summaryrefslogtreecommitdiffstats
path: root/scripts/archive-source.sh
diff options
context:
space:
mode:
Diffstat (limited to 'scripts/archive-source.sh')
0 files changed, 0 insertions, 0 deletions
href='#n54'>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
#!/usr/bin/env python3
# Copyright (c) 2018 Linaro Limited
#
# This library is free software; you can redistribute it and/or
# modify it under the terms of the GNU Lesser General Public
# License as published by the Free Software Foundation; either
# version 2.1 of the License, or (at your option) any later version.
#
# This library 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
# Lesser General Public License for more details.
#
# You should have received a copy of the GNU Lesser General Public
# License along with this library; if not, see <http://www.gnu.org/licenses/>.
#

#
# Generate a decoding tree from a specification file.
# See the syntax and semantics in docs/devel/decodetree.rst.
#

import io
import os
import re
import sys
import getopt

insnwidth = 32
bitop_width = 32
insnmask = 0xffffffff
variablewidth = False
fields = {}
arguments = {}
formats = {}
allpatterns = []
anyextern = False

translate_prefix = 'trans'
translate_scope = 'static '
input_file = ''
output_file = None
output_fd = None
insntype = 'uint32_t'
decode_function = 'decode'

# An identifier for C.
re_C_ident = '[a-zA-Z][a-zA-Z0-9_]*'

# Identifiers for Arguments, Fields, Formats and Patterns.
re_arg_ident = '&[a-zA-Z0-9_]*'
re_fld_ident = '%[a-zA-Z0-9_]*'
re_fmt_ident = '@[a-zA-Z0-9_]*'
re_pat_ident = '[a-zA-Z0-9_]*'

def error_with_file(file, lineno, *args):
    """Print an error message from file:line and args and exit."""
    global output_file
    global output_fd

    prefix = ''
    if file:
        prefix += f'{file}:'
    if lineno:
        prefix += f'{lineno}:'
    if prefix:
        prefix += ' '
    print(prefix, end='error: ', file=sys.stderr)
    print(*args, file=sys.stderr)

    if output_file and output_fd:
        output_fd.close()
        os.remove(output_file)
    exit(1)
# end error_with_file


def error(lineno, *args):
    error_with_file(input_file, lineno, *args)
# end error


def output(*args):
    global output_fd
    for a in args:
        output_fd.write(a)


def output_autogen():
    output('/* This file is autogenerated by scripts/decodetree.py.  */\n\n')


def str_indent(c):
    """Return a string with C spaces"""
    return ' ' * c


def str_fields(fields):
    """Return a string uniquely identifying FIELDS"""
    r = ''
    for n in sorted(fields.keys()):
        r += '_' + n
    return r[1:]


def whex(val):
    """Return a hex string for val padded for insnwidth"""
    global insnwidth
    return f'0x{val:0{insnwidth // 4}x}'


def whexC(val):
    """Return a hex string for val padded for insnwidth,
       and with the proper suffix for a C constant."""
    suffix = ''
    if val >= 0x100000000:
        suffix = 'ull'
    elif val >= 0x80000000:
        suffix = 'u'
    return whex(val) + suffix


def str_match_bits(bits, mask):
    """Return a string pretty-printing BITS/MASK"""
    global insnwidth

    i = 1 << (insnwidth - 1)
    space = 0x01010100
    r = ''
    while i != 0:
        if i & mask:
            if i & bits:
                r += '1'
            else:
                r += '0'
        else:
            r += '.'
        if i & space:
            r += ' '
        i >>= 1
    return r


def is_pow2(x):
    """Return true iff X is equal to a power of 2."""
    return (x & (x - 1)) == 0


def ctz(x):
    """Return the number of times 2 factors into X."""
    assert x != 0
    r = 0
    while ((x >> r) & 1) == 0:
        r += 1
    return r


def is_contiguous(bits):
    if bits == 0:
        return -1
    shift = ctz(bits)
    if is_pow2((bits >> shift) + 1):
        return shift
    else:
        return -1


def eq_fields_for_args(flds_a, arg):
    if len(flds_a) != len(arg.fields):
        return False
    # Only allow inference on default types
    for t in arg.types:
        if t != 'int':
            return False
    for k, a in flds_a.items():
        if k not in arg.fields:
            return False
    return True


def eq_fields_for_fmts(flds_a, flds_b):
    if len(flds_a) != len(flds_b):
        return False
    for k, a in flds_a.items():
        if k not in flds_b:
            return False
        b = flds_b[k]
        if a.__class__ != b.__class__ or a != b:
            return False
    return True


class Field:
    """Class representing a simple instruction field"""
    def __init__(self, sign, pos, len):
        self.sign = sign
        self.pos = pos
        self.len = len
        self.mask = ((1 << len) - 1) << pos

