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
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
|
// script.cc -- handle linker scripts for gold.
// Copyright 2006, 2007, 2008 Free Software Foundation, Inc.
// Written by Ian Lance Taylor <iant@google.com>.
// This file is part of gold.
// This program is free software; you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation; either version 3 of the License, or
// (at your option) any later version.
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU General Public License for more details.
// You should have received a copy of the GNU General Public License
// along with this program; if not, write to the Free Software
// Foundation, Inc., 51 Franklin Street - Fifth Floor, Boston,
// MA 02110-1301, USA.
#include "gold.h"
#include <fnmatch.h>
#include <string>
#include <vector>
#include <cstdio>
#include <cstdlib>
#include "filenames.h"
#include "elfcpp.h"
#include "demangle.h"
#include "dirsearch.h"
#include "options.h"
#include "fileread.h"
#include "workqueue.h"
#include "readsyms.h"
#include "parameters.h"
#include "layout.h"
#include "symtab.h"
#include "script.h"
#include "script-c.h"
namespace gold
{
// A token read from a script file. We don't implement keywords here;
// all keywords are simply represented as a string.
class Token
{
public:
// Token classification.
enum Classification
{
// Token is invalid.
TOKEN_INVALID,
// Token indicates end of input.
TOKEN_EOF,
// Token is a string of characters.
TOKEN_STRING,
// Token is a quoted string of characters.
TOKEN_QUOTED_STRING,
// Token is an operator.
TOKEN_OPERATOR,
// Token is a number (an integer).
TOKEN_INTEGER
};
// We need an empty constructor so that we can put this STL objects.
Token()
: classification_(TOKEN_INVALID), value_(NULL), value_length_(0),
opcode_(0), lineno_(0), charpos_(0)
{ }
// A general token with no value.
Token(Classification classification, int lineno, int charpos)
: classification_(classification), value_(NULL), value_length_(0),
opcode_(0), lineno_(lineno), charpos_(charpos)
{
gold_assert(classification == TOKEN_INVALID
|| classification == TOKEN_EOF);
}
// A general token with a value.
Token(Classification classification, const char* value, size_t length,
int lineno, int charpos)
: classification_(classification), value_(value), value_length_(length),
opcode_(0), lineno_(lineno), charpos_(charpos)
{
gold_assert(classification != TOKEN_INVALID
&& classification != TOKEN_EOF);
}
// A token representing an operator.
Token(int opcode, int lineno, int charpos)
: classification_(TOKEN_OPERATOR), value_(NULL), value_length_(0),
opcode_(opcode), lineno_(lineno), charpos_(charpos)
{ }
// Return whether the token is invalid.
bool
is_invalid() const
{ return this->classification_ == TOKEN_INVALID; }
// Return whether this is an EOF token.
bool
is_eof() const
{ return this->classification_ == TOKEN_EOF; }
// Return the token classification.
Classification
classification() const
{ return this->classification_; }
// Return the line number at which the token starts.
int
lineno() const
{ return this->lineno_; }
// Return the character position at this the token starts.
int
charpos() const
{ return this->charpos_; }
// Get the value of a token.
const char*
string_value(size_t* length) const
{
gold_assert(this->classification_ == TOKEN_STRING
|| this->classification_ == TOKEN_QUOTED_STRING);
*length = this->value_length_;
return this->value_;
}
int
operator_value() const
{
gold_assert(this->classification_ == TOKEN_OPERATOR);
return this->opcode_;
}
uint64_t
integer_value() const
{
gold_assert(this->classification_ == TOKEN_INTEGER);
// Null terminate.
std::string s(this->value_, this->value_length_);
return strtoull(s.c_str(), NULL, 0);
}
private:
// The token classification.
Classification classification_;
// The token value, for TOKEN_STRING or TOKEN_QUOTED_STRING or
// TOKEN_INTEGER.
const char* value_;
// The length of the token value.
size_t value_length_;
// The token value, for TOKEN_OPERATOR.
int opcode_;
// The line number where this token started (one based).
int lineno_;
// The character position within the line where this token started
// (one based).
int charpos_;
};
// This class handles lexing a file into a sequence of tokens.
class Lex
{
public:
// We unfortunately have to support different lexing modes, because
// when reading different parts of a linker script we need to parse
// things differently.
enum Mode
{
// Reading an ordinary linker script.
LINKER_SCRIPT,
// Reading an expression in a linker script.
EXPRESSION,
// Reading a version script.
VERSION_SCRIPT
};
Lex(const char* input_string, size_t input_length, int parsing_token)
: input_string_(input_string), input_length_(input_length),
current_(input_string), mode_(LINKER_SCRIPT),
first_token_(parsing_token), token_(),
lineno_(1), linestart_(input_string)
{ }
// Read a file into a string.
static void
read_file(Input_file*, std::string*);
// Return the next token.
const Token*
next_token();
// Return the current lexing mode.
Lex::Mode
mode() const
{ return this->mode_; }
// Set the lexing mode.
void
set_mode(Mode mode)
{ this->mode_ = mode; }
private:
Lex(const Lex&);
Lex& operator=(const Lex&);
// Make a general token with no value at the current location.
Token
make_token(Token::Classification c, const char* start) const
{ return Token(c, this->lineno_, start - this->linestart_ + 1); }
// Make a general token with a value at the current location.
Token
make_token(Token::Classification c, const char* v, size_t len,
const char* start)
const
{ return Token(c, v, len, this->lineno_, start - this->linestart_ + 1); }
// Make an operator token at the current location.
Token
make_token(int opcode, const char* start) const
{ return Token(opcode, this->lineno_, start - this->linestart_ + 1); }
// Make an invalid token at the current location.
Token
make_invalid_token(const char* start)
{ return this->make_token(Token::TOKEN_INVALID, start); }
// Make an EOF token at the current location.
Token
make_eof_token(const char* start)
{ return this->make_token(Token::TOKEN_EOF, start); }
// Return whether C can be the first character in a name. C2 is the
// next character, since we sometimes need that.
inline bool
can_start_name(char c, char c2);
// If C can appear in a name which has already started, return a
// pointer to a character later in the token or just past
// it. Otherwise, return NULL.
inline const char*
can_continue_name(const char* c);
// Return whether C, C2, C3 can start a hex number.
inline bool
can_start_hex(char c, char c2, char c3);
// If C can appear in a hex number which has already started, return
// a pointer to a character later in the token or just past
// it. Otherwise, return NULL.
inline const char*
can_continue_hex(const char* c);
// Return whether C can start a non-hex number.
static inline bool
can_start_number(char c);
// If C can appear in a decimal number which has already started,
// return a pointer to a character later in the token or just past
// it. Otherwise, return NULL.
inline const char*
can_continue_number(const char* c)
{ return Lex::can_start_number(*c) ? c + 1 : NULL; }
// If C1 C2 C3 form a valid three character operator, return the
// opcode. Otherwise return 0.
static inline int
three_char_operator(char c1, char c2, char c3);
// If C1 C2 form a valid two character operator, return the opcode.
