signal.c 69.3 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
/*
 *  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>
#include <linux/module.h>
#include <linux/init.h>
#include <linux/sched.h>
#include <linux/fs.h>
#include <linux/tty.h>
#include <linux/binfmts.h>
#include <linux/security.h>
#include <linux/syscalls.h>
#include <linux/ptrace.h>
23
#include <linux/signal.h>
24
#include <linux/signalfd.h>
25
#include <linux/ratelimit.h>
26
#include <linux/tracehook.h>
27
#include <linux/capability.h>
28
#include <linux/freezer.h>
29 30
#include <linux/pid_namespace.h>
#include <linux/nsproxy.h>
31 32
#define CREATE_TRACE_POINTS
#include <trace/events/signal.h>
33

L
Linus Torvalds 已提交
34 35 36 37
#include <asm/param.h>
#include <asm/uaccess.h>
#include <asm/unistd.h>
#include <asm/siginfo.h>
38
#include "audit.h"	/* audit_signal_info() */
L
Linus Torvalds 已提交
39 40 41 42 43

/*
 * SLAB caches for signal bits.
 */

44
static struct kmem_cache *sigqueue_cachep;
L
Linus Torvalds 已提交
45

46 47
int print_fatal_signals __read_mostly;

48
static void __user *sig_handler(struct task_struct *t, int sig)
49
{
50 51
	return t->sighand->action[sig - 1].sa.sa_handler;
}
52

53 54
static int sig_handler_ignored(void __user *handler, int sig)
{
55 56 57 58
	/* Is it explicitly or implicitly ignored? */
	return handler == SIG_IGN ||
		(handler == SIG_DFL && sig_kernel_ignore(sig));
}
L
Linus Torvalds 已提交
59

60 61
static int sig_task_ignored(struct task_struct *t, int sig,
		int from_ancestor_ns)
L
Linus Torvalds 已提交
62
{
63
	void __user *handler;
L
Linus Torvalds 已提交
64

65 66 67
	handler = sig_handler(t, sig);

	if (unlikely(t->signal->flags & SIGNAL_UNKILLABLE) &&
68
			handler == SIG_DFL && !from_ancestor_ns)
69 70 71 72 73
		return 1;

	return sig_handler_ignored(handler, sig);
}

74
static int sig_ignored(struct task_struct *t, int sig, int from_ancestor_ns)
75
{
L
Linus Torvalds 已提交
76 77 78 79 80
	/*
	 * Blocked signals are never ignored, since the
	 * signal handler may change by the time it is
	 * unblocked.
	 */
81
	if (sigismember(&t->blocked, sig) || sigismember(&t->real_blocked, sig))
L
Linus Torvalds 已提交
82 83
		return 0;

84
	if (!sig_task_ignored(t, sig, from_ancestor_ns))
85 86 87 88 89
		return 0;

	/*
	 * Tracers may want to know about even ignored signals.
	 */
90
	return !tracehook_consider_ignored_signal(t, sig);
L
Linus Torvalds 已提交
91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124
}

/*
 * 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
125
static int recalc_sigpending_tsk(struct task_struct *t)
L
Linus Torvalds 已提交
126 127 128
{
	if (t->signal->group_stop_count > 0 ||
	    PENDING(&t->pending, &t->blocked) ||
R
Roland McGrath 已提交
129
	    PENDING(&t->signal->shared_pending, &t->blocked)) {
L
Linus Torvalds 已提交
130
		set_tsk_thread_flag(t, TIF_SIGPENDING);
R
Roland McGrath 已提交
131 132
		return 1;
	}
133 134 135 136 137
	/*
	 * 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
138 139 140 141 142 143 144 145 146 147 148
	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
149 150 151 152
}

void recalc_sigpending(void)
{
153 154 155
	if (unlikely(tracehook_force_sigpending()))
		set_thread_flag(TIF_SIGPENDING);
	else if (!recalc_sigpending_tsk(current) && !freezing(current))
156 157
		clear_thread_flag(TIF_SIGPENDING);

L
Linus Torvalds 已提交
158 159 160 161
}

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

162 163 164 165
#define SYNCHRONOUS_MASK \
	(sigmask(SIGSEGV) | sigmask(SIGBUS) | sigmask(SIGILL) | \
	 sigmask(SIGTRAP) | sigmask(SIGFPE))

166
int next_signal(struct sigpending *pending, sigset_t *mask)
L
Linus Torvalds 已提交
167 168 169
{
	unsigned long i, *s, *m, x;
	int sig = 0;
170

L
Linus Torvalds 已提交
171 172
	s = pending->signal.sig;
	m = mask->sig;
173 174 175 176 177 178 179 180 181 182 183 184 185

	/*
	 * 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
186 187
	switch (_NSIG_WORDS) {
	default:
188 189 190 191 192 193 194
		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
195 196
		break;

197 198 199
	case 2:
		x = s[1] &~ m[1];
		if (!x)
L
Linus Torvalds 已提交
200
			break;
201
		sig = ffz(~x) + _NSIG_BPW + 1;
L
Linus Torvalds 已提交
202 203
		break;

204 205
	case 1:
		/* Nothing to do */
L
Linus Torvalds 已提交
206 207
		break;
	}
208

L
Linus Torvalds 已提交
209 210 211
	return sig;
}

212 213 214 215 216 217 218 219 220 221 222 223 224 225
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;

	printk(KERN_INFO "%s/%d: reached RLIMIT_SIGPENDING, dropped signal %d\n",
				current->comm, current->pid, sig);
}

226 227 228
/*
 * allocate a new signal queue record
 * - this may be called without locks if and only if t == current, otherwise an
D
David Howells 已提交
229
 *   appopriate lock must be held to stop the target task from exiting
230
 */
231 232
static struct sigqueue *
__sigqueue_alloc(int sig, struct task_struct *t, gfp_t flags, int override_rlimit)
L
Linus Torvalds 已提交
233 234
{
	struct sigqueue *q = NULL;
235
	struct user_struct *user;
L
Linus Torvalds 已提交
236

237
	/*
238 239
	 * Protect access to @t credentials. This can go away when all
	 * callers hold rcu read lock.
240
	 */
241
	rcu_read_lock();
D
David Howells 已提交
242
	user = get_uid(__task_cred(t)->user);
243
	atomic_inc(&user->sigpending);
244
	rcu_read_unlock();
245

L
Linus Torvalds 已提交
246
	if (override_rlimit ||
247
	    atomic_read(&user->sigpending) <=
248
			task_rlimit(t, RLIMIT_SIGPENDING)) {
L
Linus Torvalds 已提交
249
		q = kmem_cache_alloc(sigqueue_cachep, flags);
250 251 252 253
	} else {
		print_dropped_signal(sig);
	}

L
Linus Torvalds 已提交
254
	if (unlikely(q == NULL)) {
255
		atomic_dec(&user->sigpending);
D
David Howells 已提交
256
		free_uid(user);
L
Linus Torvalds 已提交
257 258 259
	} else {
		INIT_LIST_HEAD(&q->list);
		q->flags = 0;
D
David Howells 已提交
260
		q->user = user;
L
Linus Torvalds 已提交
261
	}
D
David Howells 已提交
262 263

	return q;
L
Linus Torvalds 已提交
264 265
}

266
static void __sigqueue_free(struct sigqueue *q)
L
Linus Torvalds 已提交
267 268 269 270 271 272 273 274
{
	if (q->flags & SIGQUEUE_PREALLOC)
		return;
	atomic_dec(&q->user->sigpending);
	free_uid(q->user);
	kmem_cache_free(sigqueue_cachep, q);
}

275
void flush_sigqueue(struct sigpending *queue)
L
Linus Torvalds 已提交
276 277 278 279 280 281 282 283 284 285 286 287 288 289
{
	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);
	}
}

/*
 * Flush all pending signals for a task.
 */
290 291 292 293 294 295 296
void __flush_signals(struct task_struct *t)
{
	clear_tsk_thread_flag(t, TIF_SIGPENDING);
	flush_sigqueue(&t->pending);
	flush_sigqueue(&t->signal->shared_pending);
}

297
void flush_signals(struct task_struct *t)
L
Linus Torvalds 已提交
298 299 300 301
{
	unsigned long flags;

	spin_lock_irqsave(&t->sighand->siglock, flags);
302
	__flush_signals(t);
L
Linus Torvalds 已提交
303 304 305
	spin_unlock_irqrestore(&t->sighand->siglock, flags);
}

306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339
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);
}

340 341 342 343 344 345 346 347 348 349
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
350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367
/*
 * 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;
		sigemptyset(&ka->sa.sa_mask);
		ka++;
	}
}

368 369
int unhandled_signal(struct task_struct *tsk, int sig)
{
370
	void __user *handler = tsk->sighand->action[sig-1].sa.sa_handler;
371
	if (is_global_init(tsk))
372
		return 1;
373
	if (handler != SIG_IGN && handler != SIG_DFL)
374
		return 0;
375
	return !tracehook_consider_fatal_signal(tsk, sig);
376 377
}

