exit.c 43.1 KB
Newer Older
Linus Torvalds's avatar
Linus Torvalds committed
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>
Linus Torvalds's avatar
Linus Torvalds committed
12 13 14
#include <linux/completion.h>
#include <linux/personality.h>
#include <linux/tty.h>
15
#include <linux/iocontext.h>
Linus Torvalds's avatar
Linus Torvalds committed
16 17 18 19
#include <linux/key.h>
#include <linux/security.h>
#include <linux/cpu.h>
#include <linux/acct.h>
20
#include <linux/tsacct_kern.h>
Linus Torvalds's avatar
Linus Torvalds committed
21
#include <linux/file.h>
Al Viro's avatar
Al Viro committed
22
#include <linux/fdtable.h>
23
#include <linux/freezer.h>
Linus Torvalds's avatar
Linus Torvalds committed
24
#include <linux/binfmts.h>
25
#include <linux/nsproxy.h>
26
#include <linux/pid_namespace.h>
Linus Torvalds's avatar
Linus Torvalds committed
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>
Linus Torvalds's avatar
Linus Torvalds committed
32
#include <linux/mempolicy.h>
33
#include <linux/taskstats_kern.h>
34
#include <linux/delayacct.h>
35
#include <linux/cgroup.h>
Linus Torvalds's avatar
Linus Torvalds committed
36
#include <linux/syscalls.h>
37
#include <linux/signal.h>
38
#include <linux/posix-timers.h>
39
#include <linux/cn_proc.h>
40
#include <linux/mutex.h>
41
#include <linux/futex.h>
42
#include <linux/pipe_fs_i.h>
43
#include <linux/audit.h> /* for audit_free() */
44
#include <linux/resource.h>
45
#include <linux/blkdev.h>
46
#include <linux/task_io_accounting_ops.h>
Roland McGrath's avatar
Roland McGrath committed
47
#include <linux/tracehook.h>
48
#include <linux/fs_struct.h>
49
#include <linux/init_task.h>
50
#include <linux/perf_event.h>
51
#include <trace/events/sched.h>
52
#include <linux/hw_breakpoint.h>
53
#include <linux/oom.h>
54
#include <linux/writeback.h>
Al Viro's avatar
Al Viro committed
55
#include <linux/shm.h>
Linus Torvalds's avatar
Linus Torvalds committed
56 57 58 59 60 61

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

62 63
static void exit_mm(struct task_struct * tsk);

64
static void __unhash_process(struct task_struct *p, bool group_dead)
Linus Torvalds's avatar
Linus Torvalds committed
65 66
{
	nr_threads--;
67
	detach_pid(p, PIDTYPE_PID);
68
	if (group_dead) {
Linus Torvalds's avatar
Linus Torvalds committed
69 70
		detach_pid(p, PIDTYPE_PGID);
		detach_pid(p, PIDTYPE_SID);
71

72
		list_del_rcu(&p->tasks);
73
		list_del_init(&p->sibling);
74
		__this_cpu_dec(process_counts);
Linus Torvalds's avatar
Linus Torvalds committed
75
	}
76
	list_del_rcu(&p->thread_group);
77
	list_del_rcu(&p->thread_node);
Linus Torvalds's avatar
Linus Torvalds committed
78 79
}

80 81 82 83 84 85
/*
 * This function expects the tasklist_lock write-locked.
 */
static void __exit_signal(struct task_struct *tsk)
{
	struct signal_struct *sig = tsk->signal;
86
	bool group_dead = thread_group_leader(tsk);
87
	struct sighand_struct *sighand;
88
	struct tty_struct *uninitialized_var(tty);
89
	cputime_t utime, stime;
90

91
	sighand = rcu_dereference_check(tsk->sighand,
92
					lockdep_tasklist_lock_is_held());
93 94 95
	spin_lock(&sighand->siglock);

	posix_cpu_timers_exit(tsk);
96
	if (group_dead) {
97
		posix_cpu_timers_exit_group(tsk);
98 99
		tty = sig->tty;
		sig->tty = NULL;
100
	} else {
101 102 103 104 105 106 107 108
		/*
		 * This can only happen if the caller is de_thread().
		 * FIXME: this is the temporary hack, we should teach
		 * posix-cpu-timers to handle this case correctly.
		 */
		if (unlikely(has_group_leader_pid(tsk)))
			posix_cpu_timers_exit_group(tsk);

109 110 111 112
		/*
		 * If there is any task waiting for the group exit
		 * then notify it:
		 */
113
		if (sig->notify_count > 0 && !--sig->notify_count)
114
			wake_up_process(sig->group_exit_task);
115

116 117 118 119 120 121 122 123 124 125 126 127
		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.
		 */
128 129 130 131
		task_cputime(tsk, &utime, &stime);
		sig->utime += utime;
		sig->stime += stime;
		sig->gtime += task_gtime(tsk);
132 133 134 135
		sig->min_flt += tsk->min_flt;
		sig->maj_flt += tsk->maj_flt;
		sig->nvcsw += tsk->nvcsw;
		sig->nivcsw += tsk->nivcsw;
136 137
		sig->inblock += task_io_get_inblock(tsk);
		sig->oublock += task_io_get_oublock(tsk);
138
		task_io_accounting_add(&sig->ioac, &tsk->ioac);
139
		sig->sum_sched_runtime += tsk->se.sum_exec_runtime;
140 141
	}

142
	sig->nr_threads--;
143
	__unhash_process(tsk, group_dead);
144

145 146 147 148 149
	/*
	 * Do this under ->siglock, we can race with another thread
	 * doing sigqueue_free() if we have SIGQUEUE_PREALLOC signals.
	 */
	flush_sigqueue(&tsk->pending);
150
	tsk->sighand = NULL;
151 152
	spin_unlock(&sighand->siglock);

