signal.c 104.3 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
#include <linux/livepatch.h>
44

45 46
#define CREATE_TRACE_POINTS
#include <trace/events/signal.h>
47

L
Linus Torvalds 已提交
48
#include <asm/param.h>
49
#include <linux/uaccess.h>
L
Linus Torvalds 已提交
50 51
#include <asm/unistd.h>
#include <asm/siginfo.h>
52
#include <asm/cacheflush.h>
53
#include "audit.h"	/* audit_signal_info() */
L
Linus Torvalds 已提交
54 55 56 57 58

/*
 * SLAB caches for signal bits.
 */

59
static struct kmem_cache *sigqueue_cachep;
L
Linus Torvalds 已提交
60

61 62
int print_fatal_signals __read_mostly;

63
static void __user *sig_handler(struct task_struct *t, int sig)
64
{
65 66
	return t->sighand->action[sig - 1].sa.sa_handler;
}
67

68
static inline bool sig_handler_ignored(void __user *handler, int sig)
69
{
70 71
	/* Is it explicitly or implicitly ignored? */
	return handler == SIG_IGN ||
72
	       (handler == SIG_DFL && sig_kernel_ignore(sig));
73
}
L
Linus Torvalds 已提交
74

75
static bool sig_task_ignored(struct task_struct *t, int sig, bool force)
L
Linus Torvalds 已提交
76
{
77
	void __user *handler;
L
Linus Torvalds 已提交
78

79 80 81
	handler = sig_handler(t, sig);

	if (unlikely(t->signal->flags & SIGNAL_UNKILLABLE) &&
82
	    handler == SIG_DFL && !(force && sig_kernel_only(sig)))
83
		return true;
84 85 86 87

	return sig_handler_ignored(handler, sig);
}

88
static bool sig_ignored(struct task_struct *t, int sig, bool force)
89
{
L
Linus Torvalds 已提交
90 91 92 93 94
	/*
	 * Blocked signals are never ignored, since the
	 * signal handler may change by the time it is
	 * unblocked.
	 */
95
	if (sigismember(&t->blocked, sig) || sigismember(&t->real_blocked, sig))
96
		return false;
L
Linus Torvalds 已提交
97

98
	/*
99 100 101
	 * Tracers may want to know about even ignored signal unless it
	 * is SIGKILL which can't be reported anyway but can be ignored
	 * by SIGNAL_UNKILLABLE task.
102
	 */
103
	if (t->ptrace && sig != SIGKILL)
104
		return false;
105 106

	return sig_task_ignored(t, sig, force);
L
Linus Torvalds 已提交
107 108 109 110 111 112
}

/*
 * Re-calculate pending state from the set of locally pending
 * signals, globally pending signals, and blocked signals.
 */
113
static inline bool has_pending_signals(sigset_t *signal, sigset_t *blocked)
L
Linus Torvalds 已提交
114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140
{
	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))

141
static bool recalc_sigpending_tsk(struct task_struct *t)
L
Linus Torvalds 已提交
142
{
143
	if ((t->jobctl & JOBCTL_PENDING_MASK) ||
L
Linus Torvalds 已提交
144
	    PENDING(&t->pending, &t->blocked) ||
R
Roland McGrath 已提交
145
	    PENDING(&t->signal->shared_pending, &t->blocked)) {
L
Linus Torvalds 已提交
146
		set_tsk_thread_flag(t, TIF_SIGPENDING);
147
		return true;
R
Roland McGrath 已提交
148
	}
149

150 151 152 153 154
	/*
	 * 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.
	 */
155
	return false;
R
Roland McGrath 已提交
156 157 158 159 160 161 162 163 164 165
}

/*
 * 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
166 167 168 169
}

void recalc_sigpending(void)
{
170 171
	if (!recalc_sigpending_tsk(current) && !freezing(current) &&
	    !klp_patch_pending(current))
172 173
		clear_thread_flag(TIF_SIGPENDING);

L
Linus Torvalds 已提交
174 175
}

176 177 178 179 180 181 182 183 184 185 186
void calculate_sigpending(void)
{
	/* Have any signals or users of TIF_SIGPENDING been delayed
	 * until after fork?
	 */
	spin_lock_irq(&current->sighand->siglock);
	set_tsk_thread_flag(current, TIF_SIGPENDING);
	recalc_sigpending();
	spin_unlock_irq(&current->sighand->siglock);
}

L
Linus Torvalds 已提交
187 188
/* Given the mask, find the first available signal that should be serviced. */

189 190
#define SYNCHRONOUS_MASK \
	(sigmask(SIGSEGV) | sigmask(SIGBUS) | sigmask(SIGILL) | \
191
	 sigmask(SIGTRAP) | sigmask(SIGFPE) | sigmask(SIGSYS))
192

193
int next_signal(struct sigpending *pending, sigset_t *mask)
L
Linus Torvalds 已提交
194 195 196
{
	unsigned long i, *s, *m, x;
	int sig = 0;
197

L
Linus Torvalds 已提交
198 199
	s = pending->signal.sig;
	m = mask->sig;
200 201 202 203 204 205 206 207 208 209 210 211 212

	/*
	 * 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
213 214
	switch (_NSIG_WORDS) {
	default:
215 216 217 218 219 220 221
		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
222 223
		break;

224 225 226
	case 2:
		x = s[1] &~ m[1];
		if (!x)
L
Linus Torvalds 已提交
227
			break;
228
		sig = ffz(~x) + _NSIG_BPW + 1;
L
Linus Torvalds 已提交
229 230
		break;

231 232
	case 1:
		/* Nothing to do */
L
Linus Torvalds 已提交
233 234
		break;
	}
235

L
Linus Torvalds 已提交
236 237 238
	return sig;
}

239 240 241 242 243 244 245 246 247 248
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;

249
	pr_info("%s/%d: reached RLIMIT_SIGPENDING, dropped signal %d\n",
250 251 252
				current->comm, current->pid, sig);
}

253
/**
254
 * task_set_jobctl_pending - set jobctl pending bits
255
 * @task: target task
256
 * @mask: pending bits to set
257
 *
258 259 260 261 262 263 264 265 266 267 268 269
 * 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.
 */
270
bool task_set_jobctl_pending(struct task_struct *task, unsigned long mask)
271 272 273 274 275 276 277 278 279 280 281 282 283 284 285
{
	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;
}

286
/**
287
 * task_clear_jobctl_trapping - clear jobctl trapping bit
288 289
 * @task: target task
 *
290 291 292 293
 * 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.
294 295 296 297
 *
 * CONTEXT:
 * Must be called with @task->sighand->siglock held.
 */
298
void task_clear_jobctl_trapping(struct task_struct *task)
299
{
300 301
	if (unlikely(task->jobctl & JOBCTL_TRAPPING)) {
		task->jobctl &= ~JOBCTL_TRAPPING;
302
		smp_mb();	/* advised by wake_up_bit() */
303
		wake_up_bit(&task->jobctl, JOBCTL_TRAPPING_BIT);
304 305 306
	}
}

307
/**
308
 * task_clear_jobctl_pending - clear jobctl pending bits
309
 * @task: target task
310
 * @mask: pending bits to clear
311
 *
312 313 314
 * 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.
315
 *
316 317
 * If clearing of @mask leaves no stop or trap pending, this function calls
 * task_clear_jobctl_trapping().
318 319 320 321
 *
 * CONTEXT:
 * Must be called with @task->sighand->siglock held.
 */
322
void task_clear_jobctl_pending(struct task_struct *task, unsigned long mask)
323
{
324 325 326 327 328 329
	BUG_ON(mask & ~JOBCTL_PENDING_MASK);

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

	task->jobctl &= ~mask;
330 331 332

	if (!(task->jobctl & JOBCTL_PENDING_MASK))
		task_clear_jobctl_trapping(task);
333 334 335 336 337 338
}

/**
 * task_participate_group_stop - participate in a group stop
 * @task: task participating in a group stop
 *
339
 * @task has %JOBCTL_STOP_PENDING set and is participating in a group stop.
340
 * Group stop states are cleared and the group stop count is consumed if
341
 * %JOBCTL_STOP_CONSUME was set.  If the consumption completes the group
342
 * stop, the appropriate %SIGNAL_* flags are set.
343 344 345
 *
 * CONTEXT:
 * Must be called with @task->sighand->siglock held.
346 347 348 349
 *
 * RETURNS:
 * %true if group stop completion should be notified to the parent, %false
 * otherwise.
350 351 352 353
 */
static bool task_participate_group_stop(struct task_struct *task)
{
	struct signal_struct *sig = task->signal;
354
	bool consume = task->jobctl & JOBCTL_STOP_CONSUME;
355

356
	WARN_ON_ONCE(!(task->jobctl & JOBCTL_STOP_PENDING));
357

358
	task_clear_jobctl_pending(task, JOBCTL_STOP_PENDING);
359 360 361 362 363 364 365

	if (!consume)
		return false;

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

366 367 368 369 370
	/*
	 * 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)) {
371
		signal_set_stop_flags(sig, SIGNAL_STOP_STOPPED);
372 373 374 375 376
		return true;
	}
	return false;
}

377 378 379 380 381 382 383 384 385 386 387 388 389 390
void task_join_group_stop(struct task_struct *task)
{
	/* Have the new thread join an on-going signal group stop */
	unsigned long jobctl = current->jobctl;
	if (jobctl & JOBCTL_STOP_PENDING) {
		struct signal_struct *sig = current->signal;
		unsigned long signr = jobctl & JOBCTL_STOP_SIGMASK;
		unsigned long gstop = JOBCTL_STOP_PENDING | JOBCTL_STOP_CONSUME;
		if (task_set_jobctl_pending(task, signr | gstop)) {
			sig->group_stop_count++;
		}
	}
}

391 392 393
/*
 * allocate a new signal queue record
 * - this may be called without locks if and only if t == current, otherwise an
394
 *   appropriate lock must be held to stop the target task from exiting
395
 */
396 397
static struct sigqueue *
__sigqueue_alloc(int sig, struct task_struct *t, gfp_t flags, int override_rlimit)
L
Linus Torvalds 已提交
398 399
{
	struct sigqueue *q = NULL;
400
	struct user_struct *user;
L
Linus Torvalds 已提交
401

402
	/*
403 404
	 * Protect access to @t credentials. This can go away when all
	 * callers hold rcu read lock.
405
	 */
406
	rcu_read_lock();
D
David Howells 已提交
407
	user = get_uid(__task_cred(t)->user);
408
	atomic_inc(&user->sigpending);
409
	rcu_read_unlock();
410

L
Linus Torvalds 已提交
411
	if (override_rlimit ||
412
	    atomic_read(&user->sigpending) <=
413
			task_rlimit(t, RLIMIT_SIGPENDING)) {
L
Linus Torvalds 已提交
414
		q = kmem_cache_alloc(sigqueue_cachep, flags);
415 416 417 418
	} else {
		print_dropped_signal(sig);
	}

L
Linus Torvalds 已提交
419
	if (unlikely(q == NULL)) {
420
		atomic_dec(&user->sigpending);
D
David Howells 已提交
421
		free_uid(user);
L
Linus Torvalds 已提交
422 423 424
	} else {
		INIT_LIST_HEAD(&q->list);
		q->flags = 0;
D
David Howells 已提交
425
		q->user = user;
L
Linus Torvalds 已提交
426
	}
D
David Howells 已提交
427 428

	return q;
L
Linus Torvalds 已提交
429 430
}

431
static void __sigqueue_free(struct sigqueue *q)
L
Linus Torvalds 已提交
432 433 434 435 436 437 438 439
{
	if (q->flags & SIGQUEUE_PREALLOC)
		return;
	atomic_dec(&q->user->sigpending);
	free_uid(q->user);
	kmem_cache_free(sigqueue_cachep, q);
}

440
void flush_sigqueue(struct sigpending *queue)
L
Linus Torvalds 已提交
441 442 443 444 445 446 447 448 449 450 451 452
{
	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);
	}
}

/*
453
 * Flush all pending signals for this kthread.
L
Linus Torvalds 已提交
454
 */
455
void flush_signals(struct task_struct *t)
L
Linus Torvalds 已提交
456 457 458 459
{
	unsigned long flags;

	spin_lock_irqsave(&t->sighand->siglock, flags);
460 461 462
	clear_tsk_thread_flag(t, TIF_SIGPENDING);
	flush_sigqueue(&t->pending);
	flush_sigqueue(&t->signal->shared_pending);
L
Linus Torvalds 已提交
463 464 465
	spin_unlock_irqrestore(&t->sighand->siglock, flags);
}

466
#ifdef CONFIG_POSIX_TIMERS
467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499
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);
}
500
#endif
501

502 503 504 505 506 507 508 509 510 511
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
512 513 514 515 516 517 518 519 520 521 522 523 524
/*
 * 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;
525
#ifdef __ARCH_HAS_SA_RESTORER
526 527
		ka->sa.sa_restorer = NULL;
#endif
L
Linus Torvalds 已提交
528 529 530 531 532
		sigemptyset(&ka->sa.sa_mask);
		ka++;
	}
}

533
bool unhandled_signal(struct task_struct *tsk, int sig)
534
{
535
	void __user *handler = tsk->sighand->action[sig-1].sa.sa_handler;
536
	if (is_global_init(tsk))
537 538
		return true;

539
	if (handler != SIG_IGN && handler != SIG_DFL)
540 541
		return false;

T
Tejun Heo 已提交
542 543
	/* if ptraced, let the tracer determine */
	return !tsk->ptrace;
544 545
}

546 547
static void collect_signal(int sig, struct sigpending *list, siginfo_t *info,
			   bool *resched_timer)
L
Linus Torvalds 已提交
548 549 550 551 552 553 554 555 556
{
	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) {
557 558
			if (first)
				goto still_pending;
L
Linus Torvalds 已提交
559 560 561
			first = q;
		}
	}
562 563 564

	sigdelset(&list->signal, sig);

L
Linus Torvalds 已提交
565
	if (first) {
566
still_pending:
L
Linus Torvalds 已提交
567 568
		list_del_init(&first->list);
		copy_siginfo(info, &first->info);
569 570 571 572 573 574

		*resched_timer =
			(first->flags & SIGQUEUE_PREALLOC) &&
			(info->si_code == SI_TIMER) &&
			(info->si_sys_private);

L
Linus Torvalds 已提交
575 576
		__sigqueue_free(first);
	} else {
577 578 579 580
		/*
		 * 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
581
		 */
582
		clear_siginfo(info);
L
Linus Torvalds 已提交
583 584
		info->si_signo = sig;
		info->si_errno = 0;
585
		info->si_code = SI_USER;
L
Linus Torvalds 已提交
586 587 588 589 590 591
		info->si_pid = 0;
		info->si_uid = 0;
	}
}

static int __dequeue_signal(struct sigpending *pending, sigset_t *mask,
592
			siginfo_t *info, bool *resched_timer)
L
Linus Torvalds 已提交
593
{
594
	int sig = next_signal(pending, mask);
L
Linus Torvalds 已提交
595

596
	if (sig)
597
		collect_signal(sig, pending, info, resched_timer);
L
Linus Torvalds 已提交
598 599 600 601
	return sig;
}

/*
602
 * Dequeue a signal and return the element to the caller, which is
L
Linus Torvalds 已提交
603 604 605 606 607 608
 * expected to free it.
 *
 * All callers have to hold the siglock.
 */
int dequeue_signal(struct task_struct *tsk, sigset_t *mask, siginfo_t *info)
{
609
	bool resched_timer = false;
610
	int signr;
611 612 613 614

