gendwarfksyms: Add symtypes output

Add support for producing genksyms-style symtypes files. Process
die_map to find the longest expansions for each type, and use symtypes
references in type definitions. The basic file format is similar to
genksyms, with two notable exceptions:

  1. Type names with spaces (common with Rust) in references are
     wrapped in single quotes. E.g.:

     s#'core::result::Result<u8, core::num::error::ParseIntError>'

  2. The actual type definition is the simple parsed DWARF format we
     output with --dump-dies, not the preprocessed C-style format
     genksyms produces.

Signed-off-by: Sami Tolvanen <samitolvanen@google.com>

Author: samitolvanen

scripts/gendwarfksyms/Makefile

@@ -5,6 +5,7 @@ gendwarfksyms-objs += cache.o
 gendwarfksyms-objs += die.o
 gendwarfksyms-objs += dwarf.o
 gendwarfksyms-objs += symbols.o
+gendwarfksyms-objs += types.o
 
 HOST_EXTRACFLAGS := -I $(srctree)/tools/include
 HOSTLDLIBS_gendwarfksyms := -ldw -lelf

scripts/gendwarfksyms/die.c

@@ -85,6 +85,19 @@ static void reset_die(struct die *cd)
 	cd->list = NULL;
 }
 
+int die_map_for_each(die_map_callback_t func, void *arg)
+{
+	struct die *cd;
+	struct hlist_node *tmp;
+	int i;
+
+	hash_for_each_safe(die_map, i, tmp, cd, hash) {
+		check(func(cd, arg));
+	}
+
+	return 0;
+}
+
 void die_map_free(void)
 {
 	struct hlist_node *tmp;

scripts/gendwarfksyms/dwarf.c

@@ -60,11 +60,11 @@ static bool is_export_symbol(struct state *state, Dwarf_Die *die)
 	if (get_ref_die_attr(die, DW_AT_abstract_origin, &origin))
 		source = &origin;
 
-	state->sym = symbol_get(get_name(die));
+	state->sym = symbol_get_unprocessed(get_name(die));
 
 	/* Look up using the origin name if there are no matches. */
 	if (!state->sym && source != die)
-		state->sym = symbol_get(get_name(source));
+		state->sym = symbol_get_unprocessed(get_name(source));
 
 	state->die = *source;
 	return !!state->sym;
@@ -384,6 +384,7 @@ static int process_subroutine_type(struct state *state, struct die *cache,
 	return check(__process_subroutine_type(state, cache, die,
 					       "subroutine_type"));
 }
+
 static int process_variant_type(struct state *state, struct die *cache,
 				Dwarf_Die *die)
 {
@@ -695,14 +696,16 @@ static int process_type(struct state *state, struct die *parent, Dwarf_Die *die)
 static int process_subprogram(struct state *state, Dwarf_Die *die)
 {
 	check(__process_subroutine_type(state, NULL, die, "subprogram"));
-	return check(process(state, NULL, ";\n"));
+	state->sym->state = MAPPED;
+	return 0;
 }
 
 static int process_variable(struct state *state, Dwarf_Die *die)
 {
 	check(process(state, NULL, "variable "));
 	check(process_type_attr(state, NULL, die));
-	return check(process(state, NULL, ";\n"));
+	state->sym->state = MAPPED;
+	return 0;
 }
 
 static int process_symbol_ptr(struct state *state, Dwarf_Die *die)
@@ -757,6 +760,9 @@ static int process_exported_symbols(struct state *state, struct die *cache,
 		else
 			check(process_variable(state, &state->die));
 
+		if (dump_dies)
+			fputs("\n", stderr);
+
 		cache_clear_expanded(&state->expansion_cache);
 		return 0;
 	default:

scripts/gendwarfksyms/gendwarfksyms.c

@@ -20,6 +20,11 @@ bool debug;
 bool dump_dies;
 /* Print out inline debugging information about die_map changes */
 bool dump_die_map;
+/* Print out type_map contents */
+bool dump_types;
+/* Produce a symtypes file */
+bool symtypes;
+static const char *symtypes_file;
 
 static const struct {
 	const char *arg;
@@ -29,6 +34,8 @@ static const struct {
 	{ "--debug", &debug, NULL },
 	{ "--dump-dies", &dump_dies, NULL },
 	{ "--dump-die-map", &dump_die_map, NULL },
+	{ "--dump-types", &dump_types, NULL },
+	{ "--symtypes", &symtypes, &symtypes_file },
 };
 
 static int usage(void)
@@ -79,6 +86,7 @@ static int process_modules(Dwfl_Module *mod, void **userdata, const char *name,
 	Dwarf_Die cudie;
 	Dwarf_CU *cu = NULL;
 	Dwarf *dbg;
+	FILE *symfile = arg;
 	int res;
 
 	debug("%s", name);
@@ -100,6 +108,10 @@ static int process_modules(Dwfl_Module *mod, void **userdata, const char *name,
 		check(process_module(mod, dbg, &cudie));
 	} while (cu);
 
