xref: /llvm-project/llvm/include/llvm/IR/StructuralHash.h (revision 4f41862c5a5241654a37ee994ed0074a815d3633)
1 //===- llvm/IR/StructuralHash.h - IR Hashing --------------------*- 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 provides hashing of the LLVM IR structure to be used to check
10 // Passes modification status.
11 //
12 //===----------------------------------------------------------------------===//
13 
14 #ifndef LLVM_IR_STRUCTURALHASH_H
15 #define LLVM_IR_STRUCTURALHASH_H
16 
17 #include "llvm/ADT/MapVector.h"
18 #include "llvm/ADT/StableHashing.h"
19 #include "llvm/IR/Instruction.h"
20 #include <cstdint>
21 
22 namespace llvm {
23 
24 class Function;
25 class Module;
26 
27 /// Returns a hash of the function \p F.
28 /// \param F The function to hash.
29 /// \param DetailedHash Whether or not to encode additional information in the
30 /// hash. The additional information added into the hash when this flag is set
31 /// to true includes instruction and operand type information.
32 stable_hash StructuralHash(const Function &F, bool DetailedHash = false);
33 
34 /// Returns a hash of the global variable \p G.
35 stable_hash StructuralHash(const GlobalVariable &G);
36 
37 /// Returns a hash of the module \p M by hashing all functions and global
38 /// variables contained within. \param M The module to hash. \param DetailedHash
39 /// Whether or not to encode additional information in the function hashes that
40 /// composed the module hash.
41 stable_hash StructuralHash(const Module &M, bool DetailedHash = false);
42 
43 /// The pair of an instruction index and a operand index.
44 using IndexPair = std::pair<unsigned, unsigned>;
45 
46 /// A map from an instruction index to an instruction pointer.
47 using IndexInstrMap = MapVector<unsigned, Instruction *>;
48 
49 /// A map from an IndexPair to a stable hash.
50 using IndexOperandHashMapType = DenseMap<IndexPair, stable_hash>;
51 
52 /// A function that takes an instruction and an operand index and returns true
53 /// if the operand should be ignored in the function hash computation.
54 using IgnoreOperandFunc = std::function<bool(const Instruction *, unsigned)>;
55 
56 struct FunctionHashInfo {
57   /// A hash value representing the structural content of the function
58   stable_hash FunctionHash;
59   /// A mapping from instruction indices to instruction pointers
60   std::unique_ptr<IndexInstrMap> IndexInstruction;
61   /// A mapping from pairs of instruction indices and operand indices
62   /// to the hashes of the operands. This can be used to analyze or
63   /// reconstruct the differences in ignored operands
64   std::unique_ptr<IndexOperandHashMapType> IndexOperandHashMap;
65 
66   FunctionHashInfo(stable_hash FuntionHash,
67                    std::unique_ptr<IndexInstrMap> IndexInstruction,
68                    std::unique_ptr<IndexOperandHashMapType> IndexOperandHashMap)
69       : FunctionHash(FuntionHash),
70         IndexInstruction(std::move(IndexInstruction)),
71         IndexOperandHashMap(std::move(IndexOperandHashMap)) {}
72 };
73 
74 /// Computes a structural hash of a given function, considering the structure
75 /// and content of the function's instructions while allowing for selective
76 /// ignoring of certain operands based on custom criteria. This hash can be used
77 /// to identify functions that are structurally similar or identical, which is
78 /// useful in optimizations, deduplication, or analysis tasks.
79 /// \param F The function to hash.
80 /// \param IgnoreOp A callable that takes an instruction and an operand index,
81 /// and returns true if the operand should be ignored in the hash computation.
82 /// \return A FunctionHashInfo structure
83 FunctionHashInfo StructuralHashWithDifferences(const Function &F,
84                                                IgnoreOperandFunc IgnoreOp);
85 
86 } // end namespace llvm
87 
88 #endif
89