exit.c 45.7 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10
/*
 *  linux/kernel/exit.c
 *
 *  Copyright (C) 1991, 1992  Linus Torvalds
 */

#include <linux/mm.h>
#include <linux/slab.h>
#include <linux/interrupt.h>
#include <linux/module.h>
11
#include <linux/capability.h>
L
Linus Torvalds 已提交
12 13 14
#include <linux/completion.h>
#include <linux/personality.h>
#include <linux/tty.h>
15
#include <linux/mnt_namespace.h>
16
#include <linux/iocontext.h>
L
Linus Torvalds 已提交
17 18 19 20
#include <linux/key.h>
#include <linux/security.h>
#include <linux/cpu.h>
#include <linux/acct.h>
21
#include <linux/tsacct_kern.h>
L
Linus Torvalds 已提交
22
#include <linux/file.h>
A
Al Viro 已提交
23
#include <linux/fdtable.h>
L
Linus Torvalds 已提交
24
#include <linux/binfmts.h>
S
Serge E. Hallyn 已提交
25
#include <linux/nsproxy.h>
26
#include <linux/pid_namespace.h>
L
Linus Torvalds 已提交
27 28 29 30
#include <linux/ptrace.h>
#include <linux/profile.h>
#include <linux/mount.h>
#include <linux/proc_fs.h>
31
#include <linux/kthread.h>
L
Linus Torvalds 已提交
32
#include <linux/mempolicy.h>
33
#include <linux/taskstats_kern.h>
34
#include <linux/delayacct.h>
35
#include <linux/freezer.h>
36
#include <linux/cgroup.h>
L
Linus Torvalds 已提交
37
#include <linux/syscalls.h>
38
#include <linux/signal.h>
39
#include <linux/posix-timers.h>
M
Matt Helsley 已提交
40
#include <linux/cn_proc.h>
41
#include <linux/mutex.h>
42
#include <linux/futex.h>
43
#include <linux/compat.h>
44
#include <linux/pipe_fs_i.h>
45
#include <linux/audit.h> /* for audit_free() */
A
Adrian Bunk 已提交
46
#include <linux/resource.h>
47
#include <linux/blkdev.h>
48
#include <linux/task_io_accounting_ops.h>
L
Linus Torvalds 已提交
49 50 51 52 53 54

#include <asm/uaccess.h>
#include <asm/unistd.h>
#include <asm/pgtable.h>
#include <asm/mmu_context.h>

A
Adrian Bunk 已提交
55 56
static void exit_mm(struct task_struct * tsk);

57 58 59 60 61
static inline int task_detached(struct task_struct *p)
{
	return p->exit_signal == -1;
}

L
Linus Torvalds 已提交
62 63 64 65 66 67 68
static void __unhash_process(struct task_struct *p)
{
	nr_threads--;
	detach_pid(p, PIDTYPE_PID);
	if (thread_group_leader(p)) {
		detach_pid(p, PIDTYPE_PGID);
		detach_pid(p, PIDTYPE_SID);
69

70
		list_del_rcu(&p->tasks);
71
		__get_cpu_var(process_counts)--;
L
Linus Torvalds 已提交
72
	}
O
Oleg Nesterov 已提交
73
	list_del_rcu(&p->thread_group);
R
Roland McGrath 已提交
74
	list_del_init(&p->sibling);
L
Linus Torvalds 已提交
75 76
}

77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98
/*
 * This function expects the tasklist_lock write-locked.
 */
static void __exit_signal(struct task_struct *tsk)
{
	struct signal_struct *sig = tsk->signal;
	struct sighand_struct *sighand;

	BUG_ON(!sig);
	BUG_ON(!atomic_read(&sig->count));

	sighand = rcu_dereference(tsk->sighand);
	spin_lock(&sighand->siglock);

	posix_cpu_timers_exit(tsk);
	if (atomic_dec_and_test(&sig->count))
		posix_cpu_timers_exit_group(tsk);
	else {
		/*
		 * If there is any task waiting for the group exit
		 * then notify it:
		 */
99
		if (sig->group_exit_task && atomic_read(&sig->count) == sig->notify_count)
100
			wake_up_process(sig->group_exit_task);
101

102 103 104 105 106 107 108 109 110 111 112 113 114 115
		if (tsk == sig->curr_target)
			sig->curr_target = next_thread(tsk);
		/*
		 * Accumulate here the counters for all threads but the
		 * group leader as they die, so they can be added into
		 * the process-wide totals when those are taken.
		 * The group leader stays around as a zombie as long
		 * as there are other threads.  When it gets reaped,
		 * the exit.c code will add its counts into these totals.
		 * We won't ever get here for the group leader, since it
		 * will have been the last reference on the signal_struct.
		 */
		sig->utime = cputime_add(sig->utime, tsk->utime);
		sig->stime = cputime_add(sig->stime, tsk->stime);
116
		sig->gtime = cputime_add(sig->gtime, tsk->gtime);
117 118 119 120
		sig->min_flt += tsk->min_flt;
		sig->maj_flt += tsk->maj_flt;
		sig->nvcsw += tsk->nvcsw;
		sig->nivcsw += tsk->nivcsw;
121 122
		sig->inblock += task_io_get_inblock(tsk);
		sig->oublock += task_io_get_oublock(tsk);
123
		sig->sum_sched_runtime += tsk->se.sum_exec_runtime;
124 125 126
		sig = NULL; /* Marker for below. */
	}

127 128
	__unhash_process(tsk);

129 130 131 132 133 134
	/*
	 * Do this under ->siglock, we can race with another thread
	 * doing sigqueue_free() if we have SIGQUEUE_PREALLOC signals.
	 */
	flush_sigqueue(&tsk->pending);

135
	tsk->signal = NULL;
136
	tsk->sighand = NULL;
137 138
	spin_unlock(&sighand->siglock);

139
	__cleanup_sighand(sighand);
140 141 142
	clear_tsk_thread_flag(tsk,TIF_SIGPENDING);
	if (sig) {
		flush_sigqueue(&sig->shared_pending);
143
		taskstats_tgid_free(sig);
144 145 146 147
		__cleanup_signal(sig);
	}
}

148 149 150 151 152
static void delayed_put_task_struct(struct rcu_head *rhp)
{
	put_task_struct(container_of(rhp, struct task_struct, rcu));
}

R
Roland McGrath 已提交
153 154 155 156 157 158 159 160 161 162 163 164
/*
 * Do final ptrace-related cleanup of a zombie being reaped.
 *
 * Called with write_lock(&tasklist_lock) held.
 */
static void ptrace_release_task(struct task_struct *p)
{
	BUG_ON(!list_empty(&p->ptraced));
	ptrace_unlink(p);
	BUG_ON(!list_empty(&p->ptrace_entry));
}

L
Linus Torvalds 已提交
165 166
void release_task(struct task_struct * p)
{
167
	struct task_struct *leader;
L
Linus Torvalds 已提交
168
	int zap_leader;
169
repeat:
L
Linus Torvalds 已提交
170
	atomic_dec(&p->user->processes);
171
	proc_flush_task(p);
L
Linus Torvalds 已提交
172
	write_lock_irq(&tasklist_lock);
R
Roland McGrath 已提交
173
	ptrace_release_task(p);
L
Linus Torvalds 已提交
174
	__exit_signal(p);
175

L
Linus Torvalds 已提交
176 177 178 179 180 181 182 183
	/*
	 * If we are the last non-leader member of the thread
	 * group, and the leader is zombie, then notify the
	 * group leader's parent process. (if it wants notification.)
	 */
	zap_leader = 0;
	leader = p->group_leader;
	if (leader != p && thread_group_empty(leader) && leader->exit_state == EXIT_ZOMBIE) {
184
		BUG_ON(task_detached(leader));
L
Linus Torvalds 已提交
185 186 187 188 189 190 191 192 193
		do_notify_parent(leader, leader->exit_signal);
		/*
		 * If we were the last child thread and the leader has
		 * exited already, and the leader's parent ignores SIGCHLD,
		 * then we are the one who should release the leader.
		 *
		 * do_notify_parent() will have marked it self-reaping in
		 * that case.
		 */
194
		zap_leader = task_detached(leader);
L
Linus Torvalds 已提交
195 196 197 198
	}

	write_unlock_irq(&tasklist_lock);
	release_thread(p);
199
	call_rcu(&p->rcu, delayed_put_task_struct);
L
Linus Torvalds 已提交
200 201 202 203 204 205 206 207 208 209

	p = leader;
	if (unlikely(zap_leader))
		goto repeat;
}

/*
 * This checks not only the pgrp, but falls back on the pid if no
 * satisfactory pgrp is found. I dunno - gdb doesn't work correctly
 * without this...
210 211
 *
 * The caller must hold rcu lock or the tasklist lock.
L
Linus Torvalds 已提交
212
 */
213
struct pid *session_of_pgrp(struct pid *pgrp)
L
Linus Torvalds 已提交
214 215
{
	struct task_struct *p;
216
	struct pid *sid = NULL;
217

218
	p = pid_task(pgrp, PIDTYPE_PGID);
219
	if (p == NULL)
220
		p = pid_task(pgrp, PIDTYPE_PID);
221
	if (p != NULL)
222
		sid = task_session(p);
223

