coredump.c 22.1 KB
Newer Older
1
// SPDX-License-Identifier: GPL-2.0
2 3 4
#include <linux/slab.h>
#include <linux/file.h>
#include <linux/fdtable.h>
5
#include <linux/freezer.h>
6 7 8 9
#include <linux/mm.h>
#include <linux/stat.h>
#include <linux/fcntl.h>
#include <linux/swap.h>
10
#include <linux/ctype.h>
11 12 13 14 15 16 17 18 19
#include <linux/string.h>
#include <linux/init.h>
#include <linux/pagemap.h>
#include <linux/perf_event.h>
#include <linux/highmem.h>
#include <linux/spinlock.h>
#include <linux/key.h>
#include <linux/personality.h>
#include <linux/binfmts.h>
20
#include <linux/coredump.h>
21
#include <linux/sched/coredump.h>
22
#include <linux/sched/signal.h>
23
#include <linux/sched/task_stack.h>
24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40
#include <linux/utsname.h>
#include <linux/pid_namespace.h>
#include <linux/module.h>
#include <linux/namei.h>
#include <linux/mount.h>
#include <linux/security.h>
#include <linux/syscalls.h>
#include <linux/tsacct_kern.h>
#include <linux/cn_proc.h>
#include <linux/audit.h>
#include <linux/tracehook.h>
#include <linux/kmod.h>
#include <linux/fsnotify.h>
#include <linux/fs_struct.h>
#include <linux/pipe_fs_i.h>
#include <linux/oom.h>
#include <linux/compat.h>
41 42
#include <linux/fs.h>
#include <linux/path.h>
43
#include <linux/timekeeping.h>
44

45
#include <linux/uaccess.h>
46 47 48 49 50 51 52 53 54 55 56
#include <asm/mmu_context.h>
#include <asm/tlb.h>
#include <asm/exec.h>

#include <trace/events/task.h>
#include "internal.h"

#include <trace/events/sched.h>

int core_uses_pid;
unsigned int core_pipe_limit;
57 58
char core_pattern[CORENAME_MAX_SIZE] = "core";
static int core_name_size = CORENAME_MAX_SIZE;
59 60 61 62 63 64 65 66

struct core_name {
	char *corename;
	int used, size;
};

/* The maximal length of core_pattern is also specified in sysctl.c */

67
static int expand_corename(struct core_name *cn, int size)
68
{
69
	char *corename = krealloc(cn->corename, size, GFP_KERNEL);
70

71
	if (!corename)
72 73
		return -ENOMEM;

74 75 76 77
	if (size > core_name_size) /* racy but harmless */
		core_name_size = size;

	cn->size = ksize(corename);
78
	cn->corename = corename;
79 80 81
	return 0;
}

82 83
static __printf(2, 0) int cn_vprintf(struct core_name *cn, const char *fmt,
				     va_list arg)
84
{
85
	int free, need;
E
Eric Dumazet 已提交
86
	va_list arg_copy;
87

88 89
again:
	free = cn->size - cn->used;
E
Eric Dumazet 已提交
90 91 92 93 94

	va_copy(arg_copy, arg);
	need = vsnprintf(cn->corename + cn->used, free, fmt, arg_copy);
	va_end(arg_copy);

95 96 97 98
	if (need < free) {
		cn->used += need;
		return 0;
	}
99

100
	if (!expand_corename(cn, cn->size + need - free + 1))
101
		goto again;
102

103
	return -ENOMEM;
104 105
}

106
static __printf(2, 3) int cn_printf(struct core_name *cn, const char *fmt, ...)
O
Oleg Nesterov 已提交
107 108 109 110 111 112 113 114 115 116 117
{
	va_list arg;
	int ret;

	va_start(arg, fmt);
	ret = cn_vprintf(cn, fmt, arg);
	va_end(arg);

	return ret;
}

118 119
static __printf(2, 3)
int cn_esc_printf(struct core_name *cn, const char *fmt, ...)
120
{
121 122 123 124 125 126 127 128
	int cur = cn->used;
	va_list arg;
	int ret;

	va_start(arg, fmt);
	ret = cn_vprintf(cn, fmt, arg);
	va_end(arg);