153
	__cleanup_sighand(sighand);
154
	clear_tsk_thread_flag(tsk,TIF_SIGPENDING);
155
	if (group_dead) {
156
		flush_sigqueue(&sig->shared_pending);
157
		tty_kref_put(tty);
158 159 160
	}
}

161 162
static void delayed_put_task_struct(struct rcu_head *rhp)
{
163 164
	struct task_struct *tsk = container_of(rhp, struct task_struct, rcu);

165
	perf_event_delayed_put(tsk);
166 167
	trace_sched_process_free(tsk);
	put_task_struct(tsk);
168 169
}

Roland McGrath's avatar
Roland McGrath committed
170

Linus Torvalds's avatar
Linus Torvalds committed
171 172
void release_task(struct task_struct * p)
{
173
	struct task_struct *leader;
Linus Torvalds's avatar
Linus Torvalds committed
174
	int zap_leader;
175
repeat:
176
	/* don't need to get the RCU readlock here - the process is dead and
177 178
	 * can't be modifying its own credentials. But shut RCU-lockdep up */
	rcu_read_lock();
179
	atomic_dec(&__task_cred(p)->user->processes);
180
	rcu_read_unlock();
181

182
	proc_flush_task(p);
183

Linus Torvalds's avatar
Linus Torvalds committed
184
	write_lock_irq(&tasklist_lock);
185
	ptrace_release_task(p);
Linus Torvalds's avatar
Linus Torvalds committed
186
	__exit_signal(p);
187

Linus Torvalds's avatar
Linus Torvalds committed
188 189 190 191 192 193 194 195 196 197 198 199
	/*
	 * 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) {
		/*
		 * 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.
Roland McGrath's avatar
Roland McGrath committed
200
		 */
201
		zap_leader = do_notify_parent(leader, leader->exit_signal);
Roland McGrath's avatar
Roland McGrath committed
202 203
		if (zap_leader)
			leader->exit_state = EXIT_DEAD;
Linus Torvalds's avatar
Linus Torvalds committed
204 205 206 207
	}

	write_unlock_irq(&tasklist_lock);
	release_thread(p);
208
	call_rcu(&p->rcu, delayed_put_task_struct);
Linus Torvalds's avatar
Linus Torvalds committed
209 210 211 212 213 214 215 216 217 218

	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...
219 220
 *
 * The caller must hold rcu lock or the tasklist lock.
Linus Torvalds's avatar
Linus Torvalds committed
221
 */
222
struct pid *session_of_pgrp(struct pid *pgrp)
Linus Torvalds's avatar
Linus Torvalds committed
223 224
{
	struct task_struct *p;
225
	struct pid *sid = NULL;
226

227
	p = pid_task(pgrp, PIDTYPE_PGID);
228
	if (p == NULL)
229
		p = pid_task(pgrp, PIDTYPE_PID);
230
	if (p != NULL)
231
		sid = task_session(p);
232

Linus Torvalds's avatar
Linus Torvalds committed
233 234 235 236 237 238 239 240 241 242 243
	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?"
 */
244
static int will_become_orphaned_pgrp(struct pid *pgrp, struct task_struct *ignored_task)
Linus Torvalds's avatar
Linus Torvalds committed
245 246 247
{
	struct task_struct *p;

248
	do_each_pid_task(pgrp, PIDTYPE_PGID, p) {
249 250 251
		if ((p == ignored_task) ||
		    (p->exit_state && thread_group_empty(p)) ||
		    is_global_init(p->real_parent))
Linus Torvalds's avatar
Linus Torvalds committed
252
			continue;
253

254
		if (task_pgrp(p->real_parent) != pgrp &&
255 256
		    task_session(p->real_parent) == task_session(p))
			return 0;
257
	} while_each_pid_task(pgrp, PIDTYPE_PGID, p);
258 259

	return 1;
Linus Torvalds's avatar
Linus Torvalds committed
260 261
}

262
int is_current_pgrp_orphaned(void)
Linus Torvalds's avatar
Linus Torvalds committed
263 264 265 266
{
	int retval;

	read_lock(&tasklist_lock);
267
	retval = will_become_orphaned_pgrp(task_pgrp(current), NULL);
Linus Torvalds's avatar
Linus Torvalds committed
268 269 270 271 272
	read_unlock(&tasklist_lock);

	return retval;
}

273
static bool has_stopped_jobs(struct pid *pgrp)
Linus Torvalds's avatar
Linus Torvalds committed
274 275 276
{
	struct task_struct *p;

277
	do_each_pid_task(pgrp, PIDTYPE_PGID, p) {
278 279
		if (p->signal->flags & SIGNAL_STOP_STOPPED)
			return true;
280
	} while_each_pid_task(pgrp, PIDTYPE_PGID, p);
281 282

	return false;
Linus Torvalds's avatar
Linus Torvalds committed
283 284
}

285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315
/*
 * 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);
	}
}

316
void __set_special_pids(struct pid *pid)
Linus Torvalds's avatar
Linus Torvalds committed
317
{
318
	struct task_struct *curr = current->group_leader;
Linus Torvalds's avatar
Linus Torvalds committed
319

320
	if (task_session(curr) != pid)
321
		change_pid(curr, PIDTYPE_SID, pid);
322 323

	if (task_pgrp(curr) != pid)
324
		change_pid(curr, PIDTYPE_PGID, pid);
Linus Torvalds's avatar
Linus Torvalds committed
325 326 327
}

/*
328 329
 * Let kernel threads use this to say that they allow a certain signal.
 * Must not be used if kthread was cloned with CLONE_SIGHAND.
Linus Torvalds's avatar
Linus Torvalds committed
330 331 332
 */
int allow_signal(int sig)
{
333
	if (!valid_signal(sig) || sig < 1)
Linus Torvalds's avatar
Linus Torvalds committed
334 335 336
		return -EINVAL;

	spin_lock_irq(&current->sighand->siglock);
337
	/* This is only needed for daemonize()'ed kthreads */
Linus Torvalds's avatar
Linus Torvalds committed
338
	sigdelset(&current->blocked, sig);
339 340 341 342 343 344
	/*
	 * 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;
Linus Torvalds's avatar
Linus Torvalds committed
345 346 347 348 349 350 351 352 353
	recalc_sigpending();
	spin_unlock_irq(&current->sighand->siglock);
	return 0;
}

