ptrace.c 26.4 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9
/*
 * linux/kernel/ptrace.c
 *
 * (C) Copyright 1999 Linus Torvalds
 *
 * Common interfaces for "ptrace()" which we do not want
 * to continually duplicate across every architecture.
 */

10
#include <linux/capability.h>
L
Linus Torvalds 已提交
11 12 13 14 15 16 17 18
#include <linux/module.h>
#include <linux/sched.h>
#include <linux/errno.h>
#include <linux/mm.h>
#include <linux/highmem.h>
#include <linux/pagemap.h>
#include <linux/ptrace.h>
#include <linux/security.h>
19
#include <linux/signal.h>
A
Al Viro 已提交
20
#include <linux/audit.h>
21
#include <linux/pid_namespace.h>
22
#include <linux/syscalls.h>
R
Roland McGrath 已提交
23
#include <linux/uaccess.h>
24
#include <linux/regset.h>
25
#include <linux/hw_breakpoint.h>
26
#include <linux/cn_proc.h>
L
Linus Torvalds 已提交
27

28

29 30 31 32 33 34
static int ptrace_trapping_sleep_fn(void *flags)
{
	schedule();
	return 0;
}

L
Linus Torvalds 已提交
35 36 37 38 39 40
/*
 * ptrace a task: make the debugger its new parent and
 * move it to the ptrace list.
 *
 * Must be called with the tasklist lock write-held.
 */
41
void __ptrace_link(struct task_struct *child, struct task_struct *new_parent)
L
Linus Torvalds 已提交
42
{
R
Roland McGrath 已提交
43 44
	BUG_ON(!list_empty(&child->ptrace_entry));
	list_add(&child->ptrace_entry, &new_parent->ptraced);
L
Linus Torvalds 已提交
45 46
	child->parent = new_parent;
}
R
Roland McGrath 已提交
47

48 49 50
/**
 * __ptrace_unlink - unlink ptracee and restore its execution state
 * @child: ptracee to be unlinked
L
Linus Torvalds 已提交
51
 *
52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71
 * Remove @child from the ptrace list, move it back to the original parent,
 * and restore the execution state so that it conforms to the group stop
 * state.
 *
 * Unlinking can happen via two paths - explicit PTRACE_DETACH or ptracer
 * exiting.  For PTRACE_DETACH, unless the ptracee has been killed between
 * ptrace_check_attach() and here, it's guaranteed to be in TASK_TRACED.
 * If the ptracer is exiting, the ptracee can be in any state.
 *
 * After detach, the ptracee should be in a state which conforms to the
 * group stop.  If the group is stopped or in the process of stopping, the
 * ptracee should be put into TASK_STOPPED; otherwise, it should be woken
 * up from TASK_TRACED.
 *
 * If the ptracee is in TASK_TRACED and needs to be moved to TASK_STOPPED,
 * it goes through TRACED -> RUNNING -> STOPPED transition which is similar
 * to but in the opposite direction of what happens while attaching to a
 * stopped task.  However, in this direction, the intermediate RUNNING
 * state is not hidden even from the current ptracer and if it immediately
 * re-attaches and performs a WNOHANG wait(2), it may fail.
72 73 74
 *
 * CONTEXT:
 * write_lock_irq(tasklist_lock)
L
Linus Torvalds 已提交
75
 */
76
void __ptrace_unlink(struct task_struct *child)
L
Linus Torvalds 已提交
77
{
78 79
	BUG_ON(!child->ptrace);

L
Linus Torvalds 已提交
80
	child->ptrace = 0;
R
Roland McGrath 已提交
81 82
	child->parent = child->real_parent;
	list_del_init(&child->ptrace_entry);
L
Linus Torvalds 已提交
83 84

	spin_lock(&child->sighand->siglock);
85

86 87 88 89 90 91 92
	/*
	 * Clear all pending traps and TRAPPING.  TRAPPING should be
	 * cleared regardless of JOBCTL_STOP_PENDING.  Do it explicitly.
	 */
	task_clear_jobctl_pending(child, JOBCTL_TRAP_MASK);
	task_clear_jobctl_trapping(child);

93
	/*
94
	 * Reinstate JOBCTL_STOP_PENDING if group stop is in effect and
95 96 97 98 99
	 * @child isn't dead.
	 */
	if (!(child->flags & PF_EXITING) &&
	    (child->signal->flags & SIGNAL_STOP_STOPPED ||
	     child->signal->group_stop_count))
100
		child->jobctl |= JOBCTL_STOP_PENDING;
101 102 103 104 105 106 107

	/*
	 * If transition to TASK_STOPPED is pending or in TASK_TRACED, kick
	 * @child in the butt.  Note that @resume should be used iff @child
	 * is in TASK_TRACED; otherwise, we might unduly disrupt
	 * TASK_KILLABLE sleeps.
	 */
108
	if (child->jobctl & JOBCTL_STOP_PENDING || task_is_traced(child))
109 110
		signal_wake_up(child, task_is_traced(child));