	/* We only dequeue private signals from ourselves, we don't let
	 * signalfd steal them
	 */
615
	signr = __dequeue_signal(&tsk->pending, mask, info, &resched_timer);
616
	if (!signr) {
L
Linus Torvalds 已提交
617
		signr = __dequeue_signal(&tsk->signal->shared_pending,
618
					 mask, info, &resched_timer);
619
#ifdef CONFIG_POSIX_TIMERS
620 621 622 623 624 625
		/*
		 * 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
626
		 * compliant with the old way of self-restarting
627 628 629 630 631 632 633 634 635 636
		 * 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
637
			    tsk->signal->it_real_incr != 0) {
638 639 640 641 642
				hrtimer_forward(tmr, tmr->base->get_time(),
						tsk->signal->it_real_incr);
				hrtimer_restart(tmr);
			}
		}
643
#endif
644
	}
645

D
Davide Libenzi 已提交
646
	recalc_sigpending();
647 648 649 650
	if (!signr)
		return 0;

	if (unlikely(sig_kernel_stop(signr))) {
651 652 653 654 655 656 657 658 659 660 661 662
		/*
		 * 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.
		 */
663
		current->jobctl |= JOBCTL_STOP_DEQUEUED;
664
	}
665
#ifdef CONFIG_POSIX_TIMERS
666
	if (resched_timer) {
L
Linus Torvalds 已提交
667 668 669 670 671 672 673
		/*
		 * 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);
674
		posixtimer_rearm(info);
L
Linus Torvalds 已提交
675
		spin_lock(&tsk->sighand->siglock);
676 677 678

		/* Don't expose the si_sys_private value to userspace */
		info->si_sys_private = 0;
L
Linus Torvalds 已提交
679
	}
680
#endif
L
Linus Torvalds 已提交
681 682 683
	return signr;
}

684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725
static int dequeue_synchronous_signal(siginfo_t *info)
{
	struct task_struct *tsk = current;
	struct sigpending *pending = &tsk->pending;
	struct sigqueue *q, *sync = NULL;

	/*
	 * Might a synchronous signal be in the queue?
	 */
	if (!((pending->signal.sig[0] & ~tsk->blocked.sig[0]) & SYNCHRONOUS_MASK))
		return 0;

	/*
	 * Return the first synchronous signal in the queue.
	 */
	list_for_each_entry(q, &pending->list, list) {
		/* Synchronous signals have a postive si_code */
		if ((q->info.si_code > SI_USER) &&
		    (sigmask(q->info.si_signo) & SYNCHRONOUS_MASK)) {
			sync = q;
			goto next;
		}
	}
	return 0;
next:
	/*
	 * Check if there is another siginfo for the same signal.
	 */
	list_for_each_entry_continue(q, &pending->list, list) {
		if (q->info.si_signo == sync->info.si_signo)
			goto still_pending;
	}

	sigdelset(&pending->signal, sync->info.si_signo);
	recalc_sigpending();
still_pending:
	list_del_init(&sync->list);
	copy_siginfo(info, &sync->info);
	__sigqueue_free(sync);
	return info->si_signo;
}

L
Linus Torvalds 已提交
726 727 728 729 730 731 732 733 734 735 736
/*
 * 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
 */
737
void signal_wake_up_state(struct task_struct *t, unsigned int state)
L
Linus Torvalds 已提交
738 739 740
{
	set_tsk_thread_flag(t, TIF_SIGPENDING);
	/*
741
	 * TASK_WAKEKILL also means wake it up in the stopped/traced/killable
M
Matthew Wilcox 已提交
742
	 * case. We don't check t->state here because there is a race with it
L
Linus Torvalds 已提交
743 744 745 746
	 * 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.
	 */
747
	if (!wake_up_state(t, state | TASK_INTERRUPTIBLE))
L
Linus Torvalds 已提交
748 749 750
		kick_process(t);
}

751 752 753 754 755 756
/*
 * Remove signals in mask from the pending set and queue.
 * Returns 1 if any signals were found.
 *
 * All callers must be holding the siglock.
 */
757
static void flush_sigqueue_mask(sigset_t *mask, struct sigpending *s)
758 759 760 761 762 763
{
	struct sigqueue *q, *n;
	sigset_t m;

	sigandsets(&m, mask, &s->signal);
	if (sigisemptyset(&m))
764
		return;
765

766
	sigandnsets(&s->signal, &s->signal, mask);
767 768 769 770 771 772 773
	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);
		}
	}
}
L
Linus Torvalds 已提交
774

775 776 777 778 779 780 781 782 783 784 785
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));
}

786 787 788
/*
 * called with RCU read lock from check_kill_permission()
 */
789
static bool kill_ok_by_cred(struct task_struct *t)
790 791 792 793
{
	const struct cred *cred = current_cred();
	const struct cred *tcred = __task_cred(t);

794 795 796 797 798
	return uid_eq(cred->euid, tcred->suid) ||
	       uid_eq(cred->euid, tcred->uid) ||
	       uid_eq(cred->uid, tcred->suid) ||
	       uid_eq(cred->uid, tcred->uid) ||
	       ns_capable(tcred->user_ns, CAP_KILL);
799 800
}

L
Linus Torvalds 已提交
801 802
/*
 * Bad permissions for sending the signal
803
 * - the caller must hold the RCU read lock
L
Linus Torvalds 已提交
804 805 806 807
 */
static int check_kill_permission(int sig, struct siginfo *info,
				 struct task_struct *t)
{
808
	struct pid *sid;
809 810
	int error;

811
	if (!valid_signal(sig))
812 813
		return -EINVAL;

814
	if (!si_fromuser(info))
815
		return 0;
A
Amy Griffis 已提交
816

817 818
	error = audit_signal_info(sig, t); /* Let audit system see the signal */
	if (error)
L
Linus Torvalds 已提交
819
		return error;
820

821
	if (!same_thread_group(current, t) &&
822
	    !kill_ok_by_cred(t)) {
823 824 825 826 827 828 829 830 831 832 833 834 835
		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;
		}
	}
836

837
	return security_task_kill(t, info, sig, NULL);
L
Linus Torvalds 已提交
838 839
}

840 841 842 843 844 845 846 847
/**
 * 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
848 849 850 851 852
 * 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.
853 854 855 856 857 858 859 860 861 862
 *
 * 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);
863
	ptrace_signal_wake_up(t, t->jobctl & JOBCTL_LISTENING);
864 865
}

L
Linus Torvalds 已提交
866
/*
867 868
 * 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
869 870
 * time regardless of blocking, ignoring, or handling.  This does the
 * actual continuing for SIGCONT, but not the actual stopping for stop
871 872 873 874
 * 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
875
 */
876
static bool prepare_signal(int sig, struct task_struct *p, bool force)
L
Linus Torvalds 已提交
877
{
878
	struct signal_struct *signal = p->signal;
L
Linus Torvalds 已提交
879
	struct task_struct *t;
880
	sigset_t flush;
L
Linus Torvalds 已提交
881

882
	if (signal->flags & (SIGNAL_GROUP_EXIT | SIGNAL_GROUP_COREDUMP)) {
883
		if (!(signal->flags & SIGNAL_GROUP_EXIT))
884
			return sig == SIGKILL;
L
Linus Torvalds 已提交
885
		/*
886
		 * The process is in the middle of dying, nothing to do.
L
Linus Torvalds 已提交
887
		 */
888
	} else if (sig_kernel_stop(sig)) {
L
Linus Torvalds 已提交
889 890 891
		/*
		 * This is a stop signal.  Remove SIGCONT from all queues.
		 */
892
		siginitset(&flush, sigmask(SIGCONT));
893
		flush_sigqueue_mask(&flush, &signal->shared_pending);
894
		for_each_thread(p, t)
895
			flush_sigqueue_mask(&flush, &t->pending);
L
Linus Torvalds 已提交
896
	} else if (sig == SIGCONT) {
897
		unsigned int why;
L
Linus Torvalds 已提交
898
		/*
899
		 * Remove all stop signals from all queues, wake all threads.
L
Linus Torvalds 已提交
900
		 */
901
		siginitset(&flush, SIG_KERNEL_STOP_MASK);
902
		flush_sigqueue_mask(&flush, &signal->shared_pending);
903
		for_each_thread(p, t) {
904
			flush_sigqueue_mask(&flush, &t->pending);
905
			task_clear_jobctl_pending(t, JOBCTL_STOP_PENDING);
906 907 908 909
			if (likely(!(t->ptrace & PT_SEIZED)))
				wake_up_state(t, __TASK_STOPPED);
			else
				ptrace_trap_notify(t);
910
		}
L
Linus Torvalds 已提交
911

912 913 914 915 916 917 918 919 920
		/*
		 * 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;
921
		if (signal->flags & SIGNAL_STOP_STOPPED)
922
			why |= SIGNAL_CLD_CONTINUED;
923
		else if (signal->group_stop_count)
924 925 926
			why |= SIGNAL_CLD_STOPPED;

		if (why) {
927
			/*
928
			 * The first thread which returns from do_signal_stop()
929 930 931
			 * will take ->siglock, notice SIGNAL_CLD_MASK, and
			 * notify its parent. See get_signal_to_deliver().
			 */
932
			signal_set_stop_flags(signal, why | SIGNAL_STOP_CONTINUED);
933 934
			signal->group_stop_count = 0;
			signal->group_exit_code = 0;
L
Linus Torvalds 已提交
935 936
		}
	}
937

938
	return !sig_ignored(p, sig, force);
L
Linus Torvalds 已提交
939 940
}

941 942 943 944 945 946 947 948
/*
 * 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.
 */
949
static inline bool wants_signal(int sig, struct task_struct *p)
950 951
{
	if (sigismember(&p->blocked, sig))
952 953
		return false;

954
	if (p->flags & PF_EXITING)
955 956
		return false;

957
	if (sig == SIGKILL)
958 959
		return true;

960
	if (task_is_stopped_or_traced(p))
961 962
		return false;

963 964 965
	return task_curr(p) || !signal_pending(p);
}

966
static void complete_signal(int sig, struct task_struct *p, enum pid_type type)
967 968 969 970 971 972 973 974 975 976 977 978
{
	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;
979
	else if ((type == PIDTYPE_PID) || thread_group_empty(p))
980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006
		/*
		 * 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.
	 */
1007
	if (sig_fatal(p, sig) &&
1008
	    !(signal->flags & SIGNAL_GROUP_EXIT) &&
1009
	    !sigismember(&t->real_blocked, sig) &&
1010
	    (sig == SIGKILL || !p->ptrace)) {
1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025
		/*
		 * 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 {
1026
				task_clear_jobctl_pending(t, JOBCTL_PENDING_MASK);
1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041
				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;
}

1042
static inline bool legacy_queue(struct sigpending *signals, int sig)
1043 1044 1045 1046
{
	return (sig < SIGRTMIN) && sigismember(&signals->signal, sig);
}

1047 1048 1049 1050 1051 1052 1053 1054 1055
#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;

1056 1057 1058 1059
	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();
1060 1061 1062 1063 1064 1065 1066 1067
}
#else
static inline void userns_fixup_signal_uid(struct siginfo *info, struct task_struct *t)
{
	return;
}
#endif

1068
static int __send_signal(int sig, struct siginfo *info, struct task_struct *t,
1069
			enum pid_type type, int from_ancestor_ns)
L
Linus Torvalds 已提交
1070
{
1071
	struct sigpending *pending;
1072
	struct sigqueue *q;
1073
	int override_rlimit;
1074
	int ret = 0, result;
1075

1076
	assert_spin_locked(&t->sighand->siglock);
1077

1078
	result = TRACE_SIGNAL_IGNORED;
1079
	if (!prepare_signal(sig, t,
1080
			from_ancestor_ns || (info == SEND_SIG_PRIV) || (info == SEND_SIG_FORCED)))
1081
		goto ret;
1082

1083
	pending = (type != PIDTYPE_PID) ? &t->signal->shared_pending : &t->pending;
1084 1085 1086 1087 1088
	/*
	 * 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.
	 */
1089
	result = TRACE_SIGNAL_ALREADY_PENDING;
1090
	if (legacy_queue(pending, sig))
1091 1092 1093
		goto ret;

	result = TRACE_SIGNAL_DELIVERED;
L
Linus Torvalds 已提交
1094 1095 1096 1097
	/*
	 * fast-pathed signals for kernel-internal things like SIGSTOP
	 * or SIGKILL.
	 */
1098
	if (info == SEND_SIG_FORCED)
L
Linus Torvalds 已提交
1099 1100
		goto out_set;

1101 1102 1103 1104 1105 1106 1107 1108 1109
	/*
	 * 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.
	 */
1110 1111 1112 1113 1114
	if (sig < SIGRTMIN)
		override_rlimit = (is_si_special(info) || info->si_code >= 0);
	else
		override_rlimit = 0;

1115
	q = __sigqueue_alloc(sig, t, GFP_ATOMIC, override_rlimit);
L
Linus Torvalds 已提交
1116
	if (q) {
1117
		list_add_tail(&q->list, &pending->list);
L
Linus Torvalds 已提交
1118
		switch ((unsigned long) info) {
1119
		case (unsigned long) SEND_SIG_NOINFO:
1120
			clear_siginfo(&q->info);
L
Linus Torvalds 已提交
1121 1122 1123
			q->info.si_signo = sig;
			q->info.si_errno = 0;
			q->info.si_code = SI_USER;
1124
			q->info.si_pid = task_tgid_nr_ns(current,
1125
							task_active_pid_ns(t));
1126
			q->info.si_uid = from_kuid_munged(current_user_ns(), current_uid());
L
Linus Torvalds 已提交
1127
			break;
1128
		case (unsigned long) SEND_SIG_PRIV:
1129
			clear_siginfo(&q->info);
L
Linus Torvalds 已提交
1130 1131 1132 1133 1134 1135 1136 1137
			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);
1138 1139
			if (from_ancestor_ns)
				q->info.si_pid = 0;
L
Linus Torvalds 已提交
1140 1141
			break;
		}
1142 1143 1144

		userns_fixup_signal_uid(&q->info, t);

1145
	} else if (!is_si_special(info)) {
1146 1147 1148 1149 1150 1151
		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().
			 */
1152 1153 1154
			result = TRACE_SIGNAL_OVERFLOW_FAIL;
			ret = -EAGAIN;
			goto ret;
1155 1156 1157 1158 1159
		} else {
			/*
			 * This is a silent loss of information.  We still
			 * send the signal, but the *info bits are lost.
			 */
1160
			result = TRACE_SIGNAL_LOSE_INFO;
1161
		}
L
Linus Torvalds 已提交
1162 1163 1164
	}

out_set:
1165
	signalfd_notify(t, sig);
1166
	sigaddset(&pending->signal, sig);
1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181

	/* Let multiprocess signals appear after on-going forks */
	if (type > PIDTYPE_TGID) {
		struct multiprocess_signals *delayed;
		hlist_for_each_entry(delayed, &t->signal->multiprocess, node) {
			sigset_t *signal = &delayed->signal;
			/* Can't queue both a stop and a continue signal */
			if (sig == SIGCONT)
				sigdelsetmask(signal, SIG_KERNEL_STOP_MASK);
			else if (sig_kernel_stop(sig))
				sigdelset(signal, SIGCONT);
			sigaddset(signal, sig);
		}
	}

1182
	complete_signal(sig, t, type);
1183
ret:
1184
	trace_signal_generate(sig, info, t, type != PIDTYPE_PID, result);
1185
	return ret;
L
Linus Torvalds 已提交
1186 1187
}