// Otherwise return 0.
static inline int
two_char_operator(char c1, char c2);
// If C1 is a valid one character operator, return the opcode.
// Otherwise return 0.
static inline int
one_char_operator(char c1);
// Read the next token.
Token
get_token(const char**);
// Skip a C style /* */ comment. Return false if the comment did
// not end.
bool
skip_c_comment(const char**);
// Skip a line # comment. Return false if there was no newline.
bool
skip_line_comment(const char**);
// Build a token CLASSIFICATION from all characters that match
// CAN_CONTINUE_FN. The token starts at START. Start matching from
// MATCH. Set *PP to the character following the token.
inline Token
gather_token(Token::Classification,
const char* (Lex::*can_continue_fn)(const char*),
const char* start, const char* match, const char** pp);
// Build a token from a quoted string.
Token
gather_quoted_string(const char** pp);
// The string we are tokenizing.
const char* input_string_;
// The length of the string.
size_t input_length_;
// The current offset into the string.
const char* current_;
// The current lexing mode.
Mode mode_;
// The code to use for the first token. This is set to 0 after it
// is used.
int first_token_;
// The current token.
Token token_;
// The current line number.
int lineno_;
// The start of the current line in the string.
const char* linestart_;
};
// Read the whole file into memory. We don't expect linker scripts to
// be large, so we just use a std::string as a buffer. We ignore the
// data we've already read, so that we read aligned buffers.
void
Lex::read_file(Input_file* input_file, std::string* contents)
{
off_t filesize = input_file->file().filesize();
contents->clear();
contents->reserve(filesize);
off_t off = 0;
unsigned char buf[BUFSIZ];
while (off < filesize)
{
off_t get = BUFSIZ;
if (get > filesize - off)
get = filesize - off;
input_file->file().read(off, get, buf);
contents->append(reinterpret_cast<char*>(&buf[0]), get);
off += get;
}
}
// Return whether C can be the start of a name, if the next character
// is C2. A name can being with a letter, underscore, period, or
// dollar sign. Because a name can be a file name, we also permit
// forward slash, backslash, and tilde. Tilde is the tricky case
// here; GNU ld also uses it as a bitwise not operator. It is only
// recognized as the operator if it is not immediately followed by
// some character which can appear in a symbol. That is, when we
// don't know that we are looking at an expression, "~0" is a file
// name, and "~ 0" is an expression using bitwise not. We are
// compatible.
inline bool
Lex::can_start_name(char c, char c2)
{
switch (c)
{
case 'A': case 'B': case 'C': case 'D': case 'E': case 'F':
case 'G': case 'H': case 'I': case 'J': case 'K': case 'L':
case 'M': case 'N': case 'O': case 'Q': case 'P': case 'R':
case 'S': case 'T': case 'U': case 'V': case 'W': case 'X':
case 'Y': case 'Z':
case 'a': case 'b': case 'c': case 'd': case 'e': case 'f':
case 'g': case 'h': case 'i': case 'j': case 'k': case 'l':
case 'm': case 'n': case 'o': case 'q': case 'p': case 'r':
case 's': case 't': case 'u': case 'v': case 'w': case 'x':
case 'y': case 'z':
case '_': case '.': case '$':
return true;
case '/': case '\\':
return this->mode_ == LINKER_SCRIPT;
case '~':
return this->mode_ == LINKER_SCRIPT && can_continue_name(&c2);
case '*': case '[':
return this->mode_ == VERSION_SCRIPT;
default:
return false;
}
}
// Return whether C can continue a name which has already started.
// Subsequent characters in a name are the same as the leading
// characters, plus digits and "=+-:[],?*". So in general the linker
// script language requires spaces around operators, unless we know
// that we are parsing an expression.
inline const char*
Lex::can_continue_name(const char* c)
{
switch (*c)
{
case 'A': case 'B': case 'C': case 'D': case 'E': case 'F':
case 'G': case 'H': case 'I': case 'J': case 'K': case 'L':
case 'M': case 'N': case 'O': case 'Q': case 'P': case 'R':
case 'S': case 'T': case 'U': case 'V': case 'W': case 'X':
case 'Y': case 'Z':
case 'a': case 'b': case 'c': case 'd': case 'e': case 'f':
case 'g': case 'h': case 'i': case 'j': case 'k': case 'l':
case 'm': case 'n': case 'o': case 'q': case 'p': case 'r':
case 's': case 't': case 'u': case 'v': case 'w': case 'x':
case 'y': case 'z':
case '_': case '.': case '$':
case '0': case '1': case '2': case '3': case '4':
case '5': case '6': case '7': case '8': case '9':
return c + 1;
case '/': case '\\': case '~':
case '=': case '+':
case ',': case '?':
if (this->mode_ == LINKER_SCRIPT)
return c + 1;
return NULL;
case '[': case ']': case '*': case '-':
if (this->mode_ == LINKER_SCRIPT || this->mode_ == VERSION_SCRIPT)
return c + 1;
return NULL;
case '^':
if (this->mode_ == VERSION_SCRIPT)
return c + 1;
return NULL;
case ':':
if (this->mode_ == LINKER_SCRIPT)
return c + 1;
else if (this->mode_ == VERSION_SCRIPT && (c[1] == ':'))
{
// A name can have '::' in it, as that's a c++ namespace
// separator. But a single colon is not part of a name.
return c + 2;
}
return NULL;
default:
return NULL;
}
}
// For a number we accept 0x followed by hex digits, or any sequence
// of digits. The old linker accepts leading '$' for hex, and
// trailing HXBOD. Those are for MRI compatibility and we don't
// accept them. The old linker also accepts trailing MK for mega or
// kilo. FIXME: Those are mentioned in the documentation, and we
// should accept them.
