signal.c 94.8 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13
/*
 *  linux/kernel/signal.c
 *
 *  Copyright (C) 1991, 1992  Linus Torvalds
 *
 *  1997-11-02  Modified for POSIX.1b signals by Richard Henderson
 *
 *  2003-06-02  Jim Houston - Concurrent Computer Corp.
 *		Changes to use preallocated sigqueue structures
 *		to allow signals to be sent reliably.
 */

#include <linux/slab.h>
14
#include <linux/export.h>
L
Linus Torvalds 已提交
15
#include <linux/init.h>
16
#include <linux/sched/mm.h>
17
#include <linux/sched/user.h>
18
#include <linux/sched/debug.h>
19
#include <linux/sched/task.h>
20
#include <linux/sched/task_stack.h>
21
#include <linux/sched/cputime.h>
L
Linus Torvalds 已提交
22 23 24
#include <linux/fs.h>
#include <linux/tty.h>
#include <linux/binfmts.h>
25
#include <linux/coredump.h>
L
Linus Torvalds 已提交
26 27 28
#include <linux/security.h>
#include <linux/syscalls.h>
#include <linux/ptrace.h>
29
#include <linux/signal.h>
30
#include <linux/signalfd.h>
31
#include <linux/ratelimit.h>
32
#include <linux/tracehook.h>
33
#include <linux/capability.h>
34
#include <linux/freezer.h>
35 36
#include <linux/pid_namespace.h>
#include <linux/nsproxy.h>
37
#include <linux/user_namespace.h>
38
#include <linux/uprobes.h>
A
Al Viro 已提交
39
#include <linux/compat.h>
40
#include <linux/cn_proc.h>
41
#include <linux/compiler.h>
42
#include <linux/posix-timers.h>
43

44 45
#define CREATE_TRACE_POINTS
#include <trace/events/signal.h>
46

L
Linus Torvalds 已提交
47
#include <asm/param.h>
48
#include <linux/uaccess.h>
L
Linus Torvalds 已提交
49 50
#include <asm/unistd.h>
#include <asm/siginfo.h>
51
#include <asm/cacheflush.h>
52
#include "audit.h"	/* audit_signal_info() */
L
Linus Torvalds 已提交
53 54 55 56 57

/*
 * SLAB caches for signal bits.
 */

58
static struct kmem_cache *sigqueue_cachep;
L
Linus Torvalds 已提交
59

60 61
int print_fatal_signals __read_mostly;

62
static void __user *sig_handler(struct task_struct *t, int sig)
63
{
64 65
	return t->sighand->action[sig - 1].sa.sa_handler;
}
66

67 68
static int sig_handler_ignored(void __user *handler, int sig)
{
69 70 71 72
	/* Is it explicitly or implicitly ignored? */
	return handler == SIG_IGN ||
		(handler == SIG_DFL && sig_kernel_ignore(sig));
}
L
Linus Torvalds 已提交
73

74
static int sig_task_ignored(struct task_struct *t, int sig, bool force)
L
Linus Torvalds 已提交
75
{
76
	void __user *handler;
L
Linus Torvalds 已提交
77

78 79 80
	handler = sig_handler(t, sig);

	if (unlikely(t->signal->flags & SIGNAL_UNKILLABLE) &&
81
			handler == SIG_DFL && !force)
82 83 84 85 86
		return 1;

	return sig_handler_ignored(handler, sig);
}

87
static int sig_ignored(struct task_struct *t, int sig, bool force)
88
{
L
Linus Torvalds 已提交
89 90 91 92 93
	/*
	 * Blocked signals are never ignored, since the
	 * signal handler may change by the time it is
	 * unblocked.
	 */
94
	if (sigismember(&t->blocked, sig) || sigismember(&t->real_blocked, sig))
L
Linus Torvalds 已提交
95 96
		return 0;

97
	if (!sig_task_ignored(t, sig, force))
98 99 100 101 102
		return 0;

	/*
	 * Tracers may want to know about even ignored signals.
	 */
T
Tejun Heo 已提交
103
	return !t->ptrace;
L
Linus Torvalds 已提交
104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137
}

/*
 * Re-calculate pending state from the set of locally pending
 * signals, globally pending signals, and blocked signals.
 */
static inline int has_pending_signals(sigset_t *signal, sigset_t *blocked)
{
	unsigned long ready;
	long i;

	switch (_NSIG_WORDS) {
	default:
		for (i = _NSIG_WORDS, ready = 0; --i >= 0 ;)
			ready |= signal->sig[i] &~ blocked->sig[i];
		break;

	case 4: ready  = signal->sig[3] &~ blocked->sig[3];
		ready |= signal->sig[2] &~ blocked->sig[2];
		ready |= signal->sig[1] &~ blocked->sig[1];
		ready |= signal->sig[0] &~ blocked->sig[0];
		break;

	case 2: ready  = signal->sig[1] &~ blocked->sig[1];
		ready |= signal->sig[0] &~ blocked->sig[0];
		break;

	case 1: ready  = signal->sig[0] &~ blocked->sig[0];
	}
	return ready !=	0;
}

#define PENDING(p,b) has_pending_signals(&(p)->signal, (b))

R
Roland McGrath 已提交
138
static int recalc_sigpending_tsk(struct task_struct *t)
L
Linus Torvalds 已提交
139
{
140
	if ((t->jobctl & JOBCTL_PENDING_MASK) ||
L
Linus Torvalds 已提交
141
	    PENDING(&t->pending, &t->blocked) ||
R
Roland McGrath 已提交
142
	    PENDING(&t->signal->shared_pending, &t->blocked)) {
L
Linus Torvalds 已提交
143
		set_tsk_thread_flag(t, TIF_SIGPENDING);
R
Roland McGrath 已提交
144 145
		return 1;
	}
146 147 148 149 150
	/*
	 * We must never clear the flag in another thread, or in current
	 * when it's possible the current syscall is returning -ERESTART*.
	 * So we don't clear it here, and only callers who know they should do.
	 */
R
Roland McGrath 已提交
151 152 153 154 155 156 157 158 159 160 161
	return 0;
}

/*
 * After recalculating TIF_SIGPENDING, we need to make sure the task wakes up.
 * This is superfluous when called on current, the wakeup is a harmless no-op.
 */
void recalc_sigpending_and_wake(struct task_struct *t)
{
	if (recalc_sigpending_tsk(t))
		signal_wake_up(t, 0);
L
Linus Torvalds 已提交
162 163 164 165
}

void recalc_sigpending(void)
{
T
Tejun Heo 已提交
166
	if (!recalc_sigpending_tsk(current) && !freezing(current))
167 168
		clear_thread_flag(TIF_SIGPENDING);

L
Linus Torvalds 已提交
169 170 171 172
}

/* Given the mask, find the first available signal that should be serviced. */

173 174
#define SYNCHRONOUS_MASK \
	(sigmask(SIGSEGV) | sigmask(SIGBUS) | sigmask(SIGILL) | \
175
	 sigmask(SIGTRAP) | sigmask(SIGFPE) | sigmask(SIGSYS))
176

177
int next_signal(struct sigpending *pending, sigset_t *mask)
L
Linus Torvalds 已提交
178 179 180
{
	unsigned long i, *s, *m, x;
	int sig = 0;
181

L
Linus Torvalds 已提交
182 183
	s = pending->signal.sig;
	m = mask->sig;
184 185 186 187 188 189 190 191 192 193 194 195 196

	/*
	 * Handle the first word specially: it contains the
	 * synchronous signals that need to be dequeued first.
	 */
	x = *s &~ *m;
	if (x) {
		if (x & SYNCHRONOUS_MASK)
			x &= SYNCHRONOUS_MASK;
		sig = ffz(~x) + 1;
		return sig;
	}

L
Linus Torvalds 已提交
197 198
	switch (_NSIG_WORDS) {
	default:
199 200 201 202 203 204 205
		for (i = 1; i < _NSIG_WORDS; ++i) {
			x = *++s &~ *++m;
			if (!x)
				continue;
			sig = ffz(~x) + i*_NSIG_BPW + 1;
			break;
		}
L
Linus Torvalds 已提交
206 207
		break;

208 209 210
	case 2:
		x = s[1] &~ m[1];
		if (!x)
L
Linus Torvalds 已提交
211
			break;
212
		sig = ffz(~x) + _NSIG_BPW + 1;
L
Linus Torvalds 已提交
213 214
		break;

215 216
	case 1:
		/* Nothing to do */
L
Linus Torvalds 已提交
217 218
		break;
	}
219

L
Linus Torvalds 已提交
220 221 222
	return sig;
}

223 224 225 226 227 228 229 230 231 232
static inline void print_dropped_signal(int sig)
{
	static DEFINE_RATELIMIT_STATE(ratelimit_state, 5 * HZ, 10);

	if (!print_fatal_signals)
		return;

	if (!__ratelimit(&ratelimit_state))
		return;

233
	pr_info("%s/%d: reached RLIMIT_SIGPENDING, dropped signal %d\n",
234 235 236
				current->comm, current->pid, sig);
}

237
/**
238
 * task_set_jobctl_pending - set jobctl pending bits
239
 * @task: target task
240
 * @mask: pending bits to set
241
 *
242 243 244 245 246 247 248 249 250 251 252 253
 * Clear @mask from @task->jobctl.  @mask must be subset of
 * %JOBCTL_PENDING_MASK | %JOBCTL_STOP_CONSUME | %JOBCTL_STOP_SIGMASK |
 * %JOBCTL_TRAPPING.  If stop signo is being set, the existing signo is
 * cleared.  If @task is already being killed or exiting, this function
 * becomes noop.
 *
 * CONTEXT:
 * Must be called with @task->sighand->siglock held.
 *
 * RETURNS:
 * %true if @mask is set, %false if made noop because @task was dying.
 */
254
bool task_set_jobctl_pending(struct task_struct *task, unsigned long mask)
255 256 257 258 259 260 261 262 263 264 265 266 267 268 269
{
	BUG_ON(mask & ~(JOBCTL_PENDING_MASK | JOBCTL_STOP_CONSUME |
			JOBCTL_STOP_SIGMASK | JOBCTL_TRAPPING));
	BUG_ON((mask & JOBCTL_TRAPPING) && !(mask & JOBCTL_PENDING_MASK));

	if (unlikely(fatal_signal_pending(task) || (task->flags & PF_EXITING)))
		return false;

	if (mask & JOBCTL_STOP_SIGMASK)
		task->jobctl &= ~JOBCTL_STOP_SIGMASK;

	task->jobctl |= mask;
	return true;
}

270
/**
271
 * task_clear_jobctl_trapping - clear jobctl trapping bit
272 273
 * @task: target task
 *
274 275 276 277
 * If JOBCTL_TRAPPING is set, a ptracer is waiting for us to enter TRACED.
 * Clear it and wake up the ptracer.  Note that we don't need any further
 * locking.  @task->siglock guarantees that @task->parent points to the
 * ptracer.
278 279 280 281
 *
 * CONTEXT:
 * Must be called with @task->sighand->siglock held.
 */
282
void task_clear_jobctl_trapping(struct task_struct *task)
283
{
284 285
	if (unlikely(task->jobctl & JOBCTL_TRAPPING)) {
		task->jobctl &= ~JOBCTL_TRAPPING;
286
		smp_mb();	/* advised by wake_up_bit() */
287
		wake_up_bit(&task->jobctl, JOBCTL_TRAPPING_BIT);
288 289 290
	}
}

291
/**
292
 * task_clear_jobctl_pending - clear jobctl pending bits
293
 * @task: target task
294
 * @mask: pending bits to clear
295
 *
296 297 298
 * Clear @mask from @task->jobctl.  @mask must be subset of
 * %JOBCTL_PENDING_MASK.  If %JOBCTL_STOP_PENDING is being cleared, other
 * STOP bits are cleared together.
299
 *
300 301
 * If clearing of @mask leaves no stop or trap pending, this function calls
 * task_clear_jobctl_trapping().
302 303 304 305
 *
 * CONTEXT:
 * Must be called with @task->sighand->siglock held.
 */
306
void task_clear_jobctl_pending(struct task_struct *task, unsigned long mask)
307
{
308 309 310 311 312 313
	BUG_ON(mask & ~JOBCTL_PENDING_MASK);

	if (mask & JOBCTL_STOP_PENDING)
		mask |= JOBCTL_STOP_CONSUME | JOBCTL_STOP_DEQUEUED;

	task->jobctl &= ~mask;
314 315 316

	if (!(task->jobctl & JOBCTL_PENDING_MASK))
		task_clear_jobctl_trapping(task);
317 318 319 320 321 322
}

/**
 * task_participate_group_stop - participate in a group stop
 * @task: task participating in a group stop
 *
323
 * @task has %JOBCTL_STOP_PENDING set and is participating in a group stop.
324
 * Group stop states are cleared and the group stop count is consumed if
325
 * %JOBCTL_STOP_CONSUME was set.  If the consumption completes the group
326
 * stop, the appropriate %SIGNAL_* flags are set.
327 328 329
 *
 * CONTEXT:
 * Must be called with @task->sighand->siglock held.
330 331 332 333
 *
 * RETURNS:
 * %true if group stop completion should be notified to the parent, %false
 * otherwise.
334 335 336 337
 */
static bool task_participate_group_stop(struct task_struct *task)
{
	struct signal_struct *sig = task->signal;
338
	bool consume = task->jobctl & JOBCTL_STOP_CONSUME;
339

340
	WARN_ON_ONCE(!(task->jobctl & JOBCTL_STOP_PENDING));
341

342
	task_clear_jobctl_pending(task, JOBCTL_STOP_PENDING);
343 344 345 346 347 348 349

	if (!consume)
		return false;

	if (!WARN_ON_ONCE(sig->group_stop_count == 0))
		sig->group_stop_count--;

350 351 352 353 354
	/*
	 * Tell the caller to notify completion iff we are entering into a
	 * fresh group stop.  Read comment in do_signal_stop() for details.
	 */
	if (!sig->group_stop_count && !(sig->flags & SIGNAL_STOP_STOPPED)) {
355
		signal_set_stop_flags(sig, SIGNAL_STOP_STOPPED);
356 357 358 359 360
		return true;
	}
	return false;
}

361 362 363
/*
 * allocate a new signal queue record
 * - this may be called without locks if and only if t == current, otherwise an
364
 *   appropriate lock must be held to stop the target task from exiting
365
 */
366 367
static struct sigqueue *
__sigqueue_alloc(int sig, struct task_struct *t, gfp_t flags, int override_rlimit)
L
Linus Torvalds 已提交
368 369
{
	struct sigqueue *q = NULL;
370
	struct user_struct *user;
L
Linus Torvalds 已提交
371

372
	/*
373 374
	 * Protect access to @t credentials. This can go away when all
	 * callers hold rcu read lock.
375
	 */
376
	rcu_read_lock();
D
David Howells 已提交
377
	user = get_uid(__task_cred(t)->user);
378
	atomic_inc(&user->sigpending);
379
	rcu_read_unlock();
380

L
Linus Torvalds 已提交
381
	if (override_rlimit ||
382
	    atomic_read(&user->sigpending) <=
383
			task_rlimit(t, RLIMIT_SIGPENDING)) {
L
Linus Torvalds 已提交
384
		q = kmem_cache_alloc(sigqueue_cachep, flags);
385 386 387 388
	} else {
		print_dropped_signal(sig);
	}

L
Linus Torvalds 已提交
389
	if (unlikely(q == NULL)) {
390
		atomic_dec(&user->sigpending);
D
David Howells 已提交
391
		free_uid(user);
L
Linus Torvalds 已提交
392 393 394
	} else {
		INIT_LIST_HEAD(&q->list);
		q->flags = 0;
D
David Howells 已提交
395
		q->user = user;
L
Linus Torvalds 已提交
396
	}
D
David Howells 已提交
397 398

	return q;
L
Linus Torvalds 已提交
399 400
}

401
static void __sigqueue_free(struct sigqueue *q)
L
Linus Torvalds 已提交
402 403 404 405 406 407 408 409
{
	if (q->flags & SIGQUEUE_PREALLOC)
		return;
	atomic_dec(&q->user->sigpending);
	free_uid(q->user);
	kmem_cache_free(sigqueue_cachep, q);
}

410
void flush_sigqueue(struct sigpending *queue)
L
Linus Torvalds 已提交
411 412 413 414 415 416 417 418 419 420 421 422
{
	struct sigqueue *q;

	sigemptyset(&queue->signal);
	while (!list_empty(&queue->list)) {
		q = list_entry(queue->list.next, struct sigqueue , list);
		list_del_init(&q->list);
		__sigqueue_free(q);
	}
}

/*
423
 * Flush all pending signals for this kthread.
L
Linus Torvalds 已提交
424
 */
425
void flush_signals(struct task_struct *t)
L
Linus Torvalds 已提交
426 427 428 429
{
	unsigned long flags;

	spin_lock_irqsave(&t->sighand->siglock, flags);
430 431 432
	clear_tsk_thread_flag(t, TIF_SIGPENDING);
	flush_sigqueue(&t->pending);
	flush_sigqueue(&t->signal->shared_pending);
L
Linus Torvalds 已提交
433 434 435
	spin_unlock_irqrestore(&t->sighand->siglock, flags);
}

436
#ifdef CONFIG_POSIX_TIMERS
437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469
static void __flush_itimer_signals(struct sigpending *pending)
{
	sigset_t signal, retain;
	struct sigqueue *q, *n;

	signal = pending->signal;
	sigemptyset(&retain);

	list_for_each_entry_safe(q, n, &pending->list, list) {
		int sig = q->info.si_signo;

		if (likely(q->info.si_code != SI_TIMER)) {
			sigaddset(&retain, sig);
		} else {
			sigdelset(&signal, sig);
			list_del_init(&q->list);
			__sigqueue_free(q);
		}
	}

	sigorsets(&pending->signal, &signal, &retain);
}

void flush_itimer_signals(void)
{
	struct task_struct *tsk = current;
	unsigned long flags;

	spin_lock_irqsave(&tsk->sighand->siglock, flags);
	__flush_itimer_signals(&tsk->pending);
	__flush_itimer_signals(&tsk->signal->shared_pending);
	spin_unlock_irqrestore(&tsk->sighand->siglock, flags);
}
470
#endif
471

472 473 474 475 476 477 478 479 480 481
void ignore_signals(struct task_struct *t)
{
	int i;

	for (i = 0; i < _NSIG; ++i)
		t->sighand->action[i].sa.sa_handler = SIG_IGN;

	flush_signals(t);
}

L
Linus Torvalds 已提交
482 483 484 485 486 487 488 489 490 491 492 493 494
/*
 * Flush all handlers for a task.
 */

void
flush_signal_handlers(struct task_struct *t, int force_default)
{
	int i;
	struct k_sigaction *ka = &t->sighand->action[0];
	for (i = _NSIG ; i != 0 ; i--) {
		if (force_default || ka->sa.sa_handler != SIG_IGN)
			ka->sa.sa_handler = SIG_DFL;
		ka->sa.sa_flags = 0;
495
#ifdef __ARCH_HAS_SA_RESTORER
496 497
		ka->sa.sa_restorer = NULL;
#endif
L
Linus Torvalds 已提交
498 499 500 501 502
		sigemptyset(&ka->sa.sa_mask);
		ka++;
	}
}