L
Linus Torvalds 已提交
224 225 226 227 228 229 230 231 232 233 234
	return sid;
}

/*
 * Determine if a process group is "orphaned", according to the POSIX
 * definition in 2.2.2.52.  Orphaned process groups are not to be affected
 * by terminal-generated stop signals.  Newly orphaned process groups are
 * to receive a SIGHUP and a SIGCONT.
 *
 * "I ask you, have you ever known what it is to be an orphan?"
 */
235
static int will_become_orphaned_pgrp(struct pid *pgrp, struct task_struct *ignored_task)
L
Linus Torvalds 已提交
236 237 238
{
	struct task_struct *p;

239
	do_each_pid_task(pgrp, PIDTYPE_PGID, p) {
240 241 242
		if ((p == ignored_task) ||
		    (p->exit_state && thread_group_empty(p)) ||
		    is_global_init(p->real_parent))
L
Linus Torvalds 已提交
243
			continue;
244

245
		if (task_pgrp(p->real_parent) != pgrp &&
246 247
		    task_session(p->real_parent) == task_session(p))
			return 0;
248
	} while_each_pid_task(pgrp, PIDTYPE_PGID, p);
249 250

	return 1;
L
Linus Torvalds 已提交
251 252
}

253
int is_current_pgrp_orphaned(void)
L
Linus Torvalds 已提交
254 255 256 257
{
	int retval;

	read_lock(&tasklist_lock);
258
	retval = will_become_orphaned_pgrp(task_pgrp(current), NULL);
L
Linus Torvalds 已提交
259 260 261 262 263
	read_unlock(&tasklist_lock);

	return retval;
}

264
static int has_stopped_jobs(struct pid *pgrp)
L
Linus Torvalds 已提交
265 266 267 268
{
	int retval = 0;
	struct task_struct *p;

269
	do_each_pid_task(pgrp, PIDTYPE_PGID, p) {
M
Matthew Wilcox 已提交
270
		if (!task_is_stopped(p))
L
Linus Torvalds 已提交
271 272 273
			continue;
		retval = 1;
		break;
274
	} while_each_pid_task(pgrp, PIDTYPE_PGID, p);
L
Linus Torvalds 已提交
275 276 277
	return retval;
}

278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308
/*
 * Check to see if any process groups have become orphaned as
 * a result of our exiting, and if they have any stopped jobs,
 * send them a SIGHUP and then a SIGCONT. (POSIX 3.2.2.2)
 */
static void
kill_orphaned_pgrp(struct task_struct *tsk, struct task_struct *parent)
{
	struct pid *pgrp = task_pgrp(tsk);
	struct task_struct *ignored_task = tsk;

	if (!parent)
		 /* exit: our father is in a different pgrp than
		  * we are and we were the only connection outside.
		  */
		parent = tsk->real_parent;
	else
		/* reparent: our child is in a different pgrp than
		 * we are, and it was the only connection outside.
		 */
		ignored_task = NULL;

	if (task_pgrp(parent) != pgrp &&
	    task_session(parent) == task_session(tsk) &&
	    will_become_orphaned_pgrp(pgrp, ignored_task) &&
	    has_stopped_jobs(pgrp)) {
		__kill_pgrp_info(SIGHUP, SEND_SIG_PRIV, pgrp);
		__kill_pgrp_info(SIGCONT, SEND_SIG_PRIV, pgrp);
	}
}

L
Linus Torvalds 已提交
309
/**
310
 * reparent_to_kthreadd - Reparent the calling kernel thread to kthreadd
L
Linus Torvalds 已提交
311 312
 *
 * If a kernel thread is launched as a result of a system call, or if
313 314
 * it ever exits, it should generally reparent itself to kthreadd so it
 * isn't in the way of other processes and is correctly cleaned up on exit.
L
Linus Torvalds 已提交
315 316 317 318
 *
 * The various task state such as scheduling policy and priority may have
 * been inherited from a user process, so we reset them to sane values here.
 *
319
 * NOTE that reparent_to_kthreadd() gives the caller full capabilities.
L
Linus Torvalds 已提交
320
 */
321
static void reparent_to_kthreadd(void)
L
Linus Torvalds 已提交
322 323 324 325 326
{
	write_lock_irq(&tasklist_lock);

	ptrace_unlink(current);
	/* Reparent to init */
327
	current->real_parent = current->parent = kthreadd_task;
R
Roland McGrath 已提交
328
	list_move_tail(&current->sibling, &current->real_parent->children);
L
Linus Torvalds 已提交
329 330 331 332

	/* Set the exit signal to SIGCHLD so we signal init on exit */
	current->exit_signal = SIGCHLD;

333
	if (task_nice(current) < 0)
L
Linus Torvalds 已提交
334 335 336 337 338 339 340 341 342 343 344 345
		set_user_nice(current, 0);
	/* cpus_allowed? */
	/* rt_priority? */
	/* signals? */
	security_task_reparent_to_init(current);
	memcpy(current->signal->rlim, init_task.signal->rlim,
	       sizeof(current->signal->rlim));
	atomic_inc(&(INIT_USER->__count));
	write_unlock_irq(&tasklist_lock);
	switch_uid(INIT_USER);
}

346
void __set_special_pids(struct pid *pid)
L
Linus Torvalds 已提交
347
{
348
	struct task_struct *curr = current->group_leader;
349
	pid_t nr = pid_nr(pid);
L
Linus Torvalds 已提交
350

351
	if (task_session(curr) != pid) {
352
		change_pid(curr, PIDTYPE_SID, pid);
353
		set_task_session(curr, nr);
L
Linus Torvalds 已提交
354
	}
355
	if (task_pgrp(curr) != pid) {
356
		change_pid(curr, PIDTYPE_PGID, pid);
357
		set_task_pgrp(curr, nr);
L
Linus Torvalds 已提交
358 359 360
	}
}

361
static void set_special_pids(struct pid *pid)
L
Linus Torvalds 已提交
362 363
{
	write_lock_irq(&tasklist_lock);
364
	__set_special_pids(pid);
L
Linus Torvalds 已提交
365 366 367 368 369 370 371 372 373 374
	write_unlock_irq(&tasklist_lock);
}

/*
 * Let kernel threads use this to say that they
 * allow a certain signal (since daemonize() will
 * have disabled all of them by default).
 */
int allow_signal(int sig)
{
375
	if (!valid_signal(sig) || sig < 1)
L
Linus Torvalds 已提交
376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395
		return -EINVAL;

	spin_lock_irq(&current->sighand->siglock);
	sigdelset(&current->blocked, sig);
	if (!current->mm) {
		/* Kernel threads handle their own signals.
		   Let the signal code know it'll be handled, so
		   that they don't get converted to SIGKILL or
		   just silently dropped */
		current->sighand->action[(sig)-1].sa.sa_handler = (void __user *)2;
	}
	recalc_sigpending();
	spin_unlock_irq(&current->sighand->siglock);
	return 0;
}

EXPORT_SYMBOL(allow_signal);

int disallow_signal(int sig)
{
396
	if (!valid_signal(sig) || sig < 1)
L
Linus Torvalds 已提交
397 398 399
		return -EINVAL;

	spin_lock_irq(&current->sighand->siglock);
400
	current->sighand->action[(sig)-1].sa.sa_handler = SIG_IGN;
L
Linus Torvalds 已提交
401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428
	recalc_sigpending();
	spin_unlock_irq(&current->sighand->siglock);
	return 0;
}

EXPORT_SYMBOL(disallow_signal);

/*
 *	Put all the gunge required to become a kernel thread without
 *	attached user resources in one place where it belongs.
 */

void daemonize(const char *name, ...)
{
	va_list args;
	struct fs_struct *fs;
	sigset_t blocked;

	va_start(args, name);
	vsnprintf(current->comm, sizeof(current->comm), name, args);
	va_end(args);

	/*
	 * If we were started as result of loading a module, close all of the
	 * user space pages.  We don't need them, and if we didn't close them
	 * they would be locked into memory.
	 */
	exit_mm(current);
429 430 431 432 433
	/*
	 * We don't want to have TIF_FREEZE set if the system-wide hibernation
	 * or suspend transition begins right now.
	 */
	current->flags |= PF_NOFREEZE;
L
Linus Torvalds 已提交
434

435 436 437 438
	if (current->nsproxy != &init_nsproxy) {
		get_nsproxy(&init_nsproxy);
		switch_task_namespaces(current, &init_nsproxy);
	}
439
	set_special_pids(&init_struct_pid);
440
	proc_clear_tty(current);
L
Linus Torvalds 已提交
441 442 443 444 445 446 447 448 449 450 451 452

	/* Block and flush all signals */
	sigfillset(&blocked);
	sigprocmask(SIG_BLOCK, &blocked, NULL);
	flush_signals(current);

	/* Become as one with the init task */

	exit_fs(current);	/* current->fs->count--; */
	fs = init_task.fs;
	current->fs = fs;
	atomic_inc(&fs->count);
S
Serge E. Hallyn 已提交
453

454
	exit_files(current);
L
Linus Torvalds 已提交
455 456 457
	current->files = init_task.files;
	atomic_inc(&current->files->count);

