aboutsummaryrefslogtreecommitdiffstats
path: root/erts/emulator/sys/unix/erl_child_setup.c
diff options
context:
space:
mode:
Diffstat (limited to 'erts/emulator/sys/unix/erl_child_setup.c')
-rw-r--r--erts/emulator/sys/unix/erl_child_setup.c427
1 files changed, 338 insertions, 89 deletions
diff --git a/erts/emulator/sys/unix/erl_child_setup.c b/erts/emulator/sys/unix/erl_child_setup.c
index a3c5c20641..71c7948bf0 100644
--- a/erts/emulator/sys/unix/erl_child_setup.c
+++ b/erts/emulator/sys/unix/erl_child_setup.c
@@ -1,7 +1,7 @@
/*
* %CopyrightBegin%
*
- * Copyright Ericsson AB 2002-2009. All Rights Reserved.
+ * Copyright Ericsson AB 2002-2015. All Rights Reserved.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
@@ -19,109 +19,195 @@
*/
/*
- * After a vfork() (or fork()) the child exec()s to this program which
- * sets up the child and exec()s to the user program (see spawn_start()
- * in sys.c and ticket OTP-4389).
+ * This program is started at erts startup and all fork's that
+ * have to be done are done in here. This is done for a couple
+ * of reasons:
+ * - Allow usage of fork without a memory explosion.
+ * -- we do not want to use vfork, as it blocks the VM
+ * until the execv is done, and if the program that
+ * is to be executed is on an NFS that is unavailable,
+ * the execv can block for a very long time.
+ * -- we cannot do fork inside the VM as that would temporarily
+ * duplicate the memory usage of the VM per parallel exec.
+ *
+ * Some implementation notes:
+ * - A single Unix Domain Socket is setup in between the VM and
+ * this program. Over that UDS the file descriptors that should
+ * be used to talk to the child program are sent.
+ * The actual command to execute, together with options and the
+ * environment, is sent over the pipe represented by the
+ * file descriptors mentioned above. We don't send the
+ * command over the UDS as that would increase the likely hood
+ * that it's buffer would be full.
+ *
+ * - Since it is this program that execv's, it has to take care of
+ * all the SIGCHLD signals that the child programs generate. The
+ * signals are received and the pid+exit reason is sent as data
+ * on the UDS to the VM. The VM is then able to map the pid to the
+ * port of the child program that just exited and deliver the status
+ * code if requested.
*/
#ifdef HAVE_CONFIG_H
# include "config.h"
#endif
-#define NEED_CHILD_SETUP_DEFINES
-#include "sys.h"
-#include "erl_misc_utils.h"
+#include <stdlib.h>
+#include <stdio.h>
+#include <sys/wait.h>
-#ifdef SIG_SIGSET /* Old SysV */
-void sys_sigrelease(int sig)
-{
- sigrelse(sig);
-}
-#else /* !SIG_SIGSET */
-#ifdef SIG_SIGNAL /* Old BSD */
-sys_sigrelease(int sig)
+#include "erl_driver.h"
+#include "sys_uds.h"
+
+#define SET_CLOEXEC(fd) fcntl(fd, F_SETFD, fcntl(fd, F_GETFD) | FD_CLOEXEC)
+
+#if defined(__ANDROID__)
+#define SHELL "/system/bin/sh"
+#else
+#define SHELL "/bin/sh"
+#endif /* __ANDROID__ */
+
+//#define HARD_DEBUG
+#ifdef HARD_DEBUG
+#define DEBUG_PRINT(fmt, ...) fprintf(stderr, fmt "\r\n", ##__VA_ARGS__)
+#else
+#define DEBUG_PRINT(fmt, ...)
