"), path, realpath, INT2FIX(0), parent, ISEQ_TYPE_MAIN, &COMPILE_OPTION_DEFAULT); } static inline rb_iseq_t * iseq_translate(rb_iseq_t *iseq) { if (rb_respond_to(rb_cISeq, rb_intern("translate"))) { VALUE v1 = iseqw_new(iseq); VALUE v2 = rb_funcall(rb_cISeq, rb_intern("translate"), 1, v1); if (v1 != v2 && CLASS_OF(v2) == rb_cISeq) { iseq = (rb_iseq_t *)iseqw_check(v2); } } return iseq; } rb_iseq_t * rb_iseq_new_with_opt(const rb_ast_body_t *ast, VALUE name, VALUE path, VALUE realpath, VALUE first_lineno, const rb_iseq_t *parent, enum iseq_type type, const rb_compile_option_t *option) { const NODE *node = ast ? ast->root : 0; /* TODO: argument check */ rb_iseq_t *iseq = iseq_alloc(); rb_compile_option_t new_opt; new_opt = option ? *option : COMPILE_OPTION_DEFAULT; if (ast && ast->compile_option) rb_iseq_make_compile_option(&new_opt, ast->compile_option); prepare_iseq_build(iseq, name, path, realpath, first_lineno, node ? &node->nd_loc : NULL, parent, type, &new_opt); rb_iseq_compile_node(iseq, node); finish_iseq_build(iseq); return iseq_translate(iseq); } rb_iseq_t * rb_iseq_new_ifunc(const struct vm_ifunc *ifunc, VALUE name, VALUE path, VALUE realpath, VALUE first_lineno, const rb_iseq_t *parent, enum iseq_type type, const rb_compile_option_t *option) { /* TODO: argument check */ rb_iseq_t *iseq = iseq_alloc(); if (!option) option = &COMPILE_OPTION_DEFAULT; prepare_iseq_build(iseq, name, path, realpath, first_lineno, NULL, parent, type, option); rb_iseq_compile_ifunc(iseq, ifunc); finish_iseq_build(iseq); return iseq_translate(iseq); } const rb_iseq_t * rb_iseq_load_iseq(VALUE fname) { VALUE iseqv = rb_check_funcall(rb_cISeq, rb_intern("load_iseq"), 1, &fname); if (!SPECIAL_CONST_P(iseqv) && RBASIC_CLASS(iseqv) == rb_cISeq) { return iseqw_check(iseqv); } return NULL; } #define CHECK_ARRAY(v) rb_to_array_type(v) #define CHECK_HASH(v) rb_to_hash_type(v) #define CHECK_STRING(v) rb_str_to_str(v) #define CHECK_SYMBOL(v) rb_to_symbol_type(v) static inline VALUE CHECK_INTEGER(VALUE v) {(void)NUM2LONG(v); return v;} static enum iseq_type iseq_type_from_sym(VALUE type) { const ID id_top = rb_intern("top"); const ID id_method = rb_intern("method"); const ID id_block = rb_intern("block"); const ID id_class = rb_intern("class"); const ID id_rescue = rb_intern("rescue"); const ID id_ensure = rb_intern("ensure"); const ID id_eval = rb_intern("eval"); const ID id_main = rb_intern("main"); const ID id_plain = rb_intern("plain"); /* ensure all symbols are static or pinned down before * conversion */ const ID typeid = rb_check_id(&type); if (typeid == id_top) return ISEQ_TYPE_TOP; if (typeid == id_method) return ISEQ_TYPE_METHOD; if (typeid == id_block) return ISEQ_TYPE_BLOCK; if (typeid == id_class) return ISEQ_TYPE_CLASS; if (typeid == id_rescue) return ISEQ_TYPE_RESCUE; if (typeid == id_ensure) return ISEQ_TYPE_ENSURE; if (typeid == id_eval) return ISEQ_TYPE_EVAL; if (typeid == id_main) return ISEQ_TYPE_MAIN; if (typeid == id_plain) return ISEQ_TYPE_PLAIN; return (enum iseq_type)-1; } static VALUE iseq_load(VALUE data, const rb_iseq_t *parent, VALUE opt) { rb_iseq_t *iseq = iseq_alloc(); VALUE magic, version1, version2, format_type, misc; VALUE name, path, realpath, first_lineno, code_location; VALUE type, body, locals, params, exception; st_data_t iseq_type; rb_compile_option_t option; int i = 0; rb_code_location_t tmp_loc = { {0, 0}, {-1, -1} }; /* [magic, major_version, minor_version, format_type, misc, * label, path, first_lineno, * type, locals, args, exception_table, body] */ data = CHECK_ARRAY(data); magic = CHECK_STRING(rb_ary_entry(data, i++)); version1 = CHECK_INTEGER(rb_ary_entry(data, i++)); version2 = CHECK_INTEGER(rb_ary_entry(data, i++)); format_type = CHECK_INTEGER(rb_ary_entry(data, i++)); misc = CHECK_HASH(rb_ary_entry(data, i++)); ((void)magic, (void)version1, (void)version2, (void)format_type); name = CHECK_STRING(rb_ary_entry(data, i++)); path = CHECK_STRING(rb_ary_entry(data, i++)); realpath = rb_ary_entry(data, i++); realpath = NIL_P(realpath) ? Qnil : CHECK_STRING(realpath); first_lineno = CHECK_INTEGER(rb_ary_entry(data, i++)); type = CHECK_SYMBOL(rb_ary_entry(data, i++)); locals = CHECK_ARRAY(rb_ary_entry(data, i++)); params = CHECK_HASH(rb_ary_entry(data, i++)); exception = CHECK_ARRAY(rb_ary_entry(data, i++)); body = CHECK_ARRAY(rb_ary_entry(data, i++)); iseq->body->local_iseq = iseq; iseq_type = iseq_type_from_sym(type); if (iseq_type == (enum iseq_type)-1) { rb_raise(rb_eTypeError, "unsupport type: :%"PRIsVALUE, rb_sym2str(type)); } code_location = rb_hash_aref(misc, ID2SYM(rb_intern("code_location"))); if (RB_TYPE_P(code_location, T_ARRAY) && RARRAY_LEN(code_location) == 4) { tmp_loc.beg_pos.lineno = NUM2INT(rb_ary_entry(code_location, 0)); tmp_loc.beg_pos.column = NUM2INT(rb_ary_entry(code_location, 1)); tmp_loc.end_pos.lineno = NUM2INT(rb_ary_entry(code_location, 2)); tmp_loc.end_pos.column = NUM2INT(rb_ary_entry(code_location, 3)); } make_compile_option(&option, opt); option.peephole_optimization = FALSE; /* because peephole optimization can modify original iseq */ prepare_iseq_build(iseq, name, path, realpath, first_lineno, &tmp_loc, parent, (enum iseq_type)iseq_type, &option); rb_iseq_build_from_ary(iseq, misc, locals, params, exception, body); finish_iseq_build(iseq); return iseqw_new(iseq); } /* * :nodoc: */ static VALUE iseq_s_load(int argc, VALUE *argv, VALUE self) { VALUE data, opt=Qnil; rb_scan_args(argc, argv, "11", &data, &opt); return iseq_load(data, NULL, opt); } VALUE rb_iseq_load(VALUE data, VALUE parent, VALUE opt) { return iseq_load(data, RTEST(parent) ? (rb_iseq_t *)parent : NULL, opt); } rb_iseq_t * rb_iseq_compile_with_option(VALUE src, VALUE file, VALUE realpath, VALUE line, const struct rb_block *base_block, VALUE opt) { rb_iseq_t *iseq = NULL; const rb_iseq_t *const parent = base_block ? vm_block_iseq(base_block) : NULL; rb_compile_option_t option; const enum iseq_type type = parent ? ISEQ_TYPE_EVAL : ISEQ_TYPE_TOP; #if !defined(__GNUC__) || (__GNUC__ == 4 && __GNUC_MINOR__ == 8) # define INITIALIZED volatile /* suppress warnings by gcc 4.8 */ #else # define INITIALIZED /* volatile */ #endif rb_ast_t *(*parse)(VALUE vparser, VALUE fname, VALUE file, int start); int ln; rb_ast_t *INITIALIZED ast; /* safe results first */ make_compile_option(&option, opt); ln = NUM2INT(line); StringValueCStr(file); if (RB_TYPE_P(src, T_FILE)) { parse = rb_parser_compile_file_path; } else { parse = rb_parser_compile_string_path; StringValue(src); } { const VALUE parser = rb_parser_new(); rb_parser_set_context(parser, base_block, FALSE); ast = (*parse)(parser, file, src, ln); } if (!ast->body.root) { rb_ast_dispose(ast); rb_exc_raise(GET_EC()->errinfo); } else { INITIALIZED VALUE label = parent ? parent->body->location.label : rb_fstring_cstr(""); iseq = rb_iseq_new_with_opt(&ast->body, label, file, realpath, line, parent, type, &option); rb_ast_dispose(ast); } return iseq; } rb_iseq_t * rb_iseq_compile(VALUE src, VALUE file, VALUE line) { return rb_iseq_compile_with_option(src, file, Qnil, line, 0, Qnil); } rb_iseq_t * rb_iseq_compile_on_base(VALUE src, VALUE file, VALUE line, const struct rb_block *base_block) { return rb_iseq_compile_with_option(src, file, Qnil, line, base_block, Qnil); } VALUE rb_iseq_path(const rb_iseq_t *iseq) { return pathobj_path(iseq->body->location.pathobj); } VALUE rb_iseq_realpath(const rb_iseq_t *iseq) { return pathobj_realpath(iseq->body->location.pathobj); } VALUE rb_iseq_absolute_path(const rb_iseq_t *iseq) { return rb_iseq_realpath(iseq); } VALUE rb_iseq_label(const rb_iseq_t *iseq) { return