// Return whether C1 C2 C3 can start a hex number.
inline bool
Lex::can_start_hex(char c1, char c2, char c3)
{
if (c1 == '0' && (c2 == 'x' || c2 == 'X'))
return this->can_continue_hex(&c3);
return false;
}
// Return whether C can appear in a hex number.
inline const char*
Lex::can_continue_hex(const char* c)
{
switch (*c)
{
case '0': case '1': case '2': case '3': case '4':
case '5': case '6': case '7': case '8': case '9':
case 'A': case 'B': case 'C': case 'D': case 'E': case 'F':
case 'a': case 'b': case 'c': case 'd': case 'e': case 'f':
return c + 1;
default:
return NULL;
}
}
// Return whether C can start a non-hex number.
inline bool
Lex::can_start_number(char c)
{
switch (c)
{
case '0': case '1': case '2': case '3': case '4':
case '5': case '6': case '7': case '8': case '9':
return true;
default:
return false;
}
}
// If C1 C2 C3 form a valid three character operator, return the
// opcode (defined in the yyscript.h file generated from yyscript.y).
// Otherwise return 0.
inline int
Lex::three_char_operator(char c1, char c2, char c3)
{
switch (c1)
{
case '<':
if (c2 == '<' && c3 == '=')
return LSHIFTEQ;
break;
case '>':
if (c2 == '>' && c3 == '=')
return RSHIFTEQ;
break;
default:
break;
}
return 0;
}
// If C1 C2 form a valid two character operator, return the opcode
// (defined in the yyscript.h file generated from yyscript.y).
// Otherwise return 0.
inline int
Lex::two_char_operator(char c1, char c2)
{
switch (c1)
{
case '=':
if (c2 == '=')
return EQ;
break;
case '!':
if (c2 == '=')
return NE;
break;
case '+':
if (c2 == '=')
return PLUSEQ;
break;
case '-':
if (c2 == '=')
return MINUSEQ;
break;
case '*':
if (c2 == '=')
return MULTEQ;
break;
case '/':
if (c2 == '=')
return DIVEQ;
break;
case '|':
if (c2 == '=')
return OREQ;
if (c2 == '|')
return OROR;
break;
case '&':
if (c2 == '=')
return ANDEQ;
if (c2 == '&')
return ANDAND;
break;
case '>':
if (c2 == '=')
return GE;
if (c2 == '>')
return RSHIFT;
break;
case '<':
if (c2 == '=')
return LE;
if (c2 == '<')
return LSHIFT;
break;
default:
break;
}
return 0;
}
// If C1 is a valid operator, return the opcode. Otherwise return 0.
inline int
Lex::one_char_operator(char c1)
{
switch (c1)
{
case '+':
case '-':
case '*':
case '/':
case '%':
case '!':
case '&':
case '|':
case '^':
case '~':
case '<':
case '>':
case '=':
case '?':
case ',':
case '(':
case ')':
case '{':
case '}':
case '[':
case ']':
case ':':
case ';':
return c1;
default:
return 0;
}
}
// Skip a C style comment. *PP points to just after the "/*". Return
// false if the comment did not end.
bool
Lex::skip_c_comment(const char** pp)
{
const char* p = *pp;
while (p[0] != '*' || p[1] != '/')
{
if (*p == '\0')
{
*pp = p;
return false;
}
if (*p == '\n')
{
++this->lineno_;
this->linestart_ = p + 1;
}
++p;
}
*pp = p + 2;
return true;
}
// Skip a line # comment. Return false if there was no newline.
bool
Lex::skip_line_comment(const char** pp)
{
const char* p = *pp;
size_t skip = strcspn(p, "\n");
if (p[skip] == '\0')
{
*pp = p + skip;
return false;
}
p += skip + 1;
++this->lineno_;
this->linestart_ = p;
*pp = p;
return true;
}
// Build a token CLASSIFICATION from all characters that match
// CAN_CONTINUE_FN. Update *PP.
inline Token
Lex::gather_token(Token::Classification classification,
const char* (Lex::*can_continue_fn)(const char*),
const char* start,
const char* match,
const char **pp)
{
const char* new_match = NULL;
while ((new_match = (this->*can_continue_fn)(match)))
match = new_match;
*pp = match;
return this->make_token(classification, start, match - start, start);
}
// Build a token from a quoted string.
Token
Lex::gather_quoted_string(const char** pp)
{
const char* start = *pp;
const char* p = start;
++p;
size_t skip = strcspn(p, "\"\n");
if (p[skip] != '"')
return this->make_invalid_token(start);
*pp = p + skip + 1;
return this->make_token(Token::TOKEN_QUOTED_STRING, p, skip, start);
}
// Return the next token at *PP. Update *PP. General guideline: we
// require linker scripts to be simple ASCII. No unicode linker
// scripts. In particular we can assume that any '\0' is the end of
// the input.
Token
Lex::get_token(const char** pp)
{
const char* p = *pp;
while (true)
{
if (*p == '\0')
{
*pp = p;
return this->make_eof_token(p);
}
// Skip whitespace quickly.
while (*p == ' ' || *p == '\t')
++p;
if (*p == '\n')
{
++p;
++this->lineno_;
this->linestart_ = p;
continue;
}
// Skip C style comments.
if (p[0] == '/' && p[1] == '*')
{
int lineno = this->lineno_;
int charpos = p - this->linestart_ + 1;
*pp = p + 2;
if (!this->skip_c_comment(pp))
return Token(Token::TOKEN_INVALID, lineno, charpos);
p = *pp;
continue;
}
// Skip line comments.
if (*p == '#')
{
*pp = p + 1;
if (!this->skip_line_comment(pp))
return this->make_eof_token(p);
p = *pp;
continue;
}
// Check for a name.
if (this->can_start_name(p[0], p[1]))
return this->gather_token(Token::TOKEN_STRING,
&Lex::can_continue_name,
p, p + 1, pp);
// We accept any arbitrary name in double quotes, as long as it
// does not cross a line boundary.
if (*p == '"')
{
*pp = p;
return this->gather_quoted_string(pp);
}
// Check for a number.
if (this->can_start_hex(p[0], p[1], p[2]))
return this->gather_token(Token::TOKEN_INTEGER,
&Lex::can_continue_hex,
p, p + 3, pp);
if (Lex::can_start_number(p[0]))
return this->gather_token(Token::TOKEN_INTEGER,
&Lex::can_continue_number,
p, p + 1, pp);
// Check for operators.