503 504
int unhandled_signal(struct task_struct *tsk, int sig)
{
505
	void __user *handler = tsk->sighand->action[sig-1].sa.sa_handler;
506
	if (is_global_init(tsk))
507
		return 1;
508
	if (handler != SIG_IGN && handler != SIG_DFL)
509
		return 0;
T
Tejun Heo 已提交
510 511
	/* if ptraced, let the tracer determine */
	return !tsk->ptrace;
512 513
}

514
static void collect_signal(int sig, struct sigpending *list, siginfo_t *info)
L
Linus Torvalds 已提交
515 516 517 518 519 520 521 522 523
{
	struct sigqueue *q, *first = NULL;

	/*
	 * Collect the siginfo appropriate to this signal.  Check if
	 * there is another siginfo for the same signal.
	*/
	list_for_each_entry(q, &list->list, list) {
		if (q->info.si_signo == sig) {
524 525
			if (first)
				goto still_pending;
L
Linus Torvalds 已提交
526 527 528
			first = q;
		}
	}
529 530 531

	sigdelset(&list->signal, sig);

L
Linus Torvalds 已提交
532
	if (first) {
533
still_pending:
L
Linus Torvalds 已提交
534 535 536 537
		list_del_init(&first->list);
		copy_siginfo(info, &first->info);
		__sigqueue_free(first);
	} else {
538 539 540 541
		/*
		 * Ok, it wasn't in the queue.  This must be
		 * a fast-pathed signal or we must have been
		 * out of queue space.  So zero out the info.
L
Linus Torvalds 已提交
542 543 544
		 */
		info->si_signo = sig;
		info->si_errno = 0;
545
		info->si_code = SI_USER;
L
Linus Torvalds 已提交
546 547 548 549 550 551 552 553
		info->si_pid = 0;
		info->si_uid = 0;
	}
}

static int __dequeue_signal(struct sigpending *pending, sigset_t *mask,
			siginfo_t *info)
{
554
	int sig = next_signal(pending, mask);
L
Linus Torvalds 已提交
555

556
	if (sig)
557
		collect_signal(sig, pending, info);
L
Linus Torvalds 已提交
558 559 560 561
	return sig;
}

/*
562
 * Dequeue a signal and return the element to the caller, which is
L
Linus Torvalds 已提交
563 564 565 566 567 568
 * expected to free it.
 *
 * All callers have to hold the siglock.
 */
int dequeue_signal(struct task_struct *tsk, sigset_t *mask, siginfo_t *info)
{
569
	int signr;
570 571 572 573

	/* We only dequeue private signals from ourselves, we don't let
	 * signalfd steal them
	 */
D
Davide Libenzi 已提交
574
	signr = __dequeue_signal(&tsk->pending, mask, info);
575
	if (!signr) {
L
Linus Torvalds 已提交
576 577
		signr = __dequeue_signal(&tsk->signal->shared_pending,
					 mask, info);
578
#ifdef CONFIG_POSIX_TIMERS
579 580 581 582 583 584
		/*
		 * itimer signal ?
		 *
		 * itimers are process shared and we restart periodic
		 * itimers in the signal delivery path to prevent DoS
		 * attacks in the high resolution timer case. This is
585
		 * compliant with the old way of self-restarting
586 587 588 589 590 591 592 593 594 595
		 * itimers, as the SIGALRM is a legacy signal and only
		 * queued once. Changing the restart behaviour to
		 * restart the timer in the signal dequeue path is
		 * reducing the timer noise on heavy loaded !highres
		 * systems too.
		 */
		if (unlikely(signr == SIGALRM)) {
			struct hrtimer *tmr = &tsk->signal->real_timer;

			if (!hrtimer_is_queued(tmr) &&
T
Thomas Gleixner 已提交
596
			    tsk->signal->it_real_incr != 0) {
597 598 599 600 601
				hrtimer_forward(tmr, tmr->base->get_time(),
						tsk->signal->it_real_incr);
				hrtimer_restart(tmr);
			}
		}
602
#endif
603
	}
604

D
Davide Libenzi 已提交
605
	recalc_sigpending();
606 607 608 609
	if (!signr)
		return 0;

	if (unlikely(sig_kernel_stop(signr))) {
610 611 612 613 614 615 616 617 618 619 620 621
		/*
		 * Set a marker that we have dequeued a stop signal.  Our
		 * caller might release the siglock and then the pending
		 * stop signal it is about to process is no longer in the
		 * pending bitmasks, but must still be cleared by a SIGCONT
		 * (and overruled by a SIGKILL).  So those cases clear this
		 * shared flag after we've set it.  Note that this flag may
		 * remain set after the signal we return is ignored or
		 * handled.  That doesn't matter because its only purpose
		 * is to alert stop-signal processing code when another
		 * processor has come along and cleared the flag.
		 */
622
		current->jobctl |= JOBCTL_STOP_DEQUEUED;
623
	}
624
#ifdef CONFIG_POSIX_TIMERS
625
	if ((info->si_code & __SI_MASK) == __SI_TIMER && info->si_sys_private) {
L
Linus Torvalds 已提交
626 627 628 629 630 631 632 633 634 635
		/*
		 * Release the siglock to ensure proper locking order
		 * of timer locks outside of siglocks.  Note, we leave
		 * irqs disabled here, since the posix-timers code is
		 * about to disable them again anyway.
		 */
		spin_unlock(&tsk->sighand->siglock);
		do_schedule_next_timer(info);
		spin_lock(&tsk->sighand->siglock);
	}
636
#endif
L
Linus Torvalds 已提交
637 638 639 640 641 642 643 644 645 646 647 648 649 650
	return signr;
}

/*
 * Tell a process that it has a new active signal..
 *
 * NOTE! we rely on the previous spin_lock to
 * lock interrupts for us! We can only be called with
 * "siglock" held, and the local interrupt must
 * have been disabled when that got acquired!
 *
 * No need to set need_resched since signal event passing
 * goes through ->blocked
 */
651
void signal_wake_up_state(struct task_struct *t, unsigned int state)
L
Linus Torvalds 已提交
652 653 654
{
	set_tsk_thread_flag(t, TIF_SIGPENDING);
	/*
655
	 * TASK_WAKEKILL also means wake it up in the stopped/traced/killable
M
Matthew Wilcox 已提交
656
	 * case. We don't check t->state here because there is a race with it
L
Linus Torvalds 已提交
657 658 659 660
	 * executing another processor and just now entering stopped state.
	 * By using wake_up_state, we ensure the process will wake up and
	 * handle its death signal.
	 */
661
	if (!wake_up_state(t, state | TASK_INTERRUPTIBLE))
L
Linus Torvalds 已提交
662 663 664
		kick_process(t);
}

665 666 667 668 669 670
/*
 * Remove signals in mask from the pending set and queue.
 * Returns 1 if any signals were found.
 *
 * All callers must be holding the siglock.
 */
671
static int flush_sigqueue_mask(sigset_t *mask, struct sigpending *s)
672 673 674 675 676 677 678 679
{
	struct sigqueue *q, *n;
	sigset_t m;

	sigandsets(&m, mask, &s->signal);
	if (sigisemptyset(&m))
		return 0;

680
	sigandnsets(&s->signal, &s->signal, mask);
681 682 683 684 685 686 687 688
	list_for_each_entry_safe(q, n, &s->list, list) {
		if (sigismember(mask, q->info.si_signo)) {
			list_del_init(&q->list);
			__sigqueue_free(q);
		}
	}
	return 1;
}
L
Linus Torvalds 已提交
689

690 691 692 693 694 695 696 697 698 699 700
static inline int is_si_special(const struct siginfo *info)
{
	return info <= SEND_SIG_FORCED;
}

static inline bool si_fromuser(const struct siginfo *info)
{
	return info == SEND_SIG_NOINFO ||
		(!is_si_special(info) && SI_FROMUSER(info));
}

701 702 703 704 705 706 707 708
/*
 * called with RCU read lock from check_kill_permission()
 */
static int kill_ok_by_cred(struct task_struct *t)
{
	const struct cred *cred = current_cred();
	const struct cred *tcred = __task_cred(t);

709 710 711 712
	if (uid_eq(cred->euid, tcred->suid) ||
	    uid_eq(cred->euid, tcred->uid)  ||
	    uid_eq(cred->uid,  tcred->suid) ||
	    uid_eq(cred->uid,  tcred->uid))
713 714
		return 1;

715
	if (ns_capable(tcred->user_ns, CAP_KILL))
716 717 718 719 720
		return 1;

	return 0;
}

L
Linus Torvalds 已提交
721 722
/*
 * Bad permissions for sending the signal
723
 * - the caller must hold the RCU read lock
L
Linus Torvalds 已提交
724 725 726 727
 */
static int check_kill_permission(int sig, struct siginfo *info,
				 struct task_struct *t)
{
728
	struct pid *sid;
729 730
	int error;

731
	if (!valid_signal(sig))
732 733
		return -EINVAL;

734
	if (!si_fromuser(info))
735
		return 0;
A
Amy Griffis 已提交
736

737 738
	error = audit_signal_info(sig, t); /* Let audit system see the signal */
	if (error)
L
Linus Torvalds 已提交
739
		return error;
740

741
	if (!same_thread_group(current, t) &&
742
	    !kill_ok_by_cred(t)) {
743 744 745 746 747 748 749 750 751 752 753 754 755
		switch (sig) {
		case SIGCONT:
			sid = task_session(t);
			/*
			 * We don't return the error if sid == NULL. The
			 * task was unhashed, the caller must notice this.
			 */
			if (!sid || sid == task_session(current))
				break;
		default:
			return -EPERM;
		}
	}
756

A
Amy Griffis 已提交
757
	return security_task_kill(t, info, sig, 0);
L
Linus Torvalds 已提交
758 759
}

760 761 762 763 764 765 766 767
/**
 * ptrace_trap_notify - schedule trap to notify ptracer
 * @t: tracee wanting to notify tracer
 *
 * This function schedules sticky ptrace trap which is cleared on the next
 * TRAP_STOP to notify ptracer of an event.  @t must have been seized by
 * ptracer.
 *
T
Tejun Heo 已提交
768 769 770 771 772
 * If @t is running, STOP trap will be taken.  If trapped for STOP and
 * ptracer is listening for events, tracee is woken up so that it can
 * re-trap for the new event.  If trapped otherwise, STOP trap will be
 * eventually taken without returning to userland after the existing traps
 * are finished by PTRACE_CONT.
773 774 775 776 777 778 779 780 781 782
 *
 * CONTEXT:
 * Must be called with @task->sighand->siglock held.
 */
static void ptrace_trap_notify(struct task_struct *t)
{
	WARN_ON_ONCE(!(t->ptrace & PT_SEIZED));
	assert_spin_locked(&t->sighand->siglock);

	task_set_jobctl_pending(t, JOBCTL_TRAP_NOTIFY);
783
	ptrace_signal_wake_up(t, t->jobctl & JOBCTL_LISTENING);
784 785
}

L
Linus Torvalds 已提交
786
/*
787 788
 * Handle magic process-wide effects of stop/continue signals. Unlike
 * the signal actions, these happen immediately at signal-generation
L
Linus Torvalds 已提交
789 790
 * time regardless of blocking, ignoring, or handling.  This does the
 * actual continuing for SIGCONT, but not the actual stopping for stop
791 792 793 794
 * signals. The process stop is done as a signal action for SIG_DFL.
 *
 * Returns true if the signal should be actually delivered, otherwise
 * it should be dropped.
L
Linus Torvalds 已提交
795
 */
796
static bool prepare_signal(int sig, struct task_struct *p, bool force)
L
Linus Torvalds 已提交
797
{
798
	struct signal_struct *signal = p->signal;
L
Linus Torvalds 已提交
799
	struct task_struct *t;
800
	sigset_t flush;
L
Linus Torvalds 已提交
801

802
	if (signal->flags & (SIGNAL_GROUP_EXIT | SIGNAL_GROUP_COREDUMP)) {
803
		if (!(signal->flags & SIGNAL_GROUP_EXIT))
804
			return sig == SIGKILL;
L
Linus Torvalds 已提交
805
		/*
806
		 * The process is in the middle of dying, nothing to do.
L
Linus Torvalds 已提交
807
		 */
808
	} else if (sig_kernel_stop(sig)) {
L
Linus Torvalds 已提交
809 810 811
		/*
		 * This is a stop signal.  Remove SIGCONT from all queues.
		 */
812
		siginitset(&flush, sigmask(SIGCONT));
813
		flush_sigqueue_mask(&flush, &signal->shared_pending);
814
		for_each_thread(p, t)
815
			flush_sigqueue_mask(&flush, &t->pending);
L
Linus Torvalds 已提交
816
	} else if (sig == SIGCONT) {
817
		unsigned int why;
L
Linus Torvalds 已提交
818
		/*
819
		 * Remove all stop signals from all queues, wake all threads.
L
Linus Torvalds 已提交
820
		 */
821
		siginitset(&flush, SIG_KERNEL_STOP_MASK);
822
		flush_sigqueue_mask(&flush, &signal->shared_pending);
823
		for_each_thread(p, t) {
824
			flush_sigqueue_mask(&flush, &t->pending);
825
			task_clear_jobctl_pending(t, JOBCTL_STOP_PENDING);
826 827 828 829
			if (likely(!(t->ptrace & PT_SEIZED)))
				wake_up_state(t, __TASK_STOPPED);
			else
				ptrace_trap_notify(t);
830
		}
L
Linus Torvalds 已提交
831

832 833 834 835 836 837 838 839 840
		/*
		 * Notify the parent with CLD_CONTINUED if we were stopped.
		 *
		 * If we were in the middle of a group stop, we pretend it
		 * was already finished, and then continued. Since SIGCHLD
		 * doesn't queue we report only CLD_STOPPED, as if the next
		 * CLD_CONTINUED was dropped.
		 */
		why = 0;
841
		if (signal->flags & SIGNAL_STOP_STOPPED)
842
			why |= SIGNAL_CLD_CONTINUED;
843
		else if (signal->group_stop_count)
844 845 846
			why |= SIGNAL_CLD_STOPPED;

		if (why) {
847
			/*
848
			 * The first thread which returns from do_signal_stop()
849 850 851
			 * will take ->siglock, notice SIGNAL_CLD_MASK, and
			 * notify its parent. See get_signal_to_deliver().
			 */
852
			signal_set_stop_flags(signal, why | SIGNAL_STOP_CONTINUED);
853 854
			signal->group_stop_count = 0;
			signal->group_exit_code = 0;
L
Linus Torvalds 已提交
855 856
		}
	}
857

858
	return !sig_ignored(p, sig, force);
L
Linus Torvalds 已提交
859 860
}

861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881
/*
 * Test if P wants to take SIG.  After we've checked all threads with this,
 * it's equivalent to finding no threads not blocking SIG.  Any threads not
 * blocking SIG were ruled out because they are not running and already
 * have pending signals.  Such threads will dequeue from the shared queue
 * as soon as they're available, so putting the signal on the shared queue
 * will be equivalent to sending it to one such thread.
 */
static inline int wants_signal(int sig, struct task_struct *p)
{
	if (sigismember(&p->blocked, sig))
		return 0;
	if (p->flags & PF_EXITING)
		return 0;
	if (sig == SIGKILL)
		return 1;
	if (task_is_stopped_or_traced(p))
		return 0;
	return task_curr(p) || !signal_pending(p);
}

882
static void complete_signal(int sig, struct task_struct *p, int group)
883 884 885 886 887 888 889 890 891 892 893 894
{
	struct signal_struct *signal = p->signal;
	struct task_struct *t;

	/*
	 * Now find a thread we can wake up to take the signal off the queue.
	 *
	 * If the main thread wants the signal, it gets first crack.
	 * Probably the least surprising to the average bear.
	 */
	if (wants_signal(sig, p))
		t = p;
895
	else if (!group || thread_group_empty(p))
896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922
		/*
		 * There is just one thread and it does not need to be woken.
		 * It will dequeue unblocked signals before it runs again.
		 */
		return;
	else {
		/*
		 * Otherwise try to find a suitable thread.
		 */
		t = signal->curr_target;
		while (!wants_signal(sig, t)) {
			t = next_thread(t);
			if (t == signal->curr_target)
				/*
				 * No thread needs to be woken.
				 * Any eligible threads will see
				 * the signal in the queue soon.
				 */
				return;
		}
		signal->curr_target = t;
	}

	/*
	 * Found a killable thread.  If the signal will be fatal,
	 * then start taking the whole group down immediately.
	 */
923 924
	if (sig_fatal(p, sig) &&
	    !(signal->flags & (SIGNAL_UNKILLABLE | SIGNAL_GROUP_EXIT)) &&
925
	    !sigismember(&t->real_blocked, sig) &&
T
Tejun Heo 已提交
926
	    (sig == SIGKILL || !t->ptrace)) {
927 928 929 930 931 932 933 934 935 936 937 938 939 940 941
		/*
		 * This signal will be fatal to the whole group.
		 */
		if (!sig_kernel_coredump(sig)) {
			/*
			 * Start a group exit and wake everybody up.
			 * This way we don't have other threads
			 * running and doing things after a slower
			 * thread has the fatal signal pending.
			 */
			signal->flags = SIGNAL_GROUP_EXIT;
			signal->group_exit_code = sig;
			signal->group_stop_count = 0;
			t = p;
			do {
942
				task_clear_jobctl_pending(t, JOBCTL_PENDING_MASK);
943 944 945 946 947 948 949 950 951 952 953 954 955 956 957
				sigaddset(&t->pending.signal, SIGKILL);
				signal_wake_up(t, 1);
			} while_each_thread(p, t);
			return;
		}
	}

	/*
	 * The signal is already in the shared-pending queue.
	 * Tell the chosen thread to wake up and dequeue it.
	 */
	signal_wake_up(t, sig == SIGKILL);
	return;
}

958 959 960 961 962
static inline int legacy_queue(struct sigpending *signals, int sig)
{
	return (sig < SIGRTMIN) && sigismember(&signals->signal, sig);
}

963 964 965 966 967 968 969 970 971
#ifdef CONFIG_USER_NS
static inline void userns_fixup_signal_uid(struct siginfo *info, struct task_struct *t)
{
	if (current_user_ns() == task_cred_xxx(t, user_ns))
		return;

	if (SI_FROMKERNEL(info))
		return;

972 973 974 975
	rcu_read_lock();
	info->si_uid = from_kuid_munged(task_cred_xxx(t, user_ns),
					make_kuid(current_user_ns(), info->si_uid));
	rcu_read_unlock();
976 977 978 979 980 981 982 983
}
#else
static inline void userns_fixup_signal_uid(struct siginfo *info, struct task_struct *t)
{
	return;
}
#endif

984 985
static int __send_signal(int sig, struct siginfo *info, struct task_struct *t,
			int group, int from_ancestor_ns)
L
Linus Torvalds 已提交
986
{
987
	struct sigpending *pending;
988
	struct sigqueue *q;
989
	int override_rlimit;
990
	int ret = 0, result;
991

