* signal.c (signals): pack signal names instead of references.
* signal.c (signm2signo): also reject too long signal name.
[ruby-core:87767] [Bug #14893]
reapply r63841 and r63842, which are unrelated to r63758 but had been
wrongly reverted by r63852.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63854 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
The change is unstable on Windows. Please re-commit it when it correctly
supports Windows.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63852 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* signal.c (signals): pack signal names instead of references.
* signal.c (signm2signo): also reject too long signal name.
[ruby-core:87767] [Bug #14893]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63841 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Some systems lack SIGCHLD or have incomplete SIGCHLD
implementations. So enable polling mode for them.
[ruby-core:87705] [Bug #14867]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63795 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
sig_atomic_t may not be sufficient for multi-threaded
applications if the sighandler runs on a different CPU
than timer thread.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63793 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
[Bug #14871]
Is neon down? I missed ruby-core notifications on this.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63762 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Use a global SIGCHLD handler to guard all callers of rb_waitpid.
To work safely with multi-threaded programs, we introduce a
VM-wide waitpid_lock to be acquired BEFORE fork/vfork spawns the
process. This is to be combined with the new ruby_waitpid_locked
function used by mjit.c in a non-Ruby thread.
Ruby-level SIGCHLD handlers registered with Signal.trap(:CHLD)
continues to work as before and there should be no regressions
in any existing use cases.
Splitting the wait queues for PID > 0 and groups (PID <= 0)
ensures we favor PID > 0 callers.
The disabling of SIGCHLD in rb_f_system is longer necessary,
as we use deferred signal handling and no longer make ANY
blocking waitpid syscalls in other threads which could "beat"
the waitpid call made by rb_f_system.
We prevent SIGCHLD from firing in normal Ruby Threads and only
enable it in the timer-thread, to prevent spurious wakeups
from in test/-ext-/gvl/test_last_thread.rb with MJIT enabled.
I've tried to guard as much of the code for RUBY_SIGCHLD==0
using C "if" statements rather than CPP "#if" so to reduce
the likelyhood of portability problems as the compiler will
see more code.
We also work to suppress false-positives from
Process.wait(-1, Process::WNOHANG) to quiets warnings from
spec/ruby/core/process/wait2_spec.rb with MJIT enabled.
Lastly, we must implement rb_grantpt for ext/pty. We need a
MJIT-compatible way of supporting grantpt(3) which may spawn
the `pt_chown' binary and call waitpid(2) on it.
[ruby-core:87605] [Ruby trunk Bug#14867]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63758 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Instead of allocating and registering the altstack in different
places, do it together to reduce code and improve readability.
When thread cache is enabled, storing altstack in rb_thread_t
is wasteful and we may reuse altstack in the same pthread.
This also lets us clearly allow use of xmalloc to allow GC to
recover from ENOMEM.
[ruby-core:85621] [Feature #14487]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63213 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* vm_insnhelper.c (ec_stack_overflow): renamed from threadptr_stack_overflow
and also rb_ec_stack_overflow is from rb_threadptr_stack_overflow
because they accept `ec` instead of `th`.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60464 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
to represent execution context [Feature #14038]
* vm_core.h (rb_thread_t): rb_thread_t::ec is now a pointer.
There are many code using `th` to represent execution context
(such as cfp, VM stack and so on). To access `ec`, they need to
use `th->ec->...` (adding one indirection) so that we need to
replace them by passing `ec` instead of `th`.
* vm_core.h (GET_EC()): introduced to access current ec. Also
remove `ruby_current_thread` global variable.
* cont.c (rb_context_t): introduce rb_context_t::thread_ptr instead of
rb_context_t::thread_value.
* cont.c (ec_set_vm_stack): added to update vm_stack explicitly.
* cont.c (ec_switch): added to switch ec explicitly.