1188
static int send_signal(int sig, struct siginfo *info, struct task_struct *t,
1189
			enum pid_type type)
1190
{
1191 1192 1193
	int from_ancestor_ns = 0;

#ifdef CONFIG_PID_NS
1194 1195
	from_ancestor_ns = si_fromuser(info) &&
			   !task_pid_nr_ns(current, task_active_pid_ns(t));
1196 1197
#endif

1198
	return __send_signal(sig, info, t, type, from_ancestor_ns);
1199 1200
}

1201
static void print_fatal_signal(int signr)
I
Ingo Molnar 已提交
1202
{
1203
	struct pt_regs *regs = signal_pt_regs();
1204
	pr_info("potentially unexpected fatal signal %d.\n", signr);
I
Ingo Molnar 已提交
1205

A
Al Viro 已提交
1206
#if defined(__i386__) && !defined(__arch_um__)
1207
	pr_info("code at %08lx: ", regs->ip);
I
Ingo Molnar 已提交
1208 1209 1210 1211 1212
	{
		int i;
		for (i = 0; i < 16; i++) {
			unsigned char insn;

1213 1214
			if (get_user(insn, (unsigned char *)(regs->ip + i)))
				break;
1215
			pr_cont("%02x ", insn);
I
Ingo Molnar 已提交
1216 1217
		}
	}
1218
	pr_cont("\n");
I
Ingo Molnar 已提交
1219
#endif
1220
	preempt_disable();
I
Ingo Molnar 已提交
1221
	show_regs(regs);
1222
	preempt_enable();
I
Ingo Molnar 已提交
1223 1224 1225 1226 1227 1228 1229 1230 1231 1232
}

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
1233

1234 1235 1236
int
__group_send_sig_info(int sig, struct siginfo *info, struct task_struct *p)
{
1237
	return send_signal(sig, info, p, PIDTYPE_TGID);
1238 1239
}

L
Linus Torvalds 已提交
1240 1241 1242
static int
specific_send_sig_info(int sig, struct siginfo *info, struct task_struct *t)
{
1243
	return send_signal(sig, info, t, PIDTYPE_PID);
L
Linus Torvalds 已提交
1244 1245
}

1246
int do_send_sig_info(int sig, struct siginfo *info, struct task_struct *p,
1247
			enum pid_type type)
1248 1249 1250 1251 1252
{
	unsigned long flags;
	int ret = -ESRCH;

	if (lock_task_sighand(p, &flags)) {
1253
		ret = send_signal(sig, info, p, type);
1254 1255 1256 1257 1258 1259
		unlock_task_sighand(p, &flags);
	}

	return ret;
}

L
Linus Torvalds 已提交
1260 1261 1262
/*
 * Force a signal that the process can't ignore: if necessary
 * we unblock the signal and change any SIG_IGN to SIG_DFL.
1263 1264 1265 1266 1267
 *
 * 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.
 *
1268 1269
 * 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
1270 1271 1272 1273 1274
 */
int
force_sig_info(int sig, struct siginfo *info, struct task_struct *t)
{
	unsigned long int flags;
1275 1276
	int ret, blocked, ignored;
	struct k_sigaction *action;
L
Linus Torvalds 已提交
1277 1278

	spin_lock_irqsave(&t->sighand->siglock, flags);
1279 1280 1281 1282 1283 1284 1285
	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
1286
			recalc_sigpending_and_wake(t);
1287
		}
L
Linus Torvalds 已提交
1288
	}
1289 1290 1291 1292 1293
	/*
	 * Don't clear SIGNAL_UNKILLABLE for traced tasks, users won't expect
	 * debugging to leave init killable.
	 */
	if (action->sa.sa_handler == SIG_DFL && !t->ptrace)
1294
		t->signal->flags &= ~SIGNAL_UNKILLABLE;
L
Linus Torvalds 已提交
1295 1296 1297 1298 1299 1300 1301 1302 1303
	ret = specific_send_sig_info(sig, info, t);
	spin_unlock_irqrestore(&t->sighand->siglock, flags);

	return ret;
}

/*
 * Nuke all other threads in the group.
 */
1304
int zap_other_threads(struct task_struct *p)
L
Linus Torvalds 已提交
1305
{
1306 1307
	struct task_struct *t = p;
	int count = 0;
L
Linus Torvalds 已提交
1308 1309 1310

	p->signal->group_stop_count = 0;

1311
	while_each_thread(p, t) {
1312
		task_clear_jobctl_pending(t, JOBCTL_PENDING_MASK);
1313 1314 1315
		count++;

		/* Don't bother with already dead threads */
L
Linus Torvalds 已提交
1316 1317 1318 1319 1320
		if (t->exit_state)
			continue;
		sigaddset(&t->pending.signal, SIGKILL);
		signal_wake_up(t, 1);
	}
1321 1322

	return count;
L
Linus Torvalds 已提交
1323 1324
}

1325 1326
struct sighand_struct *__lock_task_sighand(struct task_struct *tsk,
					   unsigned long *flags)
1327 1328 1329
{
	struct sighand_struct *sighand;

1330
	rcu_read_lock();
1331 1332
	for (;;) {
		sighand = rcu_dereference(tsk->sighand);
1333
		if (unlikely(sighand == NULL))
1334
			break;
1335

1336 1337
		/*
		 * This sighand can be already freed and even reused, but
1338
		 * we rely on SLAB_TYPESAFE_BY_RCU and sighand_ctor() which
1339 1340 1341 1342 1343 1344 1345 1346
		 * 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.
		 */
1347 1348
		spin_lock_irqsave(&sighand->siglock, *flags);
		if (likely(sighand == tsk->sighand))
1349
			break;
1350
		spin_unlock_irqrestore(&sighand->siglock, *flags);
1351
	}
1352
	rcu_read_unlock();
1353 1354 1355 1356

	return sighand;
}

1357 1358 1359
/*
 * send signal info to all the members of a group
 */
1360 1361
int group_send_sig_info(int sig, struct siginfo *info, struct task_struct *p,
			enum pid_type type)
L
Linus Torvalds 已提交
1362
{
1363 1364 1365 1366 1367
	int ret;

	rcu_read_lock();
	ret = check_kill_permission(sig, info, p);
	rcu_read_unlock();
1368

1369
	if (!ret && sig)
1370
		ret = do_send_sig_info(sig, info, p, type);
L
Linus Torvalds 已提交
1371 1372 1373 1374 1375

	return ret;
}

/*
1376
 * __kill_pgrp_info() sends a signal to a process group: this is what the tty
L
Linus Torvalds 已提交
1377
 * control characters do (^C, ^Z etc)
1378
 * - the caller must hold at least a readlock on tasklist_lock
L
Linus Torvalds 已提交
1379
 */
1380
int __kill_pgrp_info(int sig, struct siginfo *info, struct pid *pgrp)
L
Linus Torvalds 已提交
1381 1382 1383 1384 1385 1386
{
	struct task_struct *p = NULL;
	int retval, success;

	success = 0;
	retval = -ESRCH;
1387
	do_each_pid_task(pgrp, PIDTYPE_PGID, p) {
1388
		int err = group_send_sig_info(sig, info, p, PIDTYPE_PGID);
L
Linus Torvalds 已提交
1389 1390
		success |= !err;
		retval = err;
1391
	} while_each_pid_task(pgrp, PIDTYPE_PGID, p);
L
Linus Torvalds 已提交
1392 1393 1394
	return success ? 0 : retval;
}

1395
int kill_pid_info(int sig, struct siginfo *info, struct pid *pid)
L
Linus Torvalds 已提交
1396
{
1397
	int error = -ESRCH;
L
Linus Torvalds 已提交
1398 1399
	struct task_struct *p;

1400 1401 1402 1403
	for (;;) {
		rcu_read_lock();
		p = pid_task(pid, PIDTYPE_PID);
		if (p)
1404
			error = group_send_sig_info(sig, info, p, PIDTYPE_TGID);
1405 1406 1407
		rcu_read_unlock();
		if (likely(!p || error != -ESRCH))
			return error;
1408

1409 1410 1411 1412 1413 1414
		/*
		 * 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
1415 1416
}

1417
static int kill_proc_info(int sig, struct siginfo *info, pid_t pid)
1418 1419 1420
{
	int error;
	rcu_read_lock();
1421
	error = kill_pid_info(sig, info, find_vpid(pid));
1422 1423 1424 1425
	rcu_read_unlock();
	return error;
}

1426 1427
static inline bool kill_as_cred_perm(const struct cred *cred,
				     struct task_struct *target)
1428 1429
{
	const struct cred *pcred = __task_cred(target);
1430 1431 1432 1433 1434

	return uid_eq(cred->euid, pcred->suid) ||
	       uid_eq(cred->euid, pcred->uid) ||
	       uid_eq(cred->uid, pcred->suid) ||
	       uid_eq(cred->uid, pcred->uid);
1435 1436
}

1437
/* like kill_pid_info(), but doesn't use uid/euid of "current" */
1438
int kill_pid_info_as_cred(int sig, struct siginfo *info, struct pid *pid,
1439
			 const struct cred *cred)
1440 1441 1442
{
	int ret = -EINVAL;
	struct task_struct *p;
1443
	unsigned long flags;
1444 1445 1446 1447

	if (!valid_signal(sig))
		return ret;

1448
	rcu_read_lock();
1449
	p = pid_task(pid, PIDTYPE_PID);
1450 1451 1452 1453
	if (!p) {
		ret = -ESRCH;
		goto out_unlock;
	}
1454
	if (si_fromuser(info) && !kill_as_cred_perm(cred, p)) {
1455 1456 1457
		ret = -EPERM;
		goto out_unlock;
	}
1458
	ret = security_task_kill(p, info, sig, cred);
1459 1460
	if (ret)
		goto out_unlock;
1461 1462 1463

	if (sig) {
		if (lock_task_sighand(p, &flags)) {
1464
			ret = __send_signal(sig, info, p, PIDTYPE_TGID, 0);
1465 1466 1467
			unlock_task_sighand(p, &flags);
		} else
			ret = -ESRCH;
1468 1469
	}
out_unlock:
1470
	rcu_read_unlock();
1471 1472
	return ret;
}
1473
EXPORT_SYMBOL_GPL(kill_pid_info_as_cred);
L
Linus Torvalds 已提交
1474 1475 1476 1477 1478 1479 1480 1481

/*
 * 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.
 */

1482
static int kill_something_info(int sig, struct siginfo *info, pid_t pid)
L
Linus Torvalds 已提交
1483
{
1484
	int ret;
1485 1486 1487 1488 1489 1490 1491 1492

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

1493 1494 1495 1496
	/* -INT_MIN is undefined.  Exclude this case to avoid a UBSAN warning */
	if (pid == INT_MIN)
		return -ESRCH;

1497 1498 1499 1500 1501
	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
1502 1503 1504 1505
		int retval = 0, count = 0;
		struct task_struct * p;

		for_each_process(p) {
1506 1507
			if (task_pid_vnr(p) > 1 &&
					!same_thread_group(p, current)) {
1508 1509
				int err = group_send_sig_info(sig, info, p,
							      PIDTYPE_MAX);
L
Linus Torvalds 已提交
1510 1511 1512 1513 1514
				++count;
				if (err != -EPERM)
					retval = err;
			}
		}
1515
		ret = count ? retval : -ESRCH;
L
Linus Torvalds 已提交
1516
	}
1517 1518
	read_unlock(&tasklist_lock);

1519
	return ret;
L
Linus Torvalds 已提交
1520 1521 1522 1523 1524 1525
}

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

1526
int send_sig_info(int sig, struct siginfo *info, struct task_struct *p)
L
Linus Torvalds 已提交
1527 1528 1529 1530 1531
{
	/*
	 * Make sure legacy kernel users don't send in bad values
	 * (normal paths check this in check_kill_permission).
	 */
1532
	if (!valid_signal(sig))
L
Linus Torvalds 已提交
1533 1534
		return -EINVAL;

1535
	return do_send_sig_info(sig, info, p, PIDTYPE_PID);
L
Linus Torvalds 已提交
1536 1537
}

1538 1539 1540
#define __si_special(priv) \
	((priv) ? SEND_SIG_PRIV : SEND_SIG_NOINFO)

L
Linus Torvalds 已提交
1541 1542 1543
int
send_sig(int sig, struct task_struct *p, int priv)
{
1544
	return send_sig_info(sig, __si_special(priv), p);
L
Linus Torvalds 已提交
1545 1546
}

1547
void force_sig(int sig, struct task_struct *p)
L
Linus Torvalds 已提交
1548
{
1549
	force_sig_info(sig, SEND_SIG_PRIV, p);
L
Linus Torvalds 已提交
1550 1551 1552 1553 1554 1555 1556 1557
}

/*
 * 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..
 */
1558
void force_sigsegv(int sig, struct task_struct *p)
L
Linus Torvalds 已提交
1559 1560 1561 1562 1563 1564 1565 1566 1567 1568
{
	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);
}

1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614
int force_sig_fault(int sig, int code, void __user *addr
	___ARCH_SI_TRAPNO(int trapno)
	___ARCH_SI_IA64(int imm, unsigned int flags, unsigned long isr)
	, struct task_struct *t)
{
	struct siginfo info;

	clear_siginfo(&info);
	info.si_signo = sig;
	info.si_errno = 0;
	info.si_code  = code;
	info.si_addr  = addr;
#ifdef __ARCH_SI_TRAPNO
	info.si_trapno = trapno;
#endif
#ifdef __ia64__
	info.si_imm = imm;
	info.si_flags = flags;
	info.si_isr = isr;
#endif
	return force_sig_info(info.si_signo, &info, t);
}

int send_sig_fault(int sig, int code, void __user *addr
	___ARCH_SI_TRAPNO(int trapno)
	___ARCH_SI_IA64(int imm, unsigned int flags, unsigned long isr)
	, struct task_struct *t)
{
	struct siginfo info;

	clear_siginfo(&info);
	info.si_signo = sig;
	info.si_errno = 0;
	info.si_code  = code;
	info.si_addr  = addr;
#ifdef __ARCH_SI_TRAPNO
	info.si_trapno = trapno;
#endif
#ifdef __ia64__
	info.si_imm = imm;
	info.si_flags = flags;
	info.si_isr = isr;
#endif
	return send_sig_info(info.si_signo, &info, t);
}

1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671
int force_sig_mceerr(int code, void __user *addr, short lsb, struct task_struct *t)
{
	struct siginfo info;

	WARN_ON((code != BUS_MCEERR_AO) && (code != BUS_MCEERR_AR));
	clear_siginfo(&info);
	info.si_signo = SIGBUS;
	info.si_errno = 0;
	info.si_code = code;
	info.si_addr = addr;
	info.si_addr_lsb = lsb;
	return force_sig_info(info.si_signo, &info, t);
}

int send_sig_mceerr(int code, void __user *addr, short lsb, struct task_struct *t)
{
	struct siginfo info;

	WARN_ON((code != BUS_MCEERR_AO) && (code != BUS_MCEERR_AR));
	clear_siginfo(&info);
	info.si_signo = SIGBUS;
	info.si_errno = 0;
	info.si_code = code;
	info.si_addr = addr;
	info.si_addr_lsb = lsb;
	return send_sig_info(info.si_signo, &info, t);
}
EXPORT_SYMBOL(send_sig_mceerr);

int force_sig_bnderr(void __user *addr, void __user *lower, void __user *upper)
{
	struct siginfo info;

	clear_siginfo(&info);
	info.si_signo = SIGSEGV;
	info.si_errno = 0;
	info.si_code  = SEGV_BNDERR;
	info.si_addr  = addr;
	info.si_lower = lower;
	info.si_upper = upper;
	return force_sig_info(info.si_signo, &info, current);
}