992
	assert_spin_locked(&t->sighand->siglock);
993

994
	result = TRACE_SIGNAL_IGNORED;
995 996
	if (!prepare_signal(sig, t,
			from_ancestor_ns || (info == SEND_SIG_FORCED)))
997
		goto ret;
998 999

	pending = group ? &t->signal->shared_pending : &t->pending;
1000 1001 1002 1003 1004
	/*
	 * Short-circuit ignored signals and support queuing
	 * exactly one non-rt signal, so that we can get more
	 * detailed information about the cause of the signal.
	 */
1005
	result = TRACE_SIGNAL_ALREADY_PENDING;
1006
	if (legacy_queue(pending, sig))
1007 1008 1009
		goto ret;

	result = TRACE_SIGNAL_DELIVERED;
L
Linus Torvalds 已提交
1010 1011 1012 1013
	/*
	 * fast-pathed signals for kernel-internal things like SIGSTOP
	 * or SIGKILL.
	 */
1014
	if (info == SEND_SIG_FORCED)
L
Linus Torvalds 已提交
1015 1016
		goto out_set;

1017 1018 1019 1020 1021 1022 1023 1024 1025
	/*
	 * Real-time signals must be queued if sent by sigqueue, or
	 * some other real-time mechanism.  It is implementation
	 * defined whether kill() does so.  We attempt to do so, on
	 * the principle of least surprise, but since kill is not
	 * allowed to fail with EAGAIN when low on memory we just
	 * make sure at least one signal gets delivered and don't
	 * pass on the info struct.
	 */
1026 1027 1028 1029 1030
	if (sig < SIGRTMIN)
		override_rlimit = (is_si_special(info) || info->si_code >= 0);
	else
		override_rlimit = 0;

1031
	q = __sigqueue_alloc(sig, t, GFP_ATOMIC | __GFP_NOTRACK_FALSE_POSITIVE,
1032
		override_rlimit);
L
Linus Torvalds 已提交
1033
	if (q) {
1034
		list_add_tail(&q->list, &pending->list);
L
Linus Torvalds 已提交
1035
		switch ((unsigned long) info) {
1036
		case (unsigned long) SEND_SIG_NOINFO:
L
Linus Torvalds 已提交
1037 1038 1039
			q->info.si_signo = sig;
			q->info.si_errno = 0;
			q->info.si_code = SI_USER;
1040
			q->info.si_pid = task_tgid_nr_ns(current,
1041
							task_active_pid_ns(t));
1042
			q->info.si_uid = from_kuid_munged(current_user_ns(), current_uid());
L
Linus Torvalds 已提交
1043
			break;
1044
		case (unsigned long) SEND_SIG_PRIV:
L
Linus Torvalds 已提交
1045 1046 1047 1048 1049 1050 1051 1052
			q->info.si_signo = sig;
			q->info.si_errno = 0;
			q->info.si_code = SI_KERNEL;
			q->info.si_pid = 0;
			q->info.si_uid = 0;
			break;
		default:
			copy_siginfo(&q->info, info);
1053 1054
			if (from_ancestor_ns)
				q->info.si_pid = 0;
L
Linus Torvalds 已提交
1055 1056
			break;
		}
1057 1058 1059

		userns_fixup_signal_uid(&q->info, t);

1060
	} else if (!is_si_special(info)) {
1061 1062 1063 1064 1065 1066
		if (sig >= SIGRTMIN && info->si_code != SI_USER) {
			/*
			 * Queue overflow, abort.  We may abort if the
			 * signal was rt and sent by user using something
			 * other than kill().
			 */
1067 1068 1069
			result = TRACE_SIGNAL_OVERFLOW_FAIL;
			ret = -EAGAIN;
			goto ret;
1070 1071 1072 1073 1074
		} else {
			/*
			 * This is a silent loss of information.  We still
			 * send the signal, but the *info bits are lost.
			 */
1075
			result = TRACE_SIGNAL_LOSE_INFO;
1076
		}
L
Linus Torvalds 已提交
1077 1078 1079
	}

out_set:
1080
	signalfd_notify(t, sig);
1081
	sigaddset(&pending->signal, sig);
1082
	complete_signal(sig, t, group);
1083 1084 1085
ret:
	trace_signal_generate(sig, info, t, group, result);
	return ret;
L
Linus Torvalds 已提交
1086 1087
}

1088 1089 1090
static int send_signal(int sig, struct siginfo *info, struct task_struct *t,
			int group)
{
1091 1092 1093
	int from_ancestor_ns = 0;

#ifdef CONFIG_PID_NS
1094 1095
	from_ancestor_ns = si_fromuser(info) &&
			   !task_pid_nr_ns(current, task_active_pid_ns(t));
1096 1097 1098
#endif

	return __send_signal(sig, info, t, group, from_ancestor_ns);
1099 1100
}

1101
static void print_fatal_signal(int signr)
I
Ingo Molnar 已提交
1102
{
1103
	struct pt_regs *regs = signal_pt_regs();
1104
	pr_info("potentially unexpected fatal signal %d.\n", signr);
I
Ingo Molnar 已提交
1105

A
Al Viro 已提交
1106
#if defined(__i386__) && !defined(__arch_um__)
1107
	pr_info("code at %08lx: ", regs->ip);
I
Ingo Molnar 已提交
1108 1109 1110 1111 1112
	{
		int i;
		for (i = 0; i < 16; i++) {
			unsigned char insn;

1113 1114
			if (get_user(insn, (unsigned char *)(regs->ip + i)))
				break;
1115
			pr_cont("%02x ", insn);
I
Ingo Molnar 已提交
1116 1117
		}
	}
1118
	pr_cont("\n");
I
Ingo Molnar 已提交
1119
#endif
1120
	preempt_disable();
I
Ingo Molnar 已提交
1121
	show_regs(regs);
1122
	preempt_enable();
I
Ingo Molnar 已提交
1123 1124 1125 1126 1127 1128 1129 1130 1131 1132
}

static int __init setup_print_fatal_signals(char *str)
{
	get_option (&str, &print_fatal_signals);

	return 1;
}

__setup("print-fatal-signals=", setup_print_fatal_signals);
L
Linus Torvalds 已提交
1133

1134 1135 1136 1137 1138 1139
int
__group_send_sig_info(int sig, struct siginfo *info, struct task_struct *p)
{
	return send_signal(sig, info, p, 1);
}

L
Linus Torvalds 已提交
1140 1141 1142
static int
specific_send_sig_info(int sig, struct siginfo *info, struct task_struct *t)
{
1143
	return send_signal(sig, info, t, 0);
L
Linus Torvalds 已提交
1144 1145
}

1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159
int do_send_sig_info(int sig, struct siginfo *info, struct task_struct *p,
			bool group)
{
	unsigned long flags;
	int ret = -ESRCH;

	if (lock_task_sighand(p, &flags)) {
		ret = send_signal(sig, info, p, group);
		unlock_task_sighand(p, &flags);
	}

	return ret;
}

L
Linus Torvalds 已提交
1160 1161 1162
/*
 * Force a signal that the process can't ignore: if necessary
 * we unblock the signal and change any SIG_IGN to SIG_DFL.
1163 1164 1165 1166 1167
 *
 * Note: If we unblock the signal, we always reset it to SIG_DFL,
 * since we do not want to have a signal handler that was blocked
 * be invoked when user space had explicitly blocked it.
 *
1168 1169
 * We don't want to have recursive SIGSEGV's etc, for example,
 * that is why we also clear SIGNAL_UNKILLABLE.
L
Linus Torvalds 已提交
1170 1171 1172 1173 1174
 */
int
force_sig_info(int sig, struct siginfo *info, struct task_struct *t)
{
	unsigned long int flags;
1175 1176
	int ret, blocked, ignored;
	struct k_sigaction *action;
L
Linus Torvalds 已提交
1177 1178

	spin_lock_irqsave(&t->sighand->siglock, flags);
1179 1180 1181 1182 1183 1184 1185
	action = &t->sighand->action[sig-1];
	ignored = action->sa.sa_handler == SIG_IGN;
	blocked = sigismember(&t->blocked, sig);
	if (blocked || ignored) {
		action->sa.sa_handler = SIG_DFL;
		if (blocked) {
			sigdelset(&t->blocked, sig);
R
Roland McGrath 已提交
1186
			recalc_sigpending_and_wake(t);
1187
		}
L
Linus Torvalds 已提交
1188
	}
1189 1190
	if (action->sa.sa_handler == SIG_DFL)
		t->signal->flags &= ~SIGNAL_UNKILLABLE;
L
Linus Torvalds 已提交
1191 1192 1193 1194 1195 1196 1197 1198 1199
	ret = specific_send_sig_info(sig, info, t);
	spin_unlock_irqrestore(&t->sighand->siglock, flags);

	return ret;
}

/*
 * Nuke all other threads in the group.
 */
1200
int zap_other_threads(struct task_struct *p)
L
Linus Torvalds 已提交
1201
{
1202 1203
	struct task_struct *t = p;
	int count = 0;
L
Linus Torvalds 已提交
1204 1205 1206

	p->signal->group_stop_count = 0;

1207
	while_each_thread(p, t) {
1208
		task_clear_jobctl_pending(t, JOBCTL_PENDING_MASK);
1209 1210 1211
		count++;

		/* Don't bother with already dead threads */
L
Linus Torvalds 已提交
1212 1213 1214 1215 1216
		if (t->exit_state)
			continue;
		sigaddset(&t->pending.signal, SIGKILL);
		signal_wake_up(t, 1);
	}
1217 1218

	return count;
L
Linus Torvalds 已提交
1219 1220
}

1221 1222
struct sighand_struct *__lock_task_sighand(struct task_struct *tsk,
					   unsigned long *flags)
1223 1224 1225 1226
{
	struct sighand_struct *sighand;

	for (;;) {
1227 1228 1229 1230
		/*
		 * Disable interrupts early to avoid deadlocks.
		 * See rcu_read_unlock() comment header for details.
		 */
1231 1232
		local_irq_save(*flags);
		rcu_read_lock();
1233
		sighand = rcu_dereference(tsk->sighand);
1234 1235 1236
		if (unlikely(sighand == NULL)) {
			rcu_read_unlock();
			local_irq_restore(*flags);
1237
			break;
1238
		}
1239 1240
		/*
		 * This sighand can be already freed and even reused, but
1241
		 * we rely on SLAB_TYPESAFE_BY_RCU and sighand_ctor() which
1242 1243 1244 1245 1246 1247 1248 1249
		 * initializes ->siglock: this slab can't go away, it has
		 * the same object type, ->siglock can't be reinitialized.
		 *
		 * We need to ensure that tsk->sighand is still the same
		 * after we take the lock, we can race with de_thread() or
		 * __exit_signal(). In the latter case the next iteration
		 * must see ->sighand == NULL.
		 */
1250 1251 1252
		spin_lock(&sighand->siglock);
		if (likely(sighand == tsk->sighand)) {
			rcu_read_unlock();
1253
			break;
1254 1255 1256 1257
		}
		spin_unlock(&sighand->siglock);
		rcu_read_unlock();
		local_irq_restore(*flags);
1258 1259 1260 1261 1262
	}

	return sighand;
}

1263 1264 1265
/*
 * send signal info to all the members of a group
 */
L
Linus Torvalds 已提交
1266 1267
int group_send_sig_info(int sig, struct siginfo *info, struct task_struct *p)
{
1268 1269 1270 1271 1272
	int ret;

	rcu_read_lock();
	ret = check_kill_permission(sig, info, p);
	rcu_read_unlock();
1273

1274 1275
	if (!ret && sig)
		ret = do_send_sig_info(sig, info, p, true);
L
Linus Torvalds 已提交
1276 1277 1278 1279 1280

	return ret;
}

/*
1281
 * __kill_pgrp_info() sends a signal to a process group: this is what the tty
L
Linus Torvalds 已提交
1282
 * control characters do (^C, ^Z etc)
1283
 * - the caller must hold at least a readlock on tasklist_lock
L
Linus Torvalds 已提交
1284
 */
1285
int __kill_pgrp_info(int sig, struct siginfo *info, struct pid *pgrp)
L
Linus Torvalds 已提交
1286 1287 1288 1289 1290 1291
{
	struct task_struct *p = NULL;
	int retval, success;

	success = 0;
	retval = -ESRCH;
1292
	do_each_pid_task(pgrp, PIDTYPE_PGID, p) {
L
Linus Torvalds 已提交
1293 1294 1295
		int err = group_send_sig_info(sig, info, p);
		success |= !err;
		retval = err;
1296
	} while_each_pid_task(pgrp, PIDTYPE_PGID, p);
L
Linus Torvalds 已提交
1297 1298 1299
	return success ? 0 : retval;
}

1300
int kill_pid_info(int sig, struct siginfo *info, struct pid *pid)
L
Linus Torvalds 已提交
1301
{
1302
	int error = -ESRCH;
L
Linus Torvalds 已提交
1303 1304
	struct task_struct *p;

1305 1306 1307 1308 1309 1310 1311 1312
	for (;;) {
		rcu_read_lock();
		p = pid_task(pid, PIDTYPE_PID);
		if (p)
			error = group_send_sig_info(sig, info, p);
		rcu_read_unlock();
		if (likely(!p || error != -ESRCH))
			return error;
1313

1314 1315 1316 1317 1318 1319
		/*
		 * The task was unhashed in between, try again.  If it
		 * is dead, pid_task() will return NULL, if we race with
		 * de_thread() it will find the new leader.
		 */
	}
L
Linus Torvalds 已提交
1320 1321
}

1322
static int kill_proc_info(int sig, struct siginfo *info, pid_t pid)
1323 1324 1325
{
	int error;
	rcu_read_lock();
1326
	error = kill_pid_info(sig, info, find_vpid(pid));
1327 1328 1329 1330
	rcu_read_unlock();
	return error;
}

1331 1332 1333 1334
static int kill_as_cred_perm(const struct cred *cred,
			     struct task_struct *target)
{
	const struct cred *pcred = __task_cred(target);
1335 1336
	if (!uid_eq(cred->euid, pcred->suid) && !uid_eq(cred->euid, pcred->uid) &&
	    !uid_eq(cred->uid,  pcred->suid) && !uid_eq(cred->uid,  pcred->uid))
1337 1338 1339 1340
		return 0;
	return 1;
}

1341
/* like kill_pid_info(), but doesn't use uid/euid of "current" */
1342 1343
int kill_pid_info_as_cred(int sig, struct siginfo *info, struct pid *pid,
			 const struct cred *cred, u32 secid)
1344 1345 1346
{
	int ret = -EINVAL;
	struct task_struct *p;
1347
	unsigned long flags;
1348 1349 1350 1351

	if (!valid_signal(sig))
		return ret;

1352
	rcu_read_lock();
1353
	p = pid_task(pid, PIDTYPE_PID);
1354 1355 1356 1357
	if (!p) {
		ret = -ESRCH;
		goto out_unlock;
	}
1358
	if (si_fromuser(info) && !kill_as_cred_perm(cred, p)) {
1359 1360 1361
		ret = -EPERM;
		goto out_unlock;
	}
1362 1363 1364
	ret = security_task_kill(p, info, sig, secid);
	if (ret)
		goto out_unlock;
1365 1366 1367 1368 1369 1370 1371

	if (sig) {
		if (lock_task_sighand(p, &flags)) {
			ret = __send_signal(sig, info, p, 1, 0);
			unlock_task_sighand(p, &flags);
		} else
			ret = -ESRCH;
1372 1373
	}
out_unlock:
1374
	rcu_read_unlock();
1375 1376
	return ret;
}
1377
EXPORT_SYMBOL_GPL(kill_pid_info_as_cred);
L
Linus Torvalds 已提交
1378 1379 1380 1381 1382 1383 1384 1385

/*
 * kill_something_info() interprets pid in interesting ways just like kill(2).
 *
 * POSIX specifies that kill(-1,sig) is unspecified, but what we have
 * is probably wrong.  Should make it like BSD or SYSV.
 */

1386
static int kill_something_info(int sig, struct siginfo *info, pid_t pid)
L
Linus Torvalds 已提交
1387
{
1388
	int ret;
1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401

	if (pid > 0) {
		rcu_read_lock();
		ret = kill_pid_info(sig, info, find_vpid(pid));
		rcu_read_unlock();
		return ret;
	}

	read_lock(&tasklist_lock);
	if (pid != -1) {
		ret = __kill_pgrp_info(sig, info,
				pid ? find_vpid(-pid) : task_pgrp(current));
	} else {
L
Linus Torvalds 已提交
1402 1403 1404 1405
		int retval = 0, count = 0;
		struct task_struct * p;

		for_each_process(p) {
1406 1407
			if (task_pid_vnr(p) > 1 &&
					!same_thread_group(p, current)) {
L
Linus Torvalds 已提交
1408 1409 1410 1411 1412 1413
				int err = group_send_sig_info(sig, info, p);
				++count;
				if (err != -EPERM)
					retval = err;
			}
		}
1414
		ret = count ? retval : -ESRCH;
L
Linus Torvalds 已提交
1415
	}
1416 1417
	read_unlock(&tasklist_lock);

1418
	return ret;
L
Linus Torvalds 已提交
1419 1420 1421 1422 1423 1424
}

/*
 * These are for backward compatibility with the rest of the kernel source.
 */

1425
int send_sig_info(int sig, struct siginfo *info, struct task_struct *p)
L
Linus Torvalds 已提交
1426 1427 1428 1429 1430
{
	/*
	 * Make sure legacy kernel users don't send in bad values
	 * (normal paths check this in check_kill_permission).
	 */
1431
	if (!valid_signal(sig))
L
Linus Torvalds 已提交
1432 1433
		return -EINVAL;

1434
	return do_send_sig_info(sig, info, p, false);
L
Linus Torvalds 已提交
1435 1436
}

1437 1438 1439
#define __si_special(priv) \
	((priv) ? SEND_SIG_PRIV : SEND_SIG_NOINFO)

L
Linus Torvalds 已提交
1440 1441 1442
int
send_sig(int sig, struct task_struct *p, int priv)
{
1443
	return send_sig_info(sig, __si_special(priv), p);
L
Linus Torvalds 已提交
1444 1445 1446 1447 1448
}

void
force_sig(int sig, struct task_struct *p)
{
1449
	force_sig_info(sig, SEND_SIG_PRIV, p);
L
Linus Torvalds 已提交
1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470
}

/*
 * When things go south during signal handling, we
 * will force a SIGSEGV. And if the signal that caused
 * the problem was already a SIGSEGV, we'll want to
 * make sure we don't even try to deliver the signal..
 */
int
force_sigsegv(int sig, struct task_struct *p)
{
	if (sig == SIGSEGV) {
		unsigned long flags;
		spin_lock_irqsave(&p->sighand->siglock, flags);
		p->sighand->action[sig - 1].sa.sa_handler = SIG_DFL;
		spin_unlock_irqrestore(&p->sighand->siglock, flags);
	}
	force_sig(SIGSEGV, p);
	return 0;
}

1471 1472
int kill_pgrp(struct pid *pid, int sig, int priv)
{
1473 1474 1475 1476 1477 1478 1479
	int ret;

	read_lock(&tasklist_lock);
	ret = __kill_pgrp_info(sig, __si_special(priv), pid);
	read_unlock(&tasklist_lock);

	return ret;
1480 1481 1482 1483 1484 1485 1486 1487 1488
}
EXPORT_SYMBOL(kill_pgrp);

int kill_pid(struct pid *pid, int sig, int priv)
{
	return kill_pid_info(sig, __si_special(priv), pid);
}
EXPORT_SYMBOL(kill_pid);