EXPORT_SYMBOL(allow_signal);

int disallow_signal(int sig)
{
354
	if (!valid_signal(sig) || sig < 1)
Linus Torvalds's avatar
Linus Torvalds committed
355 356 357
		return -EINVAL;

	spin_lock_irq(&current->sighand->siglock);
358
	current->sighand->action[(sig)-1].sa.sa_handler = SIG_IGN;
Linus Torvalds's avatar
Linus Torvalds committed
359 360 361 362 363 364 365
	recalc_sigpending();
	spin_unlock_irq(&current->sighand->siglock);
	return 0;
}

EXPORT_SYMBOL(disallow_signal);

366 367
#ifdef CONFIG_MM_OWNER
/*
368
 * A task is exiting.   If it owned this mm, find a new owner for the mm.
369 370 371 372 373 374
 */
void mm_update_next_owner(struct mm_struct *mm)
{
	struct task_struct *c, *g, *p = current;

retry:
375 376 377 378 379
	/*
	 * If the exiting or execing task is not the owner, it's
	 * someone else's problem.
	 */
	if (mm->owner != p)
380
		return;
381 382 383 384 385 386 387 388 389
	/*
	 * The current owner is exiting/execing and there are no other
	 * candidates.  Do not leave the mm pointing to a possibly
	 * freed task structure.
	 */
	if (atomic_read(&mm->mm_users) <= 1) {
		mm->owner = NULL;
		return;
	}
390 391 392 393 394 395 396 397 398 399 400 401 402

	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
	 */
403
	list_for_each_entry(c, &p->real_parent->children, sibling) {
404 405 406 407 408 409 410 411 412 413 414 415 416 417
		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);
418 419 420
	/*
	 * We found no owner yet mm_users > 1: this implies that we are
	 * most likely racing with swapoff (try_to_unuse()) or /proc or
421
	 * ptrace or page migration (get_task_mm()).  Mark owner as NULL.
422 423
	 */
	mm->owner = NULL;
424 425 426 427 428 429 430 431 432 433
	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);
434 435 436 437 438
	/*
	 * Delay read_unlock() till we have the task_lock()
	 * to ensure that c does not slip away underneath us
	 */
	read_unlock(&tasklist_lock);
439 440 441 442 443 444 445 446 447 448 449
	if (c->mm != mm) {
		task_unlock(c);
		put_task_struct(c);
		goto retry;
	}
	mm->owner = c;
	task_unlock(c);
	put_task_struct(c);
}
#endif /* CONFIG_MM_OWNER */

Linus Torvalds's avatar
Linus Torvalds committed
450 451 452 453
/*
 * Turn us into a lazy TLB process if we
 * aren't already..
 */
454
static void exit_mm(struct task_struct * tsk)
Linus Torvalds's avatar
Linus Torvalds committed
455 456
{
	struct mm_struct *mm = tsk->mm;
457
	struct core_state *core_state;
Linus Torvalds's avatar
Linus Torvalds committed
458

459
	mm_release(tsk, mm);
Linus Torvalds's avatar
Linus Torvalds committed
460 461
	if (!mm)
		return;
462
	sync_mm_rss(mm);
Linus Torvalds's avatar
Linus Torvalds committed
463 464
	/*
	 * Serialize with any possible pending coredump.
465
	 * We must hold mmap_sem around checking core_state
Linus Torvalds's avatar
Linus Torvalds committed
466
	 * and clearing tsk->mm.  The core-inducing thread
467
	 * will increment ->nr_threads for each thread in the
Linus Torvalds's avatar
Linus Torvalds committed
468 469 470
	 * group with ->mm != NULL.
	 */
	down_read(&mm->mmap_sem);
471 472 473
	core_state = mm->core_state;
	if (core_state) {
		struct core_thread self;
Linus Torvalds's avatar
Linus Torvalds committed
474 475
		up_read(&mm->mmap_sem);

476 477 478 479 480 481 482 483
		self.task = tsk;
		self.next = xchg(&core_state->dumper.next, &self);
		/*
		 * Implies mb(), the result of xchg() must be visible
		 * to core_state->dumper.
		 */
		if (atomic_dec_and_test(&core_state->nr_threads))
			complete(&core_state->startup);
Linus Torvalds's avatar
Linus Torvalds committed
484

485 486 487 488
		for (;;) {
			set_task_state(tsk, TASK_UNINTERRUPTIBLE);
			if (!self.task) /* see coredump_finish() */
				break;
489
			freezable_schedule();
490 491
		}
		__set_task_state(tsk, TASK_RUNNING);
Linus Torvalds's avatar
Linus Torvalds committed
492 493 494
		down_read(&mm->mmap_sem);
	}
	atomic_inc(&mm->mm_count);
495
	BUG_ON(mm != tsk->active_mm);
Linus Torvalds's avatar
Linus Torvalds committed
496 497 498 499 500 501
	/* more a memory barrier than a real lock */
	task_lock(tsk);
	tsk->mm = NULL;
	up_read(&mm->mmap_sem);
	enter_lazy_tlb(mm, current);
	task_unlock(tsk);
502
	mm_update_next_owner(mm);
Linus Torvalds's avatar
Linus Torvalds committed
503 504 505 506
	mmput(mm);
}