#ifdef SEGV_PKUERR
int force_sig_pkuerr(void __user *addr, u32 pkey)
{
	struct siginfo info;

	clear_siginfo(&info);
	info.si_signo = SIGSEGV;
	info.si_errno = 0;
	info.si_code  = SEGV_PKUERR;
	info.si_addr  = addr;
	info.si_pkey  = pkey;
	return force_sig_info(info.si_signo, &info, current);
}
#endif
1672

1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687
/* For the crazy architectures that include trap information in
 * the errno field, instead of an actual errno value.
 */
int force_sig_ptrace_errno_trap(int errno, void __user *addr)
{
	struct siginfo info;

	clear_siginfo(&info);
	info.si_signo = SIGTRAP;
	info.si_errno = errno;
	info.si_code  = TRAP_HWBKPT;
	info.si_addr  = addr;
	return force_sig_info(info.si_signo, &info, current);
}

1688 1689
int kill_pgrp(struct pid *pid, int sig, int priv)
{
1690 1691 1692 1693 1694 1695 1696
	int ret;

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

	return ret;
1697 1698 1699 1700 1701 1702 1703 1704 1705
}
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
1706 1707 1708 1709
/*
 * 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
1710
 * expirations or I/O completions".  In the case of POSIX Timers
L
Linus Torvalds 已提交
1711 1712 1713 1714 1715 1716
 * 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)
{
1717
	struct sigqueue *q = __sigqueue_alloc(-1, current, GFP_KERNEL, 0);
L
Linus Torvalds 已提交
1718

1719
	if (q)
L
Linus Torvalds 已提交
1720
		q->flags |= SIGQUEUE_PREALLOC;
1721 1722

	return q;
L
Linus Torvalds 已提交
1723 1724 1725 1726 1727
}

void sigqueue_free(struct sigqueue *q)
{
	unsigned long flags;
1728 1729
	spinlock_t *lock = &current->sighand->siglock;

L
Linus Torvalds 已提交
1730 1731
	BUG_ON(!(q->flags & SIGQUEUE_PREALLOC));
	/*
1732 1733
	 * We must hold ->siglock while testing q->list
	 * to serialize with collect_signal() or with
1734
	 * __exit_signal()->flush_sigqueue().
L
Linus Torvalds 已提交
1735
	 */
1736
	spin_lock_irqsave(lock, flags);
1737 1738 1739 1740 1741
	q->flags &= ~SIGQUEUE_PREALLOC;
	/*
	 * If it is queued it will be freed when dequeued,
	 * like the "regular" sigqueue.
	 */
1742
	if (!list_empty(&q->list))
1743
		q = NULL;
1744 1745
	spin_unlock_irqrestore(lock, flags);

1746 1747
	if (q)
		__sigqueue_free(q);
L
Linus Torvalds 已提交
1748 1749
}

1750
int send_sigqueue(struct sigqueue *q, struct pid *pid, enum pid_type type)
1751
{
1752
	int sig = q->info.si_signo;
1753
	struct sigpending *pending;
1754
	struct task_struct *t;
1755
	unsigned long flags;
1756
	int ret, result;
1757

1758
	BUG_ON(!(q->flags & SIGQUEUE_PREALLOC));
1759 1760

	ret = -1;
1761 1762 1763
	rcu_read_lock();
	t = pid_task(pid, type);
	if (!t || !likely(lock_task_sighand(t, &flags)))
1764 1765
		goto ret;

1766
	ret = 1; /* the signal is ignored */
1767
	result = TRACE_SIGNAL_IGNORED;
1768
	if (!prepare_signal(sig, t, false))
1769 1770 1771
		goto out;

	ret = 0;
1772 1773 1774 1775 1776 1777 1778
	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++;
1779
		result = TRACE_SIGNAL_ALREADY_PENDING;
1780
		goto out;
1781
	}
1782
	q->info.si_overrun = 0;
1783 1784

	signalfd_notify(t, sig);
1785
	pending = (type != PIDTYPE_PID) ? &t->signal->shared_pending : &t->pending;
1786 1787
	list_add_tail(&q->list, &pending->list);
	sigaddset(&pending->signal, sig);
1788
	complete_signal(sig, t, type);
1789
	result = TRACE_SIGNAL_DELIVERED;
1790
out:
1791
	trace_signal_generate(sig, &q->info, t, type != PIDTYPE_PID, result);
1792 1793
	unlock_task_sighand(t, &flags);
ret:
1794
	rcu_read_unlock();
1795
	return ret;
1796 1797
}

L
Linus Torvalds 已提交
1798 1799 1800
/*
 * 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
1801
 *
1802 1803
 * Returns true if our parent ignored us and so we've switched to
 * self-reaping.
L
Linus Torvalds 已提交
1804
 */
1805
bool do_notify_parent(struct task_struct *tsk, int sig)
L
Linus Torvalds 已提交
1806 1807 1808 1809
{
	struct siginfo info;
	unsigned long flags;
	struct sighand_struct *psig;
1810
	bool autoreap = false;
1811
	u64 utime, stime;
L
Linus Torvalds 已提交
1812 1813 1814 1815

	BUG_ON(sig == -1);

 	/* do_notify_parent_cldstop should have been called instead.  */
M
Matthew Wilcox 已提交
1816
 	BUG_ON(task_is_stopped_or_traced(tsk));
L
Linus Torvalds 已提交
1817

T
Tejun Heo 已提交
1818
	BUG_ON(!tsk->ptrace &&
L
Linus Torvalds 已提交
1819 1820
	       (tsk->group_leader != tsk || !thread_group_empty(tsk)));

1821 1822 1823 1824 1825 1826 1827 1828 1829
	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;
	}

1830
	clear_siginfo(&info);
L
Linus Torvalds 已提交
1831 1832
	info.si_signo = sig;
	info.si_errno = 0;
1833
	/*
1834 1835
	 * We are under tasklist_lock here so our parent is tied to
	 * us and cannot change.
1836
	 *
1837 1838
	 * task_active_pid_ns will always return the same pid namespace
	 * until a task passes through release_task.
1839 1840 1841 1842 1843 1844
	 *
	 * 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();
1845
	info.si_pid = task_pid_nr_ns(tsk, task_active_pid_ns(tsk->parent));
1846 1847
	info.si_uid = from_kuid_munged(task_cred_xxx(tsk->parent, user_ns),
				       task_uid(tsk));
1848 1849
	rcu_read_unlock();

1850 1851 1852
	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
1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865

	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
1866
	if (!tsk->ptrace && sig == SIGCHLD &&
L
Linus Torvalds 已提交
1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883
	    (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).
		 */
1884
		autoreap = true;
L
Linus Torvalds 已提交
1885
		if (psig->action[SIGCHLD-1].sa.sa_handler == SIG_IGN)
1886
			sig = 0;
L
Linus Torvalds 已提交
1887
	}
1888
	if (valid_signal(sig) && sig)
L
Linus Torvalds 已提交
1889 1890 1891
		__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
1892

1893
	return autoreap;
L
Linus Torvalds 已提交
1894 1895
}

1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910
/**
 * 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
1911 1912 1913
{
	struct siginfo info;
	unsigned long flags;
1914
	struct task_struct *parent;
L
Linus Torvalds 已提交
1915
	struct sighand_struct *sighand;
1916
	u64 utime, stime;
L
Linus Torvalds 已提交
1917

1918
	if (for_ptracer) {
1919
		parent = tsk->parent;
1920
	} else {
1921 1922 1923 1924
		tsk = tsk->group_leader;
		parent = tsk->real_parent;
	}

1925
	clear_siginfo(&info);
L
Linus Torvalds 已提交
1926 1927
	info.si_signo = SIGCHLD;
	info.si_errno = 0;
1928
	/*
1929
	 * see comment in do_notify_parent() about the following 4 lines
1930 1931
	 */
	rcu_read_lock();
1932
	info.si_pid = task_pid_nr_ns(tsk, task_active_pid_ns(parent));
1933
	info.si_uid = from_kuid_munged(task_cred_xxx(parent, user_ns), task_uid(tsk));
1934 1935
	rcu_read_unlock();

1936 1937 1938
	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
1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966

 	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);
}

1967
static inline bool may_ptrace_stop(void)
1968
{
T
Tejun Heo 已提交
1969
	if (!likely(current->ptrace))
1970
		return false;
1971 1972 1973 1974 1975 1976
	/*
	 * 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
1977
	 * ->siglock we must see ->core_state != NULL. Otherwise it
1978
	 * is safe to enter schedule().
1979 1980 1981 1982
	 *
	 * 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.
1983
	 */
1984
	if (unlikely(current->mm->core_state) &&
1985
	    unlikely(current->mm == current->parent->mm))
1986
		return false;
1987

1988
	return true;
1989 1990
}

R
Roland McGrath 已提交
1991
/*
1992
 * Return non-zero if there is a SIGKILL that should be waking us up.
R
Roland McGrath 已提交
1993 1994
 * Called with the siglock held.
 */
1995
static bool sigkill_pending(struct task_struct *tsk)
R
Roland McGrath 已提交
1996
{
1997 1998
	return sigismember(&tsk->pending.signal, SIGKILL) ||
	       sigismember(&tsk->signal->shared_pending.signal, SIGKILL);
R
Roland McGrath 已提交
1999 2000
}

L
Linus Torvalds 已提交
2001 2002 2003 2004 2005 2006 2007 2008
/*
 * 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.
 *
2009 2010
 * 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
2011
 */
T
Tejun Heo 已提交
2012
static void ptrace_stop(int exit_code, int why, int clear_code, siginfo_t *info)
2013 2014
	__releases(&current->sighand->siglock)
	__acquires(&current->sighand->siglock)
L
Linus Torvalds 已提交
2015
{
2016 2017
	bool gstop_done = false;

R
Roland McGrath 已提交
2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032
	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);
2033 2034
		if (sigkill_pending(current))
			return;
R
Roland McGrath 已提交
2035 2036
	}

2037 2038
	set_special_state(TASK_TRACED);

L
Linus Torvalds 已提交
2039
	/*
2040 2041 2042 2043 2044
	 * 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.
2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055
	 *
	 *     TRACER				    TRACEE
	 *
	 *     ptrace_attach()
	 * [L]   wait_on_bit(JOBCTL_TRAPPING)	[S] set_special_state(TRACED)
	 *     do_wait()
	 *       set_current_state()                smp_wmb();
	 *       ptrace_do_wait()
	 *         wait_task_stopped()
	 *           task_stopped_code()
	 * [L]         task_is_traced()		[S] task_clear_jobctl_trapping();
L
Linus Torvalds 已提交
2056
	 */
2057
	smp_wmb();
L
Linus Torvalds 已提交
2058 2059 2060 2061

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

2062
	/*
2063 2064
	 * If @why is CLD_STOPPED, we're trapping to participate in a group
	 * stop.  Do the bookkeeping.  Note that if SIGCONT was delievered
2065 2066 2067
	 * 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.
2068
	 */
2069
	if (why == CLD_STOPPED && (current->jobctl & JOBCTL_STOP_PENDING))
2070
		gstop_done = task_participate_group_stop(current);
2071

2072
	/* any trap clears pending STOP trap, STOP trap clears NOTIFY */
2073
	task_clear_jobctl_pending(current, JOBCTL_TRAP_STOP);
2074 2075
	if (info && info->si_code >> 8 == PTRACE_EVENT_STOP)
		task_clear_jobctl_pending(current, JOBCTL_TRAP_NOTIFY);
2076

2077
	/* entering a trap, clear TRAPPING */
2078
	task_clear_jobctl_trapping(current);
2079

L
Linus Torvalds 已提交
2080 2081
	spin_unlock_irq(&current->sighand->siglock);
	read_lock(&tasklist_lock);
2082
	if (may_ptrace_stop()) {
2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093
		/*
		 * 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);
2094
		if (gstop_done && ptrace_reparented(current))
2095 2096
			do_notify_parent_cldstop(current, false, why);

M
Miklos Szeredi 已提交
2097 2098 2099 2100 2101 2102 2103
		/*
		 * 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
2104
		read_unlock(&tasklist_lock);
M
Miklos Szeredi 已提交
2105
		preempt_enable_no_resched();
2106
		freezable_schedule();
L
Linus Torvalds 已提交
2107 2108 2109
	} else {
		/*
		 * By the time we got the lock, our tracer went away.
2110
		 * Don't drop the lock yet, another tracer may come.
2111 2112 2113
		 *
		 * If @gstop_done, the ptracer went away between group stop
		 * completion and here.  During detach, it would have set
2114 2115 2116
		 * 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
2117
		 */
2118 2119 2120
		if (gstop_done)
			do_notify_parent_cldstop(current, false, why);

2121
		/* tasklist protects us from ptrace_freeze_traced() */
2122
		__set_current_state(TASK_RUNNING);
2123 2124
		if (clear_code)
			current->exit_code = 0;
2125
		read_unlock(&tasklist_lock);
L
Linus Torvalds 已提交
2126 2127 2128 2129 2130 2131 2132 2133 2134 2135
	}

	/*
	 * 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
2136 2137 2138
	/* LISTENING can be set only during STOP traps, clear it */
	current->jobctl &= ~JOBCTL_LISTENING;

L
Linus Torvalds 已提交
2139 2140 2141
	/*
	 * Queued signals ignored us while we were stopped for tracing.
	 * So check for any that we should take before resuming user mode.
2142
	 * This sets TIF_SIGPENDING, but never clears it.
L
Linus Torvalds 已提交
2143
	 */
2144
	recalc_sigpending_tsk(current);
L
Linus Torvalds 已提交
2145 2146
}

T
Tejun Heo 已提交
2147
static void ptrace_do_notify(int signr, int exit_code, int why)
L
Linus Torvalds 已提交
2148 2149 2150
{
	siginfo_t info;

2151
	clear_siginfo(&info);
T
Tejun Heo 已提交
2152
	info.si_signo = signr;
L
Linus Torvalds 已提交
2153
	info.si_code = exit_code;
2154
	info.si_pid = task_pid_vnr(current);
2155
	info.si_uid = from_kuid_munged(current_user_ns(), current_uid());
L
Linus Torvalds 已提交
2156 2157

	/* Let the debugger run.  */
T
Tejun Heo 已提交
2158 2159 2160 2161 2162 2163
	ptrace_stop(exit_code, why, 1, &info);
}

void ptrace_notify(int exit_code)
{
	BUG_ON((exit_code & (0x7f | ~0xffff)) != SIGTRAP);
2164 2165
	if (unlikely(current->task_works))
		task_work_run();
T
Tejun Heo 已提交
2166

L
Linus Torvalds 已提交
2167
	spin_lock_irq(&current->sighand->siglock);
T
Tejun Heo 已提交
2168
	ptrace_do_notify(SIGTRAP, exit_code, CLD_TRAPPED);
L
Linus Torvalds 已提交
2169 2170 2171
	spin_unlock_irq(&current->sighand->siglock);
}

2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192
/**
 * 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
2193
 */
2194 2195
static bool do_signal_stop(int signr)
	__releases(&current->sighand->siglock)
L
Linus Torvalds 已提交
2196 2197 2198
{
	struct signal_struct *sig = current->signal;

2199
	if (!(current->jobctl & JOBCTL_STOP_PENDING)) {
2200
		unsigned long gstop = JOBCTL_STOP_PENDING | JOBCTL_STOP_CONSUME;
2201 2202
		struct task_struct *t;