+#endif
+
+#define ABORT(fmt, ...) do { \
+ fprintf(stderr, "erl_child_setup: " fmt "\r\n", ##__VA_ARGS__); \
+ abort(); \
+ } while(0)
+
+#undef ASSERT
+#ifdef DEBUG
+#define ASSERT(cnd) do { if (!(cnd)) { ABORT("assertion %s failed", #cnd); } } while(0)
+#else
+#define ASSERT(cnd)
+#endif
+
+void sys_sigblock(int sig)
{
- sigsetmask(sigblock(0) & ~sigmask(sig));
+ sigset_t mask;
+
+ sigemptyset(&mask);
+ sigaddset(&mask, sig);
+ sigprocmask(SIG_BLOCK, &mask, (sigset_t *)NULL);
}
-#else /* !SIG_SIGNAL */ /* The True Way - POSIX!:-) */
+
void sys_sigrelease(int sig)
{
sigset_t mask;
-
sigemptyset(&mask);
sigaddset(&mask, sig);
sigprocmask(SIG_UNBLOCK, &mask, (sigset_t *)NULL);
}
-#endif /* !SIG_SIGNAL */
-#endif /* !SIG_SIGSET */
-#if defined(__ANDROID__)
-static int system_properties_fd(void);
-#endif /* __ANDROID__ */
+static int max_files = -1;
+static int sigchld_pipe[2];
-#if defined(__ANDROID__)
-#define SHELL "/system/bin/sh"
-#else
-#define SHELL "/bin/sh"
-#endif /* __ANDROID__ */
+static int
+start_new_child(int pipes[])
+{
+ int size, res, i;
+ char *buff, *o_buff;
+ char *cmd, *wd, **new_environ, **args = NULL, cbuff[1];
-int
-main(int argc, char *argv[])
-{
- int i, from, to;
- int erts_spawn_executable = 0;
-
- /* OBSERVE!
- * Keep child setup after fork() (implemented in sys.c) up to date
- * if changes are made here.
- */
-
- if (argc != CS_ARGV_NO_OF_ARGS) {
- if (argc < CS_ARGV_NO_OF_ARGS) {
- return 1;
- } else {
- erts_spawn_executable = 1;
- }
+ Sint cnt, flags;
+
+ /* only child executes here */
+
+ if ((res = read(pipes[0], (char*)&size, sizeof(size))) < 0) {
+ ABORT("Failed to read size from %d (%d)", pipes[0], errno);
}
- if (strcmp("false", argv[CS_ARGV_UNBIND_IX]) != 0)
- if (erts_unbind_from_cpu_str(argv[CS_ARGV_UNBIND_IX]) != 0)
- return 1;
-
- for (i = 0; i < CS_ARGV_NO_OF_DUP2_OPS; i++) {
- if (argv[CS_ARGV_DUP2_OP_IX(i)][0] == '-'
- && argv[CS_ARGV_DUP2_OP_IX(i)][1] == '\0')
- break;
- if (sscanf(argv[CS_ARGV_DUP2_OP_IX(i)], "%d:%d", &from, &to) != 2)
- return 1;
- if (dup2(from, to) < 0)
- return 1;
+ buff = malloc(size);
+
+ DEBUG_PRINT("size = %d", size);
+
+ if ((res = read(pipes[0], buff, size)) != size) {
+ ABORT("Failed to read %d bytes from %d (%d,%d)",
+ size, pipes[0], res, errno);
}
- if (sscanf(argv[CS_ARGV_FD_CR_IX], "%d:%d", &from, &to) != 2)
- return 1;
+ o_buff = buff;
-#if defined(HAVE_CLOSEFROM)
- closefrom(from);
-#elif defined(__ANDROID__)
- if (from <= to) {
- int spfd = system_properties_fd();
- for (i = from; i <= to; i++) {
- if (i != spfd) {
- (void) close(i);
- }
- }
+ flags = *(int*)buff;
+ buff += sizeof(int);
+
+ DEBUG_PRINT("flags = %d", flags);
+
+ cmd = buff;
+ buff += strlen(buff) + 1;
+ if (*buff == '\0') {
+ wd = NULL;
+ } else {
+ wd = buff;
+ buff += strlen(buff) + 1;
+ }
+ buff++;
+
+ DEBUG_PRINT("wd = %s", wd);
+
+ cnt = *(int*)buff;
+ buff += sizeof(int);
+ new_environ = malloc(sizeof(char*)*(cnt + 1));
+
+ for (i = 0; i < cnt; i++, buff++) {
+ new_environ[i] = buff;
+ while(*buff != '\0') buff++;
+ }
+ new_environ[cnt] = NULL;
+
+ if (o_buff + size != buff) {
+ /* This is a spawn executable call */
+ cnt = *(int*)buff;
+ buff += sizeof(int);
+ args = malloc(sizeof(char*)*(cnt + 1));
+ for (i = 0; i < cnt; i++, buff++) {
+ args[i] = buff;
+ while(*buff != '\0') buff++;
+ }
+ args[cnt] = NULL;
+ }
+
+ if (o_buff + size != buff) {
+ ABORT("Buff error: %p, %p:%p", o_buff, o_buff+size, buff);
+ }
+
+ if (read(pipes[0], cbuff, 1) < 1)
+ goto child_error;
+
+ DEBUG_PRINT("Do that forking business: '%s'\n",cmd);
+
+ /* When the dup2'ing below is done, only