/*
507 508 509 510 511
 * When we die, we re-parent all our children, and try to:
 * 1. give them to another thread in our thread group, if such a member exists
 * 2. give it to the first ancestor process which prctl'd itself as a
 *    child_subreaper for its children (like a service manager)
 * 3. give it to the init process (PID 1) in our pid namespace
Linus Torvalds's avatar
Linus Torvalds committed
512
 */
513
static struct task_struct *find_new_reaper(struct task_struct *father)
514 515
	__releases(&tasklist_lock)
	__acquires(&tasklist_lock)
Linus Torvalds's avatar
Linus Torvalds committed
516
{
517 518
	struct pid_namespace *pid_ns = task_active_pid_ns(father);
	struct task_struct *thread;
Linus Torvalds's avatar
Linus Torvalds committed
519

520 521 522 523 524 525 526 527
	thread = father;
	while_each_thread(father, thread) {
		if (thread->flags & PF_EXITING)
			continue;
		if (unlikely(pid_ns->child_reaper == father))
			pid_ns->child_reaper = thread;
		return thread;
	}
Linus Torvalds's avatar
Linus Torvalds committed
528

529 530
	if (unlikely(pid_ns->child_reaper == father)) {
		write_unlock_irq(&tasklist_lock);
531 532 533 534 535
		if (unlikely(pid_ns == &init_pid_ns)) {
			panic("Attempted to kill init! exitcode=0x%08x\n",
				father->signal->group_exit_code ?:
					father->exit_code);
		}
Linus Torvalds's avatar
Linus Torvalds committed
536

537 538
		zap_pid_ns_processes(pid_ns);
		write_lock_irq(&tasklist_lock);
539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561
	} else if (father->signal->has_child_subreaper) {
		struct task_struct *reaper;

		/*
		 * Find the first ancestor marked as child_subreaper.
		 * Note that the code below checks same_thread_group(reaper,
		 * pid_ns->child_reaper).  This is what we need to DTRT in a
		 * PID namespace. However we still need the check above, see
		 * http://marc.info/?l=linux-kernel&m=131385460420380
		 */
		for (reaper = father->real_parent;
		     reaper != &init_task;
		     reaper = reaper->real_parent) {
			if (same_thread_group(reaper, pid_ns->child_reaper))
				break;
			if (!reaper->signal->is_child_subreaper)
				continue;
			thread = reaper;
			do {
				if (!(thread->flags & PF_EXITING))
					return reaper;
			} while_each_thread(reaper, thread);
		}
Linus Torvalds's avatar
Linus Torvalds committed
562
	}
563

564 565 566
	return pid_ns->child_reaper;
}

567 568 569
/*
* Any that need to be release_task'd are put on the @dead list.
 */
570
static void reparent_leader(struct task_struct *father, struct task_struct *p,
571 572 573 574 575 576 577 578 579 580
				struct list_head *dead)
{
	list_move_tail(&p->sibling, &p->real_parent->children);
	/*
	 * If this is a threaded reparent there is no need to
	 * notify anyone anything has happened.
	 */
	if (same_thread_group(p->real_parent, father))
		return;

581 582 583 584 585 586 587 588
	/*
	 * We don't want people slaying init.
	 *
	 * Note: we do this even if it is EXIT_DEAD, wait_task_zombie()
	 * can change ->exit_state to EXIT_ZOMBIE. If this is the final
	 * state, do_notify_parent() was already called and ->exit_signal
	 * doesn't matter.
	 */
589 590
	p->exit_signal = SIGCHLD;

591 592 593
	if (p->exit_state == EXIT_DEAD)
		return;

594
	/* If it has exited notify the new parent about this child's death. */
Tejun Heo's avatar
Tejun Heo committed
595
	if (!p->ptrace &&
596
	    p->exit_state == EXIT_ZOMBIE && thread_group_empty(p)) {
597
		if (do_notify_parent(p, p->exit_signal)) {
598 599 600 601 602 603 604 605
			p->exit_state = EXIT_DEAD;
			list_move_tail(&p->sibling, dead);
		}
	}

	kill_orphaned_pgrp(p, father);
}

606
static void forget_original_parent(struct task_struct *father)
Linus Torvalds's avatar
Linus Torvalds committed
607
{
608
	struct task_struct *p, *n, *reaper;
609
	LIST_HEAD(dead_children);
610 611

	write_lock_irq(&tasklist_lock);
612 613 614 615 616
	/*
	 * Note that exit_ptrace() and find_new_reaper() might
	 * drop tasklist_lock and reacquire it.
	 */
	exit_ptrace(father);
617
	reaper = find_new_reaper(father);
Roland McGrath's avatar
Roland McGrath committed
618

619
	list_for_each_entry_safe(p, n, &father->children, sibling) {
620 621 622 623
		struct task_struct *t = p;
		do {
			t->real_parent = reaper;
			if (t->parent == father) {
Tejun Heo's avatar
Tejun Heo committed
624
				BUG_ON(t->ptrace);
625 626 627 628 629 630 631
				t->parent = t->real_parent;
			}
			if (t->pdeath_signal)
				group_send_sig_info(t->pdeath_signal,
						    SEND_SIG_NOINFO, t);
		} while_each_thread(p, t);
		reparent_leader(father, p, &dead_children);
Linus Torvalds's avatar
Linus Torvalds committed
632
	}
633
	write_unlock_irq(&tasklist_lock);
634

635 636
	BUG_ON(!list_empty(&father->children));

637 638
	list_for_each_entry_safe(p, n, &dead_children, sibling) {
		list_del_init(&p->sibling);
639 640
		release_task(p);
	}
Linus Torvalds's avatar
Linus Torvalds committed
641 642 643 644 645 646
}

/*
 * Send signals to all our closest relatives so that they know
 * to properly mourn us..
 */
647
static void exit_notify(struct task_struct *tsk, int group_dead)
Linus Torvalds's avatar
Linus Torvalds committed
648
{
649
	bool autoreap;
Linus Torvalds's avatar
Linus Torvalds committed
650 651 652 653 654 655 656 657 658

	/*
	 * 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)
	 */
659
	forget_original_parent(tsk);
Linus Torvalds's avatar
Linus Torvalds committed
660

