab, ModuleSummaryIndex &TheIndex, + StringRef ModulePath, unsigned ModuleId) + : BitcodeReaderBase(std::move(Cursor), Strtab), TheIndex(TheIndex), + ModulePath(ModulePath), ModuleId(ModuleId) {} + +void ModuleSummaryIndexBitcodeReader::addThisModule() { + TheIndex.addModule(ModulePath, ModuleId); +} + +ModuleSummaryIndex::ModuleInfo * +ModuleSummaryIndexBitcodeReader::getThisModule() { + return TheIndex.getModule(ModulePath); +} + +std::pair<ValueInfo, GlobalValue::GUID> +ModuleSummaryIndexBitcodeReader::getValueInfoFromValueId(unsigned ValueId) { + auto VGI = ValueIdToValueInfoMap[ValueId]; + assert(VGI.first); + return VGI; +} + +void ModuleSummaryIndexBitcodeReader::setValueGUID( + uint64_t ValueID, StringRef ValueName, GlobalValue::LinkageTypes Linkage, + StringRef SourceFileName) { + std::string GlobalId = + GlobalValue::getGlobalIdentifier(ValueName, Linkage, SourceFileName); + auto ValueGUID = GlobalValue::getGUID(GlobalId); + auto OriginalNameID = ValueGUID; + if (GlobalValue::isLocalLinkage(Linkage)) + OriginalNameID = GlobalValue::getGUID(ValueName); + if (PrintSummaryGUIDs) + dbgs() << "GUID " << ValueGUID << "(" << OriginalNameID << ") is " + << ValueName << "\n"; + + // UseStrtab is false for legacy summary formats and value names are + // created on stack. In that case we save the name in a string saver in + // the index so that the value name can be recorded. + ValueIdToValueInfoMap[ValueID] = std::make_pair( + TheIndex.getOrInsertValueInfo( + ValueGUID, + UseStrtab ? ValueName : TheIndex.saveString(ValueName)), + OriginalNameID); +} + +// Specialized value symbol table parser used when reading module index +// blocks where we don't actually create global values. The parsed information +// is saved in the bitcode reader for use when later parsing summaries. +Error ModuleSummaryIndexBitcodeReader::parseValueSymbolTable( + uint64_t Offset, + DenseMap<unsigned, GlobalValue::LinkageTypes> &ValueIdToLinkageMap) { + // With a strtab the VST is not required to parse the summary. + if (UseStrtab) + return Error::success(); + + assert(Offset > 0 && "Expected non-zero VST offset"); + Expected<uint64_t> MaybeCurrentBit = jumpToValueSymbolTable(Offset, Stream); + if (!MaybeCurrentBit) + return MaybeCurrentBit.takeError(); + uint64_t CurrentBit = MaybeCurrentBit.get(); + + if (Error Err = Stream.EnterSubBlock(bitc::VALUE_SYMTAB_BLOCK_ID)) + return Err; + + SmallVector<uint64_t, 64> Record; + + // Read all the records for this value table. + SmallString<128> ValueName; + + while (true) { + Expected<BitstreamEntry> MaybeEntry = Stream.advanceSkippingSubblocks(); + if (!MaybeEntry) + return MaybeEntry.takeError(); + BitstreamEntry Entry = MaybeEntry.get(); + + switch (Entry.Kind) { + case BitstreamEntry::SubBlock: // Handled for us already. + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + // Done parsing VST, jump back to wherever we came from. + if (Error JumpFailed = Stream.JumpToBit(CurrentBit)) + return JumpFailed; + return Error::success(); + case BitstreamEntry::Record: + // The interesting case. + break; + } + + // Read a record. + Record.clear(); + Expected<unsigned> MaybeRecord = Stream.readRecord(Entry.ID, Record); + if (!MaybeRecord) + return MaybeRecord.takeError(); + switch (MaybeRecord.get()) { + default: // Default behavior: ignore (e.g. VST_CODE_BBENTRY records). + break; + case bitc::VST_CODE_ENTRY: { // VST_CODE_ENTRY: [valueid, namechar x N] + if (convertToString(Record, 1, ValueName)) + return error("Invalid record"); + unsigned ValueID = Record[0]; + assert(!SourceFileName.empty()); + auto VLI = ValueIdToLinkageMap.find(ValueID); + assert(VLI != ValueIdToLinkageMap.end() && + "No linkage found for VST entry?"); + auto Linkage = VLI->second; + setValueGUID(ValueID, ValueName, Linkage, SourceFileName); + ValueName.clear(); + break; + } + case bitc::VST_CODE_FNENTRY: { + // VST_CODE_FNENTRY: [valueid, offset, namechar x N] + if (convertToString(Record, 2, ValueName)) + return error("Invalid record"); + unsigned ValueID = Record[0]; + assert(!SourceFileName.empty()); + auto VLI = ValueIdToLinkageMap.find(ValueID); + assert(VLI != ValueIdToLinkageMap.end() && + "No linkage found for VST entry?"); + auto Linkage = VLI->second; + setValueGUID(ValueID, ValueName, Linkage, SourceFileName); + ValueName.clear(); + break; + } + case bitc::VST_CODE_COMBINED_ENTRY: { + // VST_CODE_COMBINED_ENTRY: [valueid, refguid] + unsigned ValueID = Record[0]; + GlobalValue::GUID RefGUID = Record[1]; + // The "original name", which is the second value of the pair will be + // overriden later by a FS_COMBINED_ORIGINAL_NAME in the combined index. + ValueIdToValueInfoMap[ValueID] = + std::make_pair(TheIndex.getOrInsertValueInfo(RefGUID), RefGUID); + break; + } + } + } +} + +// Parse just the blocks needed for building the index out of the module. +// At the end of this routine the module Index is populated with a map +// from global value id to GlobalValueSummary objects. +Error ModuleSummaryIndexBitcodeReader::parseModule() { + if (Error Err = Stream.EnterSubBlock(bitc::MODULE_BLOCK_ID)) + return Err; + + SmallVector<uint64_t, 64> Record; + DenseMap<unsigned, GlobalValue::LinkageTypes> ValueIdToLinkageMap; + unsigned ValueId = 0; + + // Read the index for this module. + while (true) { + Expected<llvm::BitstreamEntry> MaybeEntry = Stream.advance(); + if (!MaybeEntry) + return MaybeEntry.takeError(); + llvm::BitstreamEntry Entry = MaybeEntry.get(); + + switch (Entry.Kind) { + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + return Error::success(); + + case BitstreamEntry::SubBlock: + switch (Entry.ID) { + default: // Skip unknown content. + if (Error Err = Stream.SkipBlock()) + return Err; + break; + case bitc::BLOCKINFO_BLOCK_ID: + // Need to parse these to get abbrev ids (e.g. for VST) + if (readBlockInfo()) + return error("Malformed block"); + break; + case bitc::VALUE_SYMTAB_BLOCK_ID: + // Should have been parsed earlier via VSTOffset, unless there + // is no summary section. + assert(((SeenValueSymbolTable && VSTOffset > 0) || + !SeenGlobalValSummary) && + "Expected early VST parse via VSTOffset record"); + if (Error Err = Stream.SkipBlock()) + return Err; + break; + case bitc::GLOBALVAL_SUMMARY_BLOCK_ID: + case bitc::FULL_LTO_GLOBALVAL_SUMMARY_BLOCK_ID: + // Add the module if it is a per-module index (has a source file name). + if (!SourceFileName.empty()) + addThisModule(); + assert(!SeenValueSymbolTable && + "Already read VST when parsing summary block?"); + // We might not have a VST if there were no values in the + // summary. An empty summary block generated when we are + // performing ThinLTO compiles so we don't later invoke + // the regular LTO process on them. + if (VSTOffset > 0) { + if (Error Err = parseValueSymbolTable(VSTOffset, ValueIdToLinkageMap)) + return Err; + SeenValueSymbolTable = true; + } + SeenGlobalValSummary = true; + if (Error Err = parseEntireSummary(Entry.ID)) + return Err; + break; + case bitc::MODULE_STRTAB_BLOCK_ID: + if (Error Err = parseModuleStringTable()) + return Err; + break; + } + continue; + + case BitstreamEntry::Record: { + Record.clear(); + Expected<unsigned> MaybeBitCode = Stream.readRecord(Entry.ID, Record); + if (!MaybeBitCode) + return MaybeBitCode.takeError(); + switch (MaybeBitCode.get()) { + default: + break; // Default behavior, ignore unknown content. + case bitc::MODULE_CODE_VERSION: { + if (Error Err = parseVersionRecord(Record).takeError()) + return Err; + break; + } + /// MODULE_CODE_SOURCE_FILENAME: [namechar x N] + case bitc::MODULE_CODE_SOURCE_FILENAME: { + SmallString<128> ValueName; + if (convertToString(Record, 0, ValueName)) + return error("Invalid record"); + SourceFileName = ValueName.c_str(); + break; + } + /// MODULE_CODE_HASH: [5*i32] + case bitc::MODULE_CODE_HASH: { + if (Record.size() != 5) + return error("Invalid hash length " + Twine(Record.size()).str()); + auto &Hash = getThisModule()->second.second; + int Pos = 0; + for (auto &Val : Record) { + assert(!(Val >> 32) && "Unexpected high bits set"); + Hash[Pos++] = Val; + } + break; + } + /// MODULE_CODE_VSTOFFSET: [offset] + case bitc::MODULE_CODE_VSTOFFSET: + if (Record.empty()) + return error("Invalid record"); + // Note that we subtract 1 here because the offset is relative to one + // word before the start of the identification or module block, which + // was historically always the start of the regular bitcode header. + VSTOffset = Record[0] - 1; + break; + // v1 GLOBALVAR: [pointer type, isconst, initid, linkage, ...] + // v1 FUNCTION: [type, callingconv, isproto, linkage, ...] + // v1 ALIAS: [alias type, addrspace, aliasee val#, linkage, ...] + // v2: [strtab offset, strtab size, v1] + case bitc::MODULE_CODE_GLOBALVAR: + case bitc::MODULE_CODE_FUNCTION: + case bitc::MODULE_CODE_ALIAS: { + StringRef Name; + ArrayRef<uint64_t> GVRecord; + std::tie(Name, GVRecord) = readNameFromStrtab(Record); + if (GVRecord.size() <= 3) + return error("Invalid record"); + uint64_t RawLinkage = GVRecord[3]; + GlobalValue::LinkageTypes Linkage = getDecodedLinkage(RawLinkage); + if (!UseStrtab) { + ValueIdToLinkageMap[ValueId++] = Linkage; + break; + } + + setValueGUID(ValueId++, Name, Linkage, SourceFileName); + break; + } + } + } + continue; + } + } +} + +std::vector<ValueInfo> +ModuleSummaryIndexBitcodeReader::makeRefList(ArrayRef<uint64_t> Record) { + std::vector<ValueInfo> Ret; + Ret.reserve(Record.size()); + for (uint64_t RefValueId : Record) + Ret.push_back(getValueInfoFromValueId(RefValueId).first); + return Ret; +} + +std::vector<FunctionSummary::EdgeTy> +ModuleSummaryIndexBitcodeReader::makeCallList(ArrayRef<uint64_t> Record, + bool IsOldProfileFormat, + bool HasProfile, bool HasRelBF) { + std::vector<FunctionSummary::EdgeTy> Ret; + Ret.reserve(Record.size()); + for (unsigned I = 0, E = Record.size(); I != E; ++I) { + CalleeInfo::HotnessType Hotness = CalleeInfo::HotnessType::Unknown; + uint64_t RelBF = 0; + ValueInfo Callee = getValueInfoFromValueId(Record[I]).first; + if (IsOldProfileFormat) { + I += 1; // Skip old callsitecount field + if (HasProfile) + I += 1; // Skip old profilecount field + } else if (HasProfile) + Hotness = static_cast<CalleeInfo::HotnessType>(Record[++I]); + else if (HasRelBF) + RelBF = Record[++I]; + Ret.push_back(FunctionSummary::EdgeTy{Callee, CalleeInfo(Hotness, RelBF)}); + } + return Ret; +} + +static void +parseWholeProgramDevirtResolutionByArg(ArrayRef<uint64_t> Record, size_t &Slot, + WholeProgramDevirtResolution &Wpd) { + uint64_t ArgNum = Record[Slot++]; + WholeProgramDevirtResolution::ByArg &B = + Wpd.ResByArg[{Record.begin() + Slot, Record.begin() + Slot + ArgNum}]; + Slot += ArgNum; + + B.TheKind = + static_cast<WholeProgramDevirtResolution::ByArg::Kind>(Record[Slot++]); + B.Info = Record[Slot++]; + B.Byte = Record[Slot++]; + B.Bit = Record[Slot++]; +} + +static void parseWholeProgramDevirtResolution(ArrayRef<uint64_t> Record, + StringRef Strtab, size_t &Slot, + TypeIdSummary &TypeId) { + uint64_t Id = Record[Slot++]; + WholeProgramDevirtResolution &Wpd = TypeId.WPDRes[Id]; + + Wpd.TheKind = static_cast<WholeProgramDevirtResolution::Kind>(Record[Slot++]); + Wpd.SingleImplName = {Strtab.data() + Record[Slot], + static_cast<size_t>(Record[Slot + 1])}; + Slot += 2; + + uint64_t ResByArgNum = Record[Slot++]; + for (uint64_t I = 0; I != ResByArgNum; ++I) + parseWholeProgramDevirtResolutionByArg(Record, Slot, Wpd); +} + +static void parseTypeIdSummaryRecord(ArrayRef<uint64_t> Record, + StringRef Strtab, + ModuleSummaryIndex &TheIndex) { + size_t Slot = 0; + TypeIdSummary &TypeId = TheIndex.getOrInsertTypeIdSummary( + {Strtab.data() + Record[Slot], static_cast<size_t>(Record[Slot + 1])}); + Slot += 2; + + TypeId.TTRes.TheKind = static_cast<TypeTestResolution::Kind>(Record[Slot++]); + TypeId.TTRes.SizeM1BitWidth = Record[Slot++]; + TypeId.TTRes.AlignLog2 = Record[Slot++]; + TypeId.TTRes.SizeM1 = Record[Slot++]; + TypeId.TTRes.BitMask = Record[Slot++]; + TypeId.TTRes.InlineBits = Record[Slot++]; + + while (Slot < Record.size()) + parseWholeProgramDevirtResolution(Record, Strtab, Slot, TypeId); +} + +std::vector<FunctionSummary::ParamAccess> +ModuleSummaryIndexBitcodeReader::parseParamAccesses(ArrayRef<uint64_t> Record) { + auto ReadRange = [&]() { + APInt Lower(FunctionSummary::ParamAccess::RangeWidth, + BitcodeReader::decodeSignRotatedValue(Record.front())); + Record = Record.drop_front(); + APInt Upper(FunctionSummary::ParamAccess::RangeWidth, + BitcodeReader::decodeSignRotatedValue(Record.front())); + Record = Record.drop_front(); + ConstantRange Range{Lower, Upper}; + assert(!Range.isFullSet()); + assert(!Range.isUpperSignWrapped()); + return Range; + }; + + std::vector<FunctionSummary::ParamAccess> PendingParamAccesses; + while (!Record.empty()) { + PendingParamAccesses.emplace_back(); + FunctionSummary::ParamAccess &ParamAccess = PendingParamAccesses.back(); + ParamAccess.ParamNo = Record.front(); + Record = Record.drop_front(); + ParamAccess.Use = ReadRange(); + ParamAccess.Calls.resize(Record.front()); + Record = Record.drop_front(); + for (auto &Call : ParamAccess.Calls) { + Call.ParamNo = Record.front(); + Record = Record.drop_front(); + Call.Callee = getValueInfoFromValueId(Record.front()).first; + Record = Record.drop_front(); + Call.Offsets = ReadRange(); + } + } + return PendingParamAccesses; +} + +void ModuleSummaryIndexBitcodeReader::parseTypeIdCompatibleVtableInfo( + ArrayRef<uint64_t> Record, size_t &Slot, + TypeIdCompatibleVtableInfo &TypeId) { + uint64_t Offset = Record[Slot++]; + ValueInfo Callee = getValueInfoFromValueId(Record[Slot++]).first; + TypeId.push_back({Offset, Callee}); +} + +void ModuleSummaryIndexBitcodeReader::parseTypeIdCompatibleVtableSummaryRecord( + ArrayRef<uint64_t> Record) { + size_t Slot = 0; + TypeIdCompatibleVtableInfo &TypeId = + TheIndex.getOrInsertTypeIdCompatibleVtableSummary( + {Strtab.data() + Record[Slot], + static_cast<size_t>(Record[Slot + 1])}); + Slot += 2; + + while (Slot < Record.size()) + parseTypeIdCompatibleVtableInfo(Record, Slot, TypeId); +} + +static void setSpecialRefs(std::vector<ValueInfo> &Refs, unsigned ROCnt, + unsigned WOCnt) { + // Readonly and writeonly refs are in the end of the refs list. + assert(ROCnt + WOCnt <= Refs.size()); + unsigned FirstWORef = Refs.size() - WOCnt; + unsigned RefNo = FirstWORef - ROCnt; + for (; RefNo < FirstWORef; ++RefNo) + Refs[RefNo].setReadOnly(); + for (; RefNo < Refs.size(); ++RefNo) + Refs[RefNo].setWriteOnly(); +} + +// Eagerly parse the entire summary block. This populates the GlobalValueSummary +// objects in the index. +Error ModuleSummaryIndexBitcodeReader::parseEntireSummary(unsigned ID) { + if (Error Err = Stream.EnterSubBlock(ID)) + return Err; + SmallVector<uint64_t, 64> Record; + + // Parse version + { + Expected<BitstreamEntry> MaybeEntry = Stream.advanceSkippingSubblocks(); + if (!MaybeEntry) + return MaybeEntry.takeError(); + BitstreamEntry Entry = MaybeEntry.get(); + + if (Entry.Kind != BitstreamEntry::Record) + return error("Invalid Summary Block: record for version expected"); + Expected<unsigned> MaybeRecord = Stream.readRecord(Entry.ID, Record); + if (!MaybeRecord) + return MaybeRecord.takeError(); + if (MaybeRecord.get() != bitc::FS_VERSION) + return error("Invalid Summary Block: version expected"); + } + const uint64_t Version = Record[0]; + const bool IsOldProfileFormat = Version == 1; + if (Version < 1 || Version > ModuleSummaryIndex::BitcodeSummaryVersion) + return error("Invalid summary version " + Twine(Version) + + ". Version should be in the range [1-" + + Twine(ModuleSummaryIndex::BitcodeSummaryVersion) + + "]."); + Record.clear(); + + // Keep around the last seen summary to be used when we see an optional + // "OriginalName" attachement. + GlobalValueSummary *LastSeenSummary = nullptr; + GlobalValue::GUID LastSeenGUID = 0; + + // We can expect to see any number of type ID information records before + // each function summary records; these variables store the information + // collected so far so that it can be used to create the summary object. + std::vector<GlobalValue::GUID> PendingTypeTests; + std::vector<FunctionSummary::VFuncId> PendingTypeTestAssumeVCalls, + PendingTypeCheckedLoadVCalls; + std::vector<FunctionSummary::ConstVCall> PendingTypeTestAssumeConstVCalls, + PendingTypeCheckedLoadConstVCalls; + std::vector<FunctionSummary::ParamAccess> PendingParamAccesses; + + while (true) { + Expected<BitstreamEntry> MaybeEntry = Stream.advanceSkippingSubblocks(); + if (!MaybeEntry) + return MaybeEntry.takeError(); + BitstreamEntry Entry = MaybeEntry.get(); + + switch (Entry.Kind) { + case BitstreamEntry::SubBlock: // Handled for us already. + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + return Error::success(); + case BitstreamEntry::Record: + // The interesting case. + break; + } + + // Read a record. The record format depends on whether this + // is a per-module index or a combined index file. In the per-module + // case the records contain the associated value's ID for correlation + // with VST entries. In the combined index the correlation is done + // via the bitcode offset of the summary records (which were saved + // in the combined index VST entries). The records