+ fd's 0, 1, 2 and maybe 3, 4 should survive the
+ exec. All other fds (i.e. the unix domain sockets
+ and stray pipe ends) should have CLOEXEC set on them
+ so they will be closed when the exec happens */
+ if (flags & FORKER_FLAG_USE_STDIO) {
+ /* stdin for process */
+ if (flags & FORKER_FLAG_DO_WRITE &&
+ dup2(pipes[0], 0) < 0)
+ goto child_error;
+ /* stdout for process */
+ if (flags & FORKER_FLAG_DO_READ &&
+ dup2(pipes[1], 1) < 0)
+ goto child_error;
}
-#else /* !__ANDROID__ */
- for (i = from; i <= to; i++) {
- (void) close(i);
+ else { /* XXX will fail if pipes[0] == 4 (unlikely..) */
+ if (flags & FORKER_FLAG_DO_READ && dup2(pipes[1], 4) < 0)
+ goto child_error;
+ if (flags & FORKER_FLAG_DO_WRITE && dup2(pipes[0], 3) < 0)
+ goto child_error;
}
-#endif /* HAVE_CLOSEFROM */
- if (!(argv[CS_ARGV_WD_IX][0] == '.' && argv[CS_ARGV_WD_IX][1] == '\0')
- && chdir(argv[CS_ARGV_WD_IX]) < 0)
- return 1;
+ if (dup2(pipes[2], 2) < 0)
+ goto child_error;
+
+ if (wd && chdir(wd) < 0)
+ goto child_error;
#if defined(USE_SETPGRP_NOARGS) /* SysV */
(void) setpgrp();
@@ -131,34 +217,197 @@ main(int argc, char *argv[])
(void) setsid();
#endif
+ close(pipes[0]);
+ close(pipes[1]);
+ close(pipes[2]);
+
sys_sigrelease(SIGCHLD);
- sys_sigrelease(SIGINT);
- sys_sigrelease(SIGUSR1);
-
- if (erts_spawn_executable) {
- if (argv[CS_ARGV_NO_OF_ARGS + 1] == NULL) {
- execl(argv[CS_ARGV_NO_OF_ARGS],argv[CS_ARGV_NO_OF_ARGS],
- (char *) NULL);
- } else {
- execv(argv[CS_ARGV_NO_OF_ARGS],&(argv[CS_ARGV_NO_OF_ARGS + 1]));
- }
+
+ if (args) {
+ /* spawn_executable */
+ execve(cmd, args, new_environ);
} else {
- execl(SHELL, "sh", "-c", argv[CS_ARGV_CMD_IX], (char *) NULL);
+ execle(SHELL, "sh", "-c", cmd, (char *) NULL, new_environ);
}
- return 1;
+child_error:
+ _exit(1);
+}
+
+
+/*
+ * [OTP-3906]
+ * Solaris signal management gets confused when threads are used and a
+ * lot of child processes dies. The confusion results in that SIGCHLD
+ * signals aren't delivered to the emulator which in turn results in
+ * a lot of defunct processes in the system.
+ *
+ * The problem seems to appear when a signal is frequently
+ * blocked/unblocked at the same time as the signal is frequently
+ * propagated. The child waiter thread is a workaround for this problem.
+ * The SIGCHLD signal is always blocked (in all threads), and the child
+ * waiter thread fetches the signal by a call to sigwait(). See
+ * child_waiter().
+ *
+ * This should be a non-issue since the fork:ing was moved outside of
+ * the emulator into erl_child_setup. I'm leaving the comment here
+ * for posterity. */
+
+static void handle_sigchld(int sig) {
+ int buff[2], res;
+
+ sys_sigblock(SIGCHLD);
+
+ while ((buff[0] = waitpid((pid_t)(-1), buff+1, WNOHANG)) > 0) {
+ if ((res = write(sigchld_pipe[1], buff, sizeof(buff))) <= 0)
+ ABORT("Failed to write to sigchld_pipe (%d): %d (%d)", sigchld_pipe[1], res, errno);
+ DEBUG_PRINT("Reap child %d (%d)", buff[0], buff[1]);
+ }
+
+ sys_sigrelease(SIGCHLD);
}
#if defined(__ANDROID__)
static int system_properties_fd(void)
{
- int fd;
+ static int fd = -2;
char *env;
+ if (fd != -2) return fd;
env = getenv("ANDROID_PROPERTY_WORKSPACE");
if (!env) {
+ fd = -1;
return -1;
}
fd = atoi(env);
return fd;
}
#endif /* __ANDROID__ */
+
+int
+main(int argc, char *argv[])
+{
+ /* This fd should be open from beam */
+ int uds_fd = 3, max_fd = 3;
+#ifndef HAVE_CLOSEFROM
+ int i;
+#endif
+ struct sigaction sa;
+
+ if (argc < 1 || sscanf(argv[1],"%d",&max_files) != 1) {
+ ABORT("Invalid arguments to child_setup");
+ }
+
+/* We close all fds except the uds from beam.
