diff options
Diffstat (limited to 'scripts/genksyms')
-rw-r--r-- | scripts/genksyms/.gitignore | 2 | ||||
-rw-r--r-- | scripts/genksyms/Makefile | 30 | ||||
-rw-r--r-- | scripts/genksyms/genksyms.c | 863 | ||||
-rw-r--r-- | scripts/genksyms/genksyms.h | 83 | ||||
-rw-r--r-- | scripts/genksyms/keywords.c | 78 | ||||
-rw-r--r-- | scripts/genksyms/lex.l | 440 | ||||
-rw-r--r-- | scripts/genksyms/parse.y | 510 |
7 files changed, 2006 insertions, 0 deletions
diff --git a/scripts/genksyms/.gitignore b/scripts/genksyms/.gitignore new file mode 100644 index 0000000000..0b275abf94 --- /dev/null +++ b/scripts/genksyms/.gitignore @@ -0,0 +1,2 @@ +# SPDX-License-Identifier: GPL-2.0-only +/genksyms diff --git a/scripts/genksyms/Makefile b/scripts/genksyms/Makefile new file mode 100644 index 0000000000..d6a422a63b --- /dev/null +++ b/scripts/genksyms/Makefile @@ -0,0 +1,30 @@ +# SPDX-License-Identifier: GPL-2.0 + +hostprogs-always-y += genksyms + +genksyms-objs := genksyms.o parse.tab.o lex.lex.o + +# FIXME: fix the ambiguous grammar in parse.y and delete this hack +# +# Suppress shift/reduce, reduce/reduce conflicts warnings +# unless W=1 is specified. +# +# Just in case, run "$(YACC) --version" without suppressing stderr +# so that 'bison: not found' will be displayed if it is missing. +ifeq ($(findstring 1,$(KBUILD_EXTRA_WARN)),) + +quiet_cmd_bison_no_warn = $(quiet_cmd_bison) + cmd_bison_no_warn = $(YACC) --version >/dev/null; \ + $(cmd_bison) 2>/dev/null + +$(obj)/pars%.tab.c $(obj)/pars%.tab.h: $(src)/pars%.y FORCE + $(call if_changed,bison_no_warn) + +endif + +# -I needed for generated C source to include headers in source tree +HOSTCFLAGS_parse.tab.o := -I $(srctree)/$(src) +HOSTCFLAGS_lex.lex.o := -I $(srctree)/$(src) + +# dependencies on generated files need to be listed explicitly +$(obj)/lex.lex.o: $(obj)/parse.tab.h diff --git a/scripts/genksyms/genksyms.c b/scripts/genksyms/genksyms.c new file mode 100644 index 0000000000..f5dfdb9d80 --- /dev/null +++ b/scripts/genksyms/genksyms.c @@ -0,0 +1,863 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* Generate kernel symbol version hashes. + Copyright 1996, 1997 Linux International. + + New implementation contributed by Richard Henderson <rth@tamu.edu> + Based on original work by Bjorn Ekwall <bj0rn@blox.se> + + This file was part of the Linux modutils 2.4.22: moved back into the + kernel sources by Rusty Russell/Kai Germaschewski. + + */ + +#include <stdio.h> +#include <string.h> +#include <stdlib.h> +#include <unistd.h> +#include <assert.h> +#include <stdarg.h> +#ifdef __GNU_LIBRARY__ +#include <getopt.h> +#endif /* __GNU_LIBRARY__ */ + +#include "genksyms.h" +/*----------------------------------------------------------------------*/ + +#define HASH_BUCKETS 4096 + +static struct symbol *symtab[HASH_BUCKETS]; +static FILE *debugfile; + +int cur_line = 1; +char *cur_filename; +int in_source_file; + +static int flag_debug, flag_dump_defs, flag_reference, flag_dump_types, + flag_preserve, flag_warnings; + +static int errors; +static int nsyms; + +static struct symbol *expansion_trail; +static struct symbol *visited_symbols; + +static const struct { + int n; + const char *name; +} symbol_types[] = { + [SYM_NORMAL] = { 0, NULL}, + [SYM_TYPEDEF] = {'t', "typedef"}, + [SYM_ENUM] = {'e', "enum"}, + [SYM_STRUCT] = {'s', "struct"}, + [SYM_UNION] = {'u', "union"}, + [SYM_ENUM_CONST] = {'E', "enum constant"}, +}; + +static int equal_list(struct string_list *a, struct string_list *b); +static void print_list(FILE * f, struct string_list *list); +static struct string_list *concat_list(struct string_list *start, ...); +static struct string_list *mk_node(const char *string); +static void print_location(void); +static void print_type_name(enum symbol_type type, const char *name); + +/*----------------------------------------------------------------------*/ + +static const unsigned int crctab32[] = { + 0x00000000U, 0x77073096U, 0xee0e612cU, 0x990951baU, 0x076dc419U, + 0x706af48fU, 0xe963a535U, 0x9e6495a3U, 0x0edb8832U, 0x79dcb8a4U, + 0xe0d5e91eU, 0x97d2d988U, 0x09b64c2bU, 0x7eb17cbdU, 0xe7b82d07U, + 0x90bf1d91U, 0x1db71064U, 0x6ab020f2U, 0xf3b97148U, 0x84be41deU, + 0x1adad47dU, 0x6ddde4ebU, 0xf4d4b551U, 0x83d385c7U, 0x136c9856U, + 0x646ba8c0U, 0xfd62f97aU, 0x8a65c9ecU, 0x14015c4fU, 0x63066cd9U, + 0xfa0f3d63U, 0x8d080df5U, 0x3b6e20c8U, 0x4c69105eU, 0xd56041e4U, + 0xa2677172U, 0x3c03e4d1U, 0x4b04d447U, 0xd20d85fdU, 0xa50ab56bU, + 0x35b5a8faU, 0x42b2986cU, 0xdbbbc9d6U, 0xacbcf940U, 0x32d86ce3U, + 0x45df5c75U, 0xdcd60dcfU, 0xabd13d59U, 0x26d930acU, 0x51de003aU, + 0xc8d75180U, 0xbfd06116U, 0x21b4f4b5U, 0x56b3c423U, 0xcfba9599U, + 0xb8bda50fU, 0x2802b89eU, 0x5f058808U, 0xc60cd9b2U, 0xb10be924U, + 0x2f6f7c87U, 0x58684c11U, 0xc1611dabU, 0xb6662d3dU, 0x76dc4190U, + 0x01db7106U, 0x98d220bcU, 0xefd5102aU, 0x71b18589U, 0x06b6b51fU, + 0x9fbfe4a5U, 0xe8b8d433U, 0x7807c9a2U, 0x0f00f934U, 0x9609a88eU, + 0xe10e9818U, 0x7f6a0dbbU, 0x086d3d2dU, 0x91646c97U, 0xe6635c01U, + 0x6b6b51f4U, 0x1c6c6162U, 0x856530d8U, 0xf262004eU, 0x6c0695edU, + 0x1b01a57bU, 0x8208f4c1U, 0xf50fc457U, 0x65b0d9c6U, 0x12b7e950U, + 0x8bbeb8eaU, 0xfcb9887cU, 0x62dd1ddfU, 0x15da2d49U, 0x8cd37cf3U, + 0xfbd44c65U, 0x4db26158U, 0x3ab551ceU, 0xa3bc0074U, 0xd4bb30e2U, + 0x4adfa541U, 0x3dd895d7U, 0xa4d1c46dU, 0xd3d6f4fbU, 0x4369e96aU, + 0x346ed9fcU, 0xad678846U, 0xda60b8d0U, 0x44042d73U, 0x33031de5U, + 0xaa0a4c5fU, 0xdd0d7cc9U, 0x5005713cU, 0x270241aaU, 0xbe0b1010U, + 0xc90c2086U, 0x5768b525U, 0x206f85b3U, 0xb966d409U, 0xce61e49fU, + 0x5edef90eU, 0x29d9c998U, 0xb0d09822U, 0xc7d7a8b4U, 0x59b33d17U, + 0x2eb40d81U, 0xb7bd5c3bU, 0xc0ba6cadU, 0xedb88320U, 0x9abfb3b6U, + 0x03b6e20cU, 0x74b1d29aU, 0xead54739U, 0x9dd277afU, 0x04db2615U, + 0x73dc1683U, 0xe3630b12U, 0x94643b84U, 0x0d6d6a3eU, 0x7a6a5aa8U, + 0xe40ecf0bU, 0x9309ff9dU, 0x0a00ae27U, 0x7d079eb1U, 0xf00f9344U, + 0x8708a3d2U, 0x1e01f268U, 0x6906c2feU, 0xf762575dU, 0x806567cbU, + 0x196c3671U, 0x6e6b06e7U, 0xfed41b76U, 0x89d32be0U, 0x10da7a5aU, + 