661
	write_lock_irq(&tasklist_lock);
662 663
	if (group_dead)
		kill_orphaned_pgrp(tsk->group_leader, NULL);
Linus Torvalds's avatar
Linus Torvalds committed
664

665 666 667 668 669 670 671 672 673 674 675 676
	if (unlikely(tsk->ptrace)) {
		int sig = thread_group_leader(tsk) &&
				thread_group_empty(tsk) &&
				!ptrace_reparented(tsk) ?
			tsk->exit_signal : SIGCHLD;
		autoreap = do_notify_parent(tsk, sig);
	} else if (thread_group_leader(tsk)) {
		autoreap = thread_group_empty(tsk) &&
			do_notify_parent(tsk, tsk->exit_signal);
	} else {
		autoreap = true;
	}
Linus Torvalds's avatar
Linus Torvalds committed
677

678
	tsk->exit_state = autoreap ? EXIT_DEAD : EXIT_ZOMBIE;
Linus Torvalds's avatar
Linus Torvalds committed
679

680 681
	/* mt-exec, de_thread() is waiting for group leader */
	if (unlikely(tsk->signal->notify_count < 0))
682
		wake_up_process(tsk->signal->group_exit_task);
Linus Torvalds's avatar
Linus Torvalds committed
683 684 685
	write_unlock_irq(&tasklist_lock);

	/* If the process is dead, release it - nobody will wait for it */
686
	if (autoreap)
Linus Torvalds's avatar
Linus Torvalds committed
687 688 689
		release_task(tsk);
}

690 691 692 693 694 695 696
#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 free;

697
	free = stack_not_used(current);
698 699 700 701 702 703

	if (free >= lowest_to_date)
		return;

	spin_lock(&low_water_lock);
	if (free < lowest_to_date) {
704 705 706
		printk(KERN_WARNING "%s (%d) used greatest stack depth: "
				"%lu bytes left\n",
				current->comm, task_pid_nr(current), free);
707 708 709 710 711 712 713 714
		lowest_to_date = free;
	}
	spin_unlock(&low_water_lock);
}
#else
static inline void check_stack_usage(void) {}
#endif

715
void do_exit(long code)
Linus Torvalds's avatar
Linus Torvalds committed
716 717 718 719 720 721
{
	struct task_struct *tsk = current;
	int group_dead;

	profile_task_exit(tsk);

722
	WARN_ON(blk_needs_flush_plug(tsk));
723

Linus Torvalds's avatar
Linus Torvalds committed
724 725 726 727 728
	if (unlikely(in_interrupt()))
		panic("Aiee, killing interrupt handler!");
	if (unlikely(!tsk->pid))
		panic("Attempted to kill the idle task!");

729 730 731 732 733 734 735 736 737
	/*
	 * If do_exit is called because this processes oopsed, it's possible
	 * that get_fs() was left as KERNEL_DS, so reset it to USER_DS before
	 * continuing. Amongst other possible reasons, this is to prevent
	 * mm_release()->clear_child_tid() from writing to a user-controlled
	 * kernel address.
	 */
	set_fs(USER_DS);

738
	ptrace_event(PTRACE_EVENT_EXIT, code);
Linus Torvalds's avatar
Linus Torvalds committed
739

740 741
	validate_creds_for_do_exit(tsk);

742 743 744 745 746 747 748
	/*
	 * 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");
749 750 751 752 753 754 755 756 757 758
		/*
		 * 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;
759 760 761 762
		set_current_state(TASK_UNINTERRUPTIBLE);
		schedule();
	}

763
	exit_signals(tsk);  /* sets PF_EXITING */
764 765
	/*
	 * tsk->flags are checked in the futex code to protect against
766
	 * an exiting task cleaning up the robust pi futexes.
767
	 */
768
	smp_mb();
769
	raw_spin_unlock_wait(&tsk->pi_lock);
Linus Torvalds's avatar
Linus Torvalds committed
770 771 772

	if (unlikely(in_atomic()))
		printk(KERN_INFO "note: %s[%d] exited with preempt_count %d\n",
773
				current->comm, task_pid_nr(current),
Linus Torvalds's avatar
Linus Torvalds committed
774 775 776
				preempt_count());

	acct_update_integrals(tsk);
777 778 779
	/* sync mm's RSS info before statistics gathering */
	if (tsk->mm)
		sync_mm_rss(tsk->mm);
Linus Torvalds's avatar
Linus Torvalds committed
780
	group_dead = atomic_dec_and_test(&tsk->signal->live);
781
	if (group_dead) {
782
		hrtimer_cancel(&tsk->signal->real_timer);
783
		exit_itimers(tsk->signal);
784 785
		if (tsk->mm)
			setmax_mm_hiwater_rss(&tsk->signal->maxrss, tsk->mm);
786
	}
787
	acct_collect(code, group_dead);
788 789
	if (group_dead)
		tty_audit_exit();
790
	audit_free(tsk);
791

792
	tsk->exit_code = code;
793
	taskstats_exit(tsk, group_dead);
794

Linus Torvalds's avatar
Linus Torvalds committed
795 796
	exit_mm(tsk);

797
	if (group_dead)
798
		acct_process();
799 800
	trace_sched_process_exit(tsk);

Linus Torvalds's avatar
Linus Torvalds committed
801
	exit_sem(tsk);
802
	exit_shm(tsk);
803 804
	exit_files(tsk);
	exit_fs(tsk);
805 806
	if (group_dead)
		disassociate_ctty(1);
807
	exit_task_namespaces(tsk);
808
	exit_task_work(tsk);
809
	check_stack_usage();
Linus Torvalds's avatar
Linus Torvalds committed
810
	exit_thread();
811 812 813 814 815 816 817 818 819

	/*
	 * Flush inherited counters to the parent - before the parent
	 * gets woken up by child-exit notifications.
	 *
	 * because of cgroup mode, must be called before cgroup_exit()
	 */
	perf_event_exit_task(tsk);