iseq->body->location.label; } VALUE rb_iseq_base_label(const rb_iseq_t *iseq) { return iseq->body->location.base_label; } VALUE rb_iseq_first_lineno(const rb_iseq_t *iseq) { return iseq->body->location.first_lineno; } VALUE rb_iseq_method_name(const rb_iseq_t *iseq) { const rb_iseq_t *local_iseq; local_iseq = iseq->body->local_iseq; if (local_iseq->body->type == ISEQ_TYPE_METHOD) { return local_iseq->body->location.base_label; } else { return Qnil; } } void rb_iseq_code_location(const rb_iseq_t *iseq, int *beg_pos_lineno, int *beg_pos_column, int *end_pos_lineno, int *end_pos_column) { if (beg_pos_lineno) *beg_pos_lineno = iseq->body->location.code_location.beg_pos.lineno; if (beg_pos_column) *beg_pos_column = iseq->body->location.code_location.beg_pos.column; if (end_pos_lineno) *end_pos_lineno = iseq->body->location.code_location.end_pos.lineno; if (end_pos_column) *end_pos_column = iseq->body->location.code_location.end_pos.column; } VALUE rb_iseq_coverage(const rb_iseq_t *iseq) { return ISEQ_COVERAGE(iseq); } /* define wrapper class methods (RubyVM::InstructionSequence) */ static void iseqw_mark(void *ptr) { rb_gc_mark((VALUE)ptr); } static size_t iseqw_memsize(const void *ptr) { return iseq_memsize((const rb_iseq_t *)ptr); } static const rb_data_type_t iseqw_data_type = { "T_IMEMO/iseq", {iseqw_mark, NULL, iseqw_memsize,}, 0, 0, RUBY_TYPED_FREE_IMMEDIATELY|RUBY_TYPED_WB_PROTECTED }; static VALUE iseqw_new(const rb_iseq_t *iseq) { union { const rb_iseq_t *in; void *out; } deconst; VALUE obj; deconst.in = iseq; obj = TypedData_Wrap_Struct(rb_cISeq, &iseqw_data_type, deconst.out); RB_OBJ_WRITTEN(obj, Qundef, iseq); return obj; } VALUE rb_iseqw_new(const rb_iseq_t *iseq) { return iseqw_new(iseq); } /* * call-seq: * InstructionSequence.compile(source[, file[, path[, line[, options]]]]) -> iseq * InstructionSequence.new(source[, file[, path[, line[, options]]]]) -> iseq * * Takes +source+, a String of Ruby code and compiles it to an * InstructionSequence. * * Optionally takes +file+, +path+, and +line+ which describe the filename, * absolute path and first line number of the ruby code in +source+ which are * metadata attached to the returned +iseq+. * * +options+, which can be +true+, +false+ or a +Hash+, is used to * modify the default behavior of the Ruby iseq compiler. * * For details regarding valid compile options see ::compile_option=. * * RubyVM::InstructionSequence.compile("a = 1 + 2") * #=> @> * */ static VALUE iseqw_s_compile(int argc, VALUE *argv, VALUE self) { VALUE src, file = Qnil, path = Qnil, line = INT2FIX(1), opt = Qnil; int i; rb_secure(1); i = rb_scan_args(argc, argv, "1*:", &src, NULL, &opt); if (i > 4+NIL_P(opt)) rb_error_arity(argc, 1, 5); switch (i) { case 5: opt = argv[--i]; case 4: line = argv[--i]; case 3: path = argv[--i]; case 2: file = argv[--i]; } if (NIL_P(file)) file = rb_fstring_cstr(""); if (NIL_P(line)) line = INT2FIX(1); return iseqw_new(rb_iseq_compile_with_option(src, file, path, line, 0, opt)); } /* * call-seq: * InstructionSequence.compile_file(file[, options]) -> iseq * * Takes +file+, a String with the location of a Ruby source file, reads, * parses and compiles the file, and returns +iseq+, the compiled * InstructionSequence with source location metadata set. * * Optionally takes +options+, which can be +true+, +false+ or a +Hash+, to * modify the default behavior of the Ruby iseq compiler. * * For details regarding valid compile options see ::compile_option=. * * # /tmp/hello.rb * puts "Hello, world!" * * # elsewhere * RubyVM::InstructionSequence.compile_file("/tmp/hello.rb") * #=> @/tmp/hello.rb> */ static VALUE iseqw_s_compile_file(int argc, VALUE *argv, VALUE self) { VALUE file, line = INT2FIX(1), opt = Qnil; VALUE parser, f, exc = Qnil, ret; rb_ast_t *ast; rb_compile_option_t option; int i; rb_secure(1); i = rb_scan_args(argc, argv, "1*:", &file, NULL, &opt); if (i > 1+NIL_P(opt)) rb_error_arity(argc, 1, 2); switch (i) { case 2: opt = argv[--i]; } FilePathValue(file); file = rb_fstring(file); /* rb_io_t->pathv gets frozen anyways */ f = rb_file_open_str(file, "r"); parser = rb_parser_new(); rb_parser_set_context(parser, NULL, FALSE); ast = rb_parser_compile_file_path(parser, file, f, NUM2INT(line)); if (!ast->body.root) exc = GET_EC()->errinfo; rb_io_close(f); if (!ast->body.root) { rb_ast_dispose(ast); rb_exc_raise(exc); } make_compile_option(&option, opt); ret = iseqw_new(rb_iseq_new_with_opt(&ast->body, rb_fstring_cstr("

"), file, rb_realpath_internal(Qnil, file, 1), line, NULL, ISEQ_TYPE_TOP, &option)); rb_ast_dispose(ast); return ret; } /* * call-seq: * InstructionSequence.compile_option = options * * Sets the default values for various optimizations in the Ruby iseq * compiler. * * Possible values for +options+ include +true+, which enables all options, * +false+ which disables all options, and +nil+ which leaves all options * unchanged. * * You can also pass a +Hash+ of +options+ that you want to change, any * options not present in the hash will be left unchanged. * * Possible option names (which are keys in +options+) which can be set to * +true+ or +false+ include: * * * +:inline_const_cache+ * * +:instructions_unification+ * * +:operands_unification+ * * +:peephole_optimization+ * * +:specialized_instruction+ * * +:stack_caching+ * * +:tailcall_optimization+ * * Additionally, +:debug_level+ can be set to an integer. * * These default options can be overwritten for a single run of the iseq * compiler by passing any of the above values as the +options+ parameter to * ::new, ::compile and ::compile_file. */ static VALUE iseqw_s_compile_option_set(VALUE self, VALUE opt) { rb_compile_option_t option; rb_secure(1); make_compile_option(&option, opt); COMPILE_OPTION_DEFAULT = option; return opt; } /* * call-seq: * InstructionSequence.compile_option -> options * * Returns a hash of default options used by the Ruby iseq compiler. * * For details, see InstructionSequence.compile_option=. */ static VALUE iseqw_s_compile_option_get(VALUE self) { return make_compile_option_value(&COMPILE_OPTION_DEFAULT); } static const rb_iseq_t * iseqw_check(VALUE iseqw) { rb_iseq_t *iseq = DATA_PTR(iseqw); if (!iseq->body) { ibf_load_iseq_complete(iseq); } if (!iseq->body->location.label) { rb_raise(rb_eTypeError, "uninitialized InstructionSequence"); } return iseq; } const rb_iseq_t * rb_iseqw_to_iseq(VALUE iseqw) { return iseqw_check(iseqw); } /* * call-seq: * iseq.eval -> obj * * Evaluates the instruction sequence and returns the result. * * RubyVM::InstructionSequence.compile("1 + 2").eval #=> 3 */ static VALUE iseqw_eval(VALUE self) { rb_secure(1); return rb_iseq_eval(iseqw_check(self)); } /* * Returns a human-readable string representation of this instruction * sequence, including the #label and #path. */ static VALUE iseqw_inspect(VALUE self) { const rb_iseq_t *iseq = iseqw_check(self); VALUE klass = rb_class_name(rb_obj_class(self)); if (!iseq->body->location.label) { return rb_sprintf("#<%"PRIsVALUE": uninitialized>", klass); } else { return rb_sprintf("<%"PRIsVALUE":%"PRIsVALUE"@%"PRIsVALUE":%d>", klass, iseq->body->location.label, rb_iseq_path(iseq), FIX2INT(rb_iseq_first_lineno(iseq))); } } /* * Returns the path of this instruction sequence. * * if the iseq was evaluated from a string. * * For example, using irb: * * iseq = RubyVM::InstructionSequence.compile('num = 1 + 2') * #=> @> * iseq.path * #=> "" * * Using ::compile_file: * * # /tmp/method.rb * def hello * puts "hello, world" * end * * # in irb * > iseq = RubyVM::InstructionSequence.compile_file('/tmp/method.rb') * > iseq.path #=> /tmp/method.rb */ static VALUE iseqw_path(VALUE self) { return rb_iseq_path(iseqw_check(self)); } /* * Returns the absolute path of this instruction sequence. * * +nil+ if the iseq was evaluated from a string. * * For example, using ::compile_file: * * # /tmp/method.rb * def hello * puts "hello, world" * end * * # in irb * > iseq = RubyVM::InstructionSequence.compile_file('/tmp/method.rb') * > iseq.absolute_path #=> /tmp/method.rb */ static VALUE iseqw_absolute_path(VALUE self) { return rb_iseq_realpath(iseqw_check(self)); } /* Returns the label of this instruction sequence. * *