L
Linus Torvalds 已提交
1489 1490 1491 1492
/*
 * These functions support sending signals using preallocated sigqueue
 * structures.  This is needed "because realtime applications cannot
 * afford to lose notifications of asynchronous events, like timer
1493
 * expirations or I/O completions".  In the case of POSIX Timers
L
Linus Torvalds 已提交
1494 1495 1496 1497 1498 1499
 * we allocate the sigqueue structure from the timer_create.  If this
 * allocation fails we are able to report the failure to the application
 * with an EAGAIN error.
 */
struct sigqueue *sigqueue_alloc(void)
{
1500
	struct sigqueue *q = __sigqueue_alloc(-1, current, GFP_KERNEL, 0);
L
Linus Torvalds 已提交
1501

1502
	if (q)
L
Linus Torvalds 已提交
1503
		q->flags |= SIGQUEUE_PREALLOC;
1504 1505

	return q;
L
Linus Torvalds 已提交
1506 1507 1508 1509 1510
}

void sigqueue_free(struct sigqueue *q)
{
	unsigned long flags;
1511 1512
	spinlock_t *lock = &current->sighand->siglock;

L
Linus Torvalds 已提交
1513 1514
	BUG_ON(!(q->flags & SIGQUEUE_PREALLOC));
	/*
1515 1516
	 * We must hold ->siglock while testing q->list
	 * to serialize with collect_signal() or with
1517
	 * __exit_signal()->flush_sigqueue().
L
Linus Torvalds 已提交
1518
	 */
1519
	spin_lock_irqsave(lock, flags);
1520 1521 1522 1523 1524
	q->flags &= ~SIGQUEUE_PREALLOC;
	/*
	 * If it is queued it will be freed when dequeued,
	 * like the "regular" sigqueue.
	 */
1525
	if (!list_empty(&q->list))
1526
		q = NULL;
1527 1528
	spin_unlock_irqrestore(lock, flags);

1529 1530
	if (q)
		__sigqueue_free(q);
L
Linus Torvalds 已提交
1531 1532
}

1533
int send_sigqueue(struct sigqueue *q, struct task_struct *t, int group)
1534
{
1535
	int sig = q->info.si_signo;
1536
	struct sigpending *pending;
1537
	unsigned long flags;
1538
	int ret, result;
1539

1540
	BUG_ON(!(q->flags & SIGQUEUE_PREALLOC));
1541 1542 1543 1544 1545

	ret = -1;
	if (!likely(lock_task_sighand(t, &flags)))
		goto ret;

1546
	ret = 1; /* the signal is ignored */
1547
	result = TRACE_SIGNAL_IGNORED;
1548
	if (!prepare_signal(sig, t, false))
1549 1550 1551
		goto out;

	ret = 0;
1552 1553 1554 1555 1556 1557 1558
	if (unlikely(!list_empty(&q->list))) {
		/*
		 * If an SI_TIMER entry is already queue just increment
		 * the overrun count.
		 */
		BUG_ON(q->info.si_code != SI_TIMER);
		q->info.si_overrun++;
1559
		result = TRACE_SIGNAL_ALREADY_PENDING;
1560
		goto out;
1561
	}
1562
	q->info.si_overrun = 0;
1563 1564

	signalfd_notify(t, sig);
1565
	pending = group ? &t->signal->shared_pending : &t->pending;
1566 1567
	list_add_tail(&q->list, &pending->list);
	sigaddset(&pending->signal, sig);
1568
	complete_signal(sig, t, group);
1569
	result = TRACE_SIGNAL_DELIVERED;
1570
out:
1571
	trace_signal_generate(sig, &q->info, t, group, result);
1572 1573 1574
	unlock_task_sighand(t, &flags);
ret:
	return ret;
1575 1576
}

L
Linus Torvalds 已提交
1577 1578 1579
/*
 * Let a parent know about the death of a child.
 * For a stopped/continued status change, use do_notify_parent_cldstop instead.
R
Roland McGrath 已提交
1580
 *
1581 1582
 * Returns true if our parent ignored us and so we've switched to
 * self-reaping.
L
Linus Torvalds 已提交
1583
 */
1584
bool do_notify_parent(struct task_struct *tsk, int sig)
L
Linus Torvalds 已提交
1585 1586 1587 1588
{
	struct siginfo info;
	unsigned long flags;
	struct sighand_struct *psig;
1589
	bool autoreap = false;
1590
	u64 utime, stime;
L
Linus Torvalds 已提交
1591 1592 1593 1594

	BUG_ON(sig == -1);

 	/* do_notify_parent_cldstop should have been called instead.  */
M
Matthew Wilcox 已提交
1595
 	BUG_ON(task_is_stopped_or_traced(tsk));
L
Linus Torvalds 已提交
1596

T
Tejun Heo 已提交
1597
	BUG_ON(!tsk->ptrace &&
L
Linus Torvalds 已提交
1598 1599
	       (tsk->group_leader != tsk || !thread_group_empty(tsk)));

1600 1601 1602 1603 1604 1605 1606 1607 1608
	if (sig != SIGCHLD) {
		/*
		 * This is only possible if parent == real_parent.
		 * Check if it has changed security domain.
		 */
		if (tsk->parent_exec_id != tsk->parent->self_exec_id)
			sig = SIGCHLD;
	}

L
Linus Torvalds 已提交
1609 1610
	info.si_signo = sig;
	info.si_errno = 0;
1611
	/*
1612 1613
	 * We are under tasklist_lock here so our parent is tied to
	 * us and cannot change.
1614
	 *
1615 1616
	 * task_active_pid_ns will always return the same pid namespace
	 * until a task passes through release_task.
1617 1618 1619 1620 1621 1622
	 *
	 * write_lock() currently calls preempt_disable() which is the
	 * same as rcu_read_lock(), but according to Oleg, this is not
	 * correct to rely on this
	 */
	rcu_read_lock();
1623
	info.si_pid = task_pid_nr_ns(tsk, task_active_pid_ns(tsk->parent));
1624 1625
	info.si_uid = from_kuid_munged(task_cred_xxx(tsk->parent, user_ns),
				       task_uid(tsk));
1626 1627
	rcu_read_unlock();

1628 1629 1630
	task_cputime(tsk, &utime, &stime);
	info.si_utime = nsec_to_clock_t(utime + tsk->signal->utime);
	info.si_stime = nsec_to_clock_t(stime + tsk->signal->stime);
L
Linus Torvalds 已提交
1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643

	info.si_status = tsk->exit_code & 0x7f;
	if (tsk->exit_code & 0x80)
		info.si_code = CLD_DUMPED;
	else if (tsk->exit_code & 0x7f)
		info.si_code = CLD_KILLED;
	else {
		info.si_code = CLD_EXITED;
		info.si_status = tsk->exit_code >> 8;
	}

	psig = tsk->parent->sighand;
	spin_lock_irqsave(&psig->siglock, flags);
T
Tejun Heo 已提交
1644
	if (!tsk->ptrace && sig == SIGCHLD &&
L
Linus Torvalds 已提交
1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661
	    (psig->action[SIGCHLD-1].sa.sa_handler == SIG_IGN ||
	     (psig->action[SIGCHLD-1].sa.sa_flags & SA_NOCLDWAIT))) {
		/*
		 * We are exiting and our parent doesn't care.  POSIX.1
		 * defines special semantics for setting SIGCHLD to SIG_IGN
		 * or setting the SA_NOCLDWAIT flag: we should be reaped
		 * automatically and not left for our parent's wait4 call.
		 * Rather than having the parent do it as a magic kind of
		 * signal handler, we just set this to tell do_exit that we
		 * can be cleaned up without becoming a zombie.  Note that
		 * we still call __wake_up_parent in this case, because a
		 * blocked sys_wait4 might now return -ECHILD.
		 *
		 * Whether we send SIGCHLD or not for SA_NOCLDWAIT
		 * is implementation-defined: we do (if you don't want
		 * it, just use SIG_IGN instead).
		 */
1662
		autoreap = true;
L
Linus Torvalds 已提交
1663
		if (psig->action[SIGCHLD-1].sa.sa_handler == SIG_IGN)
1664
			sig = 0;
L
Linus Torvalds 已提交
1665
	}
1666
	if (valid_signal(sig) && sig)
L
Linus Torvalds 已提交
1667 1668 1669
		__group_send_sig_info(sig, &info, tsk->parent);
	__wake_up_parent(tsk, tsk->parent);
	spin_unlock_irqrestore(&psig->siglock, flags);
R
Roland McGrath 已提交
1670

1671
	return autoreap;
L
Linus Torvalds 已提交
1672 1673
}

1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688
/**
 * do_notify_parent_cldstop - notify parent of stopped/continued state change
 * @tsk: task reporting the state change
 * @for_ptracer: the notification is for ptracer
 * @why: CLD_{CONTINUED|STOPPED|TRAPPED} to report
 *
 * Notify @tsk's parent that the stopped/continued state has changed.  If
 * @for_ptracer is %false, @tsk's group leader notifies to its real parent.
 * If %true, @tsk reports to @tsk->parent which should be the ptracer.
 *
 * CONTEXT:
 * Must be called with tasklist_lock at least read locked.
 */
static void do_notify_parent_cldstop(struct task_struct *tsk,
				     bool for_ptracer, int why)
L
Linus Torvalds 已提交
1689 1690 1691
{
	struct siginfo info;
	unsigned long flags;
1692
	struct task_struct *parent;
L
Linus Torvalds 已提交
1693
	struct sighand_struct *sighand;
1694
	u64 utime, stime;
L
Linus Torvalds 已提交
1695

1696
	if (for_ptracer) {
1697
		parent = tsk->parent;
1698
	} else {
1699 1700 1701 1702
		tsk = tsk->group_leader;
		parent = tsk->real_parent;
	}

L
Linus Torvalds 已提交
1703 1704
	info.si_signo = SIGCHLD;
	info.si_errno = 0;
1705
	/*
1706
	 * see comment in do_notify_parent() about the following 4 lines
1707 1708
	 */
	rcu_read_lock();
1709
	info.si_pid = task_pid_nr_ns(tsk, task_active_pid_ns(parent));
1710
	info.si_uid = from_kuid_munged(task_cred_xxx(parent, user_ns), task_uid(tsk));
1711 1712
	rcu_read_unlock();

1713 1714 1715
	task_cputime(tsk, &utime, &stime);
	info.si_utime = nsec_to_clock_t(utime);
	info.si_stime = nsec_to_clock_t(stime);
L
Linus Torvalds 已提交
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

 	info.si_code = why;
 	switch (why) {
 	case CLD_CONTINUED:
 		info.si_status = SIGCONT;
 		break;
 	case CLD_STOPPED:
 		info.si_status = tsk->signal->group_exit_code & 0x7f;
 		break;
 	case CLD_TRAPPED:
 		info.si_status = tsk->exit_code & 0x7f;
 		break;
 	default:
 		BUG();
 	}

	sighand = parent->sighand;
	spin_lock_irqsave(&sighand->siglock, flags);
	if (sighand->action[SIGCHLD-1].sa.sa_handler != SIG_IGN &&
	    !(sighand->action[SIGCHLD-1].sa.sa_flags & SA_NOCLDSTOP))
		__group_send_sig_info(SIGCHLD, &info, parent);
	/*
	 * Even if SIGCHLD is not generated, we must wake up wait4 calls.
	 */
	__wake_up_parent(tsk, parent);
	spin_unlock_irqrestore(&sighand->siglock, flags);
}

1744 1745
static inline int may_ptrace_stop(void)
{
T
Tejun Heo 已提交
1746
	if (!likely(current->ptrace))
1747 1748 1749 1750 1751 1752 1753
		return 0;
	/*
	 * Are we in the middle of do_coredump?
	 * If so and our tracer is also part of the coredump stopping
	 * is a deadlock situation, and pointless because our tracer
	 * is dead so don't allow us to stop.
	 * If SIGKILL was already sent before the caller unlocked
1754
	 * ->siglock we must see ->core_state != NULL. Otherwise it
1755
	 * is safe to enter schedule().
1756 1757 1758 1759
	 *
	 * This is almost outdated, a task with the pending SIGKILL can't
	 * block in TASK_TRACED. But PTRACE_EVENT_EXIT can be reported
	 * after SIGKILL was already dequeued.
1760
	 */
1761
	if (unlikely(current->mm->core_state) &&
1762 1763 1764 1765 1766 1767
	    unlikely(current->mm == current->parent->mm))
		return 0;

	return 1;
}

R
Roland McGrath 已提交
1768
/*
1769
 * Return non-zero if there is a SIGKILL that should be waking us up.
R
Roland McGrath 已提交
1770 1771 1772 1773
 * Called with the siglock held.
 */
static int sigkill_pending(struct task_struct *tsk)
{
1774 1775
	return	sigismember(&tsk->pending.signal, SIGKILL) ||
		sigismember(&tsk->signal->shared_pending.signal, SIGKILL);
R
Roland McGrath 已提交
1776 1777
}

L
Linus Torvalds 已提交
1778 1779 1780 1781 1782 1783 1784 1785
/*
 * This must be called with current->sighand->siglock held.
 *
 * This should be the path for all ptrace stops.
 * We always set current->last_siginfo while stopped here.
 * That makes it a way to test a stopped process for
 * being ptrace-stopped vs being job-control-stopped.
 *
1786 1787
 * If we actually decide not to stop at all because the tracer
 * is gone, we keep current->exit_code unless clear_code.
L
Linus Torvalds 已提交
1788
 */
T
Tejun Heo 已提交
1789
static void ptrace_stop(int exit_code, int why, int clear_code, siginfo_t *info)
1790 1791
	__releases(&current->sighand->siglock)
	__acquires(&current->sighand->siglock)
L
Linus Torvalds 已提交
1792
{
1793 1794
	bool gstop_done = false;

R
Roland McGrath 已提交
1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809
	if (arch_ptrace_stop_needed(exit_code, info)) {
		/*
		 * The arch code has something special to do before a
		 * ptrace stop.  This is allowed to block, e.g. for faults
		 * on user stack pages.  We can't keep the siglock while
		 * calling arch_ptrace_stop, so we must release it now.
		 * To preserve proper semantics, we must do this before
		 * any signal bookkeeping like checking group_stop_count.
		 * Meanwhile, a SIGKILL could come in before we retake the
		 * siglock.  That must prevent us from sleeping in TASK_TRACED.
		 * So after regaining the lock, we must check for SIGKILL.
		 */
		spin_unlock_irq(&current->sighand->siglock);
		arch_ptrace_stop(exit_code, info);
		spin_lock_irq(&current->sighand->siglock);
1810 1811
		if (sigkill_pending(current))
			return;
R
Roland McGrath 已提交
1812 1813
	}

L
Linus Torvalds 已提交
1814
	/*
1815 1816 1817 1818 1819
	 * We're committing to trapping.  TRACED should be visible before
	 * TRAPPING is cleared; otherwise, the tracer might fail do_wait().
	 * Also, transition to TRACED and updates to ->jobctl should be
	 * atomic with respect to siglock and should be done after the arch
	 * hook as siglock is released and regrabbed across it.
L
Linus Torvalds 已提交
1820
	 */
1821
	set_current_state(TASK_TRACED);
L
Linus Torvalds 已提交
1822 1823 1824 1825

	current->last_siginfo = info;
	current->exit_code = exit_code;

1826
	/*
1827 1828
	 * If @why is CLD_STOPPED, we're trapping to participate in a group
	 * stop.  Do the bookkeeping.  Note that if SIGCONT was delievered
1829 1830 1831
	 * across siglock relocks since INTERRUPT was scheduled, PENDING
	 * could be clear now.  We act as if SIGCONT is received after
	 * TASK_TRACED is entered - ignore it.
1832
	 */
1833
	if (why == CLD_STOPPED && (current->jobctl & JOBCTL_STOP_PENDING))
1834
		gstop_done = task_participate_group_stop(current);
1835

1836
	/* any trap clears pending STOP trap, STOP trap clears NOTIFY */
1837
	task_clear_jobctl_pending(current, JOBCTL_TRAP_STOP);
1838 1839
	if (info && info->si_code >> 8 == PTRACE_EVENT_STOP)
		task_clear_jobctl_pending(current, JOBCTL_TRAP_NOTIFY);
1840

1841
	/* entering a trap, clear TRAPPING */
1842
	task_clear_jobctl_trapping(current);
1843

L
Linus Torvalds 已提交
1844 1845
	spin_unlock_irq(&current->sighand->siglock);
	read_lock(&tasklist_lock);
1846
	if (may_ptrace_stop()) {
1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857
		/*
		 * Notify parents of the stop.
		 *
		 * While ptraced, there are two parents - the ptracer and
		 * the real_parent of the group_leader.  The ptracer should
		 * know about every stop while the real parent is only
		 * interested in the completion of group stop.  The states
		 * for the two don't interact with each other.  Notify
		 * separately unless they're gonna be duplicates.
		 */
		do_notify_parent_cldstop(current, true, why);
1858
		if (gstop_done && ptrace_reparented(current))
1859 1860
			do_notify_parent_cldstop(current, false, why);

M
Miklos Szeredi 已提交
1861 1862 1863 1864 1865 1866 1867
		/*
		 * Don't want to allow preemption here, because
		 * sys_ptrace() needs this task to be inactive.
		 *
		 * XXX: implement read_unlock_no_resched().
		 */
		preempt_disable();
L
Linus Torvalds 已提交
1868
		read_unlock(&tasklist_lock);
M
Miklos Szeredi 已提交
1869
		preempt_enable_no_resched();
1870
		freezable_schedule();
L
Linus Torvalds 已提交
1871 1872 1873
	} else {
		/*
		 * By the time we got the lock, our tracer went away.
1874
		 * Don't drop the lock yet, another tracer may come.
1875 1876 1877
		 *
		 * If @gstop_done, the ptracer went away between group stop
		 * completion and here.  During detach, it would have set
1878 1879 1880
		 * JOBCTL_STOP_PENDING on us and we'll re-enter
		 * TASK_STOPPED in do_signal_stop() on return, so notifying
		 * the real parent of the group stop completion is enough.
L
Linus Torvalds 已提交
1881
		 */
1882 1883 1884
		if (gstop_done)
			do_notify_parent_cldstop(current, false, why);

1885
		/* tasklist protects us from ptrace_freeze_traced() */
1886
		__set_current_state(TASK_RUNNING);
1887 1888
		if (clear_code)
			current->exit_code = 0;
1889
		read_unlock(&tasklist_lock);
L
Linus Torvalds 已提交
1890 1891 1892 1893 1894 1895 1896 1897 1898 1899
	}

	/*
	 * We are back.  Now reacquire the siglock before touching
	 * last_siginfo, so that we are sure to have synchronized with
	 * any signal-sending on another CPU that wants to examine it.
	 */
	spin_lock_irq(&current->sighand->siglock);
	current->last_siginfo = NULL;

T
Tejun Heo 已提交
1900 1901 1902
	/* LISTENING can be set only during STOP traps, clear it */
	current->jobctl &= ~JOBCTL_LISTENING;

