diff options
author | Alexey Lapshin <a.v.lapshin@mail.ru> | 2019-12-08 01:37:53 +0300 |
---|---|---|
committer | Alexey Lapshin <a.v.lapshin@mail.ru> | 2019-12-19 15:41:48 +0300 |
commit | abc7f6800df8a1f40e1e2c9ccce826abb0208284 (patch) | |
tree | 79a527c3eef37db531326b979889fe5ec0294ff1 /llvm/lib/DebugInfo/DWARF | |
parent | a59cc5e128f09ec5048c142cafaadea279406eba (diff) | |
download | bcm5719-llvm-abc7f6800df8a1f40e1e2c9ccce826abb0208284.tar.gz bcm5719-llvm-abc7f6800df8a1f40e1e2c9ccce826abb0208284.zip |
[Dsymutil][Debuginfo][NFC] Refactor dsymutil to separate DWARF optimizing part 2.
That patch is extracted from the D70709. It moves CompileUnit, DeclContext
into llvm/DebugInfo/DWARF. It also adds new file DWARFOptimizer with
AddressesMap class. AddressesMap generalizes functionality
from RelocationManager.
Differential Revision: https://reviews.llvm.org/D71271
Diffstat (limited to 'llvm/lib/DebugInfo/DWARF')
-rw-r--r-- | llvm/lib/DebugInfo/DWARF/CMakeLists.txt | 3 | ||||
-rw-r--r-- | llvm/lib/DebugInfo/DWARF/DWARFOptCompileUnit.cpp | 144 | ||||
-rw-r--r-- | llvm/lib/DebugInfo/DWARF/DWARFOptDeclContext.cpp | 209 | ||||
-rw-r--r-- | llvm/lib/DebugInfo/DWARF/DWARFOptimizer.cpp | 15 | ||||
-rw-r--r-- | llvm/lib/DebugInfo/DWARF/LLVMBuild.txt | 2 |
5 files changed, 372 insertions, 1 deletions
diff --git a/llvm/lib/DebugInfo/DWARF/CMakeLists.txt b/llvm/lib/DebugInfo/DWARF/CMakeLists.txt index 3fe9904f221..d51fe265363 100644 --- a/llvm/lib/DebugInfo/DWARF/CMakeLists.txt +++ b/llvm/lib/DebugInfo/DWARF/CMakeLists.txt @@ -27,6 +27,9 @@ add_llvm_component_library(LLVMDebugInfoDWARF DWARFUnitIndex.cpp DWARFUnit.cpp DWARFVerifier.cpp + DWARFOptCompileUnit.cpp + DWARFOptDeclContext.cpp + DWARFOptimizer.cpp ADDITIONAL_HEADER_DIRS ${LLVM_MAIN_INCLUDE_DIR}/llvm/DebugInfo/DWARF diff --git a/llvm/lib/DebugInfo/DWARF/DWARFOptCompileUnit.cpp b/llvm/lib/DebugInfo/DWARF/DWARFOptCompileUnit.cpp new file mode 100644 index 00000000000..6ca5f128928 --- /dev/null +++ b/llvm/lib/DebugInfo/DWARF/DWARFOptCompileUnit.cpp @@ -0,0 +1,144 @@ +//===- llvm/DebugInfo/DWARF/DWARFOptCompileUnit.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/DebugInfo/DWARF/DWARFOptCompileUnit.h" +#include "llvm/DebugInfo/DWARF/DWARFOptDeclContext.h" + +namespace llvm { + +/// Check if the DIE at \p Idx is in the scope of a function. +static bool inFunctionScope(CompileUnit &U, unsigned Idx) { + while (Idx) { + if (U.getOrigUnit().getDIEAtIndex(Idx).getTag() == dwarf::DW_TAG_subprogram) + return true; + Idx = U.getInfo(Idx).ParentIdx; + } + return false; +} + +uint16_t CompileUnit::getLanguage() { + if (!Language) { + DWARFDie CU = getOrigUnit().getUnitDIE(); + Language = dwarf::toUnsigned(CU.find(dwarf::DW_AT_language), 0); + } + return Language; +} + +void CompileUnit::markEverythingAsKept() { + unsigned Idx = 0; + + setHasInterestingContent(); + + for (auto &I : Info) { + // Mark everything that wasn't explicit marked for pruning. + I.Keep = !I.Prune; + auto DIE = OrigUnit.getDIEAtIndex(Idx++); + + // Try to guess which DIEs must go to the accelerator tables. We do that + // just for variables, because functions will be handled depending on + // whether they carry a DW_AT_low_pc attribute or not. + if (DIE.getTag() != dwarf::DW_TAG_variable && + DIE.getTag() != dwarf::DW_TAG_constant) + continue; + + Optional<DWARFFormValue> Value; + if (!