signal.c 102.8 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13
/*
 *  linux/kernel/signal.c
 *
 *  Copyright (C) 1991, 1992  Linus Torvalds
 *
 *  1997-11-02  Modified for POSIX.1b signals by Richard Henderson
 *
 *  2003-06-02  Jim Houston - Concurrent Computer Corp.
 *		Changes to use preallocated sigqueue structures
 *		to allow signals to be sent reliably.
 */

#include <linux/slab.h>
14
#include <linux/export.h>
L
Linus Torvalds 已提交
15
#include <linux/init.h>
16
#include <linux/sched/mm.h>
17
#include <linux/sched/user.h>
18
#include <linux/sched/debug.h>
19
#include <linux/sched/task.h>
20
#include <linux/sched/task_stack.h>
21
#include <linux/sched/cputime.h>
L
Linus Torvalds 已提交
22 23 24
#include <linux/fs.h>
#include <linux/tty.h>
#include <linux/binfmts.h>
25
#include <linux/coredump.h>
L
Linus Torvalds 已提交
26 27 28
#include <linux/security.h>
#include <linux/syscalls.h>
#include <linux/ptrace.h>
29
#include <linux/signal.h>
30
#include <linux/signalfd.h>
31
#include <linux/ratelimit.h>
32
#include <linux/tracehook.h>
33
#include <linux/capability.h>
34
#include <linux/freezer.h>
35 36
#include <linux/pid_namespace.h>
#include <linux/nsproxy.h>
37
#include <linux/user_namespace.h>
38
#include <linux/uprobes.h>
A
Al Viro 已提交
39
#include <linux/compat.h>
40
#include <linux/cn_proc.h>
41
#include <linux/compiler.h>
42
#include <linux/posix-timers.h>
43
#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
681 682 683 684 685 686 687 688 689 690 691 692 693 694
	return signr;
}

/*
 * Tell a process that it has a new active signal..
 *
 * NOTE! we rely on the previous spin_lock to
 * lock interrupts for us! We can only be called with
 * "siglock" held, and the local interrupt must
 * have been disabled when that got acquired!
 *
 * No need to set need_resched since signal event passing
 * goes through ->blocked
 */
695
void signal_wake_up_state(struct task_struct *t, unsigned int state)
L
Linus Torvalds 已提交
696 697 698
{
	set_tsk_thread_flag(t, TIF_SIGPENDING);
	/*
699
	 * TASK_WAKEKILL also means wake it up in the stopped/traced/killable
M
Matthew Wilcox 已提交
700
	 * case. We don't check t->state here because there is a race with it
L
Linus Torvalds 已提交
701 702 703 704
	 * 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.
	 */
705
	if (!wake_up_state(t, state | TASK_INTERRUPTIBLE))
L
Linus Torvalds 已提交
706 707 708
		kick_process(t);
}

709 710 711 712 713 714
/*
 * Remove signals in mask from the pending set and queue.
 * Returns 1 if any signals were found.
 *
 * All callers must be holding the siglock.
 */
715
static void flush_sigqueue_mask(sigset_t *mask, struct sigpending *s)
716 717 718 719 720 721
{
	struct sigqueue *q, *n;
	sigset_t m;

	sigandsets(&m, mask, &s->signal);
	if (sigisemptyset(&m))
722
		return;
723

724
	sigandnsets(&s->signal, &s->signal, mask);
725 726 727 728 729 730 731
	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
732

733 734 735 736 737 738 739 740 741 742 743
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));
}

744 745 746
/*
 * called with RCU read lock from check_kill_permission()
 */
747
static bool kill_ok_by_cred(struct task_struct *t)
748 749 750 751
{
	const struct cred *cred = current_cred();
	const struct cred *tcred = __task_cred(t);

752 753 754 755 756
	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);
757 758
}

L
Linus Torvalds 已提交
759 760
/*
 * Bad permissions for sending the signal
761
 * - the caller must hold the RCU read lock
L
Linus Torvalds 已提交
762 763 764 765
 */
static int check_kill_permission(int sig, struct siginfo *info,
				 struct task_struct *t)
{
766
	struct pid *sid;
767 768
	int error;

769
	if (!valid_signal(sig))
770 771
		return -EINVAL;

772
	if (!si_fromuser(info))
773
		return 0;
A
Amy Griffis 已提交
774

775 776
	error = audit_signal_info(sig, t); /* Let audit system see the signal */
	if (error)
L
Linus Torvalds 已提交
777
		return error;
778

779
	if (!same_thread_group(current, t) &&
780
	    !kill_ok_by_cred(t)) {
781 782 783 784 785 786 787 788 789 790 791 792 793
		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;
		}
	}
794

795
	return security_task_kill(t, info, sig, NULL);
L
Linus Torvalds 已提交
796 797
}

798 799 800 801 802 803 804 805
/**
 * 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
806 807 808 809 810
 * 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.
811 812 813 814 815 816 817 818 819 820
 *
 * 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);
821
	ptrace_signal_wake_up(t, t->jobctl & JOBCTL_LISTENING);
822 823
}

L
Linus Torvalds 已提交
824
/*
825 826
 * 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
827 828
 * time regardless of blocking, ignoring, or handling.  This does the
 * actual continuing for SIGCONT, but not the actual stopping for stop
829 830 831 832
 * 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
833
 */
834
static bool prepare_signal(int sig, struct task_struct *p, bool force)
L
Linus Torvalds 已提交
835
{
836
	struct signal_struct *signal = p->signal;
L
Linus Torvalds 已提交
837
	struct task_struct *t;
838
	sigset_t flush;
L
Linus Torvalds 已提交
839

840
	if (signal->flags & (SIGNAL_GROUP_EXIT | SIGNAL_GROUP_COREDUMP)) {
841
		if (!(signal->flags & SIGNAL_GROUP_EXIT))
842
			return sig == SIGKILL;
L
Linus Torvalds 已提交
843
		/*
844
		 * The process is in the middle of dying, nothing to do.
L
Linus Torvalds 已提交
845
		 */
846
	} else if (sig_kernel_stop(sig)) {
L
Linus Torvalds 已提交
847 848 849
		/*
		 * This is a stop signal.  Remove SIGCONT from all queues.
		 */
850
		siginitset(&flush, sigmask(SIGCONT));
851
		flush_sigqueue_mask(&flush, &signal->shared_pending);
852
		for_each_thread(p, t)
853
			flush_sigqueue_mask(&flush, &t->pending);
L
Linus Torvalds 已提交
854
	} else if (sig == SIGCONT) {
855
		unsigned int why;
L
Linus Torvalds 已提交
856
		/*
857
		 * Remove all stop signals from all queues, wake all threads.
L
Linus Torvalds 已提交
858
		 */
859
		siginitset(&flush, SIG_KERNEL_STOP_MASK);
860
		flush_sigqueue_mask(&flush, &signal->shared_pending);
861
		for_each_thread(p, t) {
862
			flush_sigqueue_mask(&flush, &t->pending);
863
			task_clear_jobctl_pending(t, JOBCTL_STOP_PENDING);
864 865 866 867
			if (likely(!(t->ptrace & PT_SEIZED)))
				wake_up_state(t, __TASK_STOPPED);
			else
				ptrace_trap_notify(t);
868
		}
L
Linus Torvalds 已提交
869

870 871 872 873 874 875 876 877 878
		/*
		 * 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;
879
		if (signal->flags & SIGNAL_STOP_STOPPED)
880
			why |= SIGNAL_CLD_CONTINUED;
881
		else if (signal->group_stop_count)
882 883 884
			why |= SIGNAL_CLD_STOPPED;

		if (why) {
885
			/*
886
			 * The first thread which returns from do_signal_stop()
887 888 889
			 * will take ->siglock, notice SIGNAL_CLD_MASK, and
			 * notify its parent. See get_signal_to_deliver().
			 */
890
			signal_set_stop_flags(signal, why | SIGNAL_STOP_CONTINUED);
891 892
			signal->group_stop_count = 0;
			signal->group_exit_code = 0;
L
Linus Torvalds 已提交
893 894
		}
	}
895

896
	return !sig_ignored(p, sig, force);
L
Linus Torvalds 已提交
897 898
}

899 900 901 902 903 904 905 906
/*
 * 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.
 */
907
static inline bool wants_signal(int sig, struct task_struct *p)
908 909
{
	if (sigismember(&p->blocked, sig))
910 911
		return false;

912
	if (p->flags & PF_EXITING)
913 914
		return false;

915
	if (sig == SIGKILL)
916 917
		return true;

918
	if (task_is_stopped_or_traced(p))
919 920
		return false;

921 922 923
	return task_curr(p) || !signal_pending(p);
}

924
static void complete_signal(int sig, struct task_struct *p, enum pid_type type)
925 926 927 928 929 930 931 932 933 934 935 936
{
	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;
937
	else if ((type == PIDTYPE_PID) || thread_group_empty(p))
938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964
		/*
		 * 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.
	 */
965
	if (sig_fatal(p, sig) &&
966
	    !(signal->flags & SIGNAL_GROUP_EXIT) &&
967
	    !sigismember(&t->real_blocked, sig) &&
968
	    (sig == SIGKILL || !p->ptrace)) {
969 970 971 972 973 974 975 976 977 978 979 980 981 982 983
		/*
		 * 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 {
984
				task_clear_jobctl_pending(t, JOBCTL_PENDING_MASK);
985 986 987 988 989 990 991 992 993 994 995 996 997 998 999
				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;
}

1000
static inline bool legacy_queue(struct sigpending *signals, int sig)
1001 1002 1003 1004
{
	return (sig < SIGRTMIN) && sigismember(&signals->signal, sig);
}

1005 1006 1007 1008 1009 1010 1011 1012 1013
#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;

1014 1015 1016 1017
	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();
1018 1019 1020 1021 1022 1023 1024 1025
}
#else
static inline void userns_fixup_signal_uid(struct siginfo *info, struct task_struct *t)
{
	return;
}
#endif

1026
static int __send_signal(int sig, struct siginfo *info, struct task_struct *t,
1027
			enum pid_type type, int from_ancestor_ns)
L
Linus Torvalds 已提交
1028
{
1029
	struct sigpending *pending;
1030
	struct sigqueue *q;
1031
	int override_rlimit;
1032
	int ret = 0, result;
1033

1034
	assert_spin_locked(&t->sighand->siglock);
1035

1036
	result = TRACE_SIGNAL_IGNORED;
1037 1038
	if (!prepare_signal(sig, t,
			from_ancestor_ns || (info == SEND_SIG_FORCED)))
1039
		goto ret;
1040

1041
	pending = (type != PIDTYPE_PID) ? &t->signal->shared_pending : &t->pending;
1042 1043 1044 1045 1046
	/*
	 * 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.
	 */
1047
	result = TRACE_SIGNAL_ALREADY_PENDING;
1048
	if (legacy_queue(pending, sig))
1049 1050 1051
		goto ret;

	result = TRACE_SIGNAL_DELIVERED;
L
Linus Torvalds 已提交
1052 1053 1054 1055
	/*
	 * fast-pathed signals for kernel-internal things like SIGSTOP
	 * or SIGKILL.
	 */
1056
	if (info == SEND_SIG_FORCED)
L
Linus Torvalds 已提交
1057 1058
		goto out_set;

1059 1060 1061 1062 1063 1064 1065 1066 1067
	/*
	 * 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.
	 */
1068 1069 1070 1071 1072
	if (sig < SIGRTMIN)
		override_rlimit = (is_si_special(info) || info->si_code >= 0);
	else
		override_rlimit = 0;

