exit.c 44.6 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>
L
Linus Torvalds 已提交
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>
L
Linus Torvalds 已提交
21 22
#include <linux/file.h>
#include <linux/binfmts.h>
S
Serge E. Hallyn 已提交
23
#include <linux/nsproxy.h>
24
#include <linux/pid_namespace.h>
L
Linus Torvalds 已提交
25 26 27 28
#include <linux/ptrace.h>
#include <linux/profile.h>
#include <linux/mount.h>
#include <linux/proc_fs.h>
29
#include <linux/kthread.h>
L
Linus Torvalds 已提交
30
#include <linux/mempolicy.h>
31
#include <linux/taskstats_kern.h>
32
#include <linux/delayacct.h>
33
#include <linux/freezer.h>
34
#include <linux/cgroup.h>
L
Linus Torvalds 已提交
35
#include <linux/syscalls.h>
36
#include <linux/signal.h>
37
#include <linux/posix-timers.h>
M
Matt Helsley 已提交
38
#include <linux/cn_proc.h>
39
#include <linux/mutex.h>
40
#include <linux/futex.h>
41
#include <linux/compat.h>
42
#include <linux/pipe_fs_i.h>
43
#include <linux/audit.h> /* for audit_free() */
A
Adrian Bunk 已提交
44
#include <linux/resource.h>
45
#include <linux/blkdev.h>
46
#include <linux/task_io_accounting_ops.h>
L
Linus Torvalds 已提交
47 48 49 50 51 52 53 54

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

extern void sem_exit (void);

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

L
Linus Torvalds 已提交
57 58 59 60 61 62 63
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);
64

65
		list_del_rcu(&p->tasks);
66
		__get_cpu_var(process_counts)--;
L
Linus Torvalds 已提交
67
	}
O
Oleg Nesterov 已提交
68
	list_del_rcu(&p->thread_group);
69
	remove_parent(p);
L
Linus Torvalds 已提交
70 71
}

72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94
/*
 * 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));

	rcu_read_lock();
	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:
		 */
95
		if (sig->group_exit_task && atomic_read(&sig->count) == sig->notify_count)
96
			wake_up_process(sig->group_exit_task);
97

98 99 100 101 102 103 104 105 106 107 108 109 110 111
		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);
112
		sig->gtime = cputime_add(sig->gtime, tsk->gtime);
113 114 115 116
		sig->min_flt += tsk->min_flt;
		sig->maj_flt += tsk->maj_flt;
		sig->nvcsw += tsk->nvcsw;
		sig->nivcsw += tsk->nivcsw;
117 118
		sig->inblock += task_io_get_inblock(tsk);
		sig->oublock += task_io_get_oublock(tsk);
119
		sig->sum_sched_runtime += tsk->se.sum_exec_runtime;
120 121 122
		sig = NULL; /* Marker for below. */
	}

123 124
	__unhash_process(tsk);

125
	tsk->signal = NULL;
126
	tsk->sighand = NULL;
127 128 129
	spin_unlock(&sighand->siglock);
	rcu_read_unlock();

130
	__cleanup_sighand(sighand);
131 132 133 134
	clear_tsk_thread_flag(tsk,TIF_SIGPENDING);
	flush_sigqueue(&tsk->pending);
	if (sig) {
		flush_sigqueue(&sig->shared_pending);
135
		taskstats_tgid_free(sig);
136 137 138 139
		__cleanup_signal(sig);
	}
}

140 141 142 143 144
static void delayed_put_task_struct(struct rcu_head *rhp)
{
	put_task_struct(container_of(rhp, struct task_struct, rcu));
}

L
Linus Torvalds 已提交
145 146
void release_task(struct task_struct * p)
{
147
	struct task_struct *leader;
L
Linus Torvalds 已提交
148
	int zap_leader;
149
repeat:
L
Linus Torvalds 已提交
150
	atomic_dec(&p->user->processes);
151
	proc_flush_task(p);
L
Linus Torvalds 已提交
152
	write_lock_irq(&tasklist_lock);
153
	ptrace_unlink(p);
L
Linus Torvalds 已提交
154 155
	BUG_ON(!list_empty(&p->ptrace_list) || !list_empty(&p->ptrace_children));
	__exit_signal(p);
156

L
Linus Torvalds 已提交
157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179
	/*
	 * 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) {
		BUG_ON(leader->exit_signal == -1);
		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.
		 */
		zap_leader = (leader->exit_signal == -1);
	}

	write_unlock_irq(&tasklist_lock);
	release_thread(p);
180
	call_rcu(&p->rcu, delayed_put_task_struct);
L
Linus Torvalds 已提交
181 182 183 184 185 186 187 188 189 190

	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...
191 192
 *
 * The caller must hold rcu lock or the tasklist lock.
L
Linus Torvalds 已提交
193
 */
194
struct pid *session_of_pgrp(struct pid *pgrp)
L
Linus Torvalds 已提交
195 196
{
	struct task_struct *p;
197
	struct pid *sid = NULL;
198

199
	p = pid_task(pgrp, PIDTYPE_PGID);
200
	if (p == NULL)
201
		p = pid_task(pgrp, PIDTYPE_PID);
202
	if (p != NULL)
203
		sid = task_session(p);
204

L
Linus Torvalds 已提交
205 206 207 208 209 210 211 212 213 214 215
	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?"
 */
216
static int will_become_orphaned_pgrp(struct pid *pgrp, struct task_struct *ignored_task)
L
Linus Torvalds 已提交
217 218 219 220
{
	struct task_struct *p;
	int ret = 1;

