diff options
Diffstat (limited to 'contrib/binutils/opcodes/cgen-asm.in')
-rw-r--r-- | contrib/binutils/opcodes/cgen-asm.in | 168 |
1 files changed, 151 insertions, 17 deletions
diff --git a/contrib/binutils/opcodes/cgen-asm.in b/contrib/binutils/opcodes/cgen-asm.in index 7249708..475a4f1 100644 --- a/contrib/binutils/opcodes/cgen-asm.in +++ b/contrib/binutils/opcodes/cgen-asm.in @@ -26,7 +26,6 @@ along with this program; if not, write to the Free Software Foundation, Inc., Keep that in mind. */ #include "sysdep.h" -#include <ctype.h> #include <stdio.h> #include "ansidecl.h" #include "bfd.h" @@ -34,16 +33,142 @@ along with this program; if not, write to the Free Software Foundation, Inc., #include "@prefix@-desc.h" #include "@prefix@-opc.h" #include "opintl.h" +#include "xregex.h" +#include "libiberty.h" +#include "safe-ctype.h" -#undef min +#undef min #define min(a,b) ((a) < (b) ? (a) : (b)) -#undef max +#undef max #define max(a,b) ((a) > (b) ? (a) : (b)) static const char * parse_insn_normal PARAMS ((CGEN_CPU_DESC, const CGEN_INSN *, const char **, CGEN_FIELDS *)); -/* -- assembler routines inserted here */ +/* -- assembler routines inserted here. */ + + +/* Regex construction routine. + + This translates an opcode syntax string into a regex string, + by replacing any non-character syntax element (such as an + opcode) with the pattern '.*' + + It then compiles the regex and stores it in the opcode, for + later use by @arch@_cgen_assemble_insn + + Returns NULL for success, an error message for failure. */ + +char * +@arch@_cgen_build_insn_regex (insn) + CGEN_INSN *insn; +{ + CGEN_OPCODE *opc = (CGEN_OPCODE *) CGEN_INSN_OPCODE (insn); + const char *mnem = CGEN_INSN_MNEMONIC (insn); + char rxbuf[CGEN_MAX_RX_ELEMENTS]; + char *rx = rxbuf; + const CGEN_SYNTAX_CHAR_TYPE *syn; + int reg_err; + + syn = CGEN_SYNTAX_STRING (CGEN_OPCODE_SYNTAX (opc)); + + /* Mnemonics come first in the syntax string. */ + if (! CGEN_SYNTAX_MNEMONIC_P (* syn)) + return _("missing mnemonic in syntax string"); + ++syn; + + /* Generate a case sensitive regular expression that emulates case + insensitive matching in the "C" locale. We cannot generate a case + insensitive regular expression because in Turkish locales, 'i' and 'I' + are not equal modulo case conversion. */ + + /* Copy the literal mnemonic out of the insn. */ + for (; *mnem; mnem++) + { + char c = *mnem; + + if (ISALPHA (c)) + { + *rx++ = '['; + *rx++ = TOLOWER (c); + *rx++ = TOUPPER (c); + *rx++ = ']'; + } + else + *rx++ = c; + } + + /* Copy any remaining literals from the syntax string into the rx. */ + for(; * syn != 0 && rx <= rxbuf + (CGEN_MAX_RX_ELEMENTS - 7 - 4); ++syn) + { + if (CGEN_SYNTAX_CHAR_P (* syn)) + { + char c = CGEN_SYNTAX_CHAR (* syn); + + switch (c) + { + /* Escape any regex metacharacters in the syntax. */ + case '.': case '[': case '\\': + case '*': case '^': case '$': + +#ifdef CGEN_ESCAPE_EXTENDED_REGEX + case '?': case '{': case '}': + case '(': case ')': case '*': + case '|': case '+': case ']': +#endif + *rx++ = '\\'; + *rx++ = c; + break; + + default: + if (ISALPHA (c)) + { + *rx++ = '['; + *rx++ = TOLOWER (c); + *rx++ = TOUPPER (c); + *rx++ = ']'; + } + else + *rx++ = c; + break; + } + } + else + { + /* Replace non-syntax fields with globs. */ + *rx++ = '.'; + *rx++ = '*'; + } + } + + /* Trailing whitespace ok. */ + * rx++ = '['; + * rx++ = ' '; + * rx++ = '\t'; + * rx++ = ']'; + * rx++ = '*'; + + /* But anchor it after that. */ + * rx++ = '$'; + * rx = '\0'; + + CGEN_INSN_RX (insn) = xmalloc (sizeof (regex_t)); + reg_err = regcomp ((regex_t *) CGEN_INSN_RX (insn), rxbuf, REG_NOSUB); + + if (reg_err == 0) + return NULL; + else + { + static char msg[80]; + + regerror (reg_err, (regex_t *) CGEN_INSN_RX (insn), msg, 80); + regfree ((regex_t *) CGEN_INSN_RX (insn)); + free (CGEN_INSN_RX (insn)); + (CGEN_INSN_RX (insn)) = NULL; + return msg; + } +} + /* Default insn parser. @@ -56,8 +181,7 @@ static const char * parse_insn_normal but that can be handled there. Not handling backtracking here may get expensive in the case of the m68k. Deal