458
	reparent_to_kthreadd();
L
Linus Torvalds 已提交
459 460 461 462
}

EXPORT_SYMBOL(daemonize);

463
static void close_files(struct files_struct * files)
L
Linus Torvalds 已提交
464 465
{
	int i, j;
466
	struct fdtable *fdt;
L
Linus Torvalds 已提交
467 468

	j = 0;
469 470 471 472 473 474

	/*
	 * It is safe to dereference the fd table without RCU or
	 * ->file_lock because this is the last reference to the
	 * files structure.
	 */
475
	fdt = files_fdtable(files);
L
Linus Torvalds 已提交
476 477 478
	for (;;) {
		unsigned long set;
		i = j * __NFDBITS;
479
		if (i >= fdt->max_fds)
L
Linus Torvalds 已提交
480
			break;
481
		set = fdt->open_fds->fds_bits[j++];
L
Linus Torvalds 已提交
482 483
		while (set) {
			if (set & 1) {
484
				struct file * file = xchg(&fdt->fd[i], NULL);
485
				if (file) {
L
Linus Torvalds 已提交
486
					filp_close(file, files);
487 488
					cond_resched();
				}
L
Linus Torvalds 已提交
489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508
			}
			i++;
			set >>= 1;
		}
	}
}

struct files_struct *get_files_struct(struct task_struct *task)
{
	struct files_struct *files;

	task_lock(task);
	files = task->files;
	if (files)
		atomic_inc(&files->count);
	task_unlock(task);

	return files;
}

509
void put_files_struct(struct files_struct *files)
L
Linus Torvalds 已提交
510
{
511 512
	struct fdtable *fdt;

L
Linus Torvalds 已提交
513 514 515 516
	if (atomic_dec_and_test(&files->count)) {
		close_files(files);
		/*
		 * Free the fd and fdset arrays if we expanded them.
517 518 519
		 * If the fdtable was embedded, pass files for freeing
		 * at the end of the RCU grace period. Otherwise,
		 * you can free files immediately.
L
Linus Torvalds 已提交
520
		 */
521
		fdt = files_fdtable(files);
522
		if (fdt != &files->fdtab)
523
			kmem_cache_free(files_cachep, files);
524
		free_fdtable(fdt);
L
Linus Torvalds 已提交
525 526 527
	}
}

528
void reset_files_struct(struct files_struct *files)
529
{
530
	struct task_struct *tsk = current;
531 532 533 534 535 536 537 538 539
	struct files_struct *old;

	old = tsk->files;
	task_lock(tsk);
	tsk->files = files;
	task_unlock(tsk);
	put_files_struct(old);
}

540
void exit_files(struct task_struct *tsk)
L
Linus Torvalds 已提交
541 542 543 544 545 546 547 548 549 550 551
{
	struct files_struct * files = tsk->files;

	if (files) {
		task_lock(tsk);
		tsk->files = NULL;
		task_unlock(tsk);
		put_files_struct(files);
	}
}

552
void put_fs_struct(struct fs_struct *fs)
L
Linus Torvalds 已提交
553 554 555
{
	/* No need to hold fs->lock if we are killing it */
	if (atomic_dec_and_test(&fs->count)) {
J
Jan Blunck 已提交
556 557 558 559
		path_put(&fs->root);
		path_put(&fs->pwd);
		if (fs->altroot.dentry)
			path_put(&fs->altroot);
L
Linus Torvalds 已提交
560 561 562 563
		kmem_cache_free(fs_cachep, fs);
	}
}

564
void exit_fs(struct task_struct *tsk)
L
Linus Torvalds 已提交
565 566 567 568 569 570 571
{
	struct fs_struct * fs = tsk->fs;

	if (fs) {
		task_lock(tsk);
		tsk->fs = NULL;
		task_unlock(tsk);
572
		put_fs_struct(fs);
L
Linus Torvalds 已提交
573 574 575 576 577
	}
}

EXPORT_SYMBOL_GPL(exit_fs);

578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659
#ifdef CONFIG_MM_OWNER
/*
 * Task p is exiting and it owned mm, lets find a new owner for it
 */
static inline int
mm_need_new_owner(struct mm_struct *mm, struct task_struct *p)
{
	/*
	 * If there are other users of the mm and the owner (us) is exiting
	 * we need to find a new owner to take on the responsibility.
	 */
	if (!mm)
		return 0;
	if (atomic_read(&mm->mm_users) <= 1)
		return 0;
	if (mm->owner != p)
		return 0;
	return 1;
}

void mm_update_next_owner(struct mm_struct *mm)
{
	struct task_struct *c, *g, *p = current;

retry:
	if (!mm_need_new_owner(mm, p))
		return;

	read_lock(&tasklist_lock);
	/*
	 * Search in the children
	 */
	list_for_each_entry(c, &p->children, sibling) {
		if (c->mm == mm)
			goto assign_new_owner;
	}

	/*
	 * Search in the siblings
	 */
	list_for_each_entry(c, &p->parent->children, sibling) {
		if (c->mm == mm)
			goto assign_new_owner;
	}

	/*
	 * Search through everything else. We should not get
	 * here often
	 */
	do_each_thread(g, c) {
		if (c->mm == mm)
			goto assign_new_owner;
	} while_each_thread(g, c);

	read_unlock(&tasklist_lock);
	return;

assign_new_owner:
	BUG_ON(c == p);
	get_task_struct(c);
	/*
	 * The task_lock protects c->mm from changing.
	 * We always want mm->owner->mm == mm
	 */
	task_lock(c);
	/*
	 * Delay read_unlock() till we have the task_lock()
	 * to ensure that c does not slip away underneath us
	 */
	read_unlock(&tasklist_lock);
	if (c->mm != mm) {
		task_unlock(c);
		put_task_struct(c);
		goto retry;
	}
	cgroup_mm_owner_callbacks(mm->owner, c);
	mm->owner = c;
	task_unlock(c);
	put_task_struct(c);
}
#endif /* CONFIG_MM_OWNER */

L
Linus Torvalds 已提交
660 661 662 663
/*
 * Turn us into a lazy TLB process if we
 * aren't already..
 */
A
Adrian Bunk 已提交
664
static void exit_mm(struct task_struct * tsk)
L
Linus Torvalds 已提交
665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689
{
	struct mm_struct *mm = tsk->mm;

	mm_release(tsk, mm);
	if (!mm)
		return;
	/*
	 * Serialize with any possible pending coredump.
	 * We must hold mmap_sem around checking core_waiters
	 * and clearing tsk->mm.  The core-inducing thread
	 * will increment core_waiters for each thread in the
	 * group with ->mm != NULL.
	 */
	down_read(&mm->mmap_sem);
	if (mm->core_waiters) {
		up_read(&mm->mmap_sem);
		down_write(&mm->mmap_sem);
		if (!--mm->core_waiters)
			complete(mm->core_startup_done);
		up_write(&mm->mmap_sem);

		wait_for_completion(&mm->core_done);
		down_read(&mm->mmap_sem);
	}
	atomic_inc(&mm->mm_count);
E
Eric Sesterhenn 已提交
690
	BUG_ON(mm != tsk->active_mm);
L
Linus Torvalds 已提交
691 692 693 694 695
	/* more a memory barrier than a real lock */
	task_lock(tsk);
	tsk->mm = NULL;
	up_read(&mm->mmap_sem);
	enter_lazy_tlb(mm, current);
696 697
	/* We don't want this task to be frozen prematurely */
	clear_freeze_flag(tsk);
L
Linus Torvalds 已提交
698
	task_unlock(tsk);
699
	mm_update_next_owner(mm);
L
Linus Torvalds 已提交
700 701 702
	mmput(mm);
}

703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719
/*
 * Return nonzero if @parent's children should reap themselves.
 *
 * Called with write_lock_irq(&tasklist_lock) held.
 */
static int ignoring_children(struct task_struct *parent)
{
	int ret;
	struct sighand_struct *psig = parent->sighand;
	unsigned long flags;
	spin_lock_irqsave(&psig->siglock, flags);
	ret = (psig->action[SIGCHLD-1].sa.sa_handler == SIG_IGN ||
	       (psig->action[SIGCHLD-1].sa.sa_flags & SA_NOCLDWAIT));
	spin_unlock_irqrestore(&psig->siglock, flags);
	return ret;
}

R
Roland McGrath 已提交
720 721 722 723 724 725 726
/*
 * Detach all tasks we were using ptrace on.
 * Any that need to be release_task'd are put on the @dead list.
 *
 * Called with write_lock(&tasklist_lock) held.
 */
static void ptrace_exit(struct task_struct *parent, struct list_head *dead)
L
Linus Torvalds 已提交
727
{
R
Roland McGrath 已提交
728
	struct task_struct *p, *n;
729
	int ign = -1;
730

R
Roland McGrath 已提交
731 732 733 734 735 736 737 738 739 740 741 742 743 744
	list_for_each_entry_safe(p, n, &parent->ptraced, ptrace_entry) {
		__ptrace_unlink(p);

		if (p->exit_state != EXIT_ZOMBIE)
			continue;