2203 2204
		/* signr will be recorded in task->jobctl for retries */
		WARN_ON_ONCE(signr & ~JOBCTL_STOP_SIGMASK);
2205

2206
		if (!likely(current->jobctl & JOBCTL_STOP_DEQUEUED) ||
2207
		    unlikely(signal_group_exit(sig)))
2208
			return false;
L
Linus Torvalds 已提交
2209
		/*
2210 2211 2212 2213 2214 2215 2216
		 * 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
2217 2218
		 * cause two group stops when !ptraced.  That is why we
		 * also check !task_is_stopped(t) below.
2219 2220 2221 2222 2223 2224 2225 2226
		 *
		 * 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
2227
		 */
2228 2229
		if (!(sig->flags & SIGNAL_STOP_STOPPED))
			sig->group_exit_code = signr;
L
Linus Torvalds 已提交
2230

2231 2232 2233 2234
		sig->group_stop_count = 0;

		if (task_set_jobctl_pending(current, signr | gstop))
			sig->group_stop_count++;
L
Linus Torvalds 已提交
2235

2236 2237
		t = current;
		while_each_thread(current, t) {
L
Linus Torvalds 已提交
2238
			/*
2239 2240 2241
			 * 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
2242
			 */
2243 2244
			if (!task_is_stopped(t) &&
			    task_set_jobctl_pending(t, signr | gstop)) {
2245
				sig->group_stop_count++;
2246 2247 2248 2249
				if (likely(!(t->ptrace & PT_SEIZED)))
					signal_wake_up(t, 0);
				else
					ptrace_trap_notify(t);
2250
			}
2251
		}
L
Linus Torvalds 已提交
2252
	}
2253

T
Tejun Heo 已提交
2254
	if (likely(!current->ptrace)) {
2255
		int notify = 0;
L
Linus Torvalds 已提交
2256

2257 2258 2259 2260 2261 2262 2263 2264
		/*
		 * 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;

2265
		set_special_state(TASK_STOPPED);
2266 2267
		spin_unlock_irq(&current->sighand->siglock);

2268 2269 2270 2271 2272 2273 2274 2275 2276
		/*
		 * 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.
		 */
2277 2278
		if (notify) {
			read_lock(&tasklist_lock);
2279
			do_notify_parent_cldstop(current, false, notify);
2280 2281 2282 2283
			read_unlock(&tasklist_lock);
		}

		/* Now we don't run again until woken by SIGCONT or SIGKILL */
2284
		freezable_schedule();
2285
		return true;
2286
	} else {
2287 2288 2289 2290 2291 2292
		/*
		 * 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;
2293
	}
2294
}
L
Linus Torvalds 已提交
2295

2296 2297 2298
/**
 * do_jobctl_trap - take care of ptrace jobctl traps
 *
T
Tejun Heo 已提交
2299 2300 2301 2302 2303 2304 2305
 * 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.
2306 2307 2308 2309 2310 2311 2312
 *
 * 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
2313
	struct signal_struct *signal = current->signal;
2314
	int signr = current->jobctl & JOBCTL_STOP_SIGMASK;
2315

T
Tejun Heo 已提交
2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326
	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;
2327
	}
L
Linus Torvalds 已提交
2328 2329
}

2330
static int ptrace_signal(int signr, siginfo_t *info)
R
Roland McGrath 已提交
2331
{
2332 2333 2334 2335 2336 2337 2338 2339 2340 2341
	/*
	 * 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
2342
	ptrace_stop(signr, CLD_TRAPPED, 0, info);
R
Roland McGrath 已提交
2343 2344 2345 2346 2347 2348 2349 2350

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

	current->exit_code = 0;

2351 2352 2353 2354 2355 2356
	/*
	 * 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
2357
	if (signr != info->si_signo) {
2358
		clear_siginfo(info);
R
Roland McGrath 已提交
2359 2360 2361
		info->si_signo = signr;
		info->si_errno = 0;
		info->si_code = SI_USER;
2362
		rcu_read_lock();
R
Roland McGrath 已提交
2363
		info->si_pid = task_pid_vnr(current->parent);
2364 2365
		info->si_uid = from_kuid_munged(current_user_ns(),
						task_uid(current->parent));
2366
		rcu_read_unlock();
R
Roland McGrath 已提交
2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377
	}

	/* 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;
}

2378
bool get_signal(struct ksignal *ksig)
L
Linus Torvalds 已提交
2379
{
2380 2381 2382
	struct sighand_struct *sighand = current->sighand;
	struct signal_struct *signal = current->signal;
	int signr;
L
Linus Torvalds 已提交
2383

2384 2385
	if (unlikely(current->task_works))
		task_work_run();
2386

2387
	if (unlikely(uprobe_deny_signal()))
2388
		return false;
2389

R
Roland McGrath 已提交
2390
	/*
2391 2392 2393
	 * 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
2394
	 */
2395 2396
	try_to_freeze();

2397
relock:
2398
	spin_lock_irq(&sighand->siglock);
2399 2400 2401 2402 2403
	/*
	 * 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.
	 */
2404
	if (unlikely(signal->flags & SIGNAL_CLD_MASK)) {
2405 2406 2407 2408 2409 2410 2411
		int why;

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

2412
		signal->flags &= ~SIGNAL_CLD_MASK;
2413

2414
		spin_unlock_irq(&sighand->siglock);
R
Roland McGrath 已提交
2415

2416 2417 2418 2419 2420 2421 2422 2423
		/*
		 * 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
2424
		read_lock(&tasklist_lock);
2425 2426
		do_notify_parent_cldstop(current, false, why);

2427 2428 2429
		if (ptrace_reparented(current->group_leader))
			do_notify_parent_cldstop(current->group_leader,
						true, why);
T
Tejun Heo 已提交
2430
		read_unlock(&tasklist_lock);
2431

2432 2433 2434
		goto relock;
	}

2435
	/* Has this task already been marked for death? */
2436 2437 2438 2439
	if (signal_group_exit(signal)) {
		ksig->info.si_signo = signr = SIGKILL;
		sigdelset(&current->pending.signal, SIGKILL);
		recalc_sigpending();
2440
		goto fatal;
2441
	}
2442

L
Linus Torvalds 已提交
2443 2444
	for (;;) {
		struct k_sigaction *ka;
2445

T
Tejun Heo 已提交
2446 2447
		if (unlikely(current->jobctl & JOBCTL_STOP_PENDING) &&
		    do_signal_stop(0))
2448
			goto relock;
2449

2450 2451 2452 2453 2454
		if (unlikely(current->jobctl & JOBCTL_TRAP_MASK)) {
			do_jobctl_trap();
			spin_unlock_irq(&sighand->siglock);
			goto relock;
		}
L
Linus Torvalds 已提交
2455

2456 2457 2458 2459 2460 2461 2462 2463 2464
		/*
		 * Signals generated by the execution of an instruction
		 * need to be delivered before any other pending signals
		 * so that the instruction pointer in the signal stack
		 * frame points to the faulting instruction.
		 */
		signr = dequeue_synchronous_signal(&ksig->info);
		if (!signr)
			signr = dequeue_signal(current, &current->blocked, &ksig->info);
2465

T
Tejun Heo 已提交
2466 2467
		if (!signr)
			break; /* will return 0 */
2468

2469
		if (unlikely(current->ptrace) && signr != SIGKILL) {
2470
			signr = ptrace_signal(signr, &ksig->info);
T
Tejun Heo 已提交
2471 2472
			if (!signr)
				continue;
L
Linus Torvalds 已提交
2473 2474
		}

T
Tejun Heo 已提交
2475 2476
		ka = &sighand->action[signr-1];

2477
		/* Trace actually delivered signals. */
2478
		trace_signal_deliver(signr, &ksig->info, ka);
2479

L
Linus Torvalds 已提交
2480 2481 2482 2483
		if (ka->sa.sa_handler == SIG_IGN) /* Do nothing.  */
			continue;
		if (ka->sa.sa_handler != SIG_DFL) {
			/* Run the handler.  */
2484
			ksig->ka = *ka;
L
Linus Torvalds 已提交
2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497

			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;

2498
		/*
2499
		 * Global init gets no signals it doesn't want.
2500 2501 2502 2503 2504 2505 2506
		 * 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.
2507
		 */
2508
		if (unlikely(signal->flags & SIGNAL_UNKILLABLE) &&
2509
				!sig_kernel_only(signr))
L
Linus Torvalds 已提交
2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523
			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) {
2524
				spin_unlock_irq(&sighand->siglock);
L
Linus Torvalds 已提交
2525 2526 2527

				/* signals can be posted during this window */

2528
				if (is_current_pgrp_orphaned())
L
Linus Torvalds 已提交
2529 2530
					goto relock;

2531
				spin_lock_irq(&sighand->siglock);
L
Linus Torvalds 已提交
2532 2533
			}

2534
			if (likely(do_signal_stop(ksig->info.si_signo))) {
L
Linus Torvalds 已提交
2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545
				/* It released the siglock.  */
				goto relock;
			}

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

2546
	fatal:
2547
		spin_unlock_irq(&sighand->siglock);
L
Linus Torvalds 已提交
2548 2549 2550 2551 2552

		/*
		 * Anything else is fatal, maybe with a core dump.
		 */
		current->flags |= PF_SIGNALED;
2553

L
Linus Torvalds 已提交
2554
		if (sig_kernel_coredump(signr)) {
2555
			if (print_fatal_signals)
2556
				print_fatal_signal(ksig->info.si_signo);
2557
			proc_coredump_connector(current);
L
Linus Torvalds 已提交
2558 2559 2560 2561 2562 2563 2564 2565
			/*
			 * 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.
			 */
2566
			do_coredump(&ksig->info);
L
Linus Torvalds 已提交
2567 2568 2569 2570 2571
		}

		/*
		 * Death signals, no core dump.
		 */
2572
		do_group_exit(ksig->info.si_signo);
L
Linus Torvalds 已提交
2573 2574
		/* NOTREACHED */
	}
2575
	spin_unlock_irq(&sighand->siglock);
2576 2577 2578

	ksig->sig = signr;
	return ksig->sig > 0;
L
Linus Torvalds 已提交
2579 2580
}

2581
/**
A
Al Viro 已提交
2582
 * signal_delivered - 
R
Richard Weinberger 已提交
2583
 * @ksig:		kernel signal struct
A
Al Viro 已提交
2584
 * @stepping:		nonzero if debugger single-step or block-step in use
2585
 *
2586
 * This function should be called when a signal has successfully been
R
Richard Weinberger 已提交
2587
 * delivered. It updates the blocked signals accordingly (@ksig->ka.sa.sa_mask
A
Al Viro 已提交
2588
 * is always blocked, and the signal itself is blocked unless %SA_NODEFER
R
Richard Weinberger 已提交
2589
 * is set in @ksig->ka.sa.sa_flags.  Tracing is notified.
2590
 */
R
Richard Weinberger 已提交
2591
static void signal_delivered(struct ksignal *ksig, int stepping)
2592 2593 2594
{
	sigset_t blocked;

2595 2596 2597 2598 2599 2600
	/* 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
2601 2602 2603
	sigorsets(&blocked, &current->blocked, &ksig->ka.sa.sa_mask);
	if (!(ksig->ka.sa.sa_flags & SA_NODEFER))
		sigaddset(&blocked, ksig->sig);
2604
	set_current_blocked(&blocked);
2605
	tracehook_signal_handler(stepping);
2606 2607
}

A
Al Viro 已提交
2608 2609 2610 2611 2612
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
2613
		signal_delivered(ksig, stepping);
A
Al Viro 已提交
2614 2615
}

2616 2617
/*
 * It could be that complete_signal() picked us to notify about the
2618 2619
 * group-wide signal. Other threads should be notified now to take
 * the shared signals in @which since we will not.
2620
 */
2621
static void retarget_shared_pending(struct task_struct *tsk, sigset_t *which)
2622
{
2623
	sigset_t retarget;
2624 2625
	struct task_struct *t;

2626 2627 2628 2629
	sigandsets(&retarget, &tsk->signal->shared_pending.signal, which);
	if (sigisemptyset(&retarget))
		return;

2630 2631
	t = tsk;
	while_each_thread(tsk, t) {
2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644
		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;
2645 2646 2647
	}
}

O
Oleg Nesterov 已提交
2648 2649 2650
void exit_signals(struct task_struct *tsk)
{
	int group_stop = 0;
2651
	sigset_t unblocked;
O
Oleg Nesterov 已提交
2652

2653 2654 2655 2656
	/*
	 * @tsk is about to have PF_EXITING set - lock out users which
	 * expect stable threadgroup.
	 */
2657
	cgroup_threadgroup_change_begin(tsk);
2658

2659 2660
	if (thread_group_empty(tsk) || signal_group_exit(tsk->signal)) {
		tsk->flags |= PF_EXITING;
2661
		cgroup_threadgroup_change_end(tsk);
2662
		return;
O
Oleg Nesterov 已提交
2663 2664
	}

2665
	spin_lock_irq(&tsk->sighand->siglock);
O
Oleg Nesterov 已提交
2666 2667 2668 2669 2670
	/*
	 * From now this task is not visible for group-wide signals,
	 * see wants_signal(), do_signal_stop().
	 */
	tsk->flags |= PF_EXITING;
2671

2672
	cgroup_threadgroup_change_end(tsk);
2673

2674 2675 2676
	if (!signal_pending(tsk))
		goto out;

2677 2678 2679
	unblocked = tsk->blocked;
	signotset(&unblocked);
	retarget_shared_pending(tsk, &unblocked);
2680

2681
	if (unlikely(tsk->jobctl & JOBCTL_STOP_PENDING) &&
2682
	    task_participate_group_stop(tsk))
T
Tejun Heo 已提交
2683
		group_stop = CLD_STOPPED;
2684
out:
O
Oleg Nesterov 已提交
2685 2686
	spin_unlock_irq(&tsk->sighand->siglock);

2687 2688 2689 2690
	/*
	 * If group stop has completed, deliver the notification.  This
	 * should always go to the real parent of the group leader.
	 */
2691
	if (unlikely(group_stop)) {
O
Oleg Nesterov 已提交
2692
		read_lock(&tasklist_lock);
2693
		do_notify_parent_cldstop(tsk, false, group_stop);
O
Oleg Nesterov 已提交
2694 2695 2696 2697
		read_unlock(&tasklist_lock);
	}
}

L
Linus Torvalds 已提交
2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709
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.
 */

2710 2711 2712
/**
 *  sys_restart_syscall - restart a system call
 */
2713
SYSCALL_DEFINE0(restart_syscall)
L
Linus Torvalds 已提交
2714
{
2715
	struct restart_block *restart = &current->restart_block;
L
Linus Torvalds 已提交
2716 2717 2718 2719 2720 2721 2722 2723
	return restart->fn(restart);
}

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

2724 2725 2726 2727 2728
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. */
2729
		sigandnsets(&newblocked, newset, &current->blocked);
2730 2731 2732 2733 2734 2735
		retarget_shared_pending(tsk, &newblocked);
	}
	tsk->blocked = *newset;
	recalc_sigpending();
}

2736 2737 2738 2739 2740 2741
/**
 * 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
2742
 */
2743 2744 2745
void set_current_blocked(sigset_t *newset)
{
	sigdelsetmask(newset, sigmask(SIGKILL) | sigmask(SIGSTOP));
2746
	__set_current_blocked(newset);
2747 2748 2749
}

void __set_current_blocked(const sigset_t *newset)
2750 2751 2752
{
	struct task_struct *tsk = current;