129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148
	if (ret == 0) {
		/*
		 * Ensure that this coredump name component can't cause the
		 * resulting corefile path to consist of a ".." or ".".
		 */
		if ((cn->used - cur == 1 && cn->corename[cur] == '.') ||
				(cn->used - cur == 2 && cn->corename[cur] == '.'
				&& cn->corename[cur+1] == '.'))
			cn->corename[cur] = '!';

		/*
		 * Empty names are fishy and could be used to create a "//" in a
		 * corefile name, causing the coredump to happen one directory
		 * level too high. Enforce that all components of the core
		 * pattern are at least one character long.
		 */
		if (cn->used == cur)
			ret = cn_printf(cn, "!");
	}

149 150 151 152 153
	for (; cur < cn->used; ++cur) {
		if (cn->corename[cur] == '/')
			cn->corename[cur] = '!';
	}
	return ret;
154 155 156 157 158 159 160 161 162
}

static int cn_print_exe_file(struct core_name *cn)
{
	struct file *exe_file;
	char *pathbuf, *path;
	int ret;

	exe_file = get_mm_exe_file(current->mm);
163 164
	if (!exe_file)
		return cn_esc_printf(cn, "%s (path unknown)", current->comm);
165

166
	pathbuf = kmalloc(PATH_MAX, GFP_KERNEL);
167 168 169 170 171
	if (!pathbuf) {
		ret = -ENOMEM;
		goto put_exe_file;
	}

M
Miklos Szeredi 已提交
172
	path = file_path(exe_file, pathbuf, PATH_MAX);
173 174 175 176 177
	if (IS_ERR(path)) {
		ret = PTR_ERR(path);
		goto free_buf;
	}

178
	ret = cn_esc_printf(cn, "%s", path);
179 180 181 182 183 184 185 186 187 188 189 190

free_buf:
	kfree(pathbuf);
put_exe_file:
	fput(exe_file);
	return ret;
}

/* format_corename will inspect the pattern parameter, and output a
 * name into corename, which must have space for at least
 * CORENAME_MAX_SIZE bytes plus one byte for the zero terminator.
 */
191 192
static int format_corename(struct core_name *cn, struct coredump_params *cprm,
			   size_t **argv, int *argc)
193 194 195 196
{
	const struct cred *cred = current_cred();
	const char *pat_ptr = core_pattern;
	int ispipe = (*pat_ptr == '|');
197
	bool was_space = false;
198 199 200
	int pid_in_pattern = 0;
	int err = 0;

201
	cn->used = 0;
202 203
	cn->corename = NULL;
	if (expand_corename(cn, core_name_size))
204
		return -ENOMEM;
205 206
	cn->corename[0] = '\0';

207 208 209 210 211 212
	if (ispipe) {
		int argvs = sizeof(core_pattern) / 2;
		(*argv) = kmalloc_array(argvs, sizeof(**argv), GFP_KERNEL);
		if (!(*argv))
			return -ENOMEM;
		(*argv)[(*argc)++] = 0;
213
		++pat_ptr;
214
	}
215 216 217 218