L
Linus Torvalds 已提交
111 112 113
	spin_unlock(&child->sighand->siglock);
}

114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129
/**
 * ptrace_check_attach - check whether ptracee is ready for ptrace operation
 * @child: ptracee to check for
 * @ignore_state: don't check whether @child is currently %TASK_TRACED
 *
 * Check whether @child is being ptraced by %current and ready for further
 * ptrace operations.  If @ignore_state is %false, @child also should be in
 * %TASK_TRACED state and on return the child is guaranteed to be traced
 * and not executing.  If @ignore_state is %true, @child can be in any
 * state.
 *
 * CONTEXT:
 * Grabs and releases tasklist_lock and @child->sighand->siglock.
 *
 * RETURNS:
 * 0 on success, -ESRCH if %child is not ready.
L
Linus Torvalds 已提交
130
 */
131
int ptrace_check_attach(struct task_struct *child, bool ignore_state)
L
Linus Torvalds 已提交
132 133 134 135 136 137 138 139 140 141 142
{
	int ret = -ESRCH;

	/*
	 * We take the read lock around doing both checks to close a
	 * possible race where someone else was tracing our child and
	 * detached between these two checks.  After this locked check,
	 * we are sure that this is our traced child and that can only
	 * be changed by us so it's not changing right after this.
	 */
	read_lock(&tasklist_lock);
143 144 145 146 147
	if ((child->ptrace & PT_PTRACED) && child->parent == current) {
		/*
		 * child->sighand can't be NULL, release_task()
		 * does ptrace_unlink() before __exit_signal().
		 */
L
Linus Torvalds 已提交
148
		spin_lock_irq(&child->sighand->siglock);
149
		WARN_ON_ONCE(task_is_stopped(child));
T
Tejun Heo 已提交
150 151
		if (ignore_state || (task_is_traced(child) &&
				     !(child->jobctl & JOBCTL_LISTENING)))
152
			ret = 0;
L
Linus Torvalds 已提交
153 154 155 156
		spin_unlock_irq(&child->sighand->siglock);
	}
	read_unlock(&tasklist_lock);

157
	if (!ret && !ignore_state)
R
Roland McGrath 已提交
158
		ret = wait_task_inactive(child, TASK_TRACED) ? 0 : -ESRCH;
L
Linus Torvalds 已提交
159 160 161 162 163

	/* All systems go.. */
	return ret;
}

164
int __ptrace_may_access(struct task_struct *task, unsigned int mode)
165
{
166
	const struct cred *cred = current_cred(), *tcred;
167

168 169 170 171 172 173 174 175 176 177 178 179
	/* May we inspect the given task?
	 * This check is used both for attaching with ptrace
	 * and for allowing access to sensitive information in /proc.
	 *
	 * ptrace_attach denies several cases that /proc allows
	 * because setting up the necessary parent/child relationship
	 * or halting the specified task is impossible.
	 */
	int dumpable = 0;
	/* Don't let security modules deny introspection */
	if (task == current)
		return 0;
180 181
	rcu_read_lock();
	tcred = __task_cred(task);
182 183 184 185 186 187 188 189 190 191 192 193 194
	if (cred->user->user_ns == tcred->user->user_ns &&
	    (cred->uid == tcred->euid &&
	     cred->uid == tcred->suid &&
	     cred->uid == tcred->uid  &&
	     cred->gid == tcred->egid &&
	     cred->gid == tcred->sgid &&
	     cred->gid == tcred->gid))
		goto ok;
	if (ns_capable(tcred->user->user_ns, CAP_SYS_PTRACE))
		goto ok;
	rcu_read_unlock();
	return -EPERM;
ok:
195
	rcu_read_unlock();
196
	smp_rmb();
197
	if (task->mm)
198
		dumpable = get_dumpable(task->mm);
199
	if (!dumpable && !task_ns_capable(task, CAP_SYS_PTRACE))
200 201
		return -EPERM;

202
	return security_ptrace_access_check(task, mode);
203 204
}

205
bool ptrace_may_access(struct task_struct *task, unsigned int mode)
206 207 208
{
	int err;
	task_lock(task);
209
	err = __ptrace_may_access(task, mode);
210
	task_unlock(task);
R
Roland McGrath 已提交
211
	return !err;
212 213
}

T
Tejun Heo 已提交
214 215
static int ptrace_attach(struct task_struct *task, long request,
			 unsigned long flags)
L
Linus Torvalds 已提交
216
{
T
Tejun Heo 已提交
217
	bool seize = (request == PTRACE_SEIZE);
L
Linus Torvalds 已提交
218
	int retval;
219

T
Tejun Heo 已提交
220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235
	/*
	 * SEIZE will enable new ptrace behaviors which will be implemented
	 * gradually.  SEIZE_DEVEL is used to prevent applications
	 * expecting full SEIZE behaviors trapping on kernel commits which
	 * are still in the process of implementing them.
	 *
	 * Only test programs for new ptrace behaviors being implemented
	 * should set SEIZE_DEVEL.  If unset, SEIZE will fail with -EIO.
	 *
	 * Once SEIZE behaviors are completely implemented, this flag and
	 * the following test will be removed.
	 */
	retval = -EIO;
	if (seize && !(flags & PTRACE_SEIZE_DEVEL))
		goto out;

A
Al Viro 已提交
236 237
	audit_ptrace(task);

