1 //===- DebugInfo.h - Debug Information Helpers ------------------*- C++ -*-===//
2 //
3 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
4 // See https://llvm.org/LICENSE.txt for license information.
5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
6 //
7 //===----------------------------------------------------------------------===//
8 //
9 // This file defines a bunch of datatypes that are useful for creating and
10 // walking debug info in LLVM IR form. They essentially provide wrappers around
11 // the information in the global variables that's needed when constructing the
12 // DWARF information.
13 //
14 //===----------------------------------------------------------------------===//
15 
16 #ifndef LLVM_IR_DEBUGINFO_H
17 #define LLVM_IR_DEBUGINFO_H
18 
19 #include "llvm/ADT/STLExtras.h"
20 #include "llvm/ADT/SmallPtrSet.h"
21 #include "llvm/ADT/SmallSet.h"
22 #include "llvm/ADT/SmallVector.h"
23 #include "llvm/ADT/TinyPtrVector.h"
24 #include "llvm/ADT/iterator_range.h"
25 #include "llvm/IR/DataLayout.h"
26 #include "llvm/IR/IntrinsicInst.h"
27 #include "llvm/IR/PassManager.h"
28 #include <optional>
29 
30 namespace llvm {
31 
32 class DbgDeclareInst;
33 class DbgValueInst;
34 class DbgVariableIntrinsic;
35 class Instruction;
36 class Module;
37 
38 /// Finds all intrinsics declaring local variables as living in the memory that
39 /// 'V' points to. This may include a mix of dbg.declare and
40 /// dbg.addr intrinsics.
41 TinyPtrVector<DbgVariableIntrinsic *> FindDbgAddrUses(Value *V);
42 
43 /// Like \c FindDbgAddrUses, but only returns dbg.declare intrinsics, not
44 /// dbg.addr.
45 TinyPtrVector<DbgDeclareInst *> FindDbgDeclareUses(Value *V);
46 
47 /// Finds the llvm.dbg.value intrinsics describing a value.
48 void findDbgValues(SmallVectorImpl<DbgValueInst *> &DbgValues, Value *V);
49 
50 /// Finds the debug info intrinsics describing a value.
51 void findDbgUsers(SmallVectorImpl<DbgVariableIntrinsic *> &DbgInsts, Value *V);
52 
53 /// Find subprogram that is enclosing this scope.
54 DISubprogram *getDISubprogram(const MDNode *Scope);
55 
56 /// Produce a DebugLoc to use for each dbg.declare that is promoted to a
57 /// dbg.value.
58 DebugLoc getDebugValueLoc(DbgVariableIntrinsic *DII);
59 
60 /// Strip debug info in the module if it exists.
61 ///
62 /// To do this, we remove all calls to the debugger intrinsics and any named
63 /// metadata for debugging. We also remove debug locations for instructions.
64 /// Return true if module is modified.
65 bool StripDebugInfo(Module &M);
66 bool stripDebugInfo(Function &F);
67 
68 /// Downgrade the debug info in a module to contain only line table information.
69 ///
70 /// In order to convert debug info to what -gline-tables-only would have
71 /// created, this does the following:
72 ///   1) Delete all debug intrinsics.
73 ///   2) Delete all non-CU named metadata debug info nodes.
74 ///   3) Create new DebugLocs for each instruction.
75 ///   4) Create a new CU debug info, and similarly for every metadata node
76 ///      that's reachable from the CU debug info.
77 ///   All debug type metadata nodes are unreachable and garbage collected.
78 bool stripNonLineTableDebugInfo(Module &M);
79 
80 /// Update the debug locations contained within the MD_loop metadata attached
81 /// to the instruction \p I, if one exists. \p Updater is applied to Metadata
82 /// operand in the MD_loop metadata: the returned value is included in the
83 /// updated loop metadata node if it is non-null.
84 void updateLoopMetadataDebugLocations(
85     Instruction &I, function_ref<Metadata *(Metadata *)> Updater);
86 
87 /// Return Debug Info Metadata Version by checking module flags.
88 unsigned getDebugMetadataVersionFromModule(const Module &M);
89 
90 /// Utility to find all debug info in a module.
91 ///
92 /// DebugInfoFinder tries to list all debug info MDNodes used in a module. To
93 /// list debug info MDNodes used by an instruction, DebugInfoFinder uses
94 /// processDeclare, processValue and processLocation to handle DbgDeclareInst,
95 /// DbgValueInst and DbgLoc attached to instructions. processModule will go
96 /// through all DICompileUnits in llvm.dbg.cu and list debug info MDNodes
97 /// used by the CUs.
98 class DebugInfoFinder {
99 public:
100   /// Process entire module and collect debug info anchors.
101   void processModule(const Module &M);
102   /// Process a single instruction and collect debug info anchors.
103   void processInstruction(const Module &M, const Instruction &I);
104 
105   /// Process DbgVariableIntrinsic.
106   void processVariable(const Module &M, const DbgVariableIntrinsic &DVI);
107   /// Process debug info location.