	/* Repeat as long as we have more pattern to process and more output
	   space */
	while (*pat_ptr) {
219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235
		/*
		 * Split on spaces before doing template expansion so that
		 * %e and %E don't get split if they have spaces in them
		 */
		if (ispipe) {
			if (isspace(*pat_ptr)) {
				was_space = true;
				pat_ptr++;
				continue;
			} else if (was_space) {
				was_space = false;
				err = cn_printf(cn, "%c", '\0');
				if (err)
					return err;
				(*argv)[(*argc)++] = cn->used;
			}
		}
236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252
		if (*pat_ptr != '%') {
			err = cn_printf(cn, "%c", *pat_ptr++);
		} else {
			switch (*++pat_ptr) {
			/* single % at the end, drop that */
			case 0:
				goto out;
			/* Double percent, output one percent */
			case '%':
				err = cn_printf(cn, "%c", '%');
				break;
			/* pid */
			case 'p':
				pid_in_pattern = 1;
				err = cn_printf(cn, "%d",
					      task_tgid_vnr(current));
				break;
253 254 255 256 257
			/* global pid */
			case 'P':
				err = cn_printf(cn, "%d",
					      task_tgid_nr(current));
				break;
258 259 260 261 262 263 264 265
			case 'i':
				err = cn_printf(cn, "%d",
					      task_pid_vnr(current));
				break;
			case 'I':
				err = cn_printf(cn, "%d",
					      task_pid_nr(current));
				break;
266 267
			/* uid */
			case 'u':
268 269 270
				err = cn_printf(cn, "%u",
						from_kuid(&init_user_ns,
							  cred->uid));
271 272 273
				break;
			/* gid */
			case 'g':
274 275 276
				err = cn_printf(cn, "%u",
						from_kgid(&init_user_ns,
							  cred->gid));
277
				break;
278 279 280 281
			case 'd':
				err = cn_printf(cn, "%d",
					__get_dumpable(cprm->mm_flags));
				break;
282 283
			/* signal that caused the coredump */
			case 's':
284 285
				err = cn_printf(cn, "%d",
						cprm->siginfo->si_signo);
286 287 288
				break;
			/* UNIX time of coredump */
			case 't': {
289 290 291 292
				time64_t time;

				time = ktime_get_real_seconds();
				err = cn_printf(cn, "%lld", time);
293 294 295
				break;
			}
			/* hostname */
296
			case 'h':
297
				down_read(&uts_sem);
298
				err = cn_esc_printf(cn, "%s",
299 300 301 302
					      utsname()->nodename);
				up_read(&uts_sem);
				break;
			/* executable */
303 304
			case 'e':
				err = cn_esc_printf(cn, "%s", current->comm);
305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323
				break;
			case 'E':
				err = cn_print_exe_file(cn);
				break;
			/* core limit size */
			case 'c':
				err = cn_printf(cn, "%lu",
					      rlimit(RLIMIT_CORE));
				break;
			default:
				break;
			}
			++pat_ptr;
		}

		if (err)
			return err;
	}

324
out:
325 326 327 328 329 330 331 332 333 334 335 336 337
	/* Backward compatibility with core_uses_pid:
	 *
	 * If core_pattern does not include a %p (as is the default)
	 * and core_uses_pid is set, then .%pid will be appended to
	 * the filename. Do not do this for piped commands. */
	if (!ispipe && !pid_in_pattern && core_uses_pid) {
		err = cn_printf(cn, ".%d", task_tgid_vnr(current));
		if (err)
			return err;
	}
	return ispipe;
}

338
static int zap_process(struct task_struct *start, int exit_code, int flags)
339 340 341 342
{
	struct task_struct *t;
	int nr = 0;

343 344
	/* ignore all signals except SIGKILL, see prepare_signal() */
	start->signal->flags = SIGNAL_GROUP_COREDUMP | flags;
345 346 347
	start->signal->group_exit_code = exit_code;
	start->signal->group_stop_count = 0;