221
	do_each_pid_task(pgrp, PIDTYPE_PGID, p) {
L
Linus Torvalds 已提交
222 223
		if (p == ignored_task
				|| p->exit_state
224
				|| is_global_init(p->real_parent))
L
Linus Torvalds 已提交
225
			continue;
226 227
		if (task_pgrp(p->real_parent) != pgrp &&
		    task_session(p->real_parent) == task_session(p)) {
L
Linus Torvalds 已提交
228 229 230
			ret = 0;
			break;
		}
231
	} while_each_pid_task(pgrp, PIDTYPE_PGID, p);
L
Linus Torvalds 已提交
232 233 234
	return ret;	/* (sighing) "Often!" */
}

235
int is_current_pgrp_orphaned(void)
L
Linus Torvalds 已提交
236 237 238 239
{
	int retval;

	read_lock(&tasklist_lock);
240
	retval = will_become_orphaned_pgrp(task_pgrp(current), NULL);
L
Linus Torvalds 已提交
241 242 243 244 245
	read_unlock(&tasklist_lock);

	return retval;
}

246
static int has_stopped_jobs(struct pid *pgrp)
L
Linus Torvalds 已提交
247 248 249 250
{
	int retval = 0;
	struct task_struct *p;

251
	do_each_pid_task(pgrp, PIDTYPE_PGID, p) {
L
Linus Torvalds 已提交
252 253 254 255
		if (p->state != TASK_STOPPED)
			continue;
		retval = 1;
		break;
256
	} while_each_pid_task(pgrp, PIDTYPE_PGID, p);
L
Linus Torvalds 已提交
257 258 259 260
	return retval;
}

/**
261
 * reparent_to_kthreadd - Reparent the calling kernel thread to kthreadd
L
Linus Torvalds 已提交
262 263
 *
 * If a kernel thread is launched as a result of a system call, or if
264 265
 * 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
266 267 268 269
 *
 * 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.
 *
270
 * NOTE that reparent_to_kthreadd() gives the caller full capabilities.
L
Linus Torvalds 已提交
271
 */
272
static void reparent_to_kthreadd(void)
L
Linus Torvalds 已提交
273 274 275 276 277
{
	write_lock_irq(&tasklist_lock);

	ptrace_unlink(current);
	/* Reparent to init */
278
	remove_parent(current);
279
	current->real_parent = current->parent = kthreadd_task;
280
	add_parent(current);
L
Linus Torvalds 已提交
281 282 283 284

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

285
	if (task_nice(current) < 0)
L
Linus Torvalds 已提交
286 287 288 289 290 291 292 293 294 295 296 297 298 299
		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);
}

void __set_special_pids(pid_t session, pid_t pgrp)
{
300
	struct task_struct *curr = current->group_leader;
L
Linus Torvalds 已提交
301

P
Pavel Emelianov 已提交
302
	if (task_session_nr(curr) != session) {
L
Linus Torvalds 已提交
303
		detach_pid(curr, PIDTYPE_SID);
P
Pavel Emelianov 已提交
304
		set_task_session(curr, session);
305
		attach_pid(curr, PIDTYPE_SID, find_pid(session));
L
Linus Torvalds 已提交
306
	}
P
Pavel Emelianov 已提交
307
	if (task_pgrp_nr(curr) != pgrp) {
L
Linus Torvalds 已提交
308
		detach_pid(curr, PIDTYPE_PGID);
309
		set_task_pgrp(curr, pgrp);
310
		attach_pid(curr, PIDTYPE_PGID, find_pid(pgrp));
L
Linus Torvalds 已提交
311 312 313
	}
}

314
static void set_special_pids(pid_t session, pid_t pgrp)
L
Linus Torvalds 已提交
315 316 317 318 319 320 321 322 323 324 325 326 327
{
	write_lock_irq(&tasklist_lock);
	__set_special_pids(session, pgrp);
	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)
{
328
	if (!valid_signal(sig) || sig < 1)
L
Linus Torvalds 已提交
329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348
		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)
{
349
	if (!valid_signal(sig) || sig < 1)
L
Linus Torvalds 已提交
350 351 352
		return -EINVAL;

	spin_lock_irq(&current->sighand->siglock);
353
	current->sighand->action[(sig)-1].sa.sa_handler = SIG_IGN;
L
Linus Torvalds 已提交
354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381
	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);
382 383 384 385 386
	/*
	 * 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
387 388

	set_special_pids(1, 1);
389
	proc_clear_tty(current);
L
Linus Torvalds 已提交
390 391 392 393 394 395 396 397 398 399 400 401

	/* 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
402

403 404 405 406
	if (current->nsproxy != init_task.nsproxy) {
		get_nsproxy(init_task.nsproxy);
		switch_task_namespaces(current, init_task.nsproxy);
	}
S
Serge E. Hallyn 已提交
407

408
	exit_files(current);
L
Linus Torvalds 已提交
409 410 411
	current->files = init_task.files;
	atomic_inc(&current->files->count);

412
	reparent_to_kthreadd();
L
Linus Torvalds 已提交
413 414 415 416
}

EXPORT_SYMBOL(daemonize);

417
static void close_files(struct files_struct * files)
L
Linus Torvalds 已提交
418 419
{
	int i, j;
420
	struct fdtable *fdt;
L
Linus Torvalds 已提交
421 422

	j = 0;
423 424 425 426 427 428

	/*
	 * It is safe to dereference the fd table without RCU or
	 * ->file_lock because this is the last reference to the
	 * files structure.
	 */
429
	fdt = files_fdtable(files);
L
Linus Torvalds 已提交
430 431 432
	for (;;) {
		unsigned long set;
		i = j * __NFDBITS;
433
		if (i >= fdt->max_fds)
L
Linus Torvalds 已提交
434
			break;
435
		set = fdt->open_fds->fds_bits[j++];
L
Linus Torvalds 已提交
436 437
		while (set) {
			if (set & 1) {
438
				struct file * file = xchg(&fdt->fd[i], NULL);
439
				if (file) {
L
Linus Torvalds 已提交
440
					filp_close(file, files);
441 442
					cond_resched();
				}
L
Linus Torvalds 已提交
443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464
			}
			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;
}

void fastcall put_files_struct(struct files_struct *files)
{
465 466
	struct fdtable *fdt;