		/*
		 * If it's a zombie, our attachedness prevented normal
		 * parent notification or self-reaping.  Do notification
		 * now if it would have happened earlier.  If it should
		 * reap itself, add it to the @dead list.  We can't call
		 * release_task() here because we already hold tasklist_lock.
		 *
		 * If it's our own child, there is no notification to do.
745 746
		 * But if our normal children self-reap, then this child
		 * was prevented by ptrace and we must reap it now.
L
Linus Torvalds 已提交
747
		 */
R
Roland McGrath 已提交
748 749 750
		if (!task_detached(p) && thread_group_empty(p)) {
			if (!same_thread_group(p->real_parent, parent))
				do_notify_parent(p, p->exit_signal);
751 752 753 754 755 756
			else {
				if (ign < 0)
					ign = ignoring_children(parent);
				if (ign)
					p->exit_signal = -1;
			}
R
Roland McGrath 已提交
757
		}
L
Linus Torvalds 已提交
758

R
Roland McGrath 已提交
759
		if (task_detached(p)) {
L
Linus Torvalds 已提交
760
			/*
R
Roland McGrath 已提交
761
			 * Mark it as in the process of being reaped.
L
Linus Torvalds 已提交
762
			 */
R
Roland McGrath 已提交
763 764
			p->exit_state = EXIT_DEAD;
			list_add(&p->ptrace_entry, dead);
L
Linus Torvalds 已提交
765 766
		}
	}
R
Roland McGrath 已提交
767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793
}

/*
 * Finish up exit-time ptrace cleanup.
 *
 * Called without locks.
 */
static void ptrace_exit_finish(struct task_struct *parent,
			       struct list_head *dead)
{
	struct task_struct *p, *n;

	BUG_ON(!list_empty(&parent->ptraced));

	list_for_each_entry_safe(p, n, dead, ptrace_entry) {
		list_del_init(&p->ptrace_entry);
		release_task(p);
	}
}

static void reparent_thread(struct task_struct *p, struct task_struct *father)
{
	if (p->pdeath_signal)
		/* We already hold the tasklist_lock here.  */
		group_send_sig_info(p->pdeath_signal, SEND_SIG_NOINFO, p);

	list_move_tail(&p->sibling, &p->real_parent->children);
L
Linus Torvalds 已提交
794

795 796 797
	/* If this is a threaded reparent there is no need to
	 * notify anyone anything has happened.
	 */
798
	if (same_thread_group(p->real_parent, father))
799 800 801
		return;

	/* We don't want people slaying init.  */
802
	if (!task_detached(p))
803 804 805 806 807
		p->exit_signal = SIGCHLD;

	/* If we'd notified the old parent about this child's death,
	 * also notify the new parent.
	 */
R
Roland McGrath 已提交
808 809
	if (!ptrace_reparented(p) &&
	    p->exit_state == EXIT_ZOMBIE &&
810
	    !task_detached(p) && thread_group_empty(p))
811 812
		do_notify_parent(p, p->exit_signal);

813
	kill_orphaned_pgrp(p, father);
L
Linus Torvalds 已提交
814 815 816 817 818 819
}

/*
 * When we die, we re-parent all our children.
 * Try to give them to another thread in our thread
 * group, and if no such member exists, give it to
820 821
 * the child reaper process (ie "init") in our pid
 * space.
L
Linus Torvalds 已提交
822
 */
823
static void forget_original_parent(struct task_struct *father)
L
Linus Torvalds 已提交
824
{
825
	struct task_struct *p, *n, *reaper = father;
R
Roland McGrath 已提交
826
	LIST_HEAD(ptrace_dead);
827 828

	write_lock_irq(&tasklist_lock);
L
Linus Torvalds 已提交
829

R
Roland McGrath 已提交
830 831 832 833 834
	/*
	 * First clean up ptrace if we were using it.
	 */
	ptrace_exit(father, &ptrace_dead);

L
Linus Torvalds 已提交
835 836 837
	do {
		reaper = next_thread(reaper);
		if (reaper == father) {
838
			reaper = task_child_reaper(father);
L
Linus Torvalds 已提交
839 840
			break;
		}
841
	} while (reaper->flags & PF_EXITING);
L
Linus Torvalds 已提交
842

843
	list_for_each_entry_safe(p, n, &father->children, sibling) {
844
		p->real_parent = reaper;
R
Roland McGrath 已提交
845 846 847 848 849
		if (p->parent == father) {
			BUG_ON(p->ptrace);
			p->parent = p->real_parent;
		}
		reparent_thread(p, father);
L
Linus Torvalds 已提交
850
	}
851 852 853 854

	write_unlock_irq(&tasklist_lock);
	BUG_ON(!list_empty(&father->children));

R
Roland McGrath 已提交
855
	ptrace_exit_finish(father, &ptrace_dead);
L
Linus Torvalds 已提交
856 857 858 859 860 861
}

/*
 * Send signals to all our closest relatives so that they know
 * to properly mourn us..
 */
862
static void exit_notify(struct task_struct *tsk, int group_dead)
L
Linus Torvalds 已提交
863 864 865 866 867 868 869 870 871 872 873
{
	int state;

	/*
	 * This does two things:
	 *
  	 * A.  Make init inherit all the child processes
	 * B.  Check to see if any process groups have become orphaned
	 *	as a result of our exiting, and if they have any stopped
	 *	jobs, send them a SIGHUP and then a SIGCONT.  (POSIX 3.2.2.2)
	 */
874
	forget_original_parent(tsk);
875
	exit_task_namespaces(tsk);
L
Linus Torvalds 已提交
876

877
	write_lock_irq(&tasklist_lock);
878 879
	if (group_dead)
		kill_orphaned_pgrp(tsk->group_leader, NULL);
L
Linus Torvalds 已提交
880

881
	/* Let father know we died
L
Linus Torvalds 已提交
882 883
	 *
	 * Thread signals are configurable, but you aren't going to use
884
	 * that to send signals to arbitary processes.
L
Linus Torvalds 已提交
885 886 887 888 889 890 891 892 893 894
	 * That stops right now.
	 *
	 * If the parent exec id doesn't match the exec id we saved
	 * when we started then we know the parent has changed security
	 * domain.
	 *
	 * If our self_exec id doesn't match our parent_exec_id then
	 * we have changed execution domain as these two values started
	 * the same after a fork.
	 */
895
	if (tsk->exit_signal != SIGCHLD && !task_detached(tsk) &&
896
	    (tsk->parent_exec_id != tsk->real_parent->self_exec_id ||
897 898
	     tsk->self_exec_id != tsk->parent_exec_id) &&
	    !capable(CAP_KILL))
L
Linus Torvalds 已提交
899 900 901 902 903 904
		tsk->exit_signal = SIGCHLD;

	/* If something other than our normal parent is ptracing us, then
	 * send it a SIGCHLD instead of honoring exit_signal.  exit_signal
	 * only has special meaning to our real parent.
	 */
905
	if (!task_detached(tsk) && thread_group_empty(tsk)) {
906 907
		int signal = ptrace_reparented(tsk) ?
				SIGCHLD : tsk->exit_signal;
L
Linus Torvalds 已提交
908 909 910 911 912 913
		do_notify_parent(tsk, signal);
	} else if (tsk->ptrace) {
		do_notify_parent(tsk, SIGCHLD);
	}

	state = EXIT_ZOMBIE;
914
	if (task_detached(tsk) && likely(!tsk->ptrace))
L
Linus Torvalds 已提交
915 916 917
		state = EXIT_DEAD;
	tsk->exit_state = state;

918
	/* mt-exec, de_thread() is waiting for us */
919 920 921 922 923
	if (thread_group_leader(tsk) &&
	    tsk->signal->notify_count < 0 &&
	    tsk->signal->group_exit_task)
		wake_up_process(tsk->signal->group_exit_task);

L
Linus Torvalds 已提交
924 925 926 927 928 929 930
	write_unlock_irq(&tasklist_lock);

	/* If the process is dead, release it - nobody will wait for it */
	if (state == EXIT_DEAD)
		release_task(tsk);
}

931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958
#ifdef CONFIG_DEBUG_STACK_USAGE
static void check_stack_usage(void)
{
	static DEFINE_SPINLOCK(low_water_lock);
	static int lowest_to_date = THREAD_SIZE;
	unsigned long *n = end_of_stack(current);
	unsigned long free;

	while (*n == 0)
		n++;
	free = (unsigned long)n - (unsigned long)end_of_stack(current);

	if (free >= lowest_to_date)
		return;

	spin_lock(&low_water_lock);
	if (free < lowest_to_date) {
		printk(KERN_WARNING "%s used greatest stack depth: %lu bytes "
				"left\n",
				current->comm, free);
		lowest_to_date = free;
	}
	spin_unlock(&low_water_lock);
}
#else
static inline void check_stack_usage(void) {}
#endif

959 960
static inline void exit_child_reaper(struct task_struct *tsk)
{
961
	if (likely(tsk->group_leader != task_child_reaper(tsk)))
962 963
		return;

964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989
	if (tsk->nsproxy->pid_ns == &init_pid_ns)
		panic("Attempted to kill init!");

	/*
	 * @tsk is the last thread in the 'cgroup-init' and is exiting.
	 * Terminate all remaining processes in the namespace and reap them
	 * before exiting @tsk.
	 *
	 * Note that @tsk (last thread of cgroup-init) may not necessarily
	 * be the child-reaper (i.e main thread of cgroup-init) of the
	 * namespace i.e the child_reaper may have already exited.
	 *
	 * Even after a child_reaper exits, we let it inherit orphaned children,
	 * because, pid_ns->child_reaper remains valid as long as there is
	 * at least one living sub-thread in the cgroup init.