also contain + // information used for ThinLTO renaming and importing. + Record.clear(); + Expected<unsigned> MaybeBitCode = Stream.readRecord(Entry.ID, Record); + if (!MaybeBitCode) + return MaybeBitCode.takeError(); + switch (unsigned BitCode = MaybeBitCode.get()) { + default: // Default behavior: ignore. + break; + case bitc::FS_FLAGS: { // [flags] + TheIndex.setFlags(Record[0]); + break; + } + case bitc::FS_VALUE_GUID: { // [valueid, refguid] + uint64_t ValueID = Record[0]; + GlobalValue::GUID RefGUID = Record[1]; + ValueIdToValueInfoMap[ValueID] = + std::make_pair(TheIndex.getOrInsertValueInfo(RefGUID), RefGUID); + break; + } + // FS_PERMODULE: [valueid, flags, instcount, fflags, numrefs, + // numrefs x valueid, n x (valueid)] + // FS_PERMODULE_PROFILE: [valueid, flags, instcount, fflags, numrefs, + // numrefs x valueid, + // n x (valueid, hotness)] + // FS_PERMODULE_RELBF: [valueid, flags, instcount, fflags, numrefs, + // numrefs x valueid, + // n x (valueid, relblockfreq)] + case bitc::FS_PERMODULE: + case bitc::FS_PERMODULE_RELBF: + case bitc::FS_PERMODULE_PROFILE: { + unsigned ValueID = Record[0]; + uint64_t RawFlags = Record[1]; + unsigned InstCount = Record[2]; + uint64_t RawFunFlags = 0; + unsigned NumRefs = Record[3]; + unsigned NumRORefs = 0, NumWORefs = 0; + int RefListStartIndex = 4; + if (Version >= 4) { + RawFunFlags = Record[3]; + NumRefs = Record[4]; + RefListStartIndex = 5; + if (Version >= 5) { + NumRORefs = Record[5]; + RefListStartIndex = 6; + if (Version >= 7) { + NumWORefs = Record[6]; + RefListStartIndex = 7; + } + } + } + + auto Flags = getDecodedGVSummaryFlags(RawFlags, Version); + // The module path string ref set in the summary must be owned by the + // index's module string table. Since we don't have a module path + // string table section in the per-module index, we create a single + // module path string table entry with an empty (0) ID to take + // ownership. + int CallGraphEdgeStartIndex = RefListStartIndex + NumRefs; + assert(Record.size() >= RefListStartIndex + NumRefs && + "Record size inconsistent with number of references"); + std::vector<ValueInfo> Refs = makeRefList( + ArrayRef<uint64_t>(Record).slice(RefListStartIndex, NumRefs)); + bool HasProfile = (BitCode == bitc::FS_PERMODULE_PROFILE); + bool HasRelBF = (BitCode == bitc::FS_PERMODULE_RELBF); + std::vector<FunctionSummary::EdgeTy> Calls = makeCallList( + ArrayRef<uint64_t>(Record).slice(CallGraphEdgeStartIndex), + IsOldProfileFormat, HasProfile, HasRelBF); + setSpecialRefs(Refs, NumRORefs, NumWORefs); + auto FS = std::make_unique<FunctionSummary>( + Flags, InstCount, getDecodedFFlags(RawFunFlags), /*EntryCount=*/0, + std::move(Refs), std::move(Calls), std::move(PendingTypeTests), + std::move(PendingTypeTestAssumeVCalls), + std::move(PendingTypeCheckedLoadVCalls), + std::move(PendingTypeTestAssumeConstVCalls), + std::move(PendingTypeCheckedLoadConstVCalls), + std::move(PendingParamAccesses)); + auto VIAndOriginalGUID = getValueInfoFromValueId(ValueID); + FS->setModulePath(getThisModule()->first()); + FS->setOriginalName(VIAndOriginalGUID.second); + TheIndex.addGlobalValueSummary(VIAndOriginalGUID.first, std::move(FS)); + break; + } + // FS_ALIAS: [valueid, flags, valueid] + // Aliases must be emitted (and parsed) after all FS_PERMODULE entries, as + // they expect all aliasee summaries to be available. + case bitc::FS_ALIAS: { + unsigned ValueID = Record[0]; + uint64_t RawFlags = Record[1]; + unsigned AliaseeID = Record[2]; + auto Flags = getDecodedGVSummaryFlags(RawFlags, Version); + auto AS = std::make_unique<AliasSummary>(Flags); + // The module path string ref set in the summary must be owned by the + // index's module string table. Since we don't have a module path + // string table section in the per-module index, we create a single + // module path string table entry with an empty (0) ID to take + // ownership. + AS->setModulePath(getThisModule()->first()); + + auto AliaseeVI = getValueInfoFromValueId(AliaseeID).first; + auto AliaseeInModule = TheIndex.findSummaryInModule(AliaseeVI, ModulePath); + if (!AliaseeInModule) + return error("Alias expects aliasee summary to be parsed"); + AS->setAliasee(AliaseeVI, AliaseeInModule); + + auto GUID = getValueInfoFromValueId(ValueID); + AS->setOriginalName(GUID.second); + TheIndex.addGlobalValueSummary(GUID.first, std::move(AS)); + break; + } + // FS_PERMODULE_GLOBALVAR_INIT_REFS: [valueid, flags, varflags, n x valueid] + case bitc::FS_PERMODULE_GLOBALVAR_INIT_REFS: { + unsigned ValueID = Record[0]; + uint64_t RawFlags = Record[1]; + unsigned RefArrayStart = 2; + GlobalVarSummary::GVarFlags GVF(/* ReadOnly */ false, + /* WriteOnly */ false, + /* Constant */ false, + GlobalObject::VCallVisibilityPublic); + auto Flags = getDecodedGVSummaryFlags(RawFlags, Version); + if (Version >= 5) { + GVF = getDecodedGVarFlags(Record[2]); + RefArrayStart = 3; + } + std::vector<ValueInfo> Refs = + makeRefList(ArrayRef<uint64_t>(Record).slice(RefArrayStart)); + auto FS = + std::make_unique<GlobalVarSummary>(Flags, GVF, std::move(Refs)); + FS->setModulePath(getThisModule()->first()); + auto GUID = getValueInfoFromValueId(ValueID); + FS->setOriginalName(GUID.second); + TheIndex.addGlobalValueSummary(GUID.first, std::move(FS)); + break; + } + // FS_PERMODULE_VTABLE_GLOBALVAR_INIT_REFS: [valueid, flags, varflags, + // numrefs, numrefs x valueid, + // n x (valueid, offset)] + case bitc::FS_PERMODULE_VTABLE_GLOBALVAR_INIT_REFS: { + unsigned ValueID = Record[0]; + uint64_t RawFlags = Record[1]; + GlobalVarSummary::GVarFlags GVF = getDecodedGVarFlags(Record[2]); + unsigned NumRefs = Record[3]; + unsigned RefListStartIndex = 4; + unsigned VTableListStartIndex = RefListStartIndex + NumRefs; + auto Flags = getDecodedGVSummaryFlags(RawFlags, Version); + std::vector<ValueInfo> Refs = makeRefList( + ArrayRef<uint64_t>(Record).slice(RefListStartIndex, NumRefs)); + VTableFuncList VTableFuncs; + for (unsigned I = VTableListStartIndex, E = Record.size(); I != E; ++I) { + ValueInfo Callee = getValueInfoFromValueId(Record[I]).first; + uint64_t Offset = Record[++I]; + VTableFuncs.push_back({Callee, Offset}); + } + auto VS = + std::make_unique<GlobalVarSummary>(Flags, GVF, std::move(Refs)); + VS->setModulePath(getThisModule()->first()); + VS->setVTableFuncs(VTableFuncs); + auto GUID = getValueInfoFromValueId(ValueID); + VS->setOriginalName(GUID.second); + TheIndex.addGlobalValueSummary(GUID.first, std::move(VS)); + break; + } + // FS_COMBINED: [valueid, modid, flags, instcount, fflags, numrefs, + // numrefs x valueid, n x (valueid)] + // FS_COMBINED_PROFILE: [valueid, modid, flags, instcount, fflags, numrefs, + // numrefs x valueid, n x (valueid, hotness)] + case bitc::FS_COMBINED: + case bitc::FS_COMBINED_PROFILE: { + unsigned ValueID = Record[0]; + uint64_t ModuleId = Record[1]; + uint64_t RawFlags = Record[2]; + unsigned InstCount = Record[3]; + uint64_t RawFunFlags = 0; + uint64_t EntryCount = 0; + unsigned NumRefs = Record[4]; + unsigned NumRORefs = 0, NumWORefs = 0; + int RefListStartIndex = 5; + + if (Version >= 4) { + RawFunFlags = Record[4]; + RefListStartIndex = 6; + size_t NumRefsIndex = 5; + if (Version >= 5) { + unsigned NumRORefsOffset = 1; + RefListStartIndex = 7; + if (Version >= 6) { + NumRefsIndex = 6; + EntryCount = Record[5]; + RefListStartIndex = 8; + if (Version >= 7) { + RefListStartIndex = 9; + NumWORefs = Record[8]; + NumRORefsOffset = 2; + } + } + NumRORefs = Record[RefListStartIndex - NumRORefsOffset]; + } + NumRefs = Record[NumRefsIndex]; + } + + auto Flags = getDecodedGVSummaryFlags(RawFlags, Version); + int CallGraphEdgeStartIndex = RefListStartIndex + NumRefs; + assert(Record.size() >= RefListStartIndex + NumRefs && + "Record size inconsistent with number of references"); + std::vector<ValueInfo> Refs = makeRefList( + ArrayRef<uint64_t>(Record).slice(RefListStartIndex, NumRefs)); + bool HasProfile = (BitCode == bitc::FS_COMBINED_PROFILE); + std::vector<FunctionSummary::EdgeTy> Edges = makeCallList( + ArrayRef<uint64_t>(Record).slice(CallGraphEdgeStartIndex), + IsOldProfileFormat, HasProfile, false); + ValueInfo VI = getValueInfoFromValueId(ValueID).first; + setSpecialRefs(Refs, NumRORefs, NumWORefs); + auto FS = std::make_unique<FunctionSummary>( + Flags, InstCount, getDecodedFFlags(RawFunFlags), EntryCount, + std::move(Refs), std::move(Edges), std::move(PendingTypeTests), + std::move(PendingTypeTestAssumeVCalls), + std::move(PendingTypeCheckedLoadVCalls), + std::move(PendingTypeTestAssumeConstVCalls), + std::move(PendingTypeCheckedLoadConstVCalls), + std::move(PendingParamAccesses)); + LastSeenSummary = FS.get(); + LastSeenGUID = VI.getGUID(); + FS->setModulePath(ModuleIdMap[ModuleId]); + TheIndex.addGlobalValueSummary(VI, std::move(FS)); + break; + } + // FS_COMBINED_ALIAS: [valueid, modid, flags, valueid] + // Aliases must be emitted (and parsed) after all FS_COMBINED entries, as + // they expect all aliasee summaries to be available. + case bitc::FS_COMBINED_ALIAS: { + unsigned ValueID = Record[0]; + uint64_t ModuleId = Record[1]; + uint64_t RawFlags = Record[2]; + unsigned AliaseeValueId = Record[3]; + auto Flags = getDecodedGVSummaryFlags(RawFlags, Version); + auto AS = std::make_unique<AliasSummary>(Flags); + LastSeenSummary = AS.get(); + AS->setModulePath(ModuleIdMap[ModuleId]); + + auto AliaseeVI = getValueInfoFromValueId(AliaseeValueId).first; + auto AliaseeInModule = TheIndex.findSummaryInModule(AliaseeVI, AS->modulePath()); + AS->setAliasee(AliaseeVI, AliaseeInModule); + + ValueInfo VI = getValueInfoFromValueId(ValueID).first; + LastSeenGUID = VI.getGUID(); + TheIndex.addGlobalValueSummary(VI, std::move(AS)); + break; + } + // FS_COMBINED_GLOBALVAR_INIT_REFS: [valueid, modid, flags, n x valueid] + case bitc::FS_COMBINED_GLOBALVAR_INIT_REFS: { + unsigned ValueID = Record[0]; + uint64_t ModuleId = Record[1]; + uint64_t RawFlags = Record[2]; + unsigned RefArrayStart = 3; + GlobalVarSummary::GVarFlags GVF(/* ReadOnly */ false, + /* WriteOnly */ false, + /* Constant */ false, + GlobalObject::VCallVisibilityPublic); + auto Flags = getDecodedGVSummaryFlags(RawFlags, Version); + if (Version >= 5) { + GVF = getDecodedGVarFlags(Record[3]); + RefArrayStart = 4; + } + std::vector<ValueInfo> Refs = + makeRefList(ArrayRef<uint64_t>(Record).slice(RefArrayStart)); + auto FS = + std::make_unique<GlobalVarSummary>(Flags, GVF, std::move(Refs)); + LastSeenSummary = FS.get(); + FS->setModulePath(ModuleIdMap[ModuleId]); + ValueInfo VI = getValueInfoFromValueId(ValueID).first; + LastSeenGUID = VI.getGUID(); + TheIndex.addGlobalValueSummary(VI, std::move(FS)); + break; + } + // FS_COMBINED_ORIGINAL_NAME: [original_name] + case bitc::FS_COMBINED_ORIGINAL_NAME: { + uint64_t OriginalName = Record[0]; + if (!LastSeenSummary) + return error("Name attachment that does not follow a combined record"); + LastSeenSummary->setOriginalName(OriginalName); + TheIndex.addOriginalName(LastSeenGUID, OriginalName); + // Reset the LastSeenSummary + LastSeenSummary = nullptr; + LastSeenGUID = 0; + break; + } + case bitc::FS_TYPE_TESTS: + assert(PendingTypeTests.empty()); + llvm::append_range(PendingTypeTests, Record); + break; + + case bitc::FS_TYPE_TEST_ASSUME_VCALLS: + assert(PendingTypeTestAssumeVCalls.empty()); + for (unsigned I = 0; I != Record.size(); I += 2) + PendingTypeTestAssumeVCalls.push_back({Record[I], Record[I+1]}); + break; + + case bitc::FS_TYPE_CHECKED_LOAD_VCALLS: + assert(PendingTypeCheckedLoadVCalls.empty()); + for (unsigned I = 0; I != Record.size(); I += 2) + PendingTypeCheckedLoadVCalls.push_back({Record[I], Record[I+1]}); + break; + + case bitc::FS_TYPE_TEST_ASSUME_CONST_VCALL: + PendingTypeTestAssumeConstVCalls.push_back( + {{Record[0], Record[1]}, {Record.begin() + 2, Record.end()}}); + break; + + case bitc::FS_TYPE_CHECKED_LOAD_CONST_VCALL: + PendingTypeCheckedLoadConstVCalls.push_back( + {{Record[0], Record[1]}, {Record.begin() + 2, Record.end()}}); + break; + + case bitc::FS_CFI_FUNCTION_DEFS: { + std::set<std::string> &CfiFunctionDefs = TheIndex.cfiFunctionDefs(); + for (unsigned I = 0; I != Record.size(); I += 2) + CfiFunctionDefs.insert( + {Strtab.data() + Record[I], static_cast<size_t>(Record[I + 1])}); + break; + } + + case bitc::FS_CFI_FUNCTION_DECLS: { + std::set<std::string> &CfiFunctionDecls = TheIndex.cfiFunctionDecls(); + for (unsigned I = 0; I != Record.size(); I += 2) + CfiFunctionDecls.insert( + {Strtab.data() + Record[I], static_cast<size_t>(Record[I + 1])}); + break; + } + + case bitc::FS_TYPE_ID: + parseTypeIdSummaryRecord(Record, Strtab, TheIndex); + break; + + case bitc::FS_TYPE_ID_METADATA: + parseTypeIdCompatibleVtableSummaryRecord(Record); + break; + + case bitc::FS_BLOCK_COUNT: + TheIndex.addBlockCount(Record[0]); + break; + + case bitc::FS_PARAM_ACCESS: { + PendingParamAccesses = parseParamAccesses(Record); + break; + } + } + } + llvm_unreachable("Exit infinite loop"); +} + +// Parse the module string table block into the Index. +// This populates the ModulePathStringTable map in the index. +Error ModuleSummaryIndexBitcodeReader::parseModuleStringTable() { + if (Error Err = Stream.EnterSubBlock(bitc::MODULE_STRTAB_BLOCK_ID)) + return Err; + + SmallVector<uint64_t, 64> Record; + + SmallString<128> ModulePath; + ModuleSummaryIndex::ModuleInfo *LastSeenModule = nullptr; + + while (true) { + Expected<BitstreamEntry> MaybeEntry = Stream.advanceSkippingSubblocks(); + if (!MaybeEntry) + return MaybeEntry.takeError(); + BitstreamEntry Entry = MaybeEntry.get(); + + switch (Entry.Kind) { + case BitstreamEntry::SubBlock: // Handled for us already. + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + return Error::success(); + case BitstreamEntry::Record: + // The interesting case. + break; + } + + Record.clear(); + Expected<unsigned> MaybeRecord = Stream.readRecord(Entry.ID, Record); + if (!MaybeRecord) + return MaybeRecord.takeError(); + switch (MaybeRecord.get()) { + default: // Default behavior: ignore. + break; + case bitc::MST_CODE_ENTRY: { + // MST_ENTRY: [modid, namechar x N] + uint64_t ModuleId = Record[0]; + + if (convertToString(Record, 1, ModulePath)) + return error("Invalid record"); + + LastSeenModule = TheIndex.addModule(ModulePath, ModuleId); + ModuleIdMap[ModuleId] = LastSeenModule->first(); + + ModulePath.clear(); + break; + } + /// MST_CODE_HASH: [5*i32] + case bitc::MST_CODE_HASH: { + if (Record.size() != 5) + return error("Invalid hash length " + Twine(Record.size()).str()); + if (!LastSeenModule) + return error("Invalid hash that does not follow a module path"); + int Pos = 0; + for (auto &Val : Record) { + assert(!(Val >> 32) && "Unexpected high bits set"); + LastSeenModule->second.second[Pos++] = Val; + } + // Reset LastSeenModule to avoid overriding the hash unexpectedly. + LastSeenModule = nullptr; + break; + } + } + } + llvm_unreachable("Exit infinite loop"); +} + +namespace { + +// FIXME: This class is only here to support the transition to llvm::Error. It +// will be removed once this transition is complete. Clients should prefer to +// deal with the Error value directly, rather than converting to error_code. +class BitcodeErrorCategoryType : public std::error_category { + const char *name() const noexcept override { + return "llvm.bitcode"; + } + + std::string message(int IE) const override { + BitcodeError E = static_cast<BitcodeError>(IE); + switch (E) { + case BitcodeError::CorruptedBitcode: + return "Corrupted bitcode"; + } + llvm_unreachable("Unknown error type!"); + } +}; + +} // end anonymous namespace + +static ManagedStatic<BitcodeErrorCategoryType> ErrorCategory; + +const std::error_category &llvm::BitcodeErrorCategory() { + return *ErrorCategory; +} + +static Expected<StringRef> readBlobInRecord(BitstreamCursor &Stream, + unsigned Block, unsigned RecordID) { + if (Error Err = Stream.EnterSubBlock(Block)) + return std::move(Err); + + StringRef Strtab; + while (true) { + Expected<llvm::BitstreamEntry> MaybeEntry = Stream.advance(); + if (!MaybeEntry) + return MaybeEntry.takeError(); + llvm::BitstreamEntry Entry = MaybeEntry.get(); + + switch (Entry.Kind) { + case BitstreamEntry::EndBlock: + return Strtab; + + case BitstreamEntry::Error: + return error("Malformed block"); + + case BitstreamEntry::SubBlock: + if (Error Err = Stream.SkipBlock()) + return std::move(Err); + break; + + case BitstreamEntry::Record: + StringRef Blob; + SmallVector<uint64_t, 1> Record; + Expected<unsigned> MaybeRecord = + Stream.readRecord(Entry.ID, Record, &Blob); + if (!MaybeRecord) + return MaybeRecord.takeError(); + if (MaybeRecord.get() == RecordID) + Strtab = Blob; + break; + } + } +} + +//===----------------------------------------------------------------------===// +// External interface +//===----------------------------------------------------------------------===// + +Expected<std::vector<BitcodeModule>> +llvm::getBitcodeModuleList(MemoryBufferRef Buffer) { + auto FOrErr = getBitcodeFileContents(Buffer); + if (!FOrErr) + return FOrErr.takeError(); + return std::move(FOrErr->Mods); +} + +Expected<BitcodeFileContents> +llvm::getBitcodeFileContents(MemoryBufferRef Buffer) { + Expected<BitstreamCursor> StreamOrErr = initStream(Buffer); + if (!StreamOrErr) + return StreamOrErr.takeError(); + BitstreamCursor &Stream = *StreamOrErr; + + BitcodeFileContents F; + while (true) { + uint64_t BCBegin = Stream.getCurrentByteNo(); + + // We may be consuming bitcode from a client that leaves garbage at the end + // of the bitcode stream (e.g. Apple's ar tool). If we are close enough to + // the end that there cannot possibly be another module, stop looking. + if (BCBegin + 8 >= Stream.getBitcodeBytes().size()) + return F; + + Expected<llvm::BitstreamEntry> MaybeEntry = Stream.advance(); + if (!MaybeEntry) + return MaybeEntry.takeError(); + llvm::BitstreamEntry Entry = MaybeEntry.get(); + + switch (Entry.Kind) { + case BitstreamEntry::EndBlock: + case BitstreamEntry::Error: + return error("Malformed block"); + + case BitstreamEntry::SubBlock: { + uint64_t IdentificationBit = -1ull; + if (Entry.ID == bitc::IDENTIFICATION_BLOCK_ID) { + IdentificationBit = Stream.GetCurrentBitNo() - BCBegin * 8; + if (Error Err = Stream.SkipBlock()) + return std::move(Err); + + { + Expected<llvm::BitstreamEntry> MaybeEntry = Stream.advance(); + if (!MaybeEntry) + return MaybeEntry.takeError(); + Entry = MaybeEntry.get(); + } + + if (Entry.Kind != BitstreamEntry::SubBlock || + Entry.ID != bitc::MODULE_BLOCK_ID) + return error("Malformed block"); + } + + if (Entry.ID == bitc::MODULE_BLOCK_ID) { + uint64_t ModuleBit = Stream.GetCurrentBitNo() - BCBegin * 8; + if (Error Err = Stream.SkipBlock()) + return std::move(Err); + + F.Mods.push_back({Stream.getBitcodeBytes().slice( + BCBegin, Stream.getCurrentByteNo() - BCBegin), + Buffer.getBufferIdentifier(), IdentificationBit, + ModuleBit}); + continue; + } + + if (Entry.ID == bitc::STRTAB_BLOCK_ID) { + Expected<StringRef> Strtab = + readBlobInRecord(Stream, bitc::STRTAB_BLOCK_ID, bitc::STRTAB_BLOB); + if (!Strtab) + return Strtab.takeError(); + // This string table is used by every preceding bitcode module that does + // not have its own string table. A bitcode file may have multiple + // string tables if it was created by binary concatenation, for example + // with "llvm-cat -b". + for (BitcodeModule &I : llvm::reverse(F.Mods)) { + if (!I.Strtab.empty()) + break; + I.Strtab = *Strtab; + } + // Similarly, the string table is used by every preceding symbol table; + // normally there will be just one unless the bitcode file was created + // by binary concatenation. + if (!F.Symtab.empty() && F.StrtabForSymtab.empty()) + F.StrtabForSymtab = *Strtab; + continue; + } + + if (Entry.ID == bitc::SYMTAB_BLOCK_ID) { + Expected<StringRef> SymtabOrErr = + readBlobInRecord(Stream, bitc::SYMTAB_BLOCK_ID, bitc::SYMTAB_BLOB); + if (!SymtabOrErr) + return SymtabOrErr.takeError(); + + // We can expect the bitcode file to have multiple symbol tables if it + // was created by binary concatenation. In that case we silently + // ignore any subsequent symbol tables, which is fine because this is a + // low level function. The client is expected to notice that the number + // of modules in the symbol table does not match the number of modules + // in the input file and regenerate the symbol table. + if (F.Symtab.empty()) + F.Symtab = *SymtabOrErr; + continue; + } + + if (Error Err = Stream.SkipBlock()) + return std::move(Err); + continue; + } + case BitstreamEntry::Record: + if (Error E = Stream.skipRecord(Entry.ID).takeError()) + return std::move(E); + continue; + } + } +} + +/// Get a lazy one-at-time loading module from bitcode. +/// +/// This isn't always used in a lazy context. In particular, it's also used by +/// \a parseModule(). If this is truly lazy, then we need to eagerly pull +/// in forward-referenced functions from block address references. +/// +/// \param[in] MaterializeAll Set to \c true if we should materialize +/// everything. +Expected<std::unique_ptr<Module>> +BitcodeModule::getModuleImpl(LLVMContext &Context, bool MaterializeAll, + bool ShouldLazyLoadMetadata, bool IsImporting, + DataLayoutCallbackTy DataLayoutCallback) { + BitstreamCursor Stream(Buffer); + + std::string ProducerIdentification; + if (IdentificationBit != -1ull) { + if (Error JumpFailed = Stream.JumpToBit(IdentificationBit)) + return std::move(JumpFailed); + if (Error E = + readIdentificationBlock(Stream).moveInto(ProducerIdentification)) + return std::move(E); + } + + if (Error JumpFailed = Stream.JumpToBit(ModuleBit)) + return std::move(JumpFailed); + auto *R = new BitcodeReader(std::move(Stream), Strtab, ProducerIdentification, + Context); + + std::unique_ptr<Module> M = + std::make_unique<Module>(ModuleIdentifier, Context); + M->setMaterializer(R); + + // Delay parsing Metadata if ShouldLazyLoadMetadata is true. + if (Error Err = R->parseBitcodeInto(M.get(), ShouldLazyLoadMetadata, + IsImporting, DataLayoutCallback)) + return std::move(Err); + + if (MaterializeAll) { + // Read in the entire module, and destroy the BitcodeReader. + if (Error Err = M->materializeAll()) + return std::move(Err); + } else { + // Resolve forward references from blockaddresses. + if (Error Err = R->materializeForwardReferencedFunctions()) + return std::move(Err); + } + return std::move(M); +} + +Expected<std::unique_ptr<Module>> +BitcodeModule::getLazyModule(LLVMContext &Context, bool ShouldLazyLoadMetadata, + bool IsImporting) { + return getModuleImpl(Context, false, ShouldLazyLoadMetadata, IsImporting, + [](StringRef) { return None; }); +} + +// Parse the specified bitcode buffer and merge the index into CombinedIndex. +// We don't use ModuleIdentifier here because the client may need to control the +// module path used in the combined summary (e.g. when reading summaries for +// regular LTO modules). +Error BitcodeModule::readSummary(ModuleSummaryIndex &CombinedIndex, + StringRef ModulePath, uint64_t ModuleId) { + BitstreamCursor Stream(Buffer); + if (Error JumpFailed = Stream.JumpToBit(ModuleBit)) + return JumpFailed; + + ModuleSummaryIndexBitcodeReader R(std::move(Stream), Strtab, CombinedIndex, + ModulePath, ModuleId); + return R.parseModule(); +} + +// Parse the specified bitcode buffer, returning the function info index. +Expected<std::unique_ptr<ModuleSummaryIndex>> BitcodeModule::getSummary() { + BitstreamCursor Stream(Buffer); + if (Error JumpFailed = Stream.JumpToBit(ModuleBit)) + return std::move(JumpFailed); + + auto Index = std::make_unique<ModuleSummaryIndex>(/*HaveGVs=*/false); + ModuleSummaryIndexBitcodeReader R(std::move(Stream), Strtab, *Index, + ModuleIdentifier, 0); + + if (Error Err = R.parseModule()) + return std::move(Err); + + return std::move(Index); +} + +static Expected<bool> getEnableSplitLTOUnitFlag(BitstreamCursor &Stream, + unsigned ID) { + if (Error Err = Stream.EnterSubBlock(ID)) + return std::move(Err); + SmallVector<uint64_t, 64> Record; + + while (true) { + BitstreamEntry Entry; + if (Error E = Stream.advanceSkippingSubblocks().moveInto(Entry)) + return std::move(E); + + switch (Entry.Kind) { + case BitstreamEntry::SubBlock: // Handled for us already. + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + // If no flags record found, conservatively return true to mimic + // behavior before this flag was added. + return true; + case BitstreamEntry::Record: + // The interesting case. + break; + } + + // Look for the FS_FLAGS record. + Record.clear(); + Expected<unsigned> MaybeBitCode = Stream.readRecord(Entry.ID, Record); + if (!MaybeBitCode) + return MaybeBitCode.takeError(); + switch (MaybeBitCode.get()) { + default: // Default behavior: ignore. + break; + case bitc::FS_FLAGS: { // [flags] + uint64_t Flags = Record[0]; + // Scan flags. + assert(Flags <= 0x7f && "Unexpected bits in flag"); + + return Flags & 0x8; + } + } + } + llvm_unreachable("Exit infinite loop"); +} + +// Check if the given bitcode buffer contains a global value summary block. +Expected<BitcodeLTOInfo> BitcodeModule::getLTOInfo() { + BitstreamCursor Stream(Buffer); + if (Error JumpFailed = Stream.JumpToBit(ModuleBit)) + return std::move(JumpFailed); + + if (Error Err = Stream.EnterSubBlock(bitc::MODULE_BLOCK_ID)) + return std::move(Err); + + while (true) { + llvm::BitstreamEntry Entry; + if (Error E = Stream.advance().moveInto(Entry)) + return std::move(E); + + switch (Entry.Kind) { + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + return BitcodeLTOInfo{/*IsThinLTO=*/false, /*HasSummary=*/false, + /*EnableSplitLTOUnit=*/false}; + + case BitstreamEntry::SubBlock: + if (Entry.ID == bitc::GLOBALVAL_SUMMARY_BLOCK_ID) { + Expected<bool> EnableSplitLTOUnit = + getEnableSplitLTOUnitFlag(Stream, Entry.ID); + if (!EnableSplitLTOUnit) + return EnableSplitLTOUnit.takeError(); + return BitcodeLTOInfo{/*IsThinLTO=*/true, /*HasSummary=*/true, + *EnableSplitLTOUnit}; + } + + if (Entry.ID == bitc::FULL_LTO_GLOBALVAL_SUMMARY_BLOCK_ID) { + Expected<bool> EnableSplitLTOUnit = + getEnableSplitLTOUnitFlag(Stream, Entry.ID); + if (!EnableSplitLTOUnit) + return EnableSplitLTOUnit.takeError(); + return BitcodeLTOInfo{/*IsThinLTO=*/false, /*HasSummary=*/true, + *EnableSplitLTOUnit}; + } + + // Ignore other sub-blocks. + if (Error Err = Stream.SkipBlock()) + return std::move(Err); + continue; + + case BitstreamEntry::Record: + if (Expected<unsigned> StreamFailed = Stream.skipRecord(Entry.ID)) + continue; + else + return StreamFailed.takeError(); + } + } +} + +static Expected<BitcodeModule> getSingleModule(MemoryBufferRef Buffer) { + Expected<std::vector<BitcodeModule>> MsOrErr = getBitcodeModuleList(Buffer); + if (!MsOrErr) + return MsOrErr.takeError(); + + if (MsOrErr->size() != 1) + return error("Expected a single module"); + + return (*MsOrErr)[0]; +} + +Expected<std::unique_ptr<Module>> +llvm::getLazyBitcodeModule(MemoryBufferRef Buffer, LLVMContext &Context, + bool ShouldLazyLoadMetadata, bool IsImporting) { + Expected<BitcodeModule> BM = getSingleModule(Buffer); + if (!BM) + return BM.takeError(); + + return BM->getLazyModule(Context, ShouldLazyLoadMetadata, IsImporting); +} + +Expected<std::unique_ptr<Module>> llvm::getOwningLazyBitcodeModule( + std::unique_ptr<MemoryBuffer> &&Buffer, LLVMContext &Context, + bool ShouldLazyLoadMetadata, bool IsImporting) { + auto MOrErr = getLazyBitcodeModule(*Buffer, Context, ShouldLazyLoadMetadata, + IsImporting); + if (MOrErr) + (*MOrErr)->setOwnedMemoryBuffer(std::move(Buffer)); + return MOrErr; +} + +Expected<std::unique_ptr<Module>> +BitcodeModule::parseModule(LLVMContext &Context, + DataLayoutCallbackTy DataLayoutCallback) { + return getModuleImpl(Context, true, false, false, DataLayoutCallback); + // TODO: Restore the use-lists to the in-memory state when the bitcode was + // written. We must defer until the Module has been fully materialized. +} + +Expected<std::unique_ptr<Module>> +llvm::parseBitcodeFile(MemoryBufferRef Buffer, LLVMContext &Context, + DataLayoutCallbackTy DataLayoutCallback) { + Expected<BitcodeModule> BM = getSingleModule(Buffer); + if (!BM) + return BM.takeError(); + + return BM->parseModule(Context, DataLayoutCallback); +} + +Expected<std::string> llvm::getBitcodeTargetTriple(MemoryBufferRef Buffer) { + Expected<BitstreamCursor> StreamOrErr = initStream(Buffer); + if (!StreamOrErr) + return StreamOrErr.takeError(); + + return readTriple(*StreamOrErr); +} + +Expected<bool> llvm::isBitcodeContainingObjCCategory(MemoryBufferRef Buffer) { + Expected<BitstreamCursor> StreamOrErr = initStream(Buffer); + if (!StreamOrErr) + return StreamOrErr.takeError(); + + return hasObjCCategory(*StreamOrErr); +} + +Expected<std::string> llvm::getBitcodeProducerString(MemoryBufferRef Buffer) { + Expected<BitstreamCursor> StreamOrErr = initStream(Buffer); + if (!StreamOrErr) + return StreamOrErr.takeError(); + + return readIdentificationCode(*StreamOrErr); +} + +Error llvm::readModuleSummaryIndex(MemoryBufferRef Buffer, + ModuleSummaryIndex &CombinedIndex, + uint64_t ModuleId) { + Expected<BitcodeModule> BM = getSingleModule(Buffer); + if (!BM) + return BM.takeError(); + + return BM->readSummary(CombinedIndex, BM->getModuleIdentifier(), ModuleId); +} + +Expected<std::unique_ptr<ModuleSummaryIndex>> +llvm::getModuleSummaryIndex(MemoryBufferRef Buffer) { + Expected<BitcodeModule> BM = getSingleModule(Buffer); + if (!BM) + return BM.takeError(); + + return BM->getSummary(); +} + +Expected<BitcodeLTOInfo> llvm::getBitcodeLTOInfo(MemoryBufferRef Buffer) { + Expected<BitcodeModule> BM = getSingleModule(Buffer); + if (!BM) + return BM.takeError(); + + return BM->getLTOInfo(); +} + +Expected<std::unique_ptr<ModuleSummaryIndex>> +llvm::getModuleSummaryIndexForFile(StringRef Path, + bool IgnoreEmptyThinLTOIndexFile) { + ErrorOr<std::unique_ptr<MemoryBuffer>> FileOrErr = + MemoryBuffer::getFileOrSTDIN(Path); + if (!FileOrErr) + return errorCodeToError(FileOrErr.getError()); + if (IgnoreEmptyThinLTOIndexFile && !(*FileOrErr)->getBufferSize()) + return nullptr; + return getModuleSummaryIndex(**FileOrErr); +} diff --git a/contrib/libs/llvm14/lib/Bitcode/Reader/MetadataLoader.cpp b/contrib/libs/llvm14/lib/Bitcode/Reader/MetadataLoader.cpp new file mode 100644 index 0000000000..0f41115140 --- /dev/null +++ b/contrib/libs/llvm14/lib/Bitcode/Reader/MetadataLoader.cpp @@ -0,0 +1,2379 @@ +//===- MetadataLoader.cpp - Internal BitcodeReader implementation ---------===// +// +// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. +// See https://llvm.org/LICENSE.txt for license information. +// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception +// +//===----------------------------------------------------------------------===// + +#include "MetadataLoader.h" +#include "ValueList.h" + +#include "llvm/ADT/APFloat.h" +#include "llvm/ADT/APInt.h" +#include "llvm/ADT/ArrayRef.h" +#include "llvm/ADT/DenseMap.h" +#include "llvm/ADT/DenseSet.h" +#include "llvm/ADT/None.h" +#include "llvm/ADT/STLExtras.h" +#include "llvm/ADT/SmallString.h" +#include "llvm/ADT/Statistic.h" +#include "llvm/ADT/StringRef.h" +#include "llvm/ADT/Twine.h" +#include "llvm/Bitcode/BitcodeReader.h" +#include "llvm/Bitcode/LLVMBitCodes.h" +#include "llvm/Bitstream/BitstreamReader.h" +#include "llvm/IR/Argument.h" +#include "llvm/IR/Attributes.h" +#include "llvm/IR/AutoUpgrade.h" +#include "llvm/IR/BasicBlock.h" +#include "llvm/IR/CallingConv.h" +#include "llvm/IR/Comdat.h" +#include "llvm/IR/Constant.h" +#include "llvm/IR/Constants.h" +#include "llvm/IR/DebugInfo.h" +#include "llvm/IR/DebugInfoMetadata.h" +#include "llvm/IR/DebugLoc.h" +#include "llvm/IR/DerivedTypes.h" +#include "llvm/IR/DiagnosticPrinter.h" +#include "llvm/IR/Function.h" +#include "llvm/IR/GVMaterializer.h" +#include "llvm/IR/GlobalAlias.h" +#include "llvm/IR/GlobalIFunc.h" +#include "llvm/IR/GlobalObject.h" +#include "llvm/IR/GlobalValue.h" +#include "llvm/IR/GlobalVariable.h" +#include "llvm/IR/InlineAsm.h" +#include "llvm/IR/InstrTypes.h" +#include "llvm/IR/Instruction.h" +#include "llvm/IR/Instructions.h" +#include "llvm/IR/IntrinsicInst.h" +#include "llvm/IR/Intrinsics.h" +#include "llvm/IR/LLVMContext.h" +#include "llvm/IR/Module.h" +#include "llvm/IR/ModuleSummaryIndex.h" +#include "llvm/IR/OperandTraits.h" +#include "llvm/IR/TrackingMDRef.h" +#include "llvm/IR/Type.h" +#include "llvm/IR/ValueHandle.h" +#include "llvm/Support/AtomicOrdering.h" +#include "llvm/Support/Casting.h" +#include "llvm/Support/CommandLine.h" +#include "llvm/Support/Compiler.h" +#include "llvm/Support/Debug.h" +#include "llvm/Support/ErrorHandling.h" +#include "llvm/Support/ManagedStatic.h" +#include "llvm/Support/MemoryBuffer.h" +#include "llvm/Support/raw_ostream.h" +#include <algorithm> +#include <cassert> +#include <cstddef> +#include <cstdint> +#include <deque> +#include <limits> +#include <map> +#include <string> +#include <system_error> +#include <tuple> +#include <utility> +#include <vector> + +using namespace llvm; + +#define DEBUG_TYPE "bitcode-reader" + +STATISTIC(NumMDStringLoaded, "Number of MDStrings loaded"); +STATISTIC(NumMDNodeTemporary, "Number of MDNode::Temporary created"); +STATISTIC(NumMDRecordLoaded, "Number of Metadata records loaded"); + +/// Flag whether we need to import full type definitions for ThinLTO. +/// Currently needed for Darwin and LLDB. +static cl::opt<bool> ImportFullTypeDefinitions( + "import-full-type-definitions", cl::init(false), cl::Hidden, + cl::desc("Import full type definitions for ThinLTO.")); + +static cl::opt<bool> DisableLazyLoading( + "disable-ondemand-mds-loading", cl::init(false), cl::Hidden, + cl::desc("Force disable the lazy-loading on-demand of metadata when " + "loading bitcode for importing.")); + +namespace { + +static int64_t unrotateSign(uint64_t U) { return (U & 1) ? ~(U >> 1) : U >> 1; } + +class BitcodeReaderMetadataList { + /// Array of metadata references. + /// + /// Don't use std::vector here. Some versions of libc++ copy (instead of + /// move) on resize, and TrackingMDRef is very expensive to copy. + SmallVector<TrackingMDRef, 1> MetadataPtrs; + + /// The set of indices in MetadataPtrs above of forward references that were + /// generated. + SmallDenseSet<unsigned, 1> ForwardReference; + + /// The set of indices in MetadataPtrs above of Metadata that need to be + /// resolved. + SmallDenseSet<unsigned, 1> UnresolvedNodes; + + /// Structures for resolving old type refs. + struct { + SmallDenseMap<MDString *, TempMDTuple, 1> Unknown; + SmallDenseMap<MDString *, DICompositeType *, 