* process.c (rb_f_spawn): Be more specific regarding "other
values" by having "non-zero positive integers" Add nil, the
default value, as a possible value and what it means.
Try to use more consistent language.
[Fix GH-1008]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51819 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* process.c (rb_waitpid): do not release GVL when WNOHANG is set,
it should return immediately without blocking.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51793 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* process.c (rb_execarg_parent_start1): raise with the target path
name when open() failed.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51787 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* process.c (rb_exec_fail): raise with the target directory name
when chdir() failed. pointed out by sorah.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51786 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Shutting down the timer thread now always closes pipes to free FDs.
In fact, we close the write ends of the pipes is done in the main
RubyVM to signal the timer thread shutdown.
To effectively close pipes, we implement userspace locks via
atomics to force the pipe closing thread to wait on any signal
handlers which may be waking up.
While we're at it, improve robustness during resource exhaustion and
allow it to limp along non-fatally if restarting a timer thread
fails.
This reverts r51268
Note: this change is tested with VM_CHECK_MODE 1 in vm_core.h
* process.c (close_unless_reserved): add extra check
(dup2_with_divert): remove
(redirect_dup2): use dup2 without divert
(before_exec_non_async_signal_safe): adjust call + comment
(rb_f_exec): stop timer thread for all OSes
(rb_exec_without_timer_thread): remove
* eval.c (ruby_cleanup): adjust call
* thread.c (rb_thread_stop_timer_thread): always close pipes
* thread_pthread.c (struct timer_thread_pipe): add writing field,
mark owner_process volatile for signal handlers
(rb_thread_wakeup_timer_thread_fd): check valid FD
(rb_thread_wakeup_timer_thread): set writing flag to prevent close
(rb_thread_wakeup_timer_thread_low): ditto
(CLOSE_INVALIDATE): new macro
(close_invalidate): new function
(close_communication_pipe): removed
(setup_communication_pipe_internal): make errors non-fatal
(setup_communication_pipe): ditto
(thread_timer): close reading ends inside timer thread
(rb_thread_create_timer_thread): make errors non-fatal
(native_stop_timer_thread): close write ends only, always,
wait for signal handlers to finish
(rb_divert_reserved_fd): remove
* thread_win32.c (native_stop_timer_thread): adjust (untested)
(rb_divert_reserved_fd): remove
* vm_core.h: adjust prototype
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51576 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
with one of the timer thread FDs, the internal FD is diverted.
[Bug #11336] [ruby-core:69886] [Bug #11350] [ruby-core:69961]
* process.c (dup2_with_divert): new function for the above purpose.
* thread_pthread.c (rb_divert_reserved_fd): new function for
diverting reserved FD. If the given FD is the same as one of the
reserved FDs, the reserved FD number is internally changed.
It returns -1 when error. Otherwise, returns 0. It also returns
0 if there is no need to change reserved FD number.
* thread_win32.c (rb_divert_reserved_fd): always returns 0 because
of no reserved FDs.
* internal.h (rb_divert_reserved_fd): prototype declaration.
It is Ruby internal use only.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51268 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
reserved FD. It should be closed in the exec system call due to the
O_CLOEXEC or FD_CLOEXEC flag. [Bug #11353] [ruby-core:69977]
* process.c (close_unless_reserved): new function to close FD unless
it is reserved for internal communication.
* thread_pthread.c (rb_reserved_fd_p): should check owner_process pid
to avoid false positive in forked child process.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51255 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
used on every OS, not only on Mac OS X or Haiku, to prevent
timer thread communications using file descriptor 3 after
setting redirection of the fd 3 before calling exec.
[Bug #11336] [ruby-core:69886]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51209 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
to generate envp_str anytime on Solaris 10 (or earlier version
of Solaris) to avoid calling execv() which is async-signal unsafe
on Solaris 10. [Bug #11265] [ruby-dev:49089]
* process.c (exec_with_sh, proc_exec_cmd): On Solaris 10,
because ALWAYS_NEED_ENVP is 1 and envp_str is always generated,
execv() in exec_with_sh() and proc_exec_cmd() are never called.
To guarantee this, execv() is replaced by a macro to print
out error message on Solaris 10.
