DwarfAccelTable.h revision 36c38b81f0938974c0b1b5fde0b838d51466a94f
1//==-- llvm/CodeGen/DwarfAccelTable.h - Dwarf Accelerator Tables -*- C++ -*-==//
2//
3//                     The LLVM Compiler Infrastructure
4//
5// This file is distributed under the University of Illinois Open Source
6// License. See LICENSE.TXT for details.
7//
8//===----------------------------------------------------------------------===//
9//
10// This file contains support for writing dwarf accelerator tables.
11//
12//===----------------------------------------------------------------------===//
13
14#ifndef CODEGEN_ASMPRINTER_DWARFACCELTABLE_H__
15#define CODEGEN_ASMPRINTER_DWARFACCELTABLE_H__
16
17#include "llvm/ADT/StringMap.h"
18#include "llvm/ADT/ArrayRef.h"
19#include "llvm/MC/MCSymbol.h"
20#include "llvm/Support/Dwarf.h"
21#include "llvm/Support/DataTypes.h"
22#include "llvm/Support/Debug.h"
23#include "llvm/Support/ErrorHandling.h"
24#include "llvm/Support/Format.h"
25#include "llvm/Support/FormattedStream.h"
26#include "DIE.h"
27#include <vector>
28#include <map>
29
30// The dwarf accelerator tables are an indirect hash table optimized
31// for null lookup rather than access to known data. They are output into
32// an on-disk format that looks like this:
33//
34// .-------------.
35// |  HEADER     |
36// |-------------|
37// |  BUCKETS    |
38// |-------------|
39// |  HASHES     |
40// |-------------|
41// |  OFFSETS    |
42// |-------------|
43// |  DATA       |
44// `-------------'
45//
46// where the header contains a magic number, version, type of hash function,
47// the number of buckets, total number of hashes, and room for a special
48// struct of data and the length of that struct.
49//
50// The buckets contain an index (e.g. 6) into the hashes array. The hashes
51// section contains all of the 32-bit hash values in contiguous memory, and
52// the offsets contain the offset into the data area for the particular
53// hash.
54//
55// For a lookup example, we could hash a function name and take it modulo the
56// number of buckets giving us our bucket. From there we take the bucket value
57// as an index into the hashes table and look at each successive hash as long
58// as the hash value is still the same modulo result (bucket value) as earlier.
59// If we have a match we look at that same entry in the offsets table and
60// grab the offset in the data for our final match.
61
62namespace llvm {
63
64class AsmPrinter;
65class DIE;
66class DwarfDebug;
67
68class DwarfAccelTable {
69
70  enum HashFunctionType {
71    eHashFunctionDJB = 0u
72  };
73
74  static uint32_t HashDJB (StringRef Str) {
75    uint32_t h = 5381;
76    for (unsigned i = 0, e = Str.size(); i != e; ++i)
77      h = ((h << 5) + h) + Str[i];
78    return h;
79  }
80
81  // Helper function to compute the number of buckets needed based on
82  // the number of unique hashes.
83  void ComputeBucketCount (void);
84
85  struct TableHeader {
86    uint32_t   magic;           // 'HASH' magic value to allow endian detection
87    uint16_t   version;         // Version number.
88    uint16_t   hash_function;   // The hash function enumeration that was used.
89    uint32_t   bucket_count;    // The number of buckets in this hash table.
90    uint32_t   hashes_count;    // The total number of unique hash values
91                                // and hash data offsets in this table.
92    uint32_t   header_data_len; // The bytes to skip to get to the hash
93                                // indexes (buckets) for correct alignment.
94    // Also written to disk is the implementation specific header data.
95
96    static const uint32_t MagicHash = 0x48415348;
97
98    TableHeader (uint32_t data_len) :
99      magic (MagicHash), version (1), hash_function (eHashFunctionDJB),
100      bucket_count (0), hashes_count (0), header_data_len (data_len)
101    {}
102
103#ifndef NDEBUG
104    void print(raw_ostream &O) {
105      O << "Magic: " << format("0x%x", magic) << "\n"
106        << "Version: " << version << "\n"
107        << "Hash Function: " << hash_function << "\n"
108        << "Bucket Count: " << bucket_count << "\n"
109        << "Header Data Length: " << header_data_len << "\n";
110    }
111    void dump() { print(dbgs()); }
112#endif
113  };
114
115public:
116  // The HeaderData describes the form of each set of data. In general this
117  // is as a list of atoms (atom_count) where each atom contains a type
118  // (AtomType type) of data, and an encoding form (form). In the case of
119  // data that is referenced via DW_FORM_ref_* the die_offset_base is
120  // used to describe the offset for all forms in the list of atoms.
121  // This also serves as a public interface of sorts.
122  // When written to disk this will have the form:
123  //
124  // uint32_t die_offset_base
125  // uint32_t atom_count
126  // atom_count Atoms
127  enum AtomType {
128    eAtomTypeNULL       = 0u,
129    eAtomTypeDIEOffset  = 1u,   // DIE offset, check form for encoding
130    eAtomTypeCUOffset   = 2u,   // DIE offset of the compiler unit header that
131                                // contains the item in question
132    eAtomTypeTag        = 3u,   // DW_TAG_xxx value, should be encoded as
133                                // DW_FORM_data1 (if no tags exceed 255) or
134                                // DW_FORM_data2.