0x67dd4accU, 0xf9b9df6fU, 0x8ebeeff9U, 0x17b7be43U, 0x60b08ed5U, + 0xd6d6a3e8U, 0xa1d1937eU, 0x38d8c2c4U, 0x4fdff252U, 0xd1bb67f1U, + 0xa6bc5767U, 0x3fb506ddU, 0x48b2364bU, 0xd80d2bdaU, 0xaf0a1b4cU, + 0x36034af6U, 0x41047a60U, 0xdf60efc3U, 0xa867df55U, 0x316e8eefU, + 0x4669be79U, 0xcb61b38cU, 0xbc66831aU, 0x256fd2a0U, 0x5268e236U, + 0xcc0c7795U, 0xbb0b4703U, 0x220216b9U, 0x5505262fU, 0xc5ba3bbeU, + 0xb2bd0b28U, 0x2bb45a92U, 0x5cb36a04U, 0xc2d7ffa7U, 0xb5d0cf31U, + 0x2cd99e8bU, 0x5bdeae1dU, 0x9b64c2b0U, 0xec63f226U, 0x756aa39cU, + 0x026d930aU, 0x9c0906a9U, 0xeb0e363fU, 0x72076785U, 0x05005713U, + 0x95bf4a82U, 0xe2b87a14U, 0x7bb12baeU, 0x0cb61b38U, 0x92d28e9bU, + 0xe5d5be0dU, 0x7cdcefb7U, 0x0bdbdf21U, 0x86d3d2d4U, 0xf1d4e242U, + 0x68ddb3f8U, 0x1fda836eU, 0x81be16cdU, 0xf6b9265bU, 0x6fb077e1U, + 0x18b74777U, 0x88085ae6U, 0xff0f6a70U, 0x66063bcaU, 0x11010b5cU, + 0x8f659effU, 0xf862ae69U, 0x616bffd3U, 0x166ccf45U, 0xa00ae278U, + 0xd70dd2eeU, 0x4e048354U, 0x3903b3c2U, 0xa7672661U, 0xd06016f7U, + 0x4969474dU, 0x3e6e77dbU, 0xaed16a4aU, 0xd9d65adcU, 0x40df0b66U, + 0x37d83bf0U, 0xa9bcae53U, 0xdebb9ec5U, 0x47b2cf7fU, 0x30b5ffe9U, + 0xbdbdf21cU, 0xcabac28aU, 0x53b39330U, 0x24b4a3a6U, 0xbad03605U, + 0xcdd70693U, 0x54de5729U, 0x23d967bfU, 0xb3667a2eU, 0xc4614ab8U, + 0x5d681b02U, 0x2a6f2b94U, 0xb40bbe37U, 0xc30c8ea1U, 0x5a05df1bU, + 0x2d02ef8dU +}; + +static unsigned long partial_crc32_one(unsigned char c, unsigned long crc) +{ + return crctab32[(crc ^ c) & 0xff] ^ (crc >> 8); +} + +static unsigned long partial_crc32(const char *s, unsigned long crc) +{ + while (*s) + crc = partial_crc32_one(*s++, crc); + return crc; +} + +static unsigned long crc32(const char *s) +{ + return partial_crc32(s, 0xffffffff) ^ 0xffffffff; +} + +/*----------------------------------------------------------------------*/ + +static enum symbol_type map_to_ns(enum symbol_type t) +{ + switch (t) { + case SYM_ENUM_CONST: + case SYM_NORMAL: + case SYM_TYPEDEF: + return SYM_NORMAL; + case SYM_ENUM: + case SYM_STRUCT: + case SYM_UNION: + return SYM_STRUCT; + } + return t; +} + +struct symbol *find_symbol(const char *name, enum symbol_type ns, int exact) +{ + unsigned long h = crc32(name) % HASH_BUCKETS; + struct symbol *sym; + + for (sym = symtab[h]; sym; sym = sym->hash_next) + if (map_to_ns(sym->type) == map_to_ns(ns) && + strcmp(name, sym->name) == 0 && + sym->is_declared) + break; + + if (exact && sym && sym->type != ns) + return NULL; + return sym; +} + +static int is_unknown_symbol(struct symbol *sym) +{ + struct string_list *defn; + + return ((sym->type == SYM_STRUCT || + sym->type == SYM_UNION || + sym->type == SYM_ENUM) && + (defn = sym->defn) && defn->tag == SYM_NORMAL && + strcmp(defn->string, "}") == 0 && + (defn = defn->next) && defn->tag == SYM_NORMAL && + strcmp(defn->string, "UNKNOWN") == 0 && + (defn = defn->next) && defn->tag == SYM_NORMAL && + strcmp(defn->string, "{") == 0); +} + +static struct symbol *__add_symbol(const char *name, enum symbol_type type, + struct string_list *defn, int is_extern, + int is_reference) +{ + unsigned long h; + struct symbol *sym; + enum symbol_status status = STATUS_UNCHANGED; + /* The parser adds symbols in the order their declaration completes, + * so it is safe to store the value of the previous enum constant in + * a static variable. + */ + static int enum_counter; + static struct string_list *last_enum_expr; + + if (type == SYM_ENUM_CONST) { + if (defn) { + free_list(last_enum_expr, NULL); + last_enum_expr = copy_list_range(defn, NULL); + enum_counter = 1; + } else { + struct string_list *expr; + char buf[20]; + + snprintf(buf, sizeof(buf), "%d", enum_counter++); + if (last_enum_expr) { + expr = copy_list_range(last_enum_expr, NULL); + defn = concat_list(mk_node("("), + expr, + mk_node(")"), + mk_node("+"), + mk_node(buf), NULL); + } else { + defn = mk_node(buf); + } + } + } else if (type == SYM_ENUM) { + free_list(last_enum_expr, NULL); + last_enum_expr = NULL; + enum_counter = 0; + if (!name) + /* Anonymous enum definition, nothing more to do */ + return NULL; + } + + h = crc32(name) % HASH_BUCKETS; + for (sym = symtab[h]; sym; sym = sym->hash_next) { + if (map_to_ns(sym->type) == map_to_ns(type) && + strcmp(name, sym->name) == 0) { + if (is_reference) + /* fall through */ ; + else if (sym->type == type && + equal_list(sym->defn, defn)) { + if (!sym->is_declared && sym->is_override) { + print_location(); + print_type_name(type, name); + fprintf(stderr, " modversion is " + "unchanged\n"); + } + sym->is_declared = 1; + return sym; + } else if (!sym->is_declared) { + if (sym->is_override && flag_preserve) { + print_location(); + fprintf(stderr, "ignoring "); + print_type_name(type, name); + fprintf(stderr, " modversion change\n"); + sym->is_declared = 1; + return sym; + } else { + status = is_unknown_symbol(sym) ? + STATUS_DEFINED : STATUS_MODIFIED; + } + } else { + error_with_pos("redefinition of %s", name); + return sym; + } + break; + } + } + + if (sym) { + struct symbol **psym; + + for (psym = &symtab[h]; *psym; psym = &(*psym)->hash_next) { + if (*psym == sym) { + *psym = sym->hash_next; + break; + } + } + --nsyms; + } + + sym = xmalloc(sizeof(*sym)); + sym->name = name; + sym->type = type; + sym->defn = defn; + sym->expansion_trail = NULL; + sym->visited = NULL; + sym->is_extern = is_extern; + + sym->hash_next = symtab[h]; + symtab[h] = sym; + + sym->is_declared = !is_reference; + sym->status = status; + sym->is_override = 0; + + if (flag_debug) { + if (symbol_types[type].name) + fprintf(debugfile, "Defn for %s %s == <", + symbol_types[type].name, name); + else + fprintf(debugfile, "Defn for type%d %s == <", + type, name); + if (is_extern) + fputs("extern ", debugfile); + print_list(debugfile, defn); + fputs(">\n", debugfile); + } + + ++nsyms; + return sym; +} + +struct symbol *add_symbol(const char *name, enum symbol_type type, + struct string_list *defn, int is_extern) +{ + return __add_symbol(name, type, defn, is_extern, 0); +} + +static struct