1> Final; + SmallDenseMap<MDString *, DICompositeType *, 1> FwdDecls; + SmallVector<std::pair<TrackingMDRef, TempMDTuple>, 1> Arrays; + } OldTypeRefs; + + LLVMContext &Context; + + /// Maximum number of valid references. Forward references exceeding the + /// maximum must be invalid. + unsigned RefsUpperBound; + +public: + BitcodeReaderMetadataList(LLVMContext &C, size_t RefsUpperBound) + : Context(C), + RefsUpperBound(std::min((size_t)std::numeric_limits<unsigned>::max(), + RefsUpperBound)) {} + + // vector compatibility methods + unsigned size() const { return MetadataPtrs.size(); } + void resize(unsigned N) { MetadataPtrs.resize(N); } + void push_back(Metadata *MD) { MetadataPtrs.emplace_back(MD); } + void clear() { MetadataPtrs.clear(); } + Metadata *back() const { return MetadataPtrs.back(); } + void pop_back() { MetadataPtrs.pop_back(); } + bool empty() const { return MetadataPtrs.empty(); } + + Metadata *operator[](unsigned i) const { + assert(i < MetadataPtrs.size()); + return MetadataPtrs[i]; + } + + Metadata *lookup(unsigned I) const { + if (I < MetadataPtrs.size()) + return MetadataPtrs[I]; + return nullptr; + } + + void shrinkTo(unsigned N) { + assert(N <= size() && "Invalid shrinkTo request!"); + assert(ForwardReference.empty() && "Unexpected forward refs"); + assert(UnresolvedNodes.empty() && "Unexpected unresolved node"); + MetadataPtrs.resize(N); + } + + /// Return the given metadata, creating a replaceable forward reference if + /// necessary. + Metadata *getMetadataFwdRef(unsigned Idx); + + /// Return the given metadata only if it is fully resolved. + /// + /// Gives the same result as \a lookup(), unless \a MDNode::isResolved() + /// would give \c false. + Metadata *getMetadataIfResolved(unsigned Idx); + + MDNode *getMDNodeFwdRefOrNull(unsigned Idx); + void assignValue(Metadata *MD, unsigned Idx); + void tryToResolveCycles(); + bool hasFwdRefs() const { return !ForwardReference.empty(); } + int getNextFwdRef() { + assert(hasFwdRefs()); + return *ForwardReference.begin(); + } + + /// Upgrade a type that had an MDString reference. + void addTypeRef(MDString &UUID, DICompositeType &CT); + + /// Upgrade a type that had an MDString reference. + Metadata *upgradeTypeRef(Metadata *MaybeUUID); + + /// Upgrade a type ref array that may have MDString references. + Metadata *upgradeTypeRefArray(Metadata *MaybeTuple); + +private: + Metadata *resolveTypeRefArray(Metadata *MaybeTuple); +}; + +void BitcodeReaderMetadataList::assignValue(Metadata *MD, unsigned Idx) { + if (auto *MDN = dyn_cast<MDNode>(MD)) + if (!MDN->isResolved()) + UnresolvedNodes.insert(Idx); + + if (Idx == size()) { + push_back(MD); + return; + } + + if (Idx >= size()) + resize(Idx + 1); + + TrackingMDRef &OldMD = MetadataPtrs[Idx]; + if (!OldMD) { + OldMD.reset(MD); + return; + } + + // If there was a forward reference to this value, replace it. + TempMDTuple PrevMD(cast<MDTuple>(OldMD.get())); + PrevMD->replaceAllUsesWith(MD); + ForwardReference.erase(Idx); +} + +Metadata *BitcodeReaderMetadataList::getMetadataFwdRef(unsigned Idx) { + // Bail out for a clearly invalid value. + if (Idx >= RefsUpperBound) + return nullptr; + + if (Idx >= size()) + resize(Idx + 1); + + if (Metadata *MD = MetadataPtrs[Idx]) + return MD; + + // Track forward refs to be resolved later. + ForwardReference.insert(Idx); + + // Create and return a placeholder, which will later be RAUW'd. + ++NumMDNodeTemporary; + Metadata *MD = MDNode::getTemporary(Context, None).release(); + MetadataPtrs[Idx].reset(MD); + return MD; +} + +Metadata *BitcodeReaderMetadataList::getMetadataIfResolved(unsigned Idx) { + Metadata *MD = lookup(Idx); + if (auto *N = dyn_cast_or_null<MDNode>(MD)) + if (!N->isResolved()) + return nullptr; + return MD; +} + +MDNode *BitcodeReaderMetadataList::getMDNodeFwdRefOrNull(unsigned Idx) { + return dyn_cast_or_null<MDNode>(getMetadataFwdRef(Idx)); +} + +void BitcodeReaderMetadataList::tryToResolveCycles() { + if (!ForwardReference.empty()) + // Still forward references... can't resolve cycles. + return; + + // Give up on finding a full definition for any forward decls that remain. + for (const auto &Ref : OldTypeRefs.FwdDecls) + OldTypeRefs.Final.insert(Ref); + OldTypeRefs.FwdDecls.clear(); + + // Upgrade from old type ref arrays. In strange cases, this could add to + // OldTypeRefs.Unknown. + for (const auto &Array : OldTypeRefs.Arrays) + Array.second->replaceAllUsesWith(resolveTypeRefArray(Array.first.get())); + OldTypeRefs.Arrays.clear(); + + // Replace old string-based type refs with the resolved node, if possible. + // If we haven't seen the node, leave it to the verifier to complain about + // the invalid string reference. + for (const auto &Ref : OldTypeRefs.Unknown) { + if (DICompositeType *CT = OldTypeRefs.Final.lookup(Ref.first)) + Ref.second->replaceAllUsesWith(CT); + else + Ref.second->replaceAllUsesWith(Ref.first); + } + OldTypeRefs.Unknown.clear(); + + if (UnresolvedNodes.empty()) + // Nothing to do. + return; + + // Resolve any cycles. + for (unsigned I : UnresolvedNodes) { + auto &MD = MetadataPtrs[I]; + auto *N = dyn_cast_or_null<MDNode>(MD); + if (!N) + continue; + + assert(!N->isTemporary() && "Unexpected forward reference"); + N->resolveCycles(); + } + + // Make sure we return early again until there's another unresolved ref. + UnresolvedNodes.clear(); +} + +void BitcodeReaderMetadataList::addTypeRef(MDString &UUID, + DICompositeType &CT) { + assert(CT.getRawIdentifier() == &UUID && "Mismatched UUID"); + if (CT.isForwardDecl()) + OldTypeRefs.FwdDecls.insert(std::make_pair(&UUID, &CT)); + else + OldTypeRefs.Final.insert(std::make_pair(&UUID, &CT)); +} + +Metadata *BitcodeReaderMetadataList::upgradeTypeRef(Metadata *MaybeUUID) { + auto *UUID = dyn_cast_or_null<MDString>(MaybeUUID); + if (LLVM_LIKELY(!UUID)) + return MaybeUUID; + + if (auto *CT = OldTypeRefs.Final.lookup(UUID)) + return CT; + + auto &Ref = OldTypeRefs.Unknown[UUID]; + if (!Ref) + Ref = MDNode::getTemporary(Context, None); + return Ref.get(); +} + +Metadata *BitcodeReaderMetadataList::upgradeTypeRefArray(Metadata *MaybeTuple) { + auto *Tuple = dyn_cast_or_null<MDTuple>(MaybeTuple); + if (!Tuple || Tuple->isDistinct()) + return MaybeTuple; + + // Look through the array immediately if possible. + if (!Tuple->isTemporary()) + return resolveTypeRefArray(Tuple); + + // Create and return a placeholder to use for now. Eventually + // resolveTypeRefArrays() will be resolve this forward reference. + OldTypeRefs.Arrays.emplace_back( + std::piecewise_construct, std::forward_as_tuple(Tuple), + std::forward_as_tuple(MDTuple::getTemporary(Context, None))); + return OldTypeRefs.Arrays.back().second.get(); +} + +Metadata *BitcodeReaderMetadataList::resolveTypeRefArray(Metadata *MaybeTuple) { + auto *Tuple = dyn_cast_or_null<MDTuple>(MaybeTuple); + if (!Tuple || Tuple->isDistinct()) + return MaybeTuple; + + // Look through the DITypeRefArray, upgrading each DIType *. + SmallVector<Metadata *, 32> Ops; + Ops.reserve(Tuple->getNumOperands()); + for (Metadata *MD : Tuple->operands()) + Ops.push_back(upgradeTypeRef(MD)); + + return MDTuple::get(Context, Ops); +} + +namespace { + +class PlaceholderQueue { + // Placeholders would thrash around when moved, so store in a std::deque + // instead of some sort of vector. + std::deque<DistinctMDOperandPlaceholder> PHs; + +public: + ~PlaceholderQueue() { + assert(empty() && + "PlaceholderQueue hasn't been flushed before being destroyed"); + } + bool empty() const { return PHs.empty(); } + DistinctMDOperandPlaceholder &getPlaceholderOp(unsigned ID); + void flush(BitcodeReaderMetadataList &MetadataList); + + /// Return the list of temporaries nodes in the queue, these need to be + /// loaded before we can flush the queue. + void getTemporaries(BitcodeReaderMetadataList &MetadataList, + DenseSet<unsigned> &Temporaries) { + for (auto &PH : PHs) { + auto ID = PH.getID(); + auto *MD = MetadataList.lookup(ID); + if (!MD) { + Temporaries.insert(ID); + continue; + } + auto *N = dyn_cast_or_null<MDNode>(MD); + if (N && N->isTemporary()) + Temporaries.insert(ID); + } + } +}; + +} // end anonymous namespace + +DistinctMDOperandPlaceholder &PlaceholderQueue::getPlaceholderOp(unsigned ID) { + PHs.emplace_back(ID); + return PHs.back(); +} + +void PlaceholderQueue::flush(BitcodeReaderMetadataList &MetadataList) { + while (!PHs.empty()) { + auto *MD = MetadataList.lookup(PHs.front().getID()); + assert(MD && "Flushing placeholder on unassigned MD"); +#ifndef NDEBUG + if (auto *MDN = dyn_cast<MDNode>(MD)) + assert(MDN->isResolved() && + "Flushing Placeholder while cycles aren't resolved"); +#endif + PHs.front().replaceUseWith(MD); + PHs.pop_front(); + } +} + +} // anonymous namespace + +static Error error(const Twine &Message) { + return make_error<StringError>( + Message, make_error_code(BitcodeError::CorruptedBitcode)); +} + +class MetadataLoader::MetadataLoaderImpl { + BitcodeReaderMetadataList MetadataList; + BitcodeReaderValueList &ValueList; + BitstreamCursor &Stream; + LLVMContext &Context; + Module &TheModule; + std::function<Type *(unsigned)> getTypeByID; + + /// Cursor associated with the lazy-loading of Metadata. This is the easy way + /// to keep around the right "context" (Abbrev list) to be able to jump in + /// the middle of the metadata block and load any record. + BitstreamCursor IndexCursor; + + /// Index that keeps track of MDString values. + std::vector<StringRef> MDStringRef; + + /// On-demand loading of a single MDString. Requires the index above to be + /// populated. + MDString *lazyLoadOneMDString(unsigned Idx); + + /// Index that keeps track of where to find a metadata record in the stream. + std::vector<uint64_t> GlobalMetadataBitPosIndex; + + /// Cursor position of the start of the global decl attachments, to enable + /// loading using the index built for lazy loading, instead of forward + /// references. + uint64_t GlobalDeclAttachmentPos = 0; + +#ifndef NDEBUG + /// Baisic correctness check that we end up parsing all of the global decl + /// attachments. + unsigned NumGlobalDeclAttachSkipped = 0; + unsigned NumGlobalDeclAttachParsed = 0; +#endif + + /// Load the global decl attachments, using the index built for lazy loading. + Expected<bool> loadGlobalDeclAttachments(); + + /// Populate the index above to enable lazily loading of metadata, and load + /// the named metadata as well as the transitively referenced global + /// Metadata. + Expected<bool> lazyLoadModuleMetadataBlock(); + + /// On-demand loading of a single metadata. Requires the index above to be + /// populated. + void lazyLoadOneMetadata(unsigned Idx, PlaceholderQueue &Placeholders); + + // Keep mapping of seens pair of old-style CU <-> SP, and update pointers to + // point from SP to CU after a block is completly parsed. + std::vector<std::pair<DICompileUnit *, Metadata *>> CUSubprograms; + + /// Functions that need to be matched with subprograms when upgrading old + /// metadata. + SmallDenseMap<Function *, DISubprogram *, 16> FunctionsWithSPs; + + // Map the bitcode's custom MDKind ID to the Module's MDKind ID. + DenseMap<unsigned, unsigned> MDKindMap; + + bool StripTBAA = false; + bool HasSeenOldLoopTags = false; + bool NeedUpgradeToDIGlobalVariableExpression = false; + bool NeedDeclareExpressionUpgrade = false; + + /// True if metadata is being parsed for a module being ThinLTO imported. + bool IsImporting = false; + + Error parseOneMetadata(SmallVectorImpl<uint64_t> &Record, unsigned Code, + PlaceholderQueue &Placeholders, StringRef Blob, + unsigned &NextMetadataNo); + Error parseMetadataStrings(ArrayRef<uint64_t> Record, StringRef Blob, + function_ref<void(StringRef)> CallBack); + Error parseGlobalObjectAttachment(GlobalObject &GO, + ArrayRef<uint64_t> Record); + Error parseMetadataKindRecord(SmallVectorImpl<uint64_t> &Record); + + void resolveForwardRefsAndPlaceholders(PlaceholderQueue &Placeholders); + + /// Upgrade old-style CU <-> SP pointers to point from SP to CU. + void upgradeCUSubprograms() { + for (auto CU_SP : CUSubprograms) + if (auto *SPs = dyn_cast_or_null<MDTuple>(CU_SP.second)) + for (auto &Op : SPs->operands()) + if (auto *SP = dyn_cast_or_null<DISubprogram>(Op)) + SP->replaceUnit(CU_SP.first); + CUSubprograms.clear(); + } + + /// Upgrade old-style bare DIGlobalVariables to DIGlobalVariableExpressions. + void upgradeCUVariables() { + if (!NeedUpgradeToDIGlobalVariableExpression) + return; + + // Upgrade list of variables attached to the CUs. + if (NamedMDNode *CUNodes = TheModule.getNamedMetadata("llvm.dbg.cu")) + for (unsigned I = 0, E = CUNodes->getNumOperands(); I != E; ++I) { + auto *CU = cast<DICompileUnit>(CUNodes->getOperand(I)); + if (auto *GVs = dyn_cast_or_null<MDTuple>(CU->getRawGlobalVariables())) + for (unsigned I = 0; I < GVs->getNumOperands(); I++) + if (auto *GV = + dyn_cast_or_null<DIGlobalVariable>(GVs->getOperand(I))) { + auto *DGVE = DIGlobalVariableExpression::getDistinct( + Context, GV, DIExpression::get(Context, {})); + GVs->replaceOperandWith(I, DGVE); + } + } + + // Upgrade variables attached to globals. + for (auto &GV : TheModule.globals()) { + SmallVector<MDNode *, 1> MDs; + GV.getMetadata(LLVMContext::MD_dbg, MDs); + GV.eraseMetadata(LLVMContext::MD_dbg); + for (auto *MD : MDs) + if (auto *DGV = dyn_cast<DIGlobalVariable>(MD)) { + auto *DGVE = DIGlobalVariableExpression::getDistinct( + Context, DGV, DIExpression::get(Context, {})); + GV.addMetadata(LLVMContext::MD_dbg, *DGVE); + } else + GV.addMetadata(LLVMContext::MD_dbg, *MD); + } + } + + /// Remove a leading DW_OP_deref from DIExpressions in a dbg.declare that + /// describes a function argument. + void upgradeDeclareExpressions(Function &F) { + if (!NeedDeclareExpressionUpgrade) + return; + + for (auto &BB : F) + for (auto &I : BB) + if (auto *DDI = dyn_cast<DbgDeclareInst>(&I)) + if (auto *DIExpr = DDI->getExpression()) + if (DIExpr->startsWithDeref() && + isa_and_nonnull<Argument>(DDI->getAddress())) { + SmallVector<uint64_t, 8> Ops; + Ops.append(std::next(DIExpr->elements_begin()), + DIExpr->elements_end()); + DDI->setExpression(DIExpression::get(Context, Ops)); + } + } + + /// Upgrade the expression from previous versions. + Error upgradeDIExpression(uint64_t FromVersion, + MutableArrayRef<uint64_t> &Expr, + SmallVectorImpl<uint64_t> &Buffer) { + auto N = Expr.size(); + switch (FromVersion) { + default: + return error("Invalid record"); + case 0: + if (N >= 3 && Expr[N - 3] == dwarf::DW_OP_bit_piece) + Expr[N - 3] = dwarf::DW_OP_LLVM_fragment; + LLVM_FALLTHROUGH; + case 1: + // Move DW_OP_deref to the end. + if (N && Expr[0] == dwarf::DW_OP_deref) { + auto End = Expr.end(); + if (Expr.size() >= 3 && + *std::prev(End, 3) == dwarf::DW_OP_LLVM_fragment) + End = std::prev(End, 3); + std::move(std::next(Expr.begin()), End, Expr.begin()); + *std::prev(End) = dwarf::DW_OP_deref; + } + NeedDeclareExpressionUpgrade = true; + LLVM_FALLTHROUGH; + case 2: { + // Change DW_OP_plus to DW_OP_plus_uconst. + // Change DW_OP_minus to DW_OP_uconst, DW_OP_minus + auto SubExpr = ArrayRef<uint64_t>(Expr); + while (!SubExpr.empty()) { + // Skip past other operators with their operands + // for this version of the IR, obtained from + // from historic DIExpression::ExprOperand::getSize(). + size_t HistoricSize; + switch (SubExpr.front()) { + default: + HistoricSize = 1; + break; + case dwarf::DW_OP_constu: + case dwarf::DW_OP_minus: + case dwarf::DW_OP_plus: + HistoricSize = 2; + break; + case dwarf::DW_OP_LLVM_fragment: + HistoricSize = 3; + break; + } + + // If the expression is malformed, make sure we don't + // copy more elements than we should. + HistoricSize = std::min(SubExpr.size(), HistoricSize); + ArrayRef<uint64_t> Args = SubExpr.slice(1, HistoricSize - 1); + + switch (SubExpr.front()) { + case dwarf::DW_OP_plus: + Buffer.push_back(dwarf::DW_OP_plus_uconst); + Buffer.append(Args.begin(), Args.end()); + break; + case dwarf::DW_OP_minus: + Buffer.push_back(dwarf::DW_OP_constu); + Buffer.append(Args.begin(), Args.end()); + Buffer.push_back(dwarf::DW_OP_minus); + break; + default: + Buffer.push_back(*SubExpr.begin()); + Buffer.append(Args.begin(), Args.end()); + break; + } + + // Continue with remaining elements. + SubExpr = SubExpr.slice(HistoricSize); + } + Expr = MutableArrayRef<uint64_t>(Buffer); + LLVM_FALLTHROUGH; + } + case 3: + // Up-to-date! + break; + } + + return Error::success(); + } + + void upgradeDebugInfo() { + upgradeCUSubprograms(); + upgradeCUVariables(); + } + +public: + MetadataLoaderImpl(BitstreamCursor &Stream, Module &TheModule, + BitcodeReaderValueList &ValueList, + std::function<Type *(unsigned)> getTypeByID, + bool IsImporting) + : MetadataList(TheModule.getContext(), Stream.SizeInBytes()), + ValueList(ValueList), Stream(Stream), Context(TheModule.getContext()), + TheModule(TheModule), getTypeByID(std::move(getTypeByID)), + IsImporting(IsImporting) {} + + Error parseMetadata(bool ModuleLevel); + + bool hasFwdRefs() const { return MetadataList.hasFwdRefs(); } + + Metadata *getMetadataFwdRefOrLoad(unsigned ID) { + if (ID < MDStringRef.size()) + return lazyLoadOneMDString(ID); + if (auto *MD = MetadataList.lookup(ID)) + return MD; + // If lazy-loading is enabled, we try recursively to load the operand + // instead of