    def __str__(self):
        if self.sign:
            s = 's'
        else:
            s = ''
        return str(self.pos) + ':' + s + str(self.len)

    def str_extract(self):
        global bitop_width
        s = 's' if self.sign else ''
        return f'{s}extract{bitop_width}(insn, {self.pos}, {self.len})'

    def __eq__(self, other):
        return self.sign == other.sign and self.mask == other.mask

    def __ne__(self, other):
        return not self.__eq__(other)
# end Field


class MultiField:
    """Class representing a compound instruction field"""
    def __init__(self, subs, mask):
        self.subs = subs
        self.sign = subs[0].sign
        self.mask = mask

    def __str__(self):
        return str(self.subs)

    def str_extract(self):
        global bitop_width
        ret = '0'
        pos = 0
        for f in reversed(self.subs):
            ext = f.str_extract()
            if pos == 0:
                ret = ext
            else:
                ret = f'deposit{bitop_width}({ret}, {pos}, {bitop_width - pos}, {ext})'
            pos += f.len
        return ret

    def __ne__(self, other):
        if len(self.subs) != len(other.subs):
            return True
        for a, b in zip(self.subs, other.subs):
            if a.__class__ != b.__class__ or a != b:
                return True
        return False

    def __eq__(self, other):
        return not self.__ne__(other)
# end MultiField


class ConstField:
    """Class representing an argument field with constant value"""
    def __init__(self, value):
        self.value = value
        self.mask = 0
        self.sign = value < 0

    def __str__(self):
        return str(self.value)

    def str_extract(self):
        return str(self.value)

    def __cmp__(self, other):
        return self.value - other.value
# end ConstField


class FunctionField:
    """Class representing a field passed through a function"""
    def __init__(self, func, base):
        self.mask = base.mask
        self.sign = base.sign
        self.base = base
        self.func = func

    def __str__(self):
        return self.func + '(' + str(self.base) + ')'

    def str_extract(self):
        return self.func + '(ctx, ' + self.base.str_extract() + ')'

    def __eq__(self, other):
        return self.func == other.func and self.base == other.base

    def __ne__(self, other):
        return not self.__eq__(other)
# end FunctionField


class ParameterField:
    """Class representing a pseudo-field read from a function"""
    def __init__(self, func):
        self.mask = 0
        self.sign = 0
        self.func = func

    def __str__(self):
        return self.func

    def str_extract(self):
        return self.func + '(ctx)'

    def __eq__(self, other):
        return self.func == other.func

    def __ne__(self, other):
        return not self.__eq__(other)
# end ParameterField


class Arguments:
    """Class representing the extracted fields of a format"""
    def __init__(self, nm, flds, types, extern):
        self.name = nm
        self.extern = extern
        self.fields = flds
        self.types = types

    def __str__(self):
        return self.name + ' ' + str(self.fields)

    def struct_name(self):
        return 'arg_' + self.name

    def output_def(self):
        if not self.extern:
            output('typedef struct {\n')
            for (n, t) in zip(self.fields, self.types):
                output(f'    {t} {n};\n')
            output('} ', self.struct_name(), ';\n\n')
# end Arguments


class General:
    """Common code between instruction formats and instruction patterns"""
    def __init__(self, name, lineno, base, fixb, fixm, udfm, fldm, flds, w):
        self.name = name
        self.file = input_file
        self.lineno = lineno
        self.base = base
        self.fixedbits = fixb
        self.fixedmask = fixm
        self.undefmask = udfm
        self.fieldmask = fldm
        self.fields = flds
        self.width = w

    def __str__(self):
        return self.name + ' ' + str_match_bits(self.fixedbits, self.fixedmask)

    def str1(self, i):
        return str_indent(i) + self.__str__()
# end General


class Format(General):
    """Class representing an instruction format"""

    def extract_name(self):
        global decode_function
        return decode_function + '_extract_' + self.name

    def output_extract(self):
        output('static void ', self.extract_name(), '(DisasContext *ctx, ',
               self.base.struct_name(), ' *a, ', insntype, ' insn)\n{\n')
        for n, f in self.fields.items():
            output('    a->', n, ' = ', f.str_extract(), ';\n')
        output('}\n\n')
# end Format


class Pattern(General):
    """Class representing an instruction pattern"""

    def output_decl(self):
        global translate_scope
        global translate_prefix
        output('typedef ', self.base.base.struct_name(),
               ' arg_', self.name, ';\n')
        output(translate_scope, 'bool ', translate_prefix, '_', self.name,
               '(DisasContext *ctx, arg_', self.name, ' *a);\n')

    def output_code(self, i, extracted, outerbits, outermask):
        global translate_prefix
        ind = str_indent(i)
        arg = self.base.base.name
        output(ind, '/* ', self.file, ':', str(self.lineno), ' */\n')
        if not extracted:
            output(ind, self.base.extract_name(),
                   '(ctx, &u.f_', arg, ', insn);\n')
        for n, f in self.fields.items():
            output(ind, 'u.f_', arg, '.', n, ' = ', f.str_extract(), ';\n')
        output(ind, 'if (', translate_prefix, '_', self.name,
               '(ctx, &u.f_', arg, ')) return true;\n')