* process.c (proc_exec_sh): Because proc_exec_sh() may be called
by rb_proc_exec() with envp_str = Qfalse, execl() is replaced
by a macro that calls execle() with "extern char **environ"
traditional global variable on Solaris 10.
TODO: This may be unsafe and sholud be changed
in the future.
Although rb_proc_exec() is not used from inside current version
of ruby, it may be called by third-party extensions.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@50977 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* process.c (rb_spawn_process): do not discard global escape
status. [ruby-core:69304] [Bug #11166]
* process.c (rb_execarg_spawn): extract the start procedure in a
parent process with ensuring the end procedure.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@50600 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* process.c (initgroups): not declared on cygwin if _POSIX_SOURCE
or _XOPEN_SOURCE are defined.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@50304 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
process blocking when opening a named pipe.
(open_func): New function.
(rb_execarg_parent_start1): Extracted from rb_execarg_parent_start and
use rb_thread_call_without_gvl2 to release GVL when opening a file.
(rb_execarg_parent_start): Invoke rb_execarg_parent_start1 via
rb_protect and invoke rb_execarg_parent_end when error.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@50198 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* process.c: "spawn" opens files in the parent process.
(check_exec_redirect): Add an placeholder for fd in parameters
for fd_open.
(check_exec_fds_1): Delete fd_open condition.
(check_exec_fds): Don't call check_exec_fds_1 with fd_open.
(rb_execarg_parent_start): Open files specified as "spawn" options
and add "dup2" options.
(rb_execarg_parent_end): New function to close opened fds.
(run_exec_open): Removed.
(rb_execarg_run_options): Don't call run_exec_open.
(rb_spawn_internal): Call rb_execarg_parent_end.
* io.c (pipe_open): Call rb_execarg_parent_end.
* ext/pty/pty.c (establishShell): Call rb_execarg_parent_end.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@50194 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* hash.c (env_aset): state that ENV keys must be strings.
* process.c (rb_f_spawn): ditto. [ruby-core:68146] [Bug #10859]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@49635 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
It seems waitpid() is universaly available on POSIX platforms.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@49037 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* process.c (rb_fork_async_signal_safe): remove initialized but
never assigned local variables.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48769 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
use 0 for rb_data_type_t::reserved instead of NULL, since its type
may be changed in the future and possibly not a pointer type.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48662 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* process.c (check_exec_redirect_fd, check_exec_redirect),
(rb_execarg_addopt): get rid of inadvertent ID pindown.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48552 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Note: Some of the fixes are for newlib in general but not NaCl-specific.
* include/ruby/intern.h (rb_fd_select): declare struct timeval, or the
struct gets local to the function in C99.
* file.c (#include): add nacl/stat.h for PNaCl.
(utimes): added a declaration for PNaCl.
(stat_atimespec): stat::st_atimensec is long long but
timespec::tv_nsec is long in PNaCl.
(stat_mtimespec, stat_ctimespec): ditto.
(rb_group_member): disable getgroups unless HAVE_GETGROUPS.
(eaccess): unify the fallback to generic defined(USE_GETEUID).
* io.c: include sys/time.h for struct timeval.
(rb_close_before_exec): nothing we can do if F_GETFD is not
available.
(ioctl): pnacl newlib actually doesn't have ioctl.
* process.c (maxgroups): it is used iff
defined(_SC_NGROUPS_MAX) || defined(NGROUPS_MAX) but not
defined(HAVE_GETGROUPS) || defined(HAVE_SETGROUPS).
(obj2gid): fail unless the object is a Fixnum if getgrnam is not
available.
(disable_child_handler_fork_child): sigaction is not available in
PNaCl newlib.
* configure.in (warnflags, strict_warnflags): avoid -ansi for strlcpy.
(rb_cv_gcc_atomic_builtins): also check
__atomic_or_etch because it is used in ruby_atomic.h.
(rb_cv_gcc_sync_builtins): ditto.