1073
	q = __sigqueue_alloc(sig, t, GFP_ATOMIC, override_rlimit);
L
Linus Torvalds 已提交
1074
	if (q) {
1075
		list_add_tail(&q->list, &pending->list);
L
Linus Torvalds 已提交
1076
		switch ((unsigned long) info) {
1077
		case (unsigned long) SEND_SIG_NOINFO:
1078
			clear_siginfo(&q->info);
L
Linus Torvalds 已提交
1079 1080 1081
			q->info.si_signo = sig;
			q->info.si_errno = 0;
			q->info.si_code = SI_USER;
1082
			q->info.si_pid = task_tgid_nr_ns(current,
1083
							task_active_pid_ns(t));
1084
			q->info.si_uid = from_kuid_munged(current_user_ns(), current_uid());
L
Linus Torvalds 已提交
1085
			break;
1086
		case (unsigned long) SEND_SIG_PRIV:
1087
			clear_siginfo(&q->info);
L
Linus Torvalds 已提交
1088 1089 1090 1091 1092 1093 1094 1095
			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);
1096 1097
			if (from_ancestor_ns)
				q->info.si_pid = 0;
L
Linus Torvalds 已提交
1098 1099
			break;
		}
1100 1101 1102

		userns_fixup_signal_uid(&q->info, t);

1103
	} else if (!is_si_special(info)) {
1104 1105 1106 1107 1108 1109
		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().
			 */
1110 1111 1112
			result = TRACE_SIGNAL_OVERFLOW_FAIL;
			ret = -EAGAIN;
			goto ret;
1113 1114 1115 1116 1117
		} else {
			/*
			 * This is a silent loss of information.  We still
			 * send the signal, but the *info bits are lost.
			 */
1118
			result = TRACE_SIGNAL_LOSE_INFO;
1119
		}
L
Linus Torvalds 已提交
1120 1121 1122
	}

out_set:
1123
	signalfd_notify(t, sig);
1124
	sigaddset(&pending->signal, sig);
1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139

	/* 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);
		}
	}

1140
	complete_signal(sig, t, type);
1141
ret:
1142
	trace_signal_generate(sig, info, t, type != PIDTYPE_PID, result);
1143
	return ret;
L
Linus Torvalds 已提交
1144 1145
}

1146
static int send_signal(int sig, struct siginfo *info, struct task_struct *t,
1147
			enum pid_type type)
1148
{
1149 1150 1151
	int from_ancestor_ns = 0;

#ifdef CONFIG_PID_NS
1152 1153
	from_ancestor_ns = si_fromuser(info) &&
			   !task_pid_nr_ns(current, task_active_pid_ns(t));
1154 1155
#endif

1156
	return __send_signal(sig, info, t, type, from_ancestor_ns);
1157 1158
}

1159
static void print_fatal_signal(int signr)
I
Ingo Molnar 已提交
1160
{
1161
	struct pt_regs *regs = signal_pt_regs();
1162
	pr_info("potentially unexpected fatal signal %d.\n", signr);
I
Ingo Molnar 已提交
1163

A
Al Viro 已提交
1164
#if defined(__i386__) && !defined(__arch_um__)
1165
	pr_info("code at %08lx: ", regs->ip);
I
Ingo Molnar 已提交
1166 1167 1168 1169 1170
	{
		int i;
		for (i = 0; i < 16; i++) {
			unsigned char insn;

1171 1172
			if (get_user(insn, (unsigned char *)(regs->ip + i)))
				break;
1173
			pr_cont("%02x ", insn);
I
Ingo Molnar 已提交
1174 1175
		}
	}
1176
	pr_cont("\n");
I
Ingo Molnar 已提交
1177
#endif
1178
	preempt_disable();
I
Ingo Molnar 已提交
1179
	show_regs(regs);
1180
	preempt_enable();
I
Ingo Molnar 已提交
1181 1182 1183 1184 1185 1186 1187 1188 1189 1190
}

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
1191

1192 1193 1194
int
__group_send_sig_info(int sig, struct siginfo *info, struct task_struct *p)
{
1195
	return send_signal(sig, info, p, PIDTYPE_TGID);
1196 1197
}

L
Linus Torvalds 已提交
1198 1199 1200
static int
specific_send_sig_info(int sig, struct siginfo *info, struct task_struct *t)
{
1201
	return send_signal(sig, info, t, PIDTYPE_PID);
L
Linus Torvalds 已提交
1202 1203
}

1204
int do_send_sig_info(int sig, struct siginfo *info, struct task_struct *p,
1205
			enum pid_type type)
1206 1207 1208 1209 1210
{
	unsigned long flags;
	int ret = -ESRCH;

	if (lock_task_sighand(p, &flags)) {
1211
		ret = send_signal(sig, info, p, type);
1212 1213 1214 1215 1216 1217
		unlock_task_sighand(p, &flags);
	}

	return ret;
}

L
Linus Torvalds 已提交
1218 1219 1220
/*
 * Force a signal that the process can't ignore: if necessary
 * we unblock the signal and change any SIG_IGN to SIG_DFL.
1221 1222 1223 1224 1225
 *
 * 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.
 *
1226 1227
 * 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
1228 1229 1230 1231 1232
 */
int
force_sig_info(int sig, struct siginfo *info, struct task_struct *t)
{
	unsigned long int flags;
1233 1234
	int ret, blocked, ignored;
	struct k_sigaction *action;
L
Linus Torvalds 已提交
1235 1236

	spin_lock_irqsave(&t->sighand->siglock, flags);
1237 1238 1239 1240 1241 1242 1243
	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
1244
			recalc_sigpending_and_wake(t);
1245
		}
L
Linus Torvalds 已提交
1246
	}
1247 1248 1249 1250 1251
	/*
	 * 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)
1252
		t->signal->flags &= ~SIGNAL_UNKILLABLE;
L
Linus Torvalds 已提交
1253 1254 1255 1256 1257 1258 1259 1260 1261
	ret = specific_send_sig_info(sig, info, t);
	spin_unlock_irqrestore(&t->sighand->siglock, flags);

	return ret;
}

/*
 * Nuke all other threads in the group.
 */
1262
int zap_other_threads(struct task_struct *p)
L
Linus Torvalds 已提交
1263
{
1264 1265
	struct task_struct *t = p;
	int count = 0;
L
Linus Torvalds 已提交
1266 1267 1268

	p->signal->group_stop_count = 0;

1269
	while_each_thread(p, t) {
1270
		task_clear_jobctl_pending(t, JOBCTL_PENDING_MASK);
1271 1272 1273
		count++;

		/* Don't bother with already dead threads */
L
Linus Torvalds 已提交
1274 1275 1276 1277 1278
		if (t->exit_state)
			continue;
		sigaddset(&t->pending.signal, SIGKILL);
		signal_wake_up(t, 1);
	}
1279 1280

	return count;
L
Linus Torvalds 已提交
1281 1282
}

1283 1284
struct sighand_struct *__lock_task_sighand(struct task_struct *tsk,
					   unsigned long *flags)
1285 1286 1287
{
	struct sighand_struct *sighand;

1288
	rcu_read_lock();
1289 1290
	for (;;) {
		sighand = rcu_dereference(tsk->sighand);
1291
		if (unlikely(sighand == NULL))
1292
			break;
1293

1294 1295
		/*
		 * This sighand can be already freed and even reused, but
1296
		 * we rely on SLAB_TYPESAFE_BY_RCU and sighand_ctor() which
1297 1298 1299 1300 1301 1302 1303 1304
		 * 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.
		 */
1305 1306
		spin_lock_irqsave(&sighand->siglock, *flags);
		if (likely(sighand == tsk->sighand))
1307
			break;
1308
		spin_unlock_irqrestore(&sighand->siglock, *flags);
1309
	}
1310
	rcu_read_unlock();
1311 1312 1313 1314

	return sighand;
}

1315 1316 1317
/*
 * send signal info to all the members of a group
 */
1318 1319
int group_send_sig_info(int sig, struct siginfo *info, struct task_struct *p,
			enum pid_type type)
L
Linus Torvalds 已提交
1320
{
1321 1322 1323 1324 1325
	int ret;

	rcu_read_lock();
	ret = check_kill_permission(sig, info, p);
	rcu_read_unlock();
1326

1327
	if (!ret && sig)
1328
		ret = do_send_sig_info(sig, info, p, type);
L
Linus Torvalds 已提交
1329 1330 1331 1332 1333

	return ret;
}

/*
1334
 * __kill_pgrp_info() sends a signal to a process group: this is what the tty
L
Linus Torvalds 已提交
1335
 * control characters do (^C, ^Z etc)
1336
 * - the caller must hold at least a readlock on tasklist_lock
L
Linus Torvalds 已提交
1337
 */
1338
int __kill_pgrp_info(int sig, struct siginfo *info, struct pid *pgrp)
L
Linus Torvalds 已提交
1339 1340 1341 1342 1343 1344
{
	struct task_struct *p = NULL;
	int retval, success;

	success = 0;
	retval = -ESRCH;
1345
	do_each_pid_task(pgrp, PIDTYPE_PGID, p) {
1346
		int err = group_send_sig_info(sig, info, p, PIDTYPE_PGID);
L
Linus Torvalds 已提交
1347 1348
		success |= !err;
		retval = err;
1349
	} while_each_pid_task(pgrp, PIDTYPE_PGID, p);
L
Linus Torvalds 已提交
1350 1351 1352
	return success ? 0 : retval;
}

1353
int kill_pid_info(int sig, struct siginfo *info, struct pid *pid)
L
Linus Torvalds 已提交
1354
{
1355
	int error = -ESRCH;
L
Linus Torvalds 已提交
1356 1357
	struct task_struct *p;

1358 1359 1360 1361
	for (;;) {
		rcu_read_lock();
		p = pid_task(pid, PIDTYPE_PID);
		if (p)
1362
			error = group_send_sig_info(sig, info, p, PIDTYPE_TGID);
1363 1364 1365
		rcu_read_unlock();
		if (likely(!p || error != -ESRCH))
			return error;
1366

1367 1368 1369 1370 1371 1372
		/*
		 * 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
1373 1374
}

1375
static int kill_proc_info(int sig, struct siginfo *info, pid_t pid)
1376 1377 1378
{
	int error;
	rcu_read_lock();
1379
	error = kill_pid_info(sig, info, find_vpid(pid));
1380 1381 1382 1383
	rcu_read_unlock();
	return error;
}

1384 1385
static inline bool kill_as_cred_perm(const struct cred *cred,
				     struct task_struct *target)
1386 1387
{
	const struct cred *pcred = __task_cred(target);
1388 1389 1390 1391 1392

	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);
1393 1394
}

1395
/* like kill_pid_info(), but doesn't use uid/euid of "current" */
1396
int kill_pid_info_as_cred(int sig, struct siginfo *info, struct pid *pid,
1397
			 const struct cred *cred)
1398 1399 1400
{
	int ret = -EINVAL;
	struct task_struct *p;
1401
	unsigned long flags;
1402 1403 1404 1405

	if (!valid_signal(sig))
		return ret;