2753 2754 2755 2756 2757 2758 2759
	/*
	 * 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;

2760
	spin_lock_irq(&tsk->sighand->siglock);
2761
	__set_task_blocked(tsk, newset);
2762 2763
	spin_unlock_irq(&tsk->sighand->siglock);
}
L
Linus Torvalds 已提交
2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774

/*
 * 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)
{
2775 2776
	struct task_struct *tsk = current;
	sigset_t newset;
L
Linus Torvalds 已提交
2777

2778
	/* Lockless, only current can change ->blocked, never from irq */
2779
	if (oldset)
2780
		*oldset = tsk->blocked;
2781

L
Linus Torvalds 已提交
2782 2783
	switch (how) {
	case SIG_BLOCK:
2784
		sigorsets(&newset, &tsk->blocked, set);
L
Linus Torvalds 已提交
2785 2786
		break;
	case SIG_UNBLOCK:
2787
		sigandnsets(&newset, &tsk->blocked, set);
L
Linus Torvalds 已提交
2788 2789
		break;
	case SIG_SETMASK:
2790
		newset = *set;
L
Linus Torvalds 已提交
2791 2792
		break;
	default:
2793
		return -EINVAL;
L
Linus Torvalds 已提交
2794
	}
2795

2796
	__set_current_blocked(&newset);
2797
	return 0;
L
Linus Torvalds 已提交
2798 2799
}

2800 2801 2802
/**
 *  sys_rt_sigprocmask - change the list of currently blocked signals
 *  @how: whether to add, remove, or set signals
R
Randy Dunlap 已提交
2803
 *  @nset: stores pending signals
2804 2805 2806
 *  @oset: previous value of signal mask if non-null
 *  @sigsetsize: size of sigset_t type
 */
2807
SYSCALL_DEFINE4(rt_sigprocmask, int, how, sigset_t __user *, nset,
2808
		sigset_t __user *, oset, size_t, sigsetsize)
L
Linus Torvalds 已提交
2809 2810
{
	sigset_t old_set, new_set;
2811
	int error;
L
Linus Torvalds 已提交
2812 2813 2814

	/* XXX: Don't preclude handling different sized sigset_t's.  */
	if (sigsetsize != sizeof(sigset_t))
2815
		return -EINVAL;
L
Linus Torvalds 已提交
2816

2817 2818 2819 2820 2821
	old_set = current->blocked;

	if (nset) {
		if (copy_from_user(&new_set, nset, sizeof(sigset_t)))
			return -EFAULT;
L
Linus Torvalds 已提交
2822 2823
		sigdelsetmask(&new_set, sigmask(SIGKILL)|sigmask(SIGSTOP));

2824
		error = sigprocmask(how, &new_set, NULL);
L
Linus Torvalds 已提交
2825
		if (error)
2826 2827
			return error;
	}
L
Linus Torvalds 已提交
2828

2829 2830 2831
	if (oset) {
		if (copy_to_user(oset, &old_set, sizeof(sigset_t)))
			return -EFAULT;
L
Linus Torvalds 已提交
2832
	}
2833 2834

	return 0;
L
Linus Torvalds 已提交
2835 2836
}

A
Al Viro 已提交
2837 2838 2839
#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
2840
{
A
Al Viro 已提交
2841 2842 2843 2844 2845 2846 2847 2848 2849
	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) {
		sigset_t new_set;
		int error;
A
Al Viro 已提交
2850
		if (get_compat_sigset(&new_set, nset))
A
Al Viro 已提交
2851 2852 2853 2854 2855 2856 2857
			return -EFAULT;
		sigdelsetmask(&new_set, sigmask(SIGKILL)|sigmask(SIGSTOP));

		error = sigprocmask(how, &new_set, NULL);
		if (error)
			return error;
	}
2858
	return oset ? put_compat_sigset(oset, &old_set, sizeof(*oset)) : 0;
A
Al Viro 已提交
2859 2860
}
#endif
L
Linus Torvalds 已提交
2861

2862
static void do_sigpending(sigset_t *set)
L
Linus Torvalds 已提交
2863 2864
{
	spin_lock_irq(&current->sighand->siglock);
A
Al Viro 已提交
2865
	sigorsets(set, &current->pending.signal,
L
Linus Torvalds 已提交
2866 2867 2868 2869
		  &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
2870
	sigandsets(set, &current->blocked, set);
2871
}
L
Linus Torvalds 已提交
2872

2873 2874 2875
/**
 *  sys_rt_sigpending - examine a pending signal that has been raised
 *			while blocked
2876
 *  @uset: stores pending signals
2877 2878
 *  @sigsetsize: size of sigset_t type or larger
 */
A
Al Viro 已提交
2879
SYSCALL_DEFINE2(rt_sigpending, sigset_t __user *, uset, size_t, sigsetsize)
L
Linus Torvalds 已提交
2880
{
A
Al Viro 已提交
2881
	sigset_t set;
2882 2883 2884 2885

	if (sigsetsize > sizeof(*uset))
		return -EINVAL;

2886 2887 2888 2889 2890 2891
	do_sigpending(&set);

	if (copy_to_user(uset, &set, sigsetsize))
		return -EFAULT;

	return 0;
A
Al Viro 已提交
2892 2893 2894 2895 2896
}

#ifdef CONFIG_COMPAT
COMPAT_SYSCALL_DEFINE2(rt_sigpending, compat_sigset_t __user *, uset,
		compat_size_t, sigsetsize)
L
Linus Torvalds 已提交
2897
{
A
Al Viro 已提交
2898
	sigset_t set;
2899 2900 2901 2902

	if (sigsetsize > sizeof(*uset))
		return -EINVAL;

2903 2904 2905
	do_sigpending(&set);

	return put_compat_sigset(uset, &set, sigsetsize);
L
Linus Torvalds 已提交
2906
}
A
Al Viro 已提交
2907
#endif
L
Linus Torvalds 已提交
2908

2909
enum siginfo_layout siginfo_layout(unsigned sig, int si_code)
2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920
{
	enum siginfo_layout layout = SIL_KILL;
	if ((si_code > SI_USER) && (si_code < SI_KERNEL)) {
		static const struct {
			unsigned char limit, layout;
		} filter[] = {
			[SIGILL]  = { NSIGILL,  SIL_FAULT },
			[SIGFPE]  = { NSIGFPE,  SIL_FAULT },
			[SIGSEGV] = { NSIGSEGV, SIL_FAULT },
			[SIGBUS]  = { NSIGBUS,  SIL_FAULT },
			[SIGTRAP] = { NSIGTRAP, SIL_FAULT },
2921
#if defined(SIGEMT) && defined(NSIGEMT)
2922 2923 2924 2925 2926 2927
			[SIGEMT]  = { NSIGEMT,  SIL_FAULT },
#endif
			[SIGCHLD] = { NSIGCHLD, SIL_CHLD },
			[SIGPOLL] = { NSIGPOLL, SIL_POLL },
			[SIGSYS]  = { NSIGSYS,  SIL_SYS },
		};
2928
		if ((sig < ARRAY_SIZE(filter)) && (si_code <= filter[sig].limit)) {
2929
			layout = filter[sig].layout;
2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940
			/* Handle the exceptions */
			if ((sig == SIGBUS) &&
			    (si_code >= BUS_MCEERR_AR) && (si_code <= BUS_MCEERR_AO))
				layout = SIL_FAULT_MCEERR;
			else if ((sig == SIGSEGV) && (si_code == SEGV_BNDERR))
				layout = SIL_FAULT_BNDERR;
#ifdef SEGV_PKUERR
			else if ((sig == SIGSEGV) && (si_code == SEGV_PKUERR))
				layout = SIL_FAULT_PKUERR;
#endif
		}
2941 2942 2943 2944 2945 2946 2947 2948 2949 2950 2951 2952 2953
		else if (si_code <= NSIGPOLL)
			layout = SIL_POLL;
	} else {
		if (si_code == SI_TIMER)
			layout = SIL_TIMER;
		else if (si_code == SI_SIGIO)
			layout = SIL_POLL;
		else if (si_code < 0)
			layout = SIL_RT;
	}
	return layout;
}

A
Al Viro 已提交
2954
int copy_siginfo_to_user(siginfo_t __user *to, const siginfo_t *from)
L
Linus Torvalds 已提交
2955
{
2956
	if (copy_to_user(to, from , sizeof(struct siginfo)))
L
Linus Torvalds 已提交
2957
		return -EFAULT;
2958
	return 0;
L
Linus Torvalds 已提交
2959 2960
}

2961
#ifdef CONFIG_COMPAT
2962 2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996
int copy_siginfo_to_user32(struct compat_siginfo __user *to,
			   const struct siginfo *from)
#if defined(CONFIG_X86_X32_ABI) || defined(CONFIG_IA32_EMULATION)
{
	return __copy_siginfo_to_user32(to, from, in_x32_syscall());
}
int __copy_siginfo_to_user32(struct compat_siginfo __user *to,
			     const struct siginfo *from, bool x32_ABI)
#endif
{
	struct compat_siginfo new;
	memset(&new, 0, sizeof(new));

	new.si_signo = from->si_signo;
	new.si_errno = from->si_errno;
	new.si_code  = from->si_code;
	switch(siginfo_layout(from->si_signo, from->si_code)) {
	case SIL_KILL:
		new.si_pid = from->si_pid;
		new.si_uid = from->si_uid;
		break;
	case SIL_TIMER:
		new.si_tid     = from->si_tid;
		new.si_overrun = from->si_overrun;
		new.si_int     = from->si_int;
		break;
	case SIL_POLL:
		new.si_band = from->si_band;
		new.si_fd   = from->si_fd;
		break;
	case SIL_FAULT:
		new.si_addr = ptr_to_compat(from->si_addr);
#ifdef __ARCH_SI_TRAPNO
		new.si_trapno = from->si_trapno;
#endif
2997 2998 2999 3000 3001
		break;
	case SIL_FAULT_MCEERR:
		new.si_addr = ptr_to_compat(from->si_addr);
#ifdef __ARCH_SI_TRAPNO
		new.si_trapno = from->si_trapno;
3002
#endif
3003 3004 3005 3006 3007 3008
		new.si_addr_lsb = from->si_addr_lsb;
		break;
	case SIL_FAULT_BNDERR:
		new.si_addr = ptr_to_compat(from->si_addr);
#ifdef __ARCH_SI_TRAPNO
		new.si_trapno = from->si_trapno;
3009
#endif
3010 3011 3012 3013 3014 3015 3016
		new.si_lower = ptr_to_compat(from->si_lower);
		new.si_upper = ptr_to_compat(from->si_upper);
		break;
	case SIL_FAULT_PKUERR:
		new.si_addr = ptr_to_compat(from->si_addr);
#ifdef __ARCH_SI_TRAPNO
		new.si_trapno = from->si_trapno;
3017
#endif
3018
		new.si_pkey = from->si_pkey;
3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052
		break;
	case SIL_CHLD:
		new.si_pid    = from->si_pid;
		new.si_uid    = from->si_uid;
		new.si_status = from->si_status;
#ifdef CONFIG_X86_X32_ABI
		if (x32_ABI) {
			new._sifields._sigchld_x32._utime = from->si_utime;
			new._sifields._sigchld_x32._stime = from->si_stime;
		} else
#endif
		{
			new.si_utime = from->si_utime;
			new.si_stime = from->si_stime;
		}
		break;
	case SIL_RT:
		new.si_pid = from->si_pid;
		new.si_uid = from->si_uid;
		new.si_int = from->si_int;
		break;
	case SIL_SYS:
		new.si_call_addr = ptr_to_compat(from->si_call_addr);
		new.si_syscall   = from->si_syscall;
		new.si_arch      = from->si_arch;
		break;
	}

	if (copy_to_user(to, &new, sizeof(struct compat_siginfo)))
		return -EFAULT;

	return 0;
}

3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083
int copy_siginfo_from_user32(struct siginfo *to,
			     const struct compat_siginfo __user *ufrom)
{
	struct compat_siginfo from;

	if (copy_from_user(&from, ufrom, sizeof(struct compat_siginfo)))
		return -EFAULT;

	clear_siginfo(to);
	to->si_signo = from.si_signo;
	to->si_errno = from.si_errno;
	to->si_code  = from.si_code;
	switch(siginfo_layout(from.si_signo, from.si_code)) {
	case SIL_KILL:
		to->si_pid = from.si_pid;
		to->si_uid = from.si_uid;
		break;
	case SIL_TIMER:
		to->si_tid     = from.si_tid;
		to->si_overrun = from.si_overrun;
		to->si_int     = from.si_int;
		break;
	case SIL_POLL:
		to->si_band = from.si_band;
		to->si_fd   = from.si_fd;
		break;
	case SIL_FAULT:
		to->si_addr = compat_ptr(from.si_addr);
#ifdef __ARCH_SI_TRAPNO
		to->si_trapno = from.si_trapno;
#endif
3084 3085 3086 3087 3088
		break;
	case SIL_FAULT_MCEERR:
		to->si_addr = compat_ptr(from.si_addr);
#ifdef __ARCH_SI_TRAPNO
		to->si_trapno = from.si_trapno;
3089
#endif
3090 3091 3092 3093 3094 3095
		to->si_addr_lsb = from.si_addr_lsb;
		break;
	case SIL_FAULT_BNDERR:
		to->si_addr = compat_ptr(from.si_addr);
#ifdef __ARCH_SI_TRAPNO
		to->si_trapno = from.si_trapno;
3096
#endif
3097 3098 3099 3100 3101 3102 3103
		to->si_lower = compat_ptr(from.si_lower);
		to->si_upper = compat_ptr(from.si_upper);
		break;
	case SIL_FAULT_PKUERR:
		to->si_addr = compat_ptr(from.si_addr);
#ifdef __ARCH_SI_TRAPNO
		to->si_trapno = from.si_trapno;
3104
#endif
3105
		to->si_pkey = from.si_pkey;
3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136
		break;
	case SIL_CHLD:
		to->si_pid    = from.si_pid;
		to->si_uid    = from.si_uid;
		to->si_status = from.si_status;
#ifdef CONFIG_X86_X32_ABI
		if (in_x32_syscall()) {
			to->si_utime = from._sifields._sigchld_x32._utime;
			to->si_stime = from._sifields._sigchld_x32._stime;
		} else
#endif
		{
			to->si_utime = from.si_utime;
			to->si_stime = from.si_stime;
		}
		break;
	case SIL_RT:
		to->si_pid = from.si_pid;
		to->si_uid = from.si_uid;
		to->si_int = from.si_int;
		break;
	case SIL_SYS:
		to->si_call_addr = compat_ptr(from.si_call_addr);
		to->si_syscall   = from.si_syscall;
		to->si_arch      = from.si_arch;
		break;
	}
	return 0;
}
#endif /* CONFIG_COMPAT */

3137 3138 3139 3140 3141 3142
/**
 *  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
 */
3143
static int do_sigtimedwait(const sigset_t *which, siginfo_t *info,
3144
		    const struct timespec *ts)
3145
{
T
Thomas Gleixner 已提交
3146
	ktime_t *to = NULL, timeout = KTIME_MAX;
3147 3148
	struct task_struct *tsk = current;
	sigset_t mask = *which;
3149
	int sig, ret = 0;
3150 3151 3152 3153

	if (ts) {
		if (!timespec_valid(ts))
			return -EINVAL;
3154 3155
		timeout = timespec_to_ktime(*ts);
		to = &timeout;
3156 3157 3158 3159 3160 3161 3162 3163 3164 3165
	}