    # Normal patterns do not have children.
    def build_tree(self):
        return
    def prop_masks(self):
        return
    def prop_format(self):
        return
    def prop_width(self):
        return

# end Pattern


class MultiPattern(General):
    """Class representing a set of instruction patterns"""

    def __init__(self, lineno):
        self.file = input_file
        self.lineno = lineno
        self.pats = []
        self.base = None
        self.fixedbits = 0
        self.fixedmask = 0
        self.undefmask = 0
        self.width = None

    def __str__(self):
        r = 'group'
        if self.fixedbits is not None:
            r += ' ' + str_match_bits(self.fixedbits, self.fixedmask)
        return r

    def output_decl(self):
        for p in self.pats:
            p.output_decl()

    def prop_masks(self):
        global insnmask

        fixedmask = insnmask
        undefmask = insnmask

        # Collect fixedmask/undefmask for all of the children.
        for p in self.pats:
            p.prop_masks()
            fixedmask &= p.fixedmask
            undefmask &= p.undefmask

        # Widen fixedmask until all fixedbits match
        repeat = True
        fixedbits = 0
        while repeat and fixedmask != 0:
            fixedbits = None
            for p in self.pats:
                thisbits = p.fixedbits & fixedmask
                if fixedbits is None:
                    fixedbits = thisbits
                elif fixedbits != thisbits:
                    fixedmask &= ~(fixedbits ^ thisbits)
                    break
            else:
                repeat = False

        self.fixedbits = fixedbits
        self.fixedmask = fixedmask
        self.undefmask = undefmask

    def build_tree(self):
        for p in self.pats:
            p.build_tree()

    def prop_format(self):
        for p in self.pats:
            p.build_tree()

    def prop_width(self):
        width = None
        for p in self.pats:
            p.prop_width()
            if width is None:
                width = p.width
            elif width != p.width:
                error_with_file(self.file, self.lineno,
                                'width mismatch in patterns within braces')
        self.width = width

# end MultiPattern


class IncMultiPattern(MultiPattern):
    """Class representing an overlapping set of instruction patterns"""

    def output_code(self, i, extracted, outerbits, outermask):
        global translate_prefix
        ind = str_indent(i)
        for p in self.pats:
            if outermask != p.fixedmask:
                innermask = p.fixedmask & ~outermask
                innerbits = p.fixedbits & ~outermask
                output(ind, f'if ((insn & {whexC(innermask)}) == {whexC(innerbits)}) {{\n')
                output(ind, f'    /* {str_match_bits(p.fixedbits, p.fixedmask)} */\n')
                p.output_code(i + 4, extracted, p.fixedbits, p.fixedmask)
                output(ind, '}\n')
            else:
                p.output_code(i, extracted, p.fixedbits, p.fixedmask)
#end IncMultiPattern


class Tree:
    """Class representing a node in a decode tree"""

    def __init__(self, fm, tm):
        self.fixedmask = fm
        self.thismask = tm
        self.subs = []
        self.base = None

    def str1(self, i):
        ind = str_indent(i)
        r = ind + whex(self.fixedmask)
        if self.format:
            r += ' ' + self.format.name
        r += ' [\n'
        for (b, s) in self.subs:
            r += ind + f'  {whex(b)}:\n'
            r += s.str1(i + 4) + '\n'
        r += ind + ']'
        return r

    def __str__(self):
        return self.str1(0)

    def output_code(self, i, extracted, outerbits, outermask):
        ind = str_indent(i)

        # If we identified all nodes below have the same format,
        # extract the fields now.
        if not extracted and self.base:
            output(ind, self.base.extract_name(),
                   '(ctx, &u.f_', self.base.base.name, ', insn);\n')
            extracted = True