int opcode = Lex::three_char_operator(p[0], p[1], p[2]);
if (opcode != 0)
{
*pp = p + 3;
return this->make_token(opcode, p);
}
opcode = Lex::two_char_operator(p[0], p[1]);
if (opcode != 0)
{
*pp = p + 2;
return this->make_token(opcode, p);
}
opcode = Lex::one_char_operator(p[0]);
if (opcode != 0)
{
*pp = p + 1;
return this->make_token(opcode, p);
}
return this->make_token(Token::TOKEN_INVALID, p);
}
}
// Return the next token.
const Token*
Lex::next_token()
{
// The first token is special.
if (this->first_token_ != 0)
{
this->token_ = Token(this->first_token_, 0, 0);
this->first_token_ = 0;
return &this->token_;
}
this->token_ = this->get_token(&this->current_);
// Don't let an early null byte fool us into thinking that we've
// reached the end of the file.
if (this->token_.is_eof()
&& (static_cast<size_t>(this->current_ - this->input_string_)
< this->input_length_))
this->token_ = this->make_invalid_token(this->current_);
return &this->token_;
}
// A trivial task which waits for THIS_BLOCKER to be clear and then
// clears NEXT_BLOCKER. THIS_BLOCKER may be NULL.
class Script_unblock : public Task
{
public:
Script_unblock(Task_token* this_blocker, Task_token* next_blocker)
: this_blocker_(this_blocker), next_blocker_(next_blocker)
{ }
~Script_unblock()
{
if (this->this_blocker_ != NULL)
delete this->this_blocker_;
}
Task_token*
is_runnable()
{
if (this->this_blocker_ != NULL && this->this_blocker_->is_blocked())
return this->this_blocker_;
return NULL;
}
void
locks(Task_locker* tl)
{ tl->add(this, this->next_blocker_); }
void
run(Workqueue*)
{ }
std::string
get_name() const
{ return "Script_unblock"; }
private:
Task_token* this_blocker_;
Task_token* next_blocker_;
};
// Class Script_options.
Script_options::Script_options()
: entry_(), symbol_assignments_()
{
}
// Add any symbols we are defining to the symbol table.
void
Script_options::add_symbols_to_table(Symbol_table* symtab,
const Target* target)
{
for (Symbol_assignments::iterator p = this->symbol_assignments_.begin();
p != this->symbol_assignments_.end();
++p)
{
elfcpp::STV vis = p->hidden ? elfcpp::STV_HIDDEN : elfcpp::STV_DEFAULT;
p->sym = symtab->define_as_constant(target,
p->name.c_str(),
NULL, // version
0, // value
0, // size
elfcpp::STT_NOTYPE,
elfcpp::STB_GLOBAL,
vis,
0, // nonvis
p->provide);
}
}
// Finalize symbol values.
void
Script_options::finalize_symbols(Symbol_table* symtab, const Layout* layout)
{
if (parameters->get_size() == 32)
{
#if defined(HAVE_TARGET_32_LITTLE) || defined(HAVE_TARGET_32_BIG)
this->sized_finalize_symbols<32>(symtab, layout);
#else
gold_unreachable();
#endif
}
else if (parameters->get_size() == 64)
{
#if defined(HAVE_TARGET_64_LITTLE) || defined(HAVE_TARGET_64_BIG)
this->sized_finalize_symbols<64>(symtab, layout);
#else
gold_unreachable();
#endif
}
else
gold_unreachable();
}
template<int size>
void
Script_options::sized_finalize_symbols(Symbol_table* symtab,
const Layout* layout)
{
for (Symbol_assignments::iterator p = this->symbol_assignments_.begin();
p != this->symbol_assignments_.end();
++p)
{
if (p->sym != NULL)
{
Sized_symbol<size>* ssym = symtab->get_sized_symbol<size>(p->sym);
ssym->set_value(p->value->eval(symtab, layout));
}
}
}
// This class holds data passed through the parser to the lexer and to
// the parser support functions. This avoids global variables. We
// can't use global variables because we need not be called by a
// singleton thread.
class Parser_closure
{
public:
Parser_closure(const char* filename,
const Position_dependent_options& posdep_options,
bool in_group, bool is_in_sysroot,
Command_line* command_line,
Script_options* script_options,
Lex* lex)
: filename_(filename), posdep_options_(posdep_options),
in_group_(in_group), is_in_sysroot_(is_in_sysroot),
command_line_(command_line), script_options_(script_options),
version_script_info_(script_options->version_script_info()),
lex_(lex), lineno_(0), charpos_(0), lex_mode_stack_(), inputs_(NULL)
{
// We start out processing C symbols in the default lex mode.
language_stack_.push_back("");
lex_mode_stack_.push_back(lex->mode());
}
// Return the file name.
const char*
filename() const
{ return this->filename_; }
// Return the position dependent options. The caller may modify
// this.
Position_dependent_options&
position_dependent_options()
{ return this->posdep_options_; }
// Return whether this script is being run in a group.
bool
in_group() const
{ return this->in_group_; }
// Return whether this script was found using a directory in the
// sysroot.
bool
is_in_sysroot() const
{ return this->is_in_sysroot_; }
// Returns the Command_line structure passed in at constructor time.
// This value may be NULL. The caller may modify this, which modifies
// the passed-in Command_line object (not a copy).
Command_line*
command_line()
{ return this->command_line_; }
// Return the options which may be set by a script.
Script_options*
script_options()
{ return this->script_options_; }
// Return the object in which version script information should be stored.
Version_script_info*
version_script()
{ return this->version_script_info_; }
// Return the next token, and advance.
const Token*
next_token()
{
const Token* token = this->lex_->next_token();
this->lineno_ = token->lineno();
this->charpos_ = token->charpos();
return token;
}
// Set a new lexer mode, pushing the current one.
void
push_lex_mode(Lex::Mode mode)
{
this->lex_mode_stack_.push_back(this->lex_->mode());
this->lex_->set_mode(mode);
}
// Pop the lexer mode.
void
pop_lex_mode()
{
gold_assert(!this->lex_mode_stack_.empty());
this->lex_->set_mode(this->lex_mode_stack_.back());
this->lex_mode_stack_.pop_back();
}
// Return the current lexer mode.
Lex::Mode
lex_mode() const
{ return this->lex_mode_stack_.back(); }
// Return the line number of the last token.
int
lineno() const
{ return this->lineno_; }
// Return the character position in the line of the last token.
int
charpos() const
{ return this->charpos_; }
// Return the list of input files, creating it if necessary. This
// is a space leak--we never free the INPUTS_ pointer.