creating a temporary. + if (ID < (MDStringRef.size() + GlobalMetadataBitPosIndex.size())) { + PlaceholderQueue Placeholders; + lazyLoadOneMetadata(ID, Placeholders); + resolveForwardRefsAndPlaceholders(Placeholders); + return MetadataList.lookup(ID); + } + return MetadataList.getMetadataFwdRef(ID); + } + + DISubprogram *lookupSubprogramForFunction(Function *F) { + return FunctionsWithSPs.lookup(F); + } + + bool hasSeenOldLoopTags() const { return HasSeenOldLoopTags; } + + Error parseMetadataAttachment( + Function &F, const SmallVectorImpl<Instruction *> &InstructionList); + + Error parseMetadataKinds(); + + void setStripTBAA(bool Value) { StripTBAA = Value; } + bool isStrippingTBAA() const { return StripTBAA; } + + unsigned size() const { return MetadataList.size(); } + void shrinkTo(unsigned N) { MetadataList.shrinkTo(N); } + void upgradeDebugIntrinsics(Function &F) { upgradeDeclareExpressions(F); } +}; + +Expected<bool> +MetadataLoader::MetadataLoaderImpl::lazyLoadModuleMetadataBlock() { + IndexCursor = Stream; + SmallVector<uint64_t, 64> Record; + GlobalDeclAttachmentPos = 0; + // Get the abbrevs, and preload record positions to make them lazy-loadable. + while (true) { + uint64_t SavedPos = IndexCursor.GetCurrentBitNo(); + BitstreamEntry Entry; + if (Error E = + IndexCursor + .advanceSkippingSubblocks(BitstreamCursor::AF_DontPopBlockAtEnd) + .moveInto(Entry)) + return std::move(E); + + switch (Entry.Kind) { + case BitstreamEntry::SubBlock: // Handled for us already. + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: { + return true; + } + case BitstreamEntry::Record: { + // The interesting case. + ++NumMDRecordLoaded; + uint64_t CurrentPos = IndexCursor.GetCurrentBitNo(); + unsigned Code; + if (Error E = IndexCursor.skipRecord(Entry.ID).moveInto(Code)) + return std::move(E); + switch (Code) { + case bitc::METADATA_STRINGS: { + // Rewind and parse the strings. + if (Error Err = IndexCursor.JumpToBit(CurrentPos)) + return std::move(Err); + StringRef Blob; + Record.clear(); + if (Expected<unsigned> MaybeRecord = + IndexCursor.readRecord(Entry.ID, Record, &Blob)) + ; + else + return MaybeRecord.takeError(); + unsigned NumStrings = Record[0]; + MDStringRef.reserve(NumStrings); + auto IndexNextMDString = [&](StringRef Str) { + MDStringRef.push_back(Str); + }; + if (auto Err = parseMetadataStrings(Record, Blob, IndexNextMDString)) + return std::move(Err); + break; + } + case bitc::METADATA_INDEX_OFFSET: { + // This is the offset to the index, when we see this we skip all the + // records and load only an index to these. + if (Error Err = IndexCursor.JumpToBit(CurrentPos)) + return std::move(Err); + Record.clear(); + if (Expected<unsigned> MaybeRecord = + IndexCursor.readRecord(Entry.ID, Record)) + ; + else + return MaybeRecord.takeError(); + if (Record.size() != 2) + return error("Invalid record"); + auto Offset = Record[0] + (Record[1] << 32); + auto BeginPos = IndexCursor.GetCurrentBitNo(); + if (Error Err = IndexCursor.JumpToBit(BeginPos + Offset)) + return std::move(Err); + Expected<BitstreamEntry> MaybeEntry = + IndexCursor.advanceSkippingSubblocks( + BitstreamCursor::AF_DontPopBlockAtEnd); + if (!MaybeEntry) + return MaybeEntry.takeError(); + Entry = MaybeEntry.get(); + assert(Entry.Kind == BitstreamEntry::Record && + "Corrupted bitcode: Expected `Record` when trying to find the " + "Metadata index"); + Record.clear(); + if (Expected<unsigned> MaybeCode = + IndexCursor.readRecord(Entry.ID, Record)) + assert(MaybeCode.get() == bitc::METADATA_INDEX && + "Corrupted bitcode: Expected `METADATA_INDEX` when trying to " + "find the Metadata index"); + else + return MaybeCode.takeError(); + // Delta unpack + auto CurrentValue = BeginPos; + GlobalMetadataBitPosIndex.reserve(Record.size()); + for (auto &Elt : Record) { + CurrentValue += Elt; + GlobalMetadataBitPosIndex.push_back(CurrentValue); + } + break; + } + case bitc::METADATA_INDEX: + // We don't expect to get there, the Index is loaded when we encounter + // the offset. + return error("Corrupted Metadata block"); + case bitc::METADATA_NAME: { + // Named metadata need to be materialized now and aren't deferred. + if (Error Err = IndexCursor.JumpToBit(CurrentPos)) + return std::move(Err); + Record.clear(); + + unsigned Code; + if (Expected<unsigned> MaybeCode = + IndexCursor.readRecord(Entry.ID, Record)) { + Code = MaybeCode.get(); + assert(Code == bitc::METADATA_NAME); + } else + return MaybeCode.takeError(); + + // Read name of the named metadata. + SmallString<8> Name(Record.begin(), Record.end()); + if (Expected<unsigned> MaybeCode = IndexCursor.ReadCode()) + Code = MaybeCode.get(); + else + return MaybeCode.takeError(); + + // Named Metadata comes in two parts, we expect the name to be followed + // by the node + Record.clear(); + if (Expected<unsigned> MaybeNextBitCode = + IndexCursor.readRecord(Code, Record)) + assert(MaybeNextBitCode.get() == bitc::METADATA_NAMED_NODE); + else + return MaybeNextBitCode.takeError(); + + // Read named metadata elements. + unsigned Size = Record.size(); + NamedMDNode *NMD = TheModule.getOrInsertNamedMetadata(Name); + for (unsigned i = 0; i != Size; ++i) { + // FIXME: We could use a placeholder here, however NamedMDNode are + // taking MDNode as operand and not using the Metadata infrastructure. + // It is acknowledged by 'TODO: Inherit from Metadata' in the + // NamedMDNode class definition. + MDNode *MD = MetadataList.getMDNodeFwdRefOrNull(Record[i]); + assert(MD && "Invalid metadata: expect fwd ref to MDNode"); + NMD->addOperand(MD); + } + break; + } + case bitc::METADATA_GLOBAL_DECL_ATTACHMENT: { + if (!GlobalDeclAttachmentPos) + GlobalDeclAttachmentPos = SavedPos; +#ifndef NDEBUG + NumGlobalDeclAttachSkipped++; +#endif + break; + } + case bitc::METADATA_KIND: + case bitc::METADATA_STRING_OLD: + case bitc::METADATA_OLD_FN_NODE: + case bitc::METADATA_OLD_NODE: + case bitc::METADATA_VALUE: + case bitc::METADATA_DISTINCT_NODE: + case bitc::METADATA_NODE: + case bitc::METADATA_LOCATION: + case bitc::METADATA_GENERIC_DEBUG: + case bitc::METADATA_SUBRANGE: + case bitc::METADATA_ENUMERATOR: + case bitc::METADATA_BASIC_TYPE: + case bitc::METADATA_STRING_TYPE: + case bitc::METADATA_DERIVED_TYPE: + case bitc::METADATA_COMPOSITE_TYPE: + case bitc::METADATA_SUBROUTINE_TYPE: + case bitc::METADATA_MODULE: + case bitc::METADATA_FILE: + case bitc::METADATA_COMPILE_UNIT: + case bitc::METADATA_SUBPROGRAM: + case bitc::METADATA_LEXICAL_BLOCK: + case bitc::METADATA_LEXICAL_BLOCK_FILE: + case bitc::METADATA_NAMESPACE: + case bitc::METADATA_COMMON_BLOCK: + case bitc::METADATA_MACRO: + case bitc::METADATA_MACRO_FILE: + case bitc::METADATA_TEMPLATE_TYPE: + case bitc::METADATA_TEMPLATE_VALUE: + case bitc::METADATA_GLOBAL_VAR: + case bitc::METADATA_LOCAL_VAR: + case bitc::METADATA_LABEL: + case bitc::METADATA_EXPRESSION: + case bitc::METADATA_OBJC_PROPERTY: + case bitc::METADATA_IMPORTED_ENTITY: + case bitc::METADATA_GLOBAL_VAR_EXPR: + case bitc::METADATA_GENERIC_SUBRANGE: + // We don't expect to see any of these, if we see one, give up on + // lazy-loading and fallback. + MDStringRef.clear(); + GlobalMetadataBitPosIndex.clear(); + return false; + } + break; + } + } + } +} + +// Load the global decl attachments after building the lazy loading index. +// We don't load them "lazily" - all global decl attachments must be +// parsed since they aren't materialized on demand. However, by delaying +// their parsing until after the index is created, we can use the index +// instead of creating temporaries. +Expected<bool> MetadataLoader::MetadataLoaderImpl::loadGlobalDeclAttachments() { + // Nothing to do if we didn't find any of these metadata records. + if (!GlobalDeclAttachmentPos) + return true; + // Use a temporary cursor so that we don't mess up the main Stream cursor or + // the lazy loading IndexCursor (which holds the necessary abbrev ids). + BitstreamCursor TempCursor = Stream; + SmallVector<uint64_t, 64> Record; + // Jump to the position before the first global decl attachment, so we can + // scan for the first BitstreamEntry record. + if (Error Err = TempCursor.JumpToBit(GlobalDeclAttachmentPos)) + return std::move(Err); + while (true) { + BitstreamEntry Entry; + if (Error E = + TempCursor + .advanceSkippingSubblocks(BitstreamCursor::AF_DontPopBlockAtEnd) + .moveInto(Entry)) + return std::move(E); + + switch (Entry.Kind) { + case BitstreamEntry::SubBlock: // Handled for us already. + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + // Check that we parsed them all. + assert(NumGlobalDeclAttachSkipped == NumGlobalDeclAttachParsed); + return true; + case BitstreamEntry::Record: + break; + } + uint64_t CurrentPos = TempCursor.GetCurrentBitNo(); + Expected<unsigned> MaybeCode = TempCursor.skipRecord(Entry.ID); + if (!MaybeCode) + return MaybeCode.takeError(); + if (MaybeCode.get() != bitc::METADATA_GLOBAL_DECL_ATTACHMENT) { + // Anything other than a global decl attachment signals the end of + // these records. Check that we parsed them all. + assert(NumGlobalDeclAttachSkipped == NumGlobalDeclAttachParsed); + return true; + } +#ifndef NDEBUG + NumGlobalDeclAttachParsed++; +#endif + // FIXME: we need to do this early because we don't materialize global + // value explicitly. + if (Error Err = TempCursor.JumpToBit(CurrentPos)) + return std::move(Err); + Record.clear(); + if (Expected<unsigned> MaybeRecord = + TempCursor.readRecord(Entry.ID, Record)) + ; + else + return MaybeRecord.takeError(); + if (Record.size() % 2 == 0) + return error("Invalid record"); + unsigned ValueID = Record[0]; + if (ValueID >= ValueList.size()) + return error("Invalid record"); + if (auto *GO = dyn_cast<GlobalObject>(ValueList[ValueID])) { + // Need to save and restore the current position since + // parseGlobalObjectAttachment will resolve all forward references which + // would require parsing from locations stored in the index. + CurrentPos = TempCursor.GetCurrentBitNo(); + if (Error Err = parseGlobalObjectAttachment( + *GO, ArrayRef<uint64_t>(Record).slice(1))) + return std::move(Err); + if (Error Err = TempCursor.JumpToBit(CurrentPos)) + return std::move(Err); + } + } +} + +/// Parse a METADATA_BLOCK. If ModuleLevel is true then we are parsing +/// module level metadata. +Error MetadataLoader::MetadataLoaderImpl::parseMetadata(bool ModuleLevel) { + if (!ModuleLevel && MetadataList.hasFwdRefs()) + return error("Invalid metadata: fwd refs into function blocks"); + + // Record the entry position so that we can jump back here and efficiently + // skip the whole block in case we lazy-load. + auto EntryPos = Stream.GetCurrentBitNo(); + + if (Error Err = Stream.EnterSubBlock(bitc::METADATA_BLOCK_ID)) + return Err; + + SmallVector<uint64_t, 64> Record; + PlaceholderQueue Placeholders; + + // We lazy-load module-level metadata: we build an index for each record, and + // then load individual record as needed, starting with the named metadata. + if (ModuleLevel && IsImporting && MetadataList.empty() && + !DisableLazyLoading) { + auto SuccessOrErr = lazyLoadModuleMetadataBlock(); + if (!SuccessOrErr) + return SuccessOrErr.takeError(); + if (SuccessOrErr.get()) { + // An index was successfully created and we will be able to load metadata + // on-demand. + MetadataList.resize(MDStringRef.size() + + GlobalMetadataBitPosIndex.size()); + + // Now that we have built the index, load the global decl attachments + // that were deferred during that process. This avoids creating + // temporaries. + SuccessOrErr = loadGlobalDeclAttachments(); + if (!SuccessOrErr) + return SuccessOrErr.takeError(); + assert(SuccessOrErr.get()); + + // Reading the named metadata created forward references and/or + // placeholders, that we flush here. + resolveForwardRefsAndPlaceholders(Placeholders); + upgradeDebugInfo(); + // Return at the beginning of the block, since it is easy to skip it + // entirely from there. + Stream.ReadBlockEnd(); // Pop the abbrev block context. + if (Error Err = IndexCursor.JumpToBit(EntryPos)) + return Err; + if (Error Err = Stream.SkipBlock()) { + // FIXME this drops the error on the floor, which + // ThinLTO/X86/debuginfo-cu-import.ll relies on. + consumeError(std::move(Err)); + return Error::success(); + } + return Error::success(); + } + // Couldn't load an index, fallback to loading all the block "old-style". + } + + unsigned NextMetadataNo = MetadataList.size(); + + // Read all the records. + while (true) { + BitstreamEntry Entry; + if (Error E = Stream.advanceSkippingSubblocks().moveInto(Entry)) + return E; + + switch (Entry.Kind) { + case BitstreamEntry::SubBlock: // Handled for us already. + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + resolveForwardRefsAndPlaceholders(Placeholders); + upgradeDebugInfo(); + return Error::success(); + case BitstreamEntry::Record: + // The interesting case. + break; + } + + // Read a record. + Record.clear(); + StringRef Blob; + ++NumMDRecordLoaded; + if (Expected<unsigned> MaybeCode = + Stream.readRecord(Entry.ID, Record, &Blob)) { + if (Error Err = parseOneMetadata(Record, MaybeCode.get(), Placeholders, + Blob, NextMetadataNo)) + return Err; + } else + return MaybeCode.takeError(); + } +} + +MDString *MetadataLoader::MetadataLoaderImpl::lazyLoadOneMDString(unsigned ID) { + ++NumMDStringLoaded; + if (Metadata *MD = MetadataList.lookup(ID)) + return cast<MDString>(MD); + auto MDS = MDString::get(Context, MDStringRef[ID]); + MetadataList.assignValue(MDS, ID); + return MDS; +} + +void MetadataLoader::MetadataLoaderImpl::lazyLoadOneMetadata( + unsigned ID, PlaceholderQueue &Placeholders) { + assert(ID < (MDStringRef.size()) + GlobalMetadataBitPosIndex.size()); + assert(ID >= MDStringRef.size() && "Unexpected lazy-loading of MDString"); + // Lookup first if the metadata hasn't already been loaded. + if (auto *MD = MetadataList.lookup(ID)) { + auto *N = cast<MDNode>(MD); + if (!N->isTemporary()) + return; + } + SmallVector<uint64_t, 64> Record; + StringRef Blob; + if (Error Err = IndexCursor.JumpToBit( + GlobalMetadataBitPosIndex[ID - MDStringRef.size()])) + report_fatal_error("lazyLoadOneMetadata failed jumping: " + + Twine(toString(std::move(Err)))); + BitstreamEntry Entry; + if (Error E = IndexCursor.advanceSkippingSubblocks().moveInto(Entry)) + // FIXME this drops the error on the floor. + report_fatal_error("lazyLoadOneMetadata failed advanceSkippingSubblocks: " + + Twine(toString(std::move(E)))); + ++NumMDRecordLoaded; + if (Expected<unsigned> MaybeCode = + IndexCursor.readRecord(Entry.ID, Record, &Blob)) { + if (Error Err = + parseOneMetadata(Record, MaybeCode.get(), Placeholders, Blob, ID)) + report_fatal_error("Can't lazyload MD, parseOneMetadata: " + + Twine(toString(std::move(Err)))); + } else + report_fatal_error("Can't lazyload MD: " + + Twine(toString(MaybeCode.takeError()))); +} + +/// Ensure that all forward-references and placeholders are resolved. +/// Iteratively lazy-loading metadata on-demand if needed. +void MetadataLoader::MetadataLoaderImpl::resolveForwardRefsAndPlaceholders( + PlaceholderQueue &Placeholders) { + DenseSet<unsigned> Temporaries; + while (true) { + // Populate Temporaries with the placeholders that haven't been loaded yet. + Placeholders.getTemporaries(MetadataList, Temporaries); + + // If we don't have any temporary, or FwdReference, we're done! + if (Temporaries.empty() && !MetadataList.hasFwdRefs()) + break; + + // First, load all the temporaries. This can add new placeholders or + // forward references. + for (auto ID : Temporaries) + lazyLoadOneMetadata(ID, Placeholders); + Temporaries.clear(); + + // Second, load the forward-references. This can also add new placeholders + // or forward references. + while (MetadataList.hasFwdRefs()) + lazyLoadOneMetadata(MetadataList.getNextFwdRef(), Placeholders); + } + // At this point we don't have any forward reference remaining, or temporary + // that haven't been loaded. We can safely drop RAUW support and mark cycles + // as resolved. + MetadataList.tryToResolveCycles(); + + // Finally, everything is in place, we can replace the placeholders operands + // with the final node they refer to. + Placeholders.flush(MetadataList); +} + +Error MetadataLoader::MetadataLoaderImpl::parseOneMetadata( + SmallVectorImpl<uint64_t> &Record, unsigned Code, + PlaceholderQueue &Placeholders, StringRef Blob, unsigned &NextMetadataNo) { + + bool IsDistinct = false; + auto getMD = [&](unsigned ID) -> Metadata * { + if (ID < MDStringRef.size()) + return lazyLoadOneMDString(ID); + if (!IsDistinct) { + if (auto *MD = MetadataList.lookup(ID)) + return MD; + // If lazy-loading is enabled, we try recursively to load the operand + // instead of creating a temporary. + if (ID < (MDStringRef.size() + GlobalMetadataBitPosIndex.size())) { + // Create a temporary for the node that is referencing the operand we + // will lazy-load. It is needed before recursing in case there are + // uniquing cycles. + MetadataList.getMetadataFwdRef(NextMetadataNo); + lazyLoadOneMetadata(ID, Placeholders); + return MetadataList.lookup(ID); + } + // Return a temporary. + return MetadataList.getMetadataFwdRef(ID); + } + if (auto *MD = MetadataList.getMetadataIfResolved(ID)) + return MD; + return &Placeholders.getPlaceholderOp(ID); + }; + auto