L
Linus Torvalds 已提交
378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412

/* Notify the system that a driver wants to block all signals for this
 * process, and wants to be notified if any signals at all were to be
 * sent/acted upon.  If the notifier routine returns non-zero, then the
 * signal will be acted upon after all.  If the notifier routine returns 0,
 * then then signal will be blocked.  Only one block per process is
 * allowed.  priv is a pointer to private data that the notifier routine
 * can use to determine if the signal should be blocked or not.  */

void
block_all_signals(int (*notifier)(void *priv), void *priv, sigset_t *mask)
{
	unsigned long flags;

	spin_lock_irqsave(&current->sighand->siglock, flags);
	current->notifier_mask = mask;
	current->notifier_data = priv;
	current->notifier = notifier;
	spin_unlock_irqrestore(&current->sighand->siglock, flags);
}

/* Notify the system that blocking has ended. */

void
unblock_all_signals(void)
{
	unsigned long flags;

	spin_lock_irqsave(&current->sighand->siglock, flags);
	current->notifier = NULL;
	current->notifier_data = NULL;
	recalc_sigpending();
	spin_unlock_irqrestore(&current->sighand->siglock, flags);
}

413
static void collect_signal(int sig, struct sigpending *list, siginfo_t *info)
L
Linus Torvalds 已提交
414 415 416 417 418 419 420 421 422
{
	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) {
423 424
			if (first)
				goto still_pending;
L
Linus Torvalds 已提交
425 426 427
			first = q;
		}
	}
428 429 430

	sigdelset(&list->signal, sig);

L
Linus Torvalds 已提交
431
	if (first) {
432
still_pending:
L
Linus Torvalds 已提交
433 434 435 436 437 438 439 440 441 442
		list_del_init(&first->list);
		copy_siginfo(info, &first->info);
		__sigqueue_free(first);
	} else {
		/* 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.
		 */
		info->si_signo = sig;
		info->si_errno = 0;
443
		info->si_code = SI_USER;
L
Linus Torvalds 已提交
444 445 446 447 448 449 450 451
		info->si_pid = 0;
		info->si_uid = 0;
	}
}

static int __dequeue_signal(struct sigpending *pending, sigset_t *mask,
			siginfo_t *info)
{
452
	int sig = next_signal(pending, mask);
L
Linus Torvalds 已提交
453 454 455 456 457 458 459 460 461 462 463

	if (sig) {
		if (current->notifier) {
			if (sigismember(current->notifier_mask, sig)) {
				if (!(current->notifier)(current->notifier_data)) {
					clear_thread_flag(TIF_SIGPENDING);
					return 0;
				}
			}
		}

464
		collect_signal(sig, pending, info);
L
Linus Torvalds 已提交
465 466 467 468 469 470 471 472 473 474 475 476 477
	}

	return sig;
}

/*
 * Dequeue a signal and return the element to the caller, which is 
 * expected to free it.
 *
 * All callers have to hold the siglock.
 */
int dequeue_signal(struct task_struct *tsk, sigset_t *mask, siginfo_t *info)
{
478
	int signr;
479 480 481 482

	/* We only dequeue private signals from ourselves, we don't let
	 * signalfd steal them
	 */
D
Davide Libenzi 已提交
483
	signr = __dequeue_signal(&tsk->pending, mask, info);
484
	if (!signr) {
L
Linus Torvalds 已提交
485 486
		signr = __dequeue_signal(&tsk->signal->shared_pending,
					 mask, info);
487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510
		/*
		 * 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
		 * compliant with the old way of self restarting
		 * 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) &&
			    tsk->signal->it_real_incr.tv64 != 0) {
				hrtimer_forward(tmr, tmr->base->get_time(),
						tsk->signal->it_real_incr);
				hrtimer_restart(tmr);
			}
		}
	}
511

D
Davide Libenzi 已提交
512
	recalc_sigpending();
513 514 515 516
	if (!signr)
		return 0;

	if (unlikely(sig_kernel_stop(signr))) {
517 518 519 520 521 522 523 524 525 526 527 528
		/*
		 * 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.
		 */
529
		tsk->signal->flags |= SIGNAL_STOP_DEQUEUED;
530
	}
531
	if ((info->si_code & __SI_MASK) == __SI_TIMER && info->si_sys_private) {
L
Linus Torvalds 已提交
532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562
		/*
		 * 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);
	}
	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
 */
void signal_wake_up(struct task_struct *t, int resume)
{
	unsigned int mask;

	set_tsk_thread_flag(t, TIF_SIGPENDING);

	/*
M
Matthew Wilcox 已提交
563 564
	 * For SIGKILL, we want to wake it up in the stopped/traced/killable
	 * case. We don't check t->state here because there is a race with it
L
Linus Torvalds 已提交
565 566 567 568 569 570
	 * 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.
	 */
	mask = TASK_INTERRUPTIBLE;
	if (resume)
M
Matthew Wilcox 已提交
571
		mask |= TASK_WAKEKILL;
L
Linus Torvalds 已提交
572 573 574 575
	if (!wake_up_state(t, mask))
		kick_process(t);
}

576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602
/*
 * Remove signals in mask from the pending set and queue.
 * Returns 1 if any signals were found.
 *
 * All callers must be holding the siglock.
 *
 * This version takes a sigset mask and looks at all signals,
 * not just those in the first mask word.
 */
static int rm_from_queue_full(sigset_t *mask, struct sigpending *s)
{
	struct sigqueue *q, *n;
	sigset_t m;

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

	signandsets(&s->signal, &s->signal, mask);
	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
603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626
/*
 * Remove signals in mask from the pending set and queue.
 * Returns 1 if any signals were found.
 *
 * All callers must be holding the siglock.
 */
static int rm_from_queue(unsigned long mask, struct sigpending *s)
{
	struct sigqueue *q, *n;

	if (!sigtestsetmask(&s->signal, mask))
		return 0;

	sigdelsetmask(&s->signal, mask);
	list_for_each_entry_safe(q, n, &s->list, list) {
		if (q->info.si_signo < SIGRTMIN &&
		    (mask & sigmask(q->info.si_signo))) {
			list_del_init(&q->list);
			__sigqueue_free(q);
		}
	}
	return 1;
}

627 628 629 630 631 632 633 634 635 636 637
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));
}

L
Linus Torvalds 已提交
638 639
/*
 * Bad permissions for sending the signal
640
 * - the caller must hold at least the RCU read lock
L
Linus Torvalds 已提交
641 642 643 644
 */
static int check_kill_permission(int sig, struct siginfo *info,
				 struct task_struct *t)
{
645
	const struct cred *cred, *tcred;
646
	struct pid *sid;
647 648
	int error;

649
	if (!valid_signal(sig))
650 651
		return -EINVAL;

652
	if (!si_fromuser(info))
653
		return 0;
A
Amy Griffis 已提交
654

655 656
	error = audit_signal_info(sig, t); /* Let audit system see the signal */
	if (error)
L
Linus Torvalds 已提交
657
		return error;
658

659
	cred = current_cred();
660
	tcred = __task_cred(t);
661 662
	if (!same_thread_group(current, t) &&
	    (cred->euid ^ tcred->suid) &&
663 664 665
	    (cred->euid ^ tcred->uid) &&
	    (cred->uid  ^ tcred->suid) &&
	    (cred->uid  ^ tcred->uid) &&
666 667 668 669 670 671 672 673 674 675 676 677 678 679
	    !capable(CAP_KILL)) {
		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;
		}
	}
680

A
Amy Griffis 已提交
681
	return security_task_kill(t, info, sig, 0);
L
Linus Torvalds 已提交
682 683 684
}

/*
685 686
 * 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
687 688
 * time regardless of blocking, ignoring, or handling.  This does the
 * actual continuing for SIGCONT, but not the actual stopping for stop
689 690 691 692
 * 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
693
 */
694
static int prepare_signal(int sig, struct task_struct *p, int from_ancestor_ns)
L
Linus Torvalds 已提交
695
{
696
	struct signal_struct *signal = p->signal;
L
Linus Torvalds 已提交
697 698
	struct task_struct *t;

699
	if (unlikely(signal->flags & SIGNAL_GROUP_EXIT)) {
L
Linus Torvalds 已提交
700
		/*
701
		 * The process is in the middle of dying, nothing to do.
L
Linus Torvalds 已提交
702
		 */
703
	} else if (sig_kernel_stop(sig)) {
L
Linus Torvalds 已提交
704 705 706
		/*
		 * This is a stop signal.  Remove SIGCONT from all queues.
		 */
707
		rm_from_queue(sigmask(SIGCONT), &signal->shared_pending);
L
Linus Torvalds 已提交
708 709 710
		t = p;
		do {
			rm_from_queue(sigmask(SIGCONT), &t->pending);
711
		} while_each_thread(p, t);
L
Linus Torvalds 已提交
712
	} else if (sig == SIGCONT) {
713
		unsigned int why;
L
Linus Torvalds 已提交
714 715 716 717
		/*
		 * Remove all stop signals from all queues,
		 * and wake all threads.
		 */
718
		rm_from_queue(SIG_KERNEL_STOP_MASK, &signal->shared_pending);
L
Linus Torvalds 已提交
719 720 721 722 723 724 725 726 727 728 729 730 731
		t = p;
		do {
			unsigned int state;
			rm_from_queue(SIG_KERNEL_STOP_MASK, &t->pending);
			/*
			 * If there is a handler for SIGCONT, we must make
			 * sure that no thread returns to user mode before
			 * we post the signal, in case it was the only
			 * thread eligible to run the signal handler--then
			 * it must not do anything between resuming and
			 * running the handler.  With the TIF_SIGPENDING
			 * flag set, the thread will pause and acquire the
			 * siglock that we hold now and until we've queued
732
			 * the pending signal.
L
Linus Torvalds 已提交
733 734 735 736
			 *
			 * Wake up the stopped thread _after_ setting
			 * TIF_SIGPENDING
			 */
M
Matthew Wilcox 已提交
737
			state = __TASK_STOPPED;
L
Linus Torvalds 已提交
738 739 740 741 742
			if (sig_user_defined(t, SIGCONT) && !sigismember(&t->blocked, SIGCONT)) {
				set_tsk_thread_flag(t, TIF_SIGPENDING);
				state |= TASK_INTERRUPTIBLE;
			}
			wake_up_state(t, state);
743
		} while_each_thread(p, t);
L
Linus Torvalds 已提交
744