Input_arguments*
inputs()
{
if (this->inputs_ == NULL)
this->inputs_ = new Input_arguments();
return this->inputs_;
}
// Return whether we saw any input files.
bool
saw_inputs() const
{ return this->inputs_ != NULL && !this->inputs_->empty(); }
// Return the current language being processed in a version script
// (eg, "C++"). The empty string represents unmangled C names.
const std::string&
get_current_language() const
{ return this->language_stack_.back(); }
// Push a language onto the stack when entering an extern block.
void push_language(const std::string& lang)
{ this->language_stack_.push_back(lang); }
// Pop a language off of the stack when exiting an extern block.
void pop_language()
{
gold_assert(!this->language_stack_.empty());
this->language_stack_.pop_back();
}
private:
// The name of the file we are reading.
const char* filename_;
// The position dependent options.
Position_dependent_options posdep_options_;
// Whether we are currently in a --start-group/--end-group.
bool in_group_;
// Whether the script was found in a sysrooted directory.
bool is_in_sysroot_;
// May be NULL if the user chooses not to pass one in.
Command_line* command_line_;
// Options which may be set from any linker script.
Script_options* script_options_;
// Information parsed from a version script.
Version_script_info* version_script_info_;
// The lexer.
Lex* lex_;
// The line number of the last token returned by next_token.
int lineno_;
// The column number of the last token returned by next_token.
int charpos_;
// A stack of lexer modes.
std::vector<Lex::Mode> lex_mode_stack_;
// A stack of which extern/language block we're inside. Can be C++,
// java, or empty for C.
std::vector<std::string> language_stack_;
// New input files found to add to the link.
Input_arguments* inputs_;
};
// FILE was found as an argument on the command line. Try to read it
// as a script. We've already read BYTES of data into P, but we
// ignore that. Return true if the file was handled.
bool
read_input_script(Workqueue* workqueue, const General_options& options,
Symbol_table* symtab, Layout* layout,
Dirsearch* dirsearch, Input_objects* input_objects,
Input_group* input_group,
const Input_argument* input_argument,
Input_file* input_file, const unsigned char*, off_t,
Task_token* this_blocker, Task_token* next_blocker)
{
std::string input_string;
Lex::read_file(input_file, &input_string);
Lex lex(input_string.c_str(), input_string.length(), PARSING_LINKER_SCRIPT);
Parser_closure closure(input_file->filename().c_str(),
input_argument->file().options(),
input_group != NULL,
input_file->is_in_sysroot(),
NULL,
layout->script_options(),
&lex);
if (yyparse(&closure) != 0)
return false;
// THIS_BLOCKER must be clear before we may add anything to the
// symbol table. We are responsible for unblocking NEXT_BLOCKER
// when we are done. We are responsible for deleting THIS_BLOCKER
// when it is unblocked.
if (!closure.saw_inputs())
{
// The script did not add any files to read. Note that we are
// not permitted to call NEXT_BLOCKER->unblock() here even if
// THIS_BLOCKER is NULL, as we do not hold the workqueue lock.
workqueue->queue(new Script_unblock(this_blocker, next_blocker));
return true;
}
for (Input_arguments::const_iterator p = closure.inputs()->begin();
p != closure.inputs()->end();
++p)
{
Task_token* nb;
if (p + 1 == closure.inputs()->end())
nb = next_blocker;
else
{
nb = new Task_token(true);
nb->add_blocker();
}
workqueue->queue(new Read_symbols(options, input_objects, symtab,
layout, dirsearch, &*p,
input_group, this_blocker, nb));
this_blocker = nb;
}
return true;
}
// Helper function for read_version_script() and
// read_commandline_script(). Processes the given file in the mode
// indicated by first_token and lex_mode.
static bool
read_script_file(const char* filename, Command_line* cmdline,
int first_token, Lex::Mode lex_mode)
{
// TODO: if filename is a relative filename, search for it manually
// using "." + cmdline->options()->search_path() -- not dirsearch.
Dirsearch dirsearch;
// The file locking code wants to record a Task, but we haven't
// started the workqueue yet. This is only for debugging purposes,
// so we invent a fake value.
const Task* task = reinterpret_cast<const Task*>(-1);
Input_file_argument input_argument(filename, false, "",
cmdline->position_dependent_options());
Input_file input_file(&input_argument);
if (!input_file.open(cmdline->options(), dirsearch, task))
return false;
std::string input_string;
Lex::read_file(&input_file, &input_string);
Lex lex(input_string.c_str(), input_string.length(), first_token);
lex.set_mode(lex_mode);
Parser_closure closure(filename,
cmdline->position_dependent_options(),
false,
input_file.is_in_sysroot(),
cmdline,
cmdline->script_options(),
&lex);
if (yyparse(&closure) != 0)
{
input_file.file().unlock(task);
return false;
}
input_file.file().unlock(task);
gold_assert(!closure.saw_inputs());
return true;
}
// FILENAME was found as an argument to --script (-T).
// Read it as a script, and execute its contents immediately.
bool
read_commandline_script(const char* filename, Command_line* cmdline)
{
return read_script_file(filename, cmdline,
PARSING_LINKER_SCRIPT, Lex::LINKER_SCRIPT);
}
// FILE was found as an argument to --version-script. Read it as a
// version script, and store its contents in
// cmdline->script_options()->version_script_info().
bool
read_version_script(const char* filename, Command_line* cmdline)
{
return read_script_file(filename, cmdline,
PARSING_VERSION_SCRIPT, Lex::VERSION_SCRIPT);
}
// Implement the --defsym option on the command line. Return true if
// all is well.
bool
Script_options::define_symbol(const char* definition)
{
Lex lex(definition, strlen(definition), PARSING_DEFSYM);
lex.set_mode(Lex::EXPRESSION);
// Dummy value.