(HAVE_GETGRNAM): added.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47882 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* process.c (free_exec_arg): remove
(memsize_exec_arg): ptr is never NULL
(exec_arg_data_type): use RUBY_TYPED_DEFAULT_FREE
* variable.c (autoload_i_free): remove
(autoload_data_i_type): use RUBY_TYPED_DEFAULT_FREE
(autoload_memsize): ptr is never NULL
* vm_backtrace.c (location_free): remove
(location_mark): ptr is never NULL
(location_data_type): use RUBY_TYPED_DEFAULT_FREE
(backtrace_mark): ditto
(backtrace_free): ditto
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47572 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* process.c (Init_process): subclass Thread as Process::Waiter
(rb_detach_process): use Process::Waiter instead of singleton class
* test/ruby/test_process.rb (test_process_detach): new test
* inits.c (rb_call_inits): call Init_Thread before Init_process to
ensure Process::Waiter may be a subclass of Thread
Thanks to headius for reporting [Bug #10231]
Thanks to nobu for review of my initial patch.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47561 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* process.c (retry_fork_async_signal_safe): revert r47439 but
disable use of vfork on OS X instead, as it cause hang-up at
test_process.rb:test_deadlock_by_signal_at_forking with parallel
test.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47441 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* process.c (retry_fork_async_signal_safe): vfork may not work
with pthread always even if fork works with it.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47439 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* process.c (rb_exec_without_timer_thread): add missing semicolon
to fix compilation error on OS X and Haiku.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47424 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* process.c (getresuid): Defined for AIX.
(getresgid): Ditto
AIX don't have getresuid/getresgid but getuidx/getgidx.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47397 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
vfork() is still faster than fork() especially when the parent
process uses big memory.
ruby -rbenchmark -e 'a = "a" * 1_000_000_000; puts Benchmark.measure { system("true") }'
fork: 0.000000 0.010000 0.010000 ( 0.014968)
vfork: 0.000000 0.000000 0.000000 ( 0.000912)
on Debian sid.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47363 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
retry_fork respect to rb_fork_async_signal_safe.
(retry_fork_ruby): Specialized version of retry_fork respect to
rb_fork_ruby.
(rb_fork_ruby): Removed.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47357 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
* process.c (open): use UTF-8 version function to support
non-ascii path properly. [ruby-core:63185] [Bug #9946]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@46441 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* process.c (run_exec_open): use a local variable to get rid of
RARRAY_AREF() is evaluated multiple times by RSTRING_PTR().
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@46439 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* process.c (check_exec_redirect): use RB_TYPE_P instead of single
comparison of TYPE().
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@46435 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
(OBJ2GID1): Defined even if getgrnam_r is not usable.
This fixes compilation error on Android.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45541 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* process.c (obj2uid, obj2gid): now getpwnam_r() and getgrnam_r()
may need larger buffers than sysconf values, so retry with
expanding the buffer when ERANGE is returned.
[ruby-core:61325] [Bug #9600]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45290 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* process.c (obj2uid, obj2gid): need capacity as buffer size, not
length.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45289 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* process.c (GETPW_R_SIZE_INIT, GETGR_R_SIZE_INIT): sysconf values
are not maximum sizes, but initial sizes.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45288 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* process.c (OBJ2UID1, OBJ2GID1): separate from OBJ2UID and
OBJ2GID respectively, need given buffers.
* process.c (OBJ2UID, OBJ2GID): no longer need PREPARE_GETPWNAM
and PREPARE_GETGRNAM.
* process.c (obj2uid, obj2gid): use tmp buffer instead of alloca
to get rid of potential stack overflow.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45287 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
0001-process.c-avoid-EINTR-from-Process.spawn.patch written by
Eric Wong in [Bug #8770].
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@44727 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* process.c (send_child_error): retry write on EINTR to fix
occasional Errno::EINTR from Process.spawn.
* process.c (recv_child_error): retry read on EINTR to fix
occasional Errno::EINTR from Process.spawn.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@44706 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
signal is sent to the parent process while Ruby is forking in IO.popen.
Patch by Scott Francis. Closes GH-513.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@44687 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* process.c (rb_execarg_fixup): use RHASH_TBL_RAW and insert write
barriers where appropriate
* vm.c (kwmerge_i): use RHASH_TBL_RAW
* vm.c (HASH_ASET): use rb_hash_aset instead of calling directly into
st_insert
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@44052 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* include/ruby/ruby.h (RB_BLOCK_CALL_FUNC_ARGLIST): for declaration
argument list of rb_block_call_func.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@43910 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
nobu
kosaki
normal
ngoto
hsbt
akr
yugui
ktsj
charliesome
ko1
naruse