        # Attempt to aid the compiler in producing compact switch statements.
        # If the bits in the mask are contiguous, extract them.
        sh = is_contiguous(self.thismask)
        if sh > 0:
            # Propagate SH down into the local functions.
            def str_switch(b, sh=sh):
                return f'(insn >> {sh}) & {b >> sh:#x}'

            def str_case(b, sh=sh):
                return hex(b >> sh)
        else:
            def str_switch(b):
                return f'insn & {whexC(b)}'

            def str_case(b):
                return whexC(b)

        output(ind, 'switch (', str_switch(self.thismask), ') {\n')
        for b, s in sorted(self.subs):
            assert (self.thismask & ~s.fixedmask) == 0
            innermask = outermask | self.thismask
            innerbits = outerbits | b
            output(ind, 'case ', str_case(b), ':\n')
            output(ind, '    /* ',
                   str_match_bits(innerbits, innermask), ' */\n')
            s.output_code(i + 4, extracted, innerbits, innermask)
            output(ind, '    break;\n')
        output(ind, '}\n')
# end Tree


class ExcMultiPattern(MultiPattern):
    """Class representing a non-overlapping set of instruction patterns"""

    def output_code(self, i, extracted, outerbits, outermask):
        # Defer everything to our decomposed Tree node
        self.tree.output_code(i, extracted, outerbits, outermask)

    @staticmethod
    def __build_tree(pats, outerbits, outermask):
        # Find the intersection of all remaining fixedmask.
        innermask = ~outermask & insnmask
        for i in pats:
            innermask &= i.fixedmask

        if innermask == 0:
            # Edge condition: One pattern covers the entire insnmask
            if len(pats) == 1:
                t = Tree(outermask, innermask)
                t.subs.append((0, pats[0]))
                return t

            text = 'overlapping patterns:'
            for p in pats:
                text += '\n' + p.file + ':' + str(p.lineno) + ': ' + str(p)
            error_with_file(pats[0].file, pats[0].lineno, text)

        fullmask = outermask | innermask

        # Sort each element of pats into the bin selected by the mask.
        bins = {}
        for i in pats:
            fb = i.fixedbits & innermask
            if fb in bins:
                bins[fb].append(i)
            else:
                bins[fb] = [i]

        # We must recurse if any bin has more than one element or if
        # the single element in the bin has not been fully matched.
        t = Tree(fullmask, innermask)

        for b, l in bins.items():
            s = l[0]
            if len(l) > 1 or s.fixedmask & ~fullmask != 0:
                s = ExcMultiPattern.__build_tree(l, b | outerbits, fullmask)
            t.subs.append((b, s))

        return t

    def build_tree(self):
        super().prop_format()
        self.tree = self.__build_tree(self.pats, self.fixedbits,
                                      self.fixedmask)

    @staticmethod
    def __prop_format(tree):
        """Propagate Format objects into the decode tree"""

        # Depth first search.
        for (b, s) in tree.subs:
            if isinstance(s, Tree):
                ExcMultiPattern.__prop_format(s)

        # If all entries in SUBS have the same format, then
        # propagate that into the tree.
        f = None
        for (b, s) in tree.subs:
            if f is None:
                f = s.base
                if f is None:
                    return
            if f is not s.base:
                return
        tree.base = f

    def prop_format(self):
        super().prop_format()
        self.__prop_format(self.tree)

# end ExcMultiPattern


def parse_field(lineno, name, toks):
    """Parse one instruction field from TOKS at LINENO"""
    global fields
    global insnwidth

    # A "simple" field will have only one entry;
    # a "multifield" will have several.
    subs = []
    width = 0
    func = None
    for t in toks:
        if re.match('^!function=', t):
            if func:
                error(lineno, 'duplicate function')
            func = t.split('=')
            func = func[1]
            continue

        if re.fullmatch('[0-9]+:s[0-9]+', t):
            # Signed field extract
            subtoks = t.split(':s')
            sign = True
        elif re.fullmatch('[0-9]+:[0-9]+', t):
            # Unsigned field extract
            subtoks = t.split(':')
            sign = False
        else:
            error(lineno, f'invalid field token "{t}"')
        po = int(subtoks[0])
        le = int(subtoks[1])
        if po + le > insnwidth:
            error(lineno, f'field {t} too large')
        f = Field(sign, po, le)
        subs.append(f)
        width += le

    if width > insnwidth:
        error(lineno, 'field too large')
    if len(subs) == 0:
        if func:
            f = ParameterField(func)
        else:
            error(lineno, 'field with no value')
    else:
        if len(subs) == 1:
            f = subs[0]
        else:
            mask = 0
            for s in subs:
                if mask & s.mask:
                    error(lineno, 'field components overlap')
                mask |= s.mask
            f = MultiField(subs, mask)
        if func:
            f = FunctionField(func, f)

    if name in fields:
        error(lineno, 'duplicate field', name)
    fields[name] = f
# end parse_field


def parse_arguments(lineno, name, toks):
    """Parse one argument set from TOKS at LINENO"""
    global arguments
    global re_C_ident
    global anyextern