getMDOrNull = [&](unsigned ID) -> Metadata * { + if (ID) + return getMD(ID - 1); + return nullptr; + }; + auto getMDOrNullWithoutPlaceholders = [&](unsigned ID) -> Metadata * { + if (ID) + return MetadataList.getMetadataFwdRef(ID - 1); + return nullptr; + }; + auto getMDString = [&](unsigned ID) -> MDString * { + // This requires that the ID is not really a forward reference. In + // particular, the MDString must already have been resolved. + auto MDS = getMDOrNull(ID); + return cast_or_null<MDString>(MDS); + }; + + // Support for old type refs. + auto getDITypeRefOrNull = [&](unsigned ID) { + return MetadataList.upgradeTypeRef(getMDOrNull(ID)); + }; + +#define GET_OR_DISTINCT(CLASS, ARGS) \ + (IsDistinct ? CLASS::getDistinct ARGS : CLASS::get ARGS) + + switch (Code) { + default: // Default behavior: ignore. + break; + case bitc::METADATA_NAME: { + // Read name of the named metadata. + SmallString<8> Name(Record.begin(), Record.end()); + Record.clear(); + if (Error E = Stream.ReadCode().moveInto(Code)) + return E; + + ++NumMDRecordLoaded; + if (Expected<unsigned> MaybeNextBitCode = Stream.readRecord(Code, Record)) { + if (MaybeNextBitCode.get() != bitc::METADATA_NAMED_NODE) + return error("METADATA_NAME not followed by METADATA_NAMED_NODE"); + } else + return MaybeNextBitCode.takeError(); + + // Read named metadata elements. + unsigned Size = Record.size(); + NamedMDNode *NMD = TheModule.getOrInsertNamedMetadata(Name); + for (unsigned i = 0; i != Size; ++i) { + MDNode *MD = MetadataList.getMDNodeFwdRefOrNull(Record[i]); + if (!MD) + return error("Invalid named metadata: expect fwd ref to MDNode"); + NMD->addOperand(MD); + } + break; + } + case bitc::METADATA_OLD_FN_NODE: { + // Deprecated, but still needed to read old bitcode files. + // This is a LocalAsMetadata record, the only type of function-local + // metadata. + if (Record.size() % 2 == 1) + return error("Invalid record"); + + // If this isn't a LocalAsMetadata record, we're dropping it. This used + // to be legal, but there's no upgrade path. + auto dropRecord = [&] { + MetadataList.assignValue(MDNode::get(Context, None), NextMetadataNo); + NextMetadataNo++; + }; + if (Record.size() != 2) { + dropRecord(); + break; + } + + Type *Ty = getTypeByID(Record[0]); + if (Ty->isMetadataTy() || Ty->isVoidTy()) { + dropRecord(); + break; + } + + MetadataList.assignValue( + LocalAsMetadata::get(ValueList.getValueFwdRef(Record[1], Ty)), + NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_OLD_NODE: { + // Deprecated, but still needed to read old bitcode files. + if (Record.size() % 2 == 1) + return error("Invalid record"); + + unsigned Size = Record.size(); + SmallVector<Metadata *, 8> Elts; + for (unsigned i = 0; i != Size; i += 2) { + Type *Ty = getTypeByID(Record[i]); + if (!Ty) + return error("Invalid record"); + if (Ty->isMetadataTy()) + Elts.push_back(getMD(Record[i + 1])); + else if (!Ty->isVoidTy()) { + auto *MD = + ValueAsMetadata::get(ValueList.getValueFwdRef(Record[i + 1], Ty)); + assert(isa<ConstantAsMetadata>(MD) && + "Expected non-function-local metadata"); + Elts.push_back(MD); + } else + Elts.push_back(nullptr); + } + MetadataList.assignValue(MDNode::get(Context, Elts), NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_VALUE: { + if (Record.size() != 2) + return error("Invalid record"); + + Type *Ty = getTypeByID(Record[0]); + if (Ty->isMetadataTy() || Ty->isVoidTy()) + return error("Invalid record"); + + MetadataList.assignValue( + ValueAsMetadata::get(ValueList.getValueFwdRef(Record[1], Ty)), + NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_DISTINCT_NODE: + IsDistinct = true; + LLVM_FALLTHROUGH; + case bitc::METADATA_NODE: { + SmallVector<Metadata *, 8> Elts; + Elts.reserve(Record.size()); + for (unsigned ID : Record) + Elts.push_back(getMDOrNull(ID)); + MetadataList.assignValue(IsDistinct ? MDNode::getDistinct(Context, Elts) + : MDNode::get(Context, Elts), + NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_LOCATION: { + if (Record.size() != 5 && Record.size() != 6) + return error("Invalid record"); + + IsDistinct = Record[0]; + unsigned Line = Record[1]; + unsigned Column = Record[2]; + Metadata *Scope = getMD(Record[3]); + Metadata *InlinedAt = getMDOrNull(Record[4]); + bool ImplicitCode = Record.size() == 6 && Record[5]; + MetadataList.assignValue( + GET_OR_DISTINCT(DILocation, (Context, Line, Column, Scope, InlinedAt, + ImplicitCode)), + NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_GENERIC_DEBUG: { + if (Record.size() < 4) + return error("Invalid record"); + + IsDistinct = Record[0]; + unsigned Tag = Record[1]; + unsigned Version = Record[2]; + + if (Tag >= 1u << 16 || Version != 0) + return error("Invalid record"); + + auto *Header = getMDString(Record[3]); + SmallVector<Metadata *, 8> DwarfOps; + for (unsigned I = 4, E = Record.size(); I != E; ++I) + DwarfOps.push_back(getMDOrNull(Record[I])); + MetadataList.assignValue( + GET_OR_DISTINCT(GenericDINode, (Context, Tag, Header, DwarfOps)), + NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_SUBRANGE: { + Metadata *Val = nullptr; + // Operand 'count' is interpreted as: + // - Signed integer (version 0) + // - Metadata node (version 1) + // Operand 'lowerBound' is interpreted as: + // - Signed integer (version 0 and 1) + // - Metadata node (version 2) + // Operands 'upperBound' and 'stride' are interpreted as: + // - Metadata node (version 2) + switch (Record[0] >> 1) { + case 0: + Val = GET_OR_DISTINCT(DISubrange, + (Context, Record[1], unrotateSign(Record[2]))); + break; + case 1: + Val = GET_OR_DISTINCT(DISubrange, (Context, getMDOrNull(Record[1]), + unrotateSign(Record[2]))); + break; + case 2: + Val = GET_OR_DISTINCT( + DISubrange, (Context, getMDOrNull(Record[1]), getMDOrNull(Record[2]), + getMDOrNull(Record[3]), getMDOrNull(Record[4]))); + break; + default: + return error("Invalid record: Unsupported version of DISubrange"); + } + + MetadataList.assignValue(Val, NextMetadataNo); + IsDistinct = Record[0] & 1; + NextMetadataNo++; + break; + } + case bitc::METADATA_GENERIC_SUBRANGE: { + Metadata *Val = nullptr; + Val = GET_OR_DISTINCT(DIGenericSubrange, + (Context, getMDOrNull(Record[1]), + getMDOrNull(Record[2]), getMDOrNull(Record[3]), + getMDOrNull(Record[4]))); + + MetadataList.assignValue(Val, NextMetadataNo); + IsDistinct = Record[0] & 1; + NextMetadataNo++; + break; + } + case bitc::METADATA_ENUMERATOR: { + if (Record.size() < 3) + return error("Invalid record"); + + IsDistinct = Record[0] & 1; + bool IsUnsigned = Record[0] & 2; + bool IsBigInt = Record[0] & 4; + APInt Value; + + if (IsBigInt) { + const uint64_t BitWidth = Record[1]; + const size_t NumWords = Record.size() - 3; + Value = readWideAPInt(makeArrayRef(&Record[3], NumWords), BitWidth); + } else + Value = APInt(64, unrotateSign(Record[1]), !IsUnsigned); + + MetadataList.assignValue( + GET_OR_DISTINCT(DIEnumerator, + (Context, Value, IsUnsigned, getMDString(Record[2]))), + NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_BASIC_TYPE: { + if (Record.size() < 6 || Record.size() > 7) + return error("Invalid record"); + + IsDistinct = Record[0]; + DINode::DIFlags Flags = (Record.size() > 6) + ? static_cast<DINode::DIFlags>(Record[6]) + : DINode::FlagZero; + + MetadataList.assignValue( + GET_OR_DISTINCT(DIBasicType, + (Context, Record[1], getMDString(Record[2]), Record[3], + Record[4], Record[5], Flags)), + NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_STRING_TYPE: { + if (Record.size() > 9 || Record.size() < 8) + return error("Invalid record"); + + IsDistinct = Record[0]; + bool SizeIs8 = Record.size() == 8; + // StringLocationExp (i.e. Record[5]) is added at a later time + // than the other fields. The code here enables backward compatibility. + Metadata *StringLocationExp = SizeIs8 ? nullptr : getMDOrNull(Record[5]); + unsigned Offset = SizeIs8 ? 5 : 6; + MetadataList.assignValue( + GET_OR_DISTINCT(DIStringType, + (Context, Record[1], getMDString(Record[2]), + getMDOrNull(Record[3]), getMDOrNull(Record[4]), + StringLocationExp, Record[Offset], Record[Offset + 1], + Record[Offset + 2])), + NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_DERIVED_TYPE: { + if (Record.size() < 12 || Record.size() > 14) + return error("Invalid record"); + + // DWARF address space is encoded as N->getDWARFAddressSpace() + 1. 0 means + // that there is no DWARF address space associated with DIDerivedType. + Optional<unsigned> DWARFAddressSpace; + if (Record.size() > 12 && Record[12]) + DWARFAddressSpace = Record[12] - 1; + + Metadata *Annotations = nullptr; + if (Record.size() > 13 && Record[13]) + Annotations = getMDOrNull(Record[13]); + + IsDistinct = Record[0]; + DINode::DIFlags Flags = static_cast<DINode::DIFlags>(Record[10]); + MetadataList.assignValue( + GET_OR_DISTINCT(DIDerivedType, + (Context, Record[1], getMDString(Record[2]), + getMDOrNull(Record[3]), Record[4], + getDITypeRefOrNull(Record[5]), + getDITypeRefOrNull(Record[6]), Record[7], Record[8], + Record[9], DWARFAddressSpace, Flags, + getDITypeRefOrNull(Record[11]), Annotations)), + NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_COMPOSITE_TYPE: { + if (Record.size() < 16 || Record.size() > 22) + return error("Invalid record"); + + // If we have a UUID and this is not a forward declaration, lookup the + // mapping. + IsDistinct = Record[0] & 0x1; + bool IsNotUsedInTypeRef = Record[0] >= 2; + unsigned Tag = Record[1]; + MDString *Name = getMDString(Record[2]); + Metadata *File = getMDOrNull(Record[3]); + unsigned Line = Record[4]; + Metadata *Scope = getDITypeRefOrNull(Record[5]); + Metadata *BaseType = nullptr; + uint64_t SizeInBits = Record[7]; + if (Record[8] > (uint64_t)std::numeric_limits<uint32_t>::max()) + return error("Alignment value is too large"); + uint32_t AlignInBits = Record[8]; + uint64_t OffsetInBits = 0; + DINode::DIFlags Flags = static_cast<DINode::DIFlags>(Record[10]); + Metadata *Elements = nullptr; + unsigned RuntimeLang = Record[12]; + Metadata *VTableHolder = nullptr; + Metadata *TemplateParams = nullptr; + Metadata *Discriminator = nullptr; + Metadata *DataLocation = nullptr; + Metadata *Associated = nullptr; + Metadata *Allocated = nullptr; + Metadata *Rank = nullptr; + Metadata *Annotations = nullptr; + auto *Identifier = getMDString(Record[15]); + // If this module is being parsed so that it can be ThinLTO imported + // into another module, composite types only need to be imported + // as type declarations (unless full type definitions requested). + // Create type declarations up front to save memory. Also, buildODRType + // handles the case where this is type ODRed with a definition needed + // by the importing module, in which case the existing definition is + // used. + if (IsImporting && !ImportFullTypeDefinitions && Identifier && + (Tag == dwarf::DW_TAG_enumeration_type || + Tag == dwarf::DW_TAG_class_type || + Tag == dwarf::DW_TAG_structure_type || + Tag == dwarf::DW_TAG_union_type)) { + Flags = Flags | DINode::FlagFwdDecl; + } else { + BaseType = getDITypeRefOrNull(Record[6]); + OffsetInBits = Record[9]; + Elements = getMDOrNull(Record[11]); + VTableHolder = getDITypeRefOrNull(Record[13]); + TemplateParams = getMDOrNull(Record[14]); + if (Record.size() > 16) + Discriminator = getMDOrNull(Record[16]); + if (Record.size() > 17) + DataLocation = getMDOrNull(Record[17]); + if (Record.size() > 19) { + Associated = getMDOrNull(Record[18]); + Allocated = getMDOrNull(Record[19]); + } + if (Record.size() > 20) { + Rank = getMDOrNull(Record[20]); + } + if (Record.size() > 21) { + Annotations = getMDOrNull(Record[21]); + } + } + DICompositeType *CT = nullptr; + if (Identifier) + CT = DICompositeType::buildODRType( + Context, *Identifier, Tag, Name, File, Line, Scope, BaseType, + SizeInBits, AlignInBits, OffsetInBits, Flags, Elements, RuntimeLang, + VTableHolder, TemplateParams, Discriminator, DataLocation, Associated, + Allocated, Rank, Annotations); + + // Create a node if we didn't get a lazy ODR type. + if (!CT) + CT = GET_OR_DISTINCT(DICompositeType, + (Context, Tag, Name, File, Line, Scope, BaseType, + SizeInBits, AlignInBits, OffsetInBits, Flags, + Elements, RuntimeLang, VTableHolder, TemplateParams, + Identifier, Discriminator, DataLocation, Associated, + Allocated, Rank, Annotations)); + if (!IsNotUsedInTypeRef && Identifier) + MetadataList.addTypeRef(*Identifier, *cast<DICompositeType>(CT)); + + MetadataList.assignValue(CT, NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_SUBROUTINE_TYPE: { + if (Record.size() < 3 || Record.size() > 4) + return error("Invalid record"); + bool IsOldTypeRefArray = Record[0] < 2; + unsigned CC = (Record.size() > 3) ? Record[3] : 0; + + IsDistinct = Record[0] & 0x1; + DINode::DIFlags Flags = static_cast<DINode::DIFlags>(Record[1]); + Metadata *Types = getMDOrNull(Record[2]); + if (LLVM_UNLIKELY(IsOldTypeRefArray)) + Types = MetadataList.upgradeTypeRefArray(Types); + + MetadataList.assignValue( + GET_OR_DISTINCT(DISubroutineType, (Context, Flags, CC, Types)), + NextMetadataNo); + NextMetadataNo++; + break; + } + + case bitc::METADATA_MODULE: { + if (Record.size() < 5 || Record.size() > 9) + return error("Invalid record"); + + unsigned Offset = Record.size() >= 8 ? 2 : 1; + IsDistinct = Record[0]; + MetadataList.assignValue( + GET_OR_DISTINCT( + DIModule, + (Context, Record.size() >= 8 ? getMDOrNull(Record[1]) : nullptr, + getMDOrNull(Record[0 + Offset]), getMDString(Record[1 + Offset]), + getMDString(Record[2 + Offset]), getMDString(Record[3 + Offset]), + getMDString(Record[4 + Offset]), + Record.size() <= 7 ? 0 : Record[7], + Record.size() <= 8 ? false : Record[8])), + NextMetadataNo); + NextMetadataNo++; + break; + } + + case bitc::METADATA_FILE: { + if (Record.size() != 3 && Record.size() != 5 && Record.size() != 6) + return error("Invalid record"); + + IsDistinct = Record[0]; + Optional<DIFile::ChecksumInfo<MDString *>> Checksum; + // The BitcodeWriter writes null bytes into Record[3:4] when the Checksum + // is not present. This matches up with the old internal representation, + // and the old encoding for CSK_None in the ChecksumKind. The new + // representation reserves the value 0 in the ChecksumKind to continue to + // encode None in a backwards-compatible way. + if (Record.size() > 4 && Record[3] && Record[4]) + Checksum.emplace(static_cast<DIFile::ChecksumKind>(Record[3]), + getMDString(Record[4])); + MetadataList.assignValue( + GET_OR_DISTINCT( + DIFile, + (Context, getMDString(Record[1]), getMDString(Record[2]), Checksum, + Record.size() > 5 ? Optional<MDString *>(getMDString(Record[5])) + : None)), + NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_COMPILE_UNIT: { + if (Record.size() < 14 || Record.size() > 22) + return error("Invalid record"); + + // Ignore Record[0], which indicates whether this compile unit is + // distinct. It's always distinct. + IsDistinct = true; + auto *CU = DICompileUnit::getDistinct( + Context, Record[1], getMDOrNull(Record[2]), getMDString(Record[3]), + Record[4], getMDString(Record[5]), Record[6], getMDString(Record[7]), + Record[8], getMDOrNull(Record[9]), getMDOrNull(Record[10]), + getMDOrNull(Record[12]), getMDOrNull(Record[13]), + Record.size() <= 15 ? nullptr : getMDOrNull(Record[15]), + Record.size() <= 14 ? 0 : Record[14], + Record.size() <= 16 ? true : Record[16], + Record.size() <= 17 ? false : Record[17], + Record.size() <= 18 ? 