745 746 747 748 749 750 751 752 753
		/*
		 * 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;
754
		if (signal->flags & SIGNAL_STOP_STOPPED)
755
			why |= SIGNAL_CLD_CONTINUED;
756
		else if (signal->group_stop_count)
757 758 759
			why |= SIGNAL_CLD_STOPPED;

		if (why) {
760
			/*
761
			 * The first thread which returns from do_signal_stop()
762 763 764
			 * will take ->siglock, notice SIGNAL_CLD_MASK, and
			 * notify its parent. See get_signal_to_deliver().
			 */
765 766 767
			signal->flags = why | SIGNAL_STOP_CONTINUED;
			signal->group_stop_count = 0;
			signal->group_exit_code = 0;
L
Linus Torvalds 已提交
768 769 770 771 772 773
		} else {
			/*
			 * We are not stopped, but there could be a stop
			 * signal in the middle of being processed after
			 * being removed from the queue.  Clear that too.
			 */
774
			signal->flags &= ~SIGNAL_STOP_DEQUEUED;
L
Linus Torvalds 已提交
775 776
		}
	}
777

778
	return !sig_ignored(p, sig, from_ancestor_ns);
L
Linus Torvalds 已提交
779 780
}

781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801
/*
 * 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);
}

802
static void complete_signal(int sig, struct task_struct *p, int group)
803 804 805 806 807 808 809 810 811 812 813 814
{
	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;
815
	else if (!group || thread_group_empty(p))
816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842
		/*
		 * 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.
	 */
843 844
	if (sig_fatal(p, sig) &&
	    !(signal->flags & (SIGNAL_UNKILLABLE | SIGNAL_GROUP_EXIT)) &&
845
	    !sigismember(&t->real_blocked, sig) &&
846
	    (sig == SIGKILL ||
847
	     !tracehook_consider_fatal_signal(t, sig))) {
848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877
		/*
		 * 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 {
				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;
}

878 879 880 881 882
static inline int legacy_queue(struct sigpending *signals, int sig)
{
	return (sig < SIGRTMIN) && sigismember(&signals->signal, sig);
}

883 884
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
885
{
886
	struct sigpending *pending;
887
	struct sigqueue *q;
888
	int override_rlimit;
L
Linus Torvalds 已提交
889

890
	trace_signal_generate(sig, info, t);
891

892
	assert_spin_locked(&t->sighand->siglock);
893 894

	if (!prepare_signal(sig, t, from_ancestor_ns))
895
		return 0;
896 897

	pending = group ? &t->signal->shared_pending : &t->pending;
898 899 900 901 902
	/*
	 * 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.
	 */
903
	if (legacy_queue(pending, sig))
904
		return 0;
L
Linus Torvalds 已提交
905 906 907 908
	/*
	 * fast-pathed signals for kernel-internal things like SIGSTOP
	 * or SIGKILL.
	 */
909
	if (info == SEND_SIG_FORCED)
L
Linus Torvalds 已提交
910 911 912 913 914 915 916 917 918 919
		goto out_set;

	/* 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.  */

920 921 922 923 924
	if (sig < SIGRTMIN)
		override_rlimit = (is_si_special(info) || info->si_code >= 0);
	else
		override_rlimit = 0;

925
	q = __sigqueue_alloc(sig, t, GFP_ATOMIC | __GFP_NOTRACK_FALSE_POSITIVE,
926
		override_rlimit);
L
Linus Torvalds 已提交
927
	if (q) {
928
		list_add_tail(&q->list, &pending->list);
L
Linus Torvalds 已提交
929
		switch ((unsigned long) info) {
930
		case (unsigned long) SEND_SIG_NOINFO:
L
Linus Torvalds 已提交
931 932 933
			q->info.si_signo = sig;
			q->info.si_errno = 0;
			q->info.si_code = SI_USER;
934
			q->info.si_pid = task_tgid_nr_ns(current,
935
							task_active_pid_ns(t));
936
			q->info.si_uid = current_uid();
L
Linus Torvalds 已提交
937
			break;
938
		case (unsigned long) SEND_SIG_PRIV:
L
Linus Torvalds 已提交
939 940 941 942 943 944 945 946
			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);
947 948
			if (from_ancestor_ns)
				q->info.si_pid = 0;
L
Linus Torvalds 已提交
949 950
			break;
		}
951
	} else if (!is_si_special(info)) {
952 953 954 955 956 957 958
		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().
			 */
			trace_signal_overflow_fail(sig, group, info);
L
Linus Torvalds 已提交
959
			return -EAGAIN;
960 961 962 963 964 965 966
		} else {
			/*
			 * This is a silent loss of information.  We still
			 * send the signal, but the *info bits are lost.
			 */
			trace_signal_lose_info(sig, group, info);
		}
L
Linus Torvalds 已提交
967 968 969
	}

out_set:
970
	signalfd_notify(t, sig);
971
	sigaddset(&pending->signal, sig);
972 973
	complete_signal(sig, t, group);
	return 0;
L
Linus Torvalds 已提交
974 975
}

976 977 978
static int send_signal(int sig, struct siginfo *info, struct task_struct *t,
			int group)
{
979 980 981
	int from_ancestor_ns = 0;

#ifdef CONFIG_PID_NS
982 983
	from_ancestor_ns = si_fromuser(info) &&
			   !task_pid_nr_ns(current, task_active_pid_ns(t));
984 985 986
#endif

	return __send_signal(sig, info, t, group, from_ancestor_ns);
987 988
}

I
Ingo Molnar 已提交
989 990 991
static void print_fatal_signal(struct pt_regs *regs, int signr)
{
	printk("%s/%d: potentially unexpected fatal signal %d.\n",
992
		current->comm, task_pid_nr(current), signr);
I
Ingo Molnar 已提交
993

A
Al Viro 已提交
994
#if defined(__i386__) && !defined(__arch_um__)
995
	printk("code at %08lx: ", regs->ip);
I
Ingo Molnar 已提交
996 997 998 999 1000
	{
		int i;
		for (i = 0; i < 16; i++) {
			unsigned char insn;

1001 1002
			if (get_user(insn, (unsigned char *)(regs->ip + i)))
				break;
I
Ingo Molnar 已提交
1003 1004 1005 1006 1007
			printk("%02x ", insn);
		}
	}
#endif
	printk("\n");
1008
	preempt_disable();
I
Ingo Molnar 已提交
1009
	show_regs(regs);
1010
	preempt_enable();
I
Ingo Molnar 已提交
1011 1012 1013 1014 1015 1016 1017 1018 1019 1020
}

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
1021

1022 1023 1024 1025 1026 1027
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
1028 1029 1030
static int
specific_send_sig_info(int sig, struct siginfo *info, struct task_struct *t)
{
1031
	return send_signal(sig, info, t, 0);
L
Linus Torvalds 已提交
1032 1033
}

1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047
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
1048 1049 1050
/*
 * Force a signal that the process can't ignore: if necessary
 * we unblock the signal and change any SIG_IGN to SIG_DFL.
1051 1052 1053 1054 1055
 *
 * 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.
 *
1056 1057
 * 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
1058 1059 1060 1061 1062
 */
int
force_sig_info(int sig, struct siginfo *info, struct task_struct *t)
{
	unsigned long int flags;
1063 1064
	int ret, blocked, ignored;
	struct k_sigaction *action;
L
Linus Torvalds 已提交
1065 1066

	spin_lock_irqsave(&t->sighand->siglock, flags);
1067 1068 1069 1070 1071 1072 1073
	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
1074
			recalc_sigpending_and_wake(t);
1075
		}
L
Linus Torvalds 已提交
1076
	}
1077 1078
	if (action->sa.sa_handler == SIG_DFL)
		t->signal->flags &= ~SIGNAL_UNKILLABLE;
L
Linus Torvalds 已提交
1079 1080 1081 1082 1083 1084 1085 1086 1087
	ret = specific_send_sig_info(sig, info, t);
	spin_unlock_irqrestore(&t->sighand->siglock, flags);

	return ret;
}

/*
 * Nuke all other threads in the group.
 */
1088
int zap_other_threads(struct task_struct *p)
L
Linus Torvalds 已提交
1089
{
1090 1091
	struct task_struct *t = p;
	int count = 0;
L
Linus Torvalds 已提交
1092 1093 1094

	p->signal->group_stop_count = 0;

1095 1096 1097 1098
	while_each_thread(p, t) {
		count++;