+ All other fds from now on will have the
+ CLOEXEC flags set on them. This means that we
+ only have to close a very limited number of fds
+ after we fork before the exec. */
+#if defined(HAVE_CLOSEFROM)
+ closefrom(4);
+#else
+ for (i = 4; i < max_files; i++)
+#if defined(__ANDROID__)
+ if (i != system_properties_fd())
+#endif
+ (void) close(i);
+#endif
+
+ if (pipe(sigchld_pipe) < 0) {
+ ABORT("Failed to setup sigchld pipe (%d)", errno);
+ }
+
+ SET_CLOEXEC(sigchld_pipe[0]);
+ SET_CLOEXEC(sigchld_pipe[1]);
+
+ max_fd = max_fd < sigchld_pipe[0] ? sigchld_pipe[0] : max_fd;
+
+ sa.sa_handler = &handle_sigchld;
+ sigemptyset(&sa.sa_mask);
+ sa.sa_flags = SA_RESTART | SA_NOCLDSTOP;
+ if (sigaction(SIGCHLD, &sa, 0) == -1) {
+ perror(0);
+ exit(1);
+ }
+
+ SET_CLOEXEC(uds_fd);
+
+ DEBUG_PRINT("Starting forker %d", max_files);
+
+ while (1) {
+ fd_set read_fds;
+ int res;
+ FD_ZERO(&read_fds);
+ FD_SET(uds_fd, &read_fds);
+ FD_SET(sigchld_pipe[0], &read_fds);
+ DEBUG_PRINT("child_setup selecting on %d, %d (%d)",
+ uds_fd, sigchld_pipe[0], max_fd);
+ res = select(max_fd+1, &read_fds, NULL, NULL, NULL);
+
+ if (res < 0) {
+ if (errno == EINTR) continue;
+ ABORT("Select failed: %d (%d)",res, errno);
+ }
+
+ if (FD_ISSET(uds_fd, &read_fds)) {
+ int pipes[3], res, os_pid;
+ char buff[256];
+ errno = 0;
+ if ((res = sys_uds_read(uds_fd, buff, 1, pipes, 3, MSG_DONTWAIT)) < 0) {
+ if (errno == EINTR)
+ continue;
+ DEBUG_PRINT("erl_child_setup failed to read from uds: %d, %d", res, errno);
+ _exit(0);
+ }
+
+ if (res == 0) {
+ DEBUG_PRINT("uds was closed!");
+ _exit(0);
+ }
+ /* Since we use unix domain sockets and send the entire data in
+ one go we *should* get the entire payload at once. */
+ ASSERT(res == 1);
+ ASSERT(buff[0] == 'S');
+
+ sys_sigblock(SIGCHLD);
+
+ errno = 0;
+
+ os_pid = fork();
+ if (os_pid == 0)
+ start_new_child(pipes);
+
+ /* We write an ack here, but expect the reply on
+ the pipes[0] inside the fork */
+ res = sprintf(buff,"GO:%010d:%010d", os_pid, errno);
+ if (write(pipes[1], buff, res + 1))
+ ; /* remove gcc warning */
+
+ sys_sigrelease(SIGCHLD);
+ close(pipes[0]);
+ close(pipes[1]);
+ close(pipes[2]);
+ }
+
+ if (FD_ISSET(sigchld_pipe[0], &read_fds)) {
+ int ibuff[2];
+ char buff[256];
+ res = read(sigchld_pipe[0], ibuff, sizeof(ibuff));
+ if (res < 0) {
+ if (errno == EINTR)
+ continue;
+ ABORT("Failed to read from sigchld pipe: %d (%d)", res, errno);
+ }
+ res = snprintf(buff, 256, "SIGCHLD:%010d:%010d", ibuff[0], ibuff[1]);
+ DEBUG_PRINT("send %s to %d", buff, uds_fd);
+ if (write(uds_fd, buff, res + 1) < 0) {
+ if (errno == EINTR)
+ continue;
+ /* The uds was close, which most likely means that the VM
+ has exited. This will be detected when we try to read
+ from the uds_fd. */
+ DEBUG_PRINT("Failed to write to uds: %d (%d)", uds_fd, errno);
+ }
+ }
+ }
+ return 1;
+}
generated by cgit v1.2.3 (git 2.25.1) at 2024-07-01 22:18:33 +0000