L
Linus Torvalds 已提交
1903 1904 1905
	/*
	 * Queued signals ignored us while we were stopped for tracing.
	 * So check for any that we should take before resuming user mode.
1906
	 * This sets TIF_SIGPENDING, but never clears it.
L
Linus Torvalds 已提交
1907
	 */
1908
	recalc_sigpending_tsk(current);
L
Linus Torvalds 已提交
1909 1910
}

T
Tejun Heo 已提交
1911
static void ptrace_do_notify(int signr, int exit_code, int why)
L
Linus Torvalds 已提交
1912 1913 1914 1915
{
	siginfo_t info;

	memset(&info, 0, sizeof info);
T
Tejun Heo 已提交
1916
	info.si_signo = signr;
L
Linus Torvalds 已提交
1917
	info.si_code = exit_code;
1918
	info.si_pid = task_pid_vnr(current);
1919
	info.si_uid = from_kuid_munged(current_user_ns(), current_uid());
L
Linus Torvalds 已提交
1920 1921

	/* Let the debugger run.  */
T
Tejun Heo 已提交
1922 1923 1924 1925 1926 1927
	ptrace_stop(exit_code, why, 1, &info);
}

void ptrace_notify(int exit_code)
{
	BUG_ON((exit_code & (0x7f | ~0xffff)) != SIGTRAP);
1928 1929
	if (unlikely(current->task_works))
		task_work_run();
T
Tejun Heo 已提交
1930

L
Linus Torvalds 已提交
1931
	spin_lock_irq(&current->sighand->siglock);
T
Tejun Heo 已提交
1932
	ptrace_do_notify(SIGTRAP, exit_code, CLD_TRAPPED);
L
Linus Torvalds 已提交
1933 1934 1935
	spin_unlock_irq(&current->sighand->siglock);
}

1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956
/**
 * do_signal_stop - handle group stop for SIGSTOP and other stop signals
 * @signr: signr causing group stop if initiating
 *
 * If %JOBCTL_STOP_PENDING is not set yet, initiate group stop with @signr
 * and participate in it.  If already set, participate in the existing
 * group stop.  If participated in a group stop (and thus slept), %true is
 * returned with siglock released.
 *
 * If ptraced, this function doesn't handle stop itself.  Instead,
 * %JOBCTL_TRAP_STOP is scheduled and %false is returned with siglock
 * untouched.  The caller must ensure that INTERRUPT trap handling takes
 * places afterwards.
 *
 * CONTEXT:
 * Must be called with @current->sighand->siglock held, which is released
 * on %true return.
 *
 * RETURNS:
 * %false if group stop is already cancelled or ptrace trap is scheduled.
 * %true if participated in group stop.
L
Linus Torvalds 已提交
1957
 */
1958 1959
static bool do_signal_stop(int signr)
	__releases(&current->sighand->siglock)
L
Linus Torvalds 已提交
1960 1961 1962
{
	struct signal_struct *sig = current->signal;

1963
	if (!(current->jobctl & JOBCTL_STOP_PENDING)) {
1964
		unsigned long gstop = JOBCTL_STOP_PENDING | JOBCTL_STOP_CONSUME;
1965 1966
		struct task_struct *t;

1967 1968
		/* signr will be recorded in task->jobctl for retries */
		WARN_ON_ONCE(signr & ~JOBCTL_STOP_SIGMASK);
1969

1970
		if (!likely(current->jobctl & JOBCTL_STOP_DEQUEUED) ||
1971
		    unlikely(signal_group_exit(sig)))
1972
			return false;
L
Linus Torvalds 已提交
1973
		/*
1974 1975 1976 1977 1978 1979 1980
		 * There is no group stop already in progress.  We must
		 * initiate one now.
		 *
		 * While ptraced, a task may be resumed while group stop is
		 * still in effect and then receive a stop signal and
		 * initiate another group stop.  This deviates from the
		 * usual behavior as two consecutive stop signals can't
1981 1982
		 * cause two group stops when !ptraced.  That is why we
		 * also check !task_is_stopped(t) below.
1983 1984 1985 1986 1987 1988 1989 1990
		 *
		 * The condition can be distinguished by testing whether
		 * SIGNAL_STOP_STOPPED is already set.  Don't generate
		 * group_exit_code in such case.
		 *
		 * This is not necessary for SIGNAL_STOP_CONTINUED because
		 * an intervening stop signal is required to cause two
		 * continued events regardless of ptrace.
L
Linus Torvalds 已提交
1991
		 */
1992 1993
		if (!(sig->flags & SIGNAL_STOP_STOPPED))
			sig->group_exit_code = signr;
L
Linus Torvalds 已提交
1994

1995 1996 1997 1998
		sig->group_stop_count = 0;

		if (task_set_jobctl_pending(current, signr | gstop))
			sig->group_stop_count++;
L
Linus Torvalds 已提交
1999

2000 2001
		t = current;
		while_each_thread(current, t) {
L
Linus Torvalds 已提交
2002
			/*
2003 2004 2005
			 * Setting state to TASK_STOPPED for a group
			 * stop is always done with the siglock held,
			 * so this check has no races.
L
Linus Torvalds 已提交
2006
			 */
2007 2008
			if (!task_is_stopped(t) &&
			    task_set_jobctl_pending(t, signr | gstop)) {
2009
				sig->group_stop_count++;
2010 2011 2012 2013
				if (likely(!(t->ptrace & PT_SEIZED)))
					signal_wake_up(t, 0);
				else
					ptrace_trap_notify(t);
2014
			}
2015
		}
L
Linus Torvalds 已提交
2016
	}
2017

T
Tejun Heo 已提交
2018
	if (likely(!current->ptrace)) {
2019
		int notify = 0;
L
Linus Torvalds 已提交
2020

2021 2022 2023 2024 2025 2026 2027 2028
		/*
		 * If there are no other threads in the group, or if there
		 * is a group stop in progress and we are the last to stop,
		 * report to the parent.
		 */
		if (task_participate_group_stop(current))
			notify = CLD_STOPPED;

2029
		__set_current_state(TASK_STOPPED);
2030 2031
		spin_unlock_irq(&current->sighand->siglock);

2032 2033 2034 2035 2036 2037 2038 2039 2040
		/*
		 * Notify the parent of the group stop completion.  Because
		 * we're not holding either the siglock or tasklist_lock
		 * here, ptracer may attach inbetween; however, this is for
		 * group stop and should always be delivered to the real
		 * parent of the group leader.  The new ptracer will get
		 * its notification when this task transitions into
		 * TASK_TRACED.
		 */
2041 2042
		if (notify) {
			read_lock(&tasklist_lock);
2043
			do_notify_parent_cldstop(current, false, notify);
2044 2045 2046 2047
			read_unlock(&tasklist_lock);
		}

		/* Now we don't run again until woken by SIGCONT or SIGKILL */
2048
		freezable_schedule();
2049
		return true;
2050
	} else {
2051 2052 2053 2054 2055 2056
		/*
		 * While ptraced, group stop is handled by STOP trap.
		 * Schedule it and let the caller deal with it.
		 */
		task_set_jobctl_pending(current, JOBCTL_TRAP_STOP);
		return false;
2057
	}
2058
}
L
Linus Torvalds 已提交
2059

2060 2061 2062
/**
 * do_jobctl_trap - take care of ptrace jobctl traps
 *
T
Tejun Heo 已提交
2063 2064 2065 2066 2067 2068 2069
 * When PT_SEIZED, it's used for both group stop and explicit
 * SEIZE/INTERRUPT traps.  Both generate PTRACE_EVENT_STOP trap with
 * accompanying siginfo.  If stopped, lower eight bits of exit_code contain
 * the stop signal; otherwise, %SIGTRAP.
 *
 * When !PT_SEIZED, it's used only for group stop trap with stop signal
 * number as exit_code and no siginfo.
2070 2071 2072 2073 2074 2075 2076
 *
 * CONTEXT:
 * Must be called with @current->sighand->siglock held, which may be
 * released and re-acquired before returning with intervening sleep.
 */
static void do_jobctl_trap(void)
{
T
Tejun Heo 已提交
2077
	struct signal_struct *signal = current->signal;
2078
	int signr = current->jobctl & JOBCTL_STOP_SIGMASK;
2079

T
Tejun Heo 已提交
2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090
	if (current->ptrace & PT_SEIZED) {
		if (!signal->group_stop_count &&
		    !(signal->flags & SIGNAL_STOP_STOPPED))
			signr = SIGTRAP;
		WARN_ON_ONCE(!signr);
		ptrace_do_notify(signr, signr | (PTRACE_EVENT_STOP << 8),
				 CLD_STOPPED);
	} else {
		WARN_ON_ONCE(!signr);
		ptrace_stop(signr, CLD_STOPPED, 0, NULL);
		current->exit_code = 0;
2091
	}
L
Linus Torvalds 已提交
2092 2093
}

2094
static int ptrace_signal(int signr, siginfo_t *info)
R
Roland McGrath 已提交
2095
{
2096
	ptrace_signal_deliver();
2097 2098 2099 2100 2101 2102 2103 2104 2105 2106
	/*
	 * We do not check sig_kernel_stop(signr) but set this marker
	 * unconditionally because we do not know whether debugger will
	 * change signr. This flag has no meaning unless we are going
	 * to stop after return from ptrace_stop(). In this case it will
	 * be checked in do_signal_stop(), we should only stop if it was
	 * not cleared by SIGCONT while we were sleeping. See also the
	 * comment in dequeue_signal().
	 */
	current->jobctl |= JOBCTL_STOP_DEQUEUED;
T
Tejun Heo 已提交
2107
	ptrace_stop(signr, CLD_TRAPPED, 0, info);
R
Roland McGrath 已提交
2108 2109 2110 2111 2112 2113 2114 2115

	/* We're back.  Did the debugger cancel the sig?  */
	signr = current->exit_code;
	if (signr == 0)
		return signr;

	current->exit_code = 0;

2116 2117 2118 2119 2120 2121
	/*
	 * Update the siginfo structure if the signal has
	 * changed.  If the debugger wanted something
	 * specific in the siginfo structure then it should
	 * have updated *info via PTRACE_SETSIGINFO.
	 */
R
Roland McGrath 已提交
2122 2123 2124 2125
	if (signr != info->si_signo) {
		info->si_signo = signr;
		info->si_errno = 0;
		info->si_code = SI_USER;
2126
		rcu_read_lock();
R
Roland McGrath 已提交
2127
		info->si_pid = task_pid_vnr(current->parent);
2128 2129
		info->si_uid = from_kuid_munged(current_user_ns(),
						task_uid(current->parent));
2130
		rcu_read_unlock();
R
Roland McGrath 已提交
2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141
	}

	/* If the (new) signal is now blocked, requeue it.  */
	if (sigismember(&current->blocked, signr)) {
		specific_send_sig_info(signr, info, current);
		signr = 0;
	}

	return signr;
}

2142
int get_signal(struct ksignal *ksig)
L
Linus Torvalds 已提交
2143
{
2144 2145 2146
	struct sighand_struct *sighand = current->sighand;
	struct signal_struct *signal = current->signal;
	int signr;
L
Linus Torvalds 已提交
2147

2148 2149
	if (unlikely(current->task_works))
		task_work_run();
2150

2151 2152 2153
	if (unlikely(uprobe_deny_signal()))
		return 0;

R
Roland McGrath 已提交
2154
	/*
2155 2156 2157
	 * Do this once, we can't return to user-mode if freezing() == T.
	 * do_signal_stop() and ptrace_stop() do freezable_schedule() and
	 * thus do not need another check after return.
R
Roland McGrath 已提交
2158
	 */
2159 2160
	try_to_freeze();

2161
relock:
2162
	spin_lock_irq(&sighand->siglock);
2163 2164 2165 2166 2167
	/*
	 * Every stopped thread goes here after wakeup. Check to see if
	 * we should notify the parent, prepare_signal(SIGCONT) encodes
	 * the CLD_ si_code into SIGNAL_CLD_MASK bits.
	 */
2168
	if (unlikely(signal->flags & SIGNAL_CLD_MASK)) {
2169 2170 2171 2172 2173 2174 2175
		int why;

		if (signal->flags & SIGNAL_CLD_CONTINUED)
			why = CLD_CONTINUED;
		else
			why = CLD_STOPPED;

2176
		signal->flags &= ~SIGNAL_CLD_MASK;
2177

2178
		spin_unlock_irq(&sighand->siglock);
R
Roland McGrath 已提交
2179

2180 2181 2182 2183 2184 2185 2186 2187
		/*
		 * Notify the parent that we're continuing.  This event is
		 * always per-process and doesn't make whole lot of sense
		 * for ptracers, who shouldn't consume the state via
		 * wait(2) either, but, for backward compatibility, notify
		 * the ptracer of the group leader too unless it's gonna be
		 * a duplicate.
		 */
T
Tejun Heo 已提交
2188
		read_lock(&tasklist_lock);
2189 2190
		do_notify_parent_cldstop(current, false, why);

2191 2192 2193
		if (ptrace_reparented(current->group_leader))
			do_notify_parent_cldstop(current->group_leader,
						true, why);
T
Tejun Heo 已提交
2194
		read_unlock(&tasklist_lock);
2195

2196 2197 2198
		goto relock;
	}

L
Linus Torvalds 已提交
2199 2200
	for (;;) {
		struct k_sigaction *ka;
2201

T
Tejun Heo 已提交
2202 2203
		if (unlikely(current->jobctl & JOBCTL_STOP_PENDING) &&
		    do_signal_stop(0))
2204
			goto relock;
2205

2206 2207 2208 2209 2210
		if (unlikely(current->jobctl & JOBCTL_TRAP_MASK)) {
			do_jobctl_trap();
			spin_unlock_irq(&sighand->siglock);
			goto relock;
		}
L
Linus Torvalds 已提交
2211

2212
		signr = dequeue_signal(current, &current->blocked, &ksig->info);
2213

T
Tejun Heo 已提交
2214 2215
		if (!signr)
			break; /* will return 0 */
2216

2217
		if (unlikely(current->ptrace) && signr != SIGKILL) {
2218
			signr = ptrace_signal(signr, &ksig->info);
T
Tejun Heo 已提交
2219 2220
			if (!signr)
				continue;
L
Linus Torvalds 已提交
2221 2222
		}

T
Tejun Heo 已提交
2223 2224
		ka = &sighand->action[signr-1];

2225
		/* Trace actually delivered signals. */
2226
		trace_signal_deliver(signr, &ksig->info, ka);
2227

L
Linus Torvalds 已提交
2228 2229 2230 2231
		if (ka->sa.sa_handler == SIG_IGN) /* Do nothing.  */
			continue;
		if (ka->sa.sa_handler != SIG_DFL) {
			/* Run the handler.  */
2232
			ksig->ka = *ka;
L
Linus Torvalds 已提交
2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245

			if (ka->sa.sa_flags & SA_ONESHOT)
				ka->sa.sa_handler = SIG_DFL;

			break; /* will return non-zero "signr" value */
		}

		/*
		 * Now we are doing the default action for this signal.
		 */
		if (sig_kernel_ignore(signr)) /* Default is nothing. */
			continue;

2246
		/*
2247
		 * Global init gets no signals it doesn't want.
2248 2249 2250 2251 2252 2253 2254
		 * Container-init gets no signals it doesn't want from same
		 * container.
		 *
		 * Note that if global/container-init sees a sig_kernel_only()
		 * signal here, the signal must have been generated internally
		 * or must have come from an ancestor namespace. In either
		 * case, the signal cannot be dropped.
2255
		 */
2256
		if (unlikely(signal->flags & SIGNAL_UNKILLABLE) &&
2257
				!sig_kernel_only(signr))
L
Linus Torvalds 已提交
2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271
			continue;

		if (sig_kernel_stop(signr)) {
			/*
			 * The default action is to stop all threads in
			 * the thread group.  The job control signals
			 * do nothing in an orphaned pgrp, but SIGSTOP
			 * always works.  Note that siglock needs to be
			 * dropped during the call to is_orphaned_pgrp()
			 * because of lock ordering with tasklist_lock.
			 * This allows an intervening SIGCONT to be posted.
			 * We need to check for that and bail out if necessary.
			 */
			if (signr != SIGSTOP) {
2272
				spin_unlock_irq(&sighand->siglock);
L
Linus Torvalds 已提交
2273 2274 2275

				/* signals can be posted during this window */

2276
				if (is_current_pgrp_orphaned())
L
Linus Torvalds 已提交
2277 2278
					goto relock;

2279
				spin_lock_irq(&sighand->siglock);
L
Linus Torvalds 已提交
2280 2281
			}

2282
			if (likely(do_signal_stop(ksig->info.si_signo))) {
L
Linus Torvalds 已提交
2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293
				/* It released the siglock.  */
				goto relock;
			}

			/*
			 * We didn't actually stop, due to a race
			 * with SIGCONT or something like that.
			 */
			continue;
		}

2294
		spin_unlock_irq(&sighand->siglock);
L
Linus Torvalds 已提交
2295 2296 2297 2298 2299

		/*
		 * Anything else is fatal, maybe with a core dump.
		 */
		current->flags |= PF_SIGNALED;
2300

L
Linus Torvalds 已提交
2301
		if (sig_kernel_coredump(signr)) {
2302
			if (print_fatal_signals)
2303
				print_fatal_signal(ksig->info.si_signo);
2304
			proc_coredump_connector(current);
L
Linus Torvalds 已提交
2305 2306 2307 2308 2309 2310 2311 2312
			/*
			 * If it was able to dump core, this kills all
			 * other threads in the group and synchronizes with
			 * their demise.  If we lost the race with another
			 * thread getting here, it set group_exit_code
			 * first and our do_group_exit call below will use
			 * that value and ignore the one we pass it.
			 */
2313
			do_coredump(&ksig->info);
L
Linus Torvalds 已提交
2314 2315 2316 2317 2318
		}

		/*
		 * Death signals, no core dump.
		 */
2319
		do_group_exit(ksig->info.si_signo);
L
Linus Torvalds 已提交
2320 2321
		/* NOTREACHED */
	}
2322
	spin_unlock_irq(&sighand->siglock);
2323 2324 2325

	ksig->sig = signr;
	return ksig->sig > 0;
L
Linus Torvalds 已提交
2326 2327
}

2328
/**
A
Al Viro 已提交
2329
 * signal_delivered - 
R
Richard Weinberger 已提交
2330
 * @ksig:		kernel signal struct
A
Al Viro 已提交
2331
 * @stepping:		nonzero if debugger single-step or block-step in use
2332
 *
2333
 * This function should be called when a signal has successfully been
R
Richard Weinberger 已提交
2334
 * delivered. It updates the blocked signals accordingly (@ksig->ka.sa.sa_mask
A
Al Viro 已提交
2335
 * is always blocked, and the signal itself is blocked unless %SA_NODEFER
R
Richard Weinberger 已提交
2336
 * is set in @ksig->ka.sa.sa_flags.  Tracing is notified.
2337
 */
R
Richard Weinberger 已提交
2338
static void signal_delivered(struct ksignal *ksig, int stepping)
2339 2340 2341
{
	sigset_t blocked;

2342 2343 2344 2345 2346 2347
	/* A signal was successfully delivered, and the
	   saved sigmask was stored on the signal frame,
	   and will be restored by sigreturn.  So we can
	   simply clear the restore sigmask flag.  */
	clear_restore_sigmask();