	 * This living sub-thread of the cgroup-init will be notified when
	 * a child inherited by the 'child-reaper' exits (do_notify_parent()
	 * uses __group_send_sig_info()). Further, when reaping child processes,
	 * do_wait() iterates over children of all living sub threads.

	 * i.e even though 'child_reaper' thread is listed as the parent of the
	 * orphaned children, any living sub-thread in the cgroup-init can
	 * perform the role of the child_reaper.
	 */
	zap_pid_ns_processes(tsk->nsproxy->pid_ns);
990 991
}

992
NORET_TYPE void do_exit(long code)
L
Linus Torvalds 已提交
993 994 995 996 997 998
{
	struct task_struct *tsk = current;
	int group_dead;

	profile_task_exit(tsk);

999 1000
	WARN_ON(atomic_read(&tsk->fs_excl));

L
Linus Torvalds 已提交
1001 1002 1003 1004 1005 1006 1007 1008 1009 1010
	if (unlikely(in_interrupt()))
		panic("Aiee, killing interrupt handler!");
	if (unlikely(!tsk->pid))
		panic("Attempted to kill the idle task!");

	if (unlikely(current->ptrace & PT_TRACE_EXIT)) {
		current->ptrace_message = code;
		ptrace_notify((PTRACE_EVENT_EXIT << 8) | SIGTRAP);
	}

1011 1012 1013 1014 1015 1016 1017
	/*
	 * We're taking recursive faults here in do_exit. Safest is to just
	 * leave this task alone and wait for reboot.
	 */
	if (unlikely(tsk->flags & PF_EXITING)) {
		printk(KERN_ALERT
			"Fixing recursive fault but reboot is needed!\n");
1018 1019 1020 1021 1022 1023 1024 1025 1026 1027
		/*
		 * We can do this unlocked here. The futex code uses
		 * this flag just to verify whether the pi state
		 * cleanup has been done or not. In the worst case it
		 * loops once more. We pretend that the cleanup was
		 * done as there is no way to return. Either the
		 * OWNER_DIED bit is set by now or we push the blocked
		 * task into the wait for ever nirwana as well.
		 */
		tsk->flags |= PF_EXITPIDONE;
1028 1029
		if (tsk->io_context)
			exit_io_context();
1030 1031 1032 1033
		set_current_state(TASK_UNINTERRUPTIBLE);
		schedule();
	}

O
Oleg Nesterov 已提交
1034
	exit_signals(tsk);  /* sets PF_EXITING */
1035 1036 1037 1038
	/*
	 * tsk->flags are checked in the futex code to protect against
	 * an exiting task cleaning up the robust pi futexes.
	 */
1039 1040
	smp_mb();
	spin_unlock_wait(&tsk->pi_lock);
L
Linus Torvalds 已提交
1041 1042 1043

	if (unlikely(in_atomic()))
		printk(KERN_INFO "note: %s[%d] exited with preempt_count %d\n",
1044
				current->comm, task_pid_nr(current),
L
Linus Torvalds 已提交
1045 1046 1047
				preempt_count());

	acct_update_integrals(tsk);
1048 1049 1050 1051
	if (tsk->mm) {
		update_hiwater_rss(tsk->mm);
		update_hiwater_vm(tsk->mm);
	}
L
Linus Torvalds 已提交
1052
	group_dead = atomic_dec_and_test(&tsk->signal->live);
1053
	if (group_dead) {
1054
		exit_child_reaper(tsk);
1055
		hrtimer_cancel(&tsk->signal->real_timer);
1056
		exit_itimers(tsk->signal);
1057
	}
1058
	acct_collect(code, group_dead);
1059
#ifdef CONFIG_FUTEX
1060 1061
	if (unlikely(tsk->robust_list))
		exit_robust_list(tsk);
1062
#ifdef CONFIG_COMPAT
1063 1064
	if (unlikely(tsk->compat_robust_list))
		compat_exit_robust_list(tsk);
1065
#endif
1066
#endif
M
Miloslav Trmac 已提交
1067 1068
	if (group_dead)
		tty_audit_exit();
1069 1070
	if (unlikely(tsk->audit_context))
		audit_free(tsk);
1071

1072
	tsk->exit_code = code;
1073
	taskstats_exit(tsk, group_dead);
1074

L
Linus Torvalds 已提交
1075 1076
	exit_mm(tsk);

1077
	if (group_dead)
1078
		acct_process();
L
Linus Torvalds 已提交
1079
	exit_sem(tsk);
1080 1081
	exit_files(tsk);
	exit_fs(tsk);
1082
	check_stack_usage();
L
Linus Torvalds 已提交
1083
	exit_thread();
1084
	cgroup_exit(tsk, 1);
L
Linus Torvalds 已提交
1085 1086 1087 1088 1089
	exit_keys(tsk);

	if (group_dead && tsk->signal->leader)
		disassociate_ctty(1);

A
Al Viro 已提交
1090
	module_put(task_thread_info(tsk)->exec_domain->module);
L
Linus Torvalds 已提交
1091 1092 1093
	if (tsk->binfmt)
		module_put(tsk->binfmt->module);

M
Matt Helsley 已提交
1094
	proc_exit_connector(tsk);
1095
	exit_notify(tsk, group_dead);
L
Linus Torvalds 已提交
1096
#ifdef CONFIG_NUMA
1097
	mpol_put(tsk->mempolicy);
L
Linus Torvalds 已提交
1098 1099
	tsk->mempolicy = NULL;
#endif
1100
#ifdef CONFIG_FUTEX
1101 1102 1103 1104 1105 1106 1107 1108
	/*
	 * This must happen late, after the PID is not
	 * hashed anymore:
	 */
	if (unlikely(!list_empty(&tsk->pi_state_list)))
		exit_pi_state_list(tsk);
	if (unlikely(current->pi_state_cache))
		kfree(current->pi_state_cache);
1109
#endif
1110
	/*
1111
	 * Make sure we are holding no locks:
1112
	 */
1113
	debug_check_no_locks_held(tsk);
1114 1115 1116 1117 1118 1119
	/*
	 * We can do this unlocked here. The futex code uses this flag
	 * just to verify whether the pi state cleanup has been done
	 * or not. In the worst case it loops once more.
	 */
	tsk->flags |= PF_EXITPIDONE;
L
Linus Torvalds 已提交
1120

1121 1122 1123
	if (tsk->io_context)
		exit_io_context();

1124 1125 1126
	if (tsk->splice_pipe)
		__free_pipe_info(tsk->splice_pipe);

C
Coywolf Qi Hunt 已提交
1127
	preempt_disable();
O
Oleg Nesterov 已提交
1128
	/* causes final put_task_struct in finish_task_switch(). */
1129
	tsk->state = TASK_DEAD;
C
Coywolf Qi Hunt 已提交
1130

L
Linus Torvalds 已提交
1131 1132 1133
	schedule();
	BUG();
	/* Avoid "noreturn function does return".  */
A
Alan Cox 已提交
1134 1135
	for (;;)
		cpu_relax();	/* For when BUG is null */
L
Linus Torvalds 已提交
1136 1137
}

1138 1139
EXPORT_SYMBOL_GPL(do_exit);

L
Linus Torvalds 已提交
1140 1141 1142 1143
NORET_TYPE void complete_and_exit(struct completion *comp, long code)
{
	if (comp)
		complete(comp);
O
Oleg Nesterov 已提交
1144

L
Linus Torvalds 已提交
1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161
	do_exit(code);
}

EXPORT_SYMBOL(complete_and_exit);

asmlinkage long sys_exit(int error_code)
{
	do_exit((error_code&0xff)<<8);
}

/*
 * Take down every thread in the group.  This is called by fatal signals
 * as well as by sys_exit_group (below).
 */
NORET_TYPE void
do_group_exit(int exit_code)
{
1162 1163
	struct signal_struct *sig = current->signal;

L
Linus Torvalds 已提交
1164 1165
	BUG_ON(exit_code & 0x80); /* core dumps don't get here */

1166 1167
	if (signal_group_exit(sig))
		exit_code = sig->group_exit_code;
L
Linus Torvalds 已提交
1168 1169 1170
	else if (!thread_group_empty(current)) {
		struct sighand_struct *const sighand = current->sighand;
		spin_lock_irq(&sighand->siglock);
1171
		if (signal_group_exit(sig))
L
Linus Torvalds 已提交
1172 1173 1174 1175
			/* Another thread got here before we took the lock.  */
			exit_code = sig->group_exit_code;
		else {
			sig->group_exit_code = exit_code;
1176
			sig->flags = SIGNAL_GROUP_EXIT;
L
Linus Torvalds 已提交
1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195
			zap_other_threads(current);
		}
		spin_unlock_irq(&sighand->siglock);
	}

	do_exit(exit_code);
	/* NOTREACHED */
}

/*
 * this kills every thread in the thread group. Note that any externally
 * wait4()-ing process will get the correct exit code - even if this
 * thread is not the thread group leader.
 */
asmlinkage void sys_exit_group(int error_code)
{
	do_group_exit((error_code & 0xff) << 8);
}