1406
	rcu_read_lock();
1407
	p = pid_task(pid, PIDTYPE_PID);
1408 1409 1410 1411
	if (!p) {
		ret = -ESRCH;
		goto out_unlock;
	}
1412
	if (si_fromuser(info) && !kill_as_cred_perm(cred, p)) {
1413 1414 1415
		ret = -EPERM;
		goto out_unlock;
	}
1416
	ret = security_task_kill(p, info, sig, cred);
1417 1418
	if (ret)
		goto out_unlock;
1419 1420 1421

	if (sig) {
		if (lock_task_sighand(p, &flags)) {
1422
			ret = __send_signal(sig, info, p, PIDTYPE_TGID, 0);
1423 1424 1425
			unlock_task_sighand(p, &flags);
		} else
			ret = -ESRCH;
1426 1427
	}
out_unlock:
1428
	rcu_read_unlock();
1429 1430
	return ret;
}
1431
EXPORT_SYMBOL_GPL(kill_pid_info_as_cred);
L
Linus Torvalds 已提交
1432 1433 1434 1435 1436 1437 1438 1439

/*
 * 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.
 */

1440
static int kill_something_info(int sig, struct siginfo *info, pid_t pid)
L
Linus Torvalds 已提交
1441
{
1442
	int ret;
1443 1444 1445 1446 1447 1448 1449 1450

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

1451 1452 1453 1454
	/* -INT_MIN is undefined.  Exclude this case to avoid a UBSAN warning */
	if (pid == INT_MIN)
		return -ESRCH;

1455 1456 1457 1458 1459
	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
1460 1461 1462 1463
		int retval = 0, count = 0;
		struct task_struct * p;

		for_each_process(p) {
1464 1465
			if (task_pid_vnr(p) > 1 &&
					!same_thread_group(p, current)) {
1466 1467
				int err = group_send_sig_info(sig, info, p,
							      PIDTYPE_MAX);
L
Linus Torvalds 已提交
1468 1469 1470 1471 1472
				++count;
				if (err != -EPERM)
					retval = err;
			}
		}
1473
		ret = count ? retval : -ESRCH;
L
Linus Torvalds 已提交
1474
	}
1475 1476
	read_unlock(&tasklist_lock);

1477
	return ret;
L
Linus Torvalds 已提交
1478 1479 1480 1481 1482 1483
}

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

1484
int send_sig_info(int sig, struct siginfo *info, struct task_struct *p)
L
Linus Torvalds 已提交
1485 1486 1487 1488 1489
{
	/*
	 * Make sure legacy kernel users don't send in bad values
	 * (normal paths check this in check_kill_permission).
	 */
1490
	if (!valid_signal(sig))
L
Linus Torvalds 已提交
1491 1492
		return -EINVAL;

1493
	return do_send_sig_info(sig, info, p, PIDTYPE_PID);
L
Linus Torvalds 已提交
1494 1495
}

1496 1497 1498
#define __si_special(priv) \
	((priv) ? SEND_SIG_PRIV : SEND_SIG_NOINFO)

L
Linus Torvalds 已提交
1499 1500 1501
int
send_sig(int sig, struct task_struct *p, int priv)
{
1502
	return send_sig_info(sig, __si_special(priv), p);
L
Linus Torvalds 已提交
1503 1504
}

1505
void force_sig(int sig, struct task_struct *p)
L
Linus Torvalds 已提交
1506
{
1507
	force_sig_info(sig, SEND_SIG_PRIV, p);
L
Linus Torvalds 已提交
1508 1509 1510 1511 1512 1513 1514 1515
}

/*
 * 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..
 */
1516
void force_sigsegv(int sig, struct task_struct *p)
L
Linus Torvalds 已提交
1517 1518 1519 1520 1521 1522 1523 1524 1525 1526
{
	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);
}

1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572
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);
}

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 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629
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
1630

1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645
/* 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);
}

1646 1647
int kill_pgrp(struct pid *pid, int sig, int priv)
{
1648 1649 1650 1651 1652 1653 1654
	int ret;

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

	return ret;
1655 1656 1657 1658 1659 1660 1661 1662 1663
}
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
1664 1665 1666 1667
/*
 * 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
1668
 * expirations or I/O completions".  In the case of POSIX Timers
L
Linus Torvalds 已提交
1669 1670 1671 1672 1673 1674
 * 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)
{
1675
	struct sigqueue *q = __sigqueue_alloc(-1, current, GFP_KERNEL, 0);
L
Linus Torvalds 已提交
1676

1677
	if (q)
L
Linus Torvalds 已提交
1678
		q->flags |= SIGQUEUE_PREALLOC;
1679 1680

	return q;
L
Linus Torvalds 已提交
1681 1682 1683 1684 1685
}

void sigqueue_free(struct sigqueue *q)
{
	unsigned long flags;
1686 1687
	spinlock_t *lock = &current->sighand->siglock;

L
Linus Torvalds 已提交
1688 1689
	BUG_ON(!(q->flags & SIGQUEUE_PREALLOC));
	/*
1690 1691
	 * We must hold ->siglock while testing q->list
	 * to serialize with collect_signal() or with
1692
	 * __exit_signal()->flush_sigqueue().
L
Linus Torvalds 已提交
1693
	 */
1694
	spin_lock_irqsave(lock, flags);
1695 1696 1697 1698 1699
	q->flags &= ~SIGQUEUE_PREALLOC;
	/*
	 * If it is queued it will be freed when dequeued,
	 * like the "regular" sigqueue.
	 */
1700
	if (!list_empty(&q->list))
1701
		q = NULL;
1702 1703
	spin_unlock_irqrestore(lock, flags);

1704 1705
	if (q)
		__sigqueue_free(q);
L
Linus Torvalds 已提交
1706 1707
}

1708
int send_sigqueue(struct sigqueue *q, struct pid *pid, enum pid_type type)
1709
{
1710
	int sig = q->info.si_signo;
1711
	struct sigpending *pending;
1712
	struct task_struct *t;
1713
	unsigned long flags;
1714
	int ret, result;
1715

1716
	BUG_ON(!(q->flags & SIGQUEUE_PREALLOC));
1717 1718

	ret = -1;
1719 1720 1721
	rcu_read_lock();
	t = pid_task(pid, type);
	if (!t || !likely(lock_task_sighand(t, &flags)))
1722 1723
		goto ret;

1724
	ret = 1; /* the signal is ignored */
1725
	result = TRACE_SIGNAL_IGNORED;
1726
	if (!prepare_signal(sig, t, false))
1727 1728 1729
		goto out;

	ret = 0;
1730 1731 1732 1733 1734 1735 1736
	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++;
1737
		result = TRACE_SIGNAL_ALREADY_PENDING;
1738
		goto out;
1739
	}
1740
	q->info.si_overrun = 0;
1741 1742

	signalfd_notify(t, sig);
1743
	pending = (type != PIDTYPE_PID) ? &t->signal->shared_pending : &t->pending;
1744 1745
	list_add_tail(&q->list, &pending->list);
	sigaddset(&pending->signal, sig);
1746
	complete_signal(sig, t, type);
1747
	result = TRACE_SIGNAL_DELIVERED;
1748
out:
1749
	trace_signal_generate(sig, &q->info, t, type != PIDTYPE_PID, result);
1750 1751
	unlock_task_sighand(t, &flags);
ret:
1752
	rcu_read_unlock();
1753
	return ret;
1754 1755
}

L
Linus Torvalds 已提交
1756 1757 1758
/*
 * 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
1759
 *
1760 1761
 * Returns true if our parent ignored us and so we've switched to
 * self-reaping.
L
Linus Torvalds 已提交
1762
 */
1763
bool do_notify_parent(struct task_struct *tsk, int sig)
L
Linus Torvalds 已提交
1764 1765 1766 1767
{
	struct siginfo info;
	unsigned long flags;
	struct sighand_struct *psig;
1768
	bool autoreap = false;
1769
	u64 utime, stime;
L
Linus Torvalds 已提交
1770 1771 1772 1773

	BUG_ON(sig == -1);

 	/* do_notify_parent_cldstop should have been called instead.  */
M
Matthew Wilcox 已提交
1774
 	BUG_ON(task_is_stopped_or_traced(tsk));
L
Linus Torvalds 已提交
1775

T
Tejun Heo 已提交
1776
	BUG_ON(!tsk->ptrace &&
L
Linus Torvalds 已提交
1777 1778
	       (tsk->group_leader != tsk || !thread_group_empty(tsk)));

1779 1780 1781 1782 1783 1784 1785 1786 1787
	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;
	}

1788
	clear_siginfo(&info);
L
Linus Torvalds 已提交
1789 1790
	info.si_signo = sig;
	info.si_errno = 0;
1791
	/*
1792 1793
	 * We are under tasklist_lock here so our parent is tied to
	 * us and cannot change.
1794
	 *
1795 1796
	 * task_active_pid_ns will always return the same pid namespace
	 * until a task passes through release_task.
1797 1798 1799 1800 1801 1802
	 *
	 * 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();
1803
	info.si_pid = task_pid_nr_ns(tsk, task_active_pid_ns(tsk->parent));
1804 1805
	info.si_uid = from_kuid_munged(task_cred_xxx(tsk->parent, user_ns),
				       task_uid(tsk));
1806 1807
	rcu_read_unlock();

1808 1809 1810
	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
1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823

	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
1824
	if (!tsk->ptrace && sig == SIGCHLD &&
L
Linus Torvalds 已提交
1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841
	    (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).
		 */
1842
		autoreap = true;
L
Linus Torvalds 已提交
1843
		if (psig->action[SIGCHLD-1].sa.sa_handler == SIG_IGN)
1844
			sig = 0;
L
Linus Torvalds 已提交
1845
	}
1846
	if (valid_signal(sig) && sig)
L
Linus Torvalds 已提交
1847 1848 1849
		__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
1850

1851
	return autoreap;
L
Linus Torvalds 已提交
1852 1853
}

1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868
/**
 * 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
1869 1870 1871
{
	struct siginfo info;
	unsigned long flags;
1872
	struct task_struct *parent;
L
Linus Torvalds 已提交
1873
	struct sighand_struct *sighand;
1874
	u64 utime, stime;
L
Linus Torvalds 已提交
1875

1876
	if (for_ptracer) {
1877
		parent = tsk->parent;
1878
	} else {
1879 1880 1881 1882
		tsk = tsk->group_leader;
		parent = tsk->real_parent;
	}

1883
	clear_siginfo(&info);
L
Linus Torvalds 已提交
1884 1885
	info.si_signo = SIGCHLD;
	info.si_errno = 0;
1886
	/*
1887
	 * see comment in do_notify_parent() about the following 4 lines
1888 1889
	 */
	rcu_read_lock();
1890
	info.si_pid = task_pid_nr_ns(tsk, task_active_pid_ns(parent));
1891
	info.si_uid = from_kuid_munged(task_cred_xxx(parent, user_ns), task_uid(tsk));
1892 1893
	rcu_read_unlock();

1894 1895 1896
	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
1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924

 	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);
}

1925
static inline bool may_ptrace_stop(void)
1926
{
T
Tejun Heo 已提交
1927
	if (!likely(current->ptrace))
1928
		return false;
1929 1930 1931 1932 1933 1934
	/*
	 * 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
1935
	 * ->siglock we must see ->core_state != NULL. Otherwise it
1936
	 * is safe to enter schedule().
1937 1938 1939 1940
	 *
	 * 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.
1941
	 */
1942
	if (unlikely(current->mm->core_state) &&
1943
	    unlikely(current->mm == current->parent->mm))
1944
		return false;
1945