R
Richard Weinberger 已提交
2348 2349 2350
	sigorsets(&blocked, &current->blocked, &ksig->ka.sa.sa_mask);
	if (!(ksig->ka.sa.sa_flags & SA_NODEFER))
		sigaddset(&blocked, ksig->sig);
2351
	set_current_blocked(&blocked);
2352
	tracehook_signal_handler(stepping);
2353 2354
}

A
Al Viro 已提交
2355 2356 2357 2358 2359
void signal_setup_done(int failed, struct ksignal *ksig, int stepping)
{
	if (failed)
		force_sigsegv(ksig->sig, current);
	else
R
Richard Weinberger 已提交
2360
		signal_delivered(ksig, stepping);
A
Al Viro 已提交
2361 2362
}

2363 2364
/*
 * It could be that complete_signal() picked us to notify about the
2365 2366
 * group-wide signal. Other threads should be notified now to take
 * the shared signals in @which since we will not.
2367
 */
2368
static void retarget_shared_pending(struct task_struct *tsk, sigset_t *which)
2369
{
2370
	sigset_t retarget;
2371 2372
	struct task_struct *t;

2373 2374 2375 2376
	sigandsets(&retarget, &tsk->signal->shared_pending.signal, which);
	if (sigisemptyset(&retarget))
		return;

2377 2378
	t = tsk;
	while_each_thread(tsk, t) {
2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391
		if (t->flags & PF_EXITING)
			continue;

		if (!has_pending_signals(&retarget, &t->blocked))
			continue;
		/* Remove the signals this thread can handle. */
		sigandsets(&retarget, &retarget, &t->blocked);

		if (!signal_pending(t))
			signal_wake_up(t, 0);

		if (sigisemptyset(&retarget))
			break;
2392 2393 2394
	}
}

O
Oleg Nesterov 已提交
2395 2396 2397
void exit_signals(struct task_struct *tsk)
{
	int group_stop = 0;
2398
	sigset_t unblocked;
O
Oleg Nesterov 已提交
2399

2400 2401 2402 2403
	/*
	 * @tsk is about to have PF_EXITING set - lock out users which
	 * expect stable threadgroup.
	 */
2404
	cgroup_threadgroup_change_begin(tsk);
2405

2406 2407
	if (thread_group_empty(tsk) || signal_group_exit(tsk->signal)) {
		tsk->flags |= PF_EXITING;
2408
		cgroup_threadgroup_change_end(tsk);
2409
		return;
O
Oleg Nesterov 已提交
2410 2411
	}

2412
	spin_lock_irq(&tsk->sighand->siglock);
O
Oleg Nesterov 已提交
2413 2414 2415 2416 2417
	/*
	 * From now this task is not visible for group-wide signals,
	 * see wants_signal(), do_signal_stop().
	 */
	tsk->flags |= PF_EXITING;
2418

2419
	cgroup_threadgroup_change_end(tsk);
2420

2421 2422 2423
	if (!signal_pending(tsk))
		goto out;

2424 2425 2426
	unblocked = tsk->blocked;
	signotset(&unblocked);
	retarget_shared_pending(tsk, &unblocked);
2427

2428
	if (unlikely(tsk->jobctl & JOBCTL_STOP_PENDING) &&
2429
	    task_participate_group_stop(tsk))
T
Tejun Heo 已提交
2430
		group_stop = CLD_STOPPED;
2431
out:
O
Oleg Nesterov 已提交
2432 2433
	spin_unlock_irq(&tsk->sighand->siglock);

2434 2435 2436 2437
	/*
	 * If group stop has completed, deliver the notification.  This
	 * should always go to the real parent of the group leader.
	 */
2438
	if (unlikely(group_stop)) {
O
Oleg Nesterov 已提交
2439
		read_lock(&tasklist_lock);
2440
		do_notify_parent_cldstop(tsk, false, group_stop);
O
Oleg Nesterov 已提交
2441 2442 2443 2444
		read_unlock(&tasklist_lock);
	}
}

L
Linus Torvalds 已提交
2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456
EXPORT_SYMBOL(recalc_sigpending);
EXPORT_SYMBOL_GPL(dequeue_signal);
EXPORT_SYMBOL(flush_signals);
EXPORT_SYMBOL(force_sig);
EXPORT_SYMBOL(send_sig);
EXPORT_SYMBOL(send_sig_info);
EXPORT_SYMBOL(sigprocmask);

/*
 * System call entry points.
 */

2457 2458 2459
/**
 *  sys_restart_syscall - restart a system call
 */
2460
SYSCALL_DEFINE0(restart_syscall)
L
Linus Torvalds 已提交
2461
{
2462
	struct restart_block *restart = &current->restart_block;
L
Linus Torvalds 已提交
2463 2464 2465 2466 2467 2468 2469 2470
	return restart->fn(restart);
}

long do_no_restart_syscall(struct restart_block *param)
{
	return -EINTR;
}

2471 2472 2473 2474 2475
static void __set_task_blocked(struct task_struct *tsk, const sigset_t *newset)
{
	if (signal_pending(tsk) && !thread_group_empty(tsk)) {
		sigset_t newblocked;
		/* A set of now blocked but previously unblocked signals. */
2476
		sigandnsets(&newblocked, newset, &current->blocked);
2477 2478 2479 2480 2481 2482
		retarget_shared_pending(tsk, &newblocked);
	}
	tsk->blocked = *newset;
	recalc_sigpending();
}

2483 2484 2485 2486 2487 2488
/**
 * set_current_blocked - change current->blocked mask
 * @newset: new mask
 *
 * It is wrong to change ->blocked directly, this helper should be used
 * to ensure the process can't miss a shared signal we are going to block.
L
Linus Torvalds 已提交
2489
 */
2490 2491 2492
void set_current_blocked(sigset_t *newset)
{
	sigdelsetmask(newset, sigmask(SIGKILL) | sigmask(SIGSTOP));
2493
	__set_current_blocked(newset);
2494 2495 2496
}

void __set_current_blocked(const sigset_t *newset)
2497 2498 2499
{
	struct task_struct *tsk = current;

2500 2501 2502 2503 2504 2505 2506
	/*
	 * In case the signal mask hasn't changed, there is nothing we need
	 * to do. The current->blocked shouldn't be modified by other task.
	 */
	if (sigequalsets(&tsk->blocked, newset))
		return;

2507
	spin_lock_irq(&tsk->sighand->siglock);
2508
	__set_task_blocked(tsk, newset);
2509 2510
	spin_unlock_irq(&tsk->sighand->siglock);
}
L
Linus Torvalds 已提交
2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521

/*
 * This is also useful for kernel threads that want to temporarily
 * (or permanently) block certain signals.
 *
 * NOTE! Unlike the user-mode sys_sigprocmask(), the kernel
 * interface happily blocks "unblockable" signals like SIGKILL
 * and friends.
 */
int sigprocmask(int how, sigset_t *set, sigset_t *oldset)
{
2522 2523
	struct task_struct *tsk = current;
	sigset_t newset;
L
Linus Torvalds 已提交
2524

2525
	/* Lockless, only current can change ->blocked, never from irq */
2526
	if (oldset)
2527
		*oldset = tsk->blocked;
2528

L
Linus Torvalds 已提交
2529 2530
	switch (how) {
	case SIG_BLOCK:
2531
		sigorsets(&newset, &tsk->blocked, set);
L
Linus Torvalds 已提交
2532 2533
		break;
	case SIG_UNBLOCK:
2534
		sigandnsets(&newset, &tsk->blocked, set);
L
Linus Torvalds 已提交
2535 2536
		break;
	case SIG_SETMASK:
2537
		newset = *set;
L
Linus Torvalds 已提交
2538 2539
		break;
	default:
2540
		return -EINVAL;
L
Linus Torvalds 已提交
2541
	}
2542

2543
	__set_current_blocked(&newset);
2544
	return 0;
L
Linus Torvalds 已提交
2545 2546
}

2547 2548 2549
/**
 *  sys_rt_sigprocmask - change the list of currently blocked signals
 *  @how: whether to add, remove, or set signals
R
Randy Dunlap 已提交
2550
 *  @nset: stores pending signals
2551 2552 2553
 *  @oset: previous value of signal mask if non-null
 *  @sigsetsize: size of sigset_t type
 */
2554
SYSCALL_DEFINE4(rt_sigprocmask, int, how, sigset_t __user *, nset,
2555
		sigset_t __user *, oset, size_t, sigsetsize)
L
Linus Torvalds 已提交
2556 2557
{
	sigset_t old_set, new_set;
2558
	int error;
L
Linus Torvalds 已提交
2559 2560 2561

	/* XXX: Don't preclude handling different sized sigset_t's.  */
	if (sigsetsize != sizeof(sigset_t))
2562
		return -EINVAL;
L
Linus Torvalds 已提交
2563

2564 2565 2566 2567 2568
	old_set = current->blocked;

	if (nset) {
		if (copy_from_user(&new_set, nset, sizeof(sigset_t)))
			return -EFAULT;
L
Linus Torvalds 已提交
2569 2570
		sigdelsetmask(&new_set, sigmask(SIGKILL)|sigmask(SIGSTOP));

2571
		error = sigprocmask(how, &new_set, NULL);
L
Linus Torvalds 已提交
2572
		if (error)
2573 2574
			return error;
	}
L
Linus Torvalds 已提交
2575

2576 2577 2578
	if (oset) {
		if (copy_to_user(oset, &old_set, sizeof(sigset_t)))
			return -EFAULT;
L
Linus Torvalds 已提交
2579
	}
2580 2581

	return 0;
L
Linus Torvalds 已提交
2582 2583
}

A
Al Viro 已提交
2584 2585 2586
#ifdef CONFIG_COMPAT
COMPAT_SYSCALL_DEFINE4(rt_sigprocmask, int, how, compat_sigset_t __user *, nset,
		compat_sigset_t __user *, oset, compat_size_t, sigsetsize)
L
Linus Torvalds 已提交
2587
{
A
Al Viro 已提交
2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611
#ifdef __BIG_ENDIAN
	sigset_t old_set = current->blocked;

	/* XXX: Don't preclude handling different sized sigset_t's.  */
	if (sigsetsize != sizeof(sigset_t))
		return -EINVAL;

	if (nset) {
		compat_sigset_t new32;
		sigset_t new_set;
		int error;
		if (copy_from_user(&new32, nset, sizeof(compat_sigset_t)))
			return -EFAULT;

		sigset_from_compat(&new_set, &new32);
		sigdelsetmask(&new_set, sigmask(SIGKILL)|sigmask(SIGSTOP));

		error = sigprocmask(how, &new_set, NULL);
		if (error)
			return error;
	}
	if (oset) {
		compat_sigset_t old32;
		sigset_to_compat(&old32, &old_set);
A
Al Viro 已提交
2612
		if (copy_to_user(oset, &old32, sizeof(compat_sigset_t)))
A
Al Viro 已提交
2613 2614 2615 2616 2617 2618 2619 2620 2621
			return -EFAULT;
	}
	return 0;
#else
	return sys_rt_sigprocmask(how, (sigset_t __user *)nset,
				  (sigset_t __user *)oset, sigsetsize);
#endif
}
#endif
L
Linus Torvalds 已提交
2622

A
Al Viro 已提交
2623
static int do_sigpending(void *set, unsigned long sigsetsize)
L
Linus Torvalds 已提交
2624 2625
{
	if (sigsetsize > sizeof(sigset_t))
A
Al Viro 已提交
2626
		return -EINVAL;
L
Linus Torvalds 已提交
2627 2628

	spin_lock_irq(&current->sighand->siglock);
A
Al Viro 已提交
2629
	sigorsets(set, &current->pending.signal,
L
Linus Torvalds 已提交
2630 2631 2632 2633
		  &current->signal->shared_pending.signal);
	spin_unlock_irq(&current->sighand->siglock);

	/* Outside the lock because only this thread touches it.  */
A
Al Viro 已提交
2634 2635
	sigandsets(set, &current->blocked, set);
	return 0;
2636
}
L
Linus Torvalds 已提交
2637

2638 2639 2640
/**
 *  sys_rt_sigpending - examine a pending signal that has been raised
 *			while blocked
2641
 *  @uset: stores pending signals
2642 2643
 *  @sigsetsize: size of sigset_t type or larger
 */
A
Al Viro 已提交
2644
SYSCALL_DEFINE2(rt_sigpending, sigset_t __user *, uset, size_t, sigsetsize)
L
Linus Torvalds 已提交
2645
{
A
Al Viro 已提交
2646 2647 2648 2649 2650 2651 2652 2653 2654 2655
	sigset_t set;
	int err = do_sigpending(&set, sigsetsize);
	if (!err && copy_to_user(uset, &set, sigsetsize))
		err = -EFAULT;
	return err;
}

#ifdef CONFIG_COMPAT
COMPAT_SYSCALL_DEFINE2(rt_sigpending, compat_sigset_t __user *, uset,
		compat_size_t, sigsetsize)
L
Linus Torvalds 已提交
2656
{
A
Al Viro 已提交
2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670
#ifdef __BIG_ENDIAN
	sigset_t set;
	int err = do_sigpending(&set, sigsetsize);
	if (!err) {
		compat_sigset_t set32;
		sigset_to_compat(&set32, &set);
		/* we can get here only if sigsetsize <= sizeof(set) */
		if (copy_to_user(uset, &set32, sigsetsize))
			err = -EFAULT;
	}
	return err;
#else
	return sys_rt_sigpending((sigset_t __user *)uset, sigsetsize);
#endif
L
Linus Torvalds 已提交
2671
}
A
Al Viro 已提交
2672
#endif
L
Linus Torvalds 已提交
2673 2674 2675

#ifndef HAVE_ARCH_COPY_SIGINFO_TO_USER

A
Al Viro 已提交
2676
int copy_siginfo_to_user(siginfo_t __user *to, const siginfo_t *from)
L
Linus Torvalds 已提交
2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687
{
	int err;

	if (!access_ok (VERIFY_WRITE, to, sizeof(siginfo_t)))
		return -EFAULT;
	if (from->si_code < 0)
		return __copy_to_user(to, from, sizeof(siginfo_t))
			? -EFAULT : 0;
	/*
	 * If you change siginfo_t structure, please be sure
	 * this code is fixed accordingly.
2688 2689
	 * Please remember to update the signalfd_copyinfo() function
	 * inside fs/signalfd.c too, in case siginfo_t changes.
L
Linus Torvalds 已提交
2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714
	 * It should never copy any pad contained in the structure
	 * to avoid security leaks, but must copy the generic
	 * 3 ints plus the relevant union member.
	 */
	err = __put_user(from->si_signo, &to->si_signo);
	err |= __put_user(from->si_errno, &to->si_errno);
	err |= __put_user((short)from->si_code, &to->si_code);
	switch (from->si_code & __SI_MASK) {
	case __SI_KILL:
		err |= __put_user(from->si_pid, &to->si_pid);
		err |= __put_user(from->si_uid, &to->si_uid);
		break;
	case __SI_TIMER:
		 err |= __put_user(from->si_tid, &to->si_tid);
		 err |= __put_user(from->si_overrun, &to->si_overrun);
		 err |= __put_user(from->si_ptr, &to->si_ptr);
		break;
	case __SI_POLL:
		err |= __put_user(from->si_band, &to->si_band);
		err |= __put_user(from->si_fd, &to->si_fd);
		break;
	case __SI_FAULT:
		err |= __put_user(from->si_addr, &to->si_addr);
#ifdef __ARCH_SI_TRAPNO
		err |= __put_user(from->si_trapno, &to->si_trapno);
A
Andi Kleen 已提交
2715 2716
#endif
#ifdef BUS_MCEERR_AO
2717
		/*
A
Andi Kleen 已提交
2718
		 * Other callers might not initialize the si_lsb field,
2719
		 * so check explicitly for the right codes here.
A
Andi Kleen 已提交
2720
		 */
2721 2722
		if (from->si_signo == SIGBUS &&
		    (from->si_code == BUS_MCEERR_AR || from->si_code == BUS_MCEERR_AO))
A
Andi Kleen 已提交
2723
			err |= __put_user(from->si_addr_lsb, &to->si_addr_lsb);
2724 2725
#endif
#ifdef SEGV_BNDERR
2726 2727 2728 2729
		if (from->si_signo == SIGSEGV && from->si_code == SEGV_BNDERR) {
			err |= __put_user(from->si_lower, &to->si_lower);
			err |= __put_user(from->si_upper, &to->si_upper);
		}
2730 2731 2732 2733
#endif
#ifdef SEGV_PKUERR
		if (from->si_signo == SIGSEGV && from->si_code == SEGV_PKUERR)
			err |= __put_user(from->si_pkey, &to->si_pkey);
L
Linus Torvalds 已提交
2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748
#endif
		break;
	case __SI_CHLD:
		err |= __put_user(from->si_pid, &to->si_pid);
		err |= __put_user(from->si_uid, &to->si_uid);
		err |= __put_user(from->si_status, &to->si_status);
		err |= __put_user(from->si_utime, &to->si_utime);
		err |= __put_user(from->si_stime, &to->si_stime);
		break;
	case __SI_RT: /* This is not generated by the kernel as of now. */
	case __SI_MESGQ: /* But this is */
		err |= __put_user(from->si_pid, &to->si_pid);
		err |= __put_user(from->si_uid, &to->si_uid);
		err |= __put_user(from->si_ptr, &to->si_ptr);
		break;
2749 2750 2751 2752 2753 2754 2755
#ifdef __ARCH_SIGSYS
	case __SI_SYS:
		err |= __put_user(from->si_call_addr, &to->si_call_addr);
		err |= __put_user(from->si_syscall, &to->si_syscall);
		err |= __put_user(from->si_arch, &to->si_arch);
		break;
#endif
L
Linus Torvalds 已提交
2756 2757 2758 2759 2760 2761 2762 2763 2764 2765
	default: /* this is just in case for now ... */
		err |= __put_user(from->si_pid, &to->si_pid);
		err |= __put_user(from->si_uid, &to->si_uid);
		break;
	}
	return err;
}

#endif

2766 2767 2768 2769 2770 2771 2772
/**
 *  do_sigtimedwait - wait for queued signals specified in @which
 *  @which: queued signals to wait for
 *  @info: if non-null, the signal's siginfo is returned here
 *  @ts: upper bound on process time suspension
 */
int do_sigtimedwait(const sigset_t *which, siginfo_t *info,
2773
		    const struct timespec *ts)
2774
{
T
Thomas Gleixner 已提交
2775
	ktime_t *to = NULL, timeout = KTIME_MAX;
2776 2777
	struct task_struct *tsk = current;
	sigset_t mask = *which;
2778
	int sig, ret = 0;
2779 2780 2781 2782

	if (ts) {
		if (!timespec_valid(ts))
			return -EINVAL;
2783 2784
		timeout = timespec_to_ktime(*ts);
		to = &timeout;
2785 2786 2787 2788 2789 2790 2791 2792 2793 2794
	}