		/* Don't bother with already dead threads */
L
Linus Torvalds 已提交
1099 1100 1101 1102 1103
		if (t->exit_state)
			continue;
		sigaddset(&t->pending.signal, SIGKILL);
		signal_wake_up(t, 1);
	}
1104 1105

	return count;
L
Linus Torvalds 已提交
1106 1107
}

1108 1109 1110 1111
struct sighand_struct *lock_task_sighand(struct task_struct *tsk, unsigned long *flags)
{
	struct sighand_struct *sighand;

1112
	rcu_read_lock();
1113 1114 1115 1116 1117 1118 1119 1120 1121 1122
	for (;;) {
		sighand = rcu_dereference(tsk->sighand);
		if (unlikely(sighand == NULL))
			break;

		spin_lock_irqsave(&sighand->siglock, *flags);
		if (likely(sighand == tsk->sighand))
			break;
		spin_unlock_irqrestore(&sighand->siglock, *flags);
	}
1123
	rcu_read_unlock();
1124 1125 1126 1127

	return sighand;
}

1128 1129 1130 1131
/*
 * send signal info to all the members of a group
 * - the caller must hold the RCU read lock at least
 */
L
Linus Torvalds 已提交
1132 1133
int group_send_sig_info(int sig, struct siginfo *info, struct task_struct *p)
{
1134
	int ret = check_kill_permission(sig, info, p);
1135

1136 1137
	if (!ret && sig)
		ret = do_send_sig_info(sig, info, p, true);
L
Linus Torvalds 已提交
1138 1139 1140 1141 1142

	return ret;
}

/*
1143
 * __kill_pgrp_info() sends a signal to a process group: this is what the tty
L
Linus Torvalds 已提交
1144
 * control characters do (^C, ^Z etc)
1145
 * - the caller must hold at least a readlock on tasklist_lock
L
Linus Torvalds 已提交
1146
 */
1147
int __kill_pgrp_info(int sig, struct siginfo *info, struct pid *pgrp)
L
Linus Torvalds 已提交
1148 1149 1150 1151 1152 1153
{
	struct task_struct *p = NULL;
	int retval, success;

	success = 0;
	retval = -ESRCH;
1154
	do_each_pid_task(pgrp, PIDTYPE_PGID, p) {
L
Linus Torvalds 已提交
1155 1156 1157
		int err = group_send_sig_info(sig, info, p);
		success |= !err;
		retval = err;
1158
	} while_each_pid_task(pgrp, PIDTYPE_PGID, p);
L
Linus Torvalds 已提交
1159 1160 1161
	return success ? 0 : retval;
}

1162
int kill_pid_info(int sig, struct siginfo *info, struct pid *pid)
L
Linus Torvalds 已提交
1163
{
1164
	int error = -ESRCH;
L
Linus Torvalds 已提交
1165 1166
	struct task_struct *p;

I
Ingo Molnar 已提交
1167
	rcu_read_lock();
1168
retry:
1169
	p = pid_task(pid, PIDTYPE_PID);
1170
	if (p) {
L
Linus Torvalds 已提交
1171
		error = group_send_sig_info(sig, info, p);
1172 1173 1174 1175 1176 1177 1178 1179 1180
		if (unlikely(error == -ESRCH))
			/*
			 * 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.
			 */
			goto retry;
	}
I
Ingo Molnar 已提交
1181
	rcu_read_unlock();
1182

L
Linus Torvalds 已提交
1183 1184 1185
	return error;
}

1186 1187
int
kill_proc_info(int sig, struct siginfo *info, pid_t pid)
1188 1189 1190
{
	int error;
	rcu_read_lock();
1191
	error = kill_pid_info(sig, info, find_vpid(pid));
1192 1193 1194 1195
	rcu_read_unlock();
	return error;
}

1196 1197
/* like kill_pid_info(), but doesn't use uid/euid of "current" */
int kill_pid_info_as_uid(int sig, struct siginfo *info, struct pid *pid,
1198
		      uid_t uid, uid_t euid, u32 secid)
1199 1200 1201
{
	int ret = -EINVAL;
	struct task_struct *p;
1202
	const struct cred *pcred;
1203
	unsigned long flags;
1204 1205 1206 1207

	if (!valid_signal(sig))
		return ret;

1208
	rcu_read_lock();
1209
	p = pid_task(pid, PIDTYPE_PID);
1210 1211 1212 1213
	if (!p) {
		ret = -ESRCH;
		goto out_unlock;
	}
1214
	pcred = __task_cred(p);
1215
	if (si_fromuser(info) &&
1216 1217
	    euid != pcred->suid && euid != pcred->uid &&
	    uid  != pcred->suid && uid  != pcred->uid) {
1218 1219 1220
		ret = -EPERM;
		goto out_unlock;
	}
1221 1222 1223
	ret = security_task_kill(p, info, sig, secid);
	if (ret)
		goto out_unlock;
1224 1225 1226 1227 1228 1229 1230

	if (sig) {
		if (lock_task_sighand(p, &flags)) {
			ret = __send_signal(sig, info, p, 1, 0);
			unlock_task_sighand(p, &flags);
		} else
			ret = -ESRCH;
1231 1232
	}
out_unlock:
1233
	rcu_read_unlock();
1234 1235
	return ret;
}
1236
EXPORT_SYMBOL_GPL(kill_pid_info_as_uid);
L
Linus Torvalds 已提交
1237 1238 1239 1240 1241 1242 1243 1244

/*
 * 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.
 */

1245
static int kill_something_info(int sig, struct siginfo *info, pid_t pid)
L
Linus Torvalds 已提交
1246
{
1247
	int ret;
1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260

	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
1261 1262 1263 1264
		int retval = 0, count = 0;
		struct task_struct * p;

		for_each_process(p) {
1265 1266
			if (task_pid_vnr(p) > 1 &&
					!same_thread_group(p, current)) {
L
Linus Torvalds 已提交
1267 1268 1269 1270 1271 1272
				int err = group_send_sig_info(sig, info, p);
				++count;
				if (err != -EPERM)
					retval = err;
			}
		}
1273
		ret = count ? retval : -ESRCH;
L
Linus Torvalds 已提交
1274
	}
1275 1276
	read_unlock(&tasklist_lock);

1277
	return ret;
L
Linus Torvalds 已提交
1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290
}

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

int
send_sig_info(int sig, struct siginfo *info, struct task_struct *p)
{
	/*
	 * Make sure legacy kernel users don't send in bad values
	 * (normal paths check this in check_kill_permission).
	 */
1291
	if (!valid_signal(sig))
L
Linus Torvalds 已提交
1292 1293
		return -EINVAL;

1294
	return do_send_sig_info(sig, info, p, false);
L
Linus Torvalds 已提交
1295 1296
}

1297 1298 1299
#define __si_special(priv) \
	((priv) ? SEND_SIG_PRIV : SEND_SIG_NOINFO)

L
Linus Torvalds 已提交
1300 1301 1302
int
send_sig(int sig, struct task_struct *p, int priv)
{
1303
	return send_sig_info(sig, __si_special(priv), p);
L
Linus Torvalds 已提交
1304 1305 1306 1307 1308
}

void
force_sig(int sig, struct task_struct *p)
{
1309
	force_sig_info(sig, SEND_SIG_PRIV, p);
L
Linus Torvalds 已提交
1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330
}

/*
 * 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;
}

1331 1332
int kill_pgrp(struct pid *pid, int sig, int priv)
{
1333 1334 1335 1336 1337 1338 1339
	int ret;

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

	return ret;
1340 1341 1342 1343 1344 1345 1346 1347 1348
}
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
1349 1350 1351 1352
/*
 * 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
1353
 * expirations or I/O completions".  In the case of Posix Timers
L
Linus Torvalds 已提交
1354 1355 1356 1357 1358 1359
 * 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)
{
1360
	struct sigqueue *q = __sigqueue_alloc(-1, current, GFP_KERNEL, 0);
L
Linus Torvalds 已提交
1361

1362
	if (q)
L
Linus Torvalds 已提交
1363
		q->flags |= SIGQUEUE_PREALLOC;
1364 1365

	return q;
L
Linus Torvalds 已提交
1366 1367 1368 1369 1370
}

void sigqueue_free(struct sigqueue *q)
{
	unsigned long flags;
1371 1372
	spinlock_t *lock = &current->sighand->siglock;

L
Linus Torvalds 已提交
1373 1374
	BUG_ON(!(q->flags & SIGQUEUE_PREALLOC));
	/*
1375 1376
	 * We must hold ->siglock while testing q->list
	 * to serialize with collect_signal() or with
1377
	 * __exit_signal()->flush_sigqueue().
L
Linus Torvalds 已提交
1378
	 */
1379
	spin_lock_irqsave(lock, flags);
1380 1381 1382 1383 1384
	q->flags &= ~SIGQUEUE_PREALLOC;
	/*
	 * If it is queued it will be freed when dequeued,
	 * like the "regular" sigqueue.
	 */
1385
	if (!list_empty(&q->list))
1386
		q = NULL;
1387 1388
	spin_unlock_irqrestore(lock, flags);

1389 1390
	if (q)
		__sigqueue_free(q);
L
Linus Torvalds 已提交
1391 1392
}

1393
int send_sigqueue(struct sigqueue *q, struct task_struct *t, int group)
1394
{
1395
	int sig = q->info.si_signo;
1396
	struct sigpending *pending;
1397 1398
	unsigned long flags;
	int ret;
1399