L
Linus Torvalds 已提交
467 468 469 470
	if (atomic_dec_and_test(&files->count)) {
		close_files(files);
		/*
		 * Free the fd and fdset arrays if we expanded them.
471 472 473
		 * 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
474
		 */
475
		fdt = files_fdtable(files);
476
		if (fdt != &files->fdtab)
477
			kmem_cache_free(files_cachep, files);
478
		free_fdtable(fdt);
L
Linus Torvalds 已提交
479 480 481 482 483
	}
}

EXPORT_SYMBOL(put_files_struct);

484 485 486 487 488 489 490 491 492 493 494 495
void reset_files_struct(struct task_struct *tsk, struct files_struct *files)
{
	struct files_struct *old;

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

A
Alexey Dobriyan 已提交
496
static void __exit_files(struct task_struct *tsk)
L
Linus Torvalds 已提交
497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512
{
	struct files_struct * files = tsk->files;

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

void exit_files(struct task_struct *tsk)
{
	__exit_files(tsk);
}

A
Alexey Dobriyan 已提交
513
static void __put_fs_struct(struct fs_struct *fs)
L
Linus Torvalds 已提交
514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533
{
	/* No need to hold fs->lock if we are killing it */
	if (atomic_dec_and_test(&fs->count)) {
		dput(fs->root);
		mntput(fs->rootmnt);
		dput(fs->pwd);
		mntput(fs->pwdmnt);
		if (fs->altroot) {
			dput(fs->altroot);
			mntput(fs->altrootmnt);
		}
		kmem_cache_free(fs_cachep, fs);
	}
}

void put_fs_struct(struct fs_struct *fs)
{
	__put_fs_struct(fs);
}

A
Alexey Dobriyan 已提交
534
static void __exit_fs(struct task_struct *tsk)
L
Linus Torvalds 已提交
535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556
{
	struct fs_struct * fs = tsk->fs;

	if (fs) {
		task_lock(tsk);
		tsk->fs = NULL;
		task_unlock(tsk);
		__put_fs_struct(fs);
	}
}

void exit_fs(struct task_struct *tsk)
{
	__exit_fs(tsk);
}

EXPORT_SYMBOL_GPL(exit_fs);

/*
 * Turn us into a lazy TLB process if we
 * aren't already..
 */
A
Adrian Bunk 已提交
557
static void exit_mm(struct task_struct * tsk)
L
Linus Torvalds 已提交
558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582
{
	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
583
	BUG_ON(mm != tsk->active_mm);
L
Linus Torvalds 已提交
584 585 586 587 588
	/* more a memory barrier than a real lock */
	task_lock(tsk);
	tsk->mm = NULL;
	up_read(&mm->mmap_sem);
	enter_lazy_tlb(mm, current);
589 590
	/* We don't want this task to be frozen prematurely */
	clear_freeze_flag(tsk);
L
Linus Torvalds 已提交
591 592 593 594
	task_unlock(tsk);
	mmput(mm);
}

595 596
static void
reparent_thread(struct task_struct *p, struct task_struct *father, int traced)
L
Linus Torvalds 已提交
597
{
598 599 600 601
	if (p->pdeath_signal)
		/* We already hold the tasklist_lock here.  */
		group_send_sig_info(p->pdeath_signal, SEND_SIG_NOINFO, p);

L
Linus Torvalds 已提交
602 603 604 605 606 607 608 609 610 611 612
	/* Move the child from its dying parent to the new one.  */
	if (unlikely(traced)) {
		/* Preserve ptrace links if someone else is tracing this child.  */
		list_del_init(&p->ptrace_list);
		if (p->parent != p->real_parent)
			list_add(&p->ptrace_list, &p->real_parent->ptrace_children);
	} else {
		/* If this child is being traced, then we're the one tracing it
		 * anyway, so let go of it.
		 */
		p->ptrace = 0;
613
		remove_parent(p);
L
Linus Torvalds 已提交
614
		p->parent = p->real_parent;
615
		add_parent(p);
L
Linus Torvalds 已提交
616

617
		if (p->state == TASK_TRACED) {
L
Linus Torvalds 已提交
618 619 620 621 622 623 624 625 626
			/*
			 * If it was at a trace stop, turn it into
			 * a normal stop since it's no longer being
			 * traced.
			 */
			ptrace_untrace(p);
		}
	}

627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643
	/* If this is a threaded reparent there is no need to
	 * notify anyone anything has happened.
	 */
	if (p->real_parent->group_leader == father->group_leader)
		return;

	/* We don't want people slaying init.  */
	if (p->exit_signal != -1)
		p->exit_signal = SIGCHLD;

	/* If we'd notified the old parent about this child's death,
	 * also notify the new parent.
	 */
	if (!traced && p->exit_state == EXIT_ZOMBIE &&
	    p->exit_signal != -1 && thread_group_empty(p))
		do_notify_parent(p, p->exit_signal);

L
Linus Torvalds 已提交
644 645 646 647 648 649
	/*
	 * process group orphan check
	 * Case ii: Our child is in a different pgrp
	 * than we are, and it was the only connection
	 * outside, so the child pgrp is now orphaned.
	 */
650 651 652
	if ((task_pgrp(p) != task_pgrp(father)) &&
	    (task_session(p) == task_session(father))) {
		struct pid *pgrp = task_pgrp(p);
L
Linus Torvalds 已提交
653