E
Eric W. Biederman 已提交
1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207
static struct pid *task_pid_type(struct task_struct *task, enum pid_type type)
{
	struct pid *pid = NULL;
	if (type == PIDTYPE_PID)
		pid = task->pids[type].pid;
	else if (type < PIDTYPE_MAX)
		pid = task->group_leader->pids[type].pid;
	return pid;
}

static int eligible_child(enum pid_type type, struct pid *pid, int options,
			  struct task_struct *p)
L
Linus Torvalds 已提交
1208
{
1209 1210
	int err;

E
Eric W. Biederman 已提交
1211 1212
	if (type < PIDTYPE_MAX) {
		if (task_pid_type(p, type) != pid)
L
Linus Torvalds 已提交
1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224
			return 0;
	}

	/* Wait for all children (clone and not) if __WALL is set;
	 * otherwise, wait for clone children *only* if __WCLONE is
	 * set; otherwise, wait for non-clone children *only*.  (Note:
	 * A "clone" child here is one that reports to its parent
	 * using a signal other than SIGCHLD.) */
	if (((p->exit_signal != SIGCHLD) ^ ((options & __WCLONE) != 0))
	    && !(options & __WALL))
		return 0;

1225
	err = security_task_wait(p);
1226 1227
	if (err)
		return err;
L
Linus Torvalds 已提交
1228

1229
	return 1;
L
Linus Torvalds 已提交
1230 1231
}

1232
static int wait_noreap_copyout(struct task_struct *p, pid_t pid, uid_t uid,
L
Linus Torvalds 已提交
1233 1234 1235 1236 1237
			       int why, int status,
			       struct siginfo __user *infop,
			       struct rusage __user *rusagep)
{
	int retval = rusagep ? getrusage(p, RUSAGE_BOTH, rusagep) : 0;
1238

L
Linus Torvalds 已提交
1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262
	put_task_struct(p);
	if (!retval)
		retval = put_user(SIGCHLD, &infop->si_signo);
	if (!retval)
		retval = put_user(0, &infop->si_errno);
	if (!retval)
		retval = put_user((short)why, &infop->si_code);
	if (!retval)
		retval = put_user(pid, &infop->si_pid);
	if (!retval)
		retval = put_user(uid, &infop->si_uid);
	if (!retval)
		retval = put_user(status, &infop->si_status);
	if (!retval)
		retval = pid;
	return retval;
}

/*
 * Handle sys_wait4 work for one task in state EXIT_ZOMBIE.  We hold
 * read_lock(&tasklist_lock) on entry.  If we return zero, we still hold
 * the lock and this task is uninteresting.  If we return nonzero, we have
 * released the lock and the system call should return.
 */
R
Roland McGrath 已提交
1263
static int wait_task_zombie(struct task_struct *p, int options,
L
Linus Torvalds 已提交
1264 1265 1266 1267
			    struct siginfo __user *infop,
			    int __user *stat_addr, struct rusage __user *ru)
{
	unsigned long state;
1268
	int retval, status, traced;
1269
	pid_t pid = task_pid_vnr(p);
L
Linus Torvalds 已提交
1270

R
Roland McGrath 已提交
1271 1272 1273 1274
	if (!likely(options & WEXITED))
		return 0;

	if (unlikely(options & WNOWAIT)) {
L
Linus Torvalds 已提交
1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301
		uid_t uid = p->uid;
		int exit_code = p->exit_code;
		int why, status;

		get_task_struct(p);
		read_unlock(&tasklist_lock);
		if ((exit_code & 0x7f) == 0) {
			why = CLD_EXITED;
			status = exit_code >> 8;
		} else {
			why = (exit_code & 0x80) ? CLD_DUMPED : CLD_KILLED;
			status = exit_code & 0x7f;
		}
		return wait_noreap_copyout(p, pid, uid, why,
					   status, infop, ru);
	}

	/*
	 * Try to move the task's state to DEAD
	 * only one thread is allowed to do this:
	 */
	state = xchg(&p->exit_state, EXIT_DEAD);
	if (state != EXIT_ZOMBIE) {
		BUG_ON(state != EXIT_DEAD);
		return 0;
	}

1302
	traced = ptrace_reparented(p);
1303 1304

	if (likely(!traced)) {
1305 1306 1307
		struct signal_struct *psig;
		struct signal_struct *sig;

L
Linus Torvalds 已提交
1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323
		/*
		 * The resource counters for the group leader are in its
		 * own task_struct.  Those for dead threads in the group
		 * are in its signal_struct, as are those for the child
		 * processes it has previously reaped.  All these
		 * accumulate in the parent's signal_struct c* fields.
		 *
		 * We don't bother to take a lock here to protect these
		 * p->signal fields, because they are only touched by
		 * __exit_signal, which runs with tasklist_lock
		 * write-locked anyway, and so is excluded here.  We do
		 * need to protect the access to p->parent->signal fields,
		 * as other threads in the parent group can be right
		 * here reaping other children at the same time.
		 */
		spin_lock_irq(&p->parent->sighand->siglock);
1324 1325 1326 1327
		psig = p->parent->signal;
		sig = p->signal;
		psig->cutime =
			cputime_add(psig->cutime,
L
Linus Torvalds 已提交
1328
			cputime_add(p->utime,
1329 1330 1331 1332
			cputime_add(sig->utime,
				    sig->cutime)));
		psig->cstime =
			cputime_add(psig->cstime,
L
Linus Torvalds 已提交
1333
			cputime_add(p->stime,
1334 1335
			cputime_add(sig->stime,
				    sig->cstime)));
1336 1337 1338 1339 1340
		psig->cgtime =
			cputime_add(psig->cgtime,
			cputime_add(p->gtime,
			cputime_add(sig->gtime,
				    sig->cgtime)));
1341 1342 1343 1344 1345 1346 1347 1348
		psig->cmin_flt +=
			p->min_flt + sig->min_flt + sig->cmin_flt;
		psig->cmaj_flt +=
			p->maj_flt + sig->maj_flt + sig->cmaj_flt;
		psig->cnvcsw +=
			p->nvcsw + sig->nvcsw + sig->cnvcsw;
		psig->cnivcsw +=
			p->nivcsw + sig->nivcsw + sig->cnivcsw;
1349 1350 1351 1352 1353 1354
		psig->cinblock +=
			task_io_get_inblock(p) +
			sig->inblock + sig->cinblock;
		psig->coublock +=
			task_io_get_oublock(p) +
			sig->oublock + sig->coublock;
L
Linus Torvalds 已提交
1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387
		spin_unlock_irq(&p->parent->sighand->siglock);
	}

	/*
	 * Now we are sure this task is interesting, and no other
	 * thread can reap it because we set its state to EXIT_DEAD.
	 */
	read_unlock(&tasklist_lock);

	retval = ru ? getrusage(p, RUSAGE_BOTH, ru) : 0;
	status = (p->signal->flags & SIGNAL_GROUP_EXIT)
		? p->signal->group_exit_code : p->exit_code;
	if (!retval && stat_addr)
		retval = put_user(status, stat_addr);
	if (!retval && infop)
		retval = put_user(SIGCHLD, &infop->si_signo);
	if (!retval && infop)
		retval = put_user(0, &infop->si_errno);
	if (!retval && infop) {
		int why;

		if ((status & 0x7f) == 0) {
			why = CLD_EXITED;
			status >>= 8;
		} else {
			why = (status & 0x80) ? CLD_DUMPED : CLD_KILLED;
			status &= 0x7f;
		}
		retval = put_user((short)why, &infop->si_code);
		if (!retval)
			retval = put_user(status, &infop->si_status);
	}
	if (!retval && infop)
1388
		retval = put_user(pid, &infop->si_pid);
L
Linus Torvalds 已提交
1389 1390
	if (!retval && infop)
		retval = put_user(p->uid, &infop->si_uid);
1391
	if (!retval)
1392
		retval = pid;
1393 1394

	if (traced) {
L
Linus Torvalds 已提交
1395
		write_lock_irq(&tasklist_lock);
1396 1397 1398 1399 1400 1401 1402
		/* We dropped tasklist, ptracer could die and untrace */
		ptrace_unlink(p);
		/*
		 * If this is not a detached task, notify the parent.
		 * If it's still not detached after that, don't release
		 * it now.
		 */
1403
		if (!task_detached(p)) {
1404
			do_notify_parent(p, p->exit_signal);
1405
			if (!task_detached(p)) {
1406 1407
				p->exit_state = EXIT_ZOMBIE;
				p = NULL;
L
Linus Torvalds 已提交
1408 1409 1410 1411 1412 1413
			}
		}
		write_unlock_irq(&tasklist_lock);
	}
	if (p != NULL)
		release_task(p);
1414

L
Linus Torvalds 已提交
1415 1416 1417 1418 1419 1420 1421 1422 1423
	return retval;
}