1400
	BUG_ON(!(q->flags & SIGQUEUE_PREALLOC));
1401 1402 1403 1404 1405

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

1406
	ret = 1; /* the signal is ignored */
1407
	if (!prepare_signal(sig, t, 0))
1408 1409 1410
		goto out;

	ret = 0;
1411 1412 1413 1414 1415 1416 1417
	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++;
1418
		goto out;
1419
	}
1420
	q->info.si_overrun = 0;
1421 1422

	signalfd_notify(t, sig);
1423
	pending = group ? &t->signal->shared_pending : &t->pending;
1424 1425
	list_add_tail(&q->list, &pending->list);
	sigaddset(&pending->signal, sig);
1426
	complete_signal(sig, t, group);
1427 1428 1429 1430
out:
	unlock_task_sighand(t, &flags);
ret:
	return ret;
1431 1432
}

L
Linus Torvalds 已提交
1433 1434 1435
/*
 * 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
1436 1437 1438
 *
 * Returns -1 if our parent ignored us and so we've switched to
 * self-reaping, or else @sig.
L
Linus Torvalds 已提交
1439
 */
R
Roland McGrath 已提交
1440
int do_notify_parent(struct task_struct *tsk, int sig)
L
Linus Torvalds 已提交
1441 1442 1443 1444
{
	struct siginfo info;
	unsigned long flags;
	struct sighand_struct *psig;
R
Roland McGrath 已提交
1445
	int ret = sig;
L
Linus Torvalds 已提交
1446 1447 1448 1449

	BUG_ON(sig == -1);

 	/* do_notify_parent_cldstop should have been called instead.  */
M
Matthew Wilcox 已提交
1450
 	BUG_ON(task_is_stopped_or_traced(tsk));
L
Linus Torvalds 已提交
1451

1452
	BUG_ON(!task_ptrace(tsk) &&
L
Linus Torvalds 已提交
1453 1454 1455 1456
	       (tsk->group_leader != tsk || !thread_group_empty(tsk)));

	info.si_signo = sig;
	info.si_errno = 0;
1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470
	/*
	 * we are under tasklist_lock here so our parent is tied to
	 * us and cannot exit and release its namespace.
	 *
	 * the only it can is to switch its nsproxy with sys_unshare,
	 * bu uncharing pid namespaces is not allowed, so we'll always
	 * see relevant namespace
	 *
	 * 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();
	info.si_pid = task_pid_nr_ns(tsk, tsk->parent->nsproxy->pid_ns);
1471
	info.si_uid = __task_cred(tsk)->uid;
1472 1473
	rcu_read_unlock();

1474 1475 1476 1477
	info.si_utime = cputime_to_clock_t(cputime_add(tsk->utime,
				tsk->signal->utime));
	info.si_stime = cputime_to_clock_t(cputime_add(tsk->stime,
				tsk->signal->stime));
L
Linus Torvalds 已提交
1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490

	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);
1491
	if (!task_ptrace(tsk) && sig == SIGCHLD &&
L
Linus Torvalds 已提交
1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508
	    (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).
		 */
R
Roland McGrath 已提交
1509
		ret = tsk->exit_signal = -1;
L
Linus Torvalds 已提交
1510
		if (psig->action[SIGCHLD-1].sa.sa_handler == SIG_IGN)
R
Roland McGrath 已提交
1511
			sig = -1;
L
Linus Torvalds 已提交
1512
	}
1513
	if (valid_signal(sig) && sig > 0)
L
Linus Torvalds 已提交
1514 1515 1516
		__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
1517

R
Roland McGrath 已提交
1518
	return ret;
L
Linus Torvalds 已提交
1519 1520
}

1521
static void do_notify_parent_cldstop(struct task_struct *tsk, int why)
L
Linus Torvalds 已提交
1522 1523 1524
{
	struct siginfo info;
	unsigned long flags;
1525
	struct task_struct *parent;
L
Linus Torvalds 已提交
1526 1527
	struct sighand_struct *sighand;

1528
	if (task_ptrace(tsk))
1529 1530 1531 1532 1533 1534
		parent = tsk->parent;
	else {
		tsk = tsk->group_leader;
		parent = tsk->real_parent;
	}

L
Linus Torvalds 已提交
1535 1536
	info.si_signo = SIGCHLD;
	info.si_errno = 0;
1537 1538 1539 1540
	/*
	 * see comment in do_notify_parent() abot the following 3 lines
	 */
	rcu_read_lock();
1541
	info.si_pid = task_pid_nr_ns(tsk, parent->nsproxy->pid_ns);
1542
	info.si_uid = __task_cred(tsk)->uid;
1543 1544
	rcu_read_unlock();

1545 1546
	info.si_utime = cputime_to_clock_t(tsk->utime);
	info.si_stime = cputime_to_clock_t(tsk->stime);
L
Linus Torvalds 已提交
1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574

 	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);
}

1575 1576
static inline int may_ptrace_stop(void)
{
1577
	if (!likely(task_ptrace(current)))
1578 1579 1580 1581 1582 1583 1584
		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
1585
	 * ->siglock we must see ->core_state != NULL. Otherwise it
1586 1587
	 * is safe to enter schedule().
	 */
1588
	if (unlikely(current->mm->core_state) &&
1589 1590 1591 1592 1593 1594
	    unlikely(current->mm == current->parent->mm))
		return 0;

	return 1;
}

R
Roland McGrath 已提交
1595 1596 1597 1598 1599 1600
/*
 * Return nonzero if there is a SIGKILL that should be waking us up.
 * Called with the siglock held.
 */
static int sigkill_pending(struct task_struct *tsk)
{
1601 1602
	return	sigismember(&tsk->pending.signal, SIGKILL) ||
		sigismember(&tsk->signal->shared_pending.signal, SIGKILL);
R
Roland McGrath 已提交
1603 1604
}

L
Linus Torvalds 已提交
1605 1606 1607 1608 1609 1610 1611 1612
/*
 * 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.
 *
1613 1614
 * 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
1615
 */
1616
static void ptrace_stop(int exit_code, int clear_code, siginfo_t *info)
L
Linus Torvalds 已提交
1617
{
R
Roland McGrath 已提交
1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632
	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);
1633 1634
		if (sigkill_pending(current))
			return;
R
Roland McGrath 已提交
1635 1636
	}

L
Linus Torvalds 已提交
1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647
	/*
	 * If there is a group stop in progress,
	 * we must participate in the bookkeeping.
	 */
	if (current->signal->group_stop_count > 0)
		--current->signal->group_stop_count;

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

	/* Let the debugger run.  */
1648
	__set_current_state(TASK_TRACED);
L
Linus Torvalds 已提交
1649 1650
	spin_unlock_irq(&current->sighand->siglock);
	read_lock(&tasklist_lock);
1651
	if (may_ptrace_stop()) {
1652
		do_notify_parent_cldstop(current, CLD_TRAPPED);
M
Miklos Szeredi 已提交
1653 1654 1655 1656 1657 1658 1659
		/*
		 * 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
1660
		read_unlock(&tasklist_lock);
M
Miklos Szeredi 已提交
1661
		preempt_enable_no_resched();
L
Linus Torvalds 已提交
1662 1663 1664 1665
		schedule();
	} else {
		/*
		 * By the time we got the lock, our tracer went away.
1666
		 * Don't drop the lock yet, another tracer may come.
L
Linus Torvalds 已提交
1667
		 */
1668
		__set_current_state(TASK_RUNNING);
1669 1670
		if (clear_code)
			current->exit_code = 0;
1671
		read_unlock(&tasklist_lock);
L
Linus Torvalds 已提交
1672 1673
	}

R
Roland McGrath 已提交
1674 1675 1676 1677 1678 1679 1680
	/*
	 * While in TASK_TRACED, we were considered "frozen enough".
	 * Now that we woke up, it's crucial if we're supposed to be
	 * frozen that we freeze now before running anything substantial.
	 */
	try_to_freeze();

L
Linus Torvalds 已提交
1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691
	/*
	 * 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;

	/*
	 * Queued signals ignored us while we were stopped for tracing.
	 * So check for any that we should take before resuming user mode.
1692
	 * This sets TIF_SIGPENDING, but never clears it.
L
Linus Torvalds 已提交
1693
	 */
1694
	recalc_sigpending_tsk(current);
L
Linus Torvalds 已提交
1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705
}

void ptrace_notify(int exit_code)
{
	siginfo_t info;

	BUG_ON((exit_code & (0x7f | ~0xffff)) != SIGTRAP);

	memset(&info, 0, sizeof info);
	info.si_signo = SIGTRAP;
	info.si_code = exit_code;
1706
	info.si_pid = task_pid_vnr(current);
1707
	info.si_uid = current_uid();
L
Linus Torvalds 已提交
1708 1709 1710

	/* Let the debugger run.  */
	spin_lock_irq(&current->sighand->siglock);
1711
	ptrace_stop(exit_code, 1, &info);
L
Linus Torvalds 已提交
1712 1713 1714 1715 1716 1717 1718 1719 1720
	spin_unlock_irq(&current->sighand->siglock);
}

/*
 * This performs the stopping for SIGSTOP and other stop signals.
 * We have to stop all threads in the thread group.
 * Returns nonzero if we've actually stopped and released the siglock.
 * Returns zero if we didn't stop and still hold the siglock.
 */
1721
static int do_signal_stop(int signr)
L
Linus Torvalds 已提交
1722 1723
{
	struct signal_struct *sig = current->signal;
1724
	int notify;
L
Linus Torvalds 已提交
1725