348
	for_each_thread(start, t) {
349 350 351 352 353 354
		task_clear_jobctl_pending(t, JOBCTL_PENDING_MASK);
		if (t != current && t->mm) {
			sigaddset(&t->pending.signal, SIGKILL);
			signal_wake_up(t, 1);
			nr++;
		}
355
	}
356 357 358 359

	return nr;
}

360 361
static int zap_threads(struct task_struct *tsk, struct mm_struct *mm,
			struct core_state *core_state, int exit_code)
362 363 364 365 366 367 368 369
{
	struct task_struct *g, *p;
	unsigned long flags;
	int nr = -EAGAIN;

	spin_lock_irq(&tsk->sighand->siglock);
	if (!signal_group_exit(tsk->signal)) {
		mm->core_state = core_state;
370
		tsk->signal->group_exit_task = tsk;
371
		nr = zap_process(tsk, exit_code, 0);
372
		clear_tsk_thread_flag(tsk, TIF_SIGPENDING);
373 374 375 376 377
	}
	spin_unlock_irq(&tsk->sighand->siglock);
	if (unlikely(nr < 0))
		return nr;

378
	tsk->flags |= PF_DUMPCORE;
379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416
	if (atomic_read(&mm->mm_users) == nr + 1)
		goto done;
	/*
	 * We should find and kill all tasks which use this mm, and we should
	 * count them correctly into ->nr_threads. We don't take tasklist
	 * lock, but this is safe wrt:
	 *
	 * fork:
	 *	None of sub-threads can fork after zap_process(leader). All
	 *	processes which were created before this point should be
	 *	visible to zap_threads() because copy_process() adds the new
	 *	process to the tail of init_task.tasks list, and lock/unlock
	 *	of ->siglock provides a memory barrier.
	 *
	 * do_exit:
	 *	The caller holds mm->mmap_sem. This means that the task which
	 *	uses this mm can't pass exit_mm(), so it can't exit or clear
	 *	its ->mm.
	 *
	 * de_thread:
	 *	It does list_replace_rcu(&leader->tasks, &current->tasks),
	 *	we must see either old or new leader, this does not matter.
	 *	However, it can change p->sighand, so lock_task_sighand(p)
	 *	must be used. Since p->mm != NULL and we hold ->mmap_sem
	 *	it can't fail.
	 *
	 *	Note also that "g" can be the old leader with ->mm == NULL
	 *	and already unhashed and thus removed from ->thread_group.
	 *	This is OK, __unhash_process()->list_del_rcu() does not
	 *	clear the ->next pointer, we will find the new leader via
	 *	next_thread().
	 */
	rcu_read_lock();
	for_each_process(g) {
		if (g == tsk->group_leader)
			continue;
		if (g->flags & PF_KTHREAD)
			continue;
417 418 419 420 421 422 423 424 425

		for_each_thread(g, p) {
			if (unlikely(!p->mm))
				continue;
			if (unlikely(p->mm == mm)) {
				lock_task_sighand(p, &flags);
				nr += zap_process(p, exit_code,
							SIGNAL_GROUP_EXIT);
				unlock_task_sighand(p, &flags);
426
			}
427 428
			break;
		}
429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445
	}
	rcu_read_unlock();
done:
	atomic_set(&core_state->nr_threads, nr);
	return nr;
}

static int coredump_wait(int exit_code, struct core_state *core_state)
{
	struct task_struct *tsk = current;
	struct mm_struct *mm = tsk->mm;
	int core_waiters = -EBUSY;

	init_completion(&core_state->startup);
	core_state->dumper.task = tsk;
	core_state->dumper.next = NULL;

446 447 448
	if (down_write_killable(&mm->mmap_sem))
		return -EINTR;

449 450 451 452 453 454 455
	if (!mm->core_state)
		core_waiters = zap_threads(tsk, mm, core_state, exit_code);
	up_write(&mm->mmap_sem);

	if (core_waiters > 0) {
		struct core_thread *ptr;

456
		freezer_do_not_count();
457
		wait_for_completion(&core_state->startup);
458
		freezer_count();
459 460 461 462 463 464 465 466 467 468 469 470 471 472 473
		/*
		 * Wait for all the threads to become inactive, so that
		 * all the thread context (extended register state, like
		 * fpu etc) gets copied to the memory.
		 */
		ptr = core_state->dumper.next;
		while (ptr != NULL) {
			wait_task_inactive(ptr->task, 0);
			ptr = ptr->next;
		}
	}

	return core_waiters;
}

474
static void coredump_finish(struct mm_struct *mm, bool core_dumped)
475 476 477 478
{
	struct core_thread *curr, *next;
	struct task_struct *task;