Position_dependent_options posdep_options;
Parser_closure closure("command line", posdep_options, false, false, NULL,
this, &lex);
if (yyparse(&closure) != 0)
return false;
gold_assert(!closure.saw_inputs());
return true;
}
// Manage mapping from keywords to the codes expected by the bison
// parser. We construct one global object for each lex mode with
// keywords.
class Keyword_to_parsecode
{
public:
// The structure which maps keywords to parsecodes.
struct Keyword_parsecode
{
// Keyword.
const char* keyword;
// Corresponding parsecode.
int parsecode;
};
Keyword_to_parsecode(const Keyword_parsecode* keywords,
int keyword_count)
: keyword_parsecodes_(keywords), keyword_count_(keyword_count)
{ }
// Return the parsecode corresponding KEYWORD, or 0 if it is not a
// keyword.
int
keyword_to_parsecode(const char* keyword, size_t len) const;
private:
const Keyword_parsecode* keyword_parsecodes_;
const int keyword_count_;
};
// Mapping from keyword string to keyword parsecode. This array must
// be kept in sorted order. Parsecodes are looked up using bsearch.
// This array must correspond to the list of parsecodes in yyscript.y.
static const Keyword_to_parsecode::Keyword_parsecode
script_keyword_parsecodes[] =
{
{ "ABSOLUTE", ABSOLUTE },
{ "ADDR", ADDR },
{ "ALIGN", ALIGN_K },
{ "ALIGNOF", ALIGNOF },
{ "ASSERT", ASSERT_K },
{ "AS_NEEDED", AS_NEEDED },
{ "AT", AT },
{ "BIND", BIND },
{ "BLOCK", BLOCK },
{ "BYTE", BYTE },
{ "CONSTANT", CONSTANT },
{ "CONSTRUCTORS", CONSTRUCTORS },
{ "COPY", COPY },
{ "CREATE_OBJECT_SYMBOLS", CREATE_OBJECT_SYMBOLS },
{ "DATA_SEGMENT_ALIGN", DATA_SEGMENT_ALIGN },
{ "DATA_SEGMENT_END", DATA_SEGMENT_END },
{ "DATA_SEGMENT_RELRO_END", DATA_SEGMENT_RELRO_END },
{ "DEFINED", DEFINED },
{ "DSECT", DSECT },
{ "ENTRY", ENTRY },
{ "EXCLUDE_FILE", EXCLUDE_FILE },
{ "EXTERN", EXTERN },
{ "FILL", FILL },
{ "FLOAT", FLOAT },
{ "FORCE_COMMON_ALLOCATION", FORCE_COMMON_ALLOCATION },
{ "GROUP", GROUP },
{ "HLL", HLL },
{ "INCLUDE", INCLUDE },
{ "INFO", INFO },
{ "INHIBIT_COMMON_ALLOCATION", INHIBIT_COMMON_ALLOCATION },
{ "INPUT", INPUT },
{ "KEEP", KEEP },
{ "LENGTH", LENGTH },
{ "LOADADDR", LOADADDR },
{ "LONG", LONG },
{ "MAP", MAP },
{ "MAX", MAX_K },
{ "MEMORY", MEMORY },
{ "MIN", MIN_K },
{ "NEXT", NEXT },
{ "NOCROSSREFS", NOCROSSREFS },
{ "NOFLOAT", NOFLOAT },
{ "NOLOAD", NOLOAD },
{ "ONLY_IF_RO", ONLY_IF_RO },
{ "ONLY_IF_RW", ONLY_IF_RW },
{ "OPTION", OPTION },
{ "ORIGIN", ORIGIN },
{ "OUTPUT", OUTPUT },
{ "OUTPUT_ARCH", OUTPUT_ARCH },
{ "OUTPUT_FORMAT", OUTPUT_FORMAT },
{ "OVERLAY", OVERLAY },
{ "PHDRS", PHDRS },
{ "PROVIDE", PROVIDE },
{ "PROVIDE_HIDDEN", PROVIDE_HIDDEN },
{ "QUAD", QUAD },
{ "SEARCH_DIR", SEARCH_DIR },
{ "SECTIONS", SECTIONS },
{ "SEGMENT_START", SEGMENT_START },
{ "SHORT", SHORT },
{ "SIZEOF", SIZEOF },
{ "SIZEOF_HEADERS", SIZEOF_HEADERS },
{ "SORT_BY_ALIGNMENT", SORT_BY_ALIGNMENT },
{ "SORT_BY_NAME", SORT_BY_NAME },
{ "SPECIAL", SPECIAL },
{ "SQUAD", SQUAD },
{ "STARTUP", STARTUP },
{ "SUBALIGN", SUBALIGN },
{ "SYSLIB", SYSLIB },
{ "TARGET", TARGET_K },
{ "TRUNCATE", TRUNCATE },
{ "VERSION", VERSIONK },
{ "global", GLOBAL },
{ "l", LENGTH },
{ "len", LENGTH },
{ "local", LOCAL },
{ "o", ORIGIN },
{ "org", ORIGIN },
{ "sizeof_headers", SIZEOF_HEADERS },
};
static const Keyword_to_parsecode
script_keywords(&script_keyword_parsecodes[0],
(sizeof(script_keyword_parsecodes)
/ sizeof(script_keyword_parsecodes[0])));
static const Keyword_to_parsecode::Keyword_parsecode
version_script_keyword_parsecodes[] =
{
{ "extern", EXTERN },
{ "global", GLOBAL },
{ "local", LOCAL },
};
static const Keyword_to_parsecode
version_script_keywords(&version_script_keyword_parsecodes[0],
(sizeof(version_script_keyword_parsecodes)
/ sizeof(version_script_keyword_parsecodes[0])));
// Comparison function passed to bsearch.
extern "C"
{
struct Ktt_key
{
const char* str;
size_t len;
};
static int
ktt_compare(const void* keyv, const void* kttv)
{
const Ktt_key* key = static_cast<const Ktt_key*>(keyv);
const Keyword_to_parsecode::Keyword_parsecode* ktt =
static_cast<const Keyword_to_parsecode::Keyword_parsecode*>(kttv);
int i = strncmp(key->str, ktt->keyword, key->len);
if (i != 0)
return i;
if (ktt->keyword[key->len] != '\0')
return -1;
return 0;
}
} // End extern "C".
int
Keyword_to_parsecode::keyword_to_parsecode(const char* keyword,
size_t len) const
{
Ktt_key key;
key.str = keyword;
key.len = len;
void* kttv = bsearch(&key,
this->keyword_parsecodes_,
this->keyword_count_,
sizeof(this->keyword_parsecodes_[0]),
ktt_compare);
if (kttv == NULL)
return 0;
Keyword_parsecode* ktt = static_cast<Keyword_parsecode*>(kttv);
return ktt->parsecode;
}
} // End namespace gold.