+	/*
+	 * Use die_map to expand type strings and write them to `symfile`.
+	 */
+	check(generate_symtypes(symfile));
 	die_map_free();
 
 	return DWARF_CB_OK;
@@ -112,6 +124,7 @@ static const Dwfl_Callbacks callbacks = {
 
 int main(int argc, const char **argv)
 {
+	FILE *symfile = NULL;
 	unsigned int n;
 
 	if (parse_options(argc, argv) < 0)
@@ -122,6 +135,16 @@ int main(int argc, const char **argv)
 
 	check(symbol_read_exports(stdin));
 
+	if (symtypes_file) {
+		symfile = fopen(symtypes_file, "w+");
+
+		if (!symfile) {
+			error("fopen failed for '%s': %s", symtypes_file,
+			      strerror(errno));
+			return -1;
+		}
+	}
+
 	for (n = 0; n < object_count; n++) {
 		Dwfl *dwfl;
 		int fd;
@@ -151,7 +174,7 @@ int main(int argc, const char **argv)
 
 		dwfl_report_end(dwfl, NULL, NULL);
 
-		if (dwfl_getmodules(dwfl, &process_modules, NULL, 0)) {
+		if (dwfl_getmodules(dwfl, &process_modules, symfile, 0)) {
 			error("dwfl_getmodules failed for '%s'",
 			      object_files[n]);
 			return -1;
@@ -161,5 +184,8 @@ int main(int argc, const char **argv)
 		close(fd);
 	}
 
+	if (symfile)
+		fclose(symfile);
+
 	return 0;
 }

scripts/gendwarfksyms/gendwarfksyms.h

@@ -22,6 +22,8 @@
 extern bool debug;
 extern bool dump_dies;
 extern bool dump_die_map;
+extern bool dump_types;
+extern bool symtypes;
 
 #define MAX_INPUT_FILES 128
 
@@ -89,6 +91,12 @@ extern bool dump_die_map;
 #define SYMBOL_PTR_PREFIX "__gendwarfksyms_ptr_"
 #define SYMBOL_PTR_PREFIX_LEN (sizeof(SYMBOL_PTR_PREFIX) - 1)
 
+/* See dwarf.c:is_declaration */
+#define SYMBOL_DECLONLY_PREFIX "__gendwarfksyms_declonly_"
+#define SYMBOL_DECLONLY_PREFIX_LEN (sizeof(SYMBOL_DECLONLY_PREFIX) - 1)
+
+enum symbol_state { UNPROCESSED, MAPPED };
+
 struct symbol_addr {
 	uint32_t section;
 	Elf64_Addr address;
@@ -109,12 +117,14 @@ struct symbol {
 	struct symbol_addr addr;
 	struct hlist_node addr_hash;
 	struct hlist_node name_hash;
+	enum symbol_state state;
+	uintptr_t die_addr;
 };
 
 extern bool is_symbol_ptr(const char *name);
 extern int symbol_read_exports(FILE *file);
 extern int symbol_read_symtab(int fd);
-extern struct symbol *symbol_get(const char *name);
+extern struct symbol *symbol_get_unprocessed(const char *name);
 
 /*
  * die.c
@@ -157,12 +167,15 @@ struct die {
 	struct hlist_node hash;
 };
 
+typedef int (*die_map_callback_t)(struct die *, void *arg);
+
 extern int __die_map_get(uintptr_t addr, enum die_state state,
 			 struct die **res);
 extern int die_map_get(Dwarf_Die *die, enum die_state state, struct die **res);
 extern int die_map_add_string(struct die *pd, const char *str);
 extern int die_map_add_linebreak(struct die *pd, int linebreak);
 extern int die_map_add_die(struct die *pd, struct die *child);
+extern int die_map_for_each(die_map_callback_t func, void *arg);
 extern void die_map_free(void);
 
 /*
@@ -222,4 +235,10 @@ extern int process_die_container(struct state *state, struct die *cache,
 
 extern int process_module(Dwfl_Module *mod, Dwarf *dbg, Dwarf_Die *cudie);
 
+/*
+ * types.c
+ */
+
+extern int generate_symtypes(FILE *file);
+
 #endif /* __GENDWARFKSYMS_H */

scripts/gendwarfksyms/symbols.c

@@ -117,6 +117,7 @@ int symbol_read_exports(FILE *file)
 
 		sym->name = name;
 		sym->addr.section = SHN_UNDEF;
+		sym->state = UNPROCESSED;
 		name = NULL;
 
 		hash_add(symbol_names, &sym->name_hash, name_hash(sym->name));
@@ -132,19 +133,21 @@ int symbol_read_exports(FILE *file)
 	return 0;
 }
 
-static int get_symbol(struct symbol *sym, void *arg)
+static int get_unprocessed(struct symbol *sym, void *arg)
 {
 	struct symbol **res = arg;
 
-	*res = sym;
+	if (sym->state == UNPROCESSED)
+		*res = sym;
+
 	return 0;
 }
 
-struct symbol *symbol_get(const char *name)
+struct symbol *symbol_get_unprocessed(const char *name)
 {
 	struct symbol *sym = NULL;
 
-	for_each(name, false, get_symbol, &sym);
+	for_each(name, false, get_unprocessed, &sym);
 	return sym;
 }