/*
 * Handle sys_wait4 work for one task in state TASK_STOPPED.  We hold
 * read_lock(&tasklist_lock) on entry.  If we return zero, we still hold
 * the lock and this task is uninteresting.  If we return nonzero, we have
 * released the lock and the system call should return.
 */
R
Roland McGrath 已提交
1424
static int wait_task_stopped(int ptrace, struct task_struct *p,
R
Roland McGrath 已提交
1425
			     int options, struct siginfo __user *infop,
L
Linus Torvalds 已提交
1426 1427
			     int __user *stat_addr, struct rusage __user *ru)
{
1428 1429
	int retval, exit_code, why;
	uid_t uid = 0; /* unneeded, required by compiler */
1430
	pid_t pid;
L
Linus Torvalds 已提交
1431

R
Roland McGrath 已提交
1432
	if (!(options & WUNTRACED))
R
Roland McGrath 已提交
1433 1434
		return 0;

1435 1436 1437 1438 1439 1440
	exit_code = 0;
	spin_lock_irq(&p->sighand->siglock);

	if (unlikely(!task_is_stopped_or_traced(p)))
		goto unlock_sig;

R
Roland McGrath 已提交
1441
	if (!ptrace && p->signal->group_stop_count > 0)
L
Linus Torvalds 已提交
1442 1443 1444 1445
		/*
		 * A group stop is in progress and this is the group leader.
		 * We won't report until all threads have stopped.
		 */
1446 1447 1448 1449 1450 1451
		goto unlock_sig;

	exit_code = p->exit_code;
	if (!exit_code)
		goto unlock_sig;

R
Roland McGrath 已提交
1452
	if (!unlikely(options & WNOWAIT))
1453 1454 1455 1456 1457 1458
		p->exit_code = 0;

	uid = p->uid;
unlock_sig:
	spin_unlock_irq(&p->sighand->siglock);
	if (!exit_code)
L
Linus Torvalds 已提交
1459 1460 1461 1462 1463 1464 1465 1466 1467 1468
		return 0;

	/*
	 * Now we are pretty sure this task is interesting.
	 * Make sure it doesn't get reaped out from under us while we
	 * give up the lock and then examine it below.  We don't want to
	 * keep holding onto the tasklist_lock while we call getrusage and
	 * possibly take page faults for user memory.
	 */
	get_task_struct(p);
1469
	pid = task_pid_vnr(p);
R
Roland McGrath 已提交
1470
	why = ptrace ? CLD_TRAPPED : CLD_STOPPED;
L
Linus Torvalds 已提交
1471 1472
	read_unlock(&tasklist_lock);

R
Roland McGrath 已提交
1473
	if (unlikely(options & WNOWAIT))
L
Linus Torvalds 已提交
1474
		return wait_noreap_copyout(p, pid, uid,
1475
					   why, exit_code,
L
Linus Torvalds 已提交
1476 1477 1478 1479 1480 1481 1482 1483 1484 1485
					   infop, ru);

	retval = ru ? getrusage(p, RUSAGE_BOTH, ru) : 0;
	if (!retval && stat_addr)
		retval = put_user((exit_code << 8) | 0x7f, stat_addr);
	if (!retval && infop)
		retval = put_user(SIGCHLD, &infop->si_signo);
	if (!retval && infop)
		retval = put_user(0, &infop->si_errno);
	if (!retval && infop)
R
Roland McGrath 已提交
1486
		retval = put_user((short)why, &infop->si_code);
L
Linus Torvalds 已提交
1487 1488 1489
	if (!retval && infop)
		retval = put_user(exit_code, &infop->si_status);
	if (!retval && infop)
1490
		retval = put_user(pid, &infop->si_pid);
L
Linus Torvalds 已提交
1491
	if (!retval && infop)
1492
		retval = put_user(uid, &infop->si_uid);
L
Linus Torvalds 已提交
1493
	if (!retval)
1494
		retval = pid;
L
Linus Torvalds 已提交
1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506
	put_task_struct(p);

	BUG_ON(!retval);
	return retval;
}

/*
 * Handle do_wait work for one task in a live, non-stopped state.
 * read_lock(&tasklist_lock) on entry.  If we return zero, we still hold
 * the lock and this task is uninteresting.  If we return nonzero, we have
 * released the lock and the system call should return.
 */
R
Roland McGrath 已提交
1507
static int wait_task_continued(struct task_struct *p, int options,
L
Linus Torvalds 已提交
1508 1509 1510 1511 1512 1513 1514
			       struct siginfo __user *infop,
			       int __user *stat_addr, struct rusage __user *ru)
{
	int retval;
	pid_t pid;
	uid_t uid;

R
Roland McGrath 已提交
1515 1516 1517
	if (!unlikely(options & WCONTINUED))
		return 0;

L
Linus Torvalds 已提交
1518 1519 1520 1521 1522 1523 1524 1525 1526
	if (!(p->signal->flags & SIGNAL_STOP_CONTINUED))
		return 0;

	spin_lock_irq(&p->sighand->siglock);
	/* Re-check with the lock held.  */
	if (!(p->signal->flags & SIGNAL_STOP_CONTINUED)) {
		spin_unlock_irq(&p->sighand->siglock);
		return 0;
	}
R
Roland McGrath 已提交
1527
	if (!unlikely(options & WNOWAIT))
L
Linus Torvalds 已提交
1528 1529 1530
		p->signal->flags &= ~SIGNAL_STOP_CONTINUED;
	spin_unlock_irq(&p->sighand->siglock);

1531
	pid = task_pid_vnr(p);
L
Linus Torvalds 已提交
1532 1533 1534 1535 1536 1537 1538 1539 1540 1541
	uid = p->uid;
	get_task_struct(p);
	read_unlock(&tasklist_lock);

	if (!infop) {
		retval = ru ? getrusage(p, RUSAGE_BOTH, ru) : 0;
		put_task_struct(p);
		if (!retval && stat_addr)
			retval = put_user(0xffff, stat_addr);
		if (!retval)
1542
			retval = pid;
L
Linus Torvalds 已提交
1543 1544 1545 1546 1547 1548 1549 1550 1551 1552
	} else {
		retval = wait_noreap_copyout(p, pid, uid,
					     CLD_CONTINUED, SIGCONT,
					     infop, ru);
		BUG_ON(retval == 0);
	}

	return retval;
}

R
Roland McGrath 已提交
1553 1554 1555 1556 1557 1558
/*
 * Consider @p for a wait by @parent.
 *
 * -ECHILD should be in *@notask_error before the first call.
 * Returns nonzero for a final return, when we have unlocked tasklist_lock.
 * Returns zero if the search for a child should continue;
1559 1560
 * then *@notask_error is 0 if @p is an eligible child,
 * or another error from security_task_wait(), or still -ECHILD.
R
Roland McGrath 已提交
1561
 */
R
Roland McGrath 已提交
1562
static int wait_consider_task(struct task_struct *parent, int ptrace,
R
Roland McGrath 已提交
1563 1564 1565 1566 1567 1568
			      struct task_struct *p, int *notask_error,
			      enum pid_type type, struct pid *pid, int options,
			      struct siginfo __user *infop,
			      int __user *stat_addr, struct rusage __user *ru)
{
	int ret = eligible_child(type, pid, options, p);
1569
	if (!ret)
R
Roland McGrath 已提交
1570 1571
		return ret;

1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583
	if (unlikely(ret < 0)) {
		/*
		 * If we have not yet seen any eligible child,
		 * then let this error code replace -ECHILD.
		 * A permission error will give the user a clue
		 * to look for security policy problems, rather
		 * than for mysterious wait bugs.
		 */
		if (*notask_error)
			*notask_error = ret;
	}

R
Roland McGrath 已提交
1584 1585 1586 1587 1588 1589 1590 1591 1592
	if (likely(!ptrace) && unlikely(p->ptrace)) {
		/*
		 * This child is hidden by ptrace.
		 * We aren't allowed to see it now, but eventually we will.
		 */
		*notask_error = 0;
		return 0;
	}

R
Roland McGrath 已提交
1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608
	if (p->exit_state == EXIT_DEAD)
		return 0;

	/*
	 * We don't reap group leaders with subthreads.
	 */
	if (p->exit_state == EXIT_ZOMBIE && !delay_group_leader(p))
		return wait_task_zombie(p, options, infop, stat_addr, ru);