1946
	return true;
1947 1948
}

R
Roland McGrath 已提交
1949
/*
1950
 * Return non-zero if there is a SIGKILL that should be waking us up.
R
Roland McGrath 已提交
1951 1952
 * Called with the siglock held.
 */
1953
static bool sigkill_pending(struct task_struct *tsk)
R
Roland McGrath 已提交
1954
{
1955 1956
	return sigismember(&tsk->pending.signal, SIGKILL) ||
	       sigismember(&tsk->signal->shared_pending.signal, SIGKILL);
R
Roland McGrath 已提交
1957 1958
}

L
Linus Torvalds 已提交
1959 1960 1961 1962 1963 1964 1965 1966
/*
 * 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.
 *
1967 1968
 * 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
1969
 */
T
Tejun Heo 已提交
1970
static void ptrace_stop(int exit_code, int why, int clear_code, siginfo_t *info)
1971 1972
	__releases(&current->sighand->siglock)
	__acquires(&current->sighand->siglock)
L
Linus Torvalds 已提交
1973
{
1974 1975
	bool gstop_done = false;

R
Roland McGrath 已提交
1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
	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);
1991 1992
		if (sigkill_pending(current))
			return;
R
Roland McGrath 已提交
1993 1994
	}

1995 1996
	set_special_state(TASK_TRACED);

L
Linus Torvalds 已提交
1997
	/*
1998 1999 2000 2001 2002
	 * 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.
2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
	 *
	 *     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
2014
	 */
2015
	smp_wmb();
L
Linus Torvalds 已提交
2016 2017 2018 2019

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

2020
	/*
2021 2022
	 * If @why is CLD_STOPPED, we're trapping to participate in a group
	 * stop.  Do the bookkeeping.  Note that if SIGCONT was delievered
2023 2024 2025
	 * 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.
2026
	 */
2027
	if (why == CLD_STOPPED && (current->jobctl & JOBCTL_STOP_PENDING))
2028
		gstop_done = task_participate_group_stop(current);
2029

2030
	/* any trap clears pending STOP trap, STOP trap clears NOTIFY */
2031
	task_clear_jobctl_pending(current, JOBCTL_TRAP_STOP);
2032 2033
	if (info && info->si_code >> 8 == PTRACE_EVENT_STOP)
		task_clear_jobctl_pending(current, JOBCTL_TRAP_NOTIFY);
2034

2035
	/* entering a trap, clear TRAPPING */
2036
	task_clear_jobctl_trapping(current);
2037

L
Linus Torvalds 已提交
2038 2039
	spin_unlock_irq(&current->sighand->siglock);
	read_lock(&tasklist_lock);
2040
	if (may_ptrace_stop()) {
2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051
		/*
		 * 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);
2052
		if (gstop_done && ptrace_reparented(current))
2053 2054
			do_notify_parent_cldstop(current, false, why);

M
Miklos Szeredi 已提交
2055 2056 2057 2058 2059 2060 2061
		/*
		 * 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
2062
		read_unlock(&tasklist_lock);
M
Miklos Szeredi 已提交
2063
		preempt_enable_no_resched();
2064
		freezable_schedule();
L
Linus Torvalds 已提交
2065 2066 2067
	} else {
		/*
		 * By the time we got the lock, our tracer went away.
2068
		 * Don't drop the lock yet, another tracer may come.
2069 2070 2071
		 *
		 * If @gstop_done, the ptracer went away between group stop
		 * completion and here.  During detach, it would have set
2072 2073 2074
		 * 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
2075
		 */
2076 2077 2078
		if (gstop_done)
			do_notify_parent_cldstop(current, false, why);

2079
		/* tasklist protects us from ptrace_freeze_traced() */
2080
		__set_current_state(TASK_RUNNING);
2081 2082
		if (clear_code)
			current->exit_code = 0;
2083
		read_unlock(&tasklist_lock);
L
Linus Torvalds 已提交
2084 2085 2086 2087 2088 2089 2090 2091 2092 2093
	}

	/*
	 * 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
2094 2095 2096
	/* LISTENING can be set only during STOP traps, clear it */
	current->jobctl &= ~JOBCTL_LISTENING;

L
Linus Torvalds 已提交
2097 2098 2099
	/*
	 * Queued signals ignored us while we were stopped for tracing.
	 * So check for any that we should take before resuming user mode.
2100
	 * This sets TIF_SIGPENDING, but never clears it.
L
Linus Torvalds 已提交
2101
	 */
2102
	recalc_sigpending_tsk(current);
L
Linus Torvalds 已提交
2103 2104
}

T
Tejun Heo 已提交
2105
static void ptrace_do_notify(int signr, int exit_code, int why)
L
Linus Torvalds 已提交
2106 2107 2108
{
	siginfo_t info;

2109
	clear_siginfo(&info);
T
Tejun Heo 已提交
2110
	info.si_signo = signr;
L
Linus Torvalds 已提交
2111
	info.si_code = exit_code;
2112
	info.si_pid = task_pid_vnr(current);
2113
	info.si_uid = from_kuid_munged(current_user_ns(), current_uid());
L
Linus Torvalds 已提交
2114 2115

	/* Let the debugger run.  */
T
Tejun Heo 已提交
2116 2117 2118 2119 2120 2121
	ptrace_stop(exit_code, why, 1, &info);
}

void ptrace_notify(int exit_code)
{
	BUG_ON((exit_code & (0x7f | ~0xffff)) != SIGTRAP);
2122 2123
	if (unlikely(current->task_works))
		task_work_run();
T
Tejun Heo 已提交
2124

L
Linus Torvalds 已提交
2125
	spin_lock_irq(&current->sighand->siglock);
T
Tejun Heo 已提交
2126
	ptrace_do_notify(SIGTRAP, exit_code, CLD_TRAPPED);
L
Linus Torvalds 已提交
2127 2128 2129
	spin_unlock_irq(&current->sighand->siglock);
}

2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150
/**
 * 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
2151
 */
2152 2153
static bool do_signal_stop(int signr)
	__releases(&current->sighand->siglock)
L
Linus Torvalds 已提交
2154 2155 2156
{
	struct signal_struct *sig = current->signal;

2157
	if (!(current->jobctl & JOBCTL_STOP_PENDING)) {
2158
		unsigned long gstop = JOBCTL_STOP_PENDING | JOBCTL_STOP_CONSUME;
2159 2160
		struct task_struct *t;

2161 2162
		/* signr will be recorded in task->jobctl for retries */
		WARN_ON_ONCE(signr & ~JOBCTL_STOP_SIGMASK);
2163

2164
		if (!likely(current->jobctl & JOBCTL_STOP_DEQUEUED) ||
2165
		    unlikely(signal_group_exit(sig)))
2166
			return false;
L
Linus Torvalds 已提交
2167
		/*
2168 2169 2170 2171 2172 2173 2174
		 * 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
2175 2176
		 * cause two group stops when !ptraced.  That is why we
		 * also check !task_is_stopped(t) below.
2177 2178 2179 2180 2181 2182 2183 2184
		 *
		 * 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
2185
		 */
2186 2187
		if (!(sig->flags & SIGNAL_STOP_STOPPED))
			sig->group_exit_code = signr;
L
Linus Torvalds 已提交
2188

2189 2190 2191 2192
		sig->group_stop_count = 0;

		if (task_set_jobctl_pending(current, signr | gstop))
			sig->group_stop_count++;
L
Linus Torvalds 已提交
2193

2194 2195
		t = current;
		while_each_thread(current, t) {
L
Linus Torvalds 已提交
2196
			/*
2197 2198 2199
			 * 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
2200
			 */
2201 2202
			if (!task_is_stopped(t) &&
			    task_set_jobctl_pending(t, signr | gstop)) {
2203
				sig->group_stop_count++;
2204 2205 2206 2207
				if (likely(!(t->ptrace & PT_SEIZED)))
					signal_wake_up(t, 0);
				else
					ptrace_trap_notify(t);
2208
			}
2209
		}
L
Linus Torvalds 已提交
2210
	}
2211

T
Tejun Heo 已提交
2212
	if (likely(!current->ptrace)) {
2213
		int notify = 0;
L
Linus Torvalds 已提交
2214

2215 2216 2217 2218 2219 2220 2221 2222
		/*
		 * 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;

2223
		set_special_state(TASK_STOPPED);
2224 2225
		spin_unlock_irq(&current->sighand->siglock);

2226 2227 2228 2229 2230 2231 2232 2233 2234
		/*
		 * 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.
		 */
2235 2236
		if (notify) {
			read_lock(&tasklist_lock);
2237
			do_notify_parent_cldstop(current, false, notify);
2238 2239 2240 2241
			read_unlock(&tasklist_lock);
		}

		/* Now we don't run again until woken by SIGCONT or SIGKILL */
2242
		freezable_schedule();
2243
		return true;
2244
	} else {
2245 2246 2247 2248 2249 2250
		/*
		 * 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;
2251
	}
2252
}
L
Linus Torvalds 已提交
2253

2254 2255 2256
/**
 * do_jobctl_trap - take care of ptrace jobctl traps
 *
T
Tejun Heo 已提交
2257 2258 2259 2260 2261 2262 2263
 * 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.
2264 2265 2266 2267 2268 2269 2270
 *
 * 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
2271
	struct signal_struct *signal = current->signal;
2272
	int signr = current->jobctl & JOBCTL_STOP_SIGMASK;
2273

T
Tejun Heo 已提交
2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284
	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;
2285
	}
L
Linus Torvalds 已提交
2286 2287
}

2288
static int ptrace_signal(int signr, siginfo_t *info)
R
Roland McGrath 已提交
2289
{
2290 2291 2292 2293 2294 2295 2296 2297 2298 2299
	/*
	 * 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
2300
	ptrace_stop(signr, CLD_TRAPPED, 0, info);
R
Roland McGrath 已提交
2301 2302 2303 2304 2305 2306 2307 2308

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

	current->exit_code = 0;

2309 2310 2311 2312 2313 2314
	/*
	 * 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
2315
	if (signr != info->si_signo) {
2316
		clear_siginfo(info);
R
Roland McGrath 已提交
2317 2318 2319
		info->si_signo = signr;
		info->si_errno = 0;
		info->si_code = SI_USER;
2320
		rcu_read_lock();
R
Roland McGrath 已提交
2321
		info->si_pid = task_pid_vnr(current->parent);
2322 2323
		info->si_uid = from_kuid_munged(current_user_ns(),
						task_uid(current->parent));
2324
		rcu_read_unlock();
R
Roland McGrath 已提交
2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335
	}

	/* 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;
}

2336
bool get_signal(struct ksignal *ksig)
L
Linus Torvalds 已提交
2337
{
2338 2339 2340
	struct sighand_struct *sighand = current->sighand;
	struct signal_struct *signal = current->signal;
	int signr;
L
Linus Torvalds 已提交
2341

2342 2343
	if (unlikely(current->task_works))
		task_work_run();
2344

2345
	if (unlikely(uprobe_deny_signal()))
2346
		return false;
2347

R
Roland McGrath 已提交
2348
	/*
2349 2350 2351
	 * 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
2352
	 */
2353 2354
	try_to_freeze();

2355
relock:
2356
	spin_lock_irq(&sighand->siglock);
2357 2358 2359 2360 2361
	/*
	 * 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.
	 */
2362
	if (unlikely(signal->flags & SIGNAL_CLD_MASK)) {
2363 2364 2365 2366 2367 2368 2369
		int why;

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

2370
		signal->flags &= ~SIGNAL_CLD_MASK;
2371

2372
		spin_unlock_irq(&sighand->siglock);
R
Roland McGrath 已提交
2373

2374 2375 2376 2377 2378 2379 2380 2381
		/*
		 * 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
2382
		read_lock(&tasklist_lock);
2383 2384
		do_notify_parent_cldstop(current, false, why);