with later. - Returns NULL for success, an error message for failure. -*/ + Returns NULL for success, an error message for failure. */ static const char * parse_insn_normal (cd, insn, strp, fields) @@ -82,14 +206,14 @@ parse_insn_normal (cd, insn, strp, fields) GAS's input scrubber will ensure mnemonics are lowercase, but we may not be called from GAS. */ p = CGEN_INSN_MNEMONIC (insn); - while (*p && tolower (*p) == tolower (*str)) + while (*p && TOLOWER (*p) == TOLOWER (*str)) ++p, ++str; if (* p) return _("unrecognized instruction"); #ifndef CGEN_MNEMONIC_OPERANDS - if (* str && !isspace (* str)) + if (* str && ! ISSPACE (* str)) return _("unrecognized instruction"); #endif @@ -118,7 +242,7 @@ parse_insn_normal (cd, insn, strp, fields) first char after the mnemonic part is a space. */ /* FIXME: We also take inappropriate advantage of the fact that GAS's input scrubber will remove extraneous blanks. */ - if (tolower (*str) == tolower (CGEN_SYNTAX_CHAR (* syn))) + if (TOLOWER (*str) == TOLOWER (CGEN_SYNTAX_CHAR (* syn))) { #ifdef CGEN_MNEMONIC_OPERANDS if (CGEN_SYNTAX_CHAR(* syn) == ' ') @@ -131,6 +255,7 @@ parse_insn_normal (cd, insn, strp, fields) { /* Syntax char didn't match. Can't be this insn. */ static char msg [80]; + /* xgettext:c-format */ sprintf (msg, _("syntax error (expected char `%c', found `%c')"), CGEN_SYNTAX_CHAR(*syn), *str); @@ -140,6 +265,7 @@ parse_insn_normal (cd, insn, strp, fields) { /* Ran out of input. */ static char msg [80]; + /* xgettext:c-format */ sprintf (msg, _("syntax error (expected char `%c', found end of instruction)"), CGEN_SYNTAX_CHAR(*syn)); @@ -165,7 +291,7 @@ parse_insn_normal (cd, insn, strp, fields) blanks now. IE: We needn't try again with a longer version of the insn and it is assumed that longer versions of insns appear before shorter ones (eg: lsr r2,r3,1 vs lsr r2,r3). */ - while (isspace (* str)) + while (ISSPACE (* str)) ++ str; if (* str != '\0') @@ -211,9 +337,10 @@ const CGEN_INSN * CGEN_INSN_LIST *ilist; const char *parse_errmsg = NULL; const char *insert_errmsg = NULL; + int recognized_mnemonic = 0; /* Skip leading white space. */ - while (isspace (* str)) + while (ISSPACE (* str)) ++ str; /* The instructions are stored in hashed lists. @@ -221,19 +348,19 @@ const CGEN_INSN * ilist = CGEN_ASM_LOOKUP_INSN (cd, str); /* Keep looking until we find a match. */ - start = str; for ( ; ilist != NULL ; ilist = CGEN_ASM_NEXT_INSN (ilist)) { const CGEN_INSN *insn = ilist->insn; + recognized_mnemonic = 1; #ifdef CGEN_VALIDATE_INSN_SUPPORTED - /* not usually needed as unsupported opcodes shouldn't be in the hash lists */ + /* Not usually needed as unsupported opcodes + shouldn't be in the hash lists. */ /* Is this insn supported by the selected cpu? */ if (! @arch@_cgen_insn_supported (cd, insn)) continue; #endif - /* If the RELAX attribute is set, this is an insn that shouldn't be chosen immediately. Instead, it is used during assembler/linker relaxation if possible. */ @@ -242,6 +369,11 @@ const CGEN_INSN * str = start; + /* Skip this insn if str doesn't look right lexically. */ + if (CGEN_INSN_RX (insn) != NULL && + regexec ((regex_t *) CGEN_INSN_RX (insn), str, 0, NULL, 0) == REG_NOMATCH) + continue; + /* Allow parse/insert handlers to obtain length of insn. */ CGEN_FIELDS_BITSIZE (fields) = CGEN_INSN_BITSIZE (insn); @@ -249,7 +381,7 @@ const CGEN_INSN * if (parse_errmsg != NULL) continue; - /* ??? 0 is passed for `pc' */ + /* ??? 0 is passed for `pc'. */ insert_errmsg = CGEN_INSERT_FN (cd, insn) (cd, insn, fields, buf, (bfd_vma) 0); if (insert_errmsg != NULL) @@ -262,13 +394,15 @@ const CGEN_INSN * { static char errbuf[150]; +#ifdef CGEN_VERBOSE_ASSEMBLER_ERRORS const char *tmp_errmsg; -#ifdef CGEN_VERBOSE_ASSEMBLER_ERRORS /* If requesting verbose error messages, use insert_errmsg. - Failing that, use parse_errmsg */ + Failing that, use parse_errmsg. */ tmp_errmsg = (insert_errmsg ? insert_errmsg : parse_errmsg ? parse_errmsg : + recognized_mnemonic ? + _("unrecognized form of instruction") : _("unrecognized instruction")); if (strlen (start) > 50) |