In addition to detect dead canary, we try to detect the very moment
when we smash the stack top. Requested by k0kubun:
https://twitter.com/k0kubun/status/1085180749899194368
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@66981 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* bootstraptest/runner.rb (assert_normal_exit): check MJIT first
to support btest with ruby ~2.5. btest (bootstraptest) should be
enable to run with stable ruby interpreter because modified ruby
may not able to run runner.rb and we need to know why (this is why
we introduce btest).
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@66842 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
"hash_bulk_insert" first expands the table, but the target size was
wrong: it was calculated by "num_entries + (size to buld insert)", but
it was wrong when "num_entries < entries_bound", i.e., it has a deleted
entry. "hash_bulk_insert" adds the given entries from entries_bound,
which led to out-of-bounds write access. [Bug #15536]
As a simple fix, this commit changes the calculation to "entries_bound +
size". I'm afraid if this might be inefficient, but I think it is safe
anyway.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@66832 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
because test_io.rb:33 randomly fails
http://ci.rvm.jp/results/trunk-mjit-wait@silicon-docker/1519055
checking MJIT.enabled? on driver might not make sense for target, but as
long as the CI is -DMJIT_FORCE_ENABLE, I believe it works for now.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@66344 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* hash.c (env_enc_str_new): as no locale/filesystem encoding is
available in miniruby on Windows, fallback the encoding to
ASCII-8BIT so it is valid encoding when the conversion failed.
[ruby-core:89177] [Bug #15164]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64860 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Resurrect r63754 in a 1.8-compatible way. While we're at it,
add a note to maintain 1.8 compatibility (cf. r63757).
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64120 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
```
if L1
L0:
jump L2
L1:
...
L2:
```
was wrongly optimized to:
```
unless L2
L0:
L1:
...
L2:
```
To make it conservative, this optimization is now disabled when there is
any label between `if` and `jump` instructions.
Fixes [Bug #14897].
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63868 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
I'm still using the computer from 2005, so enabling MJIT makes
some tests take longer. For test_deadlock_by_signal_at_forking
I got it down to 135s to 89s by disabling RubyGems.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63774 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
We may use IO.select to watch for process exit. This speeds up
"make test" by 2 seconds for me.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63754 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* error.c (rb_exc_new, rb_exc_new_str): instantiate exception
object directly without Exception.new method call.
Redefinition of class method `new` is an outdated style, and
internal exceptions should not be affected by it.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62500 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* vm_core.h (rb_vm_t): move `rb_execution_context_t::safe_level` to
`rb_vm_t::safe_level_` because `$SAFE` is a process (VM) global state.
* vm_core.h (rb_proc_t): remove `rb_proc_t::safe_level` because `Proc`
objects don't need to keep `$SAFE` at the creation.
Also make `is_from_method` and `is_lambda` as 1 bit fields.
* cont.c (cont_restore_thread): no need to keep `$SAFE` for Continuation.
* eval.c (ruby_cleanup): use `rb_set_safe_level_force()` instead of access
`vm->safe_level_` directly.
* eval_jump.c: End procs `END{}` doesn't keep `$SAFE`.
* proc.c (proc_dup): removed and introduce `rb_proc_dup` in vm.c.
* safe.c (rb_set_safe_level): don't check `$SAFE` 1 -> 0 changes.
* safe.c (safe_setter): use `rb_set_safe_level()`.
* thread.c (rb_thread_safe_level): `Thread#safe_level` returns `$SAFE`.
It should be obsolete.
* transcode.c (load_transcoder_entry): `rb_safe_level()` only returns
0 or 1 so that this check is not needed.
* vm.c (vm_proc_create_from_captured): don't need to keep `$SAFE` for Proc.
* vm.c (rb_proc_create): renamed to `proc_create`.
* vm.c (rb_proc_dup): moved from proc.c.
* vm.c (vm_invoke_proc): do not need to set and restore `$SAFE`
for `Proc#call`.
* vm_eval.c (rb_eval_cmd): rename a local variable to represent clearer
meaning.
* lib/drb/drb.rb: restore `$SAFE`.
* lib/erb.rb: restore `$SAFE`, too.
* test/lib/leakchecker.rb: check `$SAFE == 0` at the end of tests.
* test/rubygems/test_gem.rb: do not set `$SAFE = 1`.
* bootstraptest/test_proc.rb: catch up this change.
* spec/ruby/optional/capi/string_spec.rb: ditto.
* test/bigdecimal/test_bigdecimal.rb: ditto.
* test/fiddle/test_func.rb: ditto.
* test/fiddle/test_handle.rb: ditto.
* test/net/imap/test_imap_response_parser.rb: ditto.
* test/pathname/test_pathname.rb: ditto.
* test/readline/test_readline.rb: ditto.
* test/ruby/test_file.rb: ditto.