2385 2386 2387
		if (ptrace_reparented(current->group_leader))
			do_notify_parent_cldstop(current->group_leader,
						true, why);
T
Tejun Heo 已提交
2388
		read_unlock(&tasklist_lock);
2389

2390 2391 2392
		goto relock;
	}

L
Linus Torvalds 已提交
2393 2394
	for (;;) {
		struct k_sigaction *ka;
2395

T
Tejun Heo 已提交
2396 2397
		if (unlikely(current->jobctl & JOBCTL_STOP_PENDING) &&
		    do_signal_stop(0))
2398
			goto relock;
2399

2400 2401 2402 2403 2404
		if (unlikely(current->jobctl & JOBCTL_TRAP_MASK)) {
			do_jobctl_trap();
			spin_unlock_irq(&sighand->siglock);
			goto relock;
		}
L
Linus Torvalds 已提交
2405

2406
		signr = dequeue_signal(current, &current->blocked, &ksig->info);
2407

T
Tejun Heo 已提交
2408 2409
		if (!signr)
			break; /* will return 0 */
2410

2411
		if (unlikely(current->ptrace) && signr != SIGKILL) {
2412
			signr = ptrace_signal(signr, &ksig->info);
T
Tejun Heo 已提交
2413 2414
			if (!signr)
				continue;
L
Linus Torvalds 已提交
2415 2416
		}

T
Tejun Heo 已提交
2417 2418
		ka = &sighand->action[signr-1];

2419
		/* Trace actually delivered signals. */
2420
		trace_signal_deliver(signr, &ksig->info, ka);
2421

L
Linus Torvalds 已提交
2422 2423 2424 2425
		if (ka->sa.sa_handler == SIG_IGN) /* Do nothing.  */
			continue;
		if (ka->sa.sa_handler != SIG_DFL) {
			/* Run the handler.  */
2426
			ksig->ka = *ka;
L
Linus Torvalds 已提交
2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439

			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;

2440
		/*
2441
		 * Global init gets no signals it doesn't want.
2442 2443 2444 2445 2446 2447 2448
		 * 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.
2449
		 */
2450
		if (unlikely(signal->flags & SIGNAL_UNKILLABLE) &&
2451
				!sig_kernel_only(signr))
L
Linus Torvalds 已提交
2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465
			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) {
2466
				spin_unlock_irq(&sighand->siglock);
L
Linus Torvalds 已提交
2467 2468 2469

				/* signals can be posted during this window */

2470
				if (is_current_pgrp_orphaned())
L
Linus Torvalds 已提交
2471 2472
					goto relock;

2473
				spin_lock_irq(&sighand->siglock);
L
Linus Torvalds 已提交
2474 2475
			}

2476
			if (likely(do_signal_stop(ksig->info.si_signo))) {
L
Linus Torvalds 已提交
2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487
				/* It released the siglock.  */
				goto relock;
			}

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

2488
		spin_unlock_irq(&sighand->siglock);
L
Linus Torvalds 已提交
2489 2490 2491 2492 2493

		/*
		 * Anything else is fatal, maybe with a core dump.
		 */
		current->flags |= PF_SIGNALED;
2494

L
Linus Torvalds 已提交
2495
		if (sig_kernel_coredump(signr)) {
2496
			if (print_fatal_signals)
2497
				print_fatal_signal(ksig->info.si_signo);
2498
			proc_coredump_connector(current);
L
Linus Torvalds 已提交
2499 2500 2501 2502 2503 2504 2505 2506
			/*
			 * 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.
			 */
2507
			do_coredump(&ksig->info);
L
Linus Torvalds 已提交
2508 2509 2510 2511 2512
		}

		/*
		 * Death signals, no core dump.
		 */
2513
		do_group_exit(ksig->info.si_signo);
L
Linus Torvalds 已提交
2514 2515
		/* NOTREACHED */
	}
2516
	spin_unlock_irq(&sighand->siglock);
2517 2518 2519

	ksig->sig = signr;
	return ksig->sig > 0;
L
Linus Torvalds 已提交
2520 2521
}

2522
/**
A
Al Viro 已提交
2523
 * signal_delivered - 
R
Richard Weinberger 已提交
2524
 * @ksig:		kernel signal struct
A
Al Viro 已提交
2525
 * @stepping:		nonzero if debugger single-step or block-step in use
2526
 *
2527
 * This function should be called when a signal has successfully been
R
Richard Weinberger 已提交
2528
 * delivered. It updates the blocked signals accordingly (@ksig->ka.sa.sa_mask
A
Al Viro 已提交
2529
 * is always blocked, and the signal itself is blocked unless %SA_NODEFER
R
Richard Weinberger 已提交
2530
 * is set in @ksig->ka.sa.sa_flags.  Tracing is notified.
2531
 */
R
Richard Weinberger 已提交
2532
static void signal_delivered(struct ksignal *ksig, int stepping)
2533 2534 2535
{
	sigset_t blocked;

2536 2537 2538 2539 2540 2541
	/* 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
2542 2543 2544
	sigorsets(&blocked, &current->blocked, &ksig->ka.sa.sa_mask);
	if (!(ksig->ka.sa.sa_flags & SA_NODEFER))
		sigaddset(&blocked, ksig->sig);
2545
	set_current_blocked(&blocked);
2546
	tracehook_signal_handler(stepping);
2547 2548
}

A
Al Viro 已提交
2549 2550 2551 2552 2553
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
2554
		signal_delivered(ksig, stepping);
A
Al Viro 已提交
2555 2556
}

2557 2558
/*
 * It could be that complete_signal() picked us to notify about the
2559 2560
 * group-wide signal. Other threads should be notified now to take
 * the shared signals in @which since we will not.
2561
 */
2562
static void retarget_shared_pending(struct task_struct *tsk, sigset_t *which)
2563
{
2564
	sigset_t retarget;
2565 2566
	struct task_struct *t;

2567 2568 2569 2570
	sigandsets(&retarget, &tsk->signal->shared_pending.signal, which);
	if (sigisemptyset(&retarget))
		return;

2571 2572
	t = tsk;
	while_each_thread(tsk, t) {
2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585
		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;
2586 2587 2588
	}
}

O
Oleg Nesterov 已提交
2589 2590 2591
void exit_signals(struct task_struct *tsk)
{
	int group_stop = 0;
2592
	sigset_t unblocked;
O
Oleg Nesterov 已提交
2593

2594 2595 2596 2597
	/*
	 * @tsk is about to have PF_EXITING set - lock out users which
	 * expect stable threadgroup.
	 */
2598
	cgroup_threadgroup_change_begin(tsk);
2599

2600 2601
	if (thread_group_empty(tsk) || signal_group_exit(tsk->signal)) {
		tsk->flags |= PF_EXITING;
2602
		cgroup_threadgroup_change_end(tsk);
2603
		return;
O
Oleg Nesterov 已提交
2604 2605
	}

2606
	spin_lock_irq(&tsk->sighand->siglock);
O
Oleg Nesterov 已提交
2607 2608 2609 2610 2611
	/*
	 * From now this task is not visible for group-wide signals,
	 * see wants_signal(), do_signal_stop().
	 */
	tsk->flags |= PF_EXITING;
2612

2613
	cgroup_threadgroup_change_end(tsk);
2614

2615 2616 2617
	if (!signal_pending(tsk))
		goto out;

2618 2619 2620
	unblocked = tsk->blocked;
	signotset(&unblocked);
	retarget_shared_pending(tsk, &unblocked);
2621

2622
	if (unlikely(tsk->jobctl & JOBCTL_STOP_PENDING) &&
2623
	    task_participate_group_stop(tsk))
T
Tejun Heo 已提交
2624
		group_stop = CLD_STOPPED;
2625
out:
O
Oleg Nesterov 已提交
2626 2627
	spin_unlock_irq(&tsk->sighand->siglock);

2628 2629 2630 2631
	/*
	 * If group stop has completed, deliver the notification.  This
	 * should always go to the real parent of the group leader.
	 */
2632
	if (unlikely(group_stop)) {
O
Oleg Nesterov 已提交
2633
		read_lock(&tasklist_lock);
2634
		do_notify_parent_cldstop(tsk, false, group_stop);
O
Oleg Nesterov 已提交
2635 2636 2637 2638
		read_unlock(&tasklist_lock);
	}
}

L
Linus Torvalds 已提交
2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650
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.
 */

2651 2652 2653
/**
 *  sys_restart_syscall - restart a system call
 */
2654
SYSCALL_DEFINE0(restart_syscall)
L
Linus Torvalds 已提交
2655
{
2656
	struct restart_block *restart = &current->restart_block;
L
Linus Torvalds 已提交
2657 2658 2659 2660 2661 2662 2663 2664
	return restart->fn(restart);
}

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

2665 2666 2667 2668 2669
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. */
2670
		sigandnsets(&newblocked, newset, &current->blocked);
2671 2672 2673 2674 2675 2676
		retarget_shared_pending(tsk, &newblocked);
	}
	tsk->blocked = *newset;
	recalc_sigpending();
}

2677 2678 2679 2680 2681 2682
/**
 * 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
2683
 */
2684 2685 2686
void set_current_blocked(sigset_t *newset)
{
	sigdelsetmask(newset, sigmask(SIGKILL) | sigmask(SIGSTOP));
2687
	__set_current_blocked(newset);
2688 2689 2690
}

void __set_current_blocked(const sigset_t *newset)
2691 2692 2693
{
	struct task_struct *tsk = current;

2694 2695 2696 2697 2698 2699 2700
	/*
	 * 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;

2701
	spin_lock_irq(&tsk->sighand->siglock);
2702
	__set_task_blocked(tsk, newset);
2703 2704
	spin_unlock_irq(&tsk->sighand->siglock);
}
L
Linus Torvalds 已提交
2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715

/*
 * 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)
{
2716 2717
	struct task_struct *tsk = current;
	sigset_t newset;
L
Linus Torvalds 已提交
2718

2719
	/* Lockless, only current can change ->blocked, never from irq */
2720
	if (oldset)
2721
		*oldset = tsk->blocked;
2722

L
Linus Torvalds 已提交
2723 2724
	switch (how) {
	case SIG_BLOCK:
2725
		sigorsets(&newset, &tsk->blocked, set);
L
Linus Torvalds 已提交
2726 2727
		break;
	case SIG_UNBLOCK:
2728
		sigandnsets(&newset, &tsk->blocked, set);
L
Linus Torvalds 已提交
2729 2730
		break;
	case SIG_SETMASK:
2731
		newset = *set;
L
Linus Torvalds 已提交
2732 2733
		break;
	default:
2734
		return -EINVAL;
L
Linus Torvalds 已提交
2735
	}
2736

2737
	__set_current_blocked(&newset);
2738
	return 0;
L
Linus Torvalds 已提交
2739 2740
}

2741 2742 2743
/**
 *  sys_rt_sigprocmask - change the list of currently blocked signals
 *  @how: whether to add, remove, or set signals
R
Randy Dunlap 已提交
2744
 *  @nset: stores pending signals
2745 2746 2747
 *  @oset: previous value of signal mask if non-null
 *  @sigsetsize: size of sigset_t type
 */
2748
SYSCALL_DEFINE4(rt_sigprocmask, int, how, sigset_t __user *, nset,
2749
		sigset_t __user *, oset, size_t, sigsetsize)
L
Linus Torvalds 已提交
2750 2751
{
	sigset_t old_set, new_set;
2752
	int error;
L
Linus Torvalds 已提交
2753 2754 2755