820
	cgroup_exit(tsk, 1);
Linus Torvalds's avatar
Linus Torvalds committed
821

822
	module_put(task_thread_info(tsk)->exec_domain->module);
Linus Torvalds's avatar
Linus Torvalds committed
823

824
	proc_exit_connector(tsk);
825 826 827
	/*
	 * FIXME: do that only when needed, using sched_exit tracepoint
	 */
828
	ptrace_put_breakpoints(tsk);
829

830
	exit_notify(tsk, group_dead);
Linus Torvalds's avatar
Linus Torvalds committed
831
#ifdef CONFIG_NUMA
832
	task_lock(tsk);
833
	mpol_put(tsk->mempolicy);
Linus Torvalds's avatar
Linus Torvalds committed
834
	tsk->mempolicy = NULL;
835
	task_unlock(tsk);
Linus Torvalds's avatar
Linus Torvalds committed
836
#endif
837
#ifdef CONFIG_FUTEX
838 839
	if (unlikely(current->pi_state_cache))
		kfree(current->pi_state_cache);
840
#endif
841
	/*
842
	 * Make sure we are holding no locks:
843
	 */
Mauro Ribeiro's avatar
Mauro Ribeiro committed
844
	debug_check_no_locks_held();
845 846 847 848 849 850
	/*
	 * 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;
Linus Torvalds's avatar
Linus Torvalds committed
851

852
	if (tsk->io_context)
853
		exit_io_context(tsk);
854

855
	if (tsk->splice_pipe)
856
		free_pipe_info(tsk->splice_pipe);
857

858 859 860
	if (tsk->task_frag.page)
		put_page(tsk->task_frag.page);

861 862
	validate_creds_for_do_exit(tsk);

863
	preempt_disable();
864 865
	if (tsk->nr_dirtied)
		__this_cpu_add(dirty_throttle_leaks, tsk->nr_dirtied);
866
	exit_rcu();
867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882

	/*
	 * The setting of TASK_RUNNING by try_to_wake_up() may be delayed
	 * when the following two conditions become true.
	 *   - There is race condition of mmap_sem (It is acquired by
	 *     exit_mm()), and
	 *   - SMI occurs before setting TASK_RUNINNG.
	 *     (or hypervisor of virtual machine switches to other guest)
	 *  As a result, we may become TASK_RUNNING after becoming TASK_DEAD
	 *
	 * To avoid it, we have to wait for releasing tsk->pi_lock which
	 * is held by try_to_wake_up()
	 */
	smp_mb();
	raw_spin_unlock_wait(&tsk->pi_lock);

883
	/* causes final put_task_struct in finish_task_switch(). */
884
	tsk->state = TASK_DEAD;
885
	tsk->flags |= PF_NOFREEZE;	/* tell freezer to ignore us */
Linus Torvalds's avatar
Linus Torvalds committed
886 887 888
	schedule();
	BUG();
	/* Avoid "noreturn function does return".  */
Alan Cox's avatar
Alan Cox committed
889 890
	for (;;)
		cpu_relax();	/* For when BUG is null */
Linus Torvalds's avatar
Linus Torvalds committed
891 892
}

893 894
EXPORT_SYMBOL_GPL(do_exit);

895
void complete_and_exit(struct completion *comp, long code)
Linus Torvalds's avatar
Linus Torvalds committed
896 897 898
{
	if (comp)
		complete(comp);
899

Linus Torvalds's avatar
Linus Torvalds committed
900 901 902 903 904
	do_exit(code);
}

EXPORT_SYMBOL(complete_and_exit);

905
SYSCALL_DEFINE1(exit, int, error_code)
Linus Torvalds's avatar
Linus Torvalds committed
906 907 908 909 910 911 912 913
{
	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).
 */
914
void
Linus Torvalds's avatar
Linus Torvalds committed
915 916
do_group_exit(int exit_code)
{
917 918
	struct signal_struct *sig = current->signal;

Linus Torvalds's avatar
Linus Torvalds committed
919 920
	BUG_ON(exit_code & 0x80); /* core dumps don't get here */

921 922
	if (signal_group_exit(sig))
		exit_code = sig->group_exit_code;
Linus Torvalds's avatar
Linus Torvalds committed
923 924 925
	else if (!thread_group_empty(current)) {
		struct sighand_struct *const sighand = current->sighand;
		spin_lock_irq(&sighand->siglock);
926
		if (signal_group_exit(sig))
Linus Torvalds's avatar
Linus Torvalds committed
927 928 929 930
			/* Another thread got here before we took the lock.  */
			exit_code = sig->group_exit_code;
		else {
			sig->group_exit_code = exit_code;
931
			sig->flags = SIGNAL_GROUP_EXIT;
Linus Torvalds's avatar
Linus Torvalds committed
932 933 934 935 936 937 938 939 940 941 942 943 944 945
			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.
 */
946
SYSCALL_DEFINE1(exit_group, int, error_code)
Linus Torvalds's avatar
Linus Torvalds committed
947 948
{
	do_group_exit((error_code & 0xff) << 8);
949 950
	/* NOTREACHED */
	return 0;
Linus Torvalds's avatar
Linus Torvalds committed
951 952
}

953 954 955
struct wait_opts {
	enum pid_type		wo_type;
	int			wo_flags;
956
	struct pid		*wo_pid;
957 958 959 960 961

	struct siginfo __user	*wo_info;
	int __user		*wo_stat;
	struct rusage __user	*wo_rusage;

962
	wait_queue_t		child_wait;
963 964 965
	int			notask_error;
};

966 967
static inline
struct pid *task_pid_type(struct task_struct *task, enum pid_type type)
968
{
969 970 971
	if (type != PIDTYPE_PID)
		task = task->group_leader;
	return task->pids[type].pid;
972 973
}