135    eAtomTypeNameFlags  = 4u,   // Flags from enum NameFlags
136    eAtomTypeTypeFlags  = 5u    // Flags from enum TypeFlags
137  };
138
139  enum TypeFlags {
140    eTypeFlagClassMask = 0x0000000fu,
141
142    // Always set for C++, only set for ObjC if this is the
143    // @implementation for a class.
144    eTypeFlagClassIsImplementation  = ( 1u << 1 )
145  };
146
147  // Make these public so that they can be used as a general interface to
148  // the class.
149  struct Atom {
150    AtomType type; // enum AtomType
151    uint16_t form; // DWARF DW_FORM_ defines
152
153    Atom(AtomType type, uint16_t form) : type(type), form(form) {}
154    static const char * AtomTypeString(enum AtomType);
155#ifndef NDEBUG
156    void print(raw_ostream &O) {
157      O << "Type: " << AtomTypeString(type) << "\n"
158        << "Form: " << dwarf::FormEncodingString(form) << "\n";
159    }
160    void dump() {
161      print(dbgs());
162    }
163#endif
164  };
165
166 private:
167  struct TableHeaderData {
168    uint32_t die_offset_base;
169    SmallVector<Atom, 1> Atoms;
170
171    TableHeaderData(ArrayRef<Atom> AtomList, uint32_t offset = 0)
172      : die_offset_base(offset), Atoms(AtomList.begin(), AtomList.end()) { }
173
174#ifndef NDEBUG
175    void print (raw_ostream &O) {
176      O << "die_offset_base: " << die_offset_base << "\n";
177      for (size_t i = 0; i < Atoms.size(); i++)
178        Atoms[i].print(O);
179    }
180    void dump() {
181      print(dbgs());
182    }
183#endif
184  };
185
186  // The data itself consists of a str_offset, a count of the DIEs in the
187  // hash and the offsets to the DIEs themselves.
188  // On disk each data section is ended with a 0 KeyType as the end of the
189  // hash chain.
190  // On output this looks like:
191  // uint32_t str_offset
192  // uint32_t hash_data_count
193  // HashData[hash_data_count]
194public:
195  struct HashDataContents {
196    DIE *Die; // Offsets
197    char Flags; // Specific flags to output
198
199    HashDataContents(DIE *D, char Flags) :
200      Die(D),
201      Flags(Flags) { }
202    #ifndef NDEBUG
203    void print(raw_ostream &O) const {
204      O << "  Offset: " << Die->getOffset() << "\n";
205      O << "  Tag: " << dwarf::TagString(Die->getTag()) << "\n";
206      O << "  Flags: " << Flags << "\n";
207    }
208    #endif
209  };
210private:
211  struct HashData {
212    StringRef Str;
213    uint32_t HashValue;
214    MCSymbol *Sym;
215    ArrayRef<HashDataContents*> Data; // offsets
216    HashData(StringRef S, ArrayRef<HashDataContents*> Data)
217      : Str(S), Data(Data) {
218      HashValue = DwarfAccelTable::HashDJB(S);
219    }
220    #ifndef NDEBUG
221    void print(raw_ostream &O) {
222      O << "Name: " << Str << "\n";
223      O << "  Hash Value: " << format("0x%x", HashValue) << "\n";
224      O << "  Symbol: " ;
225      if (Sym) Sym->print(O);
226      else O << "<none>";
227      O << "\n";
228      for (size_t i = 0; i < Data.size(); i++) {
229        O << "  Offset: " << Data[i]->Die->getOffset() << "\n";
230        O << "  Tag: " << dwarf::TagString(Data[i]->Die->getTag()) << "\n";
231        O << "  Flags: " << Data[i]->Flags << "\n";
232      }
233    }
234    void dump() {
235      print(dbgs());
236    }
237    #endif
238  };
239
240  DwarfAccelTable(const DwarfAccelTable&); // DO NOT IMPLEMENT
241  void operator=(const DwarfAccelTable&);  // DO NOT IMPLEMENT
242
243  // Internal Functions
244  void EmitHeader(AsmPrinter *);
245  void EmitBuckets(AsmPrinter *);
246  void EmitHashes(AsmPrinter *);
247  void EmitOffsets(AsmPrinter *, MCSymbol *);
248  void EmitData(AsmPrinter *, DwarfDebug *D);
249
250  // Allocator for HashData and HashDataContents.
251  BumpPtrAllocator Allocator;
252
253  // Output Variables
254  TableHeader Header;
255  TableHeaderData HeaderData;
256  std::vector<HashData*> Data;
257
258  // String Data
259  typedef std::vector<HashDataContents*> DataArray;
260  typedef StringMap<DataArray, BumpPtrAllocator&> StringEntries;
261  StringEntries Entries;
262
263  // Buckets/Hashes/Offsets
264  typedef std::vector<HashData*> HashList;
265  typedef std::vector<HashList> BucketList;
266  BucketList Buckets;
267  HashList Hashes;
268
269  // Public Implementation
270 public:
271  DwarfAccelTable(ArrayRef<DwarfAccelTable::Atom>);
272  ~DwarfAccelTable();
273  void AddName(StringRef, DIE*, char = 0);
274  void FinalizeTable(AsmPrinter *, const char *);
275  void Emit(AsmPrinter *, MCSymbol *, DwarfDebug *);
276#ifndef NDEBUG
277  void print(raw_ostream &O);
278  void dump() { print(dbgs()); }
279#endif
280};
281
282}
283#endif
284