	/*
	 * It's stopped or running now, so it might
	 * later continue, exit, or stop again.
	 */
	*notask_error = 0;

	if (task_is_stopped_or_traced(p))
R
Roland McGrath 已提交
1609 1610
		return wait_task_stopped(ptrace, p, options,
					 infop, stat_addr, ru);
R
Roland McGrath 已提交
1611 1612 1613 1614 1615 1616 1617 1618 1619 1620

	return wait_task_continued(p, options, infop, stat_addr, ru);
}

/*
 * Do the work of do_wait() for one thread in the group, @tsk.
 *
 * -ECHILD should be in *@notask_error before the first call.
 * Returns nonzero for a final return, when we have unlocked tasklist_lock.
 * Returns zero if the search for a child should continue; then
1621 1622
 * *@notask_error is 0 if there were any eligible children,
 * or another error from security_task_wait(), or still -ECHILD.
R
Roland McGrath 已提交
1623 1624 1625 1626 1627 1628 1629 1630 1631
 */
static int do_wait_thread(struct task_struct *tsk, int *notask_error,
			  enum pid_type type, struct pid *pid, int options,
			  struct siginfo __user *infop, int __user *stat_addr,
			  struct rusage __user *ru)
{
	struct task_struct *p;

	list_for_each_entry(p, &tsk->children, sibling) {
R
Roland McGrath 已提交
1632 1633 1634 1635 1636 1637 1638 1639 1640 1641
		/*
		 * Do not consider detached threads.
		 */
		if (!task_detached(p)) {
			int ret = wait_consider_task(tsk, 0, p, notask_error,
						     type, pid, options,
						     infop, stat_addr, ru);
			if (ret)
				return ret;
		}
R
Roland McGrath 已提交
1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654
	}

	return 0;
}

static int ptrace_do_wait(struct task_struct *tsk, int *notask_error,
			  enum pid_type type, struct pid *pid, int options,
			  struct siginfo __user *infop, int __user *stat_addr,
			  struct rusage __user *ru)
{
	struct task_struct *p;

	/*
R
Roland McGrath 已提交
1655
	 * Traditionally we see ptrace'd stopped tasks regardless of options.
R
Roland McGrath 已提交
1656
	 */
R
Roland McGrath 已提交
1657
	options |= WUNTRACED;
R
Roland McGrath 已提交
1658

R
Roland McGrath 已提交
1659 1660 1661 1662 1663
	list_for_each_entry(p, &tsk->ptraced, ptrace_entry) {
		int ret = wait_consider_task(tsk, 1, p, notask_error,
					     type, pid, options,
					     infop, stat_addr, ru);
		if (ret)
R
Roland McGrath 已提交
1664 1665 1666 1667 1668 1669
			return ret;
	}

	return 0;
}

E
Eric W. Biederman 已提交
1670 1671 1672
static long do_wait(enum pid_type type, struct pid *pid, int options,
		    struct siginfo __user *infop, int __user *stat_addr,
		    struct rusage __user *ru)
L
Linus Torvalds 已提交
1673 1674 1675
{
	DECLARE_WAITQUEUE(wait, current);
	struct task_struct *tsk;
R
Roland McGrath 已提交
1676
	int retval;
L
Linus Torvalds 已提交
1677 1678 1679

	add_wait_queue(&current->signal->wait_chldexit,&wait);
repeat:
R
Roland McGrath 已提交
1680 1681 1682 1683 1684
	/*
	 * If there is nothing that can match our critiera just get out.
	 * We will clear @retval to zero if we see any child that might later
	 * match our criteria, even if we are not able to reap it yet.
	 */
E
Eric W. Biederman 已提交
1685 1686 1687 1688
	retval = -ECHILD;
	if ((type < PIDTYPE_MAX) && (!pid || hlist_empty(&pid->tasks[type])))
		goto end;

L
Linus Torvalds 已提交
1689 1690 1691 1692
	current->state = TASK_INTERRUPTIBLE;
	read_lock(&tasklist_lock);
	tsk = current;
	do {
R
Roland McGrath 已提交
1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705
		int tsk_result = do_wait_thread(tsk, &retval,
						type, pid, options,
						infop, stat_addr, ru);
		if (!tsk_result)
			tsk_result = ptrace_do_wait(tsk, &retval,
						    type, pid, options,
						    infop, stat_addr, ru);
		if (tsk_result) {
			/*
			 * tasklist_lock is unlocked and we have a final result.
			 */
			retval = tsk_result;
			goto end;
L
Linus Torvalds 已提交
1706
		}
R
Roland McGrath 已提交
1707

L
Linus Torvalds 已提交
1708 1709 1710
		if (options & __WNOTHREAD)
			break;
		tsk = next_thread(tsk);
E
Eric Sesterhenn 已提交
1711
		BUG_ON(tsk->signal != current->signal);
L
Linus Torvalds 已提交
1712 1713
	} while (tsk != current);
	read_unlock(&tasklist_lock);
O
Oleg Nesterov 已提交
1714

R
Roland McGrath 已提交
1715
	if (!retval && !(options & WNOHANG)) {
L
Linus Torvalds 已提交
1716
		retval = -ERESTARTSYS;
R
Roland McGrath 已提交
1717 1718 1719 1720
		if (!signal_pending(current)) {
			schedule();
			goto repeat;
		}
L
Linus Torvalds 已提交
1721
	}
R
Roland McGrath 已提交
1722

L
Linus Torvalds 已提交
1723 1724 1725 1726 1727
end:
	current->state = TASK_RUNNING;
	remove_wait_queue(&current->signal->wait_chldexit,&wait);
	if (infop) {
		if (retval > 0)
1728
			retval = 0;
L
Linus Torvalds 已提交
1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751
		else {
			/*
			 * For a WNOHANG return, clear out all the fields
			 * we would set so the user can easily tell the
			 * difference.
			 */
			if (!retval)
				retval = put_user(0, &infop->si_signo);
			if (!retval)
				retval = put_user(0, &infop->si_errno);
			if (!retval)
				retval = put_user(0, &infop->si_code);
			if (!retval)
				retval = put_user(0, &infop->si_pid);
			if (!retval)
				retval = put_user(0, &infop->si_uid);
			if (!retval)
				retval = put_user(0, &infop->si_status);
		}
	}
	return retval;
}

E
Eric W. Biederman 已提交
1752
asmlinkage long sys_waitid(int which, pid_t upid,
L
Linus Torvalds 已提交
1753 1754 1755
			   struct siginfo __user *infop, int options,
			   struct rusage __user *ru)
{
E
Eric W. Biederman 已提交
1756 1757
	struct pid *pid = NULL;
	enum pid_type type;
L
Linus Torvalds 已提交
1758 1759 1760 1761 1762 1763 1764 1765 1766
	long ret;

	if (options & ~(WNOHANG|WNOWAIT|WEXITED|WSTOPPED|WCONTINUED))
		return -EINVAL;
	if (!(options & (WEXITED|WSTOPPED|WCONTINUED)))
		return -EINVAL;

	switch (which) {
	case P_ALL:
E
Eric W. Biederman 已提交
1767
		type = PIDTYPE_MAX;
L
Linus Torvalds 已提交
1768 1769
		break;
	case P_PID:
E
Eric W. Biederman 已提交
1770 1771
		type = PIDTYPE_PID;
		if (upid <= 0)
L
Linus Torvalds 已提交
1772 1773 1774
			return -EINVAL;
		break;
	case P_PGID:
E
Eric W. Biederman 已提交
1775 1776
		type = PIDTYPE_PGID;
		if (upid <= 0)
L
Linus Torvalds 已提交
1777 1778 1779 1780 1781 1782
			return -EINVAL;
		break;
	default:
		return -EINVAL;
	}

E
Eric W. Biederman 已提交
1783 1784 1785 1786
	if (type < PIDTYPE_MAX)
		pid = find_get_pid(upid);
	ret = do_wait(type, pid, options, infop, NULL, ru);
	put_pid(pid);
L
Linus Torvalds 已提交
1787 1788

	/* avoid REGPARM breakage on x86: */
1789
	asmlinkage_protect(5, ret, which, upid, infop, options, ru);
L
Linus Torvalds 已提交
1790 1791 1792
	return ret;
}

E
Eric W. Biederman 已提交
1793
asmlinkage long sys_wait4(pid_t upid, int __user *stat_addr,
L
Linus Torvalds 已提交
1794 1795
			  int options, struct rusage __user *ru)
{
E
Eric W. Biederman 已提交
1796 1797
	struct pid *pid = NULL;
	enum pid_type type;
L
Linus Torvalds 已提交
1798 1799 1800 1801 1802
	long ret;

	if (options & ~(WNOHANG|WUNTRACED|WCONTINUED|
			__WNOTHREAD|__WCLONE|__WALL))
		return -EINVAL;
E
Eric W. Biederman 已提交
1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818

	if (upid == -1)
		type = PIDTYPE_MAX;
	else if (upid < 0) {
		type = PIDTYPE_PGID;
		pid = find_get_pid(-upid);
	} else if (upid == 0) {
		type = PIDTYPE_PGID;
		pid = get_pid(task_pgrp(current));
	} else /* upid > 0 */ {
		type = PIDTYPE_PID;
		pid = find_get_pid(upid);
	}

	ret = do_wait(type, pid, options | WEXITED, NULL, stat_addr, ru);
	put_pid(pid);
L
Linus Torvalds 已提交
1819 1820

	/* avoid REGPARM breakage on x86: */
1821
	asmlinkage_protect(4, ret, upid, stat_addr, options, ru);
L
Linus Torvalds 已提交
1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836
	return ret;
}

#ifdef __ARCH_WANT_SYS_WAITPID

/*
 * sys_waitpid() remains for compatibility. waitpid() should be
 * implemented by calling sys_wait4() from libc.a.
 */
asmlinkage long sys_waitpid(pid_t pid, int __user *stat_addr, int options)
{
	return sys_wait4(pid, stat_addr, options, NULL);
}

#endif