1726
	if (!sig->group_stop_count) {
1727 1728
		struct task_struct *t;

1729
		if (!likely(sig->flags & SIGNAL_STOP_DEQUEUED) ||
1730
		    unlikely(signal_group_exit(sig)))
1731
			return 0;
L
Linus Torvalds 已提交
1732 1733
		/*
		 * There is no group stop already in progress.
1734
		 * We must initiate one now.
L
Linus Torvalds 已提交
1735
		 */
1736
		sig->group_exit_code = signr;
L
Linus Torvalds 已提交
1737

1738
		sig->group_stop_count = 1;
1739
		for (t = next_thread(current); t != current; t = next_thread(t))
L
Linus Torvalds 已提交
1740
			/*
1741 1742 1743
			 * 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
1744
			 */
O
Oleg Nesterov 已提交
1745
			if (!(t->flags & PF_EXITING) &&
M
Matthew Wilcox 已提交
1746
			    !task_is_stopped_or_traced(t)) {
1747
				sig->group_stop_count++;
1748 1749
				signal_wake_up(t, 0);
			}
L
Linus Torvalds 已提交
1750
	}
1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769
	/*
	 * 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.  When ptraced, every thread reports itself.
	 */
	notify = sig->group_stop_count == 1 ? CLD_STOPPED : 0;
	notify = tracehook_notify_jctl(notify, CLD_STOPPED);
	/*
	 * tracehook_notify_jctl() can drop and reacquire siglock, so
	 * we keep ->group_stop_count != 0 before the call. If SIGCONT
	 * or SIGKILL comes in between ->group_stop_count == 0.
	 */
	if (sig->group_stop_count) {
		if (!--sig->group_stop_count)
			sig->flags = SIGNAL_STOP_STOPPED;
		current->exit_code = sig->group_exit_code;
		__set_current_state(TASK_STOPPED);
	}
	spin_unlock_irq(&current->sighand->siglock);
L
Linus Torvalds 已提交
1770

1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783
	if (notify) {
		read_lock(&tasklist_lock);
		do_notify_parent_cldstop(current, notify);
		read_unlock(&tasklist_lock);
	}

	/* Now we don't run again until woken by SIGCONT or SIGKILL */
	do {
		schedule();
	} while (try_to_freeze());

	tracehook_finish_jctl();
	current->exit_code = 0;
1784

L
Linus Torvalds 已提交
1785 1786 1787
	return 1;
}

R
Roland McGrath 已提交
1788 1789 1790
static int ptrace_signal(int signr, siginfo_t *info,
			 struct pt_regs *regs, void *cookie)
{
1791
	if (!task_ptrace(current))
R
Roland McGrath 已提交
1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814
		return signr;

	ptrace_signal_deliver(regs, cookie);

	/* Let the debugger run.  */
	ptrace_stop(signr, 0, info);

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

	current->exit_code = 0;

	/* 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.  */
	if (signr != info->si_signo) {
		info->si_signo = signr;
		info->si_errno = 0;
		info->si_code = SI_USER;
		info->si_pid = task_pid_vnr(current->parent);
1815
		info->si_uid = task_uid(current->parent);
R
Roland McGrath 已提交
1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826
	}

	/* 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;
}

L
Linus Torvalds 已提交
1827 1828 1829
int get_signal_to_deliver(siginfo_t *info, struct k_sigaction *return_ka,
			  struct pt_regs *regs, void *cookie)
{
1830 1831 1832
	struct sighand_struct *sighand = current->sighand;
	struct signal_struct *signal = current->signal;
	int signr;
L
Linus Torvalds 已提交
1833

R
Roland McGrath 已提交
1834 1835 1836 1837 1838 1839 1840
relock:
	/*
	 * We'll jump back here after any time we were stopped in TASK_STOPPED.
	 * While in TASK_STOPPED, we were considered "frozen enough".
	 * Now that we woke up, it's crucial if we're supposed to be
	 * frozen that we freeze now before running anything substantial.
	 */
1841 1842
	try_to_freeze();

1843
	spin_lock_irq(&sighand->siglock);
1844 1845 1846 1847 1848
	/*
	 * 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.
	 */
1849 1850
	if (unlikely(signal->flags & SIGNAL_CLD_MASK)) {
		int why = (signal->flags & SIGNAL_STOP_CONTINUED)
1851
				? CLD_CONTINUED : CLD_STOPPED;
1852
		signal->flags &= ~SIGNAL_CLD_MASK;
1853

1854 1855
		why = tracehook_notify_jctl(why, CLD_CONTINUED);
		spin_unlock_irq(&sighand->siglock);
R
Roland McGrath 已提交
1856

1857 1858 1859 1860 1861
		if (why) {
			read_lock(&tasklist_lock);
			do_notify_parent_cldstop(current->group_leader, why);
			read_unlock(&tasklist_lock);
		}
1862 1863 1864
		goto relock;
	}

L
Linus Torvalds 已提交
1865 1866
	for (;;) {
		struct k_sigaction *ka;
1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877
		/*
		 * Tracing can induce an artifical signal and choose sigaction.
		 * The return value in @signr determines the default action,
		 * but @info->si_signo is the signal number we will report.
		 */
		signr = tracehook_get_signal(current, regs, info, return_ka);
		if (unlikely(signr < 0))
			goto relock;
		if (unlikely(signr != 0))
			ka = return_ka;
		else {
1878 1879 1880 1881
			if (unlikely(signal->group_stop_count > 0) &&
			    do_signal_stop(0))
				goto relock;

1882 1883
			signr = dequeue_signal(current, &current->blocked,
					       info);
L
Linus Torvalds 已提交
1884

R
Roland McGrath 已提交
1885
			if (!signr)
1886 1887 1888 1889 1890 1891 1892 1893 1894 1895
				break; /* will return 0 */

			if (signr != SIGKILL) {
				signr = ptrace_signal(signr, info,
						      regs, cookie);
				if (!signr)
					continue;
			}

			ka = &sighand->action[signr-1];
L
Linus Torvalds 已提交
1896 1897
		}

1898 1899 1900
		/* Trace actually delivered signals. */
		trace_signal_deliver(signr, info, ka);

L
Linus Torvalds 已提交
1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918
		if (ka->sa.sa_handler == SIG_IGN) /* Do nothing.  */
			continue;
		if (ka->sa.sa_handler != SIG_DFL) {
			/* Run the handler.  */
			*return_ka = *ka;

			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;

1919
		/*
1920
		 * Global init gets no signals it doesn't want.
1921 1922 1923 1924 1925 1926 1927
		 * 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.
1928
		 */
1929
		if (unlikely(signal->flags & SIGNAL_UNKILLABLE) &&
1930
				!sig_kernel_only(signr))
L
Linus Torvalds 已提交
1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944
			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) {
1945
				spin_unlock_irq(&sighand->siglock);
L
Linus Torvalds 已提交
1946 1947 1948

				/* signals can be posted during this window */

1949
				if (is_current_pgrp_orphaned())
L
Linus Torvalds 已提交
1950 1951
					goto relock;

1952
				spin_lock_irq(&sighand->siglock);
L
Linus Torvalds 已提交
1953 1954
			}

1955
			if (likely(do_signal_stop(info->si_signo))) {
L
Linus Torvalds 已提交
1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966
				/* It released the siglock.  */
				goto relock;
			}

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

1967
		spin_unlock_irq(&sighand->siglock);
L
Linus Torvalds 已提交
1968 1969 1970 1971 1972

		/*
		 * Anything else is fatal, maybe with a core dump.
		 */
		current->flags |= PF_SIGNALED;
1973

L
Linus Torvalds 已提交
1974
		if (sig_kernel_coredump(signr)) {
1975
			if (print_fatal_signals)
1976
				print_fatal_signal(regs, info->si_signo);
L
Linus Torvalds 已提交
1977 1978 1979 1980 1981 1982 1983 1984
			/*
			 * 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.
			 */
1985
			do_coredump(info->si_signo, info->si_signo, regs);
L
Linus Torvalds 已提交
1986 1987 1988 1989 1990
		}

		/*
		 * Death signals, no core dump.
		 */
1991
		do_group_exit(info->si_signo);
L
Linus Torvalds 已提交
1992 1993
		/* NOTREACHED */
	}
1994
	spin_unlock_irq(&sighand->siglock);
L
Linus Torvalds 已提交
1995 1996 1997
	return signr;
}

O
Oleg Nesterov 已提交
1998 1999 2000
void exit_signals(struct task_struct *tsk)
{
	int group_stop = 0;
2001
	struct task_struct *t;
O
Oleg Nesterov 已提交
2002

2003 2004 2005
	if (thread_group_empty(tsk) || signal_group_exit(tsk->signal)) {
		tsk->flags |= PF_EXITING;
		return;
O
Oleg Nesterov 已提交
2006 2007
	}

2008
	spin_lock_irq(&tsk->sighand->siglock);
O
Oleg Nesterov 已提交
2009 2010 2011 2012 2013
	/*
	 * From now this task is not visible for group-wide signals,
	 * see wants_signal(), do_signal_stop().
	 */
	tsk->flags |= PF_EXITING;
2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027
	if (!signal_pending(tsk))
		goto out;