654 655
		if (will_become_orphaned_pgrp(pgrp, NULL) &&
		    has_stopped_jobs(pgrp)) {
656 657
			__kill_pgrp_info(SIGHUP, SEND_SIG_PRIV, pgrp);
			__kill_pgrp_info(SIGCONT, SEND_SIG_PRIV, pgrp);
L
Linus Torvalds 已提交
658 659 660 661 662 663 664 665
		}
	}
}

/*
 * 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
666 667
 * the child reaper process (ie "init") in our pid
 * space.
L
Linus Torvalds 已提交
668
 */
669
static void forget_original_parent(struct task_struct *father)
L
Linus Torvalds 已提交
670
{
671
	struct task_struct *p, *n, *reaper = father;
672 673 674 675 676
	struct list_head ptrace_dead;

	INIT_LIST_HEAD(&ptrace_dead);

	write_lock_irq(&tasklist_lock);
L
Linus Torvalds 已提交
677 678 679 680

	do {
		reaper = next_thread(reaper);
		if (reaper == father) {
681
			reaper = task_child_reaper(father);
L
Linus Torvalds 已提交
682 683
			break;
		}
684
	} while (reaper->flags & PF_EXITING);
L
Linus Torvalds 已提交
685 686 687 688 689 690 691 692 693

	/*
	 * There are only two places where our children can be:
	 *
	 * - in our child list
	 * - in our ptraced child list
	 *
	 * Search them and reparent children.
	 */
694
	list_for_each_entry_safe(p, n, &father->children, sibling) {
L
Linus Torvalds 已提交
695 696 697 698 699 700 701 702 703
		int ptrace;

		ptrace = p->ptrace;

		/* if father isn't the real parent, then ptrace must be enabled */
		BUG_ON(father != p->real_parent && !ptrace);

		if (father == p->real_parent) {
			/* reparent with a reaper, real father it's us */
704
			p->real_parent = reaper;
L
Linus Torvalds 已提交
705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720
			reparent_thread(p, father, 0);
		} else {
			/* reparent ptraced task to its real parent */
			__ptrace_unlink (p);
			if (p->exit_state == EXIT_ZOMBIE && p->exit_signal != -1 &&
			    thread_group_empty(p))
				do_notify_parent(p, p->exit_signal);
		}

		/*
		 * if the ptraced child is a zombie with exit_signal == -1
		 * we must collect it before we exit, or it will remain
		 * zombie forever since we prevented it from self-reap itself
		 * while it was being traced by us, to be able to see it in wait4.
		 */
		if (unlikely(ptrace && p->exit_state == EXIT_ZOMBIE && p->exit_signal == -1))
721
			list_add(&p->ptrace_list, &ptrace_dead);
L
Linus Torvalds 已提交
722
	}
723

724
	list_for_each_entry_safe(p, n, &father->ptrace_children, ptrace_list) {
725
		p->real_parent = reaper;
L
Linus Torvalds 已提交
726 727
		reparent_thread(p, father, 1);
	}
728 729 730 731 732 733 734 735 736 737

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

	list_for_each_entry_safe(p, n, &ptrace_dead, ptrace_list) {
		list_del_init(&p->ptrace_list);
		release_task(p);
	}

L
Linus Torvalds 已提交
738 739 740 741 742 743 744 745 746 747
}

/*
 * Send signals to all our closest relatives so that they know
 * to properly mourn us..
 */
static void exit_notify(struct task_struct *tsk)
{
	int state;
	struct task_struct *t;
748
	struct pid *pgrp;
L
Linus Torvalds 已提交
749 750 751 752 753 754 755 756 757 758 759 760 761 762

	if (signal_pending(tsk) && !(tsk->signal->flags & SIGNAL_GROUP_EXIT)
	    && !thread_group_empty(tsk)) {
		/*
		 * This occurs when there was a race between our exit
		 * syscall and a group signal choosing us as the one to
		 * wake up.  It could be that we are the only thread
		 * alerted to check for pending signals, but another thread
		 * should be woken now to take the signal since we will not.
		 * Now we'll wake all the threads in the group just to make
		 * sure someone gets all the pending signals.
		 */
		spin_lock_irq(&tsk->sighand->siglock);
		for (t = next_thread(tsk); t != tsk; t = next_thread(t))
R
Roland McGrath 已提交
763 764
			if (!signal_pending(t) && !(t->flags & PF_EXITING))
				recalc_sigpending_and_wake(t);
L
Linus Torvalds 已提交
765 766 767 768 769 770 771 772 773 774 775
		spin_unlock_irq(&tsk->sighand->siglock);
	}

	/*
	 * 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)
	 */
776
	forget_original_parent(tsk);
777
	exit_task_namespaces(tsk);
L
Linus Torvalds 已提交
778

779
	write_lock_irq(&tasklist_lock);
L
Linus Torvalds 已提交
780 781 782 783 784 785 786 787 788 789
	/*
	 * 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)
	 *
	 * Case i: Our father is in a different pgrp than we are
	 * and we were the only connection outside, so our pgrp
	 * is about to become orphaned.
	 */
	t = tsk->real_parent;
790

791 792
	pgrp = task_pgrp(tsk);
	if ((task_pgrp(t) != pgrp) &&
793
	    (task_session(t) == task_session(tsk)) &&
794 795 796 797
	    will_become_orphaned_pgrp(pgrp, tsk) &&
	    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
798 799
	}

