1 //===- ArchiveReader.cpp - Code to read LLVM bytecode from .a files -------===//
3 // The LLVM Compiler Infrastructure
5 // This file was developed by the LLVM research group and is distributed under
6 // the University of Illinois Open Source License. See LICENSE.TXT for details.
8 //===----------------------------------------------------------------------===//
10 // This file implements the ReadArchiveFile interface, which allows a linker to
11 // read all of the LLVM bytecode files contained in a .a file. This file
12 // understands the standard system .a file format. This can only handle the .a
13 // variant prevalent on Linux systems so far, but may be extended. See
14 // information in this source file for more information:
15 // http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/bfd/archive.c?cvsroot=src
17 //===----------------------------------------------------------------------===//
19 #include "llvm/Bytecode/Reader.h"
20 #include "llvm/Module.h"
21 #include "Support/FileUtilities.h"
33 char fmag[2]; // Always equal to '`\n'
37 UserObject, // A user .o/.bc file
38 Unknown, // Unknown file, just ignore it
39 SVR4LongFilename, // a "//" section used for long file names
40 ArchiveSymbolTable, // Symbol table produced by ranlib.
44 /// getObjectType - Determine the type of object that this header represents.
45 /// This is capable of parsing the variety of special sections used for various
48 static enum ObjectType getObjectType(ar_hdr *H, std::string MemberName,
49 unsigned char *MemberData, unsigned Size) {
50 // Check for sections with special names...
51 if (MemberName == "__.SYMDEF " || MemberName == "__.SYMDEF SORTED")
52 return ArchiveSymbolTable;
53 else if (MemberName == "// ")
54 return SVR4LongFilename;
56 // Check to see if it looks like an llvm object file...
57 if (Size >= 4 && !memcmp(MemberData, "llvm", 4))
63 static inline bool Error(std::string *ErrorStr, const char *Message) {
64 if (ErrorStr) *ErrorStr = Message;
68 static bool ParseSymbolTableSection(unsigned char *Buffer, unsigned Size,
70 // Currently not supported (succeeds without doing anything)
74 static bool ReadArchiveBuffer(const std::string &ArchiveName,
75 unsigned char *Buffer, unsigned Length,
76 std::vector<Module*> &Objects,
77 std::string *ErrorStr) {
78 if (Length < 8 || memcmp(Buffer, "!<arch>\n", 8))
79 return Error(ErrorStr, "signature incorrect for an archive file!");
80 Buffer += 8; Length -= 8; // Skip the magic string.
82 std::vector<char> LongFilenames;
84 while (Length >= sizeof(ar_hdr)) {
85 ar_hdr *Hdr = (ar_hdr*)Buffer;
86 unsigned SizeFromHeader = atoi(Hdr->size);
87 if (SizeFromHeader + sizeof(ar_hdr) > Length)
88 return Error(ErrorStr, "invalid record length in archive file!");
90 unsigned char *MemberData = Buffer + sizeof(ar_hdr);
91 unsigned MemberSize = SizeFromHeader;
92 // Get name of archive member.
93 char *startp = Hdr->name;
94 char *endp = (char *) memchr (startp, '/', sizeof(ar_hdr));
95 if (memcmp (Hdr->name, "#1/", 3) == 0) {
96 // 4.4BSD/MacOSX long filenames are abbreviated as "#1/L", where L is an
97 // ASCII-coded decimal number representing the length of the name buffer,
98 // which is prepended to the archive member's contents.
99 unsigned NameLength = atoi (&Hdr->name[3]);
100 startp = (char *) MemberData;
101 endp = startp + NameLength;
102 MemberData += NameLength;
103 MemberSize -= NameLength;
104 } else if (startp == endp && isdigit (Hdr->name[1])) {
105 // SVR4 long filenames are abbreviated as "/I", where I is
106 // an ASCII-coded decimal index into the LongFilenames vector.
107 unsigned NameIndex = atoi (&Hdr->name[1]);
108 assert (LongFilenames.size () > NameIndex
109 && "SVR4-style long filename for archive member not found");
110 startp = &LongFilenames[NameIndex];
111 endp = strchr (startp, '/');
112 } else if (startp == endp && Hdr->name[1] == '/') {
113 // This is for the SVR4 long filename table (there might be other
114 // names starting with // but I don't know about them). Make sure that
115 // getObjectType sees it.
116 endp = &Hdr->name[sizeof (Hdr->name)];
119 // 4.4BSD/MacOSX *short* filenames are not guaranteed to have a
120 // terminator. Start at the end of the field and backtrack over spaces.
121 endp = startp + sizeof(Hdr->name);
122 while (endp[-1] == ' ')
125 std::string MemberName (startp, endp);
126 std::string FullMemberName = ArchiveName + "(" + MemberName + ")";
128 switch (getObjectType(Hdr, MemberName, MemberData, MemberSize)) {
129 case SVR4LongFilename:
130 // If this is a long filename section, read all of the file names into the
131 // LongFilenames vector.
132 LongFilenames.assign (MemberData, MemberData + MemberSize);
135 Module *M = ParseBytecodeBuffer(MemberData, MemberSize,
136 FullMemberName, ErrorStr);
138 Objects.push_back(M);
141 case ArchiveSymbolTable:
142 if (ParseSymbolTableSection(MemberData, MemberSize, ErrorStr))
146 std::cerr << "ReadArchiveBuffer: WARNING: Skipping unknown file: "
147 << FullMemberName << "\n";
148 break; // Just ignore unknown files.
151 // Round SizeFromHeader up to an even number...
152 SizeFromHeader = (SizeFromHeader+1)/2*2;
153 Buffer += sizeof(ar_hdr)+SizeFromHeader; // Move to the next entry
154 Length -= sizeof(ar_hdr)+SizeFromHeader;
161 // ReadArchiveFile - Read bytecode files from the specified .a file, returning
162 // true on error, or false on success. This does not support reading files from
165 bool llvm::ReadArchiveFile(const std::string &Filename,
166 std::vector<Module*> &Objects,std::string *ErrorStr){
169 // mmap in the file all at once...
170 unsigned char *Buffer =
171 (unsigned char*)ReadFileIntoAddressSpace(Filename, Length);
173 if (ErrorStr) *ErrorStr = "Error reading file '" + Filename + "'!";
177 // Parse the archive files we mmap'ped in
178 bool Result = ReadArchiveBuffer(Filename, Buffer, Length, Objects, ErrorStr);
180 // Unmmap the archive...
181 UnmapFileFromAddressSpace(Buffer, Length);
183 if (Result) // Free any loaded objects
184 while (!Objects.empty()) {
185 delete Objects.back();