974
static int eligible_pid(struct wait_opts *wo, struct task_struct *p)
Linus Torvalds's avatar
Linus Torvalds committed
975
{
976 977 978
	return	wo->wo_type == PIDTYPE_MAX ||
		task_pid_type(p, wo->wo_type) == wo->wo_pid;
}
Linus Torvalds's avatar
Linus Torvalds committed
979

980 981 982 983
static int eligible_child(struct wait_opts *wo, struct task_struct *p)
{
	if (!eligible_pid(wo, p))
		return 0;
Linus Torvalds's avatar
Linus Torvalds committed
984 985 986 987 988
	/* 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.) */
989 990
	if (((p->exit_signal != SIGCHLD) ^ !!(wo->wo_flags & __WCLONE))
	    && !(wo->wo_flags & __WALL))
Linus Torvalds's avatar
Linus Torvalds committed
991 992
		return 0;

993
	return 1;
Linus Torvalds's avatar
Linus Torvalds committed
994 995
}

996 997
static int wait_noreap_copyout(struct wait_opts *wo, struct task_struct *p,
				pid_t pid, uid_t uid, int why, int status)
Linus Torvalds's avatar
Linus Torvalds committed
998
{
999 1000 1001
	struct siginfo __user *infop;
	int retval = wo->wo_rusage
		? getrusage(p, RUSAGE_BOTH, wo->wo_rusage) : 0;
1002

Linus Torvalds's avatar
Linus Torvalds committed
1003
	put_task_struct(p);
1004
	infop = wo->wo_info;
1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018
	if (infop) {
		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);
	}
Linus Torvalds's avatar
Linus Torvalds committed
1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029
	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.
 */
1030
static int wait_task_zombie(struct wait_opts *wo, struct task_struct *p)
Linus Torvalds's avatar
Linus Torvalds committed
1031 1032
{
	unsigned long state;
1033
	int retval, status, traced;
1034
	pid_t pid = task_pid_vnr(p);
1035
	uid_t uid = from_kuid_munged(current_user_ns(), task_uid(p));
1036
	struct siginfo __user *infop;
Linus Torvalds's avatar
Linus Torvalds committed
1037

1038
	if (!likely(wo->wo_flags & WEXITED))
Roland McGrath's avatar
Roland McGrath committed
1039 1040
		return 0;

1041
	if (unlikely(wo->wo_flags & WNOWAIT)) {
Linus Torvalds's avatar
Linus Torvalds committed
1042
		int exit_code = p->exit_code;
1043
		int why;
Linus Torvalds's avatar
Linus Torvalds committed
1044 1045 1046 1047 1048 1049 1050 1051 1052 1053

		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;
		}
1054
		return wait_noreap_copyout(wo, p, pid, uid, why, status);
Linus Torvalds's avatar
Linus Torvalds committed
1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066
	}

	/*
	 * 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;
	}

1067
	traced = ptrace_reparented(p);
1068 1069
	/*
	 * It can be ptraced but not reparented, check
Oleg Nesterov's avatar
Oleg Nesterov committed
1070
	 * thread_group_leader() to filter out sub-threads.
1071
	 */
Oleg Nesterov's avatar
Oleg Nesterov committed
1072
	if (likely(!traced) && thread_group_leader(p)) {
1073 1074
		struct signal_struct *psig;
		struct signal_struct *sig;
1075
		unsigned long maxrss;
1076
		cputime_t tgutime, tgstime;
1077

Linus Torvalds's avatar
Linus Torvalds committed
1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088
		/*
		 * 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
1089
		 * need to protect the access to parent->signal fields,
Linus Torvalds's avatar
Linus Torvalds committed
1090 1091
		 * as other threads in the parent group can be right
		 * here reaping other children at the same time.
1092
		 *
1093
		 * We use thread_group_cputime_adjusted() to get times for the thread
1094 1095
		 * group, which consolidates times for all threads in the
		 * group including the group leader.
Linus Torvalds's avatar
Linus Torvalds committed
1096
		 */
1097
		thread_group_cputime_adjusted(p, &tgutime, &tgstime);
1098 1099
		spin_lock_irq(&p->real_parent->sighand->siglock);
		psig = p->real_parent->signal;
1100
		sig = p->signal;
1101 1102
		psig->cutime += tgutime + sig->cutime;
		psig->cstime += tgstime + sig->cstime;
1103
		psig->cgtime += task_gtime(p) + sig->gtime + sig->cgtime;
1104 1105 1106 1107 1108 1109 1110 1111
		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;
1112 1113 1114 1115 1116 1117
		psig->cinblock +=
			task_io_get_inblock(p) +
			sig->inblock + sig->cinblock;
		psig->coublock +=
			task_io_get_oublock(p) +
			sig->oublock + sig->coublock;
1118 1119 1120
		maxrss = max(sig->maxrss, sig->cmaxrss);
		if (psig->cmaxrss < maxrss)
			psig->cmaxrss = maxrss;
1121 1122
		task_io_accounting_add(&psig->ioac, &p->ioac);
		task_io_accounting_add(&psig->ioac, &sig->ioac);
1123
		spin_unlock_irq(&p->real_parent->sighand->siglock);
Linus Torvalds's avatar
Linus Torvalds committed
1124 1125 1126 1127 1128 1129 1130 1131
	}

	/*
	 * 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);