if it's at the top level, if it * was evaluated from a string. * * For example, using irb: * * iseq = RubyVM::InstructionSequence.compile('num = 1 + 2') * #=> @> * iseq.label * #=> "" * * Using ::compile_file: * * # /tmp/method.rb * def hello * puts "hello, world" * end * * # in irb * > iseq = RubyVM::InstructionSequence.compile_file('/tmp/method.rb') * > iseq.label #=>

*/ static VALUE iseqw_label(VALUE self) { return rb_iseq_label(iseqw_check(self)); } /* Returns the base label of this instruction sequence. * * For example, using irb: * * iseq = RubyVM::InstructionSequence.compile('num = 1 + 2') * #=> @> * iseq.base_label * #=> "" * * Using ::compile_file: * * # /tmp/method.rb * def hello * puts "hello, world" * end * * # in irb * > iseq = RubyVM::InstructionSequence.compile_file('/tmp/method.rb') * > iseq.base_label #=>

*/ static VALUE iseqw_base_label(VALUE self) { return rb_iseq_base_label(iseqw_check(self)); } /* Returns the number of the first source line where the instruction sequence * was loaded from. * * For example, using irb: * * iseq = RubyVM::InstructionSequence.compile('num = 1 + 2') * #=> @> * iseq.first_lineno * #=> 1 */ static VALUE iseqw_first_lineno(VALUE self) { return rb_iseq_first_lineno(iseqw_check(self)); } static VALUE iseq_data_to_ary(const rb_iseq_t *iseq); /* * call-seq: * iseq.to_a -> ary * * Returns an Array with 14 elements representing the instruction sequence * with the following data: * * [magic] * A string identifying the data format. Always * +YARVInstructionSequence/SimpleDataFormat+. * * [major_version] * The major version of the instruction sequence. * * [minor_version] * The minor version of the instruction sequence. * * [format_type] * A number identifying the data format. Always 1. * * [misc] * A hash containing: * * [+:arg_size+] * the total number of arguments taken by the method or the block (0 if * _iseq_ doesn't represent a method or block) * [+:local_size+] * the number of local variables + 1 * [+:stack_max+] * used in calculating the stack depth at which a SystemStackError is * thrown. * * [#label] * The name of the context (block, method, class, module, etc.) that this * instruction sequence belongs to. * *

if it's at the top level, if * it was evaluated from a string. * * [#path] * The relative path to the Ruby file where the instruction sequence was * loaded from. * * if the iseq was evaluated from a string. * * [#absolute_path] * The absolute path to the Ruby file where the instruction sequence was * loaded from. * * +nil+ if the iseq was evaluated from a string. * * [#first_lineno] * The number of the first source line where the instruction sequence was * loaded from. * * [type] * The type of the instruction sequence. * * Valid values are +:top+, +:method+, +:block+, +:class+, +:rescue+, * +:ensure+, +:eval+, +:main+, and +plain+. * * [locals] * An array containing the names of all arguments and local variables as * symbols. * * [params] * An Hash object containing parameter information. * * More info about these values can be found in +vm_core.h+. * * [catch_table] * A list of exceptions and control flow operators (rescue, next, redo, * break, etc.). * * [bytecode] * An array of arrays containing the instruction names and operands that * make up the body of the instruction sequence. * * Note that this format is MRI specific and version dependent. * */ static VALUE iseqw_to_a(VALUE self) { const rb_iseq_t *iseq = iseqw_check(self); rb_secure(1); return iseq_data_to_ary(iseq); } #if VM_INSN_INFO_TABLE_IMPL == 1 /* binary search */ static const struct iseq_insn_info_entry * get_insn_info_binary_search(const rb_iseq_t *iseq, size_t pos) { size_t size = iseq->body->insns_info.size; const struct iseq_insn_info_entry *insns_info = iseq->body->insns_info.body; const unsigned int *positions = iseq->body->insns_info.positions; const int debug = 0; if (debug) { printf("size: %"PRIuSIZE"\n", size); printf("insns_info[%"PRIuSIZE"]: position: %d, line: %d, pos: %"PRIuSIZE"\n", (size_t)0, positions[0], insns_info[0].line_no, pos); } if (size == 0) { return NULL; } else if (size == 1) { return &insns_info[0]; } else { size_t l = 1, r = size - 1; while (l <= r) { size_t m = l + (r - l) / 2; if (positions[m] == pos) { return &insns_info[m]; } if (positions[m] < pos) { l = m + 1; } else { r = m - 1; } } if (l >= size) { return &insns_info[size-1]; } if (positions[l] > pos) { return &insns_info[l-1]; } return &insns_info[l]; } } static const struct iseq_insn_info_entry * get_insn_info(const rb_iseq_t *iseq, size_t pos) { return get_insn_info_binary_search(iseq, pos); } #endif #if VM_INSN_INFO_TABLE_IMPL == 2 /* succinct bitvector */ static const struct iseq_insn_info_entry * get_insn_info_succinct_bitvector(const rb_iseq_t *iseq, size_t pos) { size_t size = iseq->body->insns_info.size; const struct iseq_insn_info_entry *insns_info = iseq->body->insns_info.body; const unsigned int *positions = iseq->body->insns_info.positions; const int debug = 0; if (debug) { printf("size: %"PRIuSIZE"\n", size); printf("insns_info[%"PRIuSIZE"]: position: %d, line: %d, pos: %"PRIuSIZE"\n", (size_t)0, positions[0], insns_info[0].line_no, pos); } if (size == 0) { return NULL; } else if (size == 1) { return &insns_info[0]; } else { int index; VM_ASSERT(iseq->body->insns_info.succ_index_table != NULL); index = succ_index_lookup(iseq->body->insns_info.succ_index_table, (int)pos); return &insns_info[index-1]; } } static const struct iseq_insn_info_entry * get_insn_info(const rb_iseq_t *iseq, size_t pos) { return get_insn_info_succinct_bitvector(iseq, pos); } #endif #if VM_CHECK_MODE > 0 || VM_INSN_INFO_TABLE_IMPL == 0 static const struct iseq_insn_info_entry * get_insn_info_linear_search(const rb_iseq_t *iseq, size_t pos) { size_t i = 0, size = iseq->body->insns_info.size; const struct iseq_insn_info_entry *insns_info = iseq->body->insns_info.body; const unsigned int *positions = iseq->body->insns_info.positions; const int debug = 0; if (debug) { printf("size: %"PRIuSIZE"\n", size); printf("insns_info[%"PRIuSIZE"]: position: %d, line: %d, pos: %"PRIuSIZE"\n", i, positions[i], insns_info[i].line_no, pos); } if (size == 0) { return NULL; } else if (size == 1) { return &insns_info[0]; } else { for (i=1; i pos) { return &insns_info[i-1]; } } } return &insns_info[i-1]; } #endif #if VM_INSN_INFO_TABLE_IMPL == 0 /* linear search */ static const struct iseq_insn_info_entry * get_insn_info(const rb_iseq_t *iseq, size_t pos) { return get_insn_info_linear_search(iseq, pos); } #endif #if VM_CHECK_MODE > 0 && VM_INSN_INFO_TABLE_IMPL > 0 static void validate_get_insn_info(rb_iseq_t *iseq) { size_t i; for (i = 0; i < iseq->body->iseq_size; i++) { if (get_insn_info_linear_search(iseq, i) != get_insn_info(iseq, i)) { rb_bug("validate_get_insn_info: get_insn_info_linear_search(iseq, %"PRIuSIZE") != get_insn_info(iseq, %"PRIuSIZE")", i, i); } } } #endif unsigned int rb_iseq_line_no(const rb_iseq_t *iseq, size_t pos) { const struct iseq_insn_info_entry *entry = get_insn_info(iseq, pos); if (entry) { return entry->line_no; } else { return 0; } } MJIT_FUNC_EXPORTED rb_event_flag_t rb_iseq_event_flags(const rb_iseq_t *iseq, size_t pos) { const struct iseq_insn_info_entry *entry = get_insn_info(iseq, pos); if (entry) { return entry->events; } else { return 0; } } static VALUE local_var_name(const rb_iseq_t *diseq, VALUE level, VALUE op) { VALUE i; VALUE name; ID lid; for (i = 0; i < level; i++) { diseq = diseq->body->parent_iseq; } lid = diseq->body->local_table[diseq->body->local_table_size - op - 1]; name = rb_id2str(lid); if (!name) { name = rb_sprintf("?