	/*
	 * Invert the set of allowed signals to get those we want to block.
	 */
	sigdelsetmask(&mask, sigmask(SIGKILL) | sigmask(SIGSTOP));
	signotset(&mask);

	spin_lock_irq(&tsk->sighand->siglock);
	sig = dequeue_signal(tsk, &mask, info);
T
Thomas Gleixner 已提交
2795
	if (!sig && timeout) {
2796 2797 2798
		/*
		 * None ready, temporarily unblock those we're interested
		 * while we are sleeping in so that we'll be awakened when
2799 2800
		 * they arrive. Unblocking is always fine, we can avoid
		 * set_current_blocked().
2801 2802 2803 2804 2805 2806
		 */
		tsk->real_blocked = tsk->blocked;
		sigandsets(&tsk->blocked, &tsk->blocked, &mask);
		recalc_sigpending();
		spin_unlock_irq(&tsk->sighand->siglock);

2807 2808 2809
		__set_current_state(TASK_INTERRUPTIBLE);
		ret = freezable_schedule_hrtimeout_range(to, tsk->timer_slack_ns,
							 HRTIMER_MODE_REL);
2810
		spin_lock_irq(&tsk->sighand->siglock);
2811
		__set_task_blocked(tsk, &tsk->real_blocked);
2812
		sigemptyset(&tsk->real_blocked);
2813
		sig = dequeue_signal(tsk, &mask, info);
2814 2815 2816 2817 2818
	}
	spin_unlock_irq(&tsk->sighand->siglock);

	if (sig)
		return sig;
2819
	return ret ? -EINTR : -EAGAIN;
2820 2821
}

2822 2823 2824 2825 2826 2827 2828 2829
/**
 *  sys_rt_sigtimedwait - synchronously wait for queued signals specified
 *			in @uthese
 *  @uthese: queued signals to wait for
 *  @uinfo: if non-null, the signal's siginfo is returned here
 *  @uts: upper bound on process time suspension
 *  @sigsetsize: size of sigset_t type
 */
2830 2831 2832
SYSCALL_DEFINE4(rt_sigtimedwait, const sigset_t __user *, uthese,
		siginfo_t __user *, uinfo, const struct timespec __user *, uts,
		size_t, sigsetsize)
L
Linus Torvalds 已提交
2833 2834 2835 2836
{
	sigset_t these;
	struct timespec ts;
	siginfo_t info;
2837
	int ret;
L
Linus Torvalds 已提交
2838 2839 2840 2841 2842 2843 2844

	/* XXX: Don't preclude handling different sized sigset_t's.  */
	if (sigsetsize != sizeof(sigset_t))
		return -EINVAL;

	if (copy_from_user(&these, uthese, sizeof(these)))
		return -EFAULT;
2845

L
Linus Torvalds 已提交
2846 2847 2848 2849 2850
	if (uts) {
		if (copy_from_user(&ts, uts, sizeof(ts)))
			return -EFAULT;
	}

2851
	ret = do_sigtimedwait(&these, &info, uts ? &ts : NULL);
L
Linus Torvalds 已提交
2852

2853 2854 2855
	if (ret > 0 && uinfo) {
		if (copy_siginfo_to_user(uinfo, &info))
			ret = -EFAULT;
L
Linus Torvalds 已提交
2856 2857 2858 2859 2860
	}

	return ret;
}

2861 2862 2863 2864 2865
/**
 *  sys_kill - send a signal to a process
 *  @pid: the PID of the process
 *  @sig: signal to be sent
 */
2866
SYSCALL_DEFINE2(kill, pid_t, pid, int, sig)
L
Linus Torvalds 已提交
2867 2868 2869 2870 2871 2872
{
	struct siginfo info;

	info.si_signo = sig;
	info.si_errno = 0;
	info.si_code = SI_USER;
2873
	info.si_pid = task_tgid_vnr(current);
2874
	info.si_uid = from_kuid_munged(current_user_ns(), current_uid());
L
Linus Torvalds 已提交
2875 2876 2877 2878

	return kill_something_info(sig, &info, pid);
}

T
Thomas Gleixner 已提交
2879 2880
static int
do_send_specific(pid_t tgid, pid_t pid, int sig, struct siginfo *info)
L
Linus Torvalds 已提交
2881 2882
{
	struct task_struct *p;
T
Thomas Gleixner 已提交
2883
	int error = -ESRCH;
L
Linus Torvalds 已提交
2884

2885
	rcu_read_lock();
2886
	p = find_task_by_vpid(pid);
2887
	if (p && (tgid <= 0 || task_tgid_vnr(p) == tgid)) {
T
Thomas Gleixner 已提交
2888
		error = check_kill_permission(sig, info, p);
L
Linus Torvalds 已提交
2889 2890 2891 2892
		/*
		 * The null signal is a permissions and process existence
		 * probe.  No signal is actually delivered.
		 */
2893 2894 2895 2896 2897 2898 2899 2900 2901
		if (!error && sig) {
			error = do_send_sig_info(sig, info, p, false);
			/*
			 * If lock_task_sighand() failed we pretend the task
			 * dies after receiving the signal. The window is tiny,
			 * and the signal is private anyway.
			 */
			if (unlikely(error == -ESRCH))
				error = 0;
L
Linus Torvalds 已提交
2902 2903
		}
	}
2904
	rcu_read_unlock();
2905

L
Linus Torvalds 已提交
2906 2907 2908
	return error;
}

T
Thomas Gleixner 已提交
2909 2910
static int do_tkill(pid_t tgid, pid_t pid, int sig)
{
2911
	struct siginfo info = {};
T
Thomas Gleixner 已提交
2912 2913 2914 2915 2916

	info.si_signo = sig;
	info.si_errno = 0;
	info.si_code = SI_TKILL;
	info.si_pid = task_tgid_vnr(current);
2917
	info.si_uid = from_kuid_munged(current_user_ns(), current_uid());
T
Thomas Gleixner 已提交
2918 2919 2920 2921

	return do_send_specific(tgid, pid, sig, &info);
}

2922 2923 2924 2925 2926 2927
/**
 *  sys_tgkill - send signal to one specific thread
 *  @tgid: the thread group ID of the thread
 *  @pid: the PID of the thread
 *  @sig: signal to be sent
 *
2928
 *  This syscall also checks the @tgid and returns -ESRCH even if the PID
2929 2930 2931
 *  exists but it's not belonging to the target process anymore. This
 *  method solves the problem of threads exiting and PIDs getting reused.
 */
2932
SYSCALL_DEFINE3(tgkill, pid_t, tgid, pid_t, pid, int, sig)
2933 2934 2935 2936 2937 2938 2939 2940
{
	/* This is only valid for single tasks */
	if (pid <= 0 || tgid <= 0)
		return -EINVAL;

	return do_tkill(tgid, pid, sig);
}

2941 2942 2943 2944 2945
/**
 *  sys_tkill - send signal to one specific task
 *  @pid: the PID of the task
 *  @sig: signal to be sent
 *
L
Linus Torvalds 已提交
2946 2947
 *  Send a signal to only one task, even if it's a CLONE_THREAD task.
 */
2948
SYSCALL_DEFINE2(tkill, pid_t, pid, int, sig)
L
Linus Torvalds 已提交
2949 2950 2951 2952 2953
{
	/* This is only valid for single tasks */
	if (pid <= 0)
		return -EINVAL;

2954
	return do_tkill(0, pid, sig);
L
Linus Torvalds 已提交
2955 2956
}

A
Al Viro 已提交
2957 2958 2959 2960 2961
static int do_rt_sigqueueinfo(pid_t pid, int sig, siginfo_t *info)
{
	/* Not even root can pretend to send signals from the kernel.
	 * Nor can they impersonate a kill()/tgkill(), which adds source info.
	 */
2962
	if ((info->si_code >= 0 || info->si_code == SI_TKILL) &&
2963
	    (task_pid_vnr(current) != pid))
A
Al Viro 已提交
2964
		return -EPERM;
2965

A
Al Viro 已提交
2966 2967 2968 2969 2970 2971
	info->si_signo = sig;

	/* POSIX.1b doesn't mention process groups.  */
	return kill_proc_info(sig, info, pid);
}

2972 2973 2974 2975 2976 2977
/**
 *  sys_rt_sigqueueinfo - send signal information to a signal
 *  @pid: the PID of the thread
 *  @sig: signal to be sent
 *  @uinfo: signal info to be sent
 */
2978 2979
SYSCALL_DEFINE3(rt_sigqueueinfo, pid_t, pid, int, sig,
		siginfo_t __user *, uinfo)
L
Linus Torvalds 已提交
2980 2981 2982 2983
{
	siginfo_t info;
	if (copy_from_user(&info, uinfo, sizeof(siginfo_t)))
		return -EFAULT;
A
Al Viro 已提交
2984 2985
	return do_rt_sigqueueinfo(pid, sig, &info);
}
L
Linus Torvalds 已提交
2986

A
Al Viro 已提交
2987 2988 2989 2990 2991 2992
#ifdef CONFIG_COMPAT
COMPAT_SYSCALL_DEFINE3(rt_sigqueueinfo,
			compat_pid_t, pid,
			int, sig,
			struct compat_siginfo __user *, uinfo)
{
2993
	siginfo_t info = {};
A
Al Viro 已提交
2994 2995 2996 2997
	int ret = copy_siginfo_from_user32(&info, uinfo);
	if (unlikely(ret))
		return ret;
	return do_rt_sigqueueinfo(pid, sig, &info);
L
Linus Torvalds 已提交
2998
}
A
Al Viro 已提交
2999
#endif
L
Linus Torvalds 已提交
3000

3001
static int do_rt_tgsigqueueinfo(pid_t tgid, pid_t pid, int sig, siginfo_t *info)
3002 3003 3004 3005 3006 3007
{
	/* This is only valid for single tasks */
	if (pid <= 0 || tgid <= 0)
		return -EINVAL;

	/* Not even root can pretend to send signals from the kernel.
3008 3009
	 * Nor can they impersonate a kill()/tgkill(), which adds source info.
	 */
3010 3011
	if ((info->si_code >= 0 || info->si_code == SI_TKILL) &&
	    (task_pid_vnr(current) != pid))
3012
		return -EPERM;
3013

3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 3029
	info->si_signo = sig;

	return do_send_specific(tgid, pid, sig, info);
}

SYSCALL_DEFINE4(rt_tgsigqueueinfo, pid_t, tgid, pid_t, pid, int, sig,
		siginfo_t __user *, uinfo)
{
	siginfo_t info;

	if (copy_from_user(&info, uinfo, sizeof(siginfo_t)))
		return -EFAULT;

	return do_rt_tgsigqueueinfo(tgid, pid, sig, &info);
}

3030 3031 3032 3033 3034 3035 3036
#ifdef CONFIG_COMPAT
COMPAT_SYSCALL_DEFINE4(rt_tgsigqueueinfo,
			compat_pid_t, tgid,
			compat_pid_t, pid,
			int, sig,
			struct compat_siginfo __user *, uinfo)
{
3037
	siginfo_t info = {};
3038 3039 3040 3041 3042 3043 3044

	if (copy_siginfo_from_user32(&info, uinfo))
		return -EFAULT;
	return do_rt_tgsigqueueinfo(tgid, pid, sig, &info);
}
#endif

3045
/*
3046
 * For kthreads only, must not be used if cloned with CLONE_SIGHAND
3047
 */
3048
void kernel_sigaction(int sig, __sighandler_t action)
3049
{
3050
	spin_lock_irq(&current->sighand->siglock);
3051 3052 3053
	current->sighand->action[sig - 1].sa.sa_handler = action;
	if (action == SIG_IGN) {
		sigset_t mask;
3054

3055 3056
		sigemptyset(&mask);
		sigaddset(&mask, sig);
3057

3058 3059 3060 3061
		flush_sigqueue_mask(&mask, &current->signal->shared_pending);
		flush_sigqueue_mask(&mask, &current->pending);
		recalc_sigpending();
	}
3062 3063
	spin_unlock_irq(&current->sighand->siglock);
}
3064
EXPORT_SYMBOL(kernel_sigaction);
3065

3066 3067 3068 3069 3070
void __weak sigaction_compat_abi(struct k_sigaction *act,
		struct k_sigaction *oact)
{
}

3071
int do_sigaction(int sig, struct k_sigaction *act, struct k_sigaction *oact)
L
Linus Torvalds 已提交
3072
{
3073
	struct task_struct *p = current, *t;
L
Linus Torvalds 已提交
3074
	struct k_sigaction *k;
3075
	sigset_t mask;
L
Linus Torvalds 已提交
3076

3077
	if (!valid_signal(sig) || sig < 1 || (act && sig_kernel_only(sig)))
L
Linus Torvalds 已提交
3078 3079
		return -EINVAL;

3080
	k = &p->sighand->action[sig-1];
L
Linus Torvalds 已提交
3081

3082
	spin_lock_irq(&p->sighand->siglock);
L
Linus Torvalds 已提交
3083 3084 3085
	if (oact)
		*oact = *k;

3086 3087
	sigaction_compat_abi(act, oact);

L
Linus Torvalds 已提交
3088
	if (act) {
3089 3090
		sigdelsetmask(&act->sa.sa_mask,
			      sigmask(SIGKILL) | sigmask(SIGSTOP));
3091
		*k = *act;
L
Linus Torvalds 已提交
3092 3093 3094 3095 3096 3097 3098 3099 3100 3101 3102
		/*
		 * POSIX 3.3.1.3:
		 *  "Setting a signal action to SIG_IGN for a signal that is
		 *   pending shall cause the pending signal to be discarded,
		 *   whether or not it is blocked."
		 *
		 *  "Setting a signal action to SIG_DFL for a signal that is
		 *   pending and whose default action is to ignore the signal
		 *   (for example, SIGCHLD), shall cause the pending signal to
		 *   be discarded, whether or not it is blocked"
		 */
3103
		if (sig_handler_ignored(sig_handler(p, sig), sig)) {
3104 3105
			sigemptyset(&mask);
			sigaddset(&mask, sig);
3106 3107
			flush_sigqueue_mask(&mask, &p->signal->shared_pending);
			for_each_thread(p, t)
3108
				flush_sigqueue_mask(&mask, &t->pending);
L
Linus Torvalds 已提交
3109 3110 3111
		}
	}

3112
	spin_unlock_irq(&p->sighand->siglock);
L
Linus Torvalds 已提交
3113 3114 3115
	return 0;
}

3116
static int
L
Linus Torvalds 已提交
3117 3118 3119 3120 3121
do_sigaltstack (const stack_t __user *uss, stack_t __user *uoss, unsigned long sp)
{
	stack_t oss;
	int error;

3122 3123
	oss.ss_sp = (void __user *) current->sas_ss_sp;
	oss.ss_size = current->sas_ss_size;
3124 3125
	oss.ss_flags = sas_ss_flags(sp) |
		(current->sas_ss_flags & SS_FLAG_BITS);
L
Linus Torvalds 已提交
3126 3127 3128 3129

	if (uss) {
		void __user *ss_sp;
		size_t ss_size;
3130 3131
		unsigned ss_flags;
		int ss_mode;
L
Linus Torvalds 已提交
3132 3133

		error = -EFAULT;
L
Linus Torvalds 已提交
3134 3135 3136 3137 3138 3139
		if (!access_ok(VERIFY_READ, uss, sizeof(*uss)))
			goto out;
		error = __get_user(ss_sp, &uss->ss_sp) |
			__get_user(ss_flags, &uss->ss_flags) |
			__get_user(ss_size, &uss->ss_size);
		if (error)
L
Linus Torvalds 已提交
3140 3141 3142 3143 3144 3145
			goto out;

		error = -EPERM;
		if (on_sig_stack(sp))
			goto out;

3146
		ss_mode = ss_flags & ~SS_FLAG_BITS;
L
Linus Torvalds 已提交
3147
		error = -EINVAL;
3148 3149
		if (ss_mode != SS_DISABLE && ss_mode != SS_ONSTACK &&
				ss_mode != 0)
L
Linus Torvalds 已提交
3150 3151
			goto out;

3152
		if (ss_mode == SS_DISABLE) {
L
Linus Torvalds 已提交
3153 3154 3155 3156 3157 3158 3159 3160 3161 3162
			ss_size = 0;
			ss_sp = NULL;
		} else {
			error = -ENOMEM;
			if (ss_size < MINSIGSTKSZ)
				goto out;
		}

		current->sas_ss_sp = (unsigned long) ss_sp;
		current->sas_ss_size = ss_size;
3163
		current->sas_ss_flags = ss_flags;
L
Linus Torvalds 已提交
3164 3165
	}

3166
	error = 0;
L
Linus Torvalds 已提交
3167 3168
	if (uoss) {
		error = -EFAULT;
3169
		if (!access_ok(VERIFY_WRITE, uoss, sizeof(*uoss)))
L
Linus Torvalds 已提交
3170
			goto out;
3171 3172 3173
		error = __put_user(oss.ss_sp, &uoss->ss_sp) |
			__put_user(oss.ss_size, &uoss->ss_size) |
			__put_user(oss.ss_flags, &uoss->ss_flags);
L
Linus Torvalds 已提交
3174 3175 3176 3177 3178
	}

out:
	return error;
}
3179 3180 3181 3182
SYSCALL_DEFINE2(sigaltstack,const stack_t __user *,uss, stack_t __user *,uoss)
{
	return do_sigaltstack(uss, uoss, current_user_stack_pointer());
}
L
Linus Torvalds 已提交
3183

A
Al Viro 已提交
3184 3185 3186 3187 3188 3189 3190
int restore_altstack(const stack_t __user *uss)
{
	int err = do_sigaltstack(uss, NULL, current_user_stack_pointer());
	/* squash all but EFAULT for now */
	return err == -EFAULT ? err : 0;
}

3191 3192 3193
int __save_altstack(stack_t __user *uss, unsigned long sp)
{
	struct task_struct *t = current;
3194 3195
	int err = __put_user((void __user *)t->sas_ss_sp, &uss->ss_sp) |
		__put_user(t->sas_ss_flags, &uss->ss_flags) |
3196
		__put_user(t->sas_ss_size, &uss->ss_size);
3197 3198 3199 3200 3201
	if (err)
		return err;
	if (t->sas_ss_flags & SS_AUTODISARM)
		sas_ss_reset(t);
	return 0;
3202 3203
}

