1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
|
//===- AsmParser.cpp - Parser for Assembly Files --------------------------===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This class implements the parser for assembly files.
//
//===----------------------------------------------------------------------===//
#include "AsmParser.h"
#include "llvm/Support/SourceMgr.h"
#include "llvm/Support/raw_ostream.h"
using namespace llvm;
bool AsmParser::Error(SMLoc L, const char *Msg) {
Lexer.PrintMessage(L, Msg);
return true;
}
bool AsmParser::TokError(const char *Msg) {
Lexer.PrintMessage(Lexer.getLoc(), Msg);
return true;
}
bool AsmParser::Run() {
// Prime the lexer.
Lexer.Lex();
while (Lexer.isNot(asmtok::Eof))
if (ParseStatement())
return true;
return false;
}
/// EatToEndOfStatement - Throw away the rest of the line for testing purposes.
void AsmParser::EatToEndOfStatement() {
while (Lexer.isNot(asmtok::EndOfStatement) &&
Lexer.isNot(asmtok::Eof))
Lexer.Lex();
// Eat EOL.
if (Lexer.is(asmtok::EndOfStatement))
Lexer.Lex();
}
struct AsmParser::X86Operand {
enum {
Register,
Immediate,
Memory
} Kind;
union {
struct {
unsigned RegNo;
} Reg;
struct {
// FIXME: Should be a general expression.
int64_t Val;
} Imm;
struct {
unsigned SegReg;
int64_t Disp; // FIXME: Should be a general expression.
unsigned BaseReg;
unsigned Scale;
unsigned ScaleReg;
} Mem;
};
static X86Operand CreateReg(unsigned RegNo) {
X86Operand Res;
Res.Kind = Register;
Res.Reg.RegNo = RegNo;
return Res;
}
static X86Operand CreateImm(int64_t Val) {
X86Operand Res;
Res.Kind = Immediate;
Res.Imm.Val = Val;
return Res;
}
static X86Operand CreateMem(unsigned SegReg, int64_t Disp, unsigned BaseReg,
unsigned Scale, unsigned ScaleReg) {
X86Operand Res;
Res.Kind = Memory;
Res.Mem.SegReg = SegReg;
Res.Mem.Disp = Disp;
Res.Mem.BaseReg = BaseReg;
Res.Mem.Scale = Scale;
Res.Mem.ScaleReg = ScaleReg;
return Res;
}
};
bool AsmParser::ParseX86Operand(X86Operand &Op) {
switch (Lexer.getKind()) {
default:
return ParseX86MemOperand(Op);
case asmtok::Register:
// FIXME: Decode reg #.
// FIXME: if a segment register, this could either be just the seg reg, or
// the start of a memory operand.
Op = X86Operand::CreateReg(123);
Lexer.Lex(); // Eat register.
return false;
case asmtok::Dollar: {
// $42 -> immediate.
Lexer.Lex();
int64_t Val;
if (ParseExpression(Val))
return TokError("expected integer constant");
Op = X86Operand::CreateReg(Val);
return false;
case asmtok::Star:
Lexer.Lex(); // Eat the star.
if (Lexer.is(asmtok::Register)) {
Op = X86Operand::CreateReg(123);
Lexer.Lex(); // Eat register.
} else if (ParseX86MemOperand(Op))
return true;
// FIXME: Note that these are 'dereferenced' so that clients know the '*' is
// there.
return false;
}
}
}
/// ParseX86MemOperand: segment: disp(basereg, indexreg, scale)
bool AsmParser::ParseX86MemOperand(X86Operand &Op) {
// FIXME: If SegReg ':' (e.g. %gs:), eat and remember.
unsigned SegReg = 0;
// We have to disambiguate a parenthesized expression "(4+5)" from the start
// of a memory operand with a missing displacement "(%ebx)" or "(,%eax)". The
// only way to do this without lookahead is to eat the ( and see what is after
// it.
int64_t Disp = 0;
if (Lexer.isNot(asmtok::LParen)) {
if (ParseExpression(Disp)) return true;
// After parsing the base expression we could either have a parenthesized
// memory address or not. If not, return now. If so, eat the (.
if (Lexer.isNot(asmtok::LParen)) {
Op = X86Operand::CreateMem(SegReg, Disp, 0, 0, 0);
return false;
}
// Eat the '('.
Lexer.Lex();
} else {
// Okay, we have a '('. We don't know if this is an expression or not, but
// so we have to eat the ( to see beyond it.
Lexer.Lex(); // Eat the '('.
if (Lexer.is(asmtok::Register) || Lexer.is(asmtok::Comma)) {
// Nothing to do here, fall into the code below with the '(' part of the
// memory operand consumed.
} else {
// It must be an parenthesized expression, parse it now.
if (ParseParenExpr(Disp)) return true;
// After parsing the base expression we could either have a parenthesized
// memory address or not. If not, return now. If so, eat the (.
if (Lexer.isNot(asmtok::LParen)) {
Op = X86Operand::CreateMem(SegReg, Disp, 0, 0, 0);
return false;
}
// Eat the '('.