* cont.c (rb_fiber_close): added to terminate fibers explicitly.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60440 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
* signal.c (check_stack_overflow): raise fatal when the last tag
is in danger zone.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@59634 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* vm_core.h (rb_vm_struct): separate trap_list cmd and safe to
each arrays, to shrink the size.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@59434 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* vm_core.h (rb_thread_t): move several fields which are copied at cont.c
to rb_execution_context_t.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@59177 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Return value of EXEC_TAG() is saved by "int state".
Instead of "int", use "enum ruby_tag_type". First EXEC_TAG()
value should be 0, so that define TAG_NONE (= 0) and use it.
Some code used "status" instead of "state". To make them clear,
rename them to state.
We can change variable name from "state" to "tag_state", but this
ticket doesn't contain it.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@59155 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* thread.c (rbuy_kill): removed. This function is used
with SIGSEGV, SIGBUS, SIGKILL, SIGILL, SIGFPE and SIGSTOP
and these signals are affect immediately. So that `kill(2)'
is enough for them.
* signal.c (rb_f_kill): ditto.
* vm_core.h (rb_thread_t::interrupt_cond): removed because
only `ruby_kill()' uses this field.
* test/ruby/test_signal.rb: Without this patch sending SIGSTOP to own
process wait another interrupt even if another process sends SIGCONT.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@59066 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* signal.c: disable handling signals to dump core, if installing
unreserved signals failed, not valgrind to hang on Mac OS.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@59052 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* vm_insnhelper.c (rb_threadptr_stack_overflow): move from
thread.c and integrate with vm_stackoverflow.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@58379 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* signal.c (check_stack_overflow): drop the last tag when it is
close to the fault page, not same as sp page.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@58363 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* signal.c (raise_stack_overflow): unblock the received signal, to
receive the same signal again. [ruby-core:79285] [Bug #13164]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@58353 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* signal.c (check_stack_overflow): [EXPERIMENTAL] consider the
segfault a stack overflow when the fault address is between SP
and BP.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56907 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* signal.c (sig_signame): [DOC] Add documentation in the case of
return nil. [Fix GH-1449]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56280 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* vm_eval.c (rb_eval_cmd, rb_catch_obj): use TH_JUMP_TAG with the
same rb_thread_t used for TH_PUSH_TAG, instead of JUMP_TAG with
the current thread global variable.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54914 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
This allows us to reuse string objects used in symbols as well
as any string representations of signal names in source code.
* signal.c (sig_list): use fstring for hash key
* test/ruby/test_signal.rb (test_signal_list_dedupe_keys): added
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@53657 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* signal.c (rb_f_kill): should immediately deliver reserved
signals SIGILL and SIGFPE, ont only SIGSEGV and SIGBUS.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@53656 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* beos: Drop support for BeOS now that Haiku is stable.
[Fix GH-1112]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52731 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
seems that rb_bug_errno() is called in sigpipe() intentionally.
https://gist.github.com/sorah/831169
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51390 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
strncmp is unnecessary since the switch/case statement already
checks length of the string; so use memcmp.
This makes for a small reduction in binary size on 32-bit x86:
text data bss dec hex filename
2847473 12360 30632 2890465 2c1ae1 ruby.before
2847313 12328 30632 2890273 2c1a21 ruby.after
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51287 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* signal.c (sig_signame): return nil if the argument is a valid
signal number.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@49801 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* signal.c (sigbus): seems that Funtoo Linux also delivers SIGBUS
at stack overflow.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@49463 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* signal.c (check_stack_overflow): clear the received reserved
signal before raising stack overflow but not aborting.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48910 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
We only install system-level signal handlers during init (before any
threads or user code is running), so there's no need to repeatedly
enable/disable the interrupts at that time. We also do not change
system-level sig handlers once user-level code is running.