    flds = []
    types = []
    extern = False
    for n in toks:
        if re.fullmatch('!extern', n):
            extern = True
            anyextern = True
            continue
        if re.fullmatch(re_C_ident + ':' + re_C_ident, n):
            (n, t) = n.split(':')
        elif re.fullmatch(re_C_ident, n):
            t = 'int'
        else:
            error(lineno, f'invalid argument set token "{n}"')
        if n in flds:
            error(lineno, f'duplicate argument "{n}"')
        flds.append(n)
        types.append(t)

    if name in arguments:
        error(lineno, 'duplicate argument set', name)
    arguments[name] = Arguments(name, flds, types, extern)
# end parse_arguments


def lookup_field(lineno, name):
    global fields
    if name in fields:
        return fields[name]
    error(lineno, 'undefined field', name)


def add_field(lineno, flds, new_name, f):
    if new_name in flds:
        error(lineno, 'duplicate field', new_name)
    flds[new_name] = f
    return flds


def add_field_byname(lineno, flds, new_name, old_name):
    return add_field(lineno, flds, new_name, lookup_field(lineno, old_name))


def infer_argument_set(flds):
    global arguments
    global decode_function

    for arg in arguments.values():
        if eq_fields_for_args(flds, arg):
            return arg

    name = decode_function + str(len(arguments))
    arg = Arguments(name, flds.keys(), ['int'] * len(flds), False)
    arguments[name] = arg
    return arg


def infer_format(arg, fieldmask, flds, width):
    global arguments
    global formats
    global decode_function

    const_flds = {}
    var_flds = {}
    for n, c in flds.items():
        if c is ConstField:
            const_flds[n] = c
        else:
            var_flds[n] = c

    # Look for an existing format with the same argument set and fields
    for fmt in formats.values():
        if arg and fmt.base != arg:
            continue
        if fieldmask != fmt.fieldmask:
            continue
        if width != fmt.width:
            continue
        if not eq_fields_for_fmts(flds, fmt.fields):
            continue
        return (fmt, const_flds)

    name = decode_function + '_Fmt_' + str(len(formats))
    if not arg:
        arg = infer_argument_set(flds)

    fmt = Format(name, 0, arg, 0, 0, 0, fieldmask, var_flds, width)
    formats[name] = fmt

    return (fmt, const_flds)
# end infer_format


def parse_generic(lineno, parent_pat, name, toks):
    """Parse one instruction format from TOKS at LINENO"""
    global fields
    global arguments
    global formats
    global allpatterns
    global re_arg_ident
    global re_fld_ident
    global re_fmt_ident
    global re_C_ident
    global insnwidth
    global insnmask
    global variablewidth

    is_format = parent_pat is None

    fixedmask = 0
    fixedbits = 0
    undefmask = 0
    width = 0
    flds = {}
    arg = None
    fmt = None
    for t in toks:
        # '&Foo' gives a format an explicit argument set.
        if re.fullmatch(re_arg_ident, t):
            tt = t[1:]
            if arg:
                error(lineno, 'multiple argument sets')
            if tt in arguments:
                arg = arguments[tt]
            else:
                error(lineno, 'undefined argument set', t)
            continue

        # '@Foo' gives a pattern an explicit format.
        if re.fullmatch(re_fmt_ident, t):
            tt = t[1:]
            if fmt:
                error(lineno, 'multiple formats')
            if tt in formats:
                fmt = formats[tt]
            else:
                error(lineno, 'undefined format', t)
            continue

        # '%Foo' imports a field.
        if re.fullmatch(re_fld_ident, t):
            tt = t[1:]
            flds = add_field_byname(lineno, flds, tt, tt)
            continue

        # 'Foo=%Bar' imports a field with a different name.
        if re.fullmatch(re_C_ident + '=' + re_fld_ident, t):
            (fname, iname) = t.split('=%')
            flds = add_field_byname(lineno, flds, fname, iname)
            continue

        # 'Foo=number' sets an argument field to a constant value
        if re.fullmatch(re_C_ident + '=[+-]?[0-9]+', t):
            (fname, value) = t.split('=')
            value = int(value)
            flds = add_field(lineno, flds, fname, ConstField(value))
            continue