479
	spin_lock_irq(&current->sighand->siglock);
480 481
	if (core_dumped && !__fatal_signal_pending(current))
		current->signal->group_exit_code |= 0x80;
482 483 484 485
	current->signal->group_exit_task = NULL;
	current->signal->flags = SIGNAL_GROUP_EXIT;
	spin_unlock_irq(&current->sighand->siglock);

486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501
	next = mm->core_state->dumper.next;
	while ((curr = next) != NULL) {
		next = curr->next;
		task = curr->task;
		/*
		 * see exit_mm(), curr->task must not see
		 * ->task == NULL before we read ->next.
		 */
		smp_mb();
		curr->task = NULL;
		wake_up_process(task);
	}

	mm->core_state = NULL;
}

502 503 504 505 506 507 508 509 510 511 512
static bool dump_interrupted(void)
{
	/*
	 * SIGKILL or freezing() interrupt the coredumping. Perhaps we
	 * can do try_to_freeze() and check __fatal_signal_pending(),
	 * but then we need to teach dump_write() to restart and clear
	 * TIF_SIGPENDING.
	 */
	return signal_pending(current);
}

513 514
static void wait_for_dump_helpers(struct file *file)
{
515
	struct pipe_inode_info *pipe = file->private_data;
516 517 518 519

	pipe_lock(pipe);
	pipe->readers++;
	pipe->writers--;
520 521 522
	wake_up_interruptible_sync(&pipe->wait);
	kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN);
	pipe_unlock(pipe);
523

524 525 526 527 528
	/*
	 * We actually want wait_event_freezable() but then we need
	 * to clear TIF_SIGPENDING and improve dump_interrupted().
	 */
	wait_event_interruptible(pipe->wait, pipe->readers == 1);
529

530
	pipe_lock(pipe);
531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556
	pipe->readers--;
	pipe->writers++;
	pipe_unlock(pipe);
}

/*
 * umh_pipe_setup
 * helper function to customize the process used
 * to collect the core in userspace.  Specifically
 * it sets up a pipe and installs it as fd 0 (stdin)
 * for the process.  Returns 0 on success, or
 * PTR_ERR on failure.
 * Note that it also sets the core limit to 1.  This
 * is a special value that we use to trap recursive
 * core dumps
 */
static int umh_pipe_setup(struct subprocess_info *info, struct cred *new)
{
	struct file *files[2];
	struct coredump_params *cp = (struct coredump_params *)info->data;
	int err = create_pipe_files(files, 0);
	if (err)
		return err;

	cp->file = files[1];

A
Al Viro 已提交
557 558
	err = replace_fd(0, files[0], 0);
	fput(files[0]);
559 560 561
	/* and disallow core files too */
	current->signal->rlim[RLIMIT_CORE] = (struct rlimit){1, 1};

A
Al Viro 已提交
562
	return err;
563 564
}

565
void do_coredump(const kernel_siginfo_t *siginfo)
566 567 568 569 570 571 572 573 574
{
	struct core_state core_state;
	struct core_name cn;
	struct mm_struct *mm = current->mm;
	struct linux_binfmt * binfmt;
	const struct cred *old_cred;
	struct cred *cred;
	int retval = 0;
	int ispipe;
575 576
	size_t *argv = NULL;
	int argc = 0;
577
	struct files_struct *displaced;
578 579
	/* require nonrelative corefile path and be extra careful */
	bool need_suid_safe = false;
580
	bool core_dumped = false;
581 582
	static atomic_t core_dump_count = ATOMIC_INIT(0);
	struct coredump_params cprm = {
583
		.siginfo = siginfo,
584
		.regs = signal_pt_regs(),
585 586 587 588 589 590 591 592 593
		.limit = rlimit(RLIMIT_CORE),
		/*
		 * We must use the same mm->flags while dumping core to avoid
		 * inconsistency of bit flags, since this flag is not protected
		 * by any locks.
		 */
		.mm_flags = mm->flags,
	};