* test/ruby/test_optimization.rb: ditto.
* test/ruby/test_proc.rb: ditto.
* test/ruby/test_require.rb: ditto.
* test/ruby/test_thread.rb: ditto.
* test/rubygems/test_gem_specification.rb: ditto.
* test/test_tempfile.rb: ditto.
* test/test_tmpdir.rb: ditto.
* test/win32ole/test_win32ole.rb: ditto.
* test/win32ole/test_win32ole_event.rb: ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61510 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Because NaCl and PNaCl are already sunset status.
see https://bugs.chromium.org/p/chromium/issues/detail?id=239656#c160
configure.ac: Patch for this file was provided by @nobu.
[Feature #14041][ruby-core:83497][fix GH-1726]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60374 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/limits.c.tmpl (Init_limits): rename RbConfig::Limits as
RbConfig::LIMITS, constants other than class or module are all
uppercase with underscores by convention.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@58260 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
The same optimization used for "literal string".freeze
can easily apply to uminus without introducing any
compatibility problems.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57828 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Because our tests now have several places where FIXNUM_MAX is needed,
we decided to provide it along with several other constants.
* template/limits.c.tmpl: new file, defining RbConfig::Limits
* ext/rbconfig/sizeof/depend (limits.c): rule to generate limits.c
* test/-ext-/num2int/test_num2int.rb: use RbConfig::Limits
* bootstraptest/test_insns.rb: ditto.
* .gitignore: ignore new generated file.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57465 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
While I was developing my private topic branch I found that the VM
itself is not tested very much in `make test` tests. Of course
`make test-all` covers vast majority of the VM but running that task
is not an immediately possible thing when we are touching the VM. In
order to boost development in a rapid cycle I decided to add some
tests to the bootstraptest. Here it is.
* test_insns.rb: new test that covers insns.def.
* runner.rb (#assert_equal): pass extra options to the target
so that we can test frozen_string_literal: true situation.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57399 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* bootstraptest/runner.rb (pretty): remove empty line at the
beginning only.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56394 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* bootstraptest/runner.rb: do not use safe navigation operator.
this runner may run on older ruby. partially revert r53110
(GH-1142 patched by @mlarraz).
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@55526 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* [Feature #12005] Unify Fixnum and Bignum into Integer
* include/ruby/ruby.h (rb_class_of): Return rb_cInteger for fixnums.
* insns.def (INTEGER_REDEFINED_OP_FLAG): Unified from
FIXNUM_REDEFINED_OP_FLAG and BIGNUM_REDEFINED_OP_FLAG.
* vm_core.h: Ditto.
* vm_insnhelper.c (opt_eq_func): Use INTEGER_REDEFINED_OP_FLAG instead
of FIXNUM_REDEFINED_OP_FLAG.
* vm.c (vm_redefinition_check_flag): Use rb_cInteger instead of
rb_cFixnum and rb_cBignum.
(C): Use Integer instead of Fixnum and Bignum.
* numeric.c (fix_succ): Removed.
(Init_Numeric): Define Fixnum as Integer.
* bignum.c (bignew): Use rb_cInteger instead of Rb_cBignum.
(rb_int_coerce): replaced from rb_big_coerce and return fixnums
as-is.
(Init_Bignum): Define Bignum as Integer.
Don't define ===.
* error.c (builtin_class_name): Return "Integer" for fixnums.
* sprintf.c (ruby__sfvextra): Use rb_cInteger instead of rb_cFixnum.
* ext/-test-/testutil: New directory to test.
Currently it provides utilities for fixnum and bignum.
* ext/json/generator/generator.c: Define mInteger_to_json.
* lib/mathn.rb (Fixnum#/): Redefinition removed.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@55024 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
[Bug #10853][ruby-core:68110]
* bootstraptest/test_string.rb: test for above.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@49586 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* vm_insnhelper.c (vm_throw_start): search the target to break
from a block with nested rescue, from the nested blocks.
[ruby-core:67765] [Bug #10775] [Fix GH-820]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@49390 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* bootstraptest/runner.rb (exec_test, show_progress): erase with
white spaces, and adjust each results at same column.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@46396 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* bootstraptest/test_thread.rb: remove a garbage line for debug.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@46380 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* bootstraptest/runner.rb (show_progress): show each elapsed times
in verbose mode.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@46378 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* bootstraptest/test_thread.rb: String#=~ does not accept String.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@44016 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* vm.c (Init_BareVM): initialize defined_module_hash here,
Init_top_self() is too late to register core classes/modules.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@43273 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* parse.y (parser_number_literal_suffix): return bit set of found
suffixes.
* parse.y (parser_set_number_literal, parser_set_integer_literal):
split from parser_number_literal_suffix to set yyvlal.