        # Pattern of 0s, 1s, dots and dashes indicate required zeros,
        # required ones, or dont-cares.
        if re.fullmatch('[01.-]+', t):
            shift = len(t)
            fms = t.replace('0', '1')
            fms = fms.replace('.', '0')
            fms = fms.replace('-', '0')
            fbs = t.replace('.', '0')
            fbs = fbs.replace('-', '0')
            ubm = t.replace('1', '0')
            ubm = ubm.replace('.', '0')
            ubm = ubm.replace('-', '1')
            fms = int(fms, 2)
            fbs = int(fbs, 2)
            ubm = int(ubm, 2)
            fixedbits = (fixedbits << shift) | fbs
            fixedmask = (fixedmask << shift) | fms
            undefmask = (undefmask << shift) | ubm
        # Otherwise, fieldname:fieldwidth
        elif re.fullmatch(re_C_ident + ':s?[0-9]+', t):
            (fname, flen) = t.split(':')
            sign = False
            if flen[0] == 's':
                sign = True
                flen = flen[1:]
            shift = int(flen, 10)
            if shift + width > insnwidth:
                error(lineno, f'field {fname} exceeds insnwidth')
            f = Field(sign, insnwidth - width - shift, shift)
            flds = add_field(lineno, flds, fname, f)
            fixedbits <<= shift
            fixedmask <<= shift
            undefmask <<= shift
        else:
            error(lineno, f'invalid token "{t}"')
        width += shift

    if variablewidth and width < insnwidth and width % 8 == 0:
        shift = insnwidth - width
        fixedbits <<= shift
        fixedmask <<= shift
        undefmask <<= shift
        undefmask |= (1 << shift) - 1

    # We should have filled in all of the bits of the instruction.
    elif not (is_format and width == 0) and width != insnwidth:
        error(lineno, f'definition has {width} bits')

    # Do not check for fields overlapping fields; one valid usage
    # is to be able to duplicate fields via import.
    fieldmask = 0
    for f in flds.values():
        fieldmask |= f.mask

    # Fix up what we've parsed to match either a format or a pattern.
    if is_format:
        # Formats cannot reference formats.
        if fmt:
            error(lineno, 'format referencing format')
        # If an argument set is given, then there should be no fields
        # without a place to store it.
        if arg:
            for f in flds.keys():
                if f not in arg.fields:
                    error(lineno, f'field {f} not in argument set {arg.name}')
        else:
            arg = infer_argument_set(flds)
        if name in formats:
            error(lineno, 'duplicate format name', name)
        fmt = Format(name, lineno, arg, fixedbits, fixedmask,
                     undefmask, fieldmask, flds, width)
        formats[name] = fmt
    else:
        # Patterns can reference a format ...
        if fmt:
            # ... but not an argument simultaneously
            if arg:
                error(lineno, 'pattern specifies both format and argument set')
            if fixedmask & fmt.fixedmask:
                error(lineno, 'pattern fixed bits overlap format fixed bits')
            if width != fmt.width:
                error(lineno, 'pattern uses format of different width')
            fieldmask |= fmt.fieldmask
            fixedbits |= fmt.fixedbits
            fixedmask |= fmt.fixedmask
            undefmask |= fmt.undefmask
        else:
            (fmt, flds) = infer_format(arg, fieldmask, flds, width)
        arg = fmt.base
        for f in flds.keys():
            if f not in arg.fields:
                error(lineno, f'field {f} not in argument set {arg.name}')
            if f in fmt.fields.keys():
                error(lineno, f'field {f} set by format and pattern')
        for f in arg.fields:
            if f not in flds.keys() and f not in fmt.fields.keys():
                error(lineno, f'field {f} not initialized')
        pat = Pattern(name, lineno, fmt, fixedbits, fixedmask,
                      undefmask, fieldmask, flds, width)
        parent_pat.pats.append(pat)
        allpatterns.append(pat)

    # Validate the masks that we have assembled.
    if fieldmask & fixedmask:
        error(lineno, 'fieldmask overlaps fixedmask ',
              f'({whex(fieldmask)} & {whex(fixedmask)})')
    if fieldmask & undefmask:
        error(lineno, 'fieldmask overlaps undefmask ',
              f'({whex(fieldmask)} & {whex(undefmask)})')
    if fixedmask & undefmask:
        error(lineno, 'fixedmask overlaps undefmask ',
              f'({whex(fixedmask)} & {whex(undefmask)})')
    if not is_format:
        allbits = fieldmask | fixedmask | undefmask
        if allbits != insnmask:
            error(lineno, 'bits left unspecified ',
                  f'({whex(allbits ^ insnmask)})')
# end parse_general


def parse_file(f, parent_pat):
    """Parse all of the patterns within a file"""
    global re_arg_ident
    global re_fld_ident
    global re_fmt_ident
    global re_pat_ident

    # Read all of the lines of the file.  Concatenate lines
    # ending in backslash; discard empty lines and comments.
    toks = []
    lineno = 0
    nesting = 0
    nesting_pats = []

    for line in f:
        lineno += 1

        # Expand and strip spaces, to find indent.
        line = line.rstrip()
        line = line.expandtabs()
        len1 = len(line)
        line = line.lstrip()
        len2 = len(line)