0 : Record[18], + Record.size() <= 19 ? false : Record[19], + Record.size() <= 20 ? nullptr : getMDString(Record[20]), + Record.size() <= 21 ? nullptr : getMDString(Record[21])); + + MetadataList.assignValue(CU, NextMetadataNo); + NextMetadataNo++; + + // Move the Upgrade the list of subprograms. + if (Metadata *SPs = getMDOrNullWithoutPlaceholders(Record[11])) + CUSubprograms.push_back({CU, SPs}); + break; + } + case bitc::METADATA_SUBPROGRAM: { + if (Record.size() < 18 || Record.size() > 21) + return error("Invalid record"); + + bool HasSPFlags = Record[0] & 4; + + DINode::DIFlags Flags; + DISubprogram::DISPFlags SPFlags; + if (!HasSPFlags) + Flags = static_cast<DINode::DIFlags>(Record[11 + 2]); + else { + Flags = static_cast<DINode::DIFlags>(Record[11]); + SPFlags = static_cast<DISubprogram::DISPFlags>(Record[9]); + } + + // Support for old metadata when + // subprogram specific flags are placed in DIFlags. + const unsigned DIFlagMainSubprogram = 1 << 21; + bool HasOldMainSubprogramFlag = Flags & DIFlagMainSubprogram; + if (HasOldMainSubprogramFlag) + // Remove old DIFlagMainSubprogram from DIFlags. + // Note: This assumes that any future use of bit 21 defaults to it + // being 0. + Flags &= ~static_cast<DINode::DIFlags>(DIFlagMainSubprogram); + + if (HasOldMainSubprogramFlag && HasSPFlags) + SPFlags |= DISubprogram::SPFlagMainSubprogram; + else if (!HasSPFlags) + SPFlags = DISubprogram::toSPFlags( + /*IsLocalToUnit=*/Record[7], /*IsDefinition=*/Record[8], + /*IsOptimized=*/Record[14], /*Virtuality=*/Record[11], + /*IsMainSubprogram=*/HasOldMainSubprogramFlag); + + // All definitions should be distinct. + IsDistinct = (Record[0] & 1) || (SPFlags & DISubprogram::SPFlagDefinition); + // Version 1 has a Function as Record[15]. + // Version 2 has removed Record[15]. + // Version 3 has the Unit as Record[15]. + // Version 4 added thisAdjustment. + // Version 5 repacked flags into DISPFlags, changing many element numbers. + bool HasUnit = Record[0] & 2; + if (!HasSPFlags && HasUnit && Record.size() < 19) + return error("Invalid record"); + if (HasSPFlags && !HasUnit) + return error("Invalid record"); + // Accommodate older formats. + bool HasFn = false; + bool HasThisAdj = true; + bool HasThrownTypes = true; + bool HasAnnotations = false; + unsigned OffsetA = 0; + unsigned OffsetB = 0; + if (!HasSPFlags) { + OffsetA = 2; + OffsetB = 2; + if (Record.size() >= 19) { + HasFn = !HasUnit; + OffsetB++; + } + HasThisAdj = Record.size() >= 20; + HasThrownTypes = Record.size() >= 21; + } else { + HasAnnotations = Record.size() >= 19; + } + Metadata *CUorFn = getMDOrNull(Record[12 + OffsetB]); + DISubprogram *SP = GET_OR_DISTINCT( + DISubprogram, + (Context, + getDITypeRefOrNull(Record[1]), // scope + getMDString(Record[2]), // name + getMDString(Record[3]), // linkageName + getMDOrNull(Record[4]), // file + Record[5], // line + getMDOrNull(Record[6]), // type + Record[7 + OffsetA], // scopeLine + getDITypeRefOrNull(Record[8 + OffsetA]), // containingType + Record[10 + OffsetA], // virtualIndex + HasThisAdj ? Record[16 + OffsetB] : 0, // thisAdjustment + Flags, // flags + SPFlags, // SPFlags + HasUnit ? CUorFn : nullptr, // unit + getMDOrNull(Record[13 + OffsetB]), // templateParams + getMDOrNull(Record[14 + OffsetB]), // declaration + getMDOrNull(Record[15 + OffsetB]), // retainedNodes + HasThrownTypes ? getMDOrNull(Record[17 + OffsetB]) + : nullptr, // thrownTypes + HasAnnotations ? getMDOrNull(Record[18 + OffsetB]) + : nullptr // annotations + )); + MetadataList.assignValue(SP, NextMetadataNo); + NextMetadataNo++; + + // Upgrade sp->function mapping to function->sp mapping. + if (HasFn) { + if (auto *CMD = dyn_cast_or_null<ConstantAsMetadata>(CUorFn)) + if (auto *F = dyn_cast<Function>(CMD->getValue())) { + if (F->isMaterializable()) + // Defer until materialized; unmaterialized functions may not have + // metadata. + FunctionsWithSPs[F] = SP; + else if (!F->empty()) + F->setSubprogram(SP); + } + } + break; + } + case bitc::METADATA_LEXICAL_BLOCK: { + if (Record.size() != 5) + return error("Invalid record"); + + IsDistinct = Record[0]; + MetadataList.assignValue( + GET_OR_DISTINCT(DILexicalBlock, + (Context, getMDOrNull(Record[1]), + getMDOrNull(Record[2]), Record[3], Record[4])), + NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_LEXICAL_BLOCK_FILE: { + if (Record.size() != 4) + return error("Invalid record"); + + IsDistinct = Record[0]; + MetadataList.assignValue( + GET_OR_DISTINCT(DILexicalBlockFile, + (Context, getMDOrNull(Record[1]), + getMDOrNull(Record[2]), Record[3])), + NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_COMMON_BLOCK: { + IsDistinct = Record[0] & 1; + MetadataList.assignValue( + GET_OR_DISTINCT(DICommonBlock, + (Context, getMDOrNull(Record[1]), + getMDOrNull(Record[2]), getMDString(Record[3]), + getMDOrNull(Record[4]), Record[5])), + NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_NAMESPACE: { + // Newer versions of DINamespace dropped file and line. + MDString *Name; + if (Record.size() == 3) + Name = getMDString(Record[2]); + else if (Record.size() == 5) + Name = getMDString(Record[3]); + else + return error("Invalid record"); + + IsDistinct = Record[0] & 1; + bool ExportSymbols = Record[0] & 2; + MetadataList.assignValue( + GET_OR_DISTINCT(DINamespace, + (Context, getMDOrNull(Record[1]), Name, ExportSymbols)), + NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_MACRO: { + if (Record.size() != 5) + return error("Invalid record"); + + IsDistinct = Record[0]; + MetadataList.assignValue( + GET_OR_DISTINCT(DIMacro, + (Context, Record[1], Record[2], getMDString(Record[3]), + getMDString(Record[4]))), + NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_MACRO_FILE: { + if (Record.size() != 5) + return error("Invalid record"); + + IsDistinct = Record[0]; + MetadataList.assignValue( + GET_OR_DISTINCT(DIMacroFile, + (Context, Record[1], Record[2], getMDOrNull(Record[3]), + getMDOrNull(Record[4]))), + NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_TEMPLATE_TYPE: { + if (Record.size() < 3 || Record.size() > 4) + return error("Invalid record"); + + IsDistinct = Record[0]; + MetadataList.assignValue( + GET_OR_DISTINCT(DITemplateTypeParameter, + (Context, getMDString(Record[1]), + getDITypeRefOrNull(Record[2]), + (Record.size() == 4) ? getMDOrNull(Record[3]) + : getMDOrNull(false))), + NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_TEMPLATE_VALUE: { + if (Record.size() < 5 || Record.size() > 6) + return error("Invalid record"); + + IsDistinct = Record[0]; + + MetadataList.assignValue( + GET_OR_DISTINCT( + DITemplateValueParameter, + (Context, Record[1], getMDString(Record[2]), + getDITypeRefOrNull(Record[3]), + (Record.size() == 6) ? getMDOrNull(Record[4]) : getMDOrNull(false), + (Record.size() == 6) ? getMDOrNull(Record[5]) + : getMDOrNull(Record[4]))), + NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_GLOBAL_VAR: { + if (Record.size() < 11 || Record.size() > 13) + return error("Invalid record"); + + IsDistinct = Record[0] & 1; + unsigned Version = Record[0] >> 1; + + if (Version == 2) { + Metadata *Annotations = nullptr; + if (Record.size() > 12) + Annotations = getMDOrNull(Record[12]); + + MetadataList.assignValue( + GET_OR_DISTINCT(DIGlobalVariable, + (Context, getMDOrNull(Record[1]), + getMDString(Record[2]), getMDString(Record[3]), + getMDOrNull(Record[4]), Record[5], + getDITypeRefOrNull(Record[6]), Record[7], Record[8], + getMDOrNull(Record[9]), getMDOrNull(Record[10]), + Record[11], Annotations)), + NextMetadataNo); + + NextMetadataNo++; + } else if (Version == 1) { + // No upgrade necessary. A null field will be introduced to indicate + // that no parameter information is available. + MetadataList.assignValue( + GET_OR_DISTINCT( + DIGlobalVariable, + (Context, getMDOrNull(Record[1]), getMDString(Record[2]), + getMDString(Record[3]), getMDOrNull(Record[4]), Record[5], + getDITypeRefOrNull(Record[6]), Record[7], Record[8], + getMDOrNull(Record[10]), nullptr, Record[11], nullptr)), + NextMetadataNo); + + NextMetadataNo++; + } else if (Version == 0) { + // Upgrade old metadata, which stored a global variable reference or a + // ConstantInt here. + NeedUpgradeToDIGlobalVariableExpression = true; + Metadata *Expr = getMDOrNull(Record[9]); + uint32_t AlignInBits = 0; + if (Record.size() > 11) { + if (Record[11] > (uint64_t)std::numeric_limits<uint32_t>::max()) + return error("Alignment value is too large"); + AlignInBits = Record[11]; + } + GlobalVariable *Attach = nullptr; + if (auto *CMD = dyn_cast_or_null<ConstantAsMetadata>(Expr)) { + if (auto *GV = dyn_cast<GlobalVariable>(CMD->getValue())) { + Attach = GV; + Expr = nullptr; + } else if (auto *CI = dyn_cast<ConstantInt>(CMD->getValue())) { + Expr = DIExpression::get(Context, + {dwarf::DW_OP_constu, CI->getZExtValue(), + dwarf::DW_OP_stack_value}); + } else { + Expr = nullptr; + } + } + DIGlobalVariable *DGV = GET_OR_DISTINCT( + DIGlobalVariable, + (Context, getMDOrNull(Record[1]), getMDString(Record[2]), + getMDString(Record[3]), getMDOrNull(Record[4]), Record[5], + getDITypeRefOrNull(Record[6]), Record[7], Record[8], + getMDOrNull(Record[10]), nullptr, AlignInBits, nullptr)); + + DIGlobalVariableExpression *DGVE = nullptr; + if (Attach || Expr) + DGVE = DIGlobalVariableExpression::getDistinct( + Context, DGV, Expr ? Expr : DIExpression::get(Context, {})); + if (Attach) + Attach->addDebugInfo(DGVE); + + auto *MDNode = Expr ? cast<Metadata>(DGVE) : cast<Metadata>(DGV); + MetadataList.assignValue(MDNode, NextMetadataNo); + NextMetadataNo++; + } else + return error("Invalid record"); + + break; + } + case bitc::METADATA_LOCAL_VAR: { + // 10th field is for the obseleted 'inlinedAt:' field. + if (Record.size() < 8 || Record.size() > 10) + return error("Invalid record"); + + IsDistinct = Record[0] & 1; + bool HasAlignment = Record[0] & 2; + // 2nd field used to be an artificial tag, either DW_TAG_auto_variable or + // DW_TAG_arg_variable, if we have alignment flag encoded it means, that + // this is newer version of record which doesn't have artificial tag. + bool HasTag = !HasAlignment && Record.size() > 8; + DINode::DIFlags Flags = static_cast<DINode::DIFlags>(Record[7 + HasTag]); + uint32_t AlignInBits = 0; + Metadata *Annotations = nullptr; + if (HasAlignment) { + if (Record[8] > (uint64_t)std::numeric_limits<uint32_t>::max()) + return error("Alignment value is too large"); + AlignInBits = Record[8]; + if (Record.size() > 9) + Annotations = getMDOrNull(Record[9]); + } + + MetadataList.assignValue( + GET_OR_DISTINCT(DILocalVariable, + (Context, getMDOrNull(Record[1 + HasTag]), + getMDString(Record[2 + HasTag]), + getMDOrNull(Record[3 + HasTag]), Record[4 + HasTag], + getDITypeRefOrNull(Record[5 + HasTag]), + Record[6 + HasTag], Flags, AlignInBits, Annotations)), + NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_LABEL: { + if (Record.size() != 5) + return error("Invalid record"); + + IsDistinct = Record[0] & 1; + MetadataList.assignValue( + GET_OR_DISTINCT(DILabel, (Context, getMDOrNull(Record[1]), + getMDString(Record[2]), + getMDOrNull(Record[3]), Record[4])), + NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_EXPRESSION: { + if (Record.size() < 1) + return error("Invalid record"); + + IsDistinct = Record[0] & 1; + uint64_t Version = Record[0] >> 1; + auto Elts = MutableArrayRef<uint64_t>(Record).slice(1); + + SmallVector<uint64_t, 6> Buffer; + if (Error Err = upgradeDIExpression(Version, Elts, Buffer)) + return Err; + + MetadataList.assignValue(GET_OR_DISTINCT(DIExpression, (Context, Elts)), + NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_GLOBAL_VAR_EXPR: { + if (Record.size() != 3) + return error("Invalid record"); + + IsDistinct = Record[0]; + Metadata *Expr = getMDOrNull(Record[2]); + if (!Expr) + Expr = DIExpression::get(Context, {}); + MetadataList.assignValue( + GET_OR_DISTINCT(DIGlobalVariableExpression, + (Context, getMDOrNull(Record[1]), Expr)), + NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_OBJC_PROPERTY: { + if (Record.size() != 8) + return error("Invalid record"); + + IsDistinct = Record[0]; + MetadataList.assignValue( + GET_OR_DISTINCT(DIObjCProperty, + (Context, getMDString(Record[1]), + getMDOrNull(Record[2]), Record[3], + getMDString(Record[4]), getMDString(Record[5]), + Record[6], getDITypeRefOrNull(Record[7]))), + NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_IMPORTED_ENTITY: { + if (Record.size() < 6 && Record.size() > 8) + return error("Invalid record"); + + IsDistinct = Record[0]; + bool HasFile = (Record.size() >= 7); + bool HasElements = (Record.size() >= 8); + MetadataList.assignValue( + GET_OR_DISTINCT(DIImportedEntity, + (Context, Record[1], getMDOrNull(Record[2]), + getDITypeRefOrNull(Record[3]), + HasFile ? getMDOrNull(Record[6]) : nullptr, + HasFile ? Record[4] : 0, getMDString(Record[5]), + HasElements ? getMDOrNull(Record[7]) : nullptr)), + NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_STRING_OLD: { + std::string String(Record.begin(), Record.end()); + + // Test for upgrading !llvm.loop. + HasSeenOldLoopTags |= mayBeOldLoopAttachmentTag(String); + ++NumMDStringLoaded; + Metadata *MD = MDString::get(Context, String); + MetadataList.assignValue(MD, NextMetadataNo); + NextMetadataNo++; + break; + } + case bitc::METADATA_STRINGS: { + auto CreateNextMDString = [&](StringRef Str) { + ++NumMDStringLoaded; + MetadataList.assignValue(MDString::get(Context, Str), NextMetadataNo); + NextMetadataNo++; + }; + if (Error Err = parseMetadataStrings(Record, Blob, CreateNextMDString)) + return Err; + break; + } + case bitc::METADATA_GLOBAL_DECL_ATTACHMENT: { + if (Record.size() % 2 == 0) + return error("Invalid record"); + unsigned ValueID = Record[0]; + if (ValueID >= ValueList.size()) + return error("Invalid record"); + if (auto *GO = dyn_cast<GlobalObject>(ValueList[ValueID])) + if (Error Err = parseGlobalObjectAttachment( + *GO, ArrayRef<uint64_t>(Record).slice(1))) + return Err; + break; + } + case bitc::METADATA_KIND: { + // Support older bitcode files that had METADATA_KIND records in a + // block with METADATA_BLOCK_ID. + if (Error Err = parseMetadataKindRecord(Record)) + return Err; + break; + } + case bitc::METADATA_ARG_LIST: { + SmallVector<ValueAsMetadata *, 4> Elts; + Elts.reserve(Record.size()); + for (uint64_t Elt : Record) { + Metadata *MD = getMD(Elt); + if (isa<MDNode>(MD) && cast<MDNode>(MD)->isTemporary()) + return error( + "Invalid record: DIArgList should not contain forward refs"); + if (!isa<ValueAsMetadata>(MD)) + return error("Invalid record"); + Elts.push_back(cast<ValueAsMetadata>(MD)); + } + + MetadataList.assignValue(DIArgList::get(Context, Elts), NextMetadataNo); + NextMetadataNo++; + break; + } + } + return Error::success(); +#undef GET_OR_DISTINCT +} + +Error MetadataLoader::MetadataLoaderImpl::parseMetadataStrings( + ArrayRef<uint64_t> Record, StringRef Blob, + function_ref<void(StringRef)> CallBack) { + // All the MDStrings in the block are emitted together in a single + // record. The strings are concatenated and stored in a blob along with + // their sizes. + if (Record.size() != 2) + return error("Invalid record: metadata strings layout"); + + unsigned NumStrings = Record[0]; + unsigned StringsOffset = Record[1]; + if (!NumStrings) + return error("Invalid record: metadata strings with no strings"); + if (StringsOffset > Blob.size()) + return error("Invalid record: metadata strings corrupt offset"); + + StringRef Lengths = Blob.slice(0, StringsOffset); + SimpleBitstreamCursor R(Lengths); + + StringRef Strings = Blob.drop_front(StringsOffset); + do { + if (R.AtEndOfStream()) + return error("Invalid record: metadata strings bad length"); + + uint32_t Size; + if (Error E = R.ReadVBR(6).moveInto(Size)) + return E; + if (Strings.size() < Size) + return error("Invalid record: metadata strings truncated chars"); + + CallBack(Strings.slice(0, Size)); + Strings = Strings.drop_front(Size); + } while (--NumStrings); + + return Error::success(); +} + +Error MetadataLoader::MetadataLoaderImpl::parseGlobalObjectAttachment( + GlobalObject &GO, ArrayRef<uint64_t> Record) { + assert(Record.size() % 2 == 0); + for (unsigned I = 0, E = Record.size(); I != E; I += 2) { + auto K = MDKindMap.find(Record[I]); + if (K == MDKindMap.end()) + return error("Invalid ID"); + MDNode *MD = + dyn_cast_or_null<MDNode>(getMetadataFwdRefOrLoad(Record[I + 1])); + if (!MD) + return error("Invalid metadata attachment: expect fwd ref to MDNode"); + GO.addMetadata(K->second, *MD); + } + return Error::success(); +} + +/// Parse metadata attachments. +Error MetadataLoader::MetadataLoaderImpl::parseMetadataAttachment( + Function &F, const SmallVectorImpl<Instruction *> &InstructionList) { + if (Error Err = Stream.EnterSubBlock(bitc::METADATA_ATTACHMENT_ID)) + return Err; + + SmallVector<uint64_t, 64> Record; + PlaceholderQueue Placeholders; + + while (true) { + BitstreamEntry Entry; + if (Error E = Stream.advanceSkippingSubblocks().moveInto(Entry)) + return E; + + switch (Entry.Kind) { + case BitstreamEntry::SubBlock: // Handled for us already. + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + resolveForwardRefsAndPlaceholders(Placeholders); + return Error::success(); + case BitstreamEntry::Record: + // The interesting case. + break; + } + + // Read a metadata attachment record. + Record.clear(); + ++NumMDRecordLoaded; + Expected<unsigned> MaybeRecord = Stream.readRecord(Entry.ID, Record); + if (!MaybeRecord) + return MaybeRecord.takeError(); + switch (MaybeRecord.get()) { + default: // Default behavior: ignore. + break; + case bitc::METADATA_ATTACHMENT: { + unsigned RecordLength = Record.size(); + if (Record.empty()) + return error("Invalid record"); + if (RecordLength % 2 == 0) { + // A function attachment. + if (Error Err = parseGlobalObjectAttachment(F, Record)) + return Err; + continue; + } + + // An instruction attachment. + Instruction *Inst = InstructionList[Record[0]]; + for (unsigned i = 1; i != RecordLength; i = i + 2) { + unsigned Kind = Record[i]; + DenseMap<unsigned, unsigned>::iterator I = MDKindMap.find(Kind); + if (I == MDKindMap.end()) + return error("Invalid ID"); + if (I->second == LLVMContext::MD_tbaa && StripTBAA) + continue; + + auto Idx = Record[i + 1]; + if (Idx < (MDStringRef.size() + GlobalMetadataBitPosIndex.size()) && + !MetadataList.lookup(Idx)) { + // Load the attachment if it is in the lazy-loadable range and hasn't + // been loaded yet. + lazyLoadOneMetadata(Idx, Placeholders); + resolveForwardRefsAndPlaceholders(Placeholders); + } + + Metadata *Node = MetadataList.getMetadataFwdRef(Idx); + if (isa<LocalAsMetadata>(Node)) + // Drop the attachment. This used to be legal, but there's no + // upgrade path. + break; + MDNode *MD = dyn_cast_or_null<MDNode>(Node); + if (!MD) + return error("Invalid metadata attachment"); + + if (HasSeenOldLoopTags && I->second == LLVMContext::MD_loop) + MD = upgradeInstructionLoopAttachment(*MD); + + if (I->second == LLVMContext::MD_tbaa) { + assert(!MD->isTemporary() && "should load MDs before attachments"); + MD = UpgradeTBAANode(*MD); + } + Inst->setMetadata(I->second, MD); + } + break; + } + } + } +} + +/// Parse a single METADATA_KIND record, inserting result in MDKindMap. +Error MetadataLoader::MetadataLoaderImpl::parseMetadataKindRecord( + SmallVectorImpl<uint64_t> &Record) { + if (Record.size() < 2) + return error("Invalid record"); + + unsigned Kind = Record[0]; + SmallString<8> Name(Record.begin() + 1, Record.end()); + + unsigned NewKind = TheModule.getMDKindID(Name.str()); + if (!MDKindMap.insert(std::make_pair(Kind, NewKind)).second) + return error("Conflicting METADATA_KIND records"); + return Error::success(); +} + +/// Parse the metadata kinds out of the METADATA_KIND_BLOCK. +Error MetadataLoader::MetadataLoaderImpl::parseMetadataKinds() { + if (Error Err = Stream.EnterSubBlock(bitc::METADATA_KIND_BLOCK_ID)) + return Err; + + SmallVector<uint64_t, 64> Record; + + // Read all the records. + while (true) { + BitstreamEntry Entry; + if (Error E = Stream.advanceSkippingSubblocks().moveInto(Entry)) + return E; + + switch (Entry.Kind) { + case BitstreamEntry::SubBlock: // Handled for us already. + case BitstreamEntry::Error: + return error("Malformed block"); + case BitstreamEntry::EndBlock: + return Error::success(); + case BitstreamEntry::Record: + // The interesting case. + break; + } + + // Read a record. + Record.clear(); + ++NumMDRecordLoaded; + Expected<unsigned> MaybeCode = Stream.readRecord(Entry.ID, Record); + if (!MaybeCode) + return MaybeCode.takeError(); + switch (MaybeCode.get()) { + default: // Default behavior: ignore. + break; + case bitc::METADATA_KIND: { + if (Error Err = parseMetadataKindRecord(Record)) + return Err; + break; + } + } + } +} + +MetadataLoader &MetadataLoader::operator=(MetadataLoader &&RHS) { + Pimpl = std::move(RHS.Pimpl); + return *this; +} +MetadataLoader::MetadataLoader(MetadataLoader &&RHS) + : Pimpl(std::move(RHS.Pimpl)) {} + +MetadataLoader::~MetadataLoader() = default; +MetadataLoader::MetadataLoader(BitstreamCursor &Stream, Module &TheModule, + BitcodeReaderValueList &ValueList, + bool IsImporting, + std::function<Type *(unsigned)> getTypeByID) + : Pimpl(std::make_unique<MetadataLoaderImpl>( + Stream, TheModule, ValueList, std::move(getTypeByID), IsImporting)) {} + +Error MetadataLoader::parseMetadata(bool ModuleLevel) { + return Pimpl->parseMetadata(ModuleLevel); +} + +bool MetadataLoader::hasFwdRefs() const { return Pimpl->hasFwdRefs(); } + +/// Return the given metadata, creating a replaceable forward reference if +/// necessary. +Metadata *MetadataLoader::getMetadataFwdRefOrLoad(unsigned Idx) { + return Pimpl->getMetadataFwdRefOrLoad(Idx); +} + +DISubprogram *MetadataLoader::lookupSubprogramForFunction(Function *F) { + return Pimpl->lookupSubprogramForFunction(F); +} + +Error MetadataLoader::parseMetadataAttachment( + Function &F, const SmallVectorImpl<Instruction *> &InstructionList) { + return Pimpl->parseMetadataAttachment(F, InstructionList); +} + +Error MetadataLoader::parseMetadataKinds() { + return Pimpl->parseMetadataKinds(); +} + +void MetadataLoader::setStripTBAA(bool StripTBAA) { + return Pimpl->setStripTBAA(StripTBAA); +} + +bool MetadataLoader::isStrippingTBAA() { return Pimpl->isStrippingTBAA(); } + +unsigned MetadataLoader::size() const { return Pimpl->size(); } +void MetadataLoader::shrinkTo(unsigned N) { return Pimpl->shrinkTo(N); } + +void MetadataLoader::upgradeDebugIntrinsics(Function &F) { + return Pimpl->upgradeDebugIntrinsics(F); +} diff --git a/contrib/libs/llvm14/lib/Bitcode/Reader/MetadataLoader.h b/contrib/libs/llvm14/lib/Bitcode/Reader/MetadataLoader.h new file mode 100644 index 0000000000..709800850f --- /dev/null +++ b/contrib/libs/llvm14/lib/Bitcode/Reader/MetadataLoader.h @@ -0,0 +1,83 @@ +//===-- Bitcode/Reader/MetadataLoader.h - Load Metadatas -------*- C++ -*-====// +// +// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. +// See https://llvm.org/LICENSE.txt for license information. +// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception +// +//===----------------------------------------------------------------------===// +// +// This class handles loading Metadatas. +// +//===----------------------------------------------------------------------===// + +#ifndef LLVM_LIB_BITCODE_READER_METADATALOADER_H +#define LLVM_LIB_BITCODE_READER_METADATALOADER_H + +#include "llvm/ADT/SmallVector.h" +#include "llvm/Support/Error.h" + +#include <functional> +#include <memory> + +namespace llvm { +class BitcodeReaderValueList; +class BitstreamCursor; +class DISubprogram; +class Function; +class Instruction; +class Metadata; +class Module; +class Type; + +/// Helper class that handles loading Metadatas and keeping them available. +class MetadataLoader { + class MetadataLoaderImpl; + std::unique_ptr<MetadataLoaderImpl> Pimpl; + Error parseMetadata(bool ModuleLevel); + +public: + ~MetadataLoader(); + MetadataLoader(BitstreamCursor &Stream, Module &TheModule, + BitcodeReaderValueList &ValueList, bool IsImporting, + std::function<Type *(unsigned)> getTypeByID); + MetadataLoader &operator=(MetadataLoader &&); + MetadataLoader(MetadataLoader &&); + + // Parse a module metadata block + Error parseModuleMetadata() { return parseMetadata(true); } + + // Parse a function metadata block + Error parseFunctionMetadata() { return parseMetadata(false); } + + /// Set the mode to strip TBAA metadata on load. + void setStripTBAA(bool StripTBAA = true); + + /// Return true if the Loader is stripping TBAA metadata. + bool isStrippingTBAA(); + + // Return true there are remaining unresolved forward references. + bool hasFwdRefs() const; + + /// Return the given metadata, creating a replaceable forward reference if + /// necessary. + Metadata *getMetadataFwdRefOrLoad(unsigned Idx); + + /// Return the DISubprogram metadata for a Function if any, null otherwise. + DISubprogram *lookupSubprogramForFunction(Function *F); + + /// Parse a `METADATA_ATTACHMENT` block for a function. + Error parseMetadataAttachment( + Function &F, const SmallVectorImpl<Instruction *> &InstructionList); + + /// Parse a `METADATA_KIND` block for the current module. + Error parseMetadataKinds(); + + unsigned size() const; + void shrinkTo(unsigned N); + + /// Perform bitcode upgrades on llvm.dbg.* calls. + void upgradeDebugIntrinsics(Function &F); +}; +} + +#endif // LLVM_LIB_BITCODE_READER_METADATALOADER_H diff --git a/contrib/libs/llvm14/lib/Bitcode/Reader/ValueList.cpp b/contrib/libs/llvm14/lib/Bitcode/Reader/ValueList.cpp new file mode 100644 index 0000000000..86ed664070 --- /dev/null +++ b/contrib/libs/llvm14/lib/Bitcode/Reader/ValueList.cpp @@ -0,0 +1,216 @@ +//===- ValueList.cpp - Internal BitcodeReader implementation --------------===// +// +// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. +// See https://llvm.org/LICENSE.txt for license information. +// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception +// +//===----------------------------------------------------------------------===// + +#include "ValueList.h" +#include "llvm/ADT/SmallVector.h" +#include "llvm/IR/Argument.h" +#include "llvm/IR/Constant.h" +#include "llvm/IR/Constants.h" +#include "llvm/IR/GlobalValue.h" +#include "llvm/IR/Instruction.h" +#include "llvm/IR/Type.h" +#include "llvm/IR/User.h" +#include "llvm/IR/Value.h" +#include "llvm/Support/Casting.h" +#include "llvm/Support/ErrorHandling.h" +#include <algorithm> +#include <cstddef> +#include <limits> + +using namespace llvm; + +namespace llvm { + +namespace { + +/// A class for maintaining the slot number definition +/// as a placeholder for the actual definition for forward constants defs. +class ConstantPlaceHolder : public ConstantExpr { +public: + explicit ConstantPlaceHolder(Type *Ty, LLVMContext &Context) + : ConstantExpr(Ty, Instruction::UserOp1, &Op<0>(), 1) { + Op<0>() = UndefValue::get(Type::getInt32Ty(Context)); + } + + ConstantPlaceHolder &operator=(const ConstantPlaceHolder &) = delete; + + // allocate space for exactly one operand + void *operator new(size_t s) { return User::operator new(s, 1); } + + /// Methods to support type inquiry through isa, cast, and dyn_cast. + static bool classof(const Value *V) { + return isa<ConstantExpr>(V) && + cast<ConstantExpr>(V)->getOpcode() == Instruction::UserOp1; + } + + /// Provide fast operand accessors + DECLARE_TRANSPARENT_OPERAND_ACCESSORS(Value); +}; + +} // end anonymous namespace + +// FIXME: can we inherit this from ConstantExpr? +template <> +struct OperandTraits<ConstantPlaceHolder> + : public FixedNumOperandTraits<ConstantPlaceHolder, 1> {}; +DEFINE_TRANSPARENT_OPERAND_ACCESSORS(ConstantPlaceHolder, Value) + +} // end namespace llvm + +void BitcodeReaderValueList::assignValue(Value *V, unsigned Idx) { + if (Idx == size()) { + push_back(V); + return; + } + + if (Idx >= size()) + resize(Idx + 1); + + WeakTrackingVH &OldV = ValuePtrs[Idx]; + if (!OldV) { + OldV = V; + return; + } + + // Handle constants and non-constants (e.g. instrs) differently for + // efficiency. + if (Constant *PHC = dyn_cast<Constant>(&*OldV)) { + ResolveConstants.push_back(std::make_pair(PHC, Idx)); + OldV = V; + } else { + // If there was a forward reference to this value, replace it. + Value *PrevVal = OldV; + OldV->replaceAllUsesWith(V); + PrevVal->deleteValue(); + } +} + +Constant *BitcodeReaderValueList::getConstantFwdRef(unsigned Idx, Type *Ty) { + // Bail out for a clearly invalid value. + if (Idx >= RefsUpperBound) + return nullptr; + + if (Idx >= size()) + resize(Idx + 1); + + if (Value *V = ValuePtrs[Idx]) { + if (Ty != V->getType()) + report_fatal_error("Type mismatch in constant table!"); + return cast<Constant>(V); + } + + // Create and return a placeholder, which will later be RAUW'd. + Constant *C = new ConstantPlaceHolder(Ty, Context); + ValuePtrs[Idx] = C; + return C; +} + +Value *BitcodeReaderValueList::getValueFwdRef(unsigned Idx, Type *Ty) { + // Bail out for a clearly invalid value. + if (Idx >= RefsUpperBound) + return nullptr; + + if (Idx >= size()) + resize(Idx + 1); + + if (Value *V = ValuePtrs[Idx]) { + // If the types don't match, it's invalid. + if (Ty && Ty != V->getType()) + return nullptr; + return V; + } + + // No type specified, must be invalid reference. + if (!Ty) + return nullptr; + + // Create and return a placeholder, which will later be RAUW'd. + Value *V = new Argument(Ty); + ValuePtrs[Idx] = V; + return V; +} + +/// Once all constants are read, this method bulk resolves any forward +/// references. The idea behind this is that we sometimes get constants (such +/// as large arrays) which reference *many* forward ref constants. Replacing +/// each of these causes a lot of thrashing when building/reuniquing the +/// constant. Instead of doing this, we look at all the uses and rewrite all +/// the place holders at once for any constant that uses a placeholder. +void BitcodeReaderValueList::resolveConstantForwardRefs() { + // Sort the values by-pointer so that they are efficient to look up with a + // binary search. + llvm::sort(ResolveConstants); + + SmallVector<Constant *, 64> NewOps; + + while (!ResolveConstants.empty()) { + Value *RealVal = operator[](ResolveConstants.back().second); + Constant *Placeholder = ResolveConstants.back().first; + ResolveConstants.pop_back(); + + // Loop over all users of the placeholder, updating them to reference the + // new value. If they reference more than one placeholder, update them all + // at once. + while (!Placeholder->use_empty()) { + auto UI = Placeholder->user_begin(); + User *U = *UI; + + // If the using object isn't uniqued, just update the operands. This + // handles instructions and initializers for global variables. + if (!isa<Constant>(U) || isa<GlobalValue>(U)) { + UI.getUse().set(RealVal); + continue; + } + + // Otherwise, we have a constant that uses the placeholder. Replace that + // constant with a new constant that has *all* placeholder uses updated. + Constant *UserC = cast<Constant>(U); + for (User::op_iterator I = UserC->op_begin(), E = UserC->op_end(); I != E; + ++I) { + Value *NewOp; + if (!isa<ConstantPlaceHolder>(*I)) { + // Not a placeholder reference. + NewOp = *I; + } else if (*I == Placeholder) { + // Common case is that it just references this one placeholder. + NewOp = RealVal; + } else { + // Otherwise, look up the placeholder in ResolveConstants. + ResolveConstantsTy::iterator It = llvm::lower_bound( + ResolveConstants, + std::pair<Constant *, unsigned>(cast<Constant>(*I), 0)); + assert(It != ResolveConstants.end() && It->first == *I); + NewOp = operator[](It->second); + } + + NewOps.push_back(cast<Constant>(NewOp)); + } + + // Make the new constant. + Constant *NewC; + if (ConstantArray *UserCA = dyn_cast<ConstantArray>(UserC)) { + NewC = ConstantArray::get(UserCA->getType(), NewOps); + } else if (ConstantStruct *UserCS = dyn_cast<ConstantStruct>(UserC)) { + NewC = ConstantStruct::get(UserCS->getType(), NewOps); + } else if (isa<ConstantVector>(UserC)) { + NewC = ConstantVector::get(NewOps); + } else { + assert(isa<ConstantExpr>(UserC) && "Must be a ConstantExpr."); + NewC = cast<ConstantExpr>(UserC)->getWithOperands(NewOps); + } + + UserC->replaceAllUsesWith(NewC); + UserC->destroyConstant(); + NewOps.clear(); + } + + // Update all ValueHandles, they should be the only users at this point. + Placeholder->replaceAllUsesWith(RealVal); + delete cast<ConstantPlaceHolder>(Placeholder); + } +} diff --git a/contrib/libs/llvm14/lib/Bitcode/Reader/ValueList.h b/contrib/libs/llvm14/lib/Bitcode/Reader/ValueList.h new file mode 100644 index 0000000000..a39617018f --- /dev/null +++ b/contrib/libs/llvm14/lib/Bitcode/Reader/ValueList.h @@ -0,0 +1,96 @@ +//===-- Bitcode/Reader/ValueList.h - Number values --------------*- C++ -*-===// +// +// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. +// See https://llvm.org/LICENSE.txt for license information. +// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception +// +//===----------------------------------------------------------------------===// +// +// This class gives values and types Unique ID's. +// +//===----------------------------------------------------------------------===// + +#ifndef LLVM_LIB_BITCODE_READER_VALUELIST_H +#define LLVM_LIB_BITCODE_READER_VALUELIST_H + +#include "llvm/IR/ValueHandle.h" +#include <cassert> +#include <utility> +#include <vector> + +namespace llvm { + +class Constant; +class LLVMContext; +class Type; +class Value; + +class BitcodeReaderValueList { + std::vector<WeakTrackingVH> ValuePtrs; + + /// As we resolve forward-referenced constants, we add information about them + /// to this vector. This allows us to resolve them in bulk instead of + /// resolving each reference at a time. See the code in + /// ResolveConstantForwardRefs for more information about this. + /// + /// The key of this vector is the placeholder constant, the value is the slot + /// number that holds the resolved value. + using ResolveConstantsTy = std::vector<std::pair<Constant *, unsigned>>; + ResolveConstantsTy ResolveConstants; + LLVMContext &Context; + + /// Maximum number of valid references. Forward references exceeding the + /// maximum must be invalid. + unsigned RefsUpperBound; + +public: + BitcodeReaderValueList(LLVMContext &C, size_t RefsUpperBound) + : Context(C), + RefsUpperBound(std::min((size_t)std::numeric_limits<unsigned>::max(), + RefsUpperBound)) {} + + ~BitcodeReaderValueList() { + assert(ResolveConstants.empty() && "Constants not resolved?"); + } + + // vector compatibility methods + unsigned size() const { return ValuePtrs.size(); } + void resize(unsigned N) { + ValuePtrs.resize(N); + } + void push_back(Value *V) { ValuePtrs.emplace_back(V); } + + void clear() { + assert(ResolveConstants.empty() && "Constants not resolved?"); + ValuePtrs.clear(); + } + + Value *operator[](unsigned i) const { + assert(i < ValuePtrs.size()); + return ValuePtrs[i]; + } + + Value *back() const { return ValuePtrs.back(); } + void pop_back() { + ValuePtrs.pop_back(); + } + bool empty() const { return ValuePtrs.empty(); } + + void shrinkTo(unsigned N) { + assert(N <= size() && "Invalid shrinkTo request!"); + ValuePtrs.resize(N); + } + + Constant *getConstantFwdRef(unsigned Idx, Type *Ty); + Value *getValueFwdRef(unsigned Idx, Type *Ty); + + void assignValue(Value *V, unsigned Idx); + + /// Once all constants are read, this method bulk resolves any forward + /// references. + void resolveConstantForwardRefs(); +}; + +} // end namespace llvm + +#endif // LLVM_LIB_BITCODE_READER_VALUELIST_H diff --git a/contrib/libs/llvm14/lib/Bitcode/Reader/ya.make b/contrib/libs/llvm14/lib/Bitcode/Reader/ya.make new file mode 100644 index 0000000000..ee811679c8 --- /dev/null +++ b/contrib/libs/llvm14/lib/Bitcode/Reader/ya.make @@ -0,0 +1,33 @@ +# Generated by devtools/yamaker. + +LIBRARY() + +LICENSE(Apache-2.0 WITH LLVM-exception) + +LICENSE_TEXTS(.yandex_meta/licenses.list.txt) + +PEERDIR( + contrib/libs/llvm14 + contrib/libs/llvm14/include + contrib/libs/llvm14/lib/Bitstream/Reader + contrib/libs/llvm14/lib/IR + contrib/libs/llvm14/lib/Support +) + +ADDINCL( + contrib/libs/llvm14/lib/Bitcode/Reader +) + +NO_COMPILER_WARNINGS() + +NO_UTIL() + +SRCS( + BitReader.cpp + BitcodeAnalyzer.cpp + BitcodeReader.cpp + MetadataLoader.cpp + ValueList.cpp +) + +END() |