// The remaining functions are extern "C", so it's clearer to not put
// them in namespace gold.
using namespace gold;
// This function is called by the bison parser to return the next
// token.
extern "C" int
yylex(YYSTYPE* lvalp, void* closurev)
{
Parser_closure* closure = static_cast<Parser_closure*>(closurev);
const Token* token = closure->next_token();
switch (token->classification())
{
default:
gold_unreachable();
case Token::TOKEN_INVALID:
yyerror(closurev, "invalid character");
return 0;
case Token::TOKEN_EOF:
return 0;
case Token::TOKEN_STRING:
{
// This is either a keyword or a STRING.
size_t len;
const char* str = token->string_value(&len);
int parsecode = 0;
switch (closure->lex_mode())
{
case Lex::LINKER_SCRIPT:
parsecode = script_keywords.keyword_to_parsecode(str, len);
break;
case Lex::VERSION_SCRIPT:
parsecode = version_script_keywords.keyword_to_parsecode(str, len);
break;
default:
break;
}
if (parsecode != 0)
return parsecode;
lvalp->string.value = str;
lvalp->string.length = len;
return STRING;
}
case Token::TOKEN_QUOTED_STRING:
lvalp->string.value = token->string_value(&lvalp->string.length);
return STRING;
case Token::TOKEN_OPERATOR:
return token->operator_value();
case Token::TOKEN_INTEGER:
lvalp->integer = token->integer_value();
return INTEGER;
}
}
// This function is called by the bison parser to report an error.
extern "C" void
yyerror(void* closurev, const char* message)
{
Parser_closure* closure = static_cast<Parser_closure*>(closurev);
gold_error(_("%s:%d:%d: %s"), closure->filename(), closure->lineno(),
closure->charpos(), message);
}
// Called by the bison parser to add a file to the link.
extern "C" void
script_add_file(void* closurev, const char* name, size_t length)
{
Parser_closure* closure = static_cast<Parser_closure*>(closurev);
// If this is an absolute path, and we found the script in the
// sysroot, then we want to prepend the sysroot to the file name.
// For example, this is how we handle a cross link to the x86_64
// libc.so, which refers to /lib/libc.so.6.
std::string name_string(name, length);
const char* extra_search_path = ".";
std::string script_directory;
if (IS_ABSOLUTE_PATH(name_string.c_str()))
{
if (closure->is_in_sysroot())
{
const std::string& sysroot(parameters->sysroot());
gold_assert(!sysroot.empty());
name_string = sysroot + name_string;
}
}
else
{
// In addition to checking the normal library search path, we
// also want to check in the script-directory.
const char *slash = strrchr(closure->filename(), '/');
if (slash != NULL)
{
script_directory.assign(closure->filename(),
slash - closure->filename() + 1);
extra_search_path = script_directory.c_str();
}
}
Input_file_argument file(name_string.c_str(), false, extra_search_path,
closure->position_dependent_options());
closure->inputs()->add_file(file);
}
// Called by the bison parser to start a group. If we are already in
// a group, that means that this script was invoked within a
// --start-group --end-group sequence on the command line, or that
// this script was found in a GROUP of another script. In that case,
// we simply continue the existing group, rather than starting a new
// one. It is possible to construct a case in which this will do
// something other than what would happen if we did a recursive group,
// but it's hard to imagine why the different behaviour would be
// useful for a real program. Avoiding recursive groups is simpler
// and more efficient.
extern "C" void
script_start_group(void* closurev)
{
Parser_closure* closure = static_cast<Parser_closure*>(closurev);
if (!closure->in_group())
closure->inputs()->start_group();
}
// Called by the bison parser at the end of a group.
extern "C" void
script_end_group(void* closurev)
{
Parser_closure* closure = static_cast<Parser_closure*>(closurev);
if (!closure->in_group())
closure->inputs()->end_group();
}
// Called by the bison parser to start an AS_NEEDED list.
extern "C" void
script_start_as_needed(void* closurev)
{
Parser_closure* closure = static_cast<Parser_closure*>(closurev);
closure->position_dependent_options().set_as_needed();
}
// Called by the bison parser at the end of an AS_NEEDED list.
extern "C" void
script_end_as_needed(void* closurev)
{
Parser_closure* closure = static_cast<Parser_closure*>(closurev);
closure->position_dependent_options().clear_as_needed();
}
// Called by the bison parser to set the entry symbol.
extern "C" void
script_set_entry(void* closurev, const char* entry, size_t length)
{
Parser_closure* closure = static_cast<Parser_closure*>(closurev);
closure->script_options()->set_entry(entry, length);
}
// Called by the bison parser to define a symbol.
extern "C" void
script_set_symbol(void* closurev, const char* name, size_t length,
Expression* value, int providei, int hiddeni)
{
Parser_closure* closure = static_cast<Parser_closure*>(closurev);
const bool provide = providei != 0;
const bool hidden = hiddeni != 0;
closure->script_options()->add_symbol_assignment(name, length, value,
provide, hidden);
}
// Called by the bison parser to parse an OPTION.
extern "C" void
script_parse_option(void* closurev, const char* option, size_t length)
{
Parser_closure* closure = static_cast<Parser_closure*>(closurev);
// We treat the option as a single command-line option, even if
// it has internal whitespace.
if (closure->command_line() == NULL)
{
// There are some options that we could handle here--e.g.,
// -lLIBRARY. Should we bother?
gold_warning(_("%s:%d:%d: ignoring command OPTION; OPTION is only valid"
" for scripts specified via -T/--script"),
closure->filename(), closure->lineno(), closure->charpos());
}
else
{
bool past_a_double_dash_option = false;
char* mutable_option = strndup(option, length);
gold_assert(mutable_option != NULL);
closure->command_line()->process_one_option(1, &mutable_option, 0,
&past_a_double_dash_option);
free(mutable_option);
}
}
/* Called by the bison parser to push the lexer into expression
mode. */
extern void
script_push_lex_into_expression_mode(void* closurev)
{
Parser_closure* closure = static_cast<Parser_closure*>(closurev);
closure->push_lex_mode(Lex::EXPRESSION);
}
/* Called by the bison parser to push the lexer into version
mode. */
extern void
script_push_lex_into_version_mode(void* closurev)
{
Parser_closure* closure = static_cast<Parser_closure*>(closurev);
closure->push_lex_mode(Lex::VERSION_SCRIPT);
}
/* Called by the bison parser to pop the lexer mode. */
extern void
script_pop_lex_mode(void* closurev)
{
Parser_closure* closure = static_cast<Parser_closure*>(closurev);
closure->pop_lex_mode();
}
// The following structs are used within the VersionInfo class as well
// as in the bison helper functions. They store the information
// parsed from the version script.