	/* XXX: Don't preclude handling different sized sigset_t's.  */
	if (sigsetsize != sizeof(sigset_t))
2756
		return -EINVAL;
L
Linus Torvalds 已提交
2757

2758 2759 2760 2761 2762
	old_set = current->blocked;

	if (nset) {
		if (copy_from_user(&new_set, nset, sizeof(sigset_t)))
			return -EFAULT;
L
Linus Torvalds 已提交
2763 2764
		sigdelsetmask(&new_set, sigmask(SIGKILL)|sigmask(SIGSTOP));

2765
		error = sigprocmask(how, &new_set, NULL);
L
Linus Torvalds 已提交
2766
		if (error)
2767 2768
			return error;
	}
L
Linus Torvalds 已提交
2769

2770 2771 2772
	if (oset) {
		if (copy_to_user(oset, &old_set, sizeof(sigset_t)))
			return -EFAULT;
L
Linus Torvalds 已提交
2773
	}
2774 2775

	return 0;
L
Linus Torvalds 已提交
2776 2777
}

A
Al Viro 已提交
2778 2779 2780
#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
2781
{
A
Al Viro 已提交
2782 2783 2784 2785 2786 2787 2788 2789 2790
	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
2791
		if (get_compat_sigset(&new_set, nset))
A
Al Viro 已提交
2792 2793 2794 2795 2796 2797 2798
			return -EFAULT;
		sigdelsetmask(&new_set, sigmask(SIGKILL)|sigmask(SIGSTOP));

		error = sigprocmask(how, &new_set, NULL);
		if (error)
			return error;
	}
2799
	return oset ? put_compat_sigset(oset, &old_set, sizeof(*oset)) : 0;
A
Al Viro 已提交
2800 2801
}
#endif
L
Linus Torvalds 已提交
2802

2803
static void do_sigpending(sigset_t *set)
L
Linus Torvalds 已提交
2804 2805
{
	spin_lock_irq(&current->sighand->siglock);
A
Al Viro 已提交
2806
	sigorsets(set, &current->pending.signal,
L
Linus Torvalds 已提交
2807 2808 2809 2810
		  &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
2811
	sigandsets(set, &current->blocked, set);
2812
}
L
Linus Torvalds 已提交
2813

2814 2815 2816
/**
 *  sys_rt_sigpending - examine a pending signal that has been raised
 *			while blocked
2817
 *  @uset: stores pending signals
2818 2819
 *  @sigsetsize: size of sigset_t type or larger
 */
A
Al Viro 已提交
2820
SYSCALL_DEFINE2(rt_sigpending, sigset_t __user *, uset, size_t, sigsetsize)
L
Linus Torvalds 已提交
2821
{
A
Al Viro 已提交
2822
	sigset_t set;
2823 2824 2825 2826

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

2827 2828 2829 2830 2831 2832
	do_sigpending(&set);

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

	return 0;
A
Al Viro 已提交
2833 2834 2835 2836 2837
}

#ifdef CONFIG_COMPAT
COMPAT_SYSCALL_DEFINE2(rt_sigpending, compat_sigset_t __user *, uset,
		compat_size_t, sigsetsize)
L
Linus Torvalds 已提交
2838
{
A
Al Viro 已提交
2839
	sigset_t set;
2840 2841 2842 2843

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

2844 2845 2846
	do_sigpending(&set);

	return put_compat_sigset(uset, &set, sigsetsize);
L
Linus Torvalds 已提交
2847
}
A
Al Viro 已提交
2848
#endif
L
Linus Torvalds 已提交
2849

2850 2851 2852 2853 2854 2855 2856 2857 2858 2859 2860 2861
enum siginfo_layout siginfo_layout(int sig, int si_code)
{
	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 },
2862
#if defined(SIGEMT) && defined(NSIGEMT)
2863 2864 2865 2866 2867 2868
			[SIGEMT]  = { NSIGEMT,  SIL_FAULT },
#endif
			[SIGCHLD] = { NSIGCHLD, SIL_CHLD },
			[SIGPOLL] = { NSIGPOLL, SIL_POLL },
			[SIGSYS]  = { NSIGSYS,  SIL_SYS },
		};
2869
		if ((sig < ARRAY_SIZE(filter)) && (si_code <= filter[sig].limit)) {
2870
			layout = filter[sig].layout;
2871 2872 2873 2874 2875 2876 2877 2878 2879 2880 2881
			/* 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
		}
2882 2883 2884 2885 2886 2887 2888 2889 2890 2891 2892 2893 2894
		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
2895
int copy_siginfo_to_user(siginfo_t __user *to, const siginfo_t *from)
L
Linus Torvalds 已提交
2896
{
2897
	if (copy_to_user(to, from , sizeof(struct siginfo)))
L
Linus Torvalds 已提交
2898
		return -EFAULT;
2899
	return 0;
L
Linus Torvalds 已提交
2900 2901
}

2902
#ifdef CONFIG_COMPAT
2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 2937
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
2938 2939 2940 2941 2942
		break;
	case SIL_FAULT_MCEERR:
		new.si_addr = ptr_to_compat(from->si_addr);
#ifdef __ARCH_SI_TRAPNO
		new.si_trapno = from->si_trapno;
2943
#endif
2944 2945 2946 2947 2948 2949
		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;
2950
#endif
2951 2952 2953 2954 2955 2956 2957
		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;
2958
#endif
2959
		new.si_pkey = from->si_pkey;
2960 2961 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
		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;
}

2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024
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
3025 3026 3027 3028 3029
		break;
	case SIL_FAULT_MCEERR:
		to->si_addr = compat_ptr(from.si_addr);
#ifdef __ARCH_SI_TRAPNO
		to->si_trapno = from.si_trapno;
3030
#endif
3031 3032 3033 3034 3035 3036
		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;
3037
#endif
3038 3039 3040 3041 3042 3043 3044
		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;
3045
#endif
3046
		to->si_pkey = from.si_pkey;
3047 3048 3049 3050 3051 3052 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
		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 */

3078 3079 3080 3081 3082 3083
/**
 *  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
 */
3084
static int do_sigtimedwait(const sigset_t *which, siginfo_t *info,
3085
		    const struct timespec *ts)
3086
{
T
Thomas Gleixner 已提交
3087
	ktime_t *to = NULL, timeout = KTIME_MAX;
3088 3089
	struct task_struct *tsk = current;
	sigset_t mask = *which;
3090
	int sig, ret = 0;
3091 3092 3093 3094

	if (ts) {
		if (!timespec_valid(ts))
			return -EINVAL;
3095 3096
		timeout = timespec_to_ktime(*ts);
		to = &timeout;
3097 3098 3099 3100 3101 3102 3103 3104 3105 3106
	}

	/*
	 * 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
3107
	if (!sig && timeout) {
3108 3109 3110
		/*
		 * None ready, temporarily unblock those we're interested
		 * while we are sleeping in so that we'll be awakened when
3111 3112
		 * they arrive. Unblocking is always fine, we can avoid
		 * set_current_blocked().
3113 3114 3115 3116 3117 3118
		 */
		tsk->real_blocked = tsk->blocked;
		sigandsets(&tsk->blocked, &tsk->blocked, &mask);
		recalc_sigpending();
		spin_unlock_irq(&tsk->sighand->siglock);

3119 3120 3121
		__set_current_state(TASK_INTERRUPTIBLE);
		ret = freezable_schedule_hrtimeout_range(to, tsk->timer_slack_ns,
							 HRTIMER_MODE_REL);
3122
		spin_lock_irq(&tsk->sighand->siglock);
3123
		__set_task_blocked(tsk, &tsk->real_blocked);
3124
		sigemptyset(&tsk->real_blocked);
3125
		sig = dequeue_signal(tsk, &mask, info);
3126 3127 3128 3129 3130
	}
	spin_unlock_irq(&tsk->sighand->siglock);

	if (sig)
		return sig;
3131
	return ret ? -EINTR : -EAGAIN;
3132 3133
}

3134 3135 3136 3137 3138 3139 3140 3141
/**
 *  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
 */
3142 3143 3144
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
3145 3146 3147 3148
{
	sigset_t these;
	struct timespec ts;
	siginfo_t info;
3149
	int ret;
L
Linus Torvalds 已提交
3150 3151 3152 3153 3154 3155 3156

	/* 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;
3157

L
Linus Torvalds 已提交
3158 3159 3160 3161 3162
	if (uts) {
		if (copy_from_user(&ts, uts, sizeof(ts)))
			return -EFAULT;
	}

3163
	ret = do_sigtimedwait(&these, &info, uts ? &ts : NULL);
L
Linus Torvalds 已提交
3164

3165 3166 3167
	if (ret > 0 && uinfo) {
		if (copy_siginfo_to_user(uinfo, &info))
			ret = -EFAULT;
L
Linus Torvalds 已提交
3168 3169 3170 3171 3172
	}

	return ret;
}

3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 3184 3185
#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
3186
	if (get_compat_sigset(&s, uthese))
3187 3188 3189 3190 3191 3192 3193 3194 3195 3196 3197 3198 3199 3200 3201 3202 3203 3204
		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

3205 3206 3207 3208 3209
/**
 *  sys_kill - send a signal to a process
 *  @pid: the PID of the process
 *  @sig: signal to be sent
 */
3210
SYSCALL_DEFINE2(kill, pid_t, pid, int, sig)
L
Linus Torvalds 已提交
3211 3212 3213
{
	struct siginfo info;

3214
	clear_siginfo(&info);
L
Linus Torvalds 已提交
3215 3216 3217
	info.si_signo = sig;
	info.si_errno = 0;
	info.si_code = SI_USER;
3218
	info.si_pid = task_tgid_vnr(current);
3219
	info.si_uid = from_kuid_munged(current_user_ns(), current_uid());
L
Linus Torvalds 已提交
3220 3221 3222 3223

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

T
Thomas Gleixner 已提交
3224 3225
static int
do_send_specific(pid_t tgid, pid_t pid, int sig, struct siginfo *info)
L
Linus Torvalds 已提交
3226 3227
{
	struct task_struct *p;
T
Thomas Gleixner 已提交
3228
	int error = -ESRCH;
L
Linus Torvalds 已提交
3229

3230
	rcu_read_lock();
3231
	p = find_task_by_vpid(pid);
3232
	if (p && (tgid <= 0 || task_tgid_vnr(p) == tgid)) {
T
Thomas Gleixner 已提交
3233
		error = check_kill_permission(sig, info, p);
L
Linus Torvalds 已提交
3234 3235 3236 3237
		/*
		 * The null signal is a permissions and process existence
		 * probe.  No signal is actually delivered.
		 */
3238
		if (!error && sig) {
3239
			error = do_send_sig_info(sig, info, p, PIDTYPE_PID);
3240 3241 3242 3243 3244 3245 3246
			/*
			 * 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
3247 3248
		}
	}
3249
	rcu_read_unlock();
3250

L
Linus Torvalds 已提交
3251 3252 3253
	return error;
}

T
Thomas Gleixner 已提交
3254 3255
static int do_tkill(pid_t tgid, pid_t pid, int sig)
{
3256
	struct siginfo info;
T
Thomas Gleixner 已提交
3257