800
	/* Let father know we died
L
Linus Torvalds 已提交
801 802
	 *
	 * Thread signals are configurable, but you aren't going to use
803
	 * that to send signals to arbitary processes.
L
Linus Torvalds 已提交
804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832
	 * 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.
	 */
	if (tsk->exit_signal != SIGCHLD && tsk->exit_signal != -1 &&
	    ( tsk->parent_exec_id != t->self_exec_id  ||
	      tsk->self_exec_id != tsk->parent_exec_id)
	    && !capable(CAP_KILL))
		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.
	 */
	if (tsk->exit_signal != -1 && thread_group_empty(tsk)) {
		int signal = tsk->parent == tsk->real_parent ? tsk->exit_signal : SIGCHLD;
		do_notify_parent(tsk, signal);
	} else if (tsk->ptrace) {
		do_notify_parent(tsk, SIGCHLD);
	}

	state = EXIT_ZOMBIE;
833
	if (tsk->exit_signal == -1 && likely(!tsk->ptrace))
L
Linus Torvalds 已提交
834 835 836
		state = EXIT_DEAD;
	tsk->exit_state = state;

837 838 839 840 841
	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
842 843 844 845 846 847 848
	write_unlock_irq(&tasklist_lock);

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

849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876
#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

877 878
static inline void exit_child_reaper(struct task_struct *tsk)
{
879
	if (likely(tsk->group_leader != task_child_reaper(tsk)))
880 881
		return;

882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907
	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);
908 909
}

L
Linus Torvalds 已提交
910 911 912 913 914 915 916
fastcall NORET_TYPE void do_exit(long code)
{
	struct task_struct *tsk = current;
	int group_dead;

	profile_task_exit(tsk);

917 918
	WARN_ON(atomic_read(&tsk->fs_excl));

L
Linus Torvalds 已提交
919 920 921 922 923 924 925 926 927 928
	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);
	}

929 930 931 932 933 934 935
	/*
	 * 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");
936 937 938 939 940 941 942 943 944 945
		/*
		 * 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;
946 947
		if (tsk->io_context)
			exit_io_context();
948 949 950 951
		set_current_state(TASK_UNINTERRUPTIBLE);
		schedule();
	}

952
	tsk->flags |= PF_EXITING;
953 954 955 956
	/*
	 * tsk->flags are checked in the futex code to protect against
	 * an exiting task cleaning up the robust pi futexes.
	 */
957 958
	smp_mb();
	spin_unlock_wait(&tsk->pi_lock);
L
Linus Torvalds 已提交
959 960 961

	if (unlikely(in_atomic()))
		printk(KERN_INFO "note: %s[%d] exited with preempt_count %d\n",
962
				current->comm, task_pid_nr(current),
L
Linus Torvalds 已提交
963 964 965
				preempt_count());

	acct_update_integrals(tsk);
966 967 968 969
	if (tsk->mm) {
		update_hiwater_rss(tsk->mm);
		update_hiwater_vm(tsk->mm);
	}
L
Linus Torvalds 已提交
970
	group_dead = atomic_dec_and_test(&tsk->signal->live);
971
	if (group_dead) {
972
		exit_child_reaper(tsk);
973
		hrtimer_cancel(&tsk->signal->real_timer);
974
		exit_itimers(tsk->signal);
975
	}
976
	acct_collect(code, group_dead);
977
#ifdef CONFIG_FUTEX
978 979
	if (unlikely(tsk->robust_list))
		exit_robust_list(tsk);
980
#ifdef CONFIG_COMPAT
981 982
	if (unlikely(tsk->compat_robust_list))
		compat_exit_robust_list(tsk);
983
#endif
984
#endif
M
Miloslav Trmac 已提交
985 986
	if (group_dead)
		tty_audit_exit();
987 988
	if (unlikely(tsk->audit_context))
		audit_free(tsk);
989

990
	tsk->exit_code = code;
991
	taskstats_exit(tsk, group_dead);
992

L
Linus Torvalds 已提交
993 994
	exit_mm(tsk);

995
	if (group_dead)
996
		acct_process();
L
Linus Torvalds 已提交
997 998 999
	exit_sem(tsk);
	__exit_files(tsk);
	__exit_fs(tsk);
1000
	check_stack_usage();
L
Linus Torvalds 已提交
1001
	exit_thread();
1002
	cgroup_exit(tsk, 1);
L
Linus Torvalds 已提交
1003 1004 1005 1006 1007
	exit_keys(tsk);

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

A
Al Viro 已提交
1008
	module_put(task_thread_info(tsk)->exec_domain->module);
L
Linus Torvalds 已提交
1009 1010 1011
	if (tsk->binfmt)
		module_put(tsk->binfmt->module);

M
Matt Helsley 已提交
1012
	proc_exit_connector(tsk);
1013
	exit_notify(tsk);
L
Linus Torvalds 已提交
1014 1015 1016 1017
#ifdef CONFIG_NUMA
	mpol_free(tsk->mempolicy);
	tsk->mempolicy = NULL;
#endif
1018
#ifdef CONFIG_FUTEX
1019 1020 1021 1022 1023 1024 1025 1026
	/*
	 * 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);
1027
#endif
1028
	/*
1029
	 * Make sure we are holding no locks:
1030
	 */
1031
	debug_check_no_locks_held(tsk);
1032 1033 1034 1035 1036 1037
	/*
	 * 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
1038

1039 1040 1041
	if (tsk->io_context)
		exit_io_context();

1042 1043 1044
	if (tsk->splice_pipe)
		__free_pipe_info(tsk->splice_pipe);

C
Coywolf Qi Hunt 已提交
1045
	preempt_disable();
O
Oleg Nesterov 已提交
1046
	/* causes final put_task_struct in finish_task_switch(). */
1047
	tsk->state = TASK_DEAD;
C
Coywolf Qi Hunt 已提交
1048

L
Linus Torvalds 已提交
1049 1050 1051
	schedule();
	BUG();
	/* Avoid "noreturn function does return".  */
A
Alan Cox 已提交
1052 1053
	for (;;)
		cpu_relax();	/* For when BUG is null */
L
Linus Torvalds 已提交
1054 1055
}