A
Al Viro 已提交
3204
#ifdef CONFIG_COMPAT
3205 3206 3207
COMPAT_SYSCALL_DEFINE2(sigaltstack,
			const compat_stack_t __user *, uss_ptr,
			compat_stack_t __user *, uoss_ptr)
A
Al Viro 已提交
3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220 3221 3222 3223 3224 3225 3226 3227 3228 3229 3230 3231 3232 3233 3234 3235 3236 3237 3238 3239 3240 3241 3242 3243 3244
{
	stack_t uss, uoss;
	int ret;
	mm_segment_t seg;

	if (uss_ptr) {
		compat_stack_t uss32;

		memset(&uss, 0, sizeof(stack_t));
		if (copy_from_user(&uss32, uss_ptr, sizeof(compat_stack_t)))
			return -EFAULT;
		uss.ss_sp = compat_ptr(uss32.ss_sp);
		uss.ss_flags = uss32.ss_flags;
		uss.ss_size = uss32.ss_size;
	}
	seg = get_fs();
	set_fs(KERNEL_DS);
	ret = do_sigaltstack((stack_t __force __user *) (uss_ptr ? &uss : NULL),
			     (stack_t __force __user *) &uoss,
			     compat_user_stack_pointer());
	set_fs(seg);
	if (ret >= 0 && uoss_ptr)  {
		if (!access_ok(VERIFY_WRITE, uoss_ptr, sizeof(compat_stack_t)) ||
		    __put_user(ptr_to_compat(uoss.ss_sp), &uoss_ptr->ss_sp) ||
		    __put_user(uoss.ss_flags, &uoss_ptr->ss_flags) ||
		    __put_user(uoss.ss_size, &uoss_ptr->ss_size))
			ret = -EFAULT;
	}
	return ret;
}

int compat_restore_altstack(const compat_stack_t __user *uss)
{
	int err = compat_sys_sigaltstack(uss, NULL);
	/* squash all but -EFAULT for now */
	return err == -EFAULT ? err : 0;
}
3245 3246 3247

int __compat_save_altstack(compat_stack_t __user *uss, unsigned long sp)
{
3248
	int err;
3249
	struct task_struct *t = current;
3250 3251 3252
	err = __put_user(ptr_to_compat((void __user *)t->sas_ss_sp),
			 &uss->ss_sp) |
		__put_user(t->sas_ss_flags, &uss->ss_flags) |
3253
		__put_user(t->sas_ss_size, &uss->ss_size);
3254 3255 3256 3257 3258
	if (err)
		return err;
	if (t->sas_ss_flags & SS_AUTODISARM)
		sas_ss_reset(t);
	return 0;
3259
}
A
Al Viro 已提交
3260
#endif
L
Linus Torvalds 已提交
3261 3262 3263

#ifdef __ARCH_WANT_SYS_SIGPENDING

3264 3265 3266 3267
/**
 *  sys_sigpending - examine pending signals
 *  @set: where mask of pending signal is returned
 */
3268
SYSCALL_DEFINE1(sigpending, old_sigset_t __user *, set)
L
Linus Torvalds 已提交
3269
{
A
Al Viro 已提交
3270
	return sys_rt_sigpending((sigset_t __user *)set, sizeof(old_sigset_t)); 
L
Linus Torvalds 已提交
3271 3272 3273 3274 3275
}

#endif

#ifdef __ARCH_WANT_SYS_SIGPROCMASK
3276 3277 3278
/**
 *  sys_sigprocmask - examine and change blocked signals
 *  @how: whether to add, remove, or set signals
3279
 *  @nset: signals to add or remove (if non-null)
3280 3281
 *  @oset: previous value of signal mask if non-null
 *
3282 3283 3284
 * Some platforms have their own version with special arguments;
 * others support only sys_rt_sigprocmask.
 */
L
Linus Torvalds 已提交
3285

3286
SYSCALL_DEFINE3(sigprocmask, int, how, old_sigset_t __user *, nset,
3287
		old_sigset_t __user *, oset)
L
Linus Torvalds 已提交
3288 3289
{
	old_sigset_t old_set, new_set;
3290
	sigset_t new_blocked;
L
Linus Torvalds 已提交
3291

3292
	old_set = current->blocked.sig[0];
L
Linus Torvalds 已提交
3293

3294 3295 3296
	if (nset) {
		if (copy_from_user(&new_set, nset, sizeof(*nset)))
			return -EFAULT;
L
Linus Torvalds 已提交
3297

3298
		new_blocked = current->blocked;
L
Linus Torvalds 已提交
3299 3300 3301

		switch (how) {
		case SIG_BLOCK:
3302
			sigaddsetmask(&new_blocked, new_set);
L
Linus Torvalds 已提交
3303 3304
			break;
		case SIG_UNBLOCK:
3305
			sigdelsetmask(&new_blocked, new_set);
L
Linus Torvalds 已提交
3306 3307
			break;
		case SIG_SETMASK:
3308
			new_blocked.sig[0] = new_set;
L
Linus Torvalds 已提交
3309
			break;
3310 3311
		default:
			return -EINVAL;
L
Linus Torvalds 已提交
3312 3313
		}

3314
		set_current_blocked(&new_blocked);
3315 3316 3317
	}

	if (oset) {
L
Linus Torvalds 已提交
3318
		if (copy_to_user(oset, &old_set, sizeof(*oset)))
3319
			return -EFAULT;
L
Linus Torvalds 已提交
3320
	}
3321 3322

	return 0;
L
Linus Torvalds 已提交
3323 3324 3325
}
#endif /* __ARCH_WANT_SYS_SIGPROCMASK */

3326
#ifndef CONFIG_ODD_RT_SIGACTION
3327 3328 3329
/**
 *  sys_rt_sigaction - alter an action taken by a process
 *  @sig: signal to be sent
3330 3331
 *  @act: new sigaction
 *  @oact: used to save the previous sigaction
3332 3333
 *  @sigsetsize: size of sigset_t type
 */
3334 3335 3336 3337
SYSCALL_DEFINE4(rt_sigaction, int, sig,
		const struct sigaction __user *, act,
		struct sigaction __user *, oact,
		size_t, sigsetsize)
L
Linus Torvalds 已提交
3338 3339 3340 3341 3342 3343 3344 3345 3346 3347 3348 3349 3350 3351 3352 3353 3354 3355 3356 3357 3358 3359
{
	struct k_sigaction new_sa, old_sa;
	int ret = -EINVAL;

	/* XXX: Don't preclude handling different sized sigset_t's.  */
	if (sigsetsize != sizeof(sigset_t))
		goto out;

	if (act) {
		if (copy_from_user(&new_sa.sa, act, sizeof(new_sa.sa)))
			return -EFAULT;
	}

	ret = do_sigaction(sig, act ? &new_sa : NULL, oact ? &old_sa : NULL);

	if (!ret && oact) {
		if (copy_to_user(oact, &old_sa.sa, sizeof(old_sa.sa)))
			return -EFAULT;
	}
out:
	return ret;
}
A
Al Viro 已提交
3360 3361 3362 3363 3364 3365 3366 3367 3368 3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 3380 3381 3382 3383 3384 3385
#ifdef CONFIG_COMPAT
COMPAT_SYSCALL_DEFINE4(rt_sigaction, int, sig,
		const struct compat_sigaction __user *, act,
		struct compat_sigaction __user *, oact,
		compat_size_t, sigsetsize)
{
	struct k_sigaction new_ka, old_ka;
	compat_sigset_t mask;
#ifdef __ARCH_HAS_SA_RESTORER
	compat_uptr_t restorer;
#endif
	int ret;

	/* XXX: Don't preclude handling different sized sigset_t's.  */
	if (sigsetsize != sizeof(compat_sigset_t))
		return -EINVAL;

	if (act) {
		compat_uptr_t handler;
		ret = get_user(handler, &act->sa_handler);
		new_ka.sa.sa_handler = compat_ptr(handler);
#ifdef __ARCH_HAS_SA_RESTORER
		ret |= get_user(restorer, &act->sa_restorer);
		new_ka.sa.sa_restorer = compat_ptr(restorer);
#endif
		ret |= copy_from_user(&mask, &act->sa_mask, sizeof(mask));
3386
		ret |= get_user(new_ka.sa.sa_flags, &act->sa_flags);
A
Al Viro 已提交
3387 3388 3389 3390 3391 3392 3393 3394 3395 3396 3397
		if (ret)
			return -EFAULT;
		sigset_from_compat(&new_ka.sa.sa_mask, &mask);
	}

	ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL);
	if (!ret && oact) {
		sigset_to_compat(&mask, &old_ka.sa.sa_mask);
		ret = put_user(ptr_to_compat(old_ka.sa.sa_handler), 
			       &oact->sa_handler);
		ret |= copy_to_user(&oact->sa_mask, &mask, sizeof(mask));
3398
		ret |= put_user(old_ka.sa.sa_flags, &oact->sa_flags);
A
Al Viro 已提交
3399 3400 3401 3402 3403 3404 3405 3406
#ifdef __ARCH_HAS_SA_RESTORER
		ret |= put_user(ptr_to_compat(old_ka.sa.sa_restorer),
				&oact->sa_restorer);
#endif
	}
	return ret;
}
#endif
3407
#endif /* !CONFIG_ODD_RT_SIGACTION */
L
Linus Torvalds 已提交
3408

3409 3410 3411 3412 3413 3414 3415 3416 3417 3418 3419 3420 3421 3422 3423 3424 3425 3426 3427 3428 3429 3430 3431 3432 3433 3434 3435 3436 3437 3438 3439 3440 3441 3442 3443 3444 3445 3446 3447 3448 3449 3450 3451 3452 3453 3454 3455 3456 3457 3458 3459 3460 3461 3462 3463 3464 3465 3466 3467 3468 3469 3470 3471 3472 3473 3474 3475 3476 3477 3478 3479 3480 3481 3482 3483 3484 3485
#ifdef CONFIG_OLD_SIGACTION
SYSCALL_DEFINE3(sigaction, int, sig,
		const struct old_sigaction __user *, act,
	        struct old_sigaction __user *, oact)
{
	struct k_sigaction new_ka, old_ka;
	int ret;

	if (act) {
		old_sigset_t mask;
		if (!access_ok(VERIFY_READ, act, sizeof(*act)) ||
		    __get_user(new_ka.sa.sa_handler, &act->sa_handler) ||
		    __get_user(new_ka.sa.sa_restorer, &act->sa_restorer) ||
		    __get_user(new_ka.sa.sa_flags, &act->sa_flags) ||
		    __get_user(mask, &act->sa_mask))
			return -EFAULT;
#ifdef __ARCH_HAS_KA_RESTORER
		new_ka.ka_restorer = NULL;
#endif
		siginitset(&new_ka.sa.sa_mask, mask);
	}

	ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL);

	if (!ret && oact) {
		if (!access_ok(VERIFY_WRITE, oact, sizeof(*oact)) ||
		    __put_user(old_ka.sa.sa_handler, &oact->sa_handler) ||
		    __put_user(old_ka.sa.sa_restorer, &oact->sa_restorer) ||
		    __put_user(old_ka.sa.sa_flags, &oact->sa_flags) ||
		    __put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask))
			return -EFAULT;
	}

	return ret;
}
#endif
#ifdef CONFIG_COMPAT_OLD_SIGACTION
COMPAT_SYSCALL_DEFINE3(sigaction, int, sig,
		const struct compat_old_sigaction __user *, act,
	        struct compat_old_sigaction __user *, oact)
{
	struct k_sigaction new_ka, old_ka;
	int ret;
	compat_old_sigset_t mask;
	compat_uptr_t handler, restorer;

	if (act) {
		if (!access_ok(VERIFY_READ, act, sizeof(*act)) ||
		    __get_user(handler, &act->sa_handler) ||
		    __get_user(restorer, &act->sa_restorer) ||
		    __get_user(new_ka.sa.sa_flags, &act->sa_flags) ||
		    __get_user(mask, &act->sa_mask))
			return -EFAULT;

#ifdef __ARCH_HAS_KA_RESTORER
		new_ka.ka_restorer = NULL;
#endif
		new_ka.sa.sa_handler = compat_ptr(handler);
		new_ka.sa.sa_restorer = compat_ptr(restorer);
		siginitset(&new_ka.sa.sa_mask, mask);
	}

	ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL);

	if (!ret && oact) {
		if (!access_ok(VERIFY_WRITE, oact, sizeof(*oact)) ||
		    __put_user(ptr_to_compat(old_ka.sa.sa_handler),
			       &oact->sa_handler) ||
		    __put_user(ptr_to_compat(old_ka.sa.sa_restorer),
			       &oact->sa_restorer) ||
		    __put_user(old_ka.sa.sa_flags, &oact->sa_flags) ||
		    __put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask))
			return -EFAULT;
	}
	return ret;
}
#endif
L
Linus Torvalds 已提交
3486

3487
#ifdef CONFIG_SGETMASK_SYSCALL
L
Linus Torvalds 已提交
3488 3489 3490 3491

/*
 * For backwards compatibility.  Functionality superseded by sigprocmask.
 */
3492
SYSCALL_DEFINE0(sgetmask)
L
Linus Torvalds 已提交
3493 3494 3495 3496 3497
{
	/* SMP safe */
	return current->blocked.sig[0];
}

3498
SYSCALL_DEFINE1(ssetmask, int, newmask)
L
Linus Torvalds 已提交
3499
{
3500 3501
	int old = current->blocked.sig[0];
	sigset_t newset;
L
Linus Torvalds 已提交
3502

3503
	siginitset(&newset, newmask);
3504
	set_current_blocked(&newset);
L
Linus Torvalds 已提交
3505 3506 3507

	return old;
}
3508
#endif /* CONFIG_SGETMASK_SYSCALL */
L
Linus Torvalds 已提交
3509 3510 3511 3512 3513

#ifdef __ARCH_WANT_SYS_SIGNAL
/*
 * For backwards compatibility.  Functionality superseded by sigaction.
 */
3514
SYSCALL_DEFINE2(signal, int, sig, __sighandler_t, handler)
L
Linus Torvalds 已提交
3515 3516 3517 3518 3519 3520
{
	struct k_sigaction new_sa, old_sa;
	int ret;

	new_sa.sa.sa_handler = handler;
	new_sa.sa.sa_flags = SA_ONESHOT | SA_NOMASK;
3521
	sigemptyset(&new_sa.sa.sa_mask);
L
Linus Torvalds 已提交
3522 3523 3524 3525 3526 3527 3528 3529 3530

	ret = do_sigaction(sig, &new_sa, &old_sa);

	return ret ? ret : (unsigned long)old_sa.sa.sa_handler;
}
#endif /* __ARCH_WANT_SYS_SIGNAL */

#ifdef __ARCH_WANT_SYS_PAUSE

3531
SYSCALL_DEFINE0(pause)
L
Linus Torvalds 已提交
3532
{
3533
	while (!signal_pending(current)) {
3534
		__set_current_state(TASK_INTERRUPTIBLE);
3535 3536
		schedule();
	}
L
Linus Torvalds 已提交
3537 3538 3539 3540 3541
	return -ERESTARTNOHAND;
}

#endif

3542
static int sigsuspend(sigset_t *set)
A
Al Viro 已提交
3543 3544 3545 3546
{
	current->saved_sigmask = current->blocked;
	set_current_blocked(set);

3547 3548 3549 3550
	while (!signal_pending(current)) {
		__set_current_state(TASK_INTERRUPTIBLE);
		schedule();
	}
A
Al Viro 已提交
3551 3552 3553 3554
	set_restore_sigmask();
	return -ERESTARTNOHAND;
}

3555 3556 3557 3558 3559 3560
/**
 *  sys_rt_sigsuspend - replace the signal mask for a value with the
 *	@unewset value until a signal is received
 *  @unewset: new signal mask value
 *  @sigsetsize: size of sigset_t type
 */
3561
SYSCALL_DEFINE2(rt_sigsuspend, sigset_t __user *, unewset, size_t, sigsetsize)
3562 3563 3564 3565 3566 3567 3568 3569 3570
{
	sigset_t newset;

	/* XXX: Don't preclude handling different sized sigset_t's.  */
	if (sigsetsize != sizeof(sigset_t))
		return -EINVAL;

	if (copy_from_user(&newset, unewset, sizeof(newset)))
		return -EFAULT;
A
Al Viro 已提交
3571
	return sigsuspend(&newset);
3572
}
A
Al Viro 已提交
3573 3574 3575 3576 3577 3578 3579 3580 3581 3582 3583 3584 3585 3586 3587 3588 3589 3590 3591 3592 3593 3594
 
#ifdef CONFIG_COMPAT
COMPAT_SYSCALL_DEFINE2(rt_sigsuspend, compat_sigset_t __user *, unewset, compat_size_t, sigsetsize)
{
#ifdef __BIG_ENDIAN
	sigset_t newset;
	compat_sigset_t newset32;

	/* XXX: Don't preclude handling different sized sigset_t's.  */
	if (sigsetsize != sizeof(sigset_t))
		return -EINVAL;

	if (copy_from_user(&newset32, unewset, sizeof(compat_sigset_t)))
		return -EFAULT;
	sigset_from_compat(&newset, &newset32);
	return sigsuspend(&newset);
#else
	/* on little-endian bitmaps don't care about granularity */
	return sys_rt_sigsuspend((sigset_t __user *)unewset, sigsetsize);
#endif
}
#endif
3595

3596 3597 3598 3599 3600 3601 3602 3603 3604 3605 3606 3607 3608 3609 3610 3611
#ifdef CONFIG_OLD_SIGSUSPEND
SYSCALL_DEFINE1(sigsuspend, old_sigset_t, mask)
{
	sigset_t blocked;
	siginitset(&blocked, mask);
	return sigsuspend(&blocked);
}
#endif
#ifdef CONFIG_OLD_SIGSUSPEND3
SYSCALL_DEFINE3(sigsuspend, int, unused1, int, unused2, old_sigset_t, mask)
{
	sigset_t blocked;
	siginitset(&blocked, mask);
	return sigsuspend(&blocked);
}
#endif
3612

3613
__weak const char *arch_vma_name(struct vm_area_struct *vma)
3614 3615 3616 3617
{
	return NULL;
}

L
Linus Torvalds 已提交
3618 3619
void __init signals_init(void)
{
3620 3621 3622 3623
	/* If this check fails, the __ARCH_SI_PREAMBLE_SIZE value is wrong! */
	BUILD_BUG_ON(__ARCH_SI_PREAMBLE_SIZE
		!= offsetof(struct siginfo, _sifields._pad));

3624
	sigqueue_cachep = KMEM_CACHE(sigqueue, SLAB_PANIC);
L
Linus Torvalds 已提交
3625
}
3626 3627 3628 3629 3630 3631 3632 3633 3634 3635 3636 3637 3638 3639 3640 3641 3642 3643 3644 3645 3646 3647 3648 3649 3650 3651 3652 3653 3654 3655 3656 3657 3658 3659 3660 3661 3662 3663 3664 3665

#ifdef CONFIG_KGDB_KDB
#include <linux/kdb.h>
/*
 * kdb_send_sig_info - Allows kdb to send signals without exposing
 * signal internals.  This function checks if the required locks are
 * available before calling the main signal code, to avoid kdb
 * deadlocks.
 */
void
kdb_send_sig_info(struct task_struct *t, struct siginfo *info)
{
	static struct task_struct *kdb_prev_t;
	int sig, new_t;
	if (!spin_trylock(&t->sighand->siglock)) {
		kdb_printf("Can't do kill command now.\n"
			   "The sigmask lock is held somewhere else in "
			   "kernel, try again later\n");
		return;
	}
	spin_unlock(&t->sighand->siglock);
	new_t = kdb_prev_t != t;
	kdb_prev_t = t;
	if (t->state != TASK_RUNNING && new_t) {
		kdb_printf("Process is not RUNNING, sending a signal from "
			   "kdb risks deadlock\n"
			   "on the run queue locks. "
			   "The signal has _not_ been sent.\n"
			   "Reissue the kill command if you want to risk "
			   "the deadlock.\n");
		return;
	}
	sig = info->si_signo;
	if (send_sig_info(sig, info, t))
		kdb_printf("Fail to deliver Signal %d to process %d.\n",
			   sig, t->pid);
	else
		kdb_printf("Signal %d is sent to process %d.\n", sig, t->pid);
}
#endif	/* CONFIG_KGDB_KDB */