Lexer.Lex();
}
}
// If we reached here, then we just ate the ( of the memory operand. Process
// the rest of the memory operand.
unsigned BaseReg = 0, ScaleReg = 0, Scale = 0;
if (Lexer.is(asmtok::Register)) {
BaseReg = 123; // FIXME: decode reg #
Lexer.Lex(); // eat the register.
}
if (Lexer.is(asmtok::Comma)) {
Lexer.Lex(); // eat the comma.
if (Lexer.is(asmtok::Register)) {
ScaleReg = 123; // FIXME: decode reg #
Lexer.Lex(); // eat the register.
Scale = 1; // If not specified, the scale defaults to 1.
}
if (Lexer.is(asmtok::Comma)) {
Lexer.Lex(); // eat the comma.
// If present, get and validate scale amount.
if (Lexer.is(asmtok::IntVal)) {
int64_t ScaleVal = Lexer.getCurIntVal();
if (ScaleVal != 1 && ScaleVal != 2 && ScaleVal != 4 && ScaleVal != 8)
return TokError("scale factor in address must be 1, 2, 4 or 8");
Lexer.Lex(); // eat the scale.
Scale = (unsigned)ScaleVal;
}
}
}
// Ok, we've eaten the memory operand, verify we have a ')' and eat it too.
if (Lexer.isNot(asmtok::RParen))
return TokError("unexpected token in memory operand");
Lexer.Lex(); // Eat the ')'.
Op = X86Operand::CreateMem(SegReg, Disp, BaseReg, Scale, ScaleReg);
return false;
}
/// ParseParenExpr - Parse a paren expression and return it.
/// NOTE: This assumes the leading '(' has already been consumed.
///
/// parenexpr ::= expr)
///
bool AsmParser::ParseParenExpr(int64_t &Res) {
if (ParseExpression(Res)) return true;
if (Lexer.isNot(asmtok::RParen))
return TokError("expected ')' in parentheses expression");
Lexer.Lex();
return false;
}
/// ParsePrimaryExpr - Parse a primary expression and return it.
/// primaryexpr ::= (parenexpr
/// primaryexpr ::= symbol
/// primaryexpr ::= number
/// primaryexpr ::= ~,+,- primaryexpr
bool AsmParser::ParsePrimaryExpr(int64_t &Res) {
switch (Lexer.getKind()) {
default:
return TokError("unknown token in expression");
case asmtok::Identifier:
// This is a label, this should be parsed as part of an expression, to
// handle things like LFOO+4
Res = 0; // FIXME.
Lexer.Lex(); // Eat identifier.
return false;
case asmtok::IntVal:
Res = Lexer.getCurIntVal();
Lexer.Lex(); // Eat identifier.
return false;
case asmtok::LParen:
Lexer.Lex(); // Eat the '('.
return ParseParenExpr(Res);
case asmtok::Tilde:
case asmtok::Plus:
case asmtok::Minus:
Lexer.Lex(); // Eat the operator.
return ParsePrimaryExpr(Res);
}
}
/// ParseExpression - Parse an expression and return it.
///
/// expr ::= expr +,- expr -> lowest.
/// expr ::= expr |,^,&,! expr -> middle.
/// expr ::= expr *,/,%,<<,>> expr -> highest.
/// expr ::= primaryexpr
///
bool AsmParser::ParseExpression(int64_t &Res) {
return ParsePrimaryExpr(Res);
}
/// ParseStatement:
/// ::= EndOfStatement
/// ::= Label* Directive ...Operands... EndOfStatement
/// ::= Label* Identifier OperandList* EndOfStatement
bool AsmParser::ParseStatement() {
switch (Lexer.getKind()) {
default:
return TokError("unexpected token at start of statement");
case asmtok::EndOfStatement:
Lexer.Lex();
return false;
case asmtok::Identifier:
break;
// TODO: Recurse on local labels etc.
}
// If we have an identifier, handle it as the key symbol.
SMLoc IDLoc = Lexer.getLoc();
std::string IDVal = Lexer.getCurStrVal();
// Consume the identifier, see what is after it.
if (Lexer.Lex() == asmtok::Colon) {
// identifier ':' -> Label.
Lexer.Lex();
return ParseStatement();
}
// Otherwise, we have a normal instruction or directive.
if (IDVal[0] == '.') {
Lexer.PrintMessage(IDLoc, "warning: ignoring directive for now");
EatToEndOfStatement();
return false;
}
// If it's an instruction, parse an operand list.
std::vector<X86Operand> Operands;
// Read the first operand, if present. Note that we require a newline at the
// end of file, so we don't have to worry about Eof here.
if (Lexer.isNot(asmtok::EndOfStatement)) {
X86Operand Op;
if (ParseX86Operand(Op))
return true;
Operands.push_back(Op);
}
while (Lexer.is(asmtok::Comma)) {
Lexer.Lex(); // Eat the comma.
// Parse and remember the operand.
X86Operand Op;
if (ParseX86Operand(Op))
return true;
Operands.push_back(Op);
}
if (Lexer.isNot(asmtok::EndOfStatement))
return TokError("unexpected token in operand list");
// Eat the end of statement marker.
Lexer.Lex();
// Instruction is good, process it.
outs() << "Found instruction: " << IDVal << " with " << Operands.size()
<< " operands.\n";
// Skip to end of line for now.
return false;
}
|