	/*
	 * 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
3166
	if (!sig && timeout) {
3167 3168 3169
		/*
		 * None ready, temporarily unblock those we're interested
		 * while we are sleeping in so that we'll be awakened when
3170 3171
		 * they arrive. Unblocking is always fine, we can avoid
		 * set_current_blocked().
3172 3173 3174 3175 3176 3177
		 */
		tsk->real_blocked = tsk->blocked;
		sigandsets(&tsk->blocked, &tsk->blocked, &mask);
		recalc_sigpending();
		spin_unlock_irq(&tsk->sighand->siglock);

3178 3179 3180
		__set_current_state(TASK_INTERRUPTIBLE);
		ret = freezable_schedule_hrtimeout_range(to, tsk->timer_slack_ns,
							 HRTIMER_MODE_REL);
3181
		spin_lock_irq(&tsk->sighand->siglock);
3182
		__set_task_blocked(tsk, &tsk->real_blocked);
3183
		sigemptyset(&tsk->real_blocked);
3184
		sig = dequeue_signal(tsk, &mask, info);
3185 3186 3187 3188 3189
	}
	spin_unlock_irq(&tsk->sighand->siglock);

	if (sig)
		return sig;
3190
	return ret ? -EINTR : -EAGAIN;
3191 3192
}

3193 3194 3195 3196 3197 3198 3199 3200
/**
 *  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
 */
3201 3202 3203
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
3204 3205 3206 3207
{
	sigset_t these;
	struct timespec ts;
	siginfo_t info;
3208
	int ret;
L
Linus Torvalds 已提交
3209 3210 3211 3212 3213 3214 3215

	/* 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;
3216

L
Linus Torvalds 已提交
3217 3218 3219 3220 3221
	if (uts) {
		if (copy_from_user(&ts, uts, sizeof(ts)))
			return -EFAULT;
	}

3222
	ret = do_sigtimedwait(&these, &info, uts ? &ts : NULL);
L
Linus Torvalds 已提交
3223

3224 3225 3226
	if (ret > 0 && uinfo) {
		if (copy_siginfo_to_user(uinfo, &info))
			ret = -EFAULT;
L
Linus Torvalds 已提交
3227 3228 3229 3230 3231
	}

	return ret;
}

3232 3233 3234 3235 3236 3237 3238 3239 3240 3241 3242 3243 3244
#ifdef CONFIG_COMPAT
COMPAT_SYSCALL_DEFINE4(rt_sigtimedwait, compat_sigset_t __user *, uthese,
		struct compat_siginfo __user *, uinfo,
		struct compat_timespec __user *, uts, compat_size_t, sigsetsize)
{
	sigset_t s;
	struct timespec t;
	siginfo_t info;
	long ret;

	if (sigsetsize != sizeof(sigset_t))
		return -EINVAL;

A
Al Viro 已提交
3245
	if (get_compat_sigset(&s, uthese))
3246 3247 3248 3249 3250 3251 3252 3253 3254 3255 3256 3257 3258 3259 3260 3261 3262 3263
		return -EFAULT;

	if (uts) {
		if (compat_get_timespec(&t, uts))
			return -EFAULT;
	}

	ret = do_sigtimedwait(&s, &info, uts ? &t : NULL);

	if (ret > 0 && uinfo) {
		if (copy_siginfo_to_user32(uinfo, &info))
			ret = -EFAULT;
	}

	return ret;
}
#endif

3264 3265 3266 3267 3268
/**
 *  sys_kill - send a signal to a process
 *  @pid: the PID of the process
 *  @sig: signal to be sent
 */
3269
SYSCALL_DEFINE2(kill, pid_t, pid, int, sig)
L
Linus Torvalds 已提交
3270 3271 3272
{
	struct siginfo info;

3273
	clear_siginfo(&info);
L
Linus Torvalds 已提交
3274 3275 3276
	info.si_signo = sig;
	info.si_errno = 0;
	info.si_code = SI_USER;
3277
	info.si_pid = task_tgid_vnr(current);
3278
	info.si_uid = from_kuid_munged(current_user_ns(), current_uid());
L
Linus Torvalds 已提交
3279 3280 3281 3282

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

T
Thomas Gleixner 已提交
3283 3284
static int
do_send_specific(pid_t tgid, pid_t pid, int sig, struct siginfo *info)
L
Linus Torvalds 已提交
3285 3286
{
	struct task_struct *p;
T
Thomas Gleixner 已提交
3287
	int error = -ESRCH;
L
Linus Torvalds 已提交
3288

3289
	rcu_read_lock();
3290
	p = find_task_by_vpid(pid);
3291
	if (p && (tgid <= 0 || task_tgid_vnr(p) == tgid)) {
T
Thomas Gleixner 已提交
3292
		error = check_kill_permission(sig, info, p);
L
Linus Torvalds 已提交
3293 3294 3295 3296
		/*
		 * The null signal is a permissions and process existence
		 * probe.  No signal is actually delivered.
		 */
3297
		if (!error && sig) {
3298
			error = do_send_sig_info(sig, info, p, PIDTYPE_PID);
3299 3300 3301 3302 3303 3304 3305
			/*
			 * 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
3306 3307
		}
	}
3308
	rcu_read_unlock();
3309

L
Linus Torvalds 已提交
3310 3311 3312
	return error;
}

T
Thomas Gleixner 已提交
3313 3314
static int do_tkill(pid_t tgid, pid_t pid, int sig)
{
3315
	struct siginfo info;
T
Thomas Gleixner 已提交
3316

3317
	clear_siginfo(&info);
T
Thomas Gleixner 已提交
3318 3319 3320 3321
	info.si_signo = sig;
	info.si_errno = 0;
	info.si_code = SI_TKILL;
	info.si_pid = task_tgid_vnr(current);
3322
	info.si_uid = from_kuid_munged(current_user_ns(), current_uid());
T
Thomas Gleixner 已提交
3323 3324 3325 3326

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

3327 3328 3329 3330 3331 3332
/**
 *  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
 *
3333
 *  This syscall also checks the @tgid and returns -ESRCH even if the PID
3334 3335 3336
 *  exists but it's not belonging to the target process anymore. This
 *  method solves the problem of threads exiting and PIDs getting reused.
 */
3337
SYSCALL_DEFINE3(tgkill, pid_t, tgid, pid_t, pid, int, sig)
3338 3339 3340 3341 3342 3343 3344 3345
{
	/* This is only valid for single tasks */
	if (pid <= 0 || tgid <= 0)
		return -EINVAL;

	return do_tkill(tgid, pid, sig);
}

3346 3347 3348 3349 3350
/**
 *  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
3351 3352
 *  Send a signal to only one task, even if it's a CLONE_THREAD task.
 */
3353
SYSCALL_DEFINE2(tkill, pid_t, pid, int, sig)
L
Linus Torvalds 已提交
3354 3355 3356 3357 3358
{
	/* This is only valid for single tasks */
	if (pid <= 0)
		return -EINVAL;

3359
	return do_tkill(0, pid, sig);
L
Linus Torvalds 已提交
3360 3361
}

A
Al Viro 已提交
3362 3363 3364 3365 3366
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.
	 */
3367
	if ((info->si_code >= 0 || info->si_code == SI_TKILL) &&
3368
	    (task_pid_vnr(current) != pid))
A
Al Viro 已提交
3369
		return -EPERM;
3370

A
Al Viro 已提交
3371 3372 3373 3374 3375 3376
	info->si_signo = sig;

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

3377 3378 3379 3380 3381 3382
/**
 *  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
 */
3383 3384
SYSCALL_DEFINE3(rt_sigqueueinfo, pid_t, pid, int, sig,
		siginfo_t __user *, uinfo)
L
Linus Torvalds 已提交
3385 3386 3387 3388
{
	siginfo_t info;
	if (copy_from_user(&info, uinfo, sizeof(siginfo_t)))
		return -EFAULT;
A
Al Viro 已提交
3389 3390
	return do_rt_sigqueueinfo(pid, sig, &info);
}
L
Linus Torvalds 已提交
3391

A
Al Viro 已提交
3392 3393 3394 3395 3396 3397
#ifdef CONFIG_COMPAT
COMPAT_SYSCALL_DEFINE3(rt_sigqueueinfo,
			compat_pid_t, pid,
			int, sig,
			struct compat_siginfo __user *, uinfo)
{
3398
	siginfo_t info;
A
Al Viro 已提交
3399 3400 3401 3402
	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
3403
}
A
Al Viro 已提交
3404
#endif
L
Linus Torvalds 已提交
3405

3406
static int do_rt_tgsigqueueinfo(pid_t tgid, pid_t pid, int sig, siginfo_t *info)
3407 3408 3409 3410 3411 3412
{
	/* 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.
3413 3414
	 * Nor can they impersonate a kill()/tgkill(), which adds source info.
	 */
3415 3416
	if ((info->si_code >= 0 || info->si_code == SI_TKILL) &&
	    (task_pid_vnr(current) != pid))
3417
		return -EPERM;
3418

3419 3420 3421 3422 3423 3424 3425 3426 3427 3428 3429 3430 3431 3432 3433 3434
	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);
}

3435 3436 3437 3438 3439 3440 3441
#ifdef CONFIG_COMPAT
COMPAT_SYSCALL_DEFINE4(rt_tgsigqueueinfo,
			compat_pid_t, tgid,
			compat_pid_t, pid,
			int, sig,
			struct compat_siginfo __user *, uinfo)
{
3442
	siginfo_t info;
3443 3444 3445 3446 3447 3448 3449

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

3450
/*
3451
 * For kthreads only, must not be used if cloned with CLONE_SIGHAND
3452
 */
3453
void kernel_sigaction(int sig, __sighandler_t action)
3454
{
3455
	spin_lock_irq(&current->sighand->siglock);
3456 3457 3458
	current->sighand->action[sig - 1].sa.sa_handler = action;
	if (action == SIG_IGN) {
		sigset_t mask;
3459

3460 3461
		sigemptyset(&mask);
		sigaddset(&mask, sig);
3462

3463 3464 3465 3466
		flush_sigqueue_mask(&mask, &current->signal->shared_pending);
		flush_sigqueue_mask(&mask, &current->pending);
		recalc_sigpending();
	}
3467 3468
	spin_unlock_irq(&current->sighand->siglock);
}
3469
EXPORT_SYMBOL(kernel_sigaction);
3470

3471 3472 3473 3474 3475
void __weak sigaction_compat_abi(struct k_sigaction *act,
		struct k_sigaction *oact)
{
}

3476
int do_sigaction(int sig, struct k_sigaction *act, struct k_sigaction *oact)
L
Linus Torvalds 已提交
3477
{
3478
	struct task_struct *p = current, *t;
L
Linus Torvalds 已提交
3479
	struct k_sigaction *k;
3480
	sigset_t mask;
L
Linus Torvalds 已提交
3481

3482
	if (!valid_signal(sig) || sig < 1 || (act && sig_kernel_only(sig)))
L
Linus Torvalds 已提交
3483 3484
		return -EINVAL;

3485
	k = &p->sighand->action[sig-1];
L
Linus Torvalds 已提交
3486

3487
	spin_lock_irq(&p->sighand->siglock);
L
Linus Torvalds 已提交
3488 3489 3490
	if (oact)
		*oact = *k;

3491 3492
	sigaction_compat_abi(act, oact);

L
Linus Torvalds 已提交
3493
	if (act) {
3494 3495
		sigdelsetmask(&act->sa.sa_mask,
			      sigmask(SIGKILL) | sigmask(SIGSTOP));
3496
		*k = *act;
L
Linus Torvalds 已提交
3497 3498 3499 3500 3501 3502 3503 3504 3505 3506 3507
		/*
		 * 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"
		 */
3508
		if (sig_handler_ignored(sig_handler(p, sig), sig)) {
3509 3510
			sigemptyset(&mask);
			sigaddset(&mask, sig);
3511 3512
			flush_sigqueue_mask(&mask, &p->signal->shared_pending);
			for_each_thread(p, t)
3513
				flush_sigqueue_mask(&mask, &t->pending);
L
Linus Torvalds 已提交
3514 3515 3516
		}
	}

3517
	spin_unlock_irq(&p->sighand->siglock);
L
Linus Torvalds 已提交
3518 3519 3520
	return 0;
}

3521
static int
3522 3523
do_sigaltstack (const stack_t *ss, stack_t *oss, unsigned long sp,
		size_t min_ss_size)
L
Linus Torvalds 已提交
3524
{
3525
	struct task_struct *t = current;
L
Linus Torvalds 已提交
3526

3527 3528 3529 3530 3531 3532 3533
	if (oss) {
		memset(oss, 0, sizeof(stack_t));
		oss->ss_sp = (void __user *) t->sas_ss_sp;
		oss->ss_size = t->sas_ss_size;
		oss->ss_flags = sas_ss_flags(sp) |
			(current->sas_ss_flags & SS_FLAG_BITS);
	}
L
Linus Torvalds 已提交
3534

3535 3536 3537 3538
	if (ss) {
		void __user *ss_sp = ss->ss_sp;
		size_t ss_size = ss->ss_size;
		unsigned ss_flags = ss->ss_flags;
3539
		int ss_mode;
L
Linus Torvalds 已提交
3540

3541 3542
		if (unlikely(on_sig_stack(sp)))
			return -EPERM;
L
Linus Torvalds 已提交
3543

3544
		ss_mode = ss_flags & ~SS_FLAG_BITS;
3545 3546 3547
		if (unlikely(ss_mode != SS_DISABLE && ss_mode != SS_ONSTACK &&
				ss_mode != 0))
			return -EINVAL;
L
Linus Torvalds 已提交
3548

3549
		if (ss_mode == SS_DISABLE) {
L
Linus Torvalds 已提交
3550 3551 3552
			ss_size = 0;
			ss_sp = NULL;
		} else {
3553
			if (unlikely(ss_size < min_ss_size))
3554
				return -ENOMEM;
L
Linus Torvalds 已提交
3555 3556
		}

3557 3558 3559
		t->sas_ss_sp = (unsigned long) ss_sp;
		t->sas_ss_size = ss_size;
		t->sas_ss_flags = ss_flags;
L
Linus Torvalds 已提交
3560
	}
3561
	return 0;
L
Linus Torvalds 已提交
3562
}
3563

3564 3565
SYSCALL_DEFINE2(sigaltstack,const stack_t __user *,uss, stack_t __user *,uoss)
{
3566 3567 3568 3569 3570
	stack_t new, old;
	int err;
	if (uss && copy_from_user(&new, uss, sizeof(stack_t)))
		return -EFAULT;
	err = do_sigaltstack(uss ? &new : NULL, uoss ? &old : NULL,
3571 3572
			      current_user_stack_pointer(),
			      MINSIGSTKSZ);
3573 3574 3575
	if (!err && uoss && copy_to_user(uoss, &old, sizeof(stack_t)))
		err = -EFAULT;
	return err;
3576
}
L
Linus Torvalds 已提交
3577

A
Al Viro 已提交
3578 3579
int restore_altstack(const stack_t __user *uss)
{
3580 3581 3582
	stack_t new;
	if (copy_from_user(&new, uss, sizeof(stack_t)))
		return -EFAULT;
3583 3584
	(void)do_sigaltstack(&new, NULL, current_user_stack_pointer(),
			     MINSIGSTKSZ);
A
Al Viro 已提交
3585
	/* squash all but EFAULT for now */
3586
	return 0;
A
Al Viro 已提交
3587 3588
}