	/* It could be that __group_complete_signal() choose us to
	 * notify about group-wide signal. Another thread should be
	 * woken now to take the signal since we will not.
	 */
	for (t = tsk; (t = next_thread(t)) != tsk; )
		if (!signal_pending(t) && !(t->flags & PF_EXITING))
			recalc_sigpending_and_wake(t);

	if (unlikely(tsk->signal->group_stop_count) &&
			!--tsk->signal->group_stop_count) {
		tsk->signal->flags = SIGNAL_STOP_STOPPED;
2028
		group_stop = tracehook_notify_jctl(CLD_STOPPED, CLD_STOPPED);
2029 2030
	}
out:
O
Oleg Nesterov 已提交
2031 2032
	spin_unlock_irq(&tsk->sighand->siglock);

2033
	if (unlikely(group_stop)) {
O
Oleg Nesterov 已提交
2034
		read_lock(&tasklist_lock);
2035
		do_notify_parent_cldstop(tsk, group_stop);
O
Oleg Nesterov 已提交
2036 2037 2038 2039
		read_unlock(&tasklist_lock);
	}
}

L
Linus Torvalds 已提交
2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054
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);
EXPORT_SYMBOL(block_all_signals);
EXPORT_SYMBOL(unblock_all_signals);


/*
 * System call entry points.
 */

2055
SYSCALL_DEFINE0(restart_syscall)
L
Linus Torvalds 已提交
2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084
{
	struct restart_block *restart = &current_thread_info()->restart_block;
	return restart->fn(restart);
}

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

/*
 * We don't need to get the kernel lock - this is all local to this
 * particular thread.. (and that's good, because this is _heavily_
 * used by various programs)
 */

/*
 * 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)
{
	int error;

	spin_lock_irq(&current->sighand->siglock);
2085 2086 2087
	if (oldset)
		*oldset = current->blocked;

L
Linus Torvalds 已提交
2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103
	error = 0;
	switch (how) {
	case SIG_BLOCK:
		sigorsets(&current->blocked, &current->blocked, set);
		break;
	case SIG_UNBLOCK:
		signandsets(&current->blocked, &current->blocked, set);
		break;
	case SIG_SETMASK:
		current->blocked = *set;
		break;
	default:
		error = -EINVAL;
	}
	recalc_sigpending();
	spin_unlock_irq(&current->sighand->siglock);
2104

L
Linus Torvalds 已提交
2105 2106 2107
	return error;
}

2108 2109
SYSCALL_DEFINE4(rt_sigprocmask, int, how, sigset_t __user *, set,
		sigset_t __user *, oset, size_t, sigsetsize)
L
Linus Torvalds 已提交
2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167
{
	int error = -EINVAL;
	sigset_t old_set, new_set;

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

	if (set) {
		error = -EFAULT;
		if (copy_from_user(&new_set, set, sizeof(*set)))
			goto out;
		sigdelsetmask(&new_set, sigmask(SIGKILL)|sigmask(SIGSTOP));

		error = sigprocmask(how, &new_set, &old_set);
		if (error)
			goto out;
		if (oset)
			goto set_old;
	} else if (oset) {
		spin_lock_irq(&current->sighand->siglock);
		old_set = current->blocked;
		spin_unlock_irq(&current->sighand->siglock);

	set_old:
		error = -EFAULT;
		if (copy_to_user(oset, &old_set, sizeof(*oset)))
			goto out;
	}
	error = 0;
out:
	return error;
}

long do_sigpending(void __user *set, unsigned long sigsetsize)
{
	long error = -EINVAL;
	sigset_t pending;

	if (sigsetsize > sizeof(sigset_t))
		goto out;

	spin_lock_irq(&current->sighand->siglock);
	sigorsets(&pending, &current->pending.signal,
		  &current->signal->shared_pending.signal);
	spin_unlock_irq(&current->sighand->siglock);

	/* Outside the lock because only this thread touches it.  */
	sigandsets(&pending, &current->blocked, &pending);

	error = -EFAULT;
	if (!copy_to_user(set, &pending, sigsetsize))
		error = 0;

out:
	return error;
}	

2168
SYSCALL_DEFINE2(rt_sigpending, sigset_t __user *, set, size_t, sigsetsize)
L
Linus Torvalds 已提交
2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186
{
	return do_sigpending(set, sigsetsize);
}

#ifndef HAVE_ARCH_COPY_SIGINFO_TO_USER

int copy_siginfo_to_user(siginfo_t __user *to, siginfo_t *from)
{
	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.
2187 2188
	 * Please remember to update the signalfd_copyinfo() function
	 * inside fs/signalfd.c too, in case siginfo_t changes.
L
Linus Torvalds 已提交
2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238
	 * 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);
#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;
	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

2239 2240 2241
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
2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287
{
	int ret, sig;
	sigset_t these;
	struct timespec ts;
	siginfo_t info;
	long timeout = 0;

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

	if (uts) {
		if (copy_from_user(&ts, uts, sizeof(ts)))
			return -EFAULT;
		if (ts.tv_nsec >= 1000000000L || ts.tv_nsec < 0
		    || ts.tv_sec < 0)
			return -EINVAL;
	}

	spin_lock_irq(&current->sighand->siglock);
	sig = dequeue_signal(current, &these, &info);
	if (!sig) {
		timeout = MAX_SCHEDULE_TIMEOUT;
		if (uts)
			timeout = (timespec_to_jiffies(&ts)
				   + (ts.tv_sec || ts.tv_nsec));

		if (timeout) {
			/* None ready -- temporarily unblock those we're
			 * interested while we are sleeping in so that we'll
			 * be awakened when they arrive.  */
			current->real_blocked = current->blocked;
			sigandsets(&current->blocked, &current->blocked, &these);
			recalc_sigpending();
			spin_unlock_irq(&current->sighand->siglock);

2288
			timeout = schedule_timeout_interruptible(timeout);
L
Linus Torvalds 已提交
2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313

			spin_lock_irq(&current->sighand->siglock);
			sig = dequeue_signal(current, &these, &info);
			current->blocked = current->real_blocked;
			siginitset(&current->real_blocked, 0);
			recalc_sigpending();
		}
	}
	spin_unlock_irq(&current->sighand->siglock);

	if (sig) {
		ret = sig;
		if (uinfo) {
			if (copy_siginfo_to_user(uinfo, &info))
				ret = -EFAULT;
		}
	} else {
		ret = -EAGAIN;
		if (timeout)
			ret = -EINTR;
	}

	return ret;
}

2314
SYSCALL_DEFINE2(kill, pid_t, pid, int, sig)
L
Linus Torvalds 已提交
2315 2316 2317 2318 2319 2320
{
	struct siginfo info;

	info.si_signo = sig;
	info.si_errno = 0;
	info.si_code = SI_USER;
2321
	info.si_pid = task_tgid_vnr(current);
2322
	info.si_uid = current_uid();
L
Linus Torvalds 已提交
2323 2324 2325 2326

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

T
Thomas Gleixner 已提交
2327 2328
static int
do_send_specific(pid_t tgid, pid_t pid, int sig, struct siginfo *info)
L
Linus Torvalds 已提交
2329 2330
{
	struct task_struct *p;
T
Thomas Gleixner 已提交
2331
	int error = -ESRCH;
L
Linus Torvalds 已提交
2332

2333
	rcu_read_lock();
2334
	p = find_task_by_vpid(pid);
2335
	if (p && (tgid <= 0 || task_tgid_vnr(p) == tgid)) {
T
Thomas Gleixner 已提交
2336
		error = check_kill_permission(sig, info, p);
L
Linus Torvalds 已提交
2337 2338 2339 2340
		/*
		 * The null signal is a permissions and process existence
		 * probe.  No signal is actually delivered.
		 */
2341 2342 2343 2344 2345 2346 2347 2348 2349
		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
2350 2351
		}
	}
2352
	rcu_read_unlock();
2353

L
Linus Torvalds 已提交
2354 2355 2356
	return error;
}

T
Thomas Gleixner 已提交
2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369
static int do_tkill(pid_t tgid, pid_t pid, int sig)
{
	struct siginfo info;

	info.si_signo = sig;
	info.si_errno = 0;
	info.si_code = SI_TKILL;
	info.si_pid = task_tgid_vnr(current);
	info.si_uid = current_uid();

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

2370 2371 2372 2373 2374 2375
/**
 *  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
 *
2376
 *  This syscall also checks the @tgid and returns -ESRCH even if the PID
2377 2378 2379
 *  exists but it's not belonging to the target process anymore. This
 *  method solves the problem of threads exiting and PIDs getting reused.
 */
2380
SYSCALL_DEFINE3(tgkill, pid_t, tgid, pid_t, pid, int, sig)
2381 2382 2383 2384 2385 2386 2387 2388
{
	/* This is only valid for single tasks */
	if (pid <= 0 || tgid <= 0)
		return -EINVAL;

	return do_tkill(tgid, pid, sig);
}

L
Linus Torvalds 已提交
2389 2390 2391
/*
 *  Send a signal to only one task, even if it's a CLONE_THREAD task.
 */
2392
SYSCALL_DEFINE2(tkill, pid_t, pid, int, sig)
L
Linus Torvalds 已提交
2393 2394 2395 2396 2397
{
	/* This is only valid for single tasks */
	if (pid <= 0)
		return -EINVAL;