%d", diseq->body->local_table_size - (int)op); } else if (!rb_str_symname_p(name)) { name = rb_str_inspect(name); } return name; } int rb_insn_unified_local_var_level(VALUE); VALUE rb_insn_operand_intern(const rb_iseq_t *iseq, VALUE insn, int op_no, VALUE op, int len, size_t pos, const VALUE *pnop, VALUE child) { const char *types = insn_op_types(insn); char type = types[op_no]; VALUE ret = Qundef; switch (type) { case TS_OFFSET: /* LONG */ ret = rb_sprintf("%"PRIdVALUE, (VALUE)(pos + len + op)); break; case TS_NUM: /* ULONG */ if (insn == BIN(defined) && op_no == 0) { enum defined_type deftype = (enum defined_type)op; if (deftype == DEFINED_FUNC) { ret = rb_fstring_cstr("func"); break; } if (deftype == DEFINED_REF) { ret = rb_fstring_cstr("ref"); break; } ret = rb_iseq_defined_string(deftype); if (ret) break; } else if (insn == BIN(branchiftype) && op_no == 0) { const char *type_str = rb_type_str((enum ruby_value_type)op); if (type_str) { ret = rb_str_new_cstr(type_str); break; } } ret = rb_sprintf("%"PRIuVALUE, op); break; case TS_LINDEX:{ int level; if (types[op_no+1] == TS_NUM && pnop) { ret = local_var_name(iseq, *pnop, op - VM_ENV_DATA_SIZE); } else if ((level = rb_insn_unified_local_var_level(insn)) >= 0) { ret = local_var_name(iseq, (VALUE)level, op - VM_ENV_DATA_SIZE); } else { ret = rb_inspect(INT2FIX(op)); } break; } case TS_ID: /* ID (symbol) */ ret = rb_inspect(ID2SYM(op)); break; case TS_VALUE: /* VALUE */ op = obj_resurrect(op); if (insn == BIN(defined) && op_no == 1 && FIXNUM_P(op)) { /* should be DEFINED_REF */ int type = NUM2INT(op); if (type) { if (type & 1) { ret = rb_sprintf(":$%c", (type >> 1)); } else { ret = rb_sprintf(":$%d", (type >> 1)); } break; } } ret = rb_inspect(op); if (CLASS_OF(op) == rb_cISeq) { if (child) { rb_ary_push(child, op); } } break; case TS_ISEQ: /* iseq */ { if (op) { const rb_iseq_t *iseq = rb_iseq_check((rb_iseq_t *)op); ret = iseq->body->location.label; if (child) { rb_ary_push(child, (VALUE)iseq); } } else { ret = rb_str_new2("nil"); } break; } case TS_GENTRY: { struct rb_global_entry *entry = (struct rb_global_entry *)op; ret = rb_str_dup(rb_id2str(entry->id)); } break; case TS_IC: ret = rb_sprintf("", (union iseq_inline_storage_entry *)op - iseq->body->is_entries); break; case TS_CALLINFO: { struct rb_call_info *ci = (struct rb_call_info *)op; VALUE ary = rb_ary_new(); if (ci->mid) { rb_ary_push(ary, rb_sprintf("mid:%"PRIsVALUE, rb_id2str(ci->mid))); } rb_ary_push(ary, rb_sprintf("argc:%d", ci->orig_argc)); if (ci->flag & VM_CALL_KWARG) { struct rb_call_info_kw_arg *kw_args = ((struct rb_call_info_with_kwarg *)ci)->kw_arg; VALUE kw_ary = rb_ary_new_from_values(kw_args->keyword_len, kw_args->keywords); rb_ary_push(ary, rb_sprintf("kw:[%"PRIsVALUE"]", rb_ary_join(kw_ary, rb_str_new2(",")))); } if (ci->flag) { VALUE flags = rb_ary_new(); # define CALL_FLAG(n) if (ci->flag & VM_CALL_##n) rb_ary_push(flags, rb_str_new2(#n)) CALL_FLAG(ARGS_SPLAT); CALL_FLAG(ARGS_BLOCKARG); CALL_FLAG(FCALL); CALL_FLAG(VCALL); CALL_FLAG(ARGS_SIMPLE); CALL_FLAG(BLOCKISEQ); CALL_FLAG(TAILCALL); CALL_FLAG(SUPER); CALL_FLAG(KWARG); CALL_FLAG(KW_SPLAT); CALL_FLAG(OPT_SEND); /* maybe not reachable */ rb_ary_push(ary, rb_ary_join(flags, rb_str_new2("|"))); } ret = rb_sprintf("", rb_ary_join(ary, rb_str_new2(", "))); } break; case TS_CALLCACHE: ret = rb_str_new2(""); break; case TS_CDHASH: ret = rb_str_new2(""); break; case TS_FUNCPTR: { #ifdef HAVE_DLADDR Dl_info info; if (dladdr((void *)op, &info) && info.dli_sname) { ret = rb_str_new_cstr(info.dli_sname); break; } #endif ret = rb_str_new2(""); } break; default: rb_bug("insn_operand_intern: unknown operand type: %c", type); } return ret; } /** * Disassemble a instruction * Iseq -> Iseq inspect object */ int rb_iseq_disasm_insn(VALUE ret, const VALUE *code, size_t pos, const rb_iseq_t *iseq, VALUE child) { VALUE insn = code[pos]; int len = insn_len(insn); int j; const char *types = insn_op_types(insn); VALUE str = rb_str_new(0, 0); const char *insn_name_buff; insn_name_buff = insn_name(insn); if (1) { rb_str_catf(str, "%04"PRIuSIZE" %-16s ", pos, insn_name_buff); } else { rb_str_catf(str, "%04"PRIuSIZE" %-16.*s ", pos, (int)strcspn(insn_name_buff, "_"), insn_name_buff); } for (j = 0; types[j]; j++) { VALUE opstr = rb_insn_operand_intern(iseq, insn, j, code[pos + j + 1], len, pos, &code[pos + j + 2], child); rb_str_concat(str, opstr); if (types[j + 1]) { rb_str_cat2(str, ", "); } } { unsigned int line_no = rb_iseq_line_no(iseq, pos); unsigned int prev = pos == 0 ? 0 : rb_iseq_line_no(iseq, pos - 1); if (line_no && line_no != prev) { long slen = RSTRING_LEN(str); slen = (slen > 70) ? 0 : (70 - slen); str = rb_str_catf(str, "%*s(%4d)", (int)slen, "", line_no); } } { rb_event_flag_t events = rb_iseq_event_flags(iseq, pos); if (events) { str = rb_str_catf(str, "[%s%s%s%s%s%s%s%s%s]", events & RUBY_EVENT_LINE ? "Li" : "", events & RUBY_EVENT_CLASS ? "Cl" : "", events & RUBY_EVENT_END ? "En" : "", events & RUBY_EVENT_CALL ? "Ca" : "", events & RUBY_EVENT_RETURN ? "Re" : "", events & RUBY_EVENT_C_CALL ? "Cc" : "", events & RUBY_EVENT_C_RETURN ? "Cr" : "", events & RUBY_EVENT_B_CALL ? "Bc" : "", events & RUBY_EVENT_B_RETURN ? "Br" : ""); } } if (ret) { rb_str_cat2(str, "\n"); rb_str_concat(ret, str); } else { printf("%s\n", RSTRING_PTR(str)); } return len; } static const char * catch_type(int type) { switch (type) { case CATCH_TYPE_RESCUE: return "rescue"; case CATCH_TYPE_ENSURE: return "ensure"; case CATCH_TYPE_RETRY: return "retry"; case CATCH_TYPE_BREAK: return "break"; case CATCH_TYPE_REDO: return "redo"; case CATCH_TYPE_NEXT: return "next"; default: rb_bug("unknown catch type (%d)", type); return 0; } } static VALUE iseq_inspect(const rb_iseq_t *iseq) { if (!iseq->body->location.label) { return rb_sprintf("#"); } else { return rb_sprintf("#", iseq->body->location.label, rb_iseq_path(iseq), iseq->body->location.code_location.beg_pos.lineno, iseq->body->location.code_location.beg_pos.lineno, iseq->body->location.code_location.beg_pos.column, iseq->body->location.code_location.end_pos.lineno, iseq->body->location.code_location.end_pos.column); } } VALUE rb_iseq_disasm(const rb_iseq_t *iseq) { VALUE *code; VALUE str = rb_str_new(0, 0); VALUE child = rb_ary_tmp_new(3); unsigned int size; unsigned int i; long l; size_t n; enum {header_minlen = 72}; st_table *done_iseq = 0; rb_secure(1); size = iseq->body->iseq_size; rb_str_cat2(str, "== disasm: "); rb_str_concat(str, iseq_inspect(iseq)); rb_str_catf(str, " (catch: %s)", iseq->body->catch_except_p ? "TRUE" : "FALSE"); if ((l = RSTRING_LEN(str)) < header_minlen) { rb_str_resize(str, header_minlen); memset(RSTRING_PTR(str) + l, '=', header_minlen - l); } rb_str_cat2(str, "\n"); /* show catch table information */ if (iseq->body->catch_table) { rb_str_cat2(str, "== catch table\n"); } if (iseq->body->catch_table) { for (i = 0; i < iseq->body->catch_table->size; i++) { const struct iseq_catch_table_entry *entry = &iseq->body->catch_table->entries[i]; rb_str_catf(str, "| catch type: %-6s st: %04d ed: %04d sp: %04d cont: %04d\n", catch_type((int)entry->type), (int)entry->start, (int)entry->end, (int)entry->sp, (int)entry->cont); if (entry->iseq && !