3589 3590 3591
int __save_altstack(stack_t __user *uss, unsigned long sp)
{
	struct task_struct *t = current;
3592 3593
	int err = __put_user((void __user *)t->sas_ss_sp, &uss->ss_sp) |
		__put_user(t->sas_ss_flags, &uss->ss_flags) |
3594
		__put_user(t->sas_ss_size, &uss->ss_size);
3595 3596 3597 3598 3599
	if (err)
		return err;
	if (t->sas_ss_flags & SS_AUTODISARM)
		sas_ss_reset(t);
	return 0;
3600 3601
}

A
Al Viro 已提交
3602
#ifdef CONFIG_COMPAT
3603 3604
static int do_compat_sigaltstack(const compat_stack_t __user *uss_ptr,
				 compat_stack_t __user *uoss_ptr)
A
Al Viro 已提交
3605 3606 3607 3608 3609 3610 3611 3612 3613 3614 3615 3616
{
	stack_t uss, uoss;
	int ret;

	if (uss_ptr) {
		compat_stack_t uss32;
		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;
	}
3617
	ret = do_sigaltstack(uss_ptr ? &uss : NULL, &uoss,
3618 3619
			     compat_user_stack_pointer(),
			     COMPAT_MINSIGSTKSZ);
A
Al Viro 已提交
3620
	if (ret >= 0 && uoss_ptr)  {
3621 3622 3623 3624 3625 3626
		compat_stack_t old;
		memset(&old, 0, sizeof(old));
		old.ss_sp = ptr_to_compat(uoss.ss_sp);
		old.ss_flags = uoss.ss_flags;
		old.ss_size = uoss.ss_size;
		if (copy_to_user(uoss_ptr, &old, sizeof(compat_stack_t)))
A
Al Viro 已提交
3627 3628 3629 3630 3631
			ret = -EFAULT;
	}
	return ret;
}

3632 3633 3634 3635 3636 3637 3638
COMPAT_SYSCALL_DEFINE2(sigaltstack,
			const compat_stack_t __user *, uss_ptr,
			compat_stack_t __user *, uoss_ptr)
{
	return do_compat_sigaltstack(uss_ptr, uoss_ptr);
}

A
Al Viro 已提交
3639 3640
int compat_restore_altstack(const compat_stack_t __user *uss)
{
3641
	int err = do_compat_sigaltstack(uss, NULL);
A
Al Viro 已提交
3642 3643 3644
	/* squash all but -EFAULT for now */
	return err == -EFAULT ? err : 0;
}
3645 3646 3647

int __compat_save_altstack(compat_stack_t __user *uss, unsigned long sp)
{
3648
	int err;
3649
	struct task_struct *t = current;
3650 3651 3652
	err = __put_user(ptr_to_compat((void __user *)t->sas_ss_sp),
			 &uss->ss_sp) |
		__put_user(t->sas_ss_flags, &uss->ss_flags) |
3653
		__put_user(t->sas_ss_size, &uss->ss_size);
3654 3655 3656 3657 3658
	if (err)
		return err;
	if (t->sas_ss_flags & SS_AUTODISARM)
		sas_ss_reset(t);
	return 0;
3659
}
A
Al Viro 已提交
3660
#endif
L
Linus Torvalds 已提交
3661 3662 3663

#ifdef __ARCH_WANT_SYS_SIGPENDING

3664 3665
/**
 *  sys_sigpending - examine pending signals
3666
 *  @uset: where mask of pending signal is returned
3667
 */
3668
SYSCALL_DEFINE1(sigpending, old_sigset_t __user *, uset)
L
Linus Torvalds 已提交
3669
{
3670 3671 3672 3673 3674
	sigset_t set;

	if (sizeof(old_sigset_t) > sizeof(*uset))
		return -EINVAL;

3675 3676 3677 3678 3679 3680
	do_sigpending(&set);

	if (copy_to_user(uset, &set, sizeof(old_sigset_t)))
		return -EFAULT;

	return 0;
L
Linus Torvalds 已提交
3681 3682
}

A
Al Viro 已提交
3683 3684 3685 3686
#ifdef CONFIG_COMPAT
COMPAT_SYSCALL_DEFINE1(sigpending, compat_old_sigset_t __user *, set32)
{
	sigset_t set;
3687 3688 3689 3690

	do_sigpending(&set);

	return put_user(set.sig[0], set32);
A
Al Viro 已提交
3691 3692 3693
}
#endif

L
Linus Torvalds 已提交
3694 3695 3696
#endif

#ifdef __ARCH_WANT_SYS_SIGPROCMASK
3697 3698 3699
/**
 *  sys_sigprocmask - examine and change blocked signals
 *  @how: whether to add, remove, or set signals
3700
 *  @nset: signals to add or remove (if non-null)
3701 3702
 *  @oset: previous value of signal mask if non-null
 *
3703 3704 3705
 * Some platforms have their own version with special arguments;
 * others support only sys_rt_sigprocmask.
 */
L
Linus Torvalds 已提交
3706

3707
SYSCALL_DEFINE3(sigprocmask, int, how, old_sigset_t __user *, nset,
3708
		old_sigset_t __user *, oset)
L
Linus Torvalds 已提交
3709 3710
{
	old_sigset_t old_set, new_set;
3711
	sigset_t new_blocked;
L
Linus Torvalds 已提交
3712

3713
	old_set = current->blocked.sig[0];
L
Linus Torvalds 已提交
3714

3715 3716 3717
	if (nset) {
		if (copy_from_user(&new_set, nset, sizeof(*nset)))
			return -EFAULT;
L
Linus Torvalds 已提交
3718

3719
		new_blocked = current->blocked;
L
Linus Torvalds 已提交
3720 3721 3722

		switch (how) {
		case SIG_BLOCK:
3723
			sigaddsetmask(&new_blocked, new_set);
L
Linus Torvalds 已提交
3724 3725
			break;
		case SIG_UNBLOCK:
3726
			sigdelsetmask(&new_blocked, new_set);
L
Linus Torvalds 已提交
3727 3728
			break;
		case SIG_SETMASK:
3729
			new_blocked.sig[0] = new_set;
L
Linus Torvalds 已提交
3730
			break;
3731 3732
		default:
			return -EINVAL;
L
Linus Torvalds 已提交
3733 3734
		}

3735
		set_current_blocked(&new_blocked);
3736 3737 3738
	}

	if (oset) {
L
Linus Torvalds 已提交
3739
		if (copy_to_user(oset, &old_set, sizeof(*oset)))
3740
			return -EFAULT;
L
Linus Torvalds 已提交
3741
	}
3742 3743

	return 0;
L
Linus Torvalds 已提交
3744 3745 3746
}
#endif /* __ARCH_WANT_SYS_SIGPROCMASK */

3747
#ifndef CONFIG_ODD_RT_SIGACTION
3748 3749 3750
/**
 *  sys_rt_sigaction - alter an action taken by a process
 *  @sig: signal to be sent
3751 3752
 *  @act: new sigaction
 *  @oact: used to save the previous sigaction
3753 3754
 *  @sigsetsize: size of sigset_t type
 */
3755 3756 3757 3758
SYSCALL_DEFINE4(rt_sigaction, int, sig,
		const struct sigaction __user *, act,
		struct sigaction __user *, oact,
		size_t, sigsetsize)
L
Linus Torvalds 已提交
3759 3760
{
	struct k_sigaction new_sa, old_sa;
3761
	int ret;
L
Linus Torvalds 已提交
3762 3763 3764

	/* XXX: Don't preclude handling different sized sigset_t's.  */
	if (sigsetsize != sizeof(sigset_t))
3765
		return -EINVAL;
L
Linus Torvalds 已提交
3766

3767 3768
	if (act && copy_from_user(&new_sa.sa, act, sizeof(new_sa.sa)))
		return -EFAULT;
L
Linus Torvalds 已提交
3769 3770

	ret = do_sigaction(sig, act ? &new_sa : NULL, oact ? &old_sa : NULL);
3771 3772
	if (ret)
		return ret;
L
Linus Torvalds 已提交
3773

3774 3775 3776 3777
	if (oact && copy_to_user(oact, &old_sa.sa, sizeof(old_sa.sa)))
		return -EFAULT;

	return 0;
L
Linus Torvalds 已提交
3778
}
A
Al Viro 已提交
3779 3780 3781 3782 3783 3784 3785 3786 3787 3788 3789 3790 3791 3792 3793 3794 3795 3796 3797 3798 3799 3800 3801 3802
#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;
#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
A
Al Viro 已提交
3803
		ret |= get_compat_sigset(&new_ka.sa.sa_mask, &act->sa_mask);
3804
		ret |= get_user(new_ka.sa.sa_flags, &act->sa_flags);
A
Al Viro 已提交
3805 3806 3807 3808 3809 3810 3811 3812
		if (ret)
			return -EFAULT;
	}

	ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL);
	if (!ret && oact) {
		ret = put_user(ptr_to_compat(old_ka.sa.sa_handler), 
			       &oact->sa_handler);
3813 3814
		ret |= put_compat_sigset(&oact->sa_mask, &old_ka.sa.sa_mask,
					 sizeof(oact->sa_mask));
3815
		ret |= put_user(old_ka.sa.sa_flags, &oact->sa_flags);
A
Al Viro 已提交
3816 3817 3818 3819 3820 3821 3822 3823
#ifdef __ARCH_HAS_SA_RESTORER
		ret |= put_user(ptr_to_compat(old_ka.sa.sa_restorer),
				&oact->sa_restorer);
#endif
	}
	return ret;
}
#endif
3824
#endif /* !CONFIG_ODD_RT_SIGACTION */
L
Linus Torvalds 已提交
3825

3826 3827 3828 3829 3830 3831 3832 3833 3834 3835 3836 3837 3838 3839 3840 3841 3842 3843 3844 3845 3846 3847 3848 3849 3850 3851 3852 3853 3854 3855 3856 3857 3858 3859 3860 3861 3862 3863 3864 3865 3866 3867 3868 3869 3870 3871 3872 3873 3874 3875 3876 3877 3878 3879 3880 3881 3882 3883 3884 3885 3886 3887 3888 3889 3890 3891 3892 3893 3894 3895 3896 3897 3898 3899 3900 3901 3902
#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
3903

3904
#ifdef CONFIG_SGETMASK_SYSCALL
L
Linus Torvalds 已提交
3905 3906 3907 3908

/*
 * For backwards compatibility.  Functionality superseded by sigprocmask.
 */
3909
SYSCALL_DEFINE0(sgetmask)
L
Linus Torvalds 已提交
3910 3911 3912 3913 3914
{
	/* SMP safe */
	return current->blocked.sig[0];
}

3915
SYSCALL_DEFINE1(ssetmask, int, newmask)
L
Linus Torvalds 已提交
3916
{
3917 3918
	int old = current->blocked.sig[0];
	sigset_t newset;
L
Linus Torvalds 已提交
3919

3920
	siginitset(&newset, newmask);
3921
	set_current_blocked(&newset);
L
Linus Torvalds 已提交
3922 3923 3924

	return old;
}
3925
#endif /* CONFIG_SGETMASK_SYSCALL */
L
Linus Torvalds 已提交
3926 3927 3928 3929 3930

#ifdef __ARCH_WANT_SYS_SIGNAL
/*
 * For backwards compatibility.  Functionality superseded by sigaction.
 */
3931
SYSCALL_DEFINE2(signal, int, sig, __sighandler_t, handler)
L
Linus Torvalds 已提交
3932 3933 3934 3935 3936 3937
{
	struct k_sigaction new_sa, old_sa;
	int ret;

	new_sa.sa.sa_handler = handler;
	new_sa.sa.sa_flags = SA_ONESHOT | SA_NOMASK;
3938
	sigemptyset(&new_sa.sa.sa_mask);
L
Linus Torvalds 已提交
3939 3940 3941 3942 3943 3944 3945 3946 3947

	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

3948
SYSCALL_DEFINE0(pause)
L
Linus Torvalds 已提交
3949
{
3950
	while (!signal_pending(current)) {
3951
		__set_current_state(TASK_INTERRUPTIBLE);
3952 3953
		schedule();
	}
L
Linus Torvalds 已提交
3954 3955 3956 3957 3958
	return -ERESTARTNOHAND;
}

#endif

3959
static int sigsuspend(sigset_t *set)
A
Al Viro 已提交
3960 3961 3962 3963
{
	current->saved_sigmask = current->blocked;
	set_current_blocked(set);

3964 3965 3966 3967
	while (!signal_pending(current)) {
		__set_current_state(TASK_INTERRUPTIBLE);
		schedule();
	}
A
Al Viro 已提交
3968 3969 3970 3971
	set_restore_sigmask();
	return -ERESTARTNOHAND;
}

3972 3973 3974 3975 3976 3977
/**
 *  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
 */
3978
SYSCALL_DEFINE2(rt_sigsuspend, sigset_t __user *, unewset, size_t, sigsetsize)
3979 3980 3981 3982 3983 3984 3985 3986 3987
{
	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
3988
	return sigsuspend(&newset);
3989
}
A
Al Viro 已提交
3990 3991 3992 3993 3994 3995 3996 3997 3998 3999
 
#ifdef CONFIG_COMPAT
COMPAT_SYSCALL_DEFINE2(rt_sigsuspend, compat_sigset_t __user *, unewset, compat_size_t, sigsetsize)
{
	sigset_t newset;

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

A
Al Viro 已提交
4000
	if (get_compat_sigset(&newset, unewset))
A
Al Viro 已提交
4001 4002 4003 4004
		return -EFAULT;
	return sigsuspend(&newset);
}
#endif
4005

4006 4007 4008 4009 4010 4011 4012 4013 4014 4015 4016 4017 4018 4019 4020 4021
#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
4022

4023
__weak const char *arch_vma_name(struct vm_area_struct *vma)
4024 4025 4026 4027
{
	return NULL;
}

L
Linus Torvalds 已提交
4028 4029
void __init signals_init(void)
{
4030 4031 4032
	/* If this check fails, the __ARCH_SI_PREAMBLE_SIZE value is wrong! */
	BUILD_BUG_ON(__ARCH_SI_PREAMBLE_SIZE
		!= offsetof(struct siginfo, _sifields._pad));
4033
	BUILD_BUG_ON(sizeof(struct siginfo) != SI_MAX_SIZE);
4034

4035
	sigqueue_cachep = KMEM_CACHE(sigqueue, SLAB_PANIC);
L
Linus Torvalds 已提交
4036
}
4037 4038 4039 4040

#ifdef CONFIG_KGDB_KDB
#include <linux/kdb.h>
/*
4041
 * kdb_send_sig - Allows kdb to send signals without exposing
4042 4043 4044 4045
 * signal internals.  This function checks if the required locks are
 * available before calling the main signal code, to avoid kdb
 * deadlocks.
 */
4046
void kdb_send_sig(struct task_struct *t, int sig)
4047 4048
{
	static struct task_struct *kdb_prev_t;
4049
	int new_t, ret;
4050 4051 4052 4053 4054 4055 4056 4057 4058
	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;
	}
	new_t = kdb_prev_t != t;
	kdb_prev_t = t;
	if (t->state != TASK_RUNNING && new_t) {
4059
		spin_unlock(&t->sighand->siglock);
4060 4061 4062 4063 4064 4065 4066 4067
		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;
	}
4068
	ret = send_signal(sig, SEND_SIG_PRIV, t, PIDTYPE_PID);
4069 4070
	spin_unlock(&t->sighand->siglock);
	if (ret)
4071 4072 4073 4074 4075 4076
		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 */