108   void processLocation(const Module &M, const DILocation *Loc);
109 
110   /// Process subprogram.
111   void processSubprogram(DISubprogram *SP);
112 
113   /// Clear all lists.
114   void reset();
115 
116 private:
117   void processCompileUnit(DICompileUnit *CU);
118   void processScope(DIScope *Scope);
119   void processType(DIType *DT);
120   bool addCompileUnit(DICompileUnit *CU);
121   bool addGlobalVariable(DIGlobalVariableExpression *DIG);
122   bool addScope(DIScope *Scope);
123   bool addSubprogram(DISubprogram *SP);
124   bool addType(DIType *DT);
125 
126 public:
127   using compile_unit_iterator =
128       SmallVectorImpl<DICompileUnit *>::const_iterator;
129   using subprogram_iterator = SmallVectorImpl<DISubprogram *>::const_iterator;
130   using global_variable_expression_iterator =
131       SmallVectorImpl<DIGlobalVariableExpression *>::const_iterator;
132   using type_iterator = SmallVectorImpl<DIType *>::const_iterator;
133   using scope_iterator = SmallVectorImpl<DIScope *>::const_iterator;
134 
135   iterator_range<compile_unit_iterator> compile_units() const {
136     return make_range(CUs.begin(), CUs.end());
137   }
138 
139   iterator_range<subprogram_iterator> subprograms() const {
140     return make_range(SPs.begin(), SPs.end());
141   }
142 
143   iterator_range<global_variable_expression_iterator> global_variables() const {
144     return make_range(GVs.begin(), GVs.end());
145   }
146 
147   iterator_range<type_iterator> types() const {
148     return make_range(TYs.begin(), TYs.end());
149   }
150 
151   iterator_range<scope_iterator> scopes() const {
152     return make_range(Scopes.begin(), Scopes.end());
153   }
154 
155   unsigned compile_unit_count() const { return CUs.size(); }
156   unsigned global_variable_count() const { return GVs.size(); }
157   unsigned subprogram_count() const { return SPs.size(); }
158   unsigned type_count() const { return TYs.size(); }
159   unsigned scope_count() const { return Scopes.size(); }
160 
161 private:
162   SmallVector<DICompileUnit *, 8> CUs;
163   SmallVector<DISubprogram *, 8> SPs;
164   SmallVector<DIGlobalVariableExpression *, 8> GVs;
165   SmallVector<DIType *, 8> TYs;
166   SmallVector<DIScope *, 8> Scopes;
167   SmallPtrSet<const MDNode *, 32> NodesSeen;
168 };
169 
170 /// Assignment Tracking (at).
171 namespace at {
172 //
173 // Utilities for enumerating storing instructions from an assignment ID.
174 //
175 /// A range of instructions.
176 using AssignmentInstRange =
177     iterator_range<SmallVectorImpl<Instruction *>::iterator>;
178 /// Return a range of instructions (typically just one) that have \p ID
179 /// as an attachment.
180 /// Iterators invalidated by adding or removing DIAssignID metadata to/from any
181 /// instruction (including by deleting or cloning instructions).
182 AssignmentInstRange getAssignmentInsts(DIAssignID *ID);
183 /// Return a range of instructions (typically just one) that perform the
184 /// assignment that \p DAI encodes.
185 /// Iterators invalidated by adding or removing DIAssignID metadata to/from any
186 /// instruction (including by deleting or cloning instructions).
187 inline AssignmentInstRange getAssignmentInsts(const DbgAssignIntrinsic *DAI) {
188   return getAssignmentInsts(DAI->getAssignID());
189 }
190 
191 //
192 // Utilities for enumerating llvm.dbg.assign intrinsic from an assignment ID.
193 //
194 /// High level: this is an iterator for llvm.dbg.assign intrinsics.
195 /// Implementation details: this is a wrapper around Value's User iterator that
196 /// dereferences to a DbgAssignIntrinsic ptr rather than a User ptr.
197 class DbgAssignIt
198     : public iterator_adaptor_base<DbgAssignIt, Value::user_iterator,
199                                    typename std::iterator_traits<
200                                        Value::user_iterator>::iterator_category,
201                                    DbgAssignIntrinsic *, std::ptrdiff_t,
202                                    DbgAssignIntrinsic **,
203                                    DbgAssignIntrinsic *&> {
204 public:
205   DbgAssignIt(Value::user_iterator It) : iterator_adaptor_base(It) {}
206   DbgAssignIntrinsic *operator*() const { return cast<DbgAssignIntrinsic>(*I); }
207 };
208 /// A range of llvm.dbg.assign intrinsics.
209 using AssignmentMarkerRange = iterator_range<DbgAssignIt>;
210 /// Return a range of dbg.assign intrinsics which use \ID as an operand.
211 /// Iterators invalidated by deleting an intrinsic contained in this range.
212 AssignmentMarkerRange getAssignmentMarkers(DIAssignID *ID);
213 /// Return a range of dbg.assign intrinsics for which \p Inst performs the
214 /// assignment they encode.