594
	audit_core_dumps(siginfo->si_signo);
595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610

	binfmt = mm->binfmt;
	if (!binfmt || !binfmt->core_dump)
		goto fail;
	if (!__get_dumpable(cprm.mm_flags))
		goto fail;

	cred = prepare_creds();
	if (!cred)
		goto fail;
	/*
	 * We cannot trust fsuid as being the "true" uid of the process
	 * nor do we know its entire history. We only know it was tainted
	 * so we dump it as root in mode 2, and only into a controlled
	 * environment (pipe handler or fully qualified path).
	 */
611
	if (__get_dumpable(cprm.mm_flags) == SUID_DUMP_ROOT) {
612 613
		/* Setuid core dump mode */
		cred->fsuid = GLOBAL_ROOT_UID;	/* Dump root private */
614
		need_suid_safe = true;
615 616
	}

617
	retval = coredump_wait(siginfo->si_signo, &core_state);
618 619 620 621 622
	if (retval < 0)
		goto fail_creds;

	old_cred = override_creds(cred);

623
	ispipe = format_corename(&cn, &cprm, &argv, &argc);
624

625
	if (ispipe) {
626
		int argi;
627 628
		int dump_count;
		char **helper_argv;
629
		struct subprocess_info *sub_info;
630 631 632 633

		if (ispipe < 0) {
			printk(KERN_WARNING "format_corename failed\n");
			printk(KERN_WARNING "Aborting core\n");
634
			goto fail_unlock;
635 636 637 638 639 640 641
		}

		if (cprm.limit == 1) {
			/* See umh_pipe_setup() which sets RLIMIT_CORE = 1.
			 *
			 * Normally core limits are irrelevant to pipes, since
			 * we're not writing to the file system, but we use
B
Bastien Nocera 已提交
642
			 * cprm.limit of 1 here as a special value, this is a
643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668
			 * consistent way to catch recursive crashes.
			 * We can still crash if the core_pattern binary sets
			 * RLIM_CORE = !1, but it runs as root, and can do
			 * lots of stupid things.
			 *
			 * Note that we use task_tgid_vnr here to grab the pid
			 * of the process group leader.  That way we get the
			 * right pid if a thread in a multi-threaded
			 * core_pattern process dies.
			 */
			printk(KERN_WARNING
				"Process %d(%s) has RLIMIT_CORE set to 1\n",
				task_tgid_vnr(current), current->comm);
			printk(KERN_WARNING "Aborting core\n");
			goto fail_unlock;
		}
		cprm.limit = RLIM_INFINITY;

		dump_count = atomic_inc_return(&core_dump_count);
		if (core_pipe_limit && (core_pipe_limit < dump_count)) {
			printk(KERN_WARNING "Pid %d(%s) over core_pipe_limit\n",
			       task_tgid_vnr(current), current->comm);
			printk(KERN_WARNING "Skipping core dump\n");
			goto fail_dropcount;
		}

669 670
		helper_argv = kmalloc_array(argc + 1, sizeof(*helper_argv),
					    GFP_KERNEL);
671 672 673 674 675
		if (!helper_argv) {
			printk(KERN_WARNING "%s failed to allocate memory\n",
			       __func__);
			goto fail_dropcount;
		}
676 677 678
		for (argi = 0; argi < argc; argi++)
			helper_argv[argi] = cn.corename + argv[argi];
		helper_argv[argi] = NULL;
679

