1 //===- llvm/Bitcode/BitcodeAnalyzer.h - Bitcode analyzer --------*- 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 header defines interfaces to analyze LLVM bitcode files/streams.
10 //
11 //===----------------------------------------------------------------------===//
12 
13 #ifndef LLVM_BITCODE_BITCODEANALYZER_H
14 #define LLVM_BITCODE_BITCODEANALYZER_H
15 
16 #include "llvm/ADT/ArrayRef.h"
17 #include "llvm/ADT/Optional.h"
18 #include "llvm/ADT/StringRef.h"
19 #include "llvm/Bitstream/BitstreamReader.h"
20 #include "llvm/Support/Error.h"
21 #include "llvm/Support/raw_ostream.h"
22 #include <map>
23 #include <vector>
24 
25 namespace llvm {
26 
27 /// CurStreamTypeType - A type for CurStreamType
28 enum CurStreamTypeType {
29   UnknownBitstream,
30   LLVMIRBitstream,
31   ClangSerializedASTBitstream,
32   ClangSerializedDiagnosticsBitstream,
33   LLVMBitstreamRemarks
34 };
35 
36 struct BCDumpOptions {
37   /// The stream.
38   raw_ostream &OS;
39   /// Print per-code histogram.
40   bool Histogram = false;
41   /// Don't emit numeric info in dump if symbolic info is available.
42   bool Symbolic = false;
43   /// Print binary blobs using hex escapes.
44   bool ShowBinaryBlobs = false;
45   /// Print BLOCKINFO block details.
46   bool DumpBlockinfo = false;
47 
48   BCDumpOptions(raw_ostream &OS) : OS(OS) {}
49 };
50 
51 class BitcodeAnalyzer {
52   BitstreamCursor Stream;
53   BitstreamBlockInfo BlockInfo;
54   CurStreamTypeType CurStreamType;
55   Optional<BitstreamCursor> BlockInfoStream;
56   unsigned NumTopBlocks = 0;
57 
58   struct PerRecordStats {
59     unsigned NumInstances;
60     unsigned NumAbbrev;
61     uint64_t TotalBits;
62     PerRecordStats() : NumInstances(0), NumAbbrev(0), TotalBits(0) {}
63   };
64 
65   struct PerBlockIDStats {
66     /// NumInstances - This the number of times this block ID has been seen.
67     unsigned NumInstances;
68     /// NumBits - The total size in bits of all of these blocks.
69     uint64_t NumBits;
70     /// NumSubBlocks - The total number of blocks these blocks contain.
71     unsigned NumSubBlocks;
72     /// NumAbbrevs - The total number of abbreviations.
73     unsigned NumAbbrevs;
74     /// NumRecords - The total number of records these blocks contain, and the
75     /// number that are abbreviated.
76     unsigned NumRecords, NumAbbreviatedRecords;
77     /// CodeFreq - Keep track of the number of times we see each code.
78     std::vector<PerRecordStats> CodeFreq;
79     PerBlockIDStats()
80         : NumInstances(0), NumBits(0), NumSubBlocks(0), NumAbbrevs(0),
81           NumRecords(0), NumAbbreviatedRecords(0) {}
82   };
83 
84   std::map<unsigned, PerBlockIDStats> BlockIDStats;
85 
86 public:
87   BitcodeAnalyzer(StringRef Buffer, Optional<StringRef> BlockInfoBuffer = None);
88   /// Analyze the bitcode file.
89   Error analyze(Optional<BCDumpOptions> O = None,
90                 Optional<StringRef> CheckHash = None);
91   /// Print stats about the bitcode file.
92   void printStats(BCDumpOptions O, Optional<StringRef> Filename = None);
93 
94 private:
95   /// Read a block, updating statistics, etc.
96   Error parseBlock(unsigned BlockID, unsigned IndentLevel,
97                    Optional<BCDumpOptions> O = None,
98                    Optional<StringRef> CheckHash = None);
99 
100   Error decodeMetadataStringsBlob(StringRef Indent, ArrayRef<uint64_t> Record,
101                                   StringRef Blob, raw_ostream &OS);
102 };
103 
104 } // end namespace llvm
105 
106 #endif // LLVM_BITCODE_BITCODEANALYZER_H
107