1132 1133
	retval = wo->wo_rusage
		? getrusage(p, RUSAGE_BOTH, wo->wo_rusage) : 0;
Linus Torvalds's avatar
Linus Torvalds committed
1134 1135
	status = (p->signal->flags & SIGNAL_GROUP_EXIT)
		? p->signal->group_exit_code : p->exit_code;
1136 1137 1138 1139
	if (!retval && wo->wo_stat)
		retval = put_user(status, wo->wo_stat);

	infop = wo->wo_info;
Linus Torvalds's avatar
Linus Torvalds committed
1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158
	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)
1159
		retval = put_user(pid, &infop->si_pid);
Linus Torvalds's avatar
Linus Torvalds committed
1160
	if (!retval && infop)
1161
		retval = put_user(uid, &infop->si_uid);
1162
	if (!retval)
1163
		retval = pid;
1164 1165

	if (traced) {
Linus Torvalds's avatar
Linus Torvalds committed
1166
		write_lock_irq(&tasklist_lock);
1167 1168 1169
		/* We dropped tasklist, ptracer could die and untrace */
		ptrace_unlink(p);
		/*
1170 1171
		 * If this is not a sub-thread, notify the parent.
		 * If parent wants a zombie, don't release it now.
1172
		 */
1173 1174 1175 1176
		if (thread_group_leader(p) &&
		    !do_notify_parent(p, p->exit_signal)) {
			p->exit_state = EXIT_ZOMBIE;
			p = NULL;
Linus Torvalds's avatar
Linus Torvalds committed
1177 1178 1179 1180 1181
		}
		write_unlock_irq(&tasklist_lock);
	}
	if (p != NULL)
		release_task(p);
1182

Linus Torvalds's avatar
Linus Torvalds committed
1183 1184 1185
	return retval;
}

1186 1187 1188
static int *task_stopped_code(struct task_struct *p, bool ptrace)
{
	if (ptrace) {
1189 1190
		if (task_is_stopped_or_traced(p) &&
		    !(p->jobctl & JOBCTL_LISTENING))
1191 1192 1193 1194 1195 1196 1197 1198
			return &p->exit_code;
	} else {
		if (p->signal->flags & SIGNAL_STOP_STOPPED)
			return &p->signal->group_exit_code;
	}
	return NULL;
}

1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215
/**
 * wait_task_stopped - Wait for %TASK_STOPPED or %TASK_TRACED
 * @wo: wait options
 * @ptrace: is the wait for ptrace
 * @p: task to wait for
 *
 * Handle sys_wait4() work for %p in state %TASK_STOPPED or %TASK_TRACED.
 *
 * CONTEXT:
 * read_lock(&tasklist_lock), which is released if return value is
 * non-zero.  Also, grabs and releases @p->sighand->siglock.
 *
 * RETURNS:
 * 0 if wait condition didn't exist and search for other wait conditions
 * should continue.  Non-zero return, -errno on failure and @p's pid on
 * success, implies that tasklist_lock is released and wait condition
 * search should terminate.
Linus Torvalds's avatar
Linus Torvalds committed
1216
 */
1217 1218
static int wait_task_stopped(struct wait_opts *wo,
				int ptrace, struct task_struct *p)
Linus Torvalds's avatar
Linus Torvalds committed
1219
{
1220
	struct siginfo __user *infop;
1221
	int retval, exit_code, *p_code, why;
1222
	uid_t uid = 0; /* unneeded, required by compiler */
1223
	pid_t pid;
Linus Torvalds's avatar
Linus Torvalds committed
1224

1225 1226 1227
	/*
	 * Traditionally we see ptrace'd stopped tasks regardless of options.
	 */
1228
	if (!ptrace && !(wo->wo_flags & WUNTRACED))
Roland McGrath's avatar
Roland McGrath committed
1229 1230
		return 0;

1231 1232 1233
	if (!task_stopped_code(p, ptrace))
		return 0;

1234 1235 1236
	exit_code = 0;
	spin_lock_irq(&p->sighand->siglock);

1237 1238
	p_code = task_stopped_code(p, ptrace);
	if (unlikely(!p_code))
1239 1240
		goto unlock_sig;

1241
	exit_code = *p_code;
1242 1243 1244
	if (!exit_code)
		goto unlock_sig;

1245
	if (!unlikely(wo->wo_flags & WNOWAIT))
1246
		*p_code = 0;
1247

1248
	uid = from_kuid_munged(current_user_ns(), task_uid(p));
1249 1250 1251
unlock_sig:
	spin_unlock_irq(&p->sighand->siglock);
	if (!exit_code)
Linus Torvalds's avatar
Linus Torvalds committed
1252 1253 1254 1255 1256 1257 1258 1259 1260 1261
		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);
1262
	pid = task_pid_vnr(p);
Roland McGrath's avatar
Roland McGrath committed
1263
	why = ptrace ? CLD_TRAPPED : CLD_STOPPED;
Linus Torvalds's avatar
Linus Torvalds committed
1264 1265
	read_unlock(&tasklist_lock);

1266 1267 1268 1269 1270 1271 1272
	if (unlikely(wo->wo_flags & WNOWAIT))
		return wait_noreap_copyout(wo, p, pid, uid, why, exit_code);

	retval = wo->wo_rusage
		? getrusage(p, RUSAGE_BOTH, wo->wo_rusage) : 0;
	if (!retval && wo->wo_stat)
		retval = put_user((exit_code << 8) | 0x7f, wo->wo_stat);
Linus Torvalds's avatar
Linus Torvalds committed
1273

1274
	infop = wo->wo_info;
Linus Torvalds's avatar
Linus Torvalds committed
1275 1276 1277 1278 1279
	if (!retval && infop)
		retval = put_user(SIGCHLD, &infop->si_signo);
	if (!retval && infop)
		retval = put_user(0, &infop->si_errno);
	if (!retval && infop)
1280
		retval = put_user((short)why, &infop->si_code);
Linus Torvalds's avatar
Linus Torvalds committed
1281 1282 1283
	if (!retval && infop)
		retval = put_user(exit_code, &infop->si_status);
	if (!retval && infop)
1284
		retval = put_user(pid, &infop->si_pid);
Linus Torvalds's avatar
Linus Torvalds committed
1285
	if (!retval && infop)
1286
		retval = put_user(uid, &infop->si_uid);
Linus Torvalds's avatar
Linus Torvalds committed
1287
	if (!retval)
1288
		retval = pid;
Linus Torvalds's avatar
Linus Torvalds committed
1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300
	put_task_struct(p);

	BUG_ON(!retval);
	return retval;
}

/*