1056 1057
EXPORT_SYMBOL_GPL(do_exit);

L
Linus Torvalds 已提交
1058 1059 1060 1061
NORET_TYPE void complete_and_exit(struct completion *comp, long code)
{
	if (comp)
		complete(comp);
O
Oleg Nesterov 已提交
1062

L
Linus Torvalds 已提交
1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111
	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)
{
	BUG_ON(exit_code & 0x80); /* core dumps don't get here */

	if (current->signal->flags & SIGNAL_GROUP_EXIT)
		exit_code = current->signal->group_exit_code;
	else if (!thread_group_empty(current)) {
		struct signal_struct *const sig = current->signal;
		struct sighand_struct *const sighand = current->sighand;
		spin_lock_irq(&sighand->siglock);
		if (sig->flags & SIGNAL_GROUP_EXIT)
			/* Another thread got here before we took the lock.  */
			exit_code = sig->group_exit_code;
		else {
			sig->group_exit_code = exit_code;
			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);
}

1112
static int eligible_child(pid_t pid, int options, struct task_struct *p)
L
Linus Torvalds 已提交
1113
{
1114
	int err;
1115
	struct pid_namespace *ns;
1116

1117
	ns = current->nsproxy->pid_ns;
L
Linus Torvalds 已提交
1118
	if (pid > 0) {
1119
		if (task_pid_nr_ns(p, ns) != pid)
L
Linus Torvalds 已提交
1120 1121
			return 0;
	} else if (!pid) {
1122
		if (task_pgrp_nr_ns(p, ns) != task_pgrp_vnr(current))
L
Linus Torvalds 已提交
1123 1124
			return 0;
	} else if (pid != -1) {
1125
		if (task_pgrp_nr_ns(p, ns) != -pid)
L
Linus Torvalds 已提交
1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147
			return 0;
	}

	/*
	 * Do not consider detached threads that are
	 * not ptraced:
	 */
	if (p->exit_signal == -1 && !p->ptrace)
		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;
	/*
	 * Do not consider thread group leaders that are
	 * in a non-empty thread group:
	 */
1148
	if (delay_group_leader(p))
L
Linus Torvalds 已提交
1149 1150
		return 2;

1151 1152 1153
	err = security_task_wait(p);
	if (err)
		return err;
L
Linus Torvalds 已提交
1154 1155 1156 1157

	return 1;
}

1158
static int wait_noreap_copyout(struct task_struct *p, pid_t pid, uid_t uid,
L
Linus Torvalds 已提交
1159 1160 1161 1162 1163
			       int why, int status,
			       struct siginfo __user *infop,
			       struct rusage __user *rusagep)
{
	int retval = rusagep ? getrusage(p, RUSAGE_BOTH, rusagep) : 0;
1164

L
Linus Torvalds 已提交
1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188
	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.
 */
1189
static int wait_task_zombie(struct task_struct *p, int noreap,
L
Linus Torvalds 已提交
1190 1191 1192 1193
			    struct siginfo __user *infop,
			    int __user *stat_addr, struct rusage __user *ru)
{
	unsigned long state;
1194
	int retval, status, traced;
1195 1196 1197
	struct pid_namespace *ns;

	ns = current->nsproxy->pid_ns;
L
Linus Torvalds 已提交
1198 1199

	if (unlikely(noreap)) {
1200
		pid_t pid = task_pid_nr_ns(p, ns);
L
Linus Torvalds 已提交
1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231
		uid_t uid = p->uid;
		int exit_code = p->exit_code;
		int why, status;

		if (unlikely(p->exit_state != EXIT_ZOMBIE))
			return 0;
		if (unlikely(p->exit_signal == -1 && p->ptrace == 0))
			return 0;
		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;
	}

1232 1233 1234 1235
	/* traced means p->ptrace, but not vice versa */
	traced = (p->real_parent != p->parent);

	if (likely(!traced)) {
1236 1237 1238
		struct signal_struct *psig;
		struct signal_struct *sig;

L
Linus Torvalds 已提交
1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254
		/*
		 * 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);
1255 1256 1257 1258
		psig = p->parent->signal;
		sig = p->signal;
		psig->cutime =
			cputime_add(psig->cutime,
L
Linus Torvalds 已提交
1259
			cputime_add(p->utime,
1260 1261 1262 1263
			cputime_add(sig->utime,
				    sig->cutime)));
		psig->cstime =
			cputime_add(psig->cstime,
L
Linus Torvalds 已提交
1264
			cputime_add(p->stime,
1265 1266
			cputime_add(sig->stime,
				    sig->cstime)));
1267 1268 1269 1270 1271
		psig->cgtime =
			cputime_add(psig->cgtime,
			cputime_add(p->gtime,
			cputime_add(sig->gtime,
				    sig->cgtime)));
1272 1273 1274 1275 1276 1277 1278 1279
		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;
1280 1281 1282 1283 1284 1285
		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
1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318
		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)
1319
		retval = put_user(task_pid_nr_ns(p, ns), &infop->si_pid);
L
Linus Torvalds 已提交
1320 1321
	if (!retval && infop)
		retval = put_user(p->uid, &infop->si_uid);
1322
	if (!retval)
1323
		retval = task_pid_nr_ns(p, ns);
1324 1325

	if (traced) {
L
Linus Torvalds 已提交
1326
		write_lock_irq(&tasklist_lock);
1327 1328 1329 1330 1331 1332 1333 1334 1335
		/* 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.
		 */
		if (p->exit_signal != -1) {
			do_notify_parent(p, p->exit_signal);
L
Linus Torvalds 已提交
1336
			if (p->exit_signal != -1) {
1337 1338
				p->exit_state = EXIT_ZOMBIE;
				p = NULL;
L
Linus Torvalds 已提交
1339 1340 1341 1342 1343 1344
			}
		}
		write_unlock_irq(&tasklist_lock);
	}
	if (p != NULL)
		release_task(p);
1345