L
Linus Torvalds 已提交
238
	retval = -EPERM;
239 240
	if (unlikely(task->flags & PF_KTHREAD))
		goto out;
241
	if (same_thread_group(task, current))
242 243
		goto out;

244 245
	/*
	 * Protect exec's credential calculations against our interference;
246 247
	 * interference; SUID, SGID and LSM creds get determined differently
	 * under ptrace.
D
David Howells 已提交
248
	 */
249
	retval = -ERESTARTNOINTR;
250
	if (mutex_lock_interruptible(&task->signal->cred_guard_mutex))
D
David Howells 已提交
251
		goto out;
252

253
	task_lock(task);
254
	retval = __ptrace_may_access(task, PTRACE_MODE_ATTACH);
255
	task_unlock(task);
L
Linus Torvalds 已提交
256
	if (retval)
257
		goto unlock_creds;
L
Linus Torvalds 已提交
258

259
	write_lock_irq(&tasklist_lock);
260 261
	retval = -EPERM;
	if (unlikely(task->exit_state))
262
		goto unlock_tasklist;
263
	if (task->ptrace)
264
		goto unlock_tasklist;
265

266
	task->ptrace = PT_PTRACED;
T
Tejun Heo 已提交
267 268
	if (seize)
		task->ptrace |= PT_SEIZED;
269
	if (task_ns_capable(task, CAP_SYS_PTRACE))
L
Linus Torvalds 已提交
270 271 272
		task->ptrace |= PT_PTRACE_CAP;

	__ptrace_link(task, current);
T
Tejun Heo 已提交
273 274 275 276

	/* SEIZE doesn't trap tracee on attach */
	if (!seize)
		send_sig_info(SIGSTOP, SEND_SIG_FORCED, task);
277

278 279 280
	spin_lock(&task->sighand->siglock);

	/*
281
	 * If the task is already STOPPED, set JOBCTL_TRAP_STOP and
282 283 284 285 286 287 288 289 290 291 292 293 294 295 296
	 * TRAPPING, and kick it so that it transits to TRACED.  TRAPPING
	 * will be cleared if the child completes the transition or any
	 * event which clears the group stop states happens.  We'll wait
	 * for the transition to complete before returning from this
	 * function.
	 *
	 * This hides STOPPED -> RUNNING -> TRACED transition from the
	 * attaching thread but a different thread in the same group can
	 * still observe the transient RUNNING state.  IOW, if another
	 * thread's WNOHANG wait(2) on the stopped tracee races against
	 * ATTACH, the wait(2) may fail due to the transient RUNNING.
	 *
	 * The following task_is_stopped() test is safe as both transitions
	 * in and out of STOPPED are protected by siglock.
	 */
297
	if (task_is_stopped(task) &&
298
	    task_set_jobctl_pending(task, JOBCTL_TRAP_STOP | JOBCTL_TRAPPING))
299 300 301 302
		signal_wake_up(task, 1);

	spin_unlock(&task->sighand->siglock);

303
	retval = 0;
304 305 306
unlock_tasklist:
	write_unlock_irq(&tasklist_lock);
unlock_creds:
307
	mutex_unlock(&task->signal->cred_guard_mutex);
308
out:
309
	if (!retval) {
310 311
		wait_on_bit(&task->jobctl, JOBCTL_TRAPPING_BIT,
			    ptrace_trapping_sleep_fn, TASK_UNINTERRUPTIBLE);
312 313 314
		proc_ptrace_connector(task, PTRACE_ATTACH);
	}

L
Linus Torvalds 已提交
315 316 317
	return retval;
}

318 319 320 321 322 323
/**
 * ptrace_traceme  --  helper for PTRACE_TRACEME
 *
 * Performs checks and sets PT_PTRACED.
 * Should be used by all ptrace implementations for PTRACE_TRACEME.
 */
324
static int ptrace_traceme(void)
325 326 327
{
	int ret = -EPERM;

328 329
	write_lock_irq(&tasklist_lock);
	/* Are we already being traced? */
330 331 332 333 334 335 336 337 338 339 340 341
	if (!current->ptrace) {
		ret = security_ptrace_traceme(current->parent);
		/*
		 * Check PF_EXITING to ensure ->real_parent has not passed
		 * exit_ptrace(). Otherwise we don't report the error but
		 * pretend ->real_parent untraces us right after return.
		 */
		if (!ret && !(current->real_parent->flags & PF_EXITING)) {
			current->ptrace = PT_PTRACED;
			__ptrace_link(current, current->real_parent);
		}
	}
342 343
	write_unlock_irq(&tasklist_lock);

344 345 346
	return ret;
}

347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369
/*
 * Called with irqs disabled, returns true if childs should reap themselves.
 */
static int ignoring_children(struct sighand_struct *sigh)
{
	int ret;
	spin_lock(&sigh->siglock);
	ret = (sigh->action[SIGCHLD-1].sa.sa_handler == SIG_IGN) ||
	      (sigh->action[SIGCHLD-1].sa.sa_flags & SA_NOCLDWAIT);
	spin_unlock(&sigh->siglock);
	return ret;
}