This saves about 20 syscalls at startup and makes the executable
smaller (numbers from 32-bit x86):
text data bss dec hex filename
before: 2815726 12100 30552 2858378 2b9d8a ruby
after: 2815022 12100 30552 2857674 2b9aca ruby
* signal.c (install_sighandler): remove rb_disable_interrupt and
rb_enable_interrupt calls
(init_sigchld): ditto
(Init_signal): disable and enable interrupt once around all
install_sighandler and init_sigchld to reduce syscalls at start
[Feature #9345] [ruby-core:59480]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48178 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* signal.c (rb_sigaltstack_size): double default size to get rid
of heap corruption by alternate stack overflow in SEGV handler.
typically happened at fprintf() in control_frame_dump().
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47984 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* signal.c (check_reserved_signal): abort if reserved signal
received in reserved signal handlers.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47977 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* signal.c (install_signalhandler, init_sigchld): allow failure because it
always fails with ENOSYS on NaCl.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47873 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* signal.c (rb_f_kill): get rid of deadlock as unhandled and
discarded signals do not make interrupt_cond signaled.
based on the patch by Kazuki Tsujimoto at [ruby-dev:48606].
[Bug #9820]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47832 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* signal.c (rb_f_kill): should not ignore signal unless the
default handler is registered. [ruby-dev:48592] [Bug #9820]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47818 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* signal.c (trap): SIGEXIT is not a system signal and is dealt
with internally, so it should not try to register the system
signal handler by sigaction.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47669 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* signal.c (ruby_signal): return SIG_ERR as well as signal(2).
* signal.c (trap): raise SystemCallError for all failures when
called as a method.
* signal.c (Init_signal): fail by [BUG] only if initialization is
failed.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47660 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* signal.c (ruby_signal): although "EINVAL from sigaction(2) is
not a bug", but even it is a failure. pointed at toRuby/guRuby
in RubyHiroba.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47657 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* process.c (rlimit_resource_type, rlimit_resource_value): get rid
of function calls in RSTRING_PTR(), as it evaluates the argument
twice.
* re.c (match_backref_number): ditto.
* signal.c (esignal_init, rb_f_kill, trap_signm): ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47006 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
rb_vm_t::trap_list_cmds (an array) and
rb_vm_t::trap_list_safes[RUBY_NSIG]
(separate to two different array).
This modification reduce root objects.
* signal.c: ditto.
* vm.c (rb_vm_mark): remove marking code for rb_vm_t::trap_list.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@46834 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* signal.c (check_stack_overflow): drop the last tag too close to
the fault page, to get rid of stack overflow deadlock.
[Bug #9971]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@46499 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* signal.c (ruby_signal): should return either `old.sa_sigaction`
or `old.sa_handler`, depending on whether `SA_SIGINFO` is set in
`old.sa_flags`, because they may not be a union.
[ruby-core:62836] [Bug #9878]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@46233 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* signal.c (check_stack_overflow): revert part of r46100, the
previous condition was correct, and fix compilation error on
other architecture linux. [ruby-core:62746] [Bug #9862]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@46113 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* error.c (rb_bug_context): new function to report bug with
context.
* vm_dump.c (rb_vm_bugreport): accepts `ucontext_t` argument to
dump machine regisiters. based on [GH-584].
* signal.c (sigbus, sigsegv): dump machine regisiters if available.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@46106 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* signal.c (check_stack_overflow): fix condition to use ucontext
register, mcontext_t dereference, and its member names, on Mac
OS X.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@46100 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* signal.c (check_stack_overflow): use mcontext_t. the struct tag
name is not portable.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45997 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* signal.c (check_stack_overflow): raise SystemStackError if SP
register and fault address is in the same page or the next, on
x86 Mac OS X. [EXPERIMENTAL]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45995 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* signal.c (check_stack_overflow): check the next page too. SP in
ucontext is not decremented yet when `push` failed, so the fault
page can be the next.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45924 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
nobu
naruse
kazu
normal
shyouhei
ko1
hsbt
akr
usa
yugui
ktsj