(done_iseq && st_is_member(done_iseq, (st_data_t)entry->iseq))) { rb_str_concat(str, rb_iseq_disasm(rb_iseq_check(entry->iseq))); if (!done_iseq) done_iseq = st_init_numtable(); st_insert(done_iseq, (st_data_t)entry->iseq, (st_data_t)0); } } } if (iseq->body->catch_table) { rb_str_cat2(str, "|-------------------------------------" "-----------------------------------\n"); } /* show local table information */ if (iseq->body->local_table) { rb_str_catf(str, "local table (size: %d, argc: %d " "[opts: %d, rest: %d, post: %d, block: %d, kw: %d@%d, kwrest: %d])\n", iseq->body->local_table_size, iseq->body->param.lead_num, iseq->body->param.opt_num, iseq->body->param.flags.has_rest ? iseq->body->param.rest_start : -1, iseq->body->param.post_num, iseq->body->param.flags.has_block ? iseq->body->param.block_start : -1, iseq->body->param.flags.has_kw ? iseq->body->param.keyword->num : -1, iseq->body->param.flags.has_kw ? iseq->body->param.keyword->required_num : -1, iseq->body->param.flags.has_kwrest ? iseq->body->param.keyword->rest_start : -1); for (i = iseq->body->local_table_size; i > 0;) { int li = iseq->body->local_table_size - --i - 1; long width; VALUE name = local_var_name(iseq, 0, i); char argi[0x100] = ""; char opti[0x100] = ""; if (iseq->body->param.flags.has_opt) { int argc = iseq->body->param.lead_num; int opts = iseq->body->param.opt_num; if (li >= argc && li < argc + opts) { snprintf(opti, sizeof(opti), "Opt=%"PRIdVALUE, iseq->body->param.opt_table[li - argc]); } } snprintf(argi, sizeof(argi), "%s%s%s%s%s%s", /* arg, opts, rest, post, kwrest, block */ iseq->body->param.lead_num > li ? "Arg" : "", opti, (iseq->body->param.flags.has_rest && iseq->body->param.rest_start == li) ? "Rest" : "", (iseq->body->param.flags.has_post && iseq->body->param.post_start <= li && li < iseq->body->param.post_start + iseq->body->param.post_num) ? "Post" : "", (iseq->body->param.flags.has_kwrest && iseq->body->param.keyword->rest_start == li) ? "Kwrest" : "", (iseq->body->param.flags.has_block && iseq->body->param.block_start == li) ? "Block" : ""); rb_str_catf(str, "[%2d] ", i + 1); width = RSTRING_LEN(str) + 11; rb_str_append(str, name); if (*argi) rb_str_catf(str, "<%s>", argi); if ((width -= RSTRING_LEN(str)) > 0) rb_str_catf(str, "%*s", (int)width, ""); } rb_str_cat2(str, "\n"); } /* show each line */ code = rb_iseq_original_iseq(iseq); for (n = 0; n < size;) { n += rb_iseq_disasm_insn(str, code, n, iseq, child); } for (l = 0; l < RARRAY_LEN(child); l++) { VALUE isv = rb_ary_entry(child, l); if (done_iseq && st_is_member(done_iseq, (st_data_t)isv)) continue; rb_str_concat(str, rb_iseq_disasm(rb_iseq_check((rb_iseq_t *)isv))); } if (done_iseq) st_free_table(done_iseq); return str; } static VALUE rb_iseq_all_children(const rb_iseq_t *iseq) { unsigned int i; VALUE *code = rb_iseq_original_iseq(iseq); VALUE all_children = rb_obj_hide(rb_ident_hash_new()); VALUE child; if (iseq->body->catch_table) { for (i = 0; i < iseq->body->catch_table->size; i++) { const struct iseq_catch_table_entry *entry = &iseq->body->catch_table->entries[i]; child = (VALUE)entry->iseq; if (child) { rb_hash_aset(all_children, child, Qtrue); } } } for (i=0; ibody->iseq_size;) { VALUE insn = code[i]; int len = insn_len(insn); const char *types = insn_op_types(insn); int j; for (j=0; types[j]; j++) { switch (types[j]) { case TS_ISEQ: child = code[i+j+1]; if (child) { rb_hash_aset(all_children, child, Qtrue); } break; default: break; } } i += len; } return all_children; } /* * call-seq: * iseq.disasm -> str * iseq.disassemble -> str * * Returns the instruction sequence as a +String+ in human readable form. * * puts RubyVM::InstructionSequence.compile('1 + 2').disasm * * Produces: * * == disasm: @>========== * 0000 trace 1 ( 1) * 0002 putobject 1 * 0004 putobject 2 * 0006 opt_plus * 0008 leave */ static VALUE iseqw_disasm(VALUE self) { return rb_iseq_disasm(iseqw_check(self)); } static int iseqw_each_child_i(VALUE key, VALUE value, VALUE dummy) { rb_yield(iseqw_new((const rb_iseq_t *)key)); return ST_CONTINUE; } /* * call-seq: * iseq.each_child{|child_iseq| ...} -> iseq * * Iterate all direct child instruction sequences. * Iteration order is implementation/version defined * so that people should not rely on the order. */ static VALUE iseqw_each_child(VALUE self) { const rb_iseq_t *iseq = iseqw_check(self); VALUE all_children = rb_iseq_all_children(iseq); rb_hash_foreach(all_children, iseqw_each_child_i, Qnil); return self; } static void push_event_info(const rb_iseq_t *iseq, rb_event_flag_t events, int line, VALUE ary) { #define C(ev, cstr, l) if (events & ev) rb_ary_push(ary, rb_ary_new_from_args(2, l, ID2SYM(rb_intern(cstr)))); C(RUBY_EVENT_CLASS, "class", rb_iseq_first_lineno(iseq)); C(RUBY_EVENT_CALL, "call", rb_iseq_first_lineno(iseq)); C(RUBY_EVENT_B_CALL, "b_call", rb_iseq_first_lineno(iseq)); C(RUBY_EVENT_LINE, "line", INT2FIX(line)); C(RUBY_EVENT_END, "end", INT2FIX(line)); C(RUBY_EVENT_RETURN, "return", INT2FIX(line)); C(RUBY_EVENT_B_RETURN, "b_return", INT2FIX(line)); #undef C } /* * call-seq: * iseq.trace_points -> ary * * Return trace points in the instruction sequence. * Return an array of [line, event_symbol] pair. */ static VALUE iseqw_trace_points(VALUE self) { const rb_iseq_t *iseq = iseqw_check(self); unsigned int i; VALUE ary = rb_ary_new(); for (i=0; ibody->insns_info.size; i++) { const struct iseq_insn_info_entry *entry = &iseq->body->insns_info.body[i]; if (entry->events) { push_event_info(iseq, entry->events, entry->line_no, ary); } } return ary; } /* * Returns the instruction sequence containing the given proc or method. * * For example, using irb: * * # a proc * > p = proc { num = 1 + 2 } * > RubyVM::InstructionSequence.of(p) * > #=> * * # for a method * > def foo(bar); puts bar; end * > RubyVM::InstructionSequence.of(method(:foo)) * > #=> * * Using ::compile_file: * * # /tmp/iseq_of.rb * def hello * puts "hello, world" * end * * $a_global_proc = proc { str = 'a' + 'b' } * * # in irb * > require '/tmp/iseq_of.rb' * * # first the method hello * > RubyVM::InstructionSequence.of(method(:hello)) * > #=> # * * # then the global proc * > RubyVM::InstructionSequence.of($a_global_proc) * > #=> # */ static VALUE iseqw_s_of(VALUE klass, VALUE body) { const rb_iseq_t *iseq = NULL; rb_secure(1); if (rb_obj_is_proc(body)) { iseq = vm_proc_iseq(body); if (!rb_obj_is_iseq((VALUE)iseq)) { iseq = NULL; } } else { iseq = rb_method_iseq(body); } return iseq ? iseqw_new(iseq) : Qnil; } /* * call-seq: * InstructionSequence.disasm(body) -> str * InstructionSequence.disassemble(body) -> str * * Takes +body+, a Method or Proc object, and returns a String with the * human readable instructions for +body+. * * For a Method object: * * # /tmp/method.rb * def hello * puts "hello, world" * end * * puts RubyVM::InstructionSequence.disasm(method(:hello)) * * Produces: * * == disasm: ============ * 0000 trace 8 ( 1) * 0002 trace 1 ( 2) * 0004 putself * 0005 putstring "hello, world" * 0007 send :puts, 1, nil, 8, * 0013 trace 16 ( 3) * 0015 leave ( 2) * * For a Proc: * * # /tmp/proc.rb * p = proc { num = 1 + 2 } * puts RubyVM::InstructionSequence.disasm(p) * * Produces: * * == disasm: @/tmp/proc.rb>=== * == catch table * | catch type: redo st: 0000 ed: 0012 sp: 0000 cont: 0000 * | catch type: next st: 0000 ed: 0012 sp: 0000 cont: 0012 * |------------------------------------------------------------------------ * local table (size: 2, argc: 0 [opts: 