/*
 * Called with tasklist_lock held for writing.
 * Unlink a traced task, and clean it up if it was a traced zombie.
 * Return true if it needs to be reaped with release_task().
 * (We can't call release_task() here because we already hold tasklist_lock.)
 *
 * If it's a zombie, our attachedness prevented normal parent notification
 * or self-reaping.  Do notification now if it would have happened earlier.
 * If it should reap itself, return true.
 *
370 371 372 373
 * If it's our own child, there is no notification to do. But if our normal
 * children self-reap, then this child was prevented by ptrace and we must
 * reap it now, in that case we must also wake up sub-threads sleeping in
 * do_wait().
374 375 376
 */
static bool __ptrace_detach(struct task_struct *tracer, struct task_struct *p)
{
377 378
	bool dead;

379 380
	__ptrace_unlink(p);

381 382 383 384 385 386 387 388 389 390 391
	if (p->exit_state != EXIT_ZOMBIE)
		return false;

	dead = !thread_group_leader(p);

	if (!dead && thread_group_empty(p)) {
		if (!same_thread_group(p->real_parent, tracer))
			dead = do_notify_parent(p, p->exit_signal);
		else if (ignoring_children(tracer->sighand)) {
			__wake_up_parent(p, tracer);
			dead = true;
392 393
		}
	}
394 395 396 397
	/* Mark it as in the process of being reaped. */
	if (dead)
		p->exit_state = EXIT_DEAD;
	return dead;
398 399
}

400
static int ptrace_detach(struct task_struct *child, unsigned int data)
L
Linus Torvalds 已提交
401
{
402
	bool dead = false;
403

404
	if (!valid_signal(data))
405
		return -EIO;
L
Linus Torvalds 已提交
406 407 408

	/* Architecture-specific hardware disable .. */
	ptrace_disable(child);
409
	clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
L
Linus Torvalds 已提交
410

411
	write_lock_irq(&tasklist_lock);
412 413 414 415
	/*
	 * This child can be already killed. Make sure de_thread() or
	 * our sub-thread doing do_wait() didn't do release_task() yet.
	 */
416 417
	if (child->ptrace) {
		child->exit_code = data;
418
		dead = __ptrace_detach(current, child);
419
	}
L
Linus Torvalds 已提交
420 421
	write_unlock_irq(&tasklist_lock);

422
	proc_ptrace_connector(child, PTRACE_DETACH);
423 424 425
	if (unlikely(dead))
		release_task(child);

L
Linus Torvalds 已提交
426 427 428
	return 0;
}

429
/*
430 431 432
 * Detach all tasks we were using ptrace on. Called with tasklist held
 * for writing, and returns with it held too. But note it can release
 * and reacquire the lock.
433 434
 */
void exit_ptrace(struct task_struct *tracer)
435 436
	__releases(&tasklist_lock)
	__acquires(&tasklist_lock)
437 438 439 440
{
	struct task_struct *p, *n;
	LIST_HEAD(ptrace_dead);

441 442 443
	if (likely(list_empty(&tracer->ptraced)))
		return;

444 445 446 447 448
	list_for_each_entry_safe(p, n, &tracer->ptraced, ptrace_entry) {
		if (__ptrace_detach(tracer, p))
			list_add(&p->ptrace_entry, &ptrace_dead);
	}

449
	write_unlock_irq(&tasklist_lock);
450 451 452 453 454 455
	BUG_ON(!list_empty(&tracer->ptraced));

	list_for_each_entry_safe(p, n, &ptrace_dead, ptrace_entry) {
		list_del_init(&p->ptrace_entry);
		release_task(p);
	}
456 457

	write_lock_irq(&tasklist_lock);
458 459
}

L
Linus Torvalds 已提交
460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479
int ptrace_readdata(struct task_struct *tsk, unsigned long src, char __user *dst, int len)
{
	int copied = 0;

	while (len > 0) {
		char buf[128];
		int this_len, retval;

		this_len = (len > sizeof(buf)) ? sizeof(buf) : len;
		retval = access_process_vm(tsk, src, buf, this_len, 0);
		if (!retval) {
			if (copied)
				break;
			return -EIO;
		}
		if (copy_to_user(dst, buf, retval))
			return -EFAULT;
		copied += retval;
		src += retval;
		dst += retval;
R
Roland McGrath 已提交
480
		len -= retval;
L
Linus Torvalds 已提交
481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504
	}
	return copied;
}

int ptrace_writedata(struct task_struct *tsk, char __user *src, unsigned long dst, int len)
{
	int copied = 0;

	while (len > 0) {
		char buf[128];
		int this_len, retval;

		this_len = (len > sizeof(buf)) ? sizeof(buf) : len;
		if (copy_from_user(buf, src, this_len))
			return -EFAULT;
		retval = access_process_vm(tsk, dst, buf, this_len, 1);
		if (!retval) {
			if (copied)
				break;
			return -EIO;
		}
		copied += retval;
		src += retval;
		dst += retval;
R
Roland McGrath 已提交
505
		len -= retval;
L
Linus Torvalds 已提交
506 507 508 509
	}
	return copied;
}