3258
	clear_siginfo(&info);
T
Thomas Gleixner 已提交
3259 3260 3261 3262
	info.si_signo = sig;
	info.si_errno = 0;
	info.si_code = SI_TKILL;
	info.si_pid = task_tgid_vnr(current);
3263
	info.si_uid = from_kuid_munged(current_user_ns(), current_uid());
T
Thomas Gleixner 已提交
3264 3265 3266 3267

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

3268 3269 3270 3271 3272 3273
/**
 *  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
 *
3274
 *  This syscall also checks the @tgid and returns -ESRCH even if the PID
3275 3276 3277
 *  exists but it's not belonging to the target process anymore. This
 *  method solves the problem of threads exiting and PIDs getting reused.
 */
3278
SYSCALL_DEFINE3(tgkill, pid_t, tgid, pid_t, pid, int, sig)
3279 3280 3281 3282 3283 3284 3285 3286
{
	/* This is only valid for single tasks */
	if (pid <= 0 || tgid <= 0)
		return -EINVAL;

	return do_tkill(tgid, pid, sig);
}

3287 3288 3289 3290 3291
/**
 *  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
3292 3293
 *  Send a signal to only one task, even if it's a CLONE_THREAD task.
 */
3294
SYSCALL_DEFINE2(tkill, pid_t, pid, int, sig)
L
Linus Torvalds 已提交
3295 3296 3297 3298 3299
{
	/* This is only valid for single tasks */
	if (pid <= 0)
		return -EINVAL;

3300
	return do_tkill(0, pid, sig);
L
Linus Torvalds 已提交
3301 3302
}

A
Al Viro 已提交
3303 3304 3305 3306 3307
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.
	 */
3308
	if ((info->si_code >= 0 || info->si_code == SI_TKILL) &&
3309
	    (task_pid_vnr(current) != pid))
A
Al Viro 已提交
3310
		return -EPERM;
3311

A
Al Viro 已提交
3312 3313 3314 3315 3316 3317
	info->si_signo = sig;

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

3318 3319 3320 3321 3322 3323
/**
 *  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
 */
3324 3325
SYSCALL_DEFINE3(rt_sigqueueinfo, pid_t, pid, int, sig,
		siginfo_t __user *, uinfo)
L
Linus Torvalds 已提交
3326 3327 3328 3329
{
	siginfo_t info;
	if (copy_from_user(&info, uinfo, sizeof(siginfo_t)))
		return -EFAULT;
A
Al Viro 已提交
3330 3331
	return do_rt_sigqueueinfo(pid, sig, &info);
}
L
Linus Torvalds 已提交
3332

A
Al Viro 已提交
3333 3334 3335 3336 3337 3338
#ifdef CONFIG_COMPAT
COMPAT_SYSCALL_DEFINE3(rt_sigqueueinfo,
			compat_pid_t, pid,
			int, sig,
			struct compat_siginfo __user *, uinfo)
{
3339
	siginfo_t info;
A
Al Viro 已提交
3340 3341 3342 3343
	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
3344
}
A
Al Viro 已提交
3345
#endif
L
Linus Torvalds 已提交
3346

3347
static int do_rt_tgsigqueueinfo(pid_t tgid, pid_t pid, int sig, siginfo_t *info)
3348 3349 3350 3351 3352 3353
{
	/* 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.
3354 3355
	 * Nor can they impersonate a kill()/tgkill(), which adds source info.
	 */
3356 3357
	if ((info->si_code >= 0 || info->si_code == SI_TKILL) &&
	    (task_pid_vnr(current) != pid))
3358
		return -EPERM;
3359

3360 3361 3362 3363 3364 3365 3366 3367 3368 3369 3370 3371 3372 3373 3374 3375
	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);
}

3376 3377 3378 3379 3380 3381 3382
#ifdef CONFIG_COMPAT
COMPAT_SYSCALL_DEFINE4(rt_tgsigqueueinfo,
			compat_pid_t, tgid,
			compat_pid_t, pid,
			int, sig,
			struct compat_siginfo __user *, uinfo)
{
3383
	siginfo_t info;
3384 3385 3386 3387 3388 3389 3390

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

3391
/*
3392
 * For kthreads only, must not be used if cloned with CLONE_SIGHAND
3393
 */
3394
void kernel_sigaction(int sig, __sighandler_t action)
3395
{
3396
	spin_lock_irq(&current->sighand->siglock);
3397 3398 3399
	current->sighand->action[sig - 1].sa.sa_handler = action;
	if (action == SIG_IGN) {
		sigset_t mask;
3400

3401 3402
		sigemptyset(&mask);
		sigaddset(&mask, sig);
3403

3404 3405 3406 3407
		flush_sigqueue_mask(&mask, &current->signal->shared_pending);
		flush_sigqueue_mask(&mask, &current->pending);
		recalc_sigpending();
	}
3408 3409
	spin_unlock_irq(&current->sighand->siglock);
}
3410
EXPORT_SYMBOL(kernel_sigaction);
3411

3412 3413 3414 3415 3416
void __weak sigaction_compat_abi(struct k_sigaction *act,
		struct k_sigaction *oact)
{
}

3417
int do_sigaction(int sig, struct k_sigaction *act, struct k_sigaction *oact)
L
Linus Torvalds 已提交
3418
{
3419
	struct task_struct *p = current, *t;
L
Linus Torvalds 已提交
3420
	struct k_sigaction *k;
3421
	sigset_t mask;
L
Linus Torvalds 已提交
3422

3423
	if (!valid_signal(sig) || sig < 1 || (act && sig_kernel_only(sig)))
L
Linus Torvalds 已提交
3424 3425
		return -EINVAL;

3426
	k = &p->sighand->action[sig-1];
L
Linus Torvalds 已提交
3427

3428
	spin_lock_irq(&p->sighand->siglock);
L
Linus Torvalds 已提交
3429 3430 3431
	if (oact)
		*oact = *k;

3432 3433
	sigaction_compat_abi(act, oact);

L
Linus Torvalds 已提交
3434
	if (act) {
3435 3436
		sigdelsetmask(&act->sa.sa_mask,
			      sigmask(SIGKILL) | sigmask(SIGSTOP));
3437
		*k = *act;
L
Linus Torvalds 已提交
3438 3439 3440 3441 3442 3443 3444 3445 3446 3447 3448
		/*
		 * 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"
		 */
3449
		if (sig_handler_ignored(sig_handler(p, sig), sig)) {
3450 3451
			sigemptyset(&mask);
			sigaddset(&mask, sig);
3452 3453
			flush_sigqueue_mask(&mask, &p->signal->shared_pending);
			for_each_thread(p, t)
3454
				flush_sigqueue_mask(&mask, &t->pending);
L
Linus Torvalds 已提交
3455 3456 3457
		}
	}

3458
	spin_unlock_irq(&p->sighand->siglock);
L
Linus Torvalds 已提交
3459 3460 3461
	return 0;
}

3462
static int
3463 3464
do_sigaltstack (const stack_t *ss, stack_t *oss, unsigned long sp,
		size_t min_ss_size)
L
Linus Torvalds 已提交
3465
{
3466
	struct task_struct *t = current;
L
Linus Torvalds 已提交
3467

3468 3469 3470 3471 3472 3473 3474
	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
3475

3476 3477 3478 3479
	if (ss) {
		void __user *ss_sp = ss->ss_sp;
		size_t ss_size = ss->ss_size;
		unsigned ss_flags = ss->ss_flags;
3480
		int ss_mode;
L
Linus Torvalds 已提交
3481

3482 3483
		if (unlikely(on_sig_stack(sp)))
			return -EPERM;
L
Linus Torvalds 已提交
3484

3485
		ss_mode = ss_flags & ~SS_FLAG_BITS;
3486 3487 3488
		if (unlikely(ss_mode != SS_DISABLE && ss_mode != SS_ONSTACK &&
				ss_mode != 0))
			return -EINVAL;
L
Linus Torvalds 已提交
3489

3490
		if (ss_mode == SS_DISABLE) {
L
Linus Torvalds 已提交
3491 3492 3493
			ss_size = 0;
			ss_sp = NULL;
		} else {
3494
			if (unlikely(ss_size < min_ss_size))
3495
				return -ENOMEM;
L
Linus Torvalds 已提交
3496 3497
		}

3498 3499 3500
		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
3501
	}
3502
	return 0;
L
Linus Torvalds 已提交
3503
}
3504

3505 3506
SYSCALL_DEFINE2(sigaltstack,const stack_t __user *,uss, stack_t __user *,uoss)
{
3507 3508 3509 3510 3511
	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,
3512 3513
			      current_user_stack_pointer(),
			      MINSIGSTKSZ);
3514 3515 3516
	if (!err && uoss && copy_to_user(uoss, &old, sizeof(stack_t)))
		err = -EFAULT;
	return err;
3517
}
L
Linus Torvalds 已提交
3518

A
Al Viro 已提交
3519 3520
int restore_altstack(const stack_t __user *uss)
{
3521 3522 3523
	stack_t new;
	if (copy_from_user(&new, uss, sizeof(stack_t)))
		return -EFAULT;
3524 3525
	(void)do_sigaltstack(&new, NULL, current_user_stack_pointer(),
			     MINSIGSTKSZ);
A
Al Viro 已提交
3526
	/* squash all but EFAULT for now */
3527
	return 0;
A
Al Viro 已提交
3528 3529
}

3530 3531 3532
int __save_altstack(stack_t __user *uss, unsigned long sp)
{
	struct task_struct *t = current;
3533 3534
	int err = __put_user((void __user *)t->sas_ss_sp, &uss->ss_sp) |
		__put_user(t->sas_ss_flags, &uss->ss_flags) |
3535
		__put_user(t->sas_ss_size, &uss->ss_size);
3536 3537 3538 3539 3540
	if (err)
		return err;
	if (t->sas_ss_flags & SS_AUTODISARM)
		sas_ss_reset(t);
	return 0;
3541 3542
}

A
Al Viro 已提交
3543
#ifdef CONFIG_COMPAT
3544 3545
static int do_compat_sigaltstack(const compat_stack_t __user *uss_ptr,
				 compat_stack_t __user *uoss_ptr)
A
Al Viro 已提交
3546 3547 3548 3549 3550 3551 3552 3553 3554 3555 3556 3557
{
	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;
	}
3558
	ret = do_sigaltstack(uss_ptr ? &uss : NULL, &uoss,
3559 3560
			     compat_user_stack_pointer(),
			     COMPAT_MINSIGSTKSZ);
A
Al Viro 已提交
3561
	if (ret >= 0 && uoss_ptr)  {
3562 3563 3564 3565 3566 3567
		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
3568 3569 3570 3571 3572
			ret = -EFAULT;
	}
	return ret;
}

3573 3574 3575 3576 3577 3578 3579
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
3580 3581
int compat_restore_altstack(const compat_stack_t __user *uss)
{
3582
	int err = do_compat_sigaltstack(uss, NULL);
A
Al Viro 已提交
3583 3584 3585
	/* squash all but -EFAULT for now */
	return err == -EFAULT ? err : 0;
}
3586 3587 3588

int __compat_save_altstack(compat_stack_t __user *uss, unsigned long sp)
{
3589
	int err;
3590
	struct task_struct *t = current;
3591 3592 3593
	err = __put_user(ptr_to_compat((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
}
A
Al Viro 已提交
3601
#endif
L
Linus Torvalds 已提交
3602 3603 3604