680 681 682 683 684 685 686 687
		retval = -ENOMEM;
		sub_info = call_usermodehelper_setup(helper_argv[0],
						helper_argv, NULL, GFP_KERNEL,
						umh_pipe_setup, NULL, &cprm);
		if (sub_info)
			retval = call_usermodehelper_exec(sub_info,
							  UMH_WAIT_EXEC);

688
		kfree(helper_argv);
689
		if (retval) {
690
			printk(KERN_INFO "Core dump to |%s pipe failed\n",
691 692
			       cn.corename);
			goto close_fail;
693
		}
694 695
	} else {
		struct inode *inode;
696 697
		int open_flags = O_CREAT | O_RDWR | O_NOFOLLOW |
				 O_LARGEFILE | O_EXCL;
698 699 700 701

		if (cprm.limit < binfmt->min_coredump)
			goto fail_unlock;

702
		if (need_suid_safe && cn.corename[0] != '/') {
703 704 705 706 707 708 709
			printk(KERN_WARNING "Pid %d(%s) can only dump core "\
				"to fully qualified path!\n",
				task_tgid_vnr(current), current->comm);
			printk(KERN_WARNING "Skipping core dump\n");
			goto fail_unlock;
		}

710 711 712 713 714 715 716 717 718 719
		/*
		 * Unlink the file if it exists unless this is a SUID
		 * binary - in that case, we're running around with root
		 * privs and don't want to unlink another user's coredump.
		 */
		if (!need_suid_safe) {
			/*
			 * If it doesn't exist, that's fine. If there's some
			 * other problem, we'll catch it at the filp_open().
			 */
720
			do_unlinkat(AT_FDCWD, getname_kernel(cn.corename));
721 722 723 724 725 726 727 728 729 730
		}

		/*
		 * There is a race between unlinking and creating the
		 * file, but if that causes an EEXIST here, that's
		 * fine - another process raced with us while creating
		 * the corefile, and the other process won. To userspace,
		 * what matters is that at least one of the two processes
		 * writes its coredump successfully, not which one.
		 */
731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751
		if (need_suid_safe) {
			/*
			 * Using user namespaces, normal user tasks can change
			 * their current->fs->root to point to arbitrary
			 * directories. Since the intention of the "only dump
			 * with a fully qualified path" rule is to control where
			 * coredumps may be placed using root privileges,
			 * current->fs->root must not be used. Instead, use the
			 * root directory of init_task.
			 */
			struct path root;

			task_lock(&init_task);
			get_fs_root(init_task.fs, &root);
			task_unlock(&init_task);
			cprm.file = file_open_root(root.dentry, root.mnt,
				cn.corename, open_flags, 0600);
			path_put(&root);
		} else {
			cprm.file = filp_open(cn.corename, open_flags, 0600);
		}
752 753 754
		if (IS_ERR(cprm.file))
			goto fail_unlock;

A
Al Viro 已提交
755
		inode = file_inode(cprm.file);
756 757 758 759 760 761 762 763 764 765 766
		if (inode->i_nlink > 1)
			goto close_fail;
		if (d_unhashed(cprm.file->f_path.dentry))
			goto close_fail;
		/*
		 * AK: actually i see no reason to not allow this for named
		 * pipes etc, but keep the previous behaviour for now.
		 */
		if (!S_ISREG(inode->i_mode))
			goto close_fail;
		/*
767 768 769 770
		 * Don't dump core if the filesystem changed owner or mode
		 * of the file during file creation. This is an issue when
		 * a process dumps core while its cwd is e.g. on a vfat
		 * filesystem.
771 772 773
		 */
		if (!uid_eq(inode->i_uid, current_fsuid()))
			goto close_fail;
774 775
		if ((inode->i_mode & 0677) != 0600)
			goto close_fail;
A
Al Viro 已提交
776
		if (!(cprm.file->f_mode & FMODE_CAN_WRITE))
777 778 779 780 781 782 783 784 785 786 787
			goto close_fail;
		if (do_truncate(cprm.file->f_path.dentry, 0, 0, cprm.file))
			goto close_fail;
	}