510
static int ptrace_setoptions(struct task_struct *child, unsigned long data)
L
Linus Torvalds 已提交
511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537
{
	child->ptrace &= ~PT_TRACE_MASK;

	if (data & PTRACE_O_TRACESYSGOOD)
		child->ptrace |= PT_TRACESYSGOOD;

	if (data & PTRACE_O_TRACEFORK)
		child->ptrace |= PT_TRACE_FORK;

	if (data & PTRACE_O_TRACEVFORK)
		child->ptrace |= PT_TRACE_VFORK;

	if (data & PTRACE_O_TRACECLONE)
		child->ptrace |= PT_TRACE_CLONE;

	if (data & PTRACE_O_TRACEEXEC)
		child->ptrace |= PT_TRACE_EXEC;

	if (data & PTRACE_O_TRACEVFORKDONE)
		child->ptrace |= PT_TRACE_VFORK_DONE;

	if (data & PTRACE_O_TRACEEXIT)
		child->ptrace |= PT_TRACE_EXIT;

	return (data & ~PTRACE_O_MASK) ? -EINVAL : 0;
}

538
static int ptrace_getsiginfo(struct task_struct *child, siginfo_t *info)
L
Linus Torvalds 已提交
539
{
540
	unsigned long flags;
L
Linus Torvalds 已提交
541 542
	int error = -ESRCH;

543
	if (lock_task_sighand(child, &flags)) {
L
Linus Torvalds 已提交
544 545
		error = -EINVAL;
		if (likely(child->last_siginfo != NULL)) {
546
			*info = *child->last_siginfo;
L
Linus Torvalds 已提交
547 548
			error = 0;
		}
549
		unlock_task_sighand(child, &flags);
L
Linus Torvalds 已提交
550 551 552 553
	}
	return error;
}

554
static int ptrace_setsiginfo(struct task_struct *child, const siginfo_t *info)
L
Linus Torvalds 已提交
555
{
556
	unsigned long flags;
L
Linus Torvalds 已提交
557 558
	int error = -ESRCH;

559
	if (lock_task_sighand(child, &flags)) {
L
Linus Torvalds 已提交
560 561
		error = -EINVAL;
		if (likely(child->last_siginfo != NULL)) {
562
			*child->last_siginfo = *info;
L
Linus Torvalds 已提交
563 564
			error = 0;
		}
565
		unlock_task_sighand(child, &flags);
L
Linus Torvalds 已提交
566 567 568 569
	}
	return error;
}

R
Roland McGrath 已提交
570 571 572 573 574 575 576

#ifdef PTRACE_SINGLESTEP
#define is_singlestep(request)		((request) == PTRACE_SINGLESTEP)
#else
#define is_singlestep(request)		0
#endif

577 578 579 580 581 582
#ifdef PTRACE_SINGLEBLOCK
#define is_singleblock(request)		((request) == PTRACE_SINGLEBLOCK)
#else
#define is_singleblock(request)		0
#endif

R
Roland McGrath 已提交
583 584 585 586 587 588
#ifdef PTRACE_SYSEMU
#define is_sysemu_singlestep(request)	((request) == PTRACE_SYSEMU_SINGLESTEP)
#else
#define is_sysemu_singlestep(request)	0
#endif

589 590
static int ptrace_resume(struct task_struct *child, long request,
			 unsigned long data)
R
Roland McGrath 已提交
591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606
{
	if (!valid_signal(data))
		return -EIO;

	if (request == PTRACE_SYSCALL)
		set_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
	else
		clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);

#ifdef TIF_SYSCALL_EMU
	if (request == PTRACE_SYSEMU || request == PTRACE_SYSEMU_SINGLESTEP)
		set_tsk_thread_flag(child, TIF_SYSCALL_EMU);
	else
		clear_tsk_thread_flag(child, TIF_SYSCALL_EMU);
#endif

607 608 609 610 611
	if (is_singleblock(request)) {
		if (unlikely(!arch_has_block_step()))
			return -EIO;
		user_enable_block_step(child);
	} else if (is_singlestep(request) || is_sysemu_singlestep(request)) {
R
Roland McGrath 已提交
612 613 614
		if (unlikely(!arch_has_single_step()))
			return -EIO;
		user_enable_single_step(child);
R
Roland McGrath 已提交
615
	} else {
R
Roland McGrath 已提交
616
		user_disable_single_step(child);
R
Roland McGrath 已提交
617
	}
R
Roland McGrath 已提交
618 619

	child->exit_code = data;
620
	wake_up_state(child, __TASK_TRACED);
R
Roland McGrath 已提交
621 622 623 624

	return 0;
}

625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649
#ifdef CONFIG_HAVE_ARCH_TRACEHOOK

static const struct user_regset *
find_regset(const struct user_regset_view *view, unsigned int type)
{
	const struct user_regset *regset;
	int n;

	for (n = 0; n < view->n; ++n) {
		regset = view->regsets + n;
		if (regset->core_note_type == type)
			return regset;
	}

	return NULL;
}

static int ptrace_regset(struct task_struct *task, int req, unsigned int type,
			 struct iovec *kiov)
{
	const struct user_regset_view *view = task_user_regset_view(task);
	const struct user_regset *regset = find_regset(view, type);
	int regset_no;

	if (!regset || (kiov->iov_len % regset->size) != 0)
650
		return -EINVAL;
651 652 653 654 655 656 657 658 659 660 661 662 663 664 665

	regset_no = regset - view->regsets;
	kiov->iov_len = min(kiov->iov_len,
			    (__kernel_size_t) (regset->n * regset->size));

	if (req == PTRACE_GETREGSET)
		return copy_regset_to_user(task, view, regset_no, 0,
					   kiov->iov_len, kiov->iov_base);
	else
		return copy_regset_from_user(task, view, regset_no, 0,
					     kiov->iov_len, kiov->iov_base);
}