symbol *add_reference_symbol(const char *name, enum symbol_type type, + struct string_list *defn, int is_extern) +{ + return __add_symbol(name, type, defn, is_extern, 1); +} + +/*----------------------------------------------------------------------*/ + +void free_node(struct string_list *node) +{ + free(node->string); + free(node); +} + +void free_list(struct string_list *s, struct string_list *e) +{ + while (s != e) { + struct string_list *next = s->next; + free_node(s); + s = next; + } +} + +static struct string_list *mk_node(const char *string) +{ + struct string_list *newnode; + + newnode = xmalloc(sizeof(*newnode)); + newnode->string = xstrdup(string); + newnode->tag = SYM_NORMAL; + newnode->next = NULL; + + return newnode; +} + +static struct string_list *concat_list(struct string_list *start, ...) +{ + va_list ap; + struct string_list *n, *n2; + + if (!start) + return NULL; + for (va_start(ap, start); (n = va_arg(ap, struct string_list *));) { + for (n2 = n; n2->next; n2 = n2->next) + ; + n2->next = start; + start = n; + } + va_end(ap); + return start; +} + +struct string_list *copy_node(struct string_list *node) +{ + struct string_list *newnode; + + newnode = xmalloc(sizeof(*newnode)); + newnode->string = xstrdup(node->string); + newnode->tag = node->tag; + + return newnode; +} + +struct string_list *copy_list_range(struct string_list *start, + struct string_list *end) +{ + struct string_list *res, *n; + + if (start == end) + return NULL; + n = res = copy_node(start); + for (start = start->next; start != end; start = start->next) { + n->next = copy_node(start); + n = n->next; + } + n->next = NULL; + return res; +} + +static int equal_list(struct string_list *a, struct string_list *b) +{ + while (a && b) { + if (a->tag != b->tag || strcmp(a->string, b->string)) + return 0; + a = a->next; + b = b->next; + } + + return !a && !b; +} + +#define ARRAY_SIZE(arr) (sizeof(arr) / sizeof((arr)[0])) + +static struct string_list *read_node(FILE *f) +{ + char buffer[256]; + struct string_list node = { + .string = buffer, + .tag = SYM_NORMAL }; + int c, in_string = 0; + + while ((c = fgetc(f)) != EOF) { + if (!in_string && c == ' ') { + if (node.string == buffer) + continue; + break; + } else if (c == '"') { + in_string = !in_string; + } else if (c == '\n') { + if (node.string == buffer) + return NULL; + ungetc(c, f); + break; + } + if (node.string >= buffer + sizeof(buffer) - 1) { + fprintf(stderr, "Token too long\n"); + exit(1); + } + *node.string++ = c; + } + if (node.string == buffer) + return NULL; + *node.string = 0; + node.string = buffer; + + if (node.string[1] == '#') { + size_t n; + + for (n = 0; n < ARRAY_SIZE(symbol_types); n++) { + if (node.string[0] == symbol_types[n].n) { + node.tag = n; + node.string += 2; + return copy_node(&node); + } + } + fprintf(stderr, "Unknown type %c\n", node.string[0]); + exit(1); + } + return copy_node(&node); +} + +static void read_reference(FILE *f) +{ + while (!feof(f)) { + struct string_list *defn = NULL; + struct string_list *sym, *def; + int is_extern = 0, is_override = 0; + struct symbol *subsym; + + sym = read_node(f); + if (sym && sym->tag == SYM_NORMAL && + !strcmp(sym->string, "override")) { + is_override = 1; + free_node(sym); + sym = read_node(f); + } + if (!sym) + continue; + def = read_node(f); + if (def && def->tag == SYM_NORMAL && + !strcmp(def->string, "extern")) { + is_extern = 1; + free_node(def); + def = read_node(f); + } + while (def) { + def->next = defn; + defn = def; + def = read_node(f); + } + subsym = add_reference_symbol(xstrdup(sym->string), sym->tag, + defn, is_extern); + subsym->is_override = is_override; + free_node(sym); + } +} + +static void print_node(FILE * f, struct string_list *list) +{ + if (symbol_types[list->tag].n) { + putc(symbol_types[list->tag].n, f); + putc('#', f); + } + fputs(list->string, f); +} + +static void print_list(FILE * f, struct string_list *list) +{ + struct string_list **e, **b; + struct string_list *tmp, **tmp2; + int elem = 1; + + if (list == NULL) { + fputs("(nil)", f); + return; + } + + tmp = list; + while ((tmp = tmp->next) != NULL) + elem++; + + b = alloca(elem * sizeof(*e)); + e = b + elem; + tmp2 = e - 1; + + (*tmp2--) = list; + while ((list = list->next) != NULL) + *(tmp2--) = list; + + while (b != e) { + print_node(f, *b++); + putc(' ', f); + } +} + +static unsigned long expand_and_crc_sym(struct symbol *sym, unsigned long crc) +{ + struct string_list *list = sym->defn; + struct string_list **e, **b; + struct string_list *tmp, **tmp2; + int elem = 1; + + if (!list) + return crc; + + tmp = list; + while ((tmp = tmp->next) != NULL) + elem++; + + b = alloca(elem * sizeof(*e)); + e = b + elem; + tmp2 = e - 1; + + *(tmp2--) = list; + while ((list = list->next) != NULL) + *(tmp2--) = list; + + while (b != e) { + struct string_list *cur; + struct symbol *subsym; + + cur = *(b++); + switch (cur->tag) { + case SYM_NORMAL: + if (flag_dump_defs) + fprintf(debugfile, "%s ", cur->string); + crc = partial_crc32(cur->string, crc); + crc = partial_crc32_one(' ', crc); + break; + + case SYM_ENUM_CONST: + case SYM_TYPEDEF: + subsym = find_symbol(cur->string, cur->tag, 0); + /* FIXME: Bad reference files can segfault here. */ + if (subsym->expansion_trail) { + if (flag_dump_defs) + fprintf(debugfile, "%s ", cur->string); + crc = partial_crc32(cur->string, crc); + crc = partial_crc32_one(' ', crc); + } else { + subsym->expansion_trail = expansion_trail; + expansion_trail = subsym; + crc = expand_and_crc_sym(subsym, crc); + } + break; + + case SYM_STRUCT: + case SYM_UNION: + case SYM_ENUM: + subsym = find_symbol(cur->string, cur->tag, 0); + if (!subsym) { + struct string_list *n; + + error_with_pos("expand undefined %s %s", + symbol_types[cur->tag].name, + cur->string); + n = concat_list(mk_node + (symbol_types[cur->tag].name), + mk_node(cur->string), + mk_node("{"), + mk_node("UNKNOWN"), + mk_node("}"), NULL); + subsym = + add_symbol(cur->string, cur->tag, n, 0); + } + if (subsym->expansion_trail) { + if (flag_dump_defs) { + fprintf(debugfile, "%s %s ", + symbol_types[cur->tag].name, + cur->string); + } + + crc = partial_crc32(symbol_types[cur->tag].name, + crc); + crc = partial_crc32_one(' ', crc); + crc = partial_crc32(cur->string, crc); + crc = partial_crc32_one(' ', crc); + } else { + subsym->expansion_trail = expansion_trail; + expansion_trail = subsym; + crc = expand_and_crc_sym(subsym, crc); + } + break; + } + } + + { + static struct symbol **end = &visited_symbols; + + if (!sym->visited) { + *end = sym; + end = &sym->visited; + sym->visited = (struct symbol *)-1L; + } + } + + return crc; +} + +void export_symbol(const char *name) +{ + struct symbol *sym; + + sym = find_symbol(name, SYM_NORMAL, 0); + if (!sym) + error_with_pos("export undefined symbol %s", name); + else { + unsigned long crc; + int has_changed = 0; + + if (flag_dump_defs) + fprintf(debugfile, "Export %s == <", name); + + expansion_trail = (struct symbol *)-1L; + + sym->expansion_trail = expansion_trail; + expansion_trail = sym; + crc = expand_and_crc_sym(sym, 0xffffffff) ^ 0xffffffff; + + sym = expansion_trail; + while (sym != (struct symbol *)-1L) { + struct symbol *n = sym->expansion_trail; + + if (sym->status != STATUS_UNCHANGED) { + if (!has_changed) { + print_location(); + fprintf(stderr, "%s: %s: modversion " + "changed because of changes " + "in ", flag_preserve ? "error" : + "warning", name); + } else + fprintf(stderr, ", "); + print_type_name(sym->type, sym->name); + if (sym->status == STATUS_DEFINED) + fprintf(stderr, " (became defined)"); + has_changed = 1; + if (flag_preserve) + errors++; + } + sym->expansion_trail = 0; + sym = n; + } + if (has_changed) + fprintf(stderr, "\n"); + + if (flag_dump_defs) + fputs(">\n", debugfile); + + printf("#SYMVER %s 0x%08lx\n", name, crc); + } +} + +/*----------------------------------------------------------------------*/ + +static void print_location(void) +{ + fprintf(stderr, "%s:%d: ", cur_filename ? : "<stdin>", cur_line); +} + +static void print_type_name(enum symbol_type type, const char *name) +{ + if (symbol_types[type].name) + fprintf(stderr, "%s %s", symbol_types[type].name, name); + else + fprintf(stderr, "%s", name); +} + +void error_with_pos(const char *fmt, ...) +{ + va_list args; + + if (flag_warnings) { + print_location(); + + va_start(args, fmt); + vfprintf(stderr, fmt, args); + va_end(args); + putc('\n', stderr); + + errors++; + } +} + +static void genksyms_usage(void) +{ + fputs("Usage:\n" "genksyms [-adDTwqhVR] > /path/to/.tmp_obj.ver\n" "\n" +#ifdef __GNU_LIBRARY__ + " -s, --symbol-prefix Select symbol prefix\n" + " -d, --debug Increment the debug level (repeatable)\n" + " -D, --dump Dump expanded symbol defs (for debugging only)\n" + " -r, --reference file Read reference symbols from a file\n" + " -T, --dump-types file Dump expanded types into file\n" + " -p, --preserve Preserve reference modversions or fail\n" + " -w, --warnings Enable warnings\n" + " -q, --quiet Disable warnings (default)\n" + " -h, --help Print this message\n" + " -V, --version Print the release version\n" +#else /* __GNU_LIBRARY__ */ + " -s Select symbol prefix\n" + " -d Increment the debug level (repeatable)\n" + " -D Dump expanded symbol defs (for debugging only)\n" + " -r file Read reference symbols from a file\n" + " -T file Dump expanded types into file\n" + " -p Preserve reference modversions or fail\n" + " -w Enable warnings\n" + " -q Disable warnings (default)\n" + " -h Print this message\n" + " -V Print the release version\n" +#endif /* __GNU_LIBRARY__ */ + , stderr); +} + +int main(int argc, char **argv) +{ + FILE *dumpfile = NULL, *ref_file = NULL; + int o; + +#ifdef __GNU_LIBRARY__ + struct option long_opts[] = { + {"debug", 0, 0, 'd'}, + {"warnings", 0, 0, 'w'}, + {"quiet", 0, 0, 'q'}, + {"dump", 0, 0, 'D'}, + {"reference", 1, 0, 'r'}, + {"dump-types", 1, 0, 'T'}, + {"preserve", 0, 0, 'p'}, + {"version", 0, 0, 'V'}, + {"help", 0, 0, 'h'}, + {0, 0, 0, 0} + }; + + while ((o = getopt_long(argc, argv, "s:dwqVDr:T:ph", + &long_opts[0], NULL)) != EOF) +#else /* __GNU_LIBRARY__ */ + while ((o = getopt(argc, argv, "s:dwqVDr:T:ph")) != EOF) +#endif /* __GNU_LIBRARY__ */ + switch (o) { + case 'd': + flag_debug++; + break; + case 'w': + flag_warnings = 1; + break; + case 'q': + flag_warnings = 0; + break; + case 'V': + fputs("genksyms version 2.5.60\n", stderr); + break; + case 'D': + flag_dump_defs = 1; + break; + case 'r': + flag_reference = 1; + ref_file = fopen(optarg, "r"); + if (!ref_file) { + perror(optarg); + return 1; + } + break; + case 'T': + flag_dump_types = 1; + dumpfile = fopen(optarg, "w"); + if (!dumpfile) { + perror(optarg); + return 1; + } + break; + case 'p': + flag_preserve = 1; + break; + case 'h': + genksyms_usage(); + return 0; + default: + genksyms_usage(); + return 1; + } + { + extern int yydebug; + extern int yy_flex_debug; + + yydebug = (flag_debug > 1); + yy_flex_debug = (flag_debug > 2); + + debugfile = stderr; + /* setlinebuf(debugfile); */ + } + + if (flag_reference) { + read_reference(ref_file); + fclose(ref_file); + } + + yyparse(); + + if (flag_dump_types && visited_symbols) { + while (visited_symbols != (struct symbol *)-1L) { + struct symbol *sym = visited_symbols; + + if (sym->is_override) + fputs("override ", dumpfile); + if (symbol_types[sym->type].n) { + putc(symbol_types[sym->type].n, dumpfile); + putc('#', dumpfile); + } + fputs(sym->name, dumpfile); + putc(' ', dumpfile); + if (sym->is_extern) + fputs("extern ", dumpfile); + print_list(dumpfile, sym->defn); + putc('\n', dumpfile); + + visited_symbols = sym->visited; + sym->visited = NULL; + } + } + + if (flag_debug) { + fprintf(debugfile, "Hash table occupancy %d/%d = %g\n", + nsyms, HASH_BUCKETS, + (double)nsyms / (double)HASH_BUCKETS); + } + + if (dumpfile) + fclose(dumpfile); + + return errors != 0; +} diff --git a/scripts/genksyms/genksyms.h b/scripts/genksyms/genksyms.h new file mode 100644 index 0000000000..21ed2ec2d9 --- /dev/null +++ b/scripts/genksyms/genksyms.h @@ -0,0 +1,83 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* Generate kernel symbol version hashes. + Copyright 1996, 1997 Linux International. + + New implementation contributed by Richard Henderson <rth@tamu.edu> + Based on original work by Bjorn Ekwall <bj0rn@blox.se> + + This file is part of the Linux modutils. + + */ + +#ifndef MODUTILS_GENKSYMS_H +#define MODUTILS_GENKSYMS_H 1 + +#include <stdio.h> + +enum symbol_type { + SYM_NORMAL, SYM_TYPEDEF, SYM_ENUM, SYM_STRUCT, SYM_UNION, + SYM_ENUM_CONST +}; + +enum symbol_status { + STATUS_UNCHANGED, STATUS_DEFINED, STATUS_MODIFIED +}; + +struct string_list { + struct string_list *next; + enum symbol_type tag; + int in_source_file; + char *string; +}; + +struct symbol { + struct symbol *hash_next; + const char *name; + enum symbol_type type; + struct string_list *defn; + struct symbol *expansion_trail; + struct symbol *visited; + int is_extern; + int is_declared; + enum symbol_status status; + int is_override; +}; + +typedef struct string_list **yystype; +#define YYSTYPE yystype + +extern int cur_line; +extern char *cur_filename; +extern int in_source_file; + +struct symbol *find_symbol(const char *name, enum symbol_type ns, int exact); +struct symbol *add_symbol(const char *name, enum symbol_type type, + struct string_list *defn, int is_extern); +void export_symbol(const char *); + +void free_node(struct string_list *list); +void free_list(struct string_list *s, struct string_list *e); +struct string_list *copy_node(struct string_list *); +struct string_list *copy_list_range(struct string_list *start, + struct string_list *end); + +int yylex(void); +int yyparse(void); + +void error_with_pos(const char *, ...) __attribute__ ((format(printf, 1, 2))); + +/*----------------------------------------------------------------------*/ +#define xmalloc(size) ({ void *__ptr = malloc(size); \ + if(!__ptr && size != 0) { \ + fprintf(stderr, "out of memory\n"); \ + exit(1); \ + } \ + __ptr; }) +#define xstrdup(str) ({ char *__str = strdup(str); \ + if (!__str) { \ + fprintf(stderr, "out of memory\n"); \ + exit(1); \ + } \ + __str; }) + +#endif /* genksyms.h */ diff --git a/scripts/genksyms/keywords.c b/scripts/genksyms/keywords.c new file mode 100644 index 0000000000..b85e0979a0 --- /dev/null +++ b/scripts/genksyms/keywords.c @@ -0,0 +1,78 @@ +// SPDX-License-Identifier: GPL-2.0-only +static struct resword { + const char *name; + int token; +} keywords[] = { + { "__GENKSYMS_EXPORT_SYMBOL", EXPORT_SYMBOL_KEYW }, + { "__asm", ASM_KEYW }, + { "__asm__", ASM_KEYW }, + { "__attribute", ATTRIBUTE_KEYW }, + { "__attribute__", ATTRIBUTE_KEYW }, + { "__const", CONST_KEYW }, + { "__const__", CONST_KEYW }, + { "__extension__", EXTENSION_KEYW }, + { "__inline", INLINE_KEYW }, + { "__inline__", INLINE_KEYW }, + { "__signed", SIGNED_KEYW }, + { "__signed__", SIGNED_KEYW }, + { "__typeof", TYPEOF_KEYW }, + { "__typeof__", TYPEOF_KEYW }, + { "__volatile", VOLATILE_KEYW }, + { "__volatile__", VOLATILE_KEYW }, + { "__builtin_va_list", VA_LIST_KEYW }, + + { "__int128", BUILTIN_INT_KEYW }, + { "__int128_t", BUILTIN_INT_KEYW }, + { "__uint128_t", BUILTIN_INT_KEYW }, + + // According to rth, c99 defines "_Bool", "__restrict", "__restrict__", "restrict". KAO + { "_Bool", BOOL_KEYW }, + { "__restrict", RESTRICT_KEYW }, + { "__restrict__", RESTRICT_KEYW }, + { "restrict", RESTRICT_KEYW }, + { "asm", ASM_KEYW }, + + // c11 keywords that can be used at module scope + { "_Static_assert", STATIC_ASSERT_KEYW }, + + // attribute commented out in modutils 2.4.2. People are using 'attribute' as a + // field name which breaks the genksyms parser. It is not a gcc keyword anyway. + // KAO. }, + // { "attribute", ATTRIBUTE_KEYW }, + + { "auto", AUTO_KEYW }, + { "char", CHAR_KEYW }, + { "const", CONST_KEYW }, + { "double", DOUBLE_KEYW }, + { "enum", ENUM_KEYW }, + { "extern", EXTERN_KEYW }, + { "float", FLOAT_KEYW }, + { "inline", INLINE_KEYW }, + { "int", INT_KEYW }, + { "long", LONG_KEYW }, + { "register", REGISTER_KEYW }, + { "short", SHORT_KEYW }, + { "signed", SIGNED_KEYW }, + { "static", STATIC_KEYW }, + { "struct", STRUCT_KEYW }, + { "typedef", TYPEDEF_KEYW }, + { "typeof", TYPEOF_KEYW }, + { "union", UNION_KEYW }, + { "unsigned", UNSIGNED_KEYW }, + { "void", VOID_KEYW }, + { "volatile", VOLATILE_KEYW }, +}; + +#define NR_KEYWORDS (sizeof(keywords)/sizeof(struct resword)) + +static int is_reserved_word(register const char *str, register unsigned int len) +{ + int i; + for (i = 0; i < NR_KEYWORDS; i++) { + struct resword *r = keywords + i; + int l = strlen(r->name); + if (len == l && !memcmp(str, r->name, len)) + return r->token; + } + return -1; +} diff --git a/scripts/genksyms/lex.l b/scripts/genksyms/lex.l new file mode 100644 index 0000000000..a4d7495eaf --- /dev/null +++ b/scripts/genksyms/lex.l @@ -0,0 +1,440 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* + * Lexical analysis for genksyms. + * Copyright 1996, 1997 Linux International. + * + * New implementation contributed by Richard Henderson <rth@tamu.edu> + * Based on original work by Bjorn Ekwall <bj0rn@blox.se> + * + * Taken from Linux modutils 2.4.22. + */ + +%{ + +#include <limits.h> +#include <stdlib.h> +#include <string.h> +#include <ctype.h> + +#include "genksyms.h" +#include "parse.tab.h" + +/* We've got a two-level lexer here. We let flex do basic tokenization + and then we categorize those basic tokens in the second stage. */ +#define YY_DECL static int yylex1(void) + +%} + +IDENT [A-Za-z_\$][A-Za-z0-9_\$]* + +O_INT 0[0-7]* +D_INT [1-9][0-9]* +X_INT 0[Xx][0-9A-Fa-f]+ +I_SUF [Uu]|[Ll]|[Uu][Ll]|[Ll][Uu] +INT ({O_INT}|{D_INT}|{X_INT}){I_SUF}? + +FRAC ([0-9]*\.[0-9]+)|([0-9]+\.) +EXP [Ee][+-]?[0-9]+ +F_SUF [FfLl] +REAL ({FRAC}{EXP}?{F_SUF}?)|([0-9]+{EXP}{F_SUF}?) + +STRING L?\"([^\\\"]*\\.)*[^\\\"]*\" +CHAR L?\'([^\\\']*\\.)