diff options
Diffstat (limited to 'ext/objspace')
| -rw-r--r-- | ext/objspace/depend | 99 | ||||
| -rw-r--r-- | ext/objspace/lib/objspace.rb | 94 | ||||
| -rw-r--r-- | ext/objspace/object_tracing.c | 56 | ||||
| -rw-r--r-- | ext/objspace/objspace.c | 403 | ||||
| -rw-r--r-- | ext/objspace/objspace_dump.c | 315 |
5 files changed, 578 insertions, 389 deletions
diff --git a/ext/objspace/depend b/ext/objspace/depend index c4da8031cc..d9dfc0c42b 100644 --- a/ext/objspace/depend +++ b/ext/objspace/depend @@ -2,10 +2,12 @@ object_tracing.o: $(RUBY_EXTCONF_H) object_tracing.o: $(arch_hdrdir)/ruby/config.h object_tracing.o: $(hdrdir)/ruby/assert.h +object_tracing.o: $(hdrdir)/ruby/atomic.h object_tracing.o: $(hdrdir)/ruby/backward.h object_tracing.o: $(hdrdir)/ruby/backward/2/assume.h object_tracing.o: $(hdrdir)/ruby/backward/2/attributes.h object_tracing.o: $(hdrdir)/ruby/backward/2/bool.h +object_tracing.o: $(hdrdir)/ruby/backward/2/gcc_version_since.h object_tracing.o: $(hdrdir)/ruby/backward/2/inttypes.h object_tracing.o: $(hdrdir)/ruby/backward/2/limits.h object_tracing.o: $(hdrdir)/ruby/backward/2/long_long.h @@ -13,6 +15,7 @@ object_tracing.o: $(hdrdir)/ruby/backward/2/stdalign.h object_tracing.o: $(hdrdir)/ruby/backward/2/stdarg.h object_tracing.o: $(hdrdir)/ruby/debug.h object_tracing.o: $(hdrdir)/ruby/defines.h +object_tracing.o: $(hdrdir)/ruby/encoding.h object_tracing.o: $(hdrdir)/ruby/intern.h object_tracing.o: $(hdrdir)/ruby/internal/abi.h object_tracing.o: $(hdrdir)/ruby/internal/anyargs.h @@ -52,6 +55,7 @@ object_tracing.o: $(hdrdir)/ruby/internal/attr/noexcept.h object_tracing.o: $(hdrdir)/ruby/internal/attr/noinline.h object_tracing.o: $(hdrdir)/ruby/internal/attr/nonnull.h object_tracing.o: $(hdrdir)/ruby/internal/attr/noreturn.h +object_tracing.o: $(hdrdir)/ruby/internal/attr/packed_struct.h object_tracing.o: $(hdrdir)/ruby/internal/attr/pure.h object_tracing.o: $(hdrdir)/ruby/internal/attr/restrict.h object_tracing.o: $(hdrdir)/ruby/internal/attr/returns_nonnull.h @@ -84,6 +88,15 @@ object_tracing.o: $(hdrdir)/ruby/internal/core/rtypeddata.h object_tracing.o: $(hdrdir)/ruby/internal/ctype.h object_tracing.o: $(hdrdir)/ruby/internal/dllexport.h object_tracing.o: $(hdrdir)/ruby/internal/dosish.h +object_tracing.o: $(hdrdir)/ruby/internal/encoding/coderange.h +object_tracing.o: $(hdrdir)/ruby/internal/encoding/ctype.h +object_tracing.o: $(hdrdir)/ruby/internal/encoding/encoding.h +object_tracing.o: $(hdrdir)/ruby/internal/encoding/pathname.h +object_tracing.o: $(hdrdir)/ruby/internal/encoding/re.h +object_tracing.o: $(hdrdir)/ruby/internal/encoding/sprintf.h +object_tracing.o: $(hdrdir)/ruby/internal/encoding/string.h +object_tracing.o: $(hdrdir)/ruby/internal/encoding/symbol.h +object_tracing.o: $(hdrdir)/ruby/internal/encoding/transcode.h object_tracing.o: $(hdrdir)/ruby/internal/error.h object_tracing.o: $(hdrdir)/ruby/internal/eval.h object_tracing.o: $(hdrdir)/ruby/internal/event.h @@ -111,7 +124,6 @@ object_tracing.o: $(hdrdir)/ruby/internal/intern/enumerator.h object_tracing.o: $(hdrdir)/ruby/internal/intern/error.h object_tracing.o: $(hdrdir)/ruby/internal/intern/eval.h object_tracing.o: $(hdrdir)/ruby/internal/intern/file.h -object_tracing.o: $(hdrdir)/ruby/internal/intern/gc.h object_tracing.o: $(hdrdir)/ruby/internal/intern/hash.h object_tracing.o: $(hdrdir)/ruby/internal/intern/io.h object_tracing.o: $(hdrdir)/ruby/internal/intern/load.h @@ -128,6 +140,7 @@ object_tracing.o: $(hdrdir)/ruby/internal/intern/re.h object_tracing.o: $(hdrdir)/ruby/internal/intern/ruby.h object_tracing.o: $(hdrdir)/ruby/internal/intern/select.h object_tracing.o: $(hdrdir)/ruby/internal/intern/select/largesize.h +object_tracing.o: $(hdrdir)/ruby/internal/intern/set.h object_tracing.o: $(hdrdir)/ruby/internal/intern/signal.h object_tracing.o: $(hdrdir)/ruby/internal/intern/sprintf.h object_tracing.o: $(hdrdir)/ruby/internal/intern/string.h @@ -142,12 +155,12 @@ object_tracing.o: $(hdrdir)/ruby/internal/memory.h object_tracing.o: $(hdrdir)/ruby/internal/method.h object_tracing.o: $(hdrdir)/ruby/internal/module.h object_tracing.o: $(hdrdir)/ruby/internal/newobj.h -object_tracing.o: $(hdrdir)/ruby/internal/rgengc.h object_tracing.o: $(hdrdir)/ruby/internal/scan_args.h object_tracing.o: $(hdrdir)/ruby/internal/special_consts.h object_tracing.o: $(hdrdir)/ruby/internal/static_assert.h object_tracing.o: $(hdrdir)/ruby/internal/stdalign.h object_tracing.o: $(hdrdir)/ruby/internal/stdbool.h +object_tracing.o: $(hdrdir)/ruby/internal/stdckdint.h object_tracing.o: $(hdrdir)/ruby/internal/symbol.h object_tracing.o: $(hdrdir)/ruby/internal/value.h object_tracing.o: $(hdrdir)/ruby/internal/value_type.h @@ -155,15 +168,45 @@ object_tracing.o: $(hdrdir)/ruby/internal/variable.h object_tracing.o: $(hdrdir)/ruby/internal/warning_push.h object_tracing.o: $(hdrdir)/ruby/internal/xmalloc.h object_tracing.o: $(hdrdir)/ruby/missing.h +object_tracing.o: $(hdrdir)/ruby/onigmo.h +object_tracing.o: $(hdrdir)/ruby/oniguruma.h object_tracing.o: $(hdrdir)/ruby/ruby.h object_tracing.o: $(hdrdir)/ruby/st.h object_tracing.o: $(hdrdir)/ruby/subst.h +object_tracing.o: $(hdrdir)/ruby/thread_native.h +object_tracing.o: $(top_srcdir)/ccan/check_type/check_type.h +object_tracing.o: $(top_srcdir)/ccan/container_of/container_of.h +object_tracing.o: $(top_srcdir)/ccan/list/list.h +object_tracing.o: $(top_srcdir)/ccan/str/str.h +object_tracing.o: $(top_srcdir)/id_table.h object_tracing.o: $(top_srcdir)/internal.h +object_tracing.o: $(top_srcdir)/internal/array.h +object_tracing.o: $(top_srcdir)/internal/basic_operators.h +object_tracing.o: $(top_srcdir)/internal/box.h +object_tracing.o: $(top_srcdir)/internal/compilers.h +object_tracing.o: $(top_srcdir)/internal/gc.h +object_tracing.o: $(top_srcdir)/internal/imemo.h +object_tracing.o: $(top_srcdir)/internal/sanitizers.h +object_tracing.o: $(top_srcdir)/internal/serial.h +object_tracing.o: $(top_srcdir)/internal/set_table.h +object_tracing.o: $(top_srcdir)/internal/static_assert.h +object_tracing.o: $(top_srcdir)/internal/vm.h +object_tracing.o: $(top_srcdir)/internal/warnings.h +object_tracing.o: $(top_srcdir)/method.h +object_tracing.o: $(top_srcdir)/node.h +object_tracing.o: $(top_srcdir)/ruby_assert.h +object_tracing.o: $(top_srcdir)/ruby_atomic.h +object_tracing.o: $(top_srcdir)/rubyparser.h +object_tracing.o: $(top_srcdir)/thread_pthread.h +object_tracing.o: $(top_srcdir)/vm_core.h +object_tracing.o: $(top_srcdir)/vm_opts.h object_tracing.o: object_tracing.c object_tracing.o: objspace.h +object_tracing.o: {$(VPATH)}id.h objspace.o: $(RUBY_EXTCONF_H) objspace.o: $(arch_hdrdir)/ruby/config.h objspace.o: $(hdrdir)/ruby/assert.h +objspace.o: $(hdrdir)/ruby/atomic.h objspace.o: $(hdrdir)/ruby/backward.h objspace.o: $(hdrdir)/ruby/backward/2/assume.h objspace.o: $(hdrdir)/ruby/backward/2/attributes.h @@ -215,6 +258,7 @@ objspace.o: $(hdrdir)/ruby/internal/attr/noexcept.h objspace.o: $(hdrdir)/ruby/internal/attr/noinline.h objspace.o: $(hdrdir)/ruby/internal/attr/nonnull.h objspace.o: $(hdrdir)/ruby/internal/attr/noreturn.h +objspace.o: $(hdrdir)/ruby/internal/attr/packed_struct.h objspace.o: $(hdrdir)/ruby/internal/attr/pure.h objspace.o: $(hdrdir)/ruby/internal/attr/restrict.h objspace.o: $(hdrdir)/ruby/internal/attr/returns_nonnull.h @@ -284,7 +328,6 @@ objspace.o: $(hdrdir)/ruby/internal/intern/enumerator.h objspace.o: $(hdrdir)/ruby/internal/intern/error.h objspace.o: $(hdrdir)/ruby/internal/intern/eval.h objspace.o: $(hdrdir)/ruby/internal/intern/file.h -objspace.o: $(hdrdir)/ruby/internal/intern/gc.h objspace.o: $(hdrdir)/ruby/internal/intern/hash.h objspace.o: $(hdrdir)/ruby/internal/intern/io.h objspace.o: $(hdrdir)/ruby/internal/intern/load.h @@ -301,6 +344,7 @@ objspace.o: $(hdrdir)/ruby/internal/intern/re.h objspace.o: $(hdrdir)/ruby/internal/intern/ruby.h objspace.o: $(hdrdir)/ruby/internal/intern/select.h objspace.o: $(hdrdir)/ruby/internal/intern/select/largesize.h +objspace.o: $(hdrdir)/ruby/internal/intern/set.h objspace.o: $(hdrdir)/ruby/internal/intern/signal.h objspace.o: $(hdrdir)/ruby/internal/intern/sprintf.h objspace.o: $(hdrdir)/ruby/internal/intern/string.h @@ -315,12 +359,12 @@ objspace.o: $(hdrdir)/ruby/internal/memory.h objspace.o: $(hdrdir)/ruby/internal/method.h objspace.o: $(hdrdir)/ruby/internal/module.h objspace.o: $(hdrdir)/ruby/internal/newobj.h -objspace.o: $(hdrdir)/ruby/internal/rgengc.h objspace.o: $(hdrdir)/ruby/internal/scan_args.h objspace.o: $(hdrdir)/ruby/internal/special_consts.h objspace.o: $(hdrdir)/ruby/internal/static_assert.h objspace.o: $(hdrdir)/ruby/internal/stdalign.h objspace.o: $(hdrdir)/ruby/internal/stdbool.h +objspace.o: $(hdrdir)/ruby/internal/stdckdint.h objspace.o: $(hdrdir)/ruby/internal/symbol.h objspace.o: $(hdrdir)/ruby/internal/value.h objspace.o: $(hdrdir)/ruby/internal/value_type.h @@ -336,10 +380,18 @@ objspace.o: $(hdrdir)/ruby/regex.h objspace.o: $(hdrdir)/ruby/ruby.h objspace.o: $(hdrdir)/ruby/st.h objspace.o: $(hdrdir)/ruby/subst.h -objspace.o: $(top_srcdir)/gc.h +objspace.o: $(hdrdir)/ruby/thread_native.h +objspace.o: $(top_srcdir)/ccan/check_type/check_type.h +objspace.o: $(top_srcdir)/ccan/container_of/container_of.h +objspace.o: $(top_srcdir)/ccan/list/list.h +objspace.o: $(top_srcdir)/ccan/str/str.h +objspace.o: $(top_srcdir)/constant.h +objspace.o: $(top_srcdir)/debug_counter.h objspace.o: $(top_srcdir)/id_table.h objspace.o: $(top_srcdir)/internal.h objspace.o: $(top_srcdir)/internal/array.h +objspace.o: $(top_srcdir)/internal/basic_operators.h +objspace.o: $(top_srcdir)/internal/box.h objspace.o: $(top_srcdir)/internal/class.h objspace.o: $(top_srcdir)/internal/compilers.h objspace.o: $(top_srcdir)/internal/gc.h @@ -347,10 +399,24 @@ objspace.o: $(top_srcdir)/internal/hash.h objspace.o: $(top_srcdir)/internal/imemo.h objspace.o: $(top_srcdir)/internal/sanitizers.h objspace.o: $(top_srcdir)/internal/serial.h +objspace.o: $(top_srcdir)/internal/set_table.h objspace.o: $(top_srcdir)/internal/static_assert.h +objspace.o: $(top_srcdir)/internal/struct.h +objspace.o: $(top_srcdir)/internal/variable.h +objspace.o: $(top_srcdir)/internal/vm.h objspace.o: $(top_srcdir)/internal/warnings.h +objspace.o: $(top_srcdir)/method.h objspace.o: $(top_srcdir)/node.h +objspace.o: $(top_srcdir)/ruby_assert.h +objspace.o: $(top_srcdir)/ruby_atomic.h +objspace.o: $(top_srcdir)/rubyparser.h +objspace.o: $(top_srcdir)/shape.h objspace.o: $(top_srcdir)/symbol.h +objspace.o: $(top_srcdir)/thread_pthread.h +objspace.o: $(top_srcdir)/vm_core.h +objspace.o: $(top_srcdir)/vm_debug.h +objspace.o: $(top_srcdir)/vm_opts.h +objspace.o: $(top_srcdir)/vm_sync.h objspace.o: objspace.c objspace.o: {$(VPATH)}id.h objspace_dump.o: $(RUBY_EXTCONF_H) @@ -409,6 +475,7 @@ objspace_dump.o: $(hdrdir)/ruby/internal/attr/noexcept.h objspace_dump.o: $(hdrdir)/ruby/internal/attr/noinline.h objspace_dump.o: $(hdrdir)/ruby/internal/attr/nonnull.h objspace_dump.o: $(hdrdir)/ruby/internal/attr/noreturn.h +objspace_dump.o: $(hdrdir)/ruby/internal/attr/packed_struct.h objspace_dump.o: $(hdrdir)/ruby/internal/attr/pure.h objspace_dump.o: $(hdrdir)/ruby/internal/attr/restrict.h objspace_dump.o: $(hdrdir)/ruby/internal/attr/returns_nonnull.h @@ -477,7 +544,6 @@ objspace_dump.o: $(hdrdir)/ruby/internal/intern/enumerator.h objspace_dump.o: $(hdrdir)/ruby/internal/intern/error.h objspace_dump.o: $(hdrdir)/ruby/internal/intern/eval.h objspace_dump.o: $(hdrdir)/ruby/internal/intern/file.h -objspace_dump.o: $(hdrdir)/ruby/internal/intern/gc.h objspace_dump.o: $(hdrdir)/ruby/internal/intern/hash.h objspace_dump.o: $(hdrdir)/ruby/internal/intern/io.h objspace_dump.o: $(hdrdir)/ruby/internal/intern/load.h @@ -494,6 +560,7 @@ objspace_dump.o: $(hdrdir)/ruby/internal/intern/re.h objspace_dump.o: $(hdrdir)/ruby/internal/intern/ruby.h objspace_dump.o: $(hdrdir)/ruby/internal/intern/select.h objspace_dump.o: $(hdrdir)/ruby/internal/intern/select/largesize.h +objspace_dump.o: $(hdrdir)/ruby/internal/intern/set.h objspace_dump.o: $(hdrdir)/ruby/internal/intern/signal.h objspace_dump.o: $(hdrdir)/ruby/internal/intern/sprintf.h objspace_dump.o: $(hdrdir)/ruby/internal/intern/string.h @@ -508,12 +575,12 @@ objspace_dump.o: $(hdrdir)/ruby/internal/memory.h objspace_dump.o: $(hdrdir)/ruby/internal/method.h objspace_dump.o: $(hdrdir)/ruby/internal/module.h objspace_dump.o: $(hdrdir)/ruby/internal/newobj.h -objspace_dump.o: $(hdrdir)/ruby/internal/rgengc.h objspace_dump.o: $(hdrdir)/ruby/internal/scan_args.h objspace_dump.o: $(hdrdir)/ruby/internal/special_consts.h objspace_dump.o: $(hdrdir)/ruby/internal/static_assert.h objspace_dump.o: $(hdrdir)/ruby/internal/stdalign.h objspace_dump.o: $(hdrdir)/ruby/internal/stdbool.h +objspace_dump.o: $(hdrdir)/ruby/internal/stdckdint.h objspace_dump.o: $(hdrdir)/ruby/internal/symbol.h objspace_dump.o: $(hdrdir)/ruby/internal/value.h objspace_dump.o: $(hdrdir)/ruby/internal/value_type.h @@ -533,26 +600,42 @@ objspace_dump.o: $(top_srcdir)/ccan/check_type/check_type.h objspace_dump.o: $(top_srcdir)/ccan/container_of/container_of.h objspace_dump.o: $(top_srcdir)/ccan/list/list.h objspace_dump.o: $(top_srcdir)/ccan/str/str.h -objspace_dump.o: $(top_srcdir)/gc.h +objspace_dump.o: $(top_srcdir)/constant.h +objspace_dump.o: $(top_srcdir)/debug_counter.h +objspace_dump.o: $(top_srcdir)/encindex.h +objspace_dump.o: $(top_srcdir)/id_table.h objspace_dump.o: $(top_srcdir)/internal.h objspace_dump.o: $(top_srcdir)/internal/array.h +objspace_dump.o: $(top_srcdir)/internal/basic_operators.h +objspace_dump.o: $(top_srcdir)/internal/box.h +objspace_dump.o: $(top_srcdir)/internal/class.h objspace_dump.o: $(top_srcdir)/internal/compilers.h objspace_dump.o: $(top_srcdir)/internal/gc.h objspace_dump.o: $(top_srcdir)/internal/hash.h objspace_dump.o: $(top_srcdir)/internal/imemo.h +objspace_dump.o: $(top_srcdir)/internal/io.h objspace_dump.o: $(top_srcdir)/internal/sanitizers.h objspace_dump.o: $(top_srcdir)/internal/serial.h +objspace_dump.o: $(top_srcdir)/internal/set_table.h objspace_dump.o: $(top_srcdir)/internal/static_assert.h objspace_dump.o: $(top_srcdir)/internal/string.h +objspace_dump.o: $(top_srcdir)/internal/struct.h +objspace_dump.o: $(top_srcdir)/internal/variable.h objspace_dump.o: $(top_srcdir)/internal/vm.h objspace_dump.o: $(top_srcdir)/internal/warnings.h objspace_dump.o: $(top_srcdir)/method.h objspace_dump.o: $(top_srcdir)/node.h objspace_dump.o: $(top_srcdir)/ruby_assert.h objspace_dump.o: $(top_srcdir)/ruby_atomic.h +objspace_dump.o: $(top_srcdir)/rubyparser.h +objspace_dump.o: $(top_srcdir)/shape.h +objspace_dump.o: $(top_srcdir)/symbol.h objspace_dump.o: $(top_srcdir)/thread_pthread.h +objspace_dump.o: $(top_srcdir)/vm_callinfo.h objspace_dump.o: $(top_srcdir)/vm_core.h +objspace_dump.o: $(top_srcdir)/vm_debug.h objspace_dump.o: $(top_srcdir)/vm_opts.h +objspace_dump.o: $(top_srcdir)/vm_sync.h objspace_dump.o: objspace.h objspace_dump.o: objspace_dump.c objspace_dump.o: {$(VPATH)}id.h diff --git a/ext/objspace/lib/objspace.rb b/ext/objspace/lib/objspace.rb index 0298b0646c..47873f5112 100644 --- a/ext/objspace/lib/objspace.rb +++ b/ext/objspace/lib/objspace.rb @@ -6,17 +6,21 @@ module ObjectSpace class << self private :_dump private :_dump_all + private :_dump_shapes end module_function - # call-seq: - # ObjectSpace.dump(obj[, output: :string]) # => "{ ... }" - # ObjectSpace.dump(obj, output: :file) # => #<File:/tmp/rubyobj20131125-88733-1xkfmpv.json> - # ObjectSpace.dump(obj, output: :stdout) # => nil - # # Dump the contents of a ruby object as JSON. # + # _output_ can be one of: +:stdout+, +:file+, +:string+, or IO object. + # + # * +:file+ means dumping to a tempfile and returning corresponding File object; + # * +:stdout+ means printing the dump and returning +nil+; + # * +:string+ means returning a string with the dump; + # * if an instance of IO object is provided, the output goes there, and the object + # is returned. + # # This method is only expected to work with C Ruby. # This is an experimental method and is subject to change. # In particular, the function signature and output format are @@ -42,38 +46,78 @@ module ObjectSpace end - # call-seq: - # ObjectSpace.dump_all([output: :file]) # => #<File:/tmp/rubyheap20131125-88469-laoj3v.json> - # ObjectSpace.dump_all(output: :stdout) # => nil - # ObjectSpace.dump_all(output: :string) # => "{...}\n{...}\n..." - # ObjectSpace.dump_all(output: - # File.open('heap.json','w')) # => #<File:heap.json> - # ObjectSpace.dump_all(output: :string, - # since: 42) # => "{...}\n{...}\n..." + # Dump the contents of the ruby heap as JSON. + # + # _output_ argument is the same as for #dump. + # + # _full_ must be a boolean. If true, all heap slots are dumped including the empty ones (+T_NONE+). # - # Dump the contents of the ruby heap as JSON. + # _since_ must be a non-negative integer or +nil+. # - # _since_ must be a non-negative integer or +nil+. + # If _since_ is a positive integer, only objects of that generation and + # newer generations are dumped. The current generation can be accessed using + # GC::count. Objects that were allocated without object allocation tracing enabled + # are ignored. See ::trace_object_allocations for more information and + # examples. # - # If _since_ is a positive integer, only objects of that generation and - # newer generations are dumped. The current generation can be accessed using - # GC::count. + # If _since_ is omitted or is +nil+, all objects are dumped. # - # Objects that were allocated without object allocation tracing enabled - # are ignored. See ::trace_object_allocations for more information and - # examples. + # _shapes_ must be a boolean or a non-negative integer. # - # If _since_ is omitted or is +nil+, all objects are dumped. + # If _shapes_ is a positive integer, only shapes newer than the provided + # shape id are dumped. The current shape_id can be accessed using <tt>RubyVM.stat(:next_shape_id)</tt>. + # + # If _shapes_ is +false+, no shapes are dumped. + # + # To only dump objects allocated past a certain point you can combine _since_ and _shapes_: + # ObjectSpace.trace_object_allocations + # GC.start + # gc_generation = GC.count + # shape_generation = RubyVM.stat(:next_shape_id) + # call_method_to_instrument + # ObjectSpace.dump_all(since: gc_generation, shapes: shape_generation) + # + # This method is only expected to work with C Ruby. + # This is an experimental method and is subject to change. + # In particular, the function signature and output format are + # not guaranteed to be compatible in future versions of ruby. + def dump_all(output: :file, full: false, since: nil, shapes: true) + out = case output + when :file, nil + require 'tempfile' + Tempfile.create(%w(rubyheap .json)) + when :stdout + STDOUT + when :string + +'' + when IO + output + else + raise ArgumentError, "wrong output option: #{output.inspect}" + end + + shapes = 0 if shapes == true + ret = _dump_all(out, full, since, shapes) + return nil if output == :stdout + ret + end + + # Dump the contents of the ruby shape tree as JSON. + # + # _output_ argument is the same as for #dump. + # + # If _since_ is a positive integer, only shapes newer than the provided + # shape id are dumped. The current shape_id can be accessed using <tt>RubyVM.stat(:next_shape_id)</tt>. # # This method is only expected to work with C Ruby. # This is an experimental method and is subject to change. # In particular, the function signature and output format are # not guaranteed to be compatible in future versions of ruby. - def dump_all(output: :file, full: false, since: nil) + def dump_shapes(output: :file, since: 0) out = case output when :file, nil require 'tempfile' - Tempfile.create(%w(rubyheap .json)) + Tempfile.create(%w(rubyshapes .json)) when :stdout STDOUT when :string @@ -84,7 +128,7 @@ module ObjectSpace raise ArgumentError, "wrong output option: #{output.inspect}" end - ret = _dump_all(out, full, since) + ret = _dump_shapes(out, since) return nil if output == :stdout ret end diff --git a/ext/objspace/object_tracing.c b/ext/objspace/object_tracing.c index 0bf866a8f1..1c18bf02ee 100644 --- a/ext/objspace/object_tracing.c +++ b/ext/objspace/object_tracing.c @@ -14,6 +14,7 @@ **********************************************************************/ #include "internal.h" +#include "internal/gc.h" #include "ruby/debug.h" #include "objspace.h" @@ -52,6 +53,14 @@ make_unique_str(st_table *tbl, const char *str, long len) } } +static int +delete_unique_str_dec(st_data_t *key, st_data_t *value, st_data_t arg, int existing) +{ + assert(existing); + *value = arg; + return ST_CONTINUE; +} + static void delete_unique_str(st_table *tbl, const char *str) { @@ -65,7 +74,7 @@ delete_unique_str(st_table *tbl, const char *str) ruby_xfree((char *)n); } else { - st_insert(tbl, (st_data_t)str, n-1); + st_update(tbl, (st_data_t)str, delete_unique_str_dec, (st_data_t)(n-1)); } } } @@ -82,7 +91,7 @@ newobj_i(VALUE tpval, void *data) VALUE klass = rb_tracearg_defined_class(tparg); struct allocation_info *info; const char *path_cstr = RTEST(path) ? make_unique_str(arg->str_table, RSTRING_PTR(path), RSTRING_LEN(path)) : 0; - VALUE class_path = (RTEST(klass) && !OBJ_FROZEN(klass)) ? rb_class_path_cached(klass) : Qnil; + VALUE class_path = (RTEST(klass) && !OBJ_FROZEN(klass)) ? rb_mod_name(klass) : Qnil; const char *class_path_cstr = RTEST(class_path) ? make_unique_str(arg->str_table, RSTRING_PTR(class_path), RSTRING_LEN(class_path)) : 0; st_data_t v; @@ -121,6 +130,10 @@ freeobj_i(VALUE tpval, void *data) st_data_t v; struct allocation_info *info; + /* Modifying the st table can cause allocations, which can trigger GC. + * Since freeobj_i is called during GC, it must not trigger another GC. */ + VALUE gc_disabled = rb_gc_disable_no_rest(); + if (arg->keep_remains) { if (st_lookup(arg->object_table, obj, &v)) { info = (struct allocation_info *)v; @@ -135,6 +148,8 @@ freeobj_i(VALUE tpval, void *data) ruby_xfree(info); } } + + if (gc_disabled == Qfalse) rb_gc_enable(); } static int @@ -183,22 +198,25 @@ allocation_info_tracer_memsize(const void *ptr) } static int -hash_foreach_should_replace_key(st_data_t key, st_data_t value, st_data_t argp, int error) +allocation_info_tracer_compact_update_object_table_i(st_data_t key, st_data_t value, st_data_t data) { - VALUE allocated_object; + st_table *table = (st_table *)data; - allocated_object = (VALUE)value; - if (allocated_object != rb_gc_location(allocated_object)) { - return ST_REPLACE; + if (!rb_gc_pointer_to_heap_p(key)) { + struct allocation_info *info = (struct allocation_info *)value; + xfree(info); + return ST_DELETE; } - return ST_CONTINUE; -} + if (key != rb_gc_location(key)) { + DURING_GC_COULD_MALLOC_REGION_START(); + { + st_insert(table, rb_gc_location(key), value); + } + DURING_GC_COULD_MALLOC_REGION_END(); -static int -hash_replace_key(st_data_t *key, st_data_t *value, st_data_t argp, int existing) -{ - *key = rb_gc_location((VALUE)*key); + return ST_DELETE; + } return ST_CONTINUE; } @@ -209,7 +227,10 @@ allocation_info_tracer_compact(void *ptr) struct traceobj_arg *trace_arg = (struct traceobj_arg *)ptr; if (trace_arg->object_table && - st_foreach_with_replace(trace_arg->object_table, hash_foreach_should_replace_key, hash_replace_key, 0)) { + st_foreach( + trace_arg->object_table, + allocation_info_tracer_compact_update_object_table_i, + (st_data_t)trace_arg->object_table)) { rb_raise(rb_eRuntimeError, "hash modified during iteration"); } } @@ -392,6 +413,13 @@ object_allocations_reporter(FILE *out, void *ptr) fprintf(out, "== object_allocations_reporter: END\n"); } +/* + * call-seq: trace_object_allocations_debug_start + * + * Starts tracing object allocations for GC debugging. + * If you encounter the BUG "... is T_NONE" (and so on) on your + * application, please try this method at the beginning of your app. + */ static VALUE trace_object_allocations_debug_start(VALUE self) { diff --git a/ext/objspace/objspace.c b/ext/objspace/objspace.c index 0b1b094325..38bffb07f7 100644 --- a/ext/objspace/objspace.c +++ b/ext/objspace/objspace.c @@ -12,14 +12,13 @@ **********************************************************************/ -#include "gc.h" #include "internal.h" #include "internal/class.h" #include "internal/compilers.h" +#include "internal/gc.h" #include "internal/hash.h" #include "internal/imemo.h" #include "internal/sanitizers.h" -#include "node.h" #include "ruby/io.h" #include "ruby/re.h" #include "ruby/st.h" @@ -39,10 +38,11 @@ * information as only a *HINT*. Especially, the size of +T_DATA+ may not be * correct. * - * This method is only expected to work with C Ruby. + * This method is only expected to work with CRuby. * - * From Ruby 2.2, memsize_of(obj) returns a memory size includes - * sizeof(RVALUE). + * From Ruby 3.2 with Variable Width Allocation, it returns the actual slot + * size used plus any additional memory allocated outside the slot (such + * as external strings, arrays, or hash tables). */ static VALUE @@ -82,15 +82,15 @@ heap_iter(void *vstart, void *vend, size_t stride, void *ptr) VALUE v; for (v = (VALUE)vstart; v != (VALUE)vend; v += stride) { - void *poisoned = asan_poisoned_object_p(v); - asan_unpoison_object(v, false); + void *poisoned = rb_asan_poisoned_object_p(v); + rb_asan_unpoison_object(v, false); if (RBASIC(v)->flags) { (*ctx->cb)(v, ctx->data); } if (poisoned) { - asan_poison_object(v); + rb_asan_poison_object(v); } } @@ -108,28 +108,24 @@ each_object_with_flags(each_obj_with_flags cb, void *ctx) /* * call-seq: - * ObjectSpace.memsize_of_all([klass]) -> Integer - * - * Return consuming memory size of all living objects in bytes. + * ObjectSpace.memsize_of_all(klass = nil) -> integer * - * If +klass+ (should be Class object) is given, return the total memory size - * of instances of the given class. + * Returns the total memory size of all living objects in bytes. * - * Note that the returned size is incomplete. You need to deal with this - * information as only a *HINT*. Especially, the size of +T_DATA+ may not be - * correct. + * ObjectSpace.memsize_of_all # => 12502001 * - * Note that this method does *NOT* return total malloc'ed memory size. + * If +klass+ is given (which must be a Class or Module), returns the total + * memory size of objects whose class is, or is a subclass, of +klass+. * - * This method can be defined by the following Ruby code: + * class MyClass; end + * ObjectSpace.memsize_of_all(MyClass) # => 0 + * o = MyClass.new + * ObjectSpace.memsize_of_all(MyClass) # => 40 * - * def memsize_of_all klass = false - * total = 0 - * ObjectSpace.each_object{|e| - * total += ObjectSpace.memsize_of(e) if klass == false || e.kind_of?(klass) - * } - * total - * end + * Note that the value returned may be an underestimate of the actual amount + * of memory used. Therefore, the value returned should only be used as a hint, + * rather than a source of truth. In particular, the size of +T_DATA+ objects may + * not be correct. * * This method is only expected to work with C Ruby. */ @@ -141,6 +137,7 @@ memsize_of_all_m(int argc, VALUE *argv, VALUE self) if (argc > 0) { rb_scan_args(argc, argv, "01", &data.klass); + if (!NIL_P(data.klass)) rb_obj_is_kind_of(Qnil, data.klass); } each_object_with_flags(total_i, &data); @@ -170,8 +167,7 @@ setup_hash(int argc, VALUE *argv) hash = rb_hash_new(); } else if (!RHASH_EMPTY_P(hash)) { - /* WB: no new reference */ - st_foreach(RHASH_TBL_RAW(hash), set_zero_i, hash); + rb_hash_foreach(hash, set_zero_i, (st_data_t)hash); } return hash; @@ -225,23 +221,26 @@ type2sym(enum ruby_value_type i) /* * call-seq: - * ObjectSpace.count_objects_size([result_hash]) -> hash + * ObjectSpace.count_objects_size(result_hash = {}) -> result_hash * * Counts objects size (in bytes) for each type. * - * Note that this information is incomplete. You need to deal with - * this information as only a *HINT*. Especially, total size of - * T_DATA may be wrong. + * Note that the returned size may not be accurate, so it should only + * be used as a hint. Specifically, the size for +T_DATA+ may be + * inaccurate because these are custom objects defined in Ruby and + * native extensions and so they may not accurately report their + * memory size. * - * It returns a hash as: - * {:TOTAL=>1461154, :T_CLASS=>158280, :T_MODULE=>20672, :T_STRING=>527249, ...} + * It returns a hash that looks like: * - * If the optional argument, result_hash, is given, - * it is overwritten and returned. - * This is intended to avoid probe effect. + * {TOTAL: 1461154, T_CLASS: 158280, T_MODULE: 20672, T_STRING: 527249, ...