#ifdef __ARCH_WANT_SYS_SIGPENDING

3605 3606
/**
 *  sys_sigpending - examine pending signals
3607
 *  @uset: where mask of pending signal is returned
3608
 */
3609
SYSCALL_DEFINE1(sigpending, old_sigset_t __user *, uset)
L
Linus Torvalds 已提交
3610
{
3611 3612 3613 3614 3615
	sigset_t set;

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

3616 3617 3618 3619 3620 3621
	do_sigpending(&set);

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

	return 0;
L
Linus Torvalds 已提交
3622 3623
}

A
Al Viro 已提交
3624 3625 3626 3627
#ifdef CONFIG_COMPAT
COMPAT_SYSCALL_DEFINE1(sigpending, compat_old_sigset_t __user *, set32)
{
	sigset_t set;
3628 3629 3630 3631

	do_sigpending(&set);

	return put_user(set.sig[0], set32);
A
Al Viro 已提交
3632 3633 3634
}
#endif

L
Linus Torvalds 已提交
3635 3636 3637
#endif

#ifdef __ARCH_WANT_SYS_SIGPROCMASK
3638 3639 3640
/**
 *  sys_sigprocmask - examine and change blocked signals
 *  @how: whether to add, remove, or set signals
3641
 *  @nset: signals to add or remove (if non-null)
3642 3643
 *  @oset: previous value of signal mask if non-null
 *
3644 3645 3646
 * Some platforms have their own version with special arguments;
 * others support only sys_rt_sigprocmask.
 */
L
Linus Torvalds 已提交
3647

3648
SYSCALL_DEFINE3(sigprocmask, int, how, old_sigset_t __user *, nset,
3649
		old_sigset_t __user *, oset)
L
Linus Torvalds 已提交
3650 3651
{
	old_sigset_t old_set, new_set;
3652
	sigset_t new_blocked;
L
Linus Torvalds 已提交
3653

3654
	old_set = current->blocked.sig[0];
L
Linus Torvalds 已提交
3655

3656 3657 3658
	if (nset) {
		if (copy_from_user(&new_set, nset, sizeof(*nset)))
			return -EFAULT;
L
Linus Torvalds 已提交
3659

3660
		new_blocked = current->blocked;
L
Linus Torvalds 已提交
3661 3662 3663

		switch (how) {
		case SIG_BLOCK:
3664
			sigaddsetmask(&new_blocked, new_set);
L
Linus Torvalds 已提交
3665 3666
			break;
		case SIG_UNBLOCK:
3667
			sigdelsetmask(&new_blocked, new_set);
L
Linus Torvalds 已提交
3668 3669
			break;
		case SIG_SETMASK:
3670
			new_blocked.sig[0] = new_set;
L
Linus Torvalds 已提交
3671
			break;
3672 3673
		default:
			return -EINVAL;
L
Linus Torvalds 已提交
3674 3675
		}

3676
		set_current_blocked(&new_blocked);
3677 3678 3679
	}

	if (oset) {
L
Linus Torvalds 已提交
3680
		if (copy_to_user(oset, &old_set, sizeof(*oset)))
3681
			return -EFAULT;
L
Linus Torvalds 已提交
3682
	}
3683 3684

	return 0;
L
Linus Torvalds 已提交
3685 3686 3687
}
#endif /* __ARCH_WANT_SYS_SIGPROCMASK */

3688
#ifndef CONFIG_ODD_RT_SIGACTION
3689 3690 3691
/**
 *  sys_rt_sigaction - alter an action taken by a process
 *  @sig: signal to be sent
3692 3693
 *  @act: new sigaction
 *  @oact: used to save the previous sigaction
3694 3695
 *  @sigsetsize: size of sigset_t type
 */
3696 3697 3698 3699
SYSCALL_DEFINE4(rt_sigaction, int, sig,
		const struct sigaction __user *, act,
		struct sigaction __user *, oact,
		size_t, sigsetsize)
L
Linus Torvalds 已提交
3700 3701
{
	struct k_sigaction new_sa, old_sa;
3702
	int ret;
L
Linus Torvalds 已提交
3703 3704 3705

	/* XXX: Don't preclude handling different sized sigset_t's.  */
	if (sigsetsize != sizeof(sigset_t))
3706
		return -EINVAL;
L
Linus Torvalds 已提交
3707

3708 3709
	if (act && copy_from_user(&new_sa.sa, act, sizeof(new_sa.sa)))
		return -EFAULT;
L
Linus Torvalds 已提交
3710 3711

	ret = do_sigaction(sig, act ? &new_sa : NULL, oact ? &old_sa : NULL);
3712 3713
	if (ret)
		return ret;
L
Linus Torvalds 已提交
3714

3715 3716 3717 3718
	if (oact && copy_to_user(oact, &old_sa.sa, sizeof(old_sa.sa)))
		return -EFAULT;

	return 0;
L
Linus Torvalds 已提交
3719
}
A
Al Viro 已提交
3720 3721 3722 3723 3724 3725 3726 3727 3728 3729 3730 3731 3732 3733 3734 3735 3736 3737 3738 3739 3740 3741 3742 3743
#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
3744
		ret |= get_compat_sigset(&new_ka.sa.sa_mask, &act->sa_mask);
3745
		ret |= get_user(new_ka.sa.sa_flags, &act->sa_flags);
A
Al Viro 已提交
3746 3747 3748 3749 3750 3751 3752 3753
		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);
3754 3755
		ret |= put_compat_sigset(&oact->sa_mask, &old_ka.sa.sa_mask,
					 sizeof(oact->sa_mask));
3756
		ret |= put_user(old_ka.sa.sa_flags, &oact->sa_flags);
A
Al Viro 已提交
3757 3758 3759 3760 3761 3762 3763 3764
#ifdef __ARCH_HAS_SA_RESTORER
		ret |= put_user(ptr_to_compat(old_ka.sa.sa_restorer),
				&oact->sa_restorer);
#endif
	}
	return ret;
}
#endif
3765
#endif /* !CONFIG_ODD_RT_SIGACTION */
L
Linus Torvalds 已提交
3766

3767 3768 3769 3770 3771 3772 3773 3774 3775 3776 3777 3778 3779 3780 3781 3782 3783 3784 3785 3786 3787 3788 3789 3790 3791 3792 3793 3794 3795 3796 3797 3798 3799 3800 3801 3802 3803 3804 3805 3806 3807 3808 3809 3810 3811 3812 3813 3814 3815 3816 3817 3818 3819 3820 3821 3822 3823 3824 3825 3826 3827 3828 3829 3830 3831 3832 3833 3834 3835 3836 3837 3838 3839 3840 3841 3842 3843
#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
3844

3845
#ifdef CONFIG_SGETMASK_SYSCALL
L
Linus Torvalds 已提交
3846 3847 3848 3849

/*
 * For backwards compatibility.  Functionality superseded by sigprocmask.
 */
3850
SYSCALL_DEFINE0(sgetmask)
L
Linus Torvalds 已提交
3851 3852 3853 3854 3855
{
	/* SMP safe */
	return current->blocked.sig[0];
}

3856
SYSCALL_DEFINE1(ssetmask, int, newmask)
L
Linus Torvalds 已提交
3857
{
3858 3859
	int old = current->blocked.sig[0];
	sigset_t newset;
L
Linus Torvalds 已提交
3860

3861
	siginitset(&newset, newmask);
3862
	set_current_blocked(&newset);
L
Linus Torvalds 已提交
3863 3864 3865

	return old;
}
3866
#endif /* CONFIG_SGETMASK_SYSCALL */
L
Linus Torvalds 已提交
3867 3868 3869 3870 3871

#ifdef __ARCH_WANT_SYS_SIGNAL
/*
 * For backwards compatibility.  Functionality superseded by sigaction.
 */
3872
SYSCALL_DEFINE2(signal, int, sig, __sighandler_t, handler)
L
Linus Torvalds 已提交
3873 3874 3875 3876 3877 3878
{
	struct k_sigaction new_sa, old_sa;
	int ret;

	new_sa.sa.sa_handler = handler;
	new_sa.sa.sa_flags = SA_ONESHOT | SA_NOMASK;
3879
	sigemptyset(&new_sa.sa.sa_mask);
L
Linus Torvalds 已提交
3880 3881 3882 3883 3884 3885 3886 3887 3888

	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

3889
SYSCALL_DEFINE0(pause)
L
Linus Torvalds 已提交
3890
{
3891
	while (!signal_pending(current)) {
3892
		__set_current_state(TASK_INTERRUPTIBLE);
3893 3894
		schedule();
	}
L
Linus Torvalds 已提交
3895 3896 3897 3898 3899
	return -ERESTARTNOHAND;
}

#endif

3900
static int sigsuspend(sigset_t *set)
A
Al Viro 已提交
3901 3902 3903 3904
{
	current->saved_sigmask = current->blocked;
	set_current_blocked(set);

3905 3906 3907 3908
	while (!signal_pending(current)) {
		__set_current_state(TASK_INTERRUPTIBLE);
		schedule();
	}
A
Al Viro 已提交
3909 3910 3911 3912
	set_restore_sigmask();
	return -ERESTARTNOHAND;
}

3913 3914 3915 3916 3917 3918
/**
 *  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
 */
3919
SYSCALL_DEFINE2(rt_sigsuspend, sigset_t __user *, unewset, size_t, sigsetsize)
3920 3921 3922 3923 3924 3925 3926 3927 3928
{
	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
3929
	return sigsuspend(&newset);
3930
}
A
Al Viro 已提交
3931 3932 3933 3934 3935 3936 3937 3938 3939 3940
 
#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
3941
	if (get_compat_sigset(&newset, unewset))
A
Al Viro 已提交
3942 3943 3944 3945
		return -EFAULT;
	return sigsuspend(&newset);
}
#endif
3946

3947 3948 3949 3950 3951 3952 3953 3954 3955 3956 3957 3958 3959 3960 3961 3962
#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
3963

3964
__weak const char *arch_vma_name(struct vm_area_struct *vma)
3965 3966 3967 3968
{
	return NULL;
}

L
Linus Torvalds 已提交
3969 3970
void __init signals_init(void)
{
3971 3972 3973
	/* If this check fails, the __ARCH_SI_PREAMBLE_SIZE value is wrong! */
	BUILD_BUG_ON(__ARCH_SI_PREAMBLE_SIZE
		!= offsetof(struct siginfo, _sifields._pad));
3974
	BUILD_BUG_ON(sizeof(struct siginfo) != SI_MAX_SIZE);
3975

3976
	sigqueue_cachep = KMEM_CACHE(sigqueue, SLAB_PANIC);
L
Linus Torvalds 已提交
3977
}
3978 3979 3980 3981

#ifdef CONFIG_KGDB_KDB
#include <linux/kdb.h>
/*
3982
 * kdb_send_sig - Allows kdb to send signals without exposing
3983 3984 3985 3986
 * signal internals.  This function checks if the required locks are
 * available before calling the main signal code, to avoid kdb
 * deadlocks.
 */
3987
void kdb_send_sig(struct task_struct *t, int sig)
3988 3989
{
	static struct task_struct *kdb_prev_t;
3990
	int new_t, ret;
3991 3992 3993 3994 3995 3996 3997 3998 3999
	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) {
4000
		spin_unlock(&t->sighand->siglock);
4001 4002 4003 4004 4005 4006 4007 4008
		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;
	}
4009
	ret = send_signal(sig, SEND_SIG_PRIV, t, PIDTYPE_PID);
4010 4011
	spin_unlock(&t->sighand->siglock);
	if (ret)
4012 4013 4014 4015 4016 4017
		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 */