0, rest: -1, post: 0, block: -1] s1) * [ 2] num * 0000 trace 1 ( 1) * 0002 putobject 1 * 0004 putobject 2 * 0006 opt_plus * 0008 dup * 0009 setlocal num, 0 * 0012 leave * */ static VALUE iseqw_s_disasm(VALUE klass, VALUE body) { VALUE iseqw = iseqw_s_of(klass, body); return NIL_P(iseqw) ? Qnil : rb_iseq_disasm(iseqw_check(iseqw)); } const char * ruby_node_name(int node) { switch (node) { #include "node_name.inc" default: rb_bug("unknown node (%d)", node); return 0; } } #define DECL_SYMBOL(name) \ static VALUE sym_##name #define INIT_SYMBOL(name) \ sym_##name = ID2SYM(rb_intern(#name)) static VALUE register_label(struct st_table *table, unsigned long idx) { VALUE sym = rb_str_intern(rb_sprintf("label_%lu", idx)); st_insert(table, idx, sym); return sym; } static VALUE exception_type2symbol(VALUE type) { ID id; switch (type) { case CATCH_TYPE_RESCUE: CONST_ID(id, "rescue"); break; case CATCH_TYPE_ENSURE: CONST_ID(id, "ensure"); break; case CATCH_TYPE_RETRY: CONST_ID(id, "retry"); break; case CATCH_TYPE_BREAK: CONST_ID(id, "break"); break; case CATCH_TYPE_REDO: CONST_ID(id, "redo"); break; case CATCH_TYPE_NEXT: CONST_ID(id, "next"); break; default: rb_bug("exception_type2symbol: unknown type %d", (int)type); } return ID2SYM(id); } static int cdhash_each(VALUE key, VALUE value, VALUE ary) { rb_ary_push(ary, obj_resurrect(key)); rb_ary_push(ary, value); return ST_CONTINUE; } static VALUE iseq_data_to_ary(const rb_iseq_t *iseq) { unsigned int i; long l; const struct iseq_insn_info_entry *prev_insn_info; unsigned int pos; int last_line = 0; VALUE *seq, *iseq_original; VALUE val = rb_ary_new(); VALUE type; /* Symbol */ VALUE locals = rb_ary_new(); VALUE params = rb_hash_new(); VALUE body = rb_ary_new(); /* [[:insn1, ...], ...] */ VALUE nbody; VALUE exception = rb_ary_new(); /* [[....]] */ VALUE misc = rb_hash_new(); static VALUE insn_syms[VM_INSTRUCTION_SIZE/2]; /* w/o-trace only */ struct st_table *labels_table = st_init_numtable(); DECL_SYMBOL(top); DECL_SYMBOL(method); DECL_SYMBOL(block); DECL_SYMBOL(class); DECL_SYMBOL(rescue); DECL_SYMBOL(ensure); DECL_SYMBOL(eval); DECL_SYMBOL(main); DECL_SYMBOL(plain); if (sym_top == 0) { int i; for (i=0; ibody->type) { case ISEQ_TYPE_TOP: type = sym_top; break; case ISEQ_TYPE_METHOD: type = sym_method; break; case ISEQ_TYPE_BLOCK: type = sym_block; break; case ISEQ_TYPE_CLASS: type = sym_class; break; case ISEQ_TYPE_RESCUE: type = sym_rescue; break; case ISEQ_TYPE_ENSURE: type = sym_ensure; break; case ISEQ_TYPE_EVAL: type = sym_eval; break; case ISEQ_TYPE_MAIN: type = sym_main; break; case ISEQ_TYPE_PLAIN: type = sym_plain; break; default: rb_bug("unsupported iseq type"); }; /* locals */ for (i=0; ibody->local_table_size; i++) { ID lid = iseq->body->local_table[i]; if (lid) { if (rb_id2str(lid)) { rb_ary_push(locals, ID2SYM(lid)); } else { /* hidden variable from id_internal() */ rb_ary_push(locals, ULONG2NUM(iseq->body->local_table_size-i+1)); } } else { rb_ary_push(locals, ID2SYM(rb_intern("#arg_rest"))); } } /* params */ { int j; if (iseq->body->param.flags.has_opt) { int len = iseq->body->param.opt_num + 1; VALUE arg_opt_labels = rb_ary_new2(len); for (j = 0; j < len; j++) { VALUE l = register_label(labels_table, iseq->body->param.opt_table[j]); rb_ary_push(arg_opt_labels, l); } rb_hash_aset(params, ID2SYM(rb_intern("opt")), arg_opt_labels); } /* commit */ if (iseq->body->param.flags.has_lead) rb_hash_aset(params, ID2SYM(rb_intern("lead_num")), INT2FIX(iseq->body->param.lead_num)); if (iseq->body->param.flags.has_post) rb_hash_aset(params, ID2SYM(rb_intern("post_num")), INT2FIX(iseq->body->param.post_num)); if (iseq->body->param.flags.has_post) rb_hash_aset(params, ID2SYM(rb_intern("post_start")), INT2FIX(iseq->body->param.post_start)); if (iseq->body->param.flags.has_rest) rb_hash_aset(params, ID2SYM(rb_intern("rest_start")), INT2FIX(iseq->body->param.rest_start)); if (iseq->body->param.flags.has_block) rb_hash_aset(params, ID2SYM(rb_intern("block_start")), INT2FIX(iseq->body->param.block_start)); if (iseq->body->param.flags.has_kw) { VALUE keywords = rb_ary_new(); int i, j; for (i=0; ibody->param.keyword->required_num; i++) { rb_ary_push(keywords, ID2SYM(iseq->body->param.keyword->table[i])); } for (j=0; ibody->param.keyword->num; i++, j++) { VALUE key = rb_ary_new_from_args(1, ID2SYM(iseq->body->param.keyword->table[i])); if (iseq->body->param.keyword->default_values[j] != Qundef) { rb_ary_push(key, iseq->body->param.keyword->default_values[j]); } rb_ary_push(keywords, key); } rb_hash_aset(params, ID2SYM(rb_intern("kwbits")), INT2FIX(iseq->body->param.keyword->bits_start)); rb_hash_aset(params, ID2SYM(rb_intern("keyword")), keywords); } if (iseq->body->param.flags.has_kwrest) rb_hash_aset(params, ID2SYM(rb_intern("kwrest")), INT2FIX(iseq->body->param.keyword->rest_start)); if (iseq->body->param.flags.ambiguous_param0) rb_hash_aset(params, ID2SYM(rb_intern("ambiguous_param0")), Qtrue); } /* body */ iseq_original = rb_iseq_original_iseq((rb_iseq_t *)iseq); for (seq = iseq_original; seq < iseq_original + iseq->body->iseq_size; ) { VALUE insn = *seq++; int j, len = insn_len(insn); VALUE *nseq = seq + len - 1; VALUE ary = rb_ary_new2(len); rb_ary_push(ary, insn_syms[insn%numberof(insn_syms)]); for (j=0; jid)); } break; case TS_IC: { union iseq_inline_storage_entry *is = (union iseq_inline_storage_entry *)*seq; rb_ary_push(ary, INT2FIX(is - iseq->body->is_entries)); } break; case TS_CALLINFO: { struct rb_call_info *ci = (struct rb_call_info *)*seq; VALUE e = rb_hash_new(); int orig_argc = ci->orig_argc; rb_hash_aset(e, ID2SYM(rb_intern("mid")), ci->mid ? ID2SYM(ci->mid) : Qnil); rb_hash_aset(e, ID2SYM(rb_intern("flag")), UINT2NUM(ci->flag)); if (ci->flag & VM_CALL_KWARG) { struct rb_call_info_with_kwarg *ci_kw = (struct rb_call_info_with_kwarg *)ci; int i; VALUE kw = rb_ary_new2((long)ci_kw->kw_arg->keyword_len); orig_argc -= ci_kw->kw_arg->keyword_len; for (i = 0; i < ci_kw->kw_arg->keyword_len; i++) { rb_ary_push(kw, ci_kw->kw_arg->keywords[i]); } rb_hash_aset(e, ID2SYM(rb_intern("kw_arg")), kw); } rb_hash_aset(e, ID2SYM(rb_intern("orig_argc")), INT2FIX(orig_argc)); rb_ary_push(ary, e); } break; case TS_CALLCACHE: rb_ary_push(ary, Qfalse); break; case TS_ID: rb_ary_push(ary, ID2SYM(*seq)); break; case TS_CDHASH: { VALUE hash = *seq; VALUE val = rb_ary_new(); int i; rb_hash_foreach(hash, cdhash_each, val); for (i=0; ibody->catch_table) for (i=0; ibody->catch_table->size; i++) { VALUE ary = rb_ary_new(); const struct iseq_catch_table_entry *entry = &iseq->body->catch_table->entries[i]; rb_ary_push(ary, exception_type2symbol(entry->type)); if (entry->iseq) { rb_ary_push(ary, iseq_data_to_ary(rb_iseq_check(entry->iseq))); } else { rb_ary_push(ary, Qnil); } rb_ary_push(ary, register_label(labels_table, entry->start)); rb_ary_push(ary, register_label(labels_table, entry->end)); rb_ary_push(ary, register_label(labels_table, entry->cont)); rb_ary_push(ary, UINT2NUM(entry->sp)); rb_ary_push(exception, ary); } /* make body with labels and insert line number */ body = rb_ary_new(); prev_insn_info = NULL; for (l=0, pos=0; lline_no; rb_event_flag_t events = info->events; if (line > 0 && last_line != line) { rb_ary_push(body, INT2FIX(line)); last_line = line; } #define CHECK_EVENT(ev) if (events & ev) rb_ary_push(body, ID2SYM(rb_intern(#ev))); CHECK_EVENT(RUBY_EVENT_LINE); CHECK_EVENT(RUBY_EVENT_CLASS); CHECK_EVENT(RUBY_EVENT_END); CHECK_EVENT(RUBY_EVENT_CALL); CHECK_EVENT(RUBY_EVENT_RETURN); CHECK_EVENT(RUBY_EVENT_B_CALL); CHECK_EVENT(RUBY_EVENT_B_RETURN); #undef CHECK_EVENT prev_insn_info = info; } rb_ary_push(body, ary); pos += RARRAY_LENINT(ary); /* reject too huge data */ } RB_GC_GUARD(nbody); st_free_table(labels_table); rb_hash_aset(misc, ID2SYM(rb_intern("arg_size")), INT2FIX(iseq->body->param.size)); rb_hash_aset(misc, ID2SYM(rb_intern("local_size")), INT2FIX(iseq->body->local_table_size)); rb_hash_aset(misc, ID2SYM(rb_intern("stack_max")), INT2FIX(iseq->body->stack_max)); rb_hash_aset(misc, ID2SYM(rb_intern("code_location")), rb_ary_new_from_args(4, INT2FIX(iseq->body->location.code_location.beg_pos.lineno), INT2FIX(iseq->body->location.code_location.beg_pos.column), INT2FIX(iseq->body->location.code_location.end_pos.lineno), INT2FIX(iseq->body->location.code_location.end_pos.column))); /* * [:magic, :major_version, :minor_version, :format_type, :misc, * :name, :path, :absolute_path, :start_lineno, :type, :locals, :args, * :catch_table, :bytecode] */ rb_ary_push(val, rb_str_new2("YARVInstructionSequence/SimpleDataFormat")); rb_ary_push(val, INT2FIX(ISEQ_MAJOR_VERSION)); /* major */ rb_ary_push(val, INT2FIX(ISEQ_MINOR_VERSION)); /* minor */ rb_ary_push(val, INT2FIX(1)); rb_ary_push(val, misc); rb_ary_push(val, iseq->body->location.label); rb_ary_push(val, rb_iseq_path(iseq)); rb_ary_push(val, rb_iseq_realpath(iseq)); rb_ary_push(val, iseq->body->location.first_lineno); rb_ary_push(val, type); rb_ary_push(val, locals); rb_ary_push(val, params); rb_ary_push(val, exception); rb_ary_push(val, body); return val; } VALUE rb_iseq_parameters(const rb_iseq_t *iseq, int is_proc) { int i, r; VALUE a, args = rb_ary_new2(iseq->body->param.size); ID req, opt, rest, block, key, keyrest; #define PARAM_TYPE(type) rb_ary_push(a = rb_ary_new2(2), ID2SYM(type)) #define PARAM_ID(i) iseq->body->local_table[(i)] #define PARAM(i, type) ( \ PARAM_TYPE(type), \ rb_id2str(PARAM_ID(i)) ? \ rb_ary_push(a, ID2SYM(PARAM_ID(i))) : \ a) CONST_ID(req, "req"); CONST_ID(opt, "opt"); if (is_proc) { for (i = 0; i < iseq->body->param.lead_num; i++) { PARAM_TYPE(opt); rb_ary_push(a, rb_id2str(PARAM_ID(i)) ? ID2SYM(PARAM_ID(i)) : Qnil); rb_ary_push(args, a); } } else { for (i = 0; i < iseq->body->param.lead_num; i++) { rb_ary_push(args, PARAM(i, req)); } } r = iseq->body->param.lead_num + iseq->body->param.opt_num; for (; i < r; i++) { PARAM_TYPE(opt); if (rb_id2str(PARAM_ID(i))) { rb_ary_push(a, ID2SYM(PARAM_ID(i))); } rb_ary_push(args, a); } if (iseq->body->param.flags.has_rest) { CONST_ID(rest, "rest"); rb_ary_push(args, PARAM(iseq->body->param.rest_start, rest)); } r = iseq->body->param.post_start + iseq->body->param.post_num; if (is_proc) { for (i = iseq->body->param.post_start; i < r; i++) { PARAM_TYPE(opt); rb_ary_push(a, rb_id2str(PARAM_ID(i)) ? ID2SYM(PARAM_ID(i)) : Qnil); rb_ary_push(args, a); } } else { for (i = iseq->body->param.post_start; i < r; i++) { rb_ary_push(args, PARAM(i, req)); } } if (iseq->body->param.flags.has_kw) { i = 0; if (iseq->body->param.keyword->required_num > 0) { ID keyreq; CONST_ID(keyreq, "keyreq"); for (; i < iseq->body->param.keyword->required_num; i++) { PARAM_TYPE(keyreq); if (rb_id2str(iseq->body->param.keyword->table[i])) { rb_ary_push(a, ID2SYM(iseq->body->param.keyword->table[i])); } rb_ary_push(args, a); } } CONST_ID(key, "key"); for (; i < iseq->body->param.keyword->num; i++) { PARAM_TYPE(key); if (rb_id2str(iseq->body->param.keyword->table[i])) { rb_ary_push(a, ID2SYM(iseq->body->param.keyword->table[i])); } rb_ary_push(args, a); } } if (iseq->body->param.flags.has_kwrest) { CONST_ID(keyrest, "keyrest"); rb_ary_push(args, PARAM(iseq->body->param.keyword->rest_start, keyrest)); } if (iseq->body->param.flags.has_block) { CONST_ID(block, "block"); rb_ary_push(args, PARAM(iseq->body->param.block_start, block)); } return args; } VALUE rb_iseq_defined_string(enum defined_type type) { static const char expr_names[][18] = { "nil", "instance-variable", "local-variable", "global-variable", "class variable", "constant", "method", "yield", "super", "self", "true", "false", "assignment", "expression", }; const char *estr; VALUE *defs, str; if ((unsigned)(type - 1) >= (unsigned)numberof(expr_names)) return 0; estr = expr_names[type - 1]; if (!estr[0]) return 0; defs = GET_VM()->defined_strings; if (!defs) { defs = ruby_xcalloc(numberof(expr_names), sizeof(VALUE)); GET_VM()->defined_strings = defs; } str = defs[type-1]; if (!str) { str = rb_str_new_cstr(estr); OBJ_FREEZE(str); defs[type-1] = str; rb_gc_register_mark_object(str); } return str; } #define TRACE_INSN_P(insn) ((insn) >= VM_INSTRUCTION_SIZE/2) #if OPT_DIRECT_THREADED_CODE || OPT_CALL_THREADED_CODE #define INSN_CODE(insn) ((VALUE)table[insn]) #else #define INSN_CODE(insn) (insn) #endif void rb_iseq_trace_set(const rb_iseq_t *iseq, rb_event_flag_t turnon_events) { VM_ASSERT((turnon_events & ~ISEQ_TRACE_EVENTS) == 0); if (iseq->aux.trace_events == turnon_events) { return; } if (iseq->flags & ISEQ_USE_COMPILE_DATA) { /* this is building ISeq */ return; } else { unsigned int i; VALUE *iseq_encoded = (VALUE *)iseq->body->iseq_encoded; #if OPT_DIRECT_THREADED_CODE || OPT_CALL_THREADED_CODE VALUE *code = rb_iseq_original_iseq(iseq); const void * const *table = rb_vm_get_insns_address_table(); #else const VALUE *code = iseq->body->iseq_encoded; #endif ((rb_iseq_t *)iseq)->aux.trace_events = turnon_events; for (i=0; ibody->iseq_size;) { int insn = (int)code[i]; rb_event_flag_t events = rb_iseq_event_flags(iseq, i); if (events & turnon_events) { if (!TRACE_INSN_P(insn)) { iseq_encoded[i] = INSN_CODE(insn + VM_INSTRUCTION_SIZE/2); } } else if (TRACE_INSN_P(insn)) { iseq_encoded[i] = INSN_CODE(insn - VM_INSTRUCTION_SIZE/2); } i += insn_len(insn); } /* clear for debugging: ISEQ_ORIGINAL_ISEQ_CLEAR(iseq); */ } } static int trace_set_i(void *vstart, void *vend, size_t stride, void *data) { rb_event_flag_t turnon_events = *(rb_event_flag_t *)data; VALUE v = (VALUE)vstart; for (; v != (VALUE)vend; v += stride) { if (rb_obj_is_iseq(v)) { rb_iseq_trace_set(rb_iseq_check((rb_iseq_t *)v), turnon_events); } } return 0; } void rb_iseq_trace_set_all(rb_event_flag_t turnon_events) { rb_objspace_each_objects(trace_set_i, &turnon_events); } void rb_iseq_trace_on_all(void) { rb_iseq_trace_set_all(RUBY_EVENT_TRACEPOINT_ALL); } VALUE rb_iseqw_local_variables(VALUE iseqval) { return rb_iseq_local_variables(iseqw_check(iseqval)); } /* * call-seq: * iseq.to_binary(extra_data = nil) -> binary str * * Returns serialized iseq binary format data as a String object. * A corresponding iseq object is created by * RubyVM::InstructionSequence.load_from_binary() method. * * String extra_data will be saved with binary data. * You can access this data with * RubyVM::InstructionSequence.load_from_binary_extra_data(binary). * * Note that the translated binary data is not portable. * You can not move this binary data to another machine. * You can not use the binary data which is created by another * version/another architecture of Ruby. */ static VALUE iseqw_to_binary(int argc, VALUE *argv, VALUE self) { VALUE opt; rb_scan_args(argc, argv, "01", &opt); return iseq_ibf_dump(iseqw_check(self), opt); } /* * call-seq: * RubyVM::InstructionSequence.load_from_binary(binary) -> iseq * * Load an iseq object from binary format String object * created by RubyVM::InstructionSequence.to_binary. * * This loader does not have a verifier, so that loading broken/modified * binary causes critical problem. * * You should not load binary data provided by others. * You should use binary data translated by yourself. */ static VALUE iseqw_s_load_from_binary(VALUE self, VALUE str) { return iseqw_new(iseq_ibf_load(str)); } /* * call-seq: * RubyVM::InstructionSequence.load_from_binary_extra_data(binary) -> str * * Load extra data embed into binary format String object. */ static VALUE iseqw_s_load_from_binary_extra_data(VALUE self, VALUE str) { return iseq_ibf_load_extra_data(str); } #if VM_INSN_INFO_TABLE_IMPL == 2 /* An implementation of succinct bit-vector for insn_info table. * * A succinct bit-vector is a small and efficient data structure that provides * a bit-vector augmented with an index for O(1) rank operation: * * rank(bv, n): the number of 1's within a range from index 0 to index n * * This can be used to lookup insn_info table from PC. * For example, consider the following iseq and insn_info_table: * * iseq insn_info_table * PC insn+operand position lineno event * 0: insn1 0: 1 [Li] * 2: insn2 2: 2 [Li] <= (A) * 5: insn3 8: 3 [Li] <= (B) * 8: insn4 * * In this case, a succinct bit-vector whose indexes 0, 2, 8 is "1" and * other indexes is "0", i.e., "101000001", is created. * To lookup the lineno of insn2, calculate rank("10100001", 2) = 2, so * the line (A) is the entry in question. * To lookup the lineno of insn4, calculate rank("10100001", 8) = 3, so * the line (B) is the entry in question. * * A naive implementatoin of succinct bit-vector works really well * not only for large size but also for small size. However, it has * tiny overhead for very small size. So, this implementation consist * of two parts: one part is the "immediate" table that keeps rank result * as a raw table, and the other part is a normal succinct bit-vector. */ #define IMMEDIATE_TABLE_SIZE 54 /* a multiple of 9, and < 128 */ struct succ_index_table { uint64_t imm_part[IMMEDIATE_TABLE_SIZE / 9]; struct succ_dict_block { unsigned int rank; uint64_t small_block_ranks; /* 9 bits * 7 = 63 bits */ uint64_t bits[512/64]; } succ_part[FLEX_ARY_LEN]; }; #define imm_block_rank_set(v, i, r) (v) |= (uint64_t)(r) << (7 * (i)) #define imm_block_rank_get(v, i) (((int)((v) >> ((i) * 7))) & 0x7f) #define small_block_rank_set(v, i, r) (v) |= (uint64_t)(r) << (9 * ((i) - 1)) #define small_block_rank_get(v, i) ((i) == 0 ? 0 : (((int)((v) >> (((i) - 1) * 9))) & 0x1ff)) static struct succ_index_table * succ_index_table_create(int max_pos, int *data, int size) { const int imm_size = (max_pos < IMMEDIATE_TABLE_SIZE ? max_pos + 8 : IMMEDIATE_TABLE_SIZE) / 9; const int succ_size = (max_pos < IMMEDIATE_TABLE_SIZE ? 0 : (max_pos - IMMEDIATE_TABLE_SIZE + 511)) / 512; struct succ_index_table *sd = ruby_xcalloc(imm_size * sizeof(uint64_t) + succ_size * sizeof(struct succ_dict_block), 1); /* zero cleared */ int i, j, k, r; r = 0; for (j = 0; j < imm_size; j++) { for (i = 0; i < 9; i++) { if (r < size && data[r] == j * 9 + i) r++; imm_block_rank_set(sd->imm_part[j], i, r); } } for (k = 0; k < succ_size; k++) { struct succ_dict_block *sd_block = &sd->succ_part[k]; int small_rank = 0; sd_block->rank = r; for (j = 0; j < 8; j++) { uint64_t bits = 0; if (j) small_block_rank_set(sd_block->small_block_ranks, j, small_rank); for (i = 0; i < 64; i++) { if (r < size && data[r] == k * 512 + j * 64 + i + IMMEDIATE_TABLE_SIZE) { bits |= ((uint64_t)1) << i; r++; } } sd_block->bits[j] = bits; small_rank += rb_popcount64(bits); } } return sd; } static unsigned int * succ_index_table_invert(int max_pos, struct succ_index_table *sd, int size) { const int imm_size = (max_pos < IMMEDIATE_TABLE_SIZE ? max_pos + 8 : IMMEDIATE_TABLE_SIZE) / 9; const int succ_size = (max_pos < IMMEDIATE_TABLE_SIZE ? 0 : (max_pos - IMMEDIATE_TABLE_SIZE + 511)) / 512; unsigned int *positions = ruby_xmalloc(sizeof(unsigned int) * size), *p; int i, j, k, r = -1; p = positions; for (j = 0; j < imm_size; j++) { for (i = 0; i < 9; i++) { int nr = imm_block_rank_get(sd->imm_part[j], i); if (r != nr) *p++ = j * 9 + i; r = nr; } } for (k = 0; k < succ_size; k++) { for (j = 0; j < 8; j++) { for (i = 0; i < 64; i++) { if (sd->succ_part[k].bits[j] & (((uint64_t)1) << i)) { *p++ = k * 512 + j * 64 + i + IMMEDIATE_TABLE_SIZE; } } } } return positions; } static int succ_index_lookup(const struct succ_index_table *sd, int x) { if (x < IMMEDIATE_TABLE_SIZE) { const int i = x / 9; const int j = x % 9; return imm_block_rank_get(sd->imm_part[i], j); } else { const int block_index = (x - IMMEDIATE_TABLE_SIZE) / 512; const struct succ_dict_block *block = &sd->succ_part[block_index]; const int block_bit_index = (x - IMMEDIATE_TABLE_SIZE) % 512; const int small_block_index = block_bit_index / 64; const int small_block_popcount = small_block_rank_get(block->small_block_ranks, small_block_index); const int popcnt = rb_popcount64(block->bits[small_block_index] << (63 - block_bit_index % 64)); return block->rank + small_block_popcount + popcnt; } } #endif /* * Document-class: RubyVM::InstructionSequence * * The InstructionSequence class represents a compiled sequence of * instructions for the Ruby Virtual Machine. * * With it, you can get a handle to the instructions that make up a method or * a proc, compile strings of Ruby code down to VM instructions, and * disassemble instruction sequences to strings for easy inspection. It is * mostly useful if you want to learn how the Ruby VM works, but it also lets * you control various settings for the Ruby iseq compiler. * * You can find the source for the VM instructions in +insns.def+ in the Ruby * source. * * The instruction sequence results will almost certainly change as Ruby * changes, so example output in this documentation may be different from what * you see. */ void Init_ISeq(void) { /* declare ::RubyVM::InstructionSequence */ rb_cISeq = rb_define_class_under(rb_cRubyVM, "InstructionSequence", rb_cObject); rb_undef_alloc_func(rb_cISeq); rb_define_method(rb_cISeq, "inspect", iseqw_inspect, 0); rb_define_method(rb_cISeq, "disasm", iseqw_disasm, 0); rb_define_method(rb_cISeq, "disassemble", iseqw_disasm, 0); rb_define_method(rb_cISeq, "to_a", iseqw_to_a, 0); rb_define_method(rb_cISeq, "eval", iseqw_eval, 0); rb_define_method(rb_cISeq, "to_binary", iseqw_to_binary, -1); rb_define_singleton_method(rb_cISeq, "load_from_binary", iseqw_s_load_from_binary, 1); rb_define_singleton_method(rb_cISeq, "load_from_binary_extra_data", iseqw_s_load_from_binary_extra_data, 1); /* location APIs */ rb_define_method(rb_cISeq, "path", iseqw_path, 0); rb_define_method(rb_cISeq, "absolute_path", iseqw_absolute_path, 0); rb_define_method(rb_cISeq, "label", iseqw_label, 0); rb_define_method(rb_cISeq, "base_label", iseqw_base_label, 0); rb_define_method(rb_cISeq, "first_lineno", iseqw_first_lineno, 0); rb_define_method(rb_cISeq, "trace_points", iseqw_trace_points, 0); rb_define_method(rb_cISeq, "each_child", iseqw_each_child, 0); #if 0 /* TBD */ rb_define_private_method(rb_cISeq, "marshal_dump", iseqw_marshal_dump, 0); rb_define_private_method(rb_cISeq, "marshal_load", iseqw_marshal_load, 1); /* disable this feature because there is no verifier. */ rb_define_singleton_method(rb_cISeq, "load", iseq_s_load, -1); #endif (void)iseq_s_load; rb_define_singleton_method(rb_cISeq, "compile", iseqw_s_compile, -1); rb_define_singleton_method(rb_cISeq, "new", iseqw_s_compile, -1); rb_define_singleton_method(rb_cISeq, "compile_file", iseqw_s_compile_file, -1); rb_define_singleton_method(rb_cISeq, "compile_option", iseqw_s_compile_option_get, 0); rb_define_singleton_method(rb_cISeq, "compile_option=", iseqw_s_compile_option_set, 1); rb_define_singleton_method(rb_cISeq, "disasm", iseqw_s_disasm, 1); rb_define_singleton_method(rb_cISeq, "disassemble", iseqw_s_disasm, 1); rb_define_singleton_method(rb_cISeq, "of", iseqw_s_of, 1); rb_undef_method(CLASS_OF(rb_cISeq), "translate"); rb_undef_method(CLASS_OF(rb_cISeq), "load_iseq"); }