215 /// Iterators invalidated by deleting an intrinsic contained in this range.
216 inline AssignmentMarkerRange getAssignmentMarkers(const Instruction *Inst) {
217   if (auto *ID = Inst->getMetadata(LLVMContext::MD_DIAssignID))
218     return getAssignmentMarkers(cast<DIAssignID>(ID));
219   else
220     return make_range(Value::user_iterator(), Value::user_iterator());
221 }
222 
223 /// Delete the llvm.dbg.assign intrinsics linked to \p Inst.
224 void deleteAssignmentMarkers(const Instruction *Inst);
225 
226 /// Replace all uses (and attachments) of \p Old with \p New.
227 void RAUW(DIAssignID *Old, DIAssignID *New);
228 
229 /// Remove all Assignment Tracking related intrinsics and metadata from \p F.
230 void deleteAll(Function *F);
231 
232 /// Helper struct for trackAssignments, below. We don't use the similar
233 /// DebugVariable class because trackAssignments doesn't (yet?) understand
234 /// partial variables (fragment info) as input and want to make that clear and
235 /// explicit using types. In addition, eventually we will want to understand
236 /// expressions that modify the base address too, which a DebugVariable doesn't
237 /// capture.
238 struct VarRecord {
239   DILocalVariable *Var;
240   DILocation *DL;
241 
242   VarRecord(DbgVariableIntrinsic *DVI)
243       : Var(DVI->getVariable()), DL(getDebugValueLoc(DVI)) {}
244   VarRecord(DILocalVariable *Var, DILocation *DL) : Var(Var), DL(DL) {}
245   friend bool operator<(const VarRecord &LHS, const VarRecord &RHS) {
246     return std::tie(LHS.Var, LHS.DL) < std::tie(RHS.Var, RHS.DL);
247   }
248   friend bool operator==(const VarRecord &LHS, const VarRecord &RHS) {
249     return std::tie(LHS.Var, LHS.DL) == std::tie(RHS.Var, RHS.DL);
250   }
251 };
252 
253 /// Map of backing storage to a set of variables that are stored to it.
254 /// TODO: Backing storage shouldn't be limited to allocas only. Some local
255 /// variables have their storage allocated by the calling function (addresses
256 /// passed in with sret & byval parameters).
257 using StorageToVarsMap = DenseMap<const AllocaInst *, SmallSet<VarRecord, 2>>;
258 
259 /// Track assignments to \p Vars between \p Start and \p End.
260 
261 void trackAssignments(Function::iterator Start, Function::iterator End,
262                       const StorageToVarsMap &Vars, const DataLayout &DL,
263                       bool DebugPrints = false);
264 
265 /// Describes properties of a store that has a static size and offset into a
266 /// some base storage. Used by the getAssignmentInfo functions.
267 struct AssignmentInfo {
268   AllocaInst const *Base;  ///< Base storage.
269   uint64_t OffsetInBits;   ///< Offset into Base.
270   uint64_t SizeInBits;     ///< Number of bits stored.
271   bool StoreToWholeAlloca; ///< SizeInBits equals the size of the base storage.
272 
273   AssignmentInfo(const DataLayout &DL, AllocaInst const *Base,
274                  uint64_t OffsetInBits, uint64_t SizeInBits)
275       : Base(Base), OffsetInBits(OffsetInBits), SizeInBits(SizeInBits),
276         StoreToWholeAlloca(
277             OffsetInBits == 0 &&
278             SizeInBits == DL.getTypeSizeInBits(Base->getAllocatedType())) {}
279 };
280 
281 std::optional<AssignmentInfo> getAssignmentInfo(const DataLayout &DL,
282                                                 const MemIntrinsic *I);
283 std::optional<AssignmentInfo> getAssignmentInfo(const DataLayout &DL,
284                                                 const StoreInst *SI);
285 std::optional<AssignmentInfo> getAssignmentInfo(const DataLayout &DL,
286                                                 const AllocaInst *AI);
287 
288 } // end namespace at
289 
290 /// Convert @llvm.dbg.declare intrinsics into sets of @llvm.dbg.assign
291 /// intrinsics by treating stores to the dbg.declare'd address as assignments
292 /// to the variable. Not all kinds of variables are supported yet; those will
293 /// be left with their dbg.declare intrinsics.
294 /// The pass sets the debug-info-assignment-tracking module flag to true to
295 /// indicate assignment tracking has been enabled.
296 class AssignmentTrackingPass : public PassInfoMixin<AssignmentTrackingPass> {
297   /// Note: this method does not set the debug-info-assignment-tracking module
298   /// flag.
299   void runOnFunction(Function &F);
300 
301 public:
302   PreservedAnalyses run(Function &F, FunctionAnalysisManager &AM);
303   PreservedAnalyses run(Module &M, ModuleAnalysisManager &AM);
304 };
305 
306 /// Return true if assignment tracking is enabled for module \p M.
307 bool isAssignmentTrackingEnabled(const Module &M);
308 } // end namespace llvm
309 
310 #endif // LLVM_IR_DEBUGINFO_H
311