* parse.y (parser_yylex): parse rational number literal with decimal
point precisely.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@42332 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
statement, and call rb_bug() if an unknown type is passed to
negate_lit(). [ruby-core:56316] [Bug #8717]
* bootstraptest/test_literal_suffix.rb (assert_equal): add test
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@42323 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
examine that "1if true" and "1rescue nil" are recognized as 1.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@42315 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
to rationalize a Float instance with a precision.
* rational.c (rb_flt_rationalize): new public C function to
rationalize a Float instance. A precision is calculated from
the given float number.
* include/ruby/intern.h: Add rb_flt_rationalize_with_prec and
rb_flt_rationalize.
* parse.y: implement number literal suffixes, 'r' and 'i'.
[ruby-core:55096] [Feature #8430]
* bootstraptest/test_literal_suffix.rb: add tests for parser to scan
number literals with the above tsuffixes.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@42311 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
remove tests for $SAFE=4.
* lib/pp.rb: use taint instead of untrust to avoid warnings when
$VERBOSE is set to true.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@41269 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
A script "[print(1)]; print(2)" should output "12".
However, the compiler had eliminted "[print(1)]" expression
because it is void expression (unused array).
Of course, side-effect should be remained.
This issue is reported by Masaya Tarui.
* bootstraptest/test_literal.rb: add a test.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@38751 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* io.c (rb_cloexec_fcntl_dupfd) Use an emulation with dup(2) when
fcntl(2) and/or F_DUPFD is unavailable.
Suggested by akr.
* configure.in (HAVE_FCNTL): NativeClient does not provide fcntl(2).
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@38713 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
to point block.
[ruby-core:41038] [ruby-trunk - Bug #5634]
* vm.c (rb_vm_make_proc): No need to make Proc object here.
* bootstraptest/test_proc.rb: add tests.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@37898 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
threads.
This benchmark seems consuming long time on travis-ci
several times (and make `failure').
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@37859 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
ci->argc and ci->blockptr before and after method invocations
because these method dispatches override call_info.
* bootstraptest/test_method.rb: add tests for this fix.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@37641 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* vm.c (rb_vm_rewrite_ep_in_errinfo): rewrite all catch points in
errinfo, not only the topmost frame. based on the patch by
ktsj (Kazuki Tsujimoto) in [ruby-dev:45656]. [Bug #6460]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@37430 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
this method call needs splat argument because cahced functions
(vm_call_attrset, vm_call_ivar, vm_call_cfunc_fast_(unary|binary))
do not check an arity.
* bootstraptest/test_method.rb: add a test to check an above issue.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@37199 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
remove checkincludearray instruction and
add new instruction checkmatch.
This change is to solve
[Bug #4438] "rescue args type check omitted".
* iseq.c: increment ISEQ_MAJOR_VERSION because removal of
checkincludearray instruction.
* vm_core.h: add several definitions for
the checkmatch instruction.
* vm_insnhelper.c (check_match): added.
* bootstraptest/test_exception.rb: add a test.
* test/ruby/test_exception.rb: ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@36658 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* class.c (rb_special_singleton_class_of): utility function.
* vm_eval.c (eval_under): special deal for class variable scope with
instance_eval.
* vm_eval.c (rb_obj_instance_eval, rb_obj_instance_exec): allow method
definition in instance_eval of special constants. [ruby-core:28324]
[Bug #2788]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@36647 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* bootstraptest/runner.rb (show_progress): refine error output. do not
count non-empty error message, but just warn.
* bootstraptest/runner.rb (error): show errors immediately if tty.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@36319 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
th->base_block is information for (a) parsing, (b) compiling
and (c) setting up the frame to execute the program passed by
`eval' method. For example, (1) parser need to know up-level
variables to detect it is variable or method without paren.
Befor (a), (b) and (c), VM set th->base_block by passed bindng
(or previous frame information). After execute (a), (b) and (c),
VM should clear th->base_block. However, if (a), (b) or (c)
raises an exception, then th->base_block is not cleared.
Problem is that the uncleared value th->balo_block is used for
irrelevant iseq compilation. It causes SEGV or critical error.
I tried to solve this problem: to clear them before exception,
but finally I found out that it is difficult to do it (Ruby
program can be run in many places).
Because of this background, I set th->base_block before
compiling iseq and restore it after compiling.
Basically, th->base_block is dirty hack (similar to global
variable) and this patch is also dirty.
* bootstraptest/test_eval.rb: add a test for above.
* internal.h: remove unused decl.
* iseq.c (rb_iseq_compile_with_option): add base_block parameter.
set th->base_block before compation and restore it after
compilation.
* ruby.c (require_libraries): pass 0 as base_block instead of
setting th->base_block
* tool/compile_prelude.rb (prelude_eval): apply above changes.
* vm.c, vm_eval.c: ditto.