        # Discard comments
        end = line.find('#')
        if end >= 0:
            line = line[:end]

        t = line.split()
        if len(toks) != 0:
            # Next line after continuation
            toks.extend(t)
        else:
            # Allow completely blank lines.
            if len1 == 0:
                continue
            indent = len1 - len2
            # Empty line due to comment.
            if len(t) == 0:
                # Indentation must be correct, even for comment lines.
                if indent != nesting:
                    error(lineno, 'indentation ', indent, ' != ', nesting)
                continue
            start_lineno = lineno
            toks = t

        # Continuation?
        if toks[-1] == '\\':
            toks.pop()
            continue

        name = toks[0]
        del toks[0]

        # End nesting?
        if name == '}' or name == ']':
            if len(toks) != 0:
                error(start_lineno, 'extra tokens after close brace')

            # Make sure { } and [ ] nest properly.
            if (name == '}') != isinstance(parent_pat, IncMultiPattern):
                error(lineno, 'mismatched close brace')

            try:
                parent_pat = nesting_pats.pop()
            except:
                error(lineno, 'extra close brace')

            nesting -= 2
            if indent != nesting:
                error(lineno, 'indentation ', indent, ' != ', nesting)

            toks = []
            continue

        # Everything else should have current indentation.
        if indent != nesting:
            error(start_lineno, 'indentation ', indent, ' != ', nesting)

        # Start nesting?
        if name == '{' or name == '[':
            if len(toks) != 0:
                error(start_lineno, 'extra tokens after open brace')

            if name == '{':
                nested_pat = IncMultiPattern(start_lineno)
            else:
                nested_pat = ExcMultiPattern(start_lineno)
            parent_pat.pats.append(nested_pat)
            nesting_pats.append(parent_pat)
            parent_pat = nested_pat

            nesting += 2
            toks = []
            continue

        # Determine the type of object needing to be parsed.
        if re.fullmatch(re_fld_ident, name):
            parse_field(start_lineno, name[1:], toks)
        elif re.fullmatch(re_arg_ident, name):
            parse_arguments(start_lineno, name[1:], toks)
        elif re.fullmatch(re_fmt_ident, name):
            parse_generic(start_lineno, None, name[1:], toks)
        elif re.fullmatch(re_pat_ident, name):
            parse_generic(start_lineno, parent_pat, name, toks)
        else:
            error(lineno, f'invalid token "{name}"')
        toks = []

    if nesting != 0:
        error(lineno, 'missing close brace')
# end parse_file


class SizeTree:
    """Class representing a node in a size decode tree"""

    def __init__(self, m, w):
        self.mask = m
        self.subs = []
        self.base = None
        self.width = w

    def str1(self, i):
        ind = str_indent(i)
        r = ind + whex(self.mask) + ' [\n'
        for (b, s) in self.subs:
            r += ind + f'  {whex(b)}:\n'
            r += s.str1(i + 4) + '\n'
        r += ind + ']'
        return r

    def __str__(self):
        return self.str1(0)

    def output_code(self, i, extracted, outerbits, outermask):
        ind = str_indent(i)

        # If we need to load more bytes to test, do so now.
        if extracted < self.width:
            output(ind, f'insn = {decode_function}_load_bytes',
                   f'(ctx, insn, {extracted // 8}, {self.width // 8});\n')
            extracted = self.width

        # Attempt to aid the compiler in producing compact switch statements.
        # If the bits in the mask are contiguous, extract them.
        sh = is_contiguous(self.mask)
        if sh > 0:
            # Propagate SH down into the local functions.
            def str_switch(b, sh=sh):
                return f'(insn >> {sh}) & {b >> sh:#x}'

            def str_case(b, sh=sh):
                return hex(b >> sh)
        else:
            def str_switch(b):
                return f'insn & {whexC(b)}'

            def str_case(b):
                return whexC(b)

        output(ind, 'switch (', str_switch(self.mask), ') {\n')
        for b, s in sorted(self.subs):
            innermask = outermask | self.mask
            innerbits = outerbits | b
            output(ind, 'case ', str_case(b), ':\n')
            output(ind, '    /* ',
                   str_match_bits(innerbits, innermask), ' */\n')
            s.output_code(i + 4, extracted, innerbits, innermask)
        output(ind, '}\n')
        output(ind, 'return insn;\n')
# end SizeTree

class SizeLeaf:
    """Class representing a leaf node in a size decode tree"""

    def __init__(self, m, w):
        self.mask = m
        self.width = w

    def str1(self, i):
        return str_indent(i) + whex(self.mask)

    def __str__(self):
        return self.str1(0)

    def output_code(self, i, extracted, outerbits, outermask):
        global decode_function
        ind = str_indent(i)