(Value = DIE.find(dwarf::DW_AT_location))) { + if ((Value = DIE.find(dwarf::DW_AT_const_value)) && + !inFunctionScope(*this, I.ParentIdx)) + I.InDebugMap = true; + continue; + } + if (auto Block = Value->getAsBlock()) { + if (Block->size() > OrigUnit.getAddressByteSize() && + (*Block)[0] == dwarf::DW_OP_addr) + I.InDebugMap = true; + } + } +} + +uint64_t CompileUnit::computeNextUnitOffset() { + NextUnitOffset = StartOffset; + if (NewUnit) { + NextUnitOffset += 11 /* Header size */; + NextUnitOffset += NewUnit->getUnitDie().getSize(); + } + return NextUnitOffset; +} + +/// Keep track of a forward cross-cu reference from this unit +/// to \p Die that lives in \p RefUnit. +void CompileUnit::noteForwardReference(DIE *Die, const CompileUnit *RefUnit, + DeclContext *Ctxt, PatchLocation Attr) { + ForwardDIEReferences.emplace_back(Die, RefUnit, Ctxt, Attr); +} + +void CompileUnit::fixupForwardReferences() { + for (const auto &Ref : ForwardDIEReferences) { + DIE *RefDie; + const CompileUnit *RefUnit; + PatchLocation Attr; + DeclContext *Ctxt; + std::tie(RefDie, RefUnit, Ctxt, Attr) = Ref; + if (Ctxt && Ctxt->getCanonicalDIEOffset()) + Attr.set(Ctxt->getCanonicalDIEOffset()); + else + Attr.set(RefDie->getOffset() + RefUnit->getStartOffset()); + } +} + +void CompileUnit::addLabelLowPc(uint64_t LabelLowPc, int64_t PcOffset) { + Labels.insert({LabelLowPc, PcOffset}); +} + +void CompileUnit::addFunctionRange(uint64_t FuncLowPc, uint64_t FuncHighPc, + int64_t PcOffset) { + // Don't add empty ranges to the interval map. They are a problem because + // the interval map expects half open intervals. This is safe because they + // are empty anyway. + if (FuncHighPc != FuncLowPc) + Ranges.insert(FuncLowPc, FuncHighPc, PcOffset); + this->LowPc = std::min(LowPc, FuncLowPc + PcOffset); + this->HighPc = std::max(HighPc, FuncHighPc + PcOffset); +} + +void CompileUnit::noteRangeAttribute(const DIE &Die, PatchLocation Attr) { + if (Die.getTag() != dwarf::DW_TAG_compile_unit) + RangeAttributes.push_back(Attr); + else + UnitRangeAttribute = Attr; +} + +void CompileUnit::noteLocationAttribute(PatchLocation Attr, int64_t PcOffset) { + LocationAttributes.emplace_back(Attr, PcOffset); +} + +void CompileUnit::addNamespaceAccelerator(const DIE *Die, + DwarfStringPoolEntryRef Name) { + Namespaces.emplace_back(Name, Die); +} + +void CompileUnit::addObjCAccelerator(const DIE *Die, + DwarfStringPoolEntryRef Name, + bool SkipPubSection) { + ObjC.emplace_back(Name, Die, SkipPubSection); +} + +void CompileUnit::addNameAccelerator(const DIE *Die, + DwarfStringPoolEntryRef Name, + bool SkipPubSection) { + Pubnames.emplace_back(Name, Die, SkipPubSection); +} + +void CompileUnit::addTypeAccelerator(const DIE *Die, + DwarfStringPoolEntryRef Name, + bool ObjcClassImplementation, + uint32_t QualifiedNameHash) { + Pubtypes.emplace_back(Name, Die, QualifiedNameHash, ObjcClassImplementation); +} + +} // namespace llvm diff --git a/llvm/lib/DebugInfo/DWARF/DWARFOptDeclContext.cpp b/llvm/lib/DebugInfo/DWARF/DWARFOptDeclContext.cpp new file mode 100644 index 00000000000..95cf58444e1 --- /dev/null +++ b/llvm/lib/DebugInfo/DWARF/DWARFOptDeclContext.cpp @@ -0,0 +1,209 @@ +//===- llvm/DebugInfo/DWARF/DWARFOptDeclContext.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/DebugInfo/DWARF/DWARFOptDeclContext.h" +#include "llvm/DebugInfo/DWARF/DWARFContext.h" +#include "llvm/DebugInfo/DWARF/DWARFDie.h" +#include "llvm/DebugInfo/DWARF/DWARFUnit.h" + +namespace llvm { + +/// Set the last DIE/CU a context was seen in and, possibly invalidate the +/// context if it is ambiguous. +/// +/// In the current implementation, we don't handle overloaded functions well, +/// because the argument types are not taken into account when computing the +/// DeclContext tree. +/// +/// Some of this is mitigated byt using mangled names that do contain the +/// arguments types, but sometimes (e.g. with function templates) we don't have +/// that. In that case, just do not unique anything that refers to the contexts +/// we are not able to distinguish. +/// +/// If a context that is not a namespace appears twice in the same CU, we know +/// it is ambiguous. Make it invalid. +bool DeclContext::setLastSeenDIE(CompileUnit &U, const DWARFDie &Die) { + if (LastSeenCompileUnitID == U.getUniqueID()) { + DWARFUnit &OrigUnit = U.getOrigUnit(); + uint32_t FirstIdx = OrigUnit.getDIEIndex(LastSeenDIE); + U.getInfo(FirstIdx).Ctxt = nullptr; + return false; + } + + LastSeenCompileUnitID = U.getUniqueID(); + LastSeenDIE = Die; + return true; +} + +PointerIntPair<DeclContext *, 1> DeclContextTree::getChildDeclContext( + DeclContext &Context, const DWARFDie &DIE, CompileUnit &U, + UniquingStringPool &StringPool, bool InClangModule) { + unsigned Tag = DIE.getTag(); + + // FIXME: dsymutil-classic compat: We should bail out here if we + // have a specification or an abstract_origin. We will get the + // parent context wrong here. + + switch (Tag) { + default: + // By default stop gathering child contexts. + return PointerIntPair<DeclContext *, 1>(nullptr); + case dwarf::DW_TAG_module: + break; + case dwarf::DW_TAG_compile_unit: + return PointerIntPair<DeclContext *, 1>(&Context); + case dwarf::DW_TAG_subprogram: + // Do not unique anything inside CU local functions. + if ((Context.getTag() == dwarf::DW_TAG_namespace || + Context.getTag() == dwarf::DW_TAG_compile_unit) && + !dwarf::toUnsigned(DIE.find(dwarf::DW_AT_external), 0)) + return PointerIntPair<DeclContext *, 1>(nullptr); + LLVM_FALLTHROUGH; + case dwarf::DW_TAG_member: + case dwarf::DW_TAG_namespace: + case dwarf::DW_TAG_structure_type: + case dwarf::DW_TAG_class_type: + case dwarf::DW_TAG_union_type: + case dwarf::DW_TAG_enumeration_type: + case dwarf::DW_TAG_typedef: + // Artificial things might be ambiguous, because they might be created on + // demand. For example implicitly defined constructors are ambiguous + // because of the way we identify contexts, and they won't be generated + // every time everywhere. + if (dwarf::toUnsigned(DIE.find(dwarf::DW_AT_artificial), 0)) + return PointerIntPair<DeclContext *, 1>(nullptr); + break; + } + + const char *Name = DIE.getName(DINameKind::LinkageName); + const char *ShortName = DIE.getName(DINameKind::ShortName); + StringRef NameRef; + StringRef ShortNameRef; + StringRef FileRef; + + if (Name) + NameRef = StringPool.internString(Name); + else if (Tag == dwarf::DW_TAG_namespace) + // FIXME: For dsymutil-classic compatibility. I think uniquing within + // anonymous namespaces is wrong. There is no ODR guarantee there. + NameRef = StringPool.internString("(anonymous namespace)"); + + if (ShortName && ShortName != Name) + ShortNameRef = StringPool.internString(ShortName); + else + ShortNameRef = NameRef; + + if (Tag != dwarf::DW_TAG_class_type && Tag != dwarf::DW_TAG_structure_type && + Tag != dwarf::DW_TAG_union_type && + Tag != dwarf::DW_TAG_enumeration_type && NameRef.empty()) + return PointerIntPair<DeclContext *, 1>(nullptr); + + unsigned Line = 0; + unsigned ByteSize = std::numeric_limits<uint32_t>::max(); + + if (!InClangModule) { + // Gather some discriminating data about the DeclContext we will be + // creating: File, line number and byte size. This shouldn't be necessary, + // because the ODR is just about names, but given that we do some + // approximations with overloaded functions and anonymous namespaces, use + // these additional data points to make the process safer. + // + // This is disabled for clang modules, because forward declarations of + // module-defined types do not have a file and line. + ByteSize = dwarf::toUnsigned(DIE.find(dwarf::DW_AT_byte_size), + std::numeric_limits<uint64_t>::max()); + if (Tag != dwarf::DW_TAG_namespace || !Name) { + if (unsigned FileNum = + dwarf::toUnsigned(DIE.find(dwarf::DW_AT_decl_file), 0)) { + if (const auto *LT = U.getOrigUnit().getContext().getLineTableForUnit( + &U.getOrigUnit())) { + // FIXME: dsymutil-classic compatibility. I'd rather not + // unique anything in anonymous namespaces, but if we do, then + // verify that the file and line correspond. + if (!Name && Tag == dwarf::DW_TAG_namespace) + FileNum = 1; + + if (LT->hasFileAtIndex(FileNum)) { + Line = dwarf::toUnsigned(DIE.find(dwarf::DW_AT_decl_line), 0); + // Cache the resolved paths based on the index in the line table, + // because calling realpath is expansive. + StringRef ResolvedPath = U.getResolvedPath(FileNum); + if (!ResolvedPath.empty()) { + FileRef = ResolvedPath; + } else { + std::string File; + bool FoundFileName = LT->getFileNameByIndex( + FileNum, U.getOrigUnit().getCompilationDir(), + DILineInfoSpecifier::FileLineInfoKind::AbsoluteFilePath, + File); + (void)FoundFileName; + assert(FoundFileName && "Must get file name from line table"); + // Second level of caching, this time based on the file's parent + // path. + FileRef = PathResolver.resolve(File, StringPool); + U.setResolvedPath(FileNum, FileRef); + } + } + } + } + } + } + + if (!Line && NameRef.empty()) + return PointerIntPair<DeclContext *, 1>(nullptr); + + // We hash NameRef, which is the mangled name, in order to get most + // overloaded functions resolve correctly. + // + // Strictly speaking, hashing the Tag is only necessary for a + // DW_TAG_module, to prevent uniquing of a module and a namespace + // with the same name. + // + // FIXME: dsymutil-classic won't unique the same type presented + // once as a struct and once as a class. Using the Tag in the fully + // qualified name hash to get the same effect. + unsigned Hash = hash_combine(Context.getQualifiedNameHash(), Tag, NameRef); + + // FIXME: dsymutil-classic compatibility: when we don't have a name, + // use the filename. + if (Tag == dwarf::DW_TAG_namespace && NameRef == "(anonymous namespace)") + Hash = hash_combine(Hash, FileRef); + + // Now look if this context already exists. + DeclContext Key(Hash, Line, ByteSize, Tag, NameRef, FileRef, Context); + auto ContextIter = Contexts.find(&Key); + + if (ContextIter == Contexts.end()) { + // The context wasn't found. + bool Inserted; + DeclContext *NewContext = + new (Allocator) DeclContext(Hash, Line, ByteSize, Tag, NameRef, FileRef, + Context, DIE, U.getUniqueID()); + std::tie(ContextIter, Inserted) = Contexts.insert(NewContext); + assert(Inserted && "Failed to insert DeclContext"); + (void)Inserted; + } else if (Tag != dwarf::DW_TAG_namespace && + !(*ContextIter)->setLastSeenDIE(U, DIE)) { + // The context was found, but it is ambiguous with another context + // in the same file. Mark it invalid. + return PointerIntPair<DeclContext *, 1>(*ContextIter, /* Invalid= */ 1); + } + + assert(ContextIter != Contexts.end()); + // FIXME: dsymutil-classic compatibility. Union types aren't + // uniques, but their children might be. + if ((Tag == dwarf::DW_TAG_subprogram && + Context.getTag() != dwarf::DW_TAG_structure_type && + Context.getTag() != dwarf::DW_TAG_class_type) || + (Tag == dwarf::DW_TAG_union_type)) + return PointerIntPair<DeclContext *, 1>(*ContextIter, /* Invalid= */ 1); + + return PointerIntPair<DeclContext *, 1>(*ContextIter); +} + +} // namespace llvm diff --git a/llvm/lib/DebugInfo/DWARF/DWARFOptimizer.cpp b/llvm/lib/DebugInfo/DWARF/DWARFOptimizer.cpp new file mode 100644 index 00000000000..535d2a04707 --- /dev/null +++ b/llvm/lib/DebugInfo/DWARF/DWARFOptimizer.cpp @@ -0,0 +1,15 @@ +//=== llvm/DebugInfo/DWARF/DWARFOptimizer.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/DebugInfo/DWARF/DWARFOptimizer.h" + +namespace llvm { + +AddressesMap::~AddressesMap() {} + +} // namespace llvm diff --git a/llvm/lib/DebugInfo/DWARF/LLVMBuild.txt b/llvm/lib/DebugInfo/DWARF/LLVMBuild.txt index 182be10f0b4..e72360deaf8 100644 --- a/llvm/lib/DebugInfo/DWARF/LLVMBuild.txt +++ b/llvm/lib/DebugInfo/DWARF/LLVMBuild.txt @@ -18,4 +18,4 @@ type = Library name = DebugInfoDWARF parent = DebugInfo -required_libraries = BinaryFormat Object MC Support +required_libraries = BinaryFormat Object MC Support CodeGen |