L
Linus Torvalds 已提交
1346 1347 1348 1349 1350 1351 1352 1353 1354
	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.
 */
1355 1356
static int wait_task_stopped(struct task_struct *p, int delayed_group_leader,
			     int noreap, struct siginfo __user *infop,
L
Linus Torvalds 已提交
1357 1358 1359
			     int __user *stat_addr, struct rusage __user *ru)
{
	int retval, exit_code;
1360
	pid_t pid;
L
Linus Torvalds 已提交
1361 1362 1363 1364

	if (!p->exit_code)
		return 0;
	if (delayed_group_leader && !(p->ptrace & PT_PTRACED) &&
1365
	    p->signal->group_stop_count > 0)
L
Linus Torvalds 已提交
1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378
		/*
		 * A group stop is in progress and this is the group leader.
		 * We won't report until all threads have stopped.
		 */
		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.
	 */
1379
	pid = task_pid_nr_ns(p, current->nsproxy->pid_ns);
L
Linus Torvalds 已提交
1380 1381 1382 1383 1384 1385 1386 1387
	get_task_struct(p);
	read_unlock(&tasklist_lock);

	if (unlikely(noreap)) {
		uid_t uid = p->uid;
		int why = (p->ptrace & PT_PTRACED) ? CLD_TRAPPED : CLD_STOPPED;

		exit_code = p->exit_code;
1388
		if (unlikely(!exit_code) || unlikely(p->exit_state))
L
Linus Torvalds 已提交
1389 1390
			goto bail_ref;
		return wait_noreap_copyout(p, pid, uid,
1391
					   why, exit_code,
L
Linus Torvalds 已提交
1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434
					   infop, ru);
	}

	write_lock_irq(&tasklist_lock);

	/*
	 * This uses xchg to be atomic with the thread resuming and setting
	 * it.  It must also be done with the write lock held to prevent a
	 * race with the EXIT_ZOMBIE case.
	 */
	exit_code = xchg(&p->exit_code, 0);
	if (unlikely(p->exit_state)) {
		/*
		 * The task resumed and then died.  Let the next iteration
		 * catch it in EXIT_ZOMBIE.  Note that exit_code might
		 * already be zero here if it resumed and did _exit(0).
		 * The task itself is dead and won't touch exit_code again;
		 * other processors in this function are locked out.
		 */
		p->exit_code = exit_code;
		exit_code = 0;
	}
	if (unlikely(exit_code == 0)) {
		/*
		 * Another thread in this function got to it first, or it
		 * resumed, or it resumed and then died.
		 */
		write_unlock_irq(&tasklist_lock);
bail_ref:
		put_task_struct(p);
		/*
		 * We are returning to the wait loop without having successfully
		 * removed the process and having released the lock. We cannot
		 * continue, since the "p" task pointer is potentially stale.
		 *
		 * Return -EAGAIN, and do_wait() will restart the loop from the
		 * beginning. Do _not_ re-acquire the lock.
		 */
		return -EAGAIN;
	}

	/* move to end of parent's list to avoid starvation */
	remove_parent(p);
1435
	add_parent(p);
L
Linus Torvalds 已提交
1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452

	write_unlock_irq(&tasklist_lock);

	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)
		retval = put_user((short)((p->ptrace & PT_PTRACED)
					  ? CLD_TRAPPED : CLD_STOPPED),
				  &infop->si_code);
	if (!retval && infop)
		retval = put_user(exit_code, &infop->si_status);
	if (!retval && infop)
1453
		retval = put_user(pid, &infop->si_pid);
L
Linus Torvalds 已提交
1454 1455 1456
	if (!retval && infop)
		retval = put_user(p->uid, &infop->si_uid);
	if (!retval)
1457
		retval = pid;
L
Linus Torvalds 已提交
1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469
	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.
 */
1470
static int wait_task_continued(struct task_struct *p, int noreap,
L
Linus Torvalds 已提交
1471 1472 1473 1474 1475 1476
			       struct siginfo __user *infop,
			       int __user *stat_addr, struct rusage __user *ru)
{
	int retval;
	pid_t pid;
	uid_t uid;
1477
	struct pid_namespace *ns;
L
Linus Torvalds 已提交
1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491

	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;
	}
	if (!noreap)
		p->signal->flags &= ~SIGNAL_STOP_CONTINUED;
	spin_unlock_irq(&p->sighand->siglock);

1492 1493
	ns = current->nsproxy->pid_ns;
	pid = task_pid_nr_ns(p, ns);
L
Linus Torvalds 已提交
1494 1495 1496 1497 1498 1499 1500 1501 1502 1503
	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)
1504
			retval = task_pid_nr_ns(p, ns);
L
Linus Torvalds 已提交
1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537
	} else {
		retval = wait_noreap_copyout(p, pid, uid,
					     CLD_CONTINUED, SIGCONT,
					     infop, ru);
		BUG_ON(retval == 0);
	}

	return retval;
}


static inline int my_ptrace_child(struct task_struct *p)
{
	if (!(p->ptrace & PT_PTRACED))
		return 0;
	if (!(p->ptrace & PT_ATTACHED))
		return 1;
	/*
	 * This child was PTRACE_ATTACH'd.  We should be seeing it only if
	 * we are the attacher.  If we are the real parent, this is a race
	 * inside ptrace_attach.  It is waiting for the tasklist_lock,
	 * which we have to switch the parent links, but has already set
	 * the flags in p->ptrace.
	 */
	return (p->parent != p->real_parent);
}