#endif

L
Linus Torvalds 已提交
666
int ptrace_request(struct task_struct *child, long request,
667
		   unsigned long addr, unsigned long data)
L
Linus Torvalds 已提交
668
{
T
Tejun Heo 已提交
669
	bool seized = child->ptrace & PT_SEIZED;
L
Linus Torvalds 已提交
670
	int ret = -EIO;
T
Tejun Heo 已提交
671
	siginfo_t siginfo, *si;
N
Namhyung Kim 已提交
672 673
	void __user *datavp = (void __user *) data;
	unsigned long __user *datalp = datavp;
T
Tejun Heo 已提交
674
	unsigned long flags;
L
Linus Torvalds 已提交
675 676

	switch (request) {
677 678 679 680 681 682 683
	case PTRACE_PEEKTEXT:
	case PTRACE_PEEKDATA:
		return generic_ptrace_peekdata(child, addr, data);
	case PTRACE_POKETEXT:
	case PTRACE_POKEDATA:
		return generic_ptrace_pokedata(child, addr, data);

L
Linus Torvalds 已提交
684 685 686 687 688 689 690
#ifdef PTRACE_OLDSETOPTIONS
	case PTRACE_OLDSETOPTIONS:
#endif
	case PTRACE_SETOPTIONS:
		ret = ptrace_setoptions(child, data);
		break;
	case PTRACE_GETEVENTMSG:
N
Namhyung Kim 已提交
691
		ret = put_user(child->ptrace_message, datalp);
L
Linus Torvalds 已提交
692
		break;
693

L
Linus Torvalds 已提交
694
	case PTRACE_GETSIGINFO:
695 696
		ret = ptrace_getsiginfo(child, &siginfo);
		if (!ret)
N
Namhyung Kim 已提交
697
			ret = copy_siginfo_to_user(datavp, &siginfo);
L
Linus Torvalds 已提交
698
		break;
699

L
Linus Torvalds 已提交
700
	case PTRACE_SETSIGINFO:
N
Namhyung Kim 已提交
701
		if (copy_from_user(&siginfo, datavp, sizeof siginfo))
702 703 704
			ret = -EFAULT;
		else
			ret = ptrace_setsiginfo(child, &siginfo);
L
Linus Torvalds 已提交
705
		break;
706

T
Tejun Heo 已提交
707 708 709 710 711 712 713 714 715 716 717 718 719 720
	case PTRACE_INTERRUPT:
		/*
		 * Stop tracee without any side-effect on signal or job
		 * control.  At least one trap is guaranteed to happen
		 * after this request.  If @child is already trapped, the
		 * current trap is not disturbed and another trap will
		 * happen after the current trap is ended with PTRACE_CONT.
		 *
		 * The actual trap might not be PTRACE_EVENT_STOP trap but
		 * the pending condition is cleared regardless.
		 */
		if (unlikely(!seized || !lock_task_sighand(child, &flags)))
			break;

T
Tejun Heo 已提交
721 722 723 724 725 726
		/*
		 * INTERRUPT doesn't disturb existing trap sans one
		 * exception.  If ptracer issued LISTEN for the current
		 * STOP, this INTERRUPT should clear LISTEN and re-trap
		 * tracee into STOP.
		 */
T
Tejun Heo 已提交
727
		if (likely(task_set_jobctl_pending(child, JOBCTL_TRAP_STOP)))
T
Tejun Heo 已提交
728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757
			signal_wake_up(child, child->jobctl & JOBCTL_LISTENING);

		unlock_task_sighand(child, &flags);
		ret = 0;
		break;

	case PTRACE_LISTEN:
		/*
		 * Listen for events.  Tracee must be in STOP.  It's not
		 * resumed per-se but is not considered to be in TRACED by
		 * wait(2) or ptrace(2).  If an async event (e.g. group
		 * stop state change) happens, tracee will enter STOP trap
		 * again.  Alternatively, ptracer can issue INTERRUPT to
		 * finish listening and re-trap tracee into STOP.
		 */
		if (unlikely(!seized || !lock_task_sighand(child, &flags)))
			break;

		si = child->last_siginfo;
		if (unlikely(!si || si->si_code >> 8 != PTRACE_EVENT_STOP))
			break;

		child->jobctl |= JOBCTL_LISTENING;

		/*
		 * If NOTIFY is set, it means event happened between start
		 * of this trap and now.  Trigger re-trap immediately.
		 */
		if (child->jobctl & JOBCTL_TRAP_NOTIFY)
			signal_wake_up(child, true);
T
Tejun Heo 已提交
758 759 760 761 762

		unlock_task_sighand(child, &flags);
		ret = 0;
		break;