} + * + * The contents of the returned hash are implementation specific and + * may be changed in future versions without notice. * - * The contents of the returned hash is implementation defined. - * It may be changed in future. + * If the optional argument, +result_hash+, is given, + * it is overwritten and returned. + * This is intended to avoid the probe effect. * * This method is only expected to work with C Ruby. */ @@ -296,28 +295,27 @@ size_t rb_sym_immortal_count(void); /* * call-seq: - * ObjectSpace.count_symbols([result_hash]) -> hash + * ObjectSpace.count_symbols(result_hash = nil) -> hash * - * Counts symbols for each Symbol type. + * Returns a hash containing the number of objects for each Symbol type. * - * This method is only for MRI developers interested in performance and memory - * usage of Ruby programs. + * The types of Symbols are the following: * - * If the optional argument, result_hash, is given, it is overwritten and - * returned. This is intended to avoid probe effect. + * - +mortal_dynamic_symbol+: Symbols that are garbage collectable. + * - +immortal_dynamic_symbol+: Symbols that are objects allocated from the + * garbage collector, but are not garbage collectable. + * - +immortal_static_symbol+: Symbols that are not allocated from the + * garbage collector, and are thus not garbage collectable. + * - +immortal_symbol+: the sum of +immortal_dynamic_symbol+ and +immortal_static_symbol+. * - * Note: - * The contents of the returned hash is implementation defined. - * It may be changed in future. + * If the optional argument +result_hash+ is given, it is overwritten and + * returned. This is intended to avoid the probe effect. * - * This method is only expected to work with C Ruby. - * - * On this version of MRI, they have 3 types of Symbols (and 1 total counts). + * This method is intended for developers interested in performance and memory + * usage of Ruby programs. The contents of the returned hash is implementation + * specific and may change in the future. * - * * mortal_dynamic_symbol: GC target symbols (collected by GC) - * * immortal_dynamic_symbol: Immortal symbols promoted from dynamic symbols (do not collected by GC) - * * immortal_static_symbol: Immortal symbols (do not collected by GC) - * * immortal_symbol: total immortal symbols (immortal_dynamic_symbol+immortal_static_symbol) + * This method is only expected to work with C Ruby. */ static VALUE @@ -338,173 +336,6 @@ count_symbols(int argc, VALUE *argv, VALUE os) } static void -cn_i(VALUE v, void *n) -{ - size_t *nodes = (size_t *)n; - - if (BUILTIN_TYPE(v) == T_NODE) { - size_t s = nd_type((NODE *)v); - nodes[s]++; - } -} - -/* - * call-seq: - * ObjectSpace.count_nodes([result_hash]) -> hash - * - * Counts nodes for each node type. - * - * This method is only for MRI developers interested in performance and memory - * usage of Ruby programs. - * - * It returns a hash as: - * - * {:NODE_METHOD=>2027, :NODE_FBODY=>1927, :NODE_CFUNC=>1798, ...} - * - * If the optional argument, result_hash, is given, it is overwritten and - * returned. This is intended to avoid probe effect. - * - * Note: - * The contents of the returned hash is implementation defined. - * It may be changed in future. - * - * This method is only expected to work with C Ruby. - */ - -static VALUE -count_nodes(int argc, VALUE *argv, VALUE os) -{ - size_t nodes[NODE_LAST+1]; - enum node_type i; - VALUE hash = setup_hash(argc, argv); - - for (i = 0; i <= NODE_LAST; i++) { - nodes[i] = 0; - } - - each_object_with_flags(cn_i, &nodes[0]); - - for (i=0; i<NODE_LAST; i++) { - if (nodes[i] != 0) { - VALUE node; - switch (i) { -#define COUNT_NODE(n) case n: node = ID2SYM(rb_intern(#n)); goto set - COUNT_NODE(NODE_SCOPE); - COUNT_NODE(NODE_BLOCK); - COUNT_NODE(NODE_IF); - COUNT_NODE(NODE_UNLESS); - COUNT_NODE(NODE_CASE); - COUNT_NODE(NODE_CASE2); - COUNT_NODE(NODE_CASE3); - COUNT_NODE(NODE_WHEN); - COUNT_NODE(NODE_IN); - COUNT_NODE(NODE_WHILE); - COUNT_NODE(NODE_UNTIL); - COUNT_NODE(NODE_ITER); - COUNT_NODE(NODE_FOR); - COUNT_NODE(NODE_FOR_MASGN); - COUNT_NODE(NODE_BREAK); - COUNT_NODE(NODE_NEXT); - COUNT_NODE(NODE_REDO); - COUNT_NODE(NODE_RETRY); - COUNT_NODE(NODE_BEGIN); - COUNT_NODE(NODE_RESCUE); - COUNT_NODE(NODE_RESBODY); - COUNT_NODE(NODE_ENSURE); - COUNT_NODE(NODE_AND); - COUNT_NODE(NODE_OR); - COUNT_NODE(NODE_MASGN); - COUNT_NODE(NODE_LASGN); - COUNT_NODE(NODE_DASGN); - COUNT_NODE(NODE_GASGN); - COUNT_NODE(NODE_IASGN); - COUNT_NODE(NODE_CDECL); - COUNT_NODE(NODE_CVASGN); - COUNT_NODE(NODE_OP_ASGN1); - COUNT_NODE(NODE_OP_ASGN2); - COUNT_NODE(NODE_OP_ASGN_AND); - COUNT_NODE(NODE_OP_ASGN_OR); - COUNT_NODE(NODE_OP_CDECL); - COUNT_NODE(NODE_CALL); - COUNT_NODE(NODE_OPCALL); - COUNT_NODE(NODE_FCALL); - COUNT_NODE(NODE_VCALL); - COUNT_NODE(NODE_QCALL); - COUNT_NODE(NODE_SUPER); - COUNT_NODE(NODE_ZSUPER); - COUNT_NODE(NODE_LIST); - COUNT_NODE(NODE_ZLIST); - COUNT_NODE(NODE_VALUES); - COUNT_NODE(NODE_HASH); - COUNT_NODE(NODE_RETURN); - COUNT_NODE(NODE_YIELD); - COUNT_NODE(NODE_LVAR); - COUNT_NODE(NODE_DVAR); - COUNT_NODE(NODE_GVAR); - COUNT_NODE(NODE_IVAR); - COUNT_NODE(NODE_CONST); - COUNT_NODE(NODE_CVAR); - COUNT_NODE(NODE_NTH_REF); - COUNT_NODE(NODE_BACK_REF); - COUNT_NODE(NODE_MATCH); - COUNT_NODE(NODE_MATCH2); - COUNT_NODE(NODE_MATCH3); - COUNT_NODE(NODE_LIT); - COUNT_NODE(NODE_STR); - COUNT_NODE(NODE_DSTR); - COUNT_NODE(NODE_XSTR); - COUNT_NODE(NODE_DXSTR); - COUNT_NODE(NODE_EVSTR); - COUNT_NODE(NODE_DREGX); - COUNT_NODE(NODE_ONCE); - COUNT_NODE(NODE_ARGS); - COUNT_NODE(NODE_ARGS_AUX); - COUNT_NODE(NODE_OPT_ARG); - COUNT_NODE(NODE_KW_ARG); - COUNT_NODE(NODE_POSTARG); - COUNT_NODE(NODE_ARGSCAT); - COUNT_NODE(NODE_ARGSPUSH); - COUNT_NODE(NODE_SPLAT); - COUNT_NODE(NODE_BLOCK_PASS); - COUNT_NODE(NODE_DEFN); - COUNT_NODE(NODE_DEFS); - COUNT_NODE(NODE_ALIAS); - COUNT_NODE(NODE_VALIAS); - COUNT_NODE(NODE_UNDEF); - COUNT_NODE(NODE_CLASS); - COUNT_NODE(NODE_MODULE); - COUNT_NODE(NODE_SCLASS); - COUNT_NODE(NODE_COLON2); - COUNT_NODE(NODE_COLON3); - COUNT_NODE(NODE_DOT2); - COUNT_NODE(NODE_DOT3); - COUNT_NODE(NODE_FLIP2); - COUNT_NODE(NODE_FLIP3); - COUNT_NODE(NODE_SELF); - COUNT_NODE(NODE_NIL); - COUNT_NODE(NODE_TRUE); - COUNT_NODE(NODE_FALSE); - COUNT_NODE(NODE_ERRINFO); - COUNT_NODE(NODE_DEFINED); - COUNT_NODE(NODE_POSTEXE); - COUNT_NODE(NODE_DSYM); - COUNT_NODE(NODE_ATTRASGN); - COUNT_NODE(NODE_LAMBDA); - COUNT_NODE(NODE_ARYPTN); - COUNT_NODE(NODE_FNDPTN); - COUNT_NODE(NODE_HSHPTN); -#undef COUNT_NODE - case NODE_LAST: break; - } - UNREACHABLE; - set: - rb_hash_aset(hash, node, SIZET2NUM(nodes[i])); - } - } - return hash; -} - -static void cto_i(VALUE v, void *data) { VALUE hash = (VALUE)data; @@ -533,32 +364,22 @@ cto_i(VALUE v, void *data) /* * call-seq: - * ObjectSpace.count_tdata_objects([result_hash]) -> hash - * - * Counts objects for each +T_DATA+ type. - * - * This method is only for MRI developers interested in performance and memory - * usage of Ruby programs. + * ObjectSpace.count_tdata_objects(result_hash = nil) -> hash * - * It returns a hash as: + * Returns a hash containing the number of objects for each +T_DATA+ type. + * The keys are Class objects when the +T_DATA+ object has an associated class, + * or Symbol objects of the name defined in the +rb_data_type_struct+ for internal + * +T_DATA+ objects. * - * {RubyVM::InstructionSequence=>504, :parser=>5, :barrier=>6, - * :mutex=>6, Proc=>60, RubyVM::Env=>57, Mutex=>1, Encoding=>99, - * ThreadGroup=>1, Binding=>1, Thread=>1, RubyVM=>1, :iseq=>1, - * Random=>1, ARGF.class=>1, Data=>1, :autoload=>3, Time=>2} - * # T_DATA objects existing at startup on r32276. + * ObjectSpace.count_tdata_objects + * # => {RBS::Location => 39255, marshal_compat_table: 1, Encoding => 103, mutex: 1, ... } * - * If the optional