        # If we need to load more bytes, do so now.
        if extracted < self.width:
            output(ind, f'insn = {decode_function}_load_bytes',
                   f'(ctx, insn, {extracted // 8}, {self.width // 8});\n')
            extracted = self.width
        output(ind, 'return insn;\n')
# end SizeLeaf


def build_size_tree(pats, width, outerbits, outermask):
    global insnwidth

    # Collect the mask of bits that are fixed in this width
    innermask = 0xff << (insnwidth - width)
    innermask &= ~outermask
    minwidth = None
    onewidth = True
    for i in pats:
        innermask &= i.fixedmask
        if minwidth is None:
            minwidth = i.width
        elif minwidth != i.width:
            onewidth = False;
            if minwidth < i.width:
                minwidth = i.width

    if onewidth:
        return SizeLeaf(innermask, minwidth)

    if innermask == 0:
        if width < minwidth:
            return build_size_tree(pats, width + 8, outerbits, outermask)

        pnames = []
        for p in pats:
            pnames.append(p.name + ':' + p.file + ':' + str(p.lineno))
        error_with_file(pats[0].file, pats[0].lineno,
                        f'overlapping patterns size {width}:', pnames)

    bins = {}
    for i in pats:
        fb = i.fixedbits & innermask
        if fb in bins:
            bins[fb].append(i)
        else:
            bins[fb] = [i]

    fullmask = outermask | innermask
    lens = sorted(bins.keys())
    if len(lens) == 1:
        b = lens[0]
        return build_size_tree(bins[b], width + 8, b | outerbits, fullmask)

    r = SizeTree(innermask, width)
    for b, l in bins.items():
        s = build_size_tree(l, width, b | outerbits, fullmask)
        r.subs.append((b, s))
    return r
# end build_size_tree


def prop_size(tree):
    """Propagate minimum widths up the decode size tree"""

    if isinstance(tree, SizeTree):
        min = None
        for (b, s) in tree.subs:
            width = prop_size(s)
            if min is None or min > width:
                min = width
        assert min >= tree.width
        tree.width = min
    else:
        min = tree.width
    return min
# end prop_size


def main():
    global arguments
    global formats
    global allpatterns
    global translate_scope
    global translate_prefix
    global output_fd
    global output_file
    global input_file
    global insnwidth
    global insntype
    global insnmask
    global decode_function
    global bitop_width
    global variablewidth
    global anyextern

    decode_scope = 'static '

    long_opts = ['decode=', 'translate=', 'output=', 'insnwidth=',
                 'static-decode=', 'varinsnwidth=']
    try:
        (opts, args) = getopt.gnu_getopt(sys.argv[1:], 'o:vw:', long_opts)
    except getopt.GetoptError as err:
        error(0, err)
    for o, a in opts:
        if o in ('-o', '--output'):
            output_file = a
        elif o == '--decode':
            decode_function = a
            decode_scope = ''
        elif o == '--static-decode':
            decode_function = a
        elif o == '--translate':
            translate_prefix = a
            translate_scope = ''
        elif o in ('-w', '--insnwidth', '--varinsnwidth'):
            if o == '--varinsnwidth':
                variablewidth = True
            insnwidth = int(a)
            if insnwidth == 16:
                insntype = 'uint16_t'
                insnmask = 0xffff
            elif insnwidth == 64:
                insntype = 'uint64_t'
                insnmask = 0xffffffffffffffff
                bitop_width = 64
            elif insnwidth != 32:
                error(0, 'cannot handle insns of width', insnwidth)
        else:
            assert False, 'unhandled option'

    if len(args) < 1:
        error(0, 'missing input file')

    toppat = ExcMultiPattern(0)

    for filename in args:
        input_file = filename
        f = open(filename, 'rt', encoding='utf-8')
        parse_file(f, toppat)
        f.close()

    # We do not want to compute masks for toppat, because those masks
    # are used as a starting point for build_tree.  For toppat, we must
    # insist that decode begins from naught.
    for i in toppat.pats:
        i.prop_masks()

    toppat.build_tree()
    toppat.prop_format()

    if variablewidth:
        for i in toppat.pats:
            i.prop_width()
        stree = build_size_tree(toppat.pats, 8, 0, 0)
        prop_size(stree)

    if output_file:
        output_fd = open(output_file, 'wt', encoding='utf-8')
    else:
        output_fd = io.TextIOWrapper(sys.stdout.buffer,
                                     encoding=sys.stdout.encoding,
                                     errors="ignore")

    output_autogen()
    for n in sorted(arguments.keys()):
        f = arguments[n]
        f.output_def()

    # A single translate function can be invoked for different patterns.
    # Make sure that the argument sets are the same, and declare the
    # function only once.
    #
    # If we're sharing formats, we're likely also sharing trans_* functions,
    # but we can't tell which ones.  Prevent issues from the compiler by
    # suppressing redundant declaration warnings.
    if anyextern:
        output("#pragma GCC diagnostic push\n",