hijack SIGCHLD handler for internal use

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

1 files changed, 37 insertions, 11 deletions

diff --git a/mjit.c b/mjit.c
index 07e9328aff..10aa15fb7f 100644
--- a/mjit.c
+++ b/mjit.c
@@ -80,6 +80,7 @@
 #include "constant.h"
 #include "id_table.h"
 #include "ruby_assert.h"
+#include "ruby/thread.h"
 #include "ruby/util.h"
 #include "ruby/version.h"
 
@@ -118,6 +119,10 @@ extern void rb_native_cond_wait(rb_nativethread_cond_t *cond, rb_nativethread_lo
 
 extern int rb_thread_create_mjit_thread(void (*child_hook)(void), void (*worker_func)(void));
 
+/* process.c */
+pid_t ruby_waitpid_locked(rb_vm_t *, rb_pid_t, int *status, int options,
+                          rb_nativethread_cond_t *cond);
+
 #define RB_CONDATTR_CLOCK_MONOTONIC 1
 
 #ifdef _WIN32
@@ -401,22 +406,40 @@ start_process(const char *path, char *const *argv)
 static int
 exec_process(const char *path, char *const argv[])
 {
-    int stat, exit_code;
+    int stat, exit_code = -2;
     pid_t pid;
+    rb_vm_t *vm = RUBY_SIGCHLD ? GET_VM() : 0;
+    rb_nativethread_cond_t cond;
 
-    pid = start_process(path, argv);
-    if (pid <= 0)
-        return -2;
+    if (vm) {
+        rb_native_cond_initialize(&cond);
+        rb_native_mutex_lock(&vm->waitpid_lock);
+    }
 
-    for (;;) {
-        waitpid(pid, &stat, 0);
-        if (WIFEXITED(stat)) {
-            exit_code = WEXITSTATUS(stat);
-            break;
-        } else if (WIFSIGNALED(stat)) {
-            exit_code = -1;
+    pid = start_process(path, argv);
+    for (;pid > 0;) {
+        pid_t r = vm ? ruby_waitpid_locked(vm, pid, &stat, 0, &cond)
+                     : waitpid(pid, &stat, 0);
+        if (r == -1) {
+            if (errno == EINTR) continue;
+            fprintf(stderr, "[%d] waitpid(%d): %s\n",
+                    getpid(), pid, strerror(errno));
             break;
         }
+        else if (r == pid) {
+            if (WIFEXITED(stat)) {
+                exit_code = WEXITSTATUS(stat);
+                break;
+            } else if (WIFSIGNALED(stat)) {
+                exit_code = -1;
+                break;
+            }
+        }
+    }
+
+    if (vm) {
+        rb_native_mutex_unlock(&vm->waitpid_lock);
+        rb_native_cond_destroy(&cond);
     }
     return exit_code;
 }
@@ -1491,12 +1514,15 @@ mjit_init(struct mjit_options *opts)
 static void
 stop_worker(void)
 {
+    rb_execution_context_t *ec = GET_EC();
+
     stop_worker_p = TRUE;
     while (!worker_stopped) {
         verbose(3, "Sending cancel signal to worker");
         CRITICAL_SECTION_START(3, "in stop_worker");
         rb_native_cond_broadcast(&mjit_worker_wakeup);
         CRITICAL_SECTION_FINISH(3, "in stop_worker");
+        RUBY_VM_CHECK_INTS(ec);
     }
 }