argument, result_hash, is given, it is overwritten and - * returned. This is intended to avoid probe effect. + * If the optional argument +result_hash+ is given, it is overwritten and + * returned. This is intended to avoid the probe effect. * - * The contents of the returned hash is implementation specific and may change - * in the future. - * - * In this version, keys are Class object or Symbol object. - * - * If object is kind of normal (accessible) object, the key is Class object. - * If object is not a kind of normal (internal) object, the key is symbol - * name, registered by rb_data_type_struct. + * This method is intended for developers interested in performance and memory + * usage of Ruby programs. The contents of the returned hash is implementation + * specific and may change in the future. * * This method is only expected to work with C Ruby. */ @@ -597,28 +418,22 @@ count_imemo_objects_i(VALUE v, void *data) /* * call-seq: - * ObjectSpace.count_imemo_objects([result_hash]) -> hash - * - * Counts objects for each +T_IMEMO+ type. + * ObjectSpace.count_imemo_objects(result_hash = nil) -> hash * - * This method is only for MRI developers interested in performance and memory - * usage of Ruby programs. + * Returns a hash containing the number of objects for each +T_IMEMO+ type. + * The keys are Symbol objects of the +T_IMEMO+ type name. + * +T_IMEMO+ objects are Ruby internal objects that are not visible to Ruby + * programs. * - * It returns a hash as: + * ObjectSpace.count_imemo_objects + * # => {imemo_callcache: 5482, imemo_constcache: 1258, imemo_ment: 13906, ... } * - * {:imemo_ifunc=>8, - * :imemo_svar=>7, - * :imemo_cref=>509, - * :imemo_memo=>1, - * :imemo_throw_data=>1} + * If the optional argument +result_hash+ is given, it is overwritten and + * returned. This is intended to avoid the probe effect. * - * If the optional argument, result_hash, is given, it is overwritten and - * returned. This is intended to avoid probe effect. - * - * The contents of the returned hash is implementation specific and may change - * in the future. - * - * In this version, keys are symbol objects. + * This method is intended for developers interested in performance and memory + * usage of Ruby programs. The contents of the returned hash is implementation + * specific and may change in the future. * * This method is only expected to work with C Ruby. */ @@ -639,11 +454,13 @@ count_imemo_objects(int argc, VALUE *argv, VALUE self) INIT_IMEMO_TYPE_ID(imemo_ment); INIT_IMEMO_TYPE_ID(imemo_iseq); INIT_IMEMO_TYPE_ID(imemo_tmpbuf); - INIT_IMEMO_TYPE_ID(imemo_ast); - INIT_IMEMO_TYPE_ID(imemo_parser_strterm); + INIT_IMEMO_TYPE_ID(imemo_cvar_entry); INIT_IMEMO_TYPE_ID(imemo_callinfo); INIT_IMEMO_TYPE_ID(imemo_callcache); INIT_IMEMO_TYPE_ID(imemo_constcache); + INIT_IMEMO_TYPE_ID(imemo_fields); + INIT_IMEMO_TYPE_ID(imemo_subclasses); + INIT_IMEMO_TYPE_ID(imemo_cdhash); #undef INIT_IMEMO_TYPE_ID } @@ -717,7 +534,7 @@ reachable_object_from_i(VALUE obj, void *data_ptr) VALUE key = obj; VALUE val = obj; - if (rb_objspace_markable_object_p(obj)) { + if (!rb_objspace_garbage_object_p(obj)) { if (NIL_P(rb_hash_lookup(data->refs, key))) { rb_hash_aset(data->refs, key, Qtrue); @@ -742,48 +559,43 @@ collect_values(st_data_t key, st_data_t value, st_data_t data) * call-seq: * ObjectSpace.reachable_objects_from(obj) -> array or nil * - * [MRI specific feature] Return all reachable objects from `obj'. - * - * This method returns all reachable objects from `obj'. + * Returns all reachable objects from +obj+ as an array: * - * If `obj' has two or more references to the same object `x', then returned - * array only includes one `x' object. + * ObjectSpace.reachable_objects_from(['a', 'b', 'c']) + * #=> [Array, 'a', 'b', 'c'] * - * If `obj' is a non-markable (non-heap management) object such as true, - * false, nil, symbols and Fixnums (and Flonum) then it simply returns nil. + * The returned array is deduplicated, meaning that if +obj+ refers + * to another object more than once, it will only be added to the array + * once: * - * If `obj' has references to an internal object, then it returns instances of - * ObjectSpace::InternalObjectWrapper class. This object contains a reference - * to an internal object and you can check the type of internal object with - * `type' method. + * ObjectSpace.reachable_objects_from([v = 'a', v, v]) + * #=> [Array, 'a'] * - * If `obj' is instance of ObjectSpace::InternalObjectWrapper class, then this - * method returns all reachable object from an internal object, which is - * pointed by `obj'. + * Returns +nil+ if +obj+ is not a markable object (i.e. non-heap + * managed) object. Non-markable objects include +true+, +false+, + * +nil+, certain symbols, small integers, and floats: * - * With this method, you can find memory leaks. + * ObjectSpace.reachable_objects_from(1) + * #=> nil * - * This method is only expected to work except with C Ruby. + * All references to internal objects in the returned array are wrapped + * using ObjectSpace::InternalObjectWrapper objects. This object contains + * a reference to the internal object and the type of the object can + * be accessed using the ObjectSpace::InternalObjectWrapper#type method. * - * Example: - * ObjectSpace.reachable_objects_from(['a', 'b', 'c']) - * #=> [Array, 'a', 'b', 'c'] + * If +obj+ is instance of ObjectSpace::InternalObjectWrapper, then this + * method returns all reachable object from the internal object. * - * ObjectSpace.reachable_objects_from(['a', 'a', 'a']) - * #=> [Array, 'a', 'a', 'a'] # all 'a' strings have different object id - * - * ObjectSpace.reachable_objects_from([v = 'a', v, v]) - * #=> [Array, 'a'] - * - * ObjectSpace.reachable_objects_from(1) - * #=> nil # 1 is not markable (heap managed) object + * This method is useful for debugging purposes, such as finding + * memory leaks. * + * This method is only expected to work with C Ruby. */ static VALUE reachable_objects_from(VALUE self, VALUE obj) { - if (rb_objspace_markable_object_p(obj)) { + if (!RB_SPECIAL_CONST_P(obj)) { struct rof_data data; if (rb_typeddata_is_kind_of(obj, &iow_data_type)) { @@ -830,7 +642,7 @@ reachable_object_from_root_i(const char *category, VALUE obj, void *ptr) rb_hash_aset(data->categories, category_str, category_objects); } - if (rb_objspace_markable_object_p(obj) && + if (!rb_objspace_garbage_object_p(obj) && obj != data->categories && obj != data->last_category_objects) { if (rb_objspace_internal_object_p(obj)) { @@ -932,7 +744,7 @@ objspace_internal_super_of(VALUE self, VALUE obj) case T_MODULE: case T_CLASS: case T_ICLASS: - super = RCLASS_SUPER(obj); + super = rb_class_super_of(obj); break; default: rb_raise(rb_eArgError, "class or module is expected"); @@ -953,7 +765,7 @@ void Init_objspace_dump(VALUE rb_mObjSpace); * * You need to <code>require 'objspace'</code> to use this extension module. * - * Generally, you *SHOULD NOT* use this library if you do not know + * Generally, you *SHOULD* *NOT* use this library if you do not know * about the MRI implementation. Mainly, this library is for (memory) * profiler developers and MRI developers who need to know about MRI * memory usage. @@ -974,7 +786,6 @@ Init_objspace(void) rb_define_module_function(rb_mObjSpace, "count_objects_size", count_objects_size, -1); rb_define_module_function(rb_mObjSpace, "count_symbols", count_symbols, -1); - rb_define_module_function(rb_mObjSpace, "count_nodes", count_nodes, -1); rb_define_module_function(rb_mObjSpace, "count_tdata_objects", count_tdata_objects, -1); rb_define_module_function(rb_mObjSpace, "count_imemo_objects", count_imemo_objects, -1); diff --git a/ext/objspace/objspace_dump.c b/ext/objspace/objspace_dump.c index 2917d49331..68ada01af3 100644 --- a/ext/objspace/objspace_dump.c +++ b/ext/objspace/objspace_dump.c @@ -12,26 +12,34 @@ **********************************************************************/ -#include "gc.h" +#include "id_table.h" #include "internal.h" +#include "internal/array.h" +#include "internal/class.h" +#include "internal/gc.h" #include "internal/hash.h" +#include "internal/io.h" #include "internal/string.h" #include "internal/sanitizers.h" +#include "symbol.h" +#include "shape.h" #include "node.h" #include "objspace.h" #include "ruby/debug.h" #include "ruby/util.h" #include "ruby/io.h" -#include "vm_core.h" +#include "vm_callinfo.h" +#include "vm_sync.h" RUBY_EXTERN const char ruby_hexdigits[]; #define BUFFER_CAPACITY 4096 struct dump_config { - VALUE type; - VALUE stream; + VALUE given_output; + VALUE output_io; VALUE string; + FILE *stream; const char *root_category; VALUE cur_obj; VALUE cur_obj_klass; @@ -41,6 +49,7 @@ struct dump_config { unsigned int full_heap: 1; unsigned int partial_dump; size_t since; + size_t shapes_since; unsigned long buffer_len; char buffer[BUFFER_CAPACITY]; }; @@ -50,7 +59,7 @@ dump_flush(struct dump_config *dc) { if (dc->buffer_len) { if (dc->stream) { - size_t written = rb_io_bufwrite(dc->stream, dc->buffer, dc->buffer_len); + size_t written = fwrite(dc->buffer, sizeof(dc->buffer[0]), dc->buffer_len, dc->stream); if (written < dc->buffer_len) { MEMMOVE(dc->buffer, dc->buffer + written, char, dc->buffer_len - written); dc->buffer_len -= written; @@ -160,10 +169,8 @@ dump_append_c(struct dump_config *dc, unsigned char c) } static void -dump_append_ref(struct dump_config *dc, VALUE ref) +dump_append_ptr(struct dump_config *dc, VALUE ref) { - RUBY_ASSERT(ref > 0); - char buffer[roomof(sizeof(VALUE) * CHAR_BIT, 4) + rb_strlen_lit("\"0x\"")]; char *buffer_start, *buffer_end; @@ -180,6 +187,14 @@ dump_append_ref(struct dump_config *dc, VALUE ref) } static void +dump_append_ref(struct dump_config *dc, VALUE ref) +{ + RUBY_ASSERT(ref > 0); + dump_append_ptr(dc, ref); +} + + +static void dump_append_string_value(struct dump_config *dc, VALUE obj) { long i; @@ -349,14 +364,28 @@ dump_append_string_content(struct dump_config *dc, VALUE obj) } } +static inline void +dump_append_id(struct dump_config *dc, ID id) +{ + VALUE str = rb_sym2str(ID2SYM(id)); + if (RTEST(str)) { + dump_append_string_value(dc, str); + } + else { + dump_append(dc, "\"ID_INTERNAL("); + dump_append_sizet(dc, rb_id_to_serial(id)); + dump_append(dc, ")\""); + } +} + + static void dump_object(VALUE obj, struct dump_config *dc) { size_t memsize; struct allocation_info *ainfo = objspace_lookup_allocation_info(obj); rb_io_t *fptr; - ID flags[RB_OBJ_GC_FLAGS_MAX]; - size_t n, i; + ID mid; if (SPECIAL_CONST_P(obj)) { dump_append_special_const(dc, obj); @@ -365,7 +394,12 @@ dump_object(VALUE obj, struct dump_config *dc) dc->cur_obj = obj; dc->cur_obj_references = 0; - dc->cur_obj_klass = BUILTIN_TYPE(obj) == T_NODE ? 0 : RBASIC_CLASS(obj); + if (BUILTIN_TYPE(obj) == T_NODE || (BUILTIN_TYPE(obj) == T_IMEMO && !IMEMO_TYPE_P(obj, imemo_fields))) { + dc->cur_obj_klass = 0; + } + else { + dc->cur_obj_klass = RBASIC_CLASS(obj); + } if (dc->partial_dump && (!ainfo || ainfo->generation < dc->since)) { return; @@ -381,6 +415,12 @@ dump_object(VALUE obj, struct dump_config *dc) dump_append(dc, obj_type(obj)); dump_append(dc, "\""); + if (BUILTIN_TYPE(obj) != T_IMEMO || IMEMO_TYPE_P(obj, imemo_fields)) { + size_t shape_id = rb_obj_shape_id(obj) & SHAPE_ID_OFFSET_MASK; + dump_append(dc, ", \"shape_id\":"); + dump_append_sizet(dc, shape_id); + } + dump_append(dc, ", \"slot_size\":"); dump_append_sizet(dc, dc->cur_page_slot_size); @@ -400,6 +440,36 @@ dump_object(VALUE obj, struct dump_config *dc) dump_append(dc, ", \"imemo_type\":\""); dump_append(dc, rb_imemo_name(imemo_type(obj))); dump_append(dc, "\""); + + switch (imemo_type(obj)) { + case imemo_callinfo: + mid = vm_ci_mid((const struct rb_callinfo *)obj); + if (mid != 0) { + dump_append(dc, ", \"mid\":"); + dump_append_id(dc, mid); + } + break; + + case imemo_callcache: + { + VALUE klass = ((const struct rb_callcache *)obj)->klass; + if (klass != Qundef) { + mid = vm_cc_cme((const struct rb_callcache *)obj)->called_id; + if (mid != 0) { + dump_append(dc, ", \"called_id\":"); + dump_append_id(dc, mid); + + } + + dump_append(dc, ", \"receiver_class\":"); + dump_append_ref(dc, klass); + } + } + break; + + default: + break; + } break; case T_SYMBOL: @@ -411,6 +481,8 @@ dump_object(VALUE obj, struct dump_config *dc) dump_append(dc, ", \"embedded\":true"); if (FL_TEST(obj, RSTRING_FSTR)) dump_append(dc, ", \"fstring\":true"); + if (CHILLED_STRING_P(obj)) + dump_append(dc, ", \"chilled\":true"); if (STR_SHARED_P(obj)) dump_append(dc, ", \"shared\":true"); else @@ -456,7 +528,7 @@ dump_object(VALUE obj, struct dump_config *dc) case T_ARRAY: dump_append(dc, ", \"length\":"); dump_append_ld(dc, RARRAY_LEN(obj)); - if (RARRAY_LEN(obj) > 0 && FL_TEST(obj, ELTS_SHARED)) + if (RARRAY_LEN(obj) > 0 && FL_TEST(obj, RARRAY_SHARED_FLAG)) dump_append(dc, ", \"shared\":true"); if (FL_TEST(obj, RARRAY_EMBED_FLAG)) dump_append(dc, ", \"embedded\":true"); @@ -470,6 +542,9 @@ dump_object(VALUE obj, struct dump_config *dc) break; case T_CLASS: + dump_append(dc, ", \"variation_count\":"); + dump_append_d(dc, rb_class_variation_count(obj)); + case T_MODULE: if (rb_class_get_superclass(obj)) { dump_append(dc, ", \"superclass\":"); @@ -479,9 +554,8 @@ dump_object(VALUE obj, struct dump_config *dc) if (dc->cur_obj_klass) { VALUE mod_name = rb_mod_name(obj); if (!NIL_P(mod_name)) { - dump_append(dc, ", \"name\":\""); - dump_append(dc, RSTRING_PTR(mod_name)); - dump_append(dc, "\""); + dump_append(dc, ", \"name\":"); + dump_append_string_value(dc, mod_name); } else { VALUE real_mod_name = rb_mod_name(rb_class_real(obj)); @@ -492,17 +566,22 @@ dump_object(VALUE obj, struct dump_config *dc) } } - if (FL_TEST(obj, FL_SINGLETON)) { + if (rb_class_singleton_p(obj)) { dump_append(dc, ", \"singleton\":true"); } } break; case T_DATA: - if (RTYPEDDATA_P(obj)) { + { + const rb_data_type_t *type = RTYPEDDATA_TYPE(obj); dump_append(dc, ", \"struct\":\""); - dump_append(dc, RTYPEDDATA_TYPE(obj)->wrap_struct_name); + dump_append(dc, type->wrap_struct_name); dump_append(dc, "\""); + if (!