	/* get us an unshared descriptor table; almost always a no-op */
	retval = unshare_files(&displaced);
	if (retval)
		goto close_fail;
	if (displaced)
		put_files_struct(displaced);
788 789 790 791 792
	if (!dump_interrupted()) {
		file_start_write(cprm.file);
		core_dumped = binfmt->core_dump(&cprm);
		file_end_write(cprm.file);
	}
793 794 795 796 797 798 799 800 801
	if (ispipe && core_pipe_limit)
		wait_for_dump_helpers(cprm.file);
close_fail:
	if (cprm.file)
		filp_close(cprm.file, NULL);
fail_dropcount:
	if (ispipe)
		atomic_dec(&core_dump_count);
fail_unlock:
802
	kfree(argv);
803
	kfree(cn.corename);
804
	coredump_finish(mm, core_dumped);
805 806 807 808 809 810 811 812 813 814 815 816
	revert_creds(old_cred);
fail_creds:
	put_cred(cred);
fail:
	return;
}

/*
 * Core dumping helper functions.  These are the only things you should
 * do on a core-file: use only these functions to write out all the
 * necessary info.
 */
A
Al Viro 已提交
817 818 819
int dump_emit(struct coredump_params *cprm, const void *addr, int nr)
{
	struct file *file = cprm->file;
820 821
	loff_t pos = file->f_pos;
	ssize_t n;
822
	if (cprm->written + nr > cprm->limit)
A
Al Viro 已提交
823
		return 0;
824 825 826
	while (nr) {
		if (dump_interrupted())
			return 0;
827
		n = __kernel_write(file, addr, nr, &pos);
828 829 830
		if (n <= 0)
			return 0;
		file->f_pos = pos;
831
		cprm->written += n;
832
		cprm->pos += n;
833 834
		nr -= n;
	}
A
Al Viro 已提交
835 836 837 838
	return 1;
}
EXPORT_SYMBOL(dump_emit);

839
int dump_skip(struct coredump_params *cprm, size_t nr)
840
{
841 842
	static char zeroes[PAGE_SIZE];
	struct file *file = cprm->file;
843
	if (file->f_op->llseek && file->f_op->llseek != no_llseek) {
844
		if (dump_interrupted() ||
845
		    file->f_op->llseek(file, nr, SEEK_CUR) < 0)
846
			return 0;
847
		cprm->pos += nr;
848
		return 1;
849
	} else {
850 851 852 853
		while (nr > PAGE_SIZE) {
			if (!dump_emit(cprm, zeroes, PAGE_SIZE))
				return 0;
			nr -= PAGE_SIZE;
854
		}
855
		return dump_emit(cprm, zeroes, nr);
856 857
	}
}
858
EXPORT_SYMBOL(dump_skip);
A
Al Viro 已提交
859 860 861

int dump_align(struct coredump_params *cprm, int align)
{
862
	unsigned mod = cprm->pos & (align - 1);
A
Al Viro 已提交
863
	if (align & (align - 1))
A
Al Viro 已提交
864 865
		return 0;
	return mod ? dump_skip(cprm, align - mod) : 1;
A
Al Viro 已提交
866 867
}
EXPORT_SYMBOL(dump_align);
868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885

/*
 * Ensures that file size is big enough to contain the current file
 * postion. This prevents gdb from complaining about a truncated file
 * if the last "write" to the file was dump_skip.
 */
void dump_truncate(struct coredump_params *cprm)
{
	struct file *file = cprm->file;
	loff_t offset;

	if (file->f_op->llseek && file->f_op->llseek != no_llseek) {
		offset = file->f_op->llseek(file, 0, SEEK_CUR);
		if (i_size_read(file->f_mapping->host) < offset)
			do_truncate(file->f_path.dentry, offset, 0, file);
	}
}
EXPORT_SYMBOL(dump_truncate);