2398
	return do_tkill(0, pid, sig);
L
Linus Torvalds 已提交
2399 2400
}

2401 2402
SYSCALL_DEFINE3(rt_sigqueueinfo, pid_t, pid, int, sig,
		siginfo_t __user *, uinfo)
L
Linus Torvalds 已提交
2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418
{
	siginfo_t info;

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

	/* Not even root can pretend to send signals from the kernel.
	   Nor can they impersonate a kill(), which adds source info.  */
	if (info.si_code >= 0)
		return -EPERM;
	info.si_signo = sig;

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

2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444
long do_rt_tgsigqueueinfo(pid_t tgid, pid_t pid, int sig, siginfo_t *info)
{
	/* 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.
	   Nor can they impersonate a kill(), which adds source info.  */
	if (info->si_code >= 0)
		return -EPERM;
	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);
}

2445
int do_sigaction(int sig, struct k_sigaction *act, struct k_sigaction *oact)
L
Linus Torvalds 已提交
2446
{
2447
	struct task_struct *t = current;
L
Linus Torvalds 已提交
2448
	struct k_sigaction *k;
2449
	sigset_t mask;
L
Linus Torvalds 已提交
2450

2451
	if (!valid_signal(sig) || sig < 1 || (act && sig_kernel_only(sig)))
L
Linus Torvalds 已提交
2452 2453
		return -EINVAL;

2454
	k = &t->sighand->action[sig-1];
L
Linus Torvalds 已提交
2455 2456 2457 2458 2459 2460

	spin_lock_irq(&current->sighand->siglock);
	if (oact)
		*oact = *k;

	if (act) {
2461 2462
		sigdelsetmask(&act->sa.sa_mask,
			      sigmask(SIGKILL) | sigmask(SIGSTOP));
2463
		*k = *act;
L
Linus Torvalds 已提交
2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474
		/*
		 * 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"
		 */
2475
		if (sig_handler_ignored(sig_handler(t, sig), sig)) {
2476 2477 2478
			sigemptyset(&mask);
			sigaddset(&mask, sig);
			rm_from_queue_full(&mask, &t->signal->shared_pending);
L
Linus Torvalds 已提交
2479
			do {
2480
				rm_from_queue_full(&mask, &t->pending);
L
Linus Torvalds 已提交
2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495
				t = next_thread(t);
			} while (t != current);
		}
	}

	spin_unlock_irq(&current->sighand->siglock);
	return 0;
}

int 
do_sigaltstack (const stack_t __user *uss, stack_t __user *uoss, unsigned long sp)
{
	stack_t oss;
	int error;

2496 2497 2498
	oss.ss_sp = (void __user *) current->sas_ss_sp;
	oss.ss_size = current->sas_ss_size;
	oss.ss_flags = sas_ss_flags(sp);
L
Linus Torvalds 已提交
2499 2500 2501 2502 2503 2504 2505

	if (uss) {
		void __user *ss_sp;
		size_t ss_size;
		int ss_flags;

		error = -EFAULT;
L
Linus Torvalds 已提交
2506 2507 2508 2509 2510 2511
		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
2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542
			goto out;

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

		error = -EINVAL;
		/*
		 *
		 * Note - this code used to test ss_flags incorrectly
		 *  	  old code may have been written using ss_flags==0
		 *	  to mean ss_flags==SS_ONSTACK (as this was the only
		 *	  way that worked) - this fix preserves that older
		 *	  mechanism
		 */
		if (ss_flags != SS_DISABLE && ss_flags != SS_ONSTACK && ss_flags != 0)
			goto out;

		if (ss_flags == SS_DISABLE) {
			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;
	}

2543
	error = 0;
L
Linus Torvalds 已提交
2544 2545
	if (uoss) {
		error = -EFAULT;
2546
		if (!access_ok(VERIFY_WRITE, uoss, sizeof(*uoss)))
L
Linus Torvalds 已提交
2547
			goto out;
2548 2549 2550
		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
2551 2552 2553 2554 2555 2556 2557 2558
	}

out:
	return error;
}

#ifdef __ARCH_WANT_SYS_SIGPENDING

2559
SYSCALL_DEFINE1(sigpending, old_sigset_t __user *, set)
L
Linus Torvalds 已提交
2560 2561 2562 2563 2564 2565 2566 2567 2568 2569
{
	return do_sigpending(set, sizeof(*set));
}

#endif

#ifdef __ARCH_WANT_SYS_SIGPROCMASK
/* Some platforms have their own version with special arguments others
   support only sys_rt_sigprocmask.  */

2570 2571
SYSCALL_DEFINE3(sigprocmask, int, how, old_sigset_t __user *, set,
		old_sigset_t __user *, oset)
L
Linus Torvalds 已提交
2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620
{
	int error;
	old_sigset_t old_set, new_set;

	if (set) {
		error = -EFAULT;
		if (copy_from_user(&new_set, set, sizeof(*set)))
			goto out;
		new_set &= ~(sigmask(SIGKILL) | sigmask(SIGSTOP));

		spin_lock_irq(&current->sighand->siglock);
		old_set = current->blocked.sig[0];

		error = 0;
		switch (how) {
		default:
			error = -EINVAL;
			break;
		case SIG_BLOCK:
			sigaddsetmask(&current->blocked, new_set);
			break;
		case SIG_UNBLOCK:
			sigdelsetmask(&current->blocked, new_set);
			break;
		case SIG_SETMASK:
			current->blocked.sig[0] = new_set;
			break;
		}

		recalc_sigpending();
		spin_unlock_irq(&current->sighand->siglock);
		if (error)
			goto out;
		if (oset)
			goto set_old;
	} else if (oset) {
		old_set = current->blocked.sig[0];
	set_old:
		error = -EFAULT;
		if (copy_to_user(oset, &old_set, sizeof(*oset)))
			goto out;
	}
	error = 0;
out:
	return error;
}
#endif /* __ARCH_WANT_SYS_SIGPROCMASK */

#ifdef __ARCH_WANT_SYS_RT_SIGACTION
2621 2622 2623 2624
SYSCALL_DEFINE4(rt_sigaction, int, sig,
		const struct sigaction __user *, act,
		struct sigaction __user *, oact,
		size_t, sigsetsize)
L
Linus Torvalds 已提交
2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653
{
	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;
}
#endif /* __ARCH_WANT_SYS_RT_SIGACTION */

#ifdef __ARCH_WANT_SYS_SGETMASK

/*
 * For backwards compatibility.  Functionality superseded by sigprocmask.
 */
2654
SYSCALL_DEFINE0(sgetmask)
L
Linus Torvalds 已提交
2655 2656 2657 2658 2659
{
	/* SMP safe */
	return current->blocked.sig[0];
}

2660
SYSCALL_DEFINE1(ssetmask, int, newmask)
L
Linus Torvalds 已提交
2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679
{
	int old;

	spin_lock_irq(&current->sighand->siglock);
	old = current->blocked.sig[0];

	siginitset(&current->blocked, newmask & ~(sigmask(SIGKILL)|
						  sigmask(SIGSTOP)));
	recalc_sigpending();
	spin_unlock_irq(&current->sighand->siglock);

	return old;
}
#endif /* __ARCH_WANT_SGETMASK */

#ifdef __ARCH_WANT_SYS_SIGNAL
/*
 * For backwards compatibility.  Functionality superseded by sigaction.
 */
2680
SYSCALL_DEFINE2(signal, int, sig, __sighandler_t, handler)
L
Linus Torvalds 已提交
2681 2682 2683 2684 2685 2686
{
	struct k_sigaction new_sa, old_sa;
	int ret;

	new_sa.sa.sa_handler = handler;
	new_sa.sa.sa_flags = SA_ONESHOT | SA_NOMASK;
2687
	sigemptyset(&new_sa.sa.sa_mask);
L
Linus Torvalds 已提交
2688 2689 2690 2691 2692 2693 2694 2695 2696

	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

2697
SYSCALL_DEFINE0(pause)
L
Linus Torvalds 已提交
2698 2699 2700 2701 2702 2703 2704 2705
{
	current->state = TASK_INTERRUPTIBLE;
	schedule();
	return -ERESTARTNOHAND;
}

#endif

2706
#ifdef __ARCH_WANT_SYS_RT_SIGSUSPEND
2707
SYSCALL_DEFINE2(rt_sigsuspend, sigset_t __user *, unewset, size_t, sigsetsize)
2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722 2723 2724 2725 2726
{
	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;
	sigdelsetmask(&newset, sigmask(SIGKILL)|sigmask(SIGSTOP));

	spin_lock_irq(&current->sighand->siglock);
	current->saved_sigmask = current->blocked;
	current->blocked = newset;
	recalc_sigpending();
	spin_unlock_irq(&current->sighand->siglock);

	current->state = TASK_INTERRUPTIBLE;
	schedule();
2727
	set_restore_sigmask();
2728 2729 2730 2731
	return -ERESTARTNOHAND;
}
#endif /* __ARCH_WANT_SYS_RT_SIGSUSPEND */

2732 2733 2734 2735 2736
__attribute__((weak)) const char *arch_vma_name(struct vm_area_struct *vma)
{
	return NULL;
}

L
Linus Torvalds 已提交
2737 2738
void __init signals_init(void)
{
2739
	sigqueue_cachep = KMEM_CACHE(sigqueue, SLAB_PANIC);
L
Linus Torvalds 已提交
2740
}
2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780

#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 */