// A single version expression.
// For example, pattern="std::map*" and language="C++".
// pattern and language should be from the stringpool
struct Version_expression {
Version_expression(const std::string& pattern,
const std::string& language)
: pattern(pattern), language(language) {}
std::string pattern;
std::string language;
};
// A list of expressions.
struct Version_expression_list {
std::vector<struct Version_expression> expressions;
};
// A list of which versions upon which another version depends.
// Strings should be from the Stringpool.
struct Version_dependency_list {
std::vector<std::string> dependencies;
};
// The total definition of a version. It includes the tag for the
// version, its global and local expressions, and any dependencies.
struct Version_tree {
Version_tree()
: tag(), global(NULL), local(NULL), dependencies(NULL) {}
std::string tag;
const struct Version_expression_list* global;
const struct Version_expression_list* local;
const struct Version_dependency_list* dependencies;
};
Version_script_info::~Version_script_info()
{
for (size_t k = 0; k < dependency_lists_.size(); ++k)
delete dependency_lists_[k];
for (size_t k = 0; k < version_trees_.size(); ++k)
delete version_trees_[k];
for (size_t k = 0; k < expression_lists_.size(); ++k)
delete expression_lists_[k];
}
std::vector<std::string>
Version_script_info::get_versions() const
{
std::vector<std::string> ret;
for (size_t j = 0; j < version_trees_.size(); ++j)
ret.push_back(version_trees_[j]->tag);
return ret;
}
std::vector<std::string>
Version_script_info::get_dependencies(const char* version) const
{
std::vector<std::string> ret;
for (size_t j = 0; j < version_trees_.size(); ++j)
if (version_trees_[j]->tag == version)
{
const struct Version_dependency_list* deps =
version_trees_[j]->dependencies;
if (deps != NULL)
for (size_t k = 0; k < deps->dependencies.size(); ++k)
ret.push_back(deps->dependencies[k]);
return ret;
}
return ret;
}
const std::string&
Version_script_info::get_symbol_version_helper(const char* symbol_name,
bool check_global) const
{
for (size_t j = 0; j < version_trees_.size(); ++j)
{
// Is it a global symbol for this version?
const Version_expression_list* exp =
check_global ? version_trees_[j]->global : version_trees_[j]->local;
if (exp != NULL)
for (size_t k = 0; k < exp->expressions.size(); ++k)
{
const char* name_to_match = symbol_name;
char* demangled_name = NULL;
if (exp->expressions[k].language == "C++")
{
demangled_name = cplus_demangle(symbol_name,
DMGL_ANSI | DMGL_PARAMS);
// This isn't a C++ symbol.
if (demangled_name == NULL)
continue;
name_to_match = demangled_name;
}
else if (exp->expressions[k].language == "Java")
{
demangled_name = cplus_demangle(symbol_name,
(DMGL_ANSI | DMGL_PARAMS
| DMGL_JAVA));
// This isn't a Java symbol.
if (demangled_name == NULL)
continue;
name_to_match = demangled_name;
}
bool matched = fnmatch(exp->expressions[k].pattern.c_str(),
name_to_match, FNM_NOESCAPE) == 0;
if (demangled_name != NULL)
free(demangled_name);
if (matched)
return version_trees_[j]->tag;
}
}
static const std::string empty = "";
return empty;
}
struct Version_dependency_list*
Version_script_info::allocate_dependency_list()
{
dependency_lists_.push_back(new Version_dependency_list);
return dependency_lists_.back();
}
struct Version_expression_list*
Version_script_info::allocate_expression_list()
{
expression_lists_.push_back(new Version_expression_list);
return expression_lists_.back();
}
struct Version_tree*
Version_script_info::allocate_version_tree()
{
version_trees_.push_back(new Version_tree);
return version_trees_.back();
}
// Register an entire version node. For example:
//
// GLIBC_2.1 {
// global: foo;
// } GLIBC_2.0;
//
// - tag is "GLIBC_2.1"
// - tree contains the information "global: foo"
// - deps contains "GLIBC_2.0"
extern "C" void
script_register_vers_node(void*,
const char* tag,
int taglen,
struct Version_tree *tree,
struct Version_dependency_list *deps)
{
gold_assert(tree != NULL);
gold_assert(tag != NULL);
tree->dependencies = deps;
tree->tag = std::string(tag, taglen);
}
// Add a dependencies to the list of existing dependencies, if any,
// and return the expanded list.
extern "C" struct Version_dependency_list *
script_add_vers_depend(void* closurev,
struct Version_dependency_list *all_deps,
const char *depend_to_add, int deplen)
{
Parser_closure* closure = static_cast<Parser_closure*>(closurev);
if (all_deps == NULL)
all_deps = closure->version_script()->allocate_dependency_list();
all_deps->dependencies.push_back(std::string(depend_to_add, deplen));
return all_deps;
}
// Add a pattern expression to an existing list of expressions, if any.
// TODO: In the old linker, the last argument used to be a bool, but I
// don't know what it meant.
extern "C" struct Version_expression_list *
script_new_vers_pattern(void* closurev,
struct Version_expression_list *expressions,
const char *pattern, int patlen)
{
Parser_closure* closure = static_cast<Parser_closure*>(closurev);
if (expressions == NULL)
expressions = closure->version_script()->allocate_expression_list();
expressions->expressions.push_back(
Version_expression(std::string(pattern, patlen),
closure->get_current_language()));
return expressions;
}
// Combine the global and local expressions into a a Version_tree.
extern "C" struct Version_tree *
script_new_vers_node(void* closurev,
struct Version_expression_list *global,
struct Version_expression_list *local)
{
Parser_closure* closure = static_cast<Parser_closure*>(closurev);
Version_tree* tree = closure->version_script()->allocate_version_tree();
tree->global = global;
tree->local = local;
return tree;
}
// Handle a transition in language, such as at the
// start or end of 'extern "C++"'
extern "C" void
version_script_push_lang(void* closurev, const char* lang, int langlen)
{
Parser_closure* closure = static_cast<Parser_closure*>(closurev);
closure->push_language(std::string(lang, langlen));
}
extern "C" void
version_script_pop_lang(void* closurev)
{
Parser_closure* closure = static_cast<Parser_closure*>(closurev);
closure->pop_language();
}
|