(type->flags & RUBY_TYPED_FREE_IMMEDIATELY) && + type->function.dfree != RUBY_DEFAULT_FREE) { + dump_append(dc, ", \"free_immediately\":false"); + } } break; @@ -513,8 +592,15 @@ dump_object(VALUE obj, struct dump_config *dc) break; case T_OBJECT: + if (!FL_TEST_RAW(obj, ROBJECT_HEAP)) { + dump_append(dc, ", \"embedded\":true"); + } + dump_append(dc, ", \"ivars\":"); - dump_append_lu(dc, ROBJECT_NUMIV(obj)); + dump_append_lu(dc, ROBJECT_FIELDS_COUNT(obj)); + if (rb_obj_shape_complex_p(obj)) { + dump_append(dc, ", \"complex_shape\":true"); + } break; case T_FILE: @@ -561,14 +647,24 @@ dump_object(VALUE obj, struct dump_config *dc) dump_append_sizet(dc, memsize); } - if ((n = rb_obj_gc_flags(obj, flags, sizeof(flags))) > 0) { - dump_append(dc, ", \"flags\":{"); - for (i=0; i<n; i++) { - dump_append(dc, "\""); - dump_append(dc, rb_id2name(flags[i])); - dump_append(dc, "\":true"); - if (i != n-1) dump_append(dc, ", "); + struct rb_gc_object_metadata_entry *gc_metadata = rb_gc_object_metadata(obj); + for (int i = 0; gc_metadata[i].name != 0; i++) { + if (i == 0) { + dump_append(dc, ", \"flags\":{"); } + else { + dump_append(dc, ", "); + } + + dump_append(dc, "\""); + dump_append(dc, rb_id2name(gc_metadata[i].name)); + dump_append(dc, "\":"); + dump_append_special_const(dc, gc_metadata[i].val); + } + + /* If rb_gc_object_metadata had any entries, we need to close the opening + * `"flags":{`. */ + if (gc_metadata[0].name != 0) { dump_append(dc, "}"); } @@ -581,15 +677,15 @@ heap_i(void *vstart, void *vend, size_t stride, void *data) struct dump_config *dc = (struct dump_config *)data; VALUE v = (VALUE)vstart; for (; v != (VALUE)vend; v += stride) { - void *ptr = asan_poisoned_object_p(v); - asan_unpoison_object(v, false); + void *ptr = rb_asan_poisoned_object_p(v); + rb_asan_unpoison_object(v, false); dc->cur_page_slot_size = stride; if (dc->full_heap || RBASIC(v)->flags) dump_object(v, dc); if (ptr) { - asan_poison_object(v); + rb_asan_poison_object(v); } } return 0; @@ -618,18 +714,36 @@ root_obj_i(const char *category, VALUE obj, void *data) } static void -dump_output(struct dump_config *dc, VALUE output, VALUE full, VALUE since) +dump_output(struct dump_config *dc, VALUE output, VALUE full, VALUE since, VALUE shapes) { - + dc->given_output = output; dc->full_heap = 0; dc->buffer_len = 0; if (TYPE(output) == T_STRING) { - dc->stream = Qfalse; + dc->stream = NULL; dc->string = output; } else { - dc->stream = output; + rb_io_t *fptr; + // Output should be an IO, typecheck and get a FILE* for writing. + // We cannot write with the usual IO code here because writes + // interleave with calls to rb_gc_mark(). The usual IO code can + // cause a thread switch, raise exceptions, and even run arbitrary + // ruby code through the fiber scheduler. + // + // Mark functions generally can't handle these possibilities so + // the usual IO code is unsafe in this context. (For example, + // there are many ways to crash when ruby code runs and mutates + // the execution context while rb_execution_context_mark() is in + // progress.) + // + // Using FILE* isn't perfect, but it avoids the most acute problems. + output = rb_io_get_io(output); + dc->output_io = rb_io_get_write_io(output); + rb_io_flush(dc->output_io); + GetOpenFile(dc->output_io, fptr); + dc->stream = rb_io_stdio_file(fptr); dc->string = Qfalse; } @@ -644,6 +758,8 @@ dump_output(struct dump_config *dc, VALUE output, VALUE full, VALUE since) else { dc->partial_dump = 0; } + + dc->shapes_since = RTEST(shapes) ? NUM2SIZET(shapes) : 0; } static VALUE @@ -651,35 +767,100 @@ dump_result(struct dump_config *dc) { dump_flush(dc); + if (dc->stream) { + fflush(dc->stream); + } if (dc->string) { return dc->string; } - else { - rb_io_flush(dc->stream); - return dc->stream; - } + return dc->given_output; } static VALUE -objspace_dump(VALUE os, VALUE obj, VALUE output) +dump_locked(void *args_p) { struct dump_config dc = {0,}; + VALUE obj = ((VALUE*)args_p)[0]; + VALUE output = ((VALUE*)args_p)[1]; + if (!RB_SPECIAL_CONST_P(obj)) { dc.cur_page_slot_size = rb_gc_obj_slot_size(obj); } - - dump_output(&dc, output, Qnil, Qnil); + dump_output(&dc, output, Qnil, Qnil, Qnil); dump_object(obj, &dc); return dump_result(&dc); } +/* :nodoc: */ +static VALUE +objspace_dump(VALUE os, VALUE obj, VALUE output) +{ + VALUE args[2]; + args[0] = obj; + args[1] = output; + return rb_vm_lock_with_barrier(dump_locked, (void*)args); +} + +static void +shape_id_i(shape_id_t shape_id, void *data) +{ + struct dump_config *dc = (struct dump_config *)data; + + if (shape_id < dc->shapes_since) { + return; + } + + dump_append(dc, "{\"address\":"); + dump_append_ref(dc, (VALUE)RSHAPE(shape_id)); + + dump_append(dc, ", \"type\":\"SHAPE\", \"id\":"); + dump_append_sizet(dc, shape_id); + + if (RSHAPE_TYPE(shape_id) != SHAPE_ROOT) { + dump_append(dc, ", \"parent_id\":"); + dump_append_lu(dc, RSHAPE_PARENT_OFFSET(shape_id)); + } + + dump_append(dc, ", \"depth\":"); + dump_append_sizet(dc, rb_shape_depth(shape_id)); + + switch (RSHAPE_TYPE(shape_id)) { + case SHAPE_ROOT: + dump_append(dc, ", \"shape_type\":\"ROOT\""); + break; + case SHAPE_IVAR: + dump_append(dc, ", \"shape_type\":\"IVAR\""); + + dump_append(dc, ",\"edge_name\":"); + dump_append_id(dc, RSHAPE_EDGE_NAME(shape_id)); + + break; + case SHAPE_OBJ_ID: + dump_append(dc, ", \"shape_type\":\"OBJ_ID\""); + break; + } + + dump_append(dc, ", \"edges\":"); + dump_append_sizet(dc, rb_shape_edges_count(shape_id)); + + dump_append(dc, ", \"memsize\":"); + dump_append_sizet(dc, rb_shape_memsize(shape_id)); + + dump_append(dc, "}\n"); +} + static VALUE -objspace_dump_all(VALUE os, VALUE output, VALUE full, VALUE since) +dump_all_locked(void *args_p) { struct dump_config dc = {0,}; - dump_output(&dc, output, full, since); + VALUE output = ((VALUE*)args_p)[0]; + VALUE full = ((VALUE*)args_p)[1]; + VALUE since = ((VALUE*)args_p)[2]; + VALUE shapes = ((VALUE*)args_p)[3]; + + dump_output(&dc, output, full, since, shapes); if (!dc.partial_dump || dc.since == 0) { /* dump roots */ @@ -687,12 +868,53 @@ objspace_dump_all(VALUE os, VALUE output, VALUE full, VALUE since) if (dc.roots) dump_append(&dc, "]}\n"); } + if (RTEST(shapes)) { + rb_shape_each_shape_id(shape_id_i, &dc); + } + /* dump all objects */ rb_objspace_each_objects(heap_i, &dc); return dump_result(&dc); } +/* :nodoc: */ +static VALUE +objspace_dump_all(VALUE os, VALUE output, VALUE full, VALUE since, VALUE shapes) +{ + VALUE args[4]; + args[0] = output; + args[1] = full; + args[2] = since; + args[3] = shapes; + return rb_vm_lock_with_barrier(dump_all_locked, (void*)args); +} + +static VALUE +dump_shapes_locked(void *args_p) +{ + struct dump_config dc = {0,}; + VALUE output = ((VALUE*)args_p)[0]; + VALUE shapes = ((VALUE*)args_p)[1]; + + dump_output(&dc, output, Qfalse, Qnil, shapes); + + if (RTEST(shapes)) { + rb_shape_each_shape_id(shape_id_i, &dc); + } + return dump_result(&dc); +} + +/* :nodoc: */ +static VALUE +objspace_dump_shapes(VALUE os, VALUE output, VALUE shapes) +{ + VALUE args[2]; + args[0] = output; + args[1] = shapes; + return rb_vm_lock_with_barrier(dump_shapes_locked, (void*)args); +} + void Init_objspace_dump(VALUE rb_mObjSpace) { @@ -700,10 +922,11 @@ Init_objspace_dump(VALUE rb_mObjSpace) #if 0 rb_mObjSpace = rb_define_module("ObjectSpace"); /* let rdoc know */ #endif +#ifdef HAVE_RB_EXT_RACTOR_SAFE + RB_EXT_RACTOR_SAFE(true); +#endif rb_define_module_function(rb_mObjSpace, "_dump", objspace_dump, 2); - rb_define_module_function(rb_mObjSpace, "_dump_all", objspace_dump_all, 3); - - /* force create static IDs */ - rb_obj_gc_flags(rb_mObjSpace, 0, 0); + rb_define_module_function(rb_mObjSpace, "_dump_all", objspace_dump_all, 4); + rb_define_module_function(rb_mObjSpace, "_dump_shapes", objspace_dump_shapes, 2); } |