A
Alexey Dobriyan 已提交
763 764 765
	case PTRACE_DETACH:	 /* detach a process that was attached. */
		ret = ptrace_detach(child, data);
		break;
R
Roland McGrath 已提交
766

767 768
#ifdef CONFIG_BINFMT_ELF_FDPIC
	case PTRACE_GETFDPIC: {
769
		struct mm_struct *mm = get_task_mm(child);
770 771
		unsigned long tmp = 0;

772 773 774 775
		ret = -ESRCH;
		if (!mm)
			break;

776 777
		switch (addr) {
		case PTRACE_GETFDPIC_EXEC:
778
			tmp = mm->context.exec_fdpic_loadmap;
779 780
			break;
		case PTRACE_GETFDPIC_INTERP:
781
			tmp = mm->context.interp_fdpic_loadmap;
782 783 784 785
			break;
		default:
			break;
		}
786
		mmput(mm);
787

N
Namhyung Kim 已提交
788
		ret = put_user(tmp, datalp);
789 790 791 792
		break;
	}
#endif

R
Roland McGrath 已提交
793 794 795
#ifdef PTRACE_SINGLESTEP
	case PTRACE_SINGLESTEP:
#endif
796 797 798
#ifdef PTRACE_SINGLEBLOCK
	case PTRACE_SINGLEBLOCK:
#endif
R
Roland McGrath 已提交
799 800 801 802 803 804 805 806 807 808 809 810 811
#ifdef PTRACE_SYSEMU
	case PTRACE_SYSEMU:
	case PTRACE_SYSEMU_SINGLESTEP:
#endif
	case PTRACE_SYSCALL:
	case PTRACE_CONT:
		return ptrace_resume(child, request, data);

	case PTRACE_KILL:
		if (child->exit_state)	/* already dead */
			return 0;
		return ptrace_resume(child, request, SIGKILL);

812 813 814 815 816
#ifdef CONFIG_HAVE_ARCH_TRACEHOOK
	case PTRACE_GETREGSET:
	case PTRACE_SETREGSET:
	{
		struct iovec kiov;
N
Namhyung Kim 已提交
817
		struct iovec __user *uiov = datavp;
818 819 820 821 822 823 824 825 826 827 828 829 830 831

		if (!access_ok(VERIFY_WRITE, uiov, sizeof(*uiov)))
			return -EFAULT;

		if (__get_user(kiov.iov_base, &uiov->iov_base) ||
		    __get_user(kiov.iov_len, &uiov->iov_len))
			return -EFAULT;

		ret = ptrace_regset(child, request, addr, &kiov);
		if (!ret)
			ret = __put_user(kiov.iov_len, &uiov->iov_len);
		break;
	}
#endif
L
Linus Torvalds 已提交
832 833 834 835 836 837
	default:
		break;
	}

	return ret;
}
838

839
static struct task_struct *ptrace_get_task_struct(pid_t pid)
840 841
{
	struct task_struct *child;
842

843
	rcu_read_lock();
844
	child = find_task_by_vpid(pid);
845 846
	if (child)
		get_task_struct(child);
847
	rcu_read_unlock();
848

849
	if (!child)
850 851
		return ERR_PTR(-ESRCH);
	return child;
852 853
}

854 855 856 857
#ifndef arch_ptrace_attach
#define arch_ptrace_attach(child)	do { } while (0)
#endif

858 859
SYSCALL_DEFINE4(ptrace, long, request, long, pid, unsigned long, addr,
		unsigned long, data)
860 861 862 863
{
	struct task_struct *child;
	long ret;

864 865
	if (request == PTRACE_TRACEME) {
		ret = ptrace_traceme();
866 867
		if (!ret)
			arch_ptrace_attach(current);
868
		goto out;
869 870 871 872 873 874 875
	}

	child = ptrace_get_task_struct(pid);
	if (IS_ERR(child)) {
		ret = PTR_ERR(child);
		goto out;
	}
876

T
Tejun Heo 已提交
877 878
	if (request == PTRACE_ATTACH || request == PTRACE_SEIZE) {
		ret = ptrace_attach(child, request, data);
879 880 881 882 883 884
		/*
		 * Some architectures need to do book-keeping after
		 * a ptrace attach.
		 */
		if (!ret)
			arch_ptrace_attach(child);
885
		goto out_put_task_struct;
886 887
	}

T
Tejun Heo 已提交
888 889
	ret = ptrace_check_attach(child, request == PTRACE_KILL ||
				  request == PTRACE_INTERRUPT);
890 891 892 893 894 895 896 897 898 899
	if (ret < 0)
		goto out_put_task_struct;

	ret = arch_ptrace(child, request, addr, data);

 out_put_task_struct:
	put_task_struct(child);
 out:
	return ret;
}
A
Alexey Dobriyan 已提交
900

901 902
int generic_ptrace_peekdata(struct task_struct *tsk, unsigned long addr,
			    unsigned long data)
A
Alexey Dobriyan 已提交
903 904 905 906 907 908 909 910 911
{
	unsigned long tmp;
	int copied;

	copied = access_process_vm(tsk, addr, &tmp, sizeof(tmp), 0);
	if (copied != sizeof(tmp))
		return -EIO;
	return put_user(tmp, (unsigned long __user *)data);
}
A
Alexey Dobriyan 已提交
912