static long do_wait(pid_t pid, int options, struct siginfo __user *infop,
		    int __user *stat_addr, struct rusage __user *ru)
{
	DECLARE_WAITQUEUE(wait, current);
	struct task_struct *tsk;
	int flag, retval;
1538
	int allowed, denied;
L
Linus Torvalds 已提交
1539 1540 1541 1542 1543 1544 1545 1546

	add_wait_queue(&current->signal->wait_chldexit,&wait);
repeat:
	/*
	 * We will set this flag if we see any child that might later
	 * match our criteria, even if we are not able to reap it yet.
	 */
	flag = 0;
1547
	allowed = denied = 0;
L
Linus Torvalds 已提交
1548 1549 1550 1551 1552 1553 1554
	current->state = TASK_INTERRUPTIBLE;
	read_lock(&tasklist_lock);
	tsk = current;
	do {
		struct task_struct *p;
		int ret;

1555
		list_for_each_entry(p, &tsk->children, sibling) {
L
Linus Torvalds 已提交
1556 1557 1558 1559
			ret = eligible_child(pid, options, p);
			if (!ret)
				continue;

1560 1561 1562 1563 1564 1565
			if (unlikely(ret < 0)) {
				denied = ret;
				continue;
			}
			allowed = 1;

L
Linus Torvalds 已提交
1566 1567
			switch (p->state) {
			case TASK_TRACED:
1568 1569 1570 1571 1572 1573 1574 1575 1576
				/*
				 * When we hit the race with PTRACE_ATTACH,
				 * we will not report this child.  But the
				 * race means it has not yet been moved to
				 * our ptrace_children list, so we need to
				 * set the flag here to avoid a spurious ECHILD
				 * when the race happens with the only child.
				 */
				flag = 1;
L
Linus Torvalds 已提交
1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636
				if (!my_ptrace_child(p))
					continue;
				/*FALLTHROUGH*/
			case TASK_STOPPED:
				/*
				 * It's stopped now, so it might later
				 * continue, exit, or stop again.
				 */
				flag = 1;
				if (!(options & WUNTRACED) &&
				    !my_ptrace_child(p))
					continue;
				retval = wait_task_stopped(p, ret == 2,
							   (options & WNOWAIT),
							   infop,
							   stat_addr, ru);
				if (retval == -EAGAIN)
					goto repeat;
				if (retval != 0) /* He released the lock.  */
					goto end;
				break;
			default:
			// case EXIT_DEAD:
				if (p->exit_state == EXIT_DEAD)
					continue;
			// case EXIT_ZOMBIE:
				if (p->exit_state == EXIT_ZOMBIE) {
					/*
					 * Eligible but we cannot release
					 * it yet:
					 */
					if (ret == 2)
						goto check_continued;
					if (!likely(options & WEXITED))
						continue;
					retval = wait_task_zombie(
						p, (options & WNOWAIT),
						infop, stat_addr, ru);
					/* He released the lock.  */
					if (retval != 0)
						goto end;
					break;
				}
check_continued:
				/*
				 * It's running now, so it might later
				 * exit, stop, or stop and then continue.
				 */
				flag = 1;
				if (!unlikely(options & WCONTINUED))
					continue;
				retval = wait_task_continued(
					p, (options & WNOWAIT),
					infop, stat_addr, ru);
				if (retval != 0) /* He released the lock.  */
					goto end;
				break;
			}
		}
		if (!flag) {
1637 1638
			list_for_each_entry(p, &tsk->ptrace_children,
					    ptrace_list) {
L
Linus Torvalds 已提交
1639 1640 1641 1642 1643 1644 1645 1646 1647
				if (!eligible_child(pid, options, p))
					continue;
				flag = 1;
				break;
			}
		}
		if (options & __WNOTHREAD)
			break;
		tsk = next_thread(tsk);
E
Eric Sesterhenn 已提交
1648
		BUG_ON(tsk->signal != current->signal);
L
Linus Torvalds 已提交
1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662
	} while (tsk != current);

	read_unlock(&tasklist_lock);
	if (flag) {
		retval = 0;
		if (options & WNOHANG)
			goto end;
		retval = -ERESTARTSYS;
		if (signal_pending(current))
			goto end;
		schedule();
		goto repeat;
	}
	retval = -ECHILD;
1663 1664
	if (unlikely(denied) && !allowed)
		retval = denied;
L
Linus Torvalds 已提交
1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755
end:
	current->state = TASK_RUNNING;
	remove_wait_queue(&current->signal->wait_chldexit,&wait);
	if (infop) {
		if (retval > 0)
		retval = 0;
		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;
}

asmlinkage long sys_waitid(int which, pid_t pid,
			   struct siginfo __user *infop, int options,
			   struct rusage __user *ru)
{
	long ret;

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

	switch (which) {
	case P_ALL:
		pid = -1;
		break;
	case P_PID:
		if (pid <= 0)
			return -EINVAL;
		break;
	case P_PGID:
		if (pid <= 0)
			return -EINVAL;
		pid = -pid;
		break;
	default:
		return -EINVAL;
	}

	ret = do_wait(pid, options, infop, NULL, ru);

	/* avoid REGPARM breakage on x86: */
	prevent_tail_call(ret);
	return ret;
}

asmlinkage long sys_wait4(pid_t pid, int __user *stat_addr,
			  int options, struct rusage __user *ru)
{
	long ret;

	if (options & ~(WNOHANG|WUNTRACED|WCONTINUED|
			__WNOTHREAD|__WCLONE|__WALL))
		return -EINVAL;
	ret = do_wait(pid, options | WEXITED, NULL, stat_addr, ru);

	/* avoid REGPARM breakage on x86: */
	prevent_tail_call(ret);
	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