*[^\\\']*\' + +MC_TOKEN ([~%^&*+=|<>/-]=)|(&&)|("||")|(->)|(<<)|(>>) + +/* We don't do multiple input files. */ +%option noyywrap + +%option noinput + +%% + + + /* Keep track of our location in the original source files. */ +^#[ \t]+{INT}[ \t]+\"[^\"\n]+\".*\n return FILENAME; +^#.*\n cur_line++; +\n cur_line++; + + /* Ignore all other whitespace. */ +[ \t\f\v\r]+ ; + + +{STRING} return STRING; +{CHAR} return CHAR; +{IDENT} return IDENT; + + /* The Pedant requires that the other C multi-character tokens be + recognized as tokens. We don't actually use them since we don't + parse expressions, but we do want whitespace to be arranged + around them properly. */ +{MC_TOKEN} return OTHER; +{INT} return INT; +{REAL} return REAL; + +"..." return DOTS; + + /* All other tokens are single characters. */ +. return yytext[0]; + + +%% + +/* Bring in the keyword recognizer. */ + +#include "keywords.c" + + +/* Macros to append to our phrase collection list. */ + +/* + * We mark any token, that that equals to a known enumerator, as + * SYM_ENUM_CONST. The parser will change this for struct and union tags later, + * the only problem is struct and union members: + * enum e { a, b }; struct s { int a, b; } + * but in this case, the only effect will be, that the ABI checksums become + * more volatile, which is acceptable. Also, such collisions are quite rare, + * so far it was only observed in include/linux/telephony.h. + */ +#define _APP(T,L) do { \ + cur_node = next_node; \ + next_node = xmalloc(sizeof(*next_node)); \ + next_node->next = cur_node; \ + cur_node->string = memcpy(xmalloc(L+1), T, L+1); \ + cur_node->tag = \ + find_symbol(cur_node->string, SYM_ENUM_CONST, 1)?\ + SYM_ENUM_CONST : SYM_NORMAL ; \ + cur_node->in_source_file = in_source_file; \ + } while (0) + +#define APP _APP(yytext, yyleng) + + +/* The second stage lexer. Here we incorporate knowledge of the state + of the parser to tailor the tokens that are returned. */ + +int +yylex(void) +{ + static enum { + ST_NOTSTARTED, ST_NORMAL, ST_ATTRIBUTE, ST_ASM, ST_TYPEOF, ST_TYPEOF_1, + ST_BRACKET, ST_BRACE, ST_EXPRESSION, ST_STATIC_ASSERT, + } lexstate = ST_NOTSTARTED; + + static int suppress_type_lookup, dont_want_brace_phrase; + static struct string_list *next_node; + static char *source_file; + + int token, count = 0; + struct string_list *cur_node; + + if (lexstate == ST_NOTSTARTED) + { + next_node = xmalloc(sizeof(*next_node)); + next_node->next = NULL; + lexstate = ST_NORMAL; + } + +repeat: + token = yylex1(); + + if (token == 0) + return 0; + else if (token == FILENAME) + { + char *file, *e; + + /* Save the filename and line number for later error messages. */ + + if (cur_filename) + free(cur_filename); + + file = strchr(yytext, '\"')+1; + e = strchr(file, '\"'); + *e = '\0'; + cur_filename = memcpy(xmalloc(e-file+1), file, e-file+1); + cur_line = atoi(yytext+2); + + if (!source_file) { + source_file = xstrdup(cur_filename); + in_source_file = 1; + } else { + in_source_file = (strcmp(cur_filename, source_file) == 0); + } + + goto repeat; + } + + switch (lexstate) + { + case ST_NORMAL: + switch (token) + { + case IDENT: + APP; + { + int r = is_reserved_word(yytext, yyleng); + if (r >= 0) + { + switch (token = r) + { + case ATTRIBUTE_KEYW: + lexstate = ST_ATTRIBUTE; + count = 0; + goto repeat; + case ASM_KEYW: + lexstate = ST_ASM; + count = 0; + goto repeat; + case TYPEOF_KEYW: + lexstate = ST_TYPEOF; + count = 0; + goto repeat; + + case STRUCT_KEYW: + case UNION_KEYW: + case ENUM_KEYW: + dont_want_brace_phrase = 3; + suppress_type_lookup = 2; + goto fini; + + case EXPORT_SYMBOL_KEYW: + goto fini; + + case STATIC_ASSERT_KEYW: + lexstate = ST_STATIC_ASSERT; + count = 0; + goto repeat; + } + } + if (!suppress_type_lookup) + { + if (find_symbol(yytext, SYM_TYPEDEF, 1)) + token = TYPE; + } + } + break; + + case '[': + APP; + lexstate = ST_BRACKET; + count = 1; + goto repeat; + + case '{': + APP; + if (dont_want_brace_phrase) + break; + lexstate = ST_BRACE; + count = 1; + goto repeat; + + case '=': case ':': + APP; + lexstate = ST_EXPRESSION; + break; + + default: + APP; + break; + } + break; + + case ST_ATTRIBUTE: + APP; + switch (token) + { + case '(': + ++count; + goto repeat; + case ')': + if (--count == 0) + { + lexstate = ST_NORMAL; + token = ATTRIBUTE_PHRASE; + break; + } + goto repeat; + default: + goto repeat; + } + break; + + case ST_ASM: + APP; + switch (token) + { + case '(': + ++count; + goto repeat; + case ')': + if (--count == 0) + { + lexstate = ST_NORMAL; + token = ASM_PHRASE; + break; + } + goto repeat; + default: + goto repeat; + } + break; + + case ST_TYPEOF_1: + if (token == IDENT) + { + if (is_reserved_word(yytext, yyleng) >= 0 + || find_symbol(yytext, SYM_TYPEDEF, 1)) + { + yyless(0); + unput('('); + lexstate = ST_NORMAL; + token = TYPEOF_KEYW; + break; + } + _APP("(", 1); + } + lexstate = ST_TYPEOF; + /* FALLTHRU */ + + case ST_TYPEOF: + switch (token) + { + case '(': + if ( ++count == 1 ) + lexstate = ST_TYPEOF_1; + else + APP; + goto repeat; + case ')': + APP; + if (--count == 0) + { + lexstate = ST_NORMAL; + token = TYPEOF_PHRASE; + break; + } + goto repeat; + default: + APP; + goto repeat; + } + break; + + case ST_BRACKET: + APP; + switch (token) + { + case '[': + ++count; + goto repeat; + case ']': + if (--count == 0) + { + lexstate = ST_NORMAL; + token = BRACKET_PHRASE; + break; + } + goto repeat; + default: + goto repeat; + } + break; + + case ST_BRACE: + APP; + switch (token) + { + case '{': + ++count; + goto repeat; + case '}': + if (--count == 0) + { + lexstate = ST_NORMAL; + token = BRACE_PHRASE; + break; + } + goto repeat; + default: + goto repeat; + } + break; + + case ST_EXPRESSION: + switch (token) + { + case '(': case '[': case '{': + ++count; + APP; + goto repeat; + case '}': + /* is this the last line of an enum declaration? */ + if (count == 0) + { + /* Put back the token we just read so's we can find it again + after registering the expression. */ + unput(token); + + lexstate = ST_NORMAL; + token = EXPRESSION_PHRASE; + break; + } + /* FALLTHRU */ + case ')': case ']': + --count; + APP; + goto repeat; + case ',': case ';': + if (count == 0) + { + /* Put back the token we just read so's we can find it again + after registering the expression. */ + unput(token); + + lexstate = ST_NORMAL; + token = EXPRESSION_PHRASE; + break; + } + APP; + goto repeat; + default: + APP; + goto repeat; + } + break; + + case ST_STATIC_ASSERT: + APP; + switch (token) + { + case '(': + ++count; + goto repeat; + case ')': + if (--count == 0) + { + lexstate = ST_NORMAL; + token = STATIC_ASSERT_PHRASE; + break; + } + goto repeat; + default: + goto repeat; + } + break; + + default: + exit(1); + } +fini: + + if (suppress_type_lookup > 0) + --suppress_type_lookup; + if (dont_want_brace_phrase > 0) + --dont_want_brace_phrase; + + yylval = &next_node->next; + + return token; +} diff --git a/scripts/genksyms/parse.y b/scripts/genksyms/parse.y new file mode 100644 index 0000000000..8e9b5e69e8 --- /dev/null +++ b/scripts/genksyms/parse.y @@ -0,0 +1,510 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* + * C global declaration parser for genksyms. + * Copyright 1996, 1997 Linux International. + * + * New implementation contributed by Richard Henderson <rth@tamu.edu> + * Based on original work by Bjorn Ekwall <bj0rn@blox.se> + * + * This file is part of the Linux modutils. + */ + +%{ + +#include <assert.h> +#include <stdlib.h> +#include <string.h> +#include "genksyms.h" + +static int is_typedef; +static int is_extern; +static char *current_name; +static struct string_list *decl_spec; + +static void yyerror(const char *); + +static inline void +remove_node(struct string_list **p) +{ + struct string_list *node = *p; + *p = node->next; + free_node(node); +} + +static inline void +remove_list(struct string_list **pb, struct string_list **pe) +{ + struct string_list *b = *pb, *e = *pe; + *pb = e; + free_list(b, e); +} + +/* Record definition of a struct/union/enum */ +static void record_compound(struct string_list **keyw, + struct string_list **ident, + struct string_list **body, + enum symbol_type type) +{ + struct string_list *b = *body, *i = *ident, *r; + + if (i->in_source_file) { + remove_node(keyw); + (*ident)->tag = type; + remove_list(body, ident); + return; + } + r = copy_node(i); r->tag = type; + r->next = (*keyw)->next; *body = r; (*keyw)->next = NULL; + add_symbol(i->string, type, b, is_extern); +} + +%} + +%token ASM_KEYW +%token ATTRIBUTE_KEYW +%token AUTO_KEYW +%token BOOL_KEYW +%token BUILTIN_INT_KEYW +%token CHAR_KEYW +%token CONST_KEYW +%token DOUBLE_KEYW +%token ENUM_KEYW +%token EXTERN_KEYW +%token EXTENSION_KEYW +%token FLOAT_KEYW +%token INLINE_KEYW +%token INT_KEYW +%token LONG_KEYW +%token REGISTER_KEYW +%token RESTRICT_KEYW +%token SHORT_KEYW +%token SIGNED_KEYW +%token STATIC_KEYW +%token STATIC_ASSERT_KEYW +%token STRUCT_KEYW +%token TYPEDEF_KEYW +%token UNION_KEYW +%token UNSIGNED_KEYW +%token VOID_KEYW +%token VOLATILE_KEYW +%token TYPEOF_KEYW +%token VA_LIST_KEYW + +%token EXPORT_SYMBOL_KEYW + +%token ASM_PHRASE +%token ATTRIBUTE_PHRASE +%token TYPEOF_PHRASE +%token BRACE_PHRASE +%token BRACKET_PHRASE +%token EXPRESSION_PHRASE +%token STATIC_ASSERT_PHRASE + +%token CHAR +%token DOTS +%token IDENT +%token INT +%token REAL +%token STRING +%token TYPE +%token OTHER +%token FILENAME + +%% + +declaration_seq: + declaration + | declaration_seq declaration + ; + +declaration: + { is_typedef = 0; is_extern = 0; current_name = NULL; decl_spec = NULL; } + declaration1 + { free_list(*$2, NULL); *$2 = NULL; } + ; + +declaration1: + EXTENSION_KEYW TYPEDEF_KEYW { is_typedef = 1; } simple_declaration + { $$ = $4; } + | TYPEDEF_KEYW { is_typedef = 1; } simple_declaration + { $$ = $3; } + | simple_declaration + | function_definition + | asm_definition + | export_definition + | static_assert + | error ';' { $$ = $2; } + | error '}' { $$ = $2; } + ; + +simple_declaration: + decl_specifier_seq_opt init_declarator_list_opt ';' + { if (current_name) { + struct string_list *decl = (*$3)->next; + (*$3)->next = NULL; + add_symbol(current_name, + is_typedef ? SYM_TYPEDEF : SYM_NORMAL, + decl, is_extern); + current_name = NULL; + } + $$ = $3; + } + ; + +init_declarator_list_opt: + /* empty */ { $$ = NULL; } + | init_declarator_list + ; + +init_declarator_list: + init_declarator + { struct string_list *decl = *$1; + *$1 = NULL; + add_symbol(current_name, + is_typedef ? SYM_TYPEDEF : SYM_NORMAL, decl, is_extern); + current_name = NULL; + $$ = $1; + } + | init_declarator_list ',' init_declarator + { struct string_list *decl = *$3; + *$3 = NULL; + free_list(*$2, NULL); + *$2 = decl_spec; + add_symbol(current_name, + is_typedef ? SYM_TYPEDEF : SYM_NORMAL, decl, is_extern); + current_name = NULL; + $$ = $3; + } + ; + +init_declarator: + declarator asm_phrase_opt attribute_opt initializer_opt + { $$ = $4 ? $4 : $3 ? $3 : $2 ? $2 : $1; } + ; + +/* Hang on to the specifiers so that we can reuse them. */ +decl_specifier_seq_opt: + /* empty */ { decl_spec = NULL; } + | decl_specifier_seq + ; + +decl_specifier_seq: + decl_specifier { decl_spec = *$1; } + | decl_specifier_seq decl_specifier { decl_spec = *$2; } + ; + +decl_specifier: + storage_class_specifier + { /* Version 2 checksumming ignores storage class, as that + is really irrelevant to the linkage. */ + remove_node($1); + $$ = $1; + } + | type_specifier + ; + +storage_class_specifier: + AUTO_KEYW + | REGISTER_KEYW + | STATIC_KEYW + | EXTERN_KEYW { is_extern = 1; $$ = $1; } + | INLINE_KEYW { is_extern = 0; $$ = $1; } + ; + +type_specifier: + simple_type_specifier + | cvar_qualifier + | TYPEOF_KEYW '(' parameter_declaration ')' + | TYPEOF_PHRASE + + /* References to s/u/e's defined elsewhere. Rearrange things + so that it is easier to expand the definition fully later. */ + | STRUCT_KEYW IDENT + { remove_node($1); (*$2)->tag = SYM_STRUCT; $$ = $2; } + | UNION_KEYW IDENT + { remove_node($1); (*$2)->tag = SYM_UNION; $$ = $2; } + | ENUM_KEYW IDENT + { remove_node($1); (*$2)->tag = SYM_ENUM; $$ = $2; } + + /* Full definitions of an s/u/e. Record it. */ + | STRUCT_KEYW IDENT class_body + { record_compound($1, $2, $3, SYM_STRUCT); $$ = $3; } + | UNION_KEYW IDENT class_body + { record_compound($1, $2, $3, SYM_UNION); $$ = $3; } + | ENUM_KEYW IDENT enum_body + { record_compound($1, $2, $3, SYM_ENUM); $$ = $3; } + /* + * Anonymous enum definition. Tell add_symbol() to restart its counter. + */ + | ENUM_KEYW enum_body + { add_symbol(NULL, SYM_ENUM, NULL, 0); $$ = $2; } + /* Anonymous s/u definitions. Nothing needs doing. */ + | STRUCT_KEYW class_body { $$ = $2; } + | UNION_KEYW class_body { $$ = $2; } + ; + +simple_type_specifier: + CHAR_KEYW + | SHORT_KEYW + | INT_KEYW + | LONG_KEYW + | SIGNED_KEYW + | UNSIGNED_KEYW + | FLOAT_KEYW + | DOUBLE_KEYW + | VOID_KEYW + | BOOL_KEYW + | VA_LIST_KEYW + | BUILTIN_INT_KEYW + | TYPE { (*$1)->tag = SYM_TYPEDEF; $$ = $1; } + ; + +ptr_operator: + '*' cvar_qualifier_seq_opt + { $$ = $2 ? $2 : $1; } + ; + +cvar_qualifier_seq_opt: + /* empty */ { $$ = NULL; } + | cvar_qualifier_seq + ; + +cvar_qualifier_seq: + cvar_qualifier + | cvar_qualifier_seq cvar_qualifier { $$ = $2; } + ; + +cvar_qualifier: + CONST_KEYW | VOLATILE_KEYW | ATTRIBUTE_PHRASE + | RESTRICT_KEYW + { /* restrict has no effect in prototypes so ignore it */ + remove_node($1); + $$ = $1; + } + ; + +declarator: + ptr_operator declarator { $$ = $2; } + | direct_declarator + ; + +direct_declarator: + IDENT + { if (current_name != NULL) { + error_with_pos("unexpected second declaration name"); + YYERROR; + } else { + current_name = (*$1)->string; + $$ = $1; + } + } + | TYPE + { if (current_name != NULL) { + error_with_pos("unexpected second declaration name"); + YYERROR; + } else { + current_name = (*$1)->string; + $$ = $1; + } + } + | direct_declarator '(' parameter_declaration_clause ')' + { $$ = $4; } + | direct_declarator '(' error ')' + { $$ = $4; } + | direct_declarator BRACKET_PHRASE + { $$ = $2; } + | '(' declarator ')' + { $$ = $3; } + ; + +/* Nested declarators differ from regular declarators in that they do + not record the symbols they find in the global symbol table. */ +nested_declarator: + ptr_operator nested_declarator { $$ = $2; } + | direct_nested_declarator + ; + +direct_nested_declarator: + IDENT + | TYPE + | direct_nested_declarator '(' parameter_declaration_clause ')' + { $$ = $4; } + | direct_nested_declarator '(' error ')' + { $$ = $4; } + | direct_nested_declarator BRACKET_PHRASE + { $$ = $2; } + | '(' nested_declarator ')' + { $$ = $3; } + | '(' error ')' + { $$ = $3; } + ; + +parameter_declaration_clause: + parameter_declaration_list_opt DOTS { $$ = $2; } + | parameter_declaration_list_opt + | parameter_declaration_list ',' DOTS { $$ = $3; } + ; + +parameter_declaration_list_opt: + /* empty */ { $$ = NULL; } + | parameter_declaration_list + ; + +parameter_declaration_list: + parameter_declaration + | parameter_declaration_list ',' parameter_declaration + { $$ = $3; } + ; + +parameter_declaration: + decl_specifier_seq m_abstract_declarator + { $$ = $2 ? $2 : $1; } + ; + +m_abstract_declarator: + ptr_operator m_abstract_declarator + { $$ = $2 ? $2 : $1; } + | direct_m_abstract_declarator + ; + +direct_m_abstract_declarator: + /* empty */ { $$ = NULL; } + | IDENT + { /* For version 2 checksums, we don't want to remember + private parameter names. */ + remove_node($1); + $$ = $1; + } + /* This wasn't really a typedef name but an identifier that + shadows one. */ + | TYPE + { remove_node($1); + $$ = $1; + } + | direct_m_abstract_declarator '(' parameter_declaration_clause ')' + { $$ = $4; } + | direct_m_abstract_declarator '(' error ')' + { $$ = $4; } + | direct_m_abstract_declarator BRACKET_PHRASE + { $$ = $2; } + | '(' m_abstract_declarator ')' + { $$ = $3; } + | '(' error ')' + { $$ = $3; } + ; + +function_definition: + decl_specifier_seq_opt declarator BRACE_PHRASE + { struct string_list *decl = *$2; + *$2 = NULL; + add_symbol(current_name, SYM_NORMAL, decl, is_extern); + $$ = $3; + } + ; + +initializer_opt: + /* empty */ { $$ = NULL; } + | initializer + ; + +/* We never care about the contents of an initializer. */ +initializer: + '=' EXPRESSION_PHRASE + { remove_list($2, &(*$1)->next); $$ = $2; } + ; + +class_body: + '{' member_specification_opt '}' { $$ = $3; } + | '{' error '}' { $$ = $3; } + ; + +member_specification_opt: + /* empty */ { $$ = NULL; } + | member_specification + ; + +member_specification: + member_declaration + | member_specification member_declaration { $$ = $2; } + ; + +member_declaration: + decl_specifier_seq_opt member_declarator_list_opt ';' + { $$ = $3; } + | error ';' + { $$ = $2; } + ; + +member_declarator_list_opt: + /* empty */ { $$ = NULL; } + | member_declarator_list + ; + +member_declarator_list: + member_declarator + | member_declarator_list ',' member_declarator { $$ = $3; } + ; + +member_declarator: + nested_declarator attribute_opt { $$ = $2 ? $2 : $1; } + | IDENT member_bitfield_declarator { $$ = $2; } + | member_bitfield_declarator + ; + +member_bitfield_declarator: + ':' EXPRESSION_PHRASE { $$ = $2; } + ; + +attribute_opt: + /* empty */ { $$ = NULL; } + | attribute_opt ATTRIBUTE_PHRASE + ; + +enum_body: + '{' enumerator_list '}' { $$ = $3; } + | '{' enumerator_list ',' '}' { $$ = $4; } + ; + +enumerator_list: + enumerator + | enumerator_list ',' enumerator + +enumerator: + IDENT + { + const char *name = strdup((*$1)->string); + add_symbol(name, SYM_ENUM_CONST, NULL, 0); + } + | IDENT '=' EXPRESSION_PHRASE + { + const char *name = strdup((*$1)->string); + struct string_list *expr = copy_list_range(*$3, *$2); + add_symbol(name, SYM_ENUM_CONST, expr, 0); + } + +asm_definition: + ASM_PHRASE ';' { $$ = $2; } + ; + +asm_phrase_opt: + /* empty */ { $$ = NULL; } + | ASM_PHRASE + ; + +export_definition: + EXPORT_SYMBOL_KEYW '(' IDENT ')' ';' + { export_symbol((*$3)->string); $$ = $5; } + ; + +/* Ignore any module scoped _Static_assert(...) */ +static_assert: + STATIC_ASSERT_PHRASE ';' { $$ = $2; } + ; + +%% + +static void +yyerror(const char *e) +{ + error_with_pos("%s", e); +} |