913 914
int generic_ptrace_pokedata(struct task_struct *tsk, unsigned long addr,
			    unsigned long data)
A
Alexey Dobriyan 已提交
915 916 917 918 919 920
{
	int copied;

	copied = access_process_vm(tsk, addr, &data, sizeof(data), 1);
	return (copied == sizeof(data)) ? 0 : -EIO;
}
R
Roland McGrath 已提交
921

922
#if defined CONFIG_COMPAT
R
Roland McGrath 已提交
923 924 925 926 927 928 929
#include <linux/compat.h>

int compat_ptrace_request(struct task_struct *child, compat_long_t request,
			  compat_ulong_t addr, compat_ulong_t data)
{
	compat_ulong_t __user *datap = compat_ptr(data);
	compat_ulong_t word;
930
	siginfo_t siginfo;
R
Roland McGrath 已提交
931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952
	int ret;

	switch (request) {
	case PTRACE_PEEKTEXT:
	case PTRACE_PEEKDATA:
		ret = access_process_vm(child, addr, &word, sizeof(word), 0);
		if (ret != sizeof(word))
			ret = -EIO;
		else
			ret = put_user(word, datap);
		break;

	case PTRACE_POKETEXT:
	case PTRACE_POKEDATA:
		ret = access_process_vm(child, addr, &data, sizeof(data), 1);
		ret = (ret != sizeof(data) ? -EIO : 0);
		break;

	case PTRACE_GETEVENTMSG:
		ret = put_user((compat_ulong_t) child->ptrace_message, datap);
		break;

953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968
	case PTRACE_GETSIGINFO:
		ret = ptrace_getsiginfo(child, &siginfo);
		if (!ret)
			ret = copy_siginfo_to_user32(
				(struct compat_siginfo __user *) datap,
				&siginfo);
		break;

	case PTRACE_SETSIGINFO:
		memset(&siginfo, 0, sizeof siginfo);
		if (copy_siginfo_from_user32(
			    &siginfo, (struct compat_siginfo __user *) datap))
			ret = -EFAULT;
		else
			ret = ptrace_setsiginfo(child, &siginfo);
		break;
969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994
#ifdef CONFIG_HAVE_ARCH_TRACEHOOK
	case PTRACE_GETREGSET:
	case PTRACE_SETREGSET:
	{
		struct iovec kiov;
		struct compat_iovec __user *uiov =
			(struct compat_iovec __user *) datap;
		compat_uptr_t ptr;
		compat_size_t len;

		if (!access_ok(VERIFY_WRITE, uiov, sizeof(*uiov)))
			return -EFAULT;

		if (__get_user(ptr, &uiov->iov_base) ||
		    __get_user(len, &uiov->iov_len))
			return -EFAULT;

		kiov.iov_base = compat_ptr(ptr);
		kiov.iov_len = len;

		ret = ptrace_regset(child, request, addr, &kiov);
		if (!ret)
			ret = __put_user(kiov.iov_len, &uiov->iov_len);
		break;
	}
#endif
995

R
Roland McGrath 已提交
996 997 998 999 1000 1001
	default:
		ret = ptrace_request(child, request, addr, data);
	}

	return ret;
}
R
Roland McGrath 已提交
1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019

asmlinkage long compat_sys_ptrace(compat_long_t request, compat_long_t pid,
				  compat_long_t addr, compat_long_t data)
{
	struct task_struct *child;
	long ret;

	if (request == PTRACE_TRACEME) {
		ret = ptrace_traceme();
		goto out;
	}

	child = ptrace_get_task_struct(pid);
	if (IS_ERR(child)) {
		ret = PTR_ERR(child);
		goto out;
	}

T
Tejun Heo 已提交
1020 1021
	if (request == PTRACE_ATTACH || request == PTRACE_SEIZE) {
		ret = ptrace_attach(child, request, data);
R
Roland McGrath 已提交
1022 1023 1024 1025 1026 1027 1028 1029 1030
		/*
		 * Some architectures need to do book-keeping after
		 * a ptrace attach.
		 */
		if (!ret)
			arch_ptrace_attach(child);
		goto out_put_task_struct;
	}

T
Tejun Heo 已提交
1031 1032
	ret = ptrace_check_attach(child, request == PTRACE_KILL ||
				  request == PTRACE_INTERRUPT);
R
Roland McGrath 已提交
1033 1034 1035 1036 1037 1038 1039 1040
	if (!ret)
		ret = compat_arch_ptrace(child, request, addr, data);

 out_put_task_struct:
	put_task_struct(child);
 out:
	return ret;
}
1041
#endif	/* CONFIG_COMPAT */
1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057

#ifdef CONFIG_HAVE_HW_BREAKPOINT
int ptrace_get_breakpoints(struct task_struct *tsk)
{
	if (atomic_inc_not_zero(&tsk->ptrace_bp_refcnt))
		return 0;

	return -1;
}

void ptrace_put_breakpoints(struct task_struct *tsk)
{
	if (atomic_dec_and_test(&tsk->ptrace_bp_refcnt))
		flush_ptrace_hw_breakpoint(tsk);
}
#endif /* CONFIG_HAVE_HW_BREAKPOINT */