* vm_core.h: add comments.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@36179 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* bootstraptest/runner.rb (main): add --color option to the help
message.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@36093 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Divide big array (or hash) literals into several blocks and
concatetene them. There was a problem that a big array (hash)
literal causes SystemStackError exception (stack overflow)
because VM push all contents of the literal onto VM stack to
make an array (or hash). To solve this issue, we make several
arrays (hashes) and concatenate them to make a big array (hash)
object.
??
* compile.c (iseq_compile_each, setup_args): use modified
compile_array.
* vm.c (m_core_hash_from_ary, m_core_hash_merge_ary,
m_core_hash_merge_ptr): added for above change.
* id.c (Init_id), parse.y: add core method ids.
* bootstraptest/test_literal.rb: add simple tests.
* bootstraptest/test_eval.rb: remove rescue clause to catch
SystemStackError exception.
* test/ruby/test_literal.rb: add tests to check no stack overflow.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@35306 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
just before calling rb_call0.
* bootstraptest/test_flow.rb: add a test for above.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@34113 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
fixes: Bug #2729
a patch from Kazuki Tsujimoto <kazuki@callcc.net>
This problem is caused by changing dfp (dynamic env pointer)
from saved dfp. Saved dfp is pointed env in VM stack. However,
the dfp can be moved because VM copies env from VM stack to
the heap. At this copying, dfp was also changed. To solve this
problem, I'll try to change throw mechanism (not save target dfp,
but save target cfp).
* bootstraptest/test_flow.rb: add a test for above.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@33064 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Fixes Bug #4648.
The patch is contributed by Kazuki Tsujimoto.
* bootstraptest/test_proc.rb: add tests for above.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32495 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
'inhibit_thread_createion' field.
* thread.c (rb_thread_terminate_all): set inhibit_thread_creation.
* thread.c (thread_s_new): don't permit to create new thread
if the VM is under destruction. Otherwise evil finalizer code
can make SEGV. [Bug #4992][ruby-core:37858]
* bootstraptest/test_objectspace.rb: new test for this fix.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32492 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* thread.c (rb_threadptr_execute_interrupts_common): check signal
deliverly if it is main thread.
fixes [ruby-dev:44005] [Ruby 1.9 - Bug #4950]
* bootstraptest/test_fork.rb: add a test for above.
* signal.c (rb_get_next_signal): skip if signal_buff is empty.
(check signal_buff.size first)
* vm_core.h: remove unused variable rb_thread_t::exec_signal.
* thread.c (rb_thread_check_trap_pending): check
rb_signal_buff_size() because rb_thread_t::exec_signal
is no longer available.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32345 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
consistency. [ruby-core:30293]
Now, lfinish[0] of defined_expr seems not to be used. Refactoring
may be needed.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@27892 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
vm_throw): allow "return" and "yield" even in singleton class
definition. based on a patch from wanabe <s.wanabe AT gmail.com>
for "return". [ruby-core:21379] [ruby-dev:40975]
* insns.def (defineclass): ditto (straightforwardly push block ptr,
instead of dfp ptr with special flag).
* vm_core.h (RUBY_VM_CLASS_SPECIAL_P): ditto (no longer needed).
* proc.c (proc_new): ditto (remove handling for special flag).
* bootstraptest/test_jump.rb: add tests for above.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@27714 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
added. Unlinked method entries are collected to
vm->unlinked_method_entry_list. On the GC timing, mark all method
entries which are on all living threads. Only non-marked method
entries are collected. This hack prevents releasing living method
entry.
[Performance Consideration] Since this Method Entry GC (MEGC)
doesn't occuer frequently, MEGC will not be a performance bottleneck.
However, to traverse living method entries, every control frame push
needs to clear cfp->me field. This will be a performance issue
(because pushing control frame is occurred frequently).
Bug #2777 [ruby-dev:40457]
* cont.c (fiber_init): init cfp->me.
* gc.c (garbage_collect): kick rb_sweep_method_entry().
* method.h (rb_method_entry_t): add a mark field.
* vm.c (invoke_block_from_c): set passed me.
* vm.c (rb_thread_mark): mark cfp->me.
* vm_core.h (rb_thread_t): add a field passed_me.
* vm_core.h (rb_vm_t): add a field unlinked_method_entry_list.
* vm_insnhelper.c (vm_push_frame): clear cfp->me at all times.
* vm_insnhelper.c (vm_call_bmethod): pass me.
* bootstraptest/test_method.rb: add a test.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@27634 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Kazuhiro NISHIYAMA
Nobuyoshi Nakada
Samuel Williams
git
nobu
shyouhei
naruse
ko1
mame
k0kubun
usa
kazu
normal
hsbt
akr
ngoto
charliesome
mrkn
shugo
yugui
ktsj
kosaki
eban