exec.c 42.0 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
/*
 *  linux/fs/exec.c
 *
 *  Copyright (C) 1991, 1992  Linus Torvalds
 */

/*
 * #!-checking implemented by tytso.
 */
/*
 * Demand-loading implemented 01.12.91 - no need to read anything but
 * the header into memory. The inode of the executable is put into
 * "current->executable", and page faults do the actual loading. Clean.
 *
 * Once more I can proudly say that linux stood up to being changed: it
 * was less than 2 hours work to get demand-loading completely implemented.
 *
 * Demand loading changed July 1993 by Eric Youngdale.   Use mmap instead,
 * current->executable is only used by the procfs.  This allows a dispatch
 * table to check for several different types  of binary formats.  We keep
 * trying until we recognize the file or we run out of supported binary
 * formats. 
 */

#include <linux/slab.h>
#include <linux/file.h>
A
Al Viro 已提交
27
#include <linux/fdtable.h>
H
Hugh Dickins 已提交
28
#include <linux/mm.h>
L
Linus Torvalds 已提交
29 30 31
#include <linux/stat.h>
#include <linux/fcntl.h>
#include <linux/smp_lock.h>
H
Hugh Dickins 已提交
32
#include <linux/swap.h>
33
#include <linux/string.h>
L
Linus Torvalds 已提交
34
#include <linux/init.h>
35
#include <linux/pagemap.h>
L
Linus Torvalds 已提交
36 37 38 39 40 41
#include <linux/highmem.h>
#include <linux/spinlock.h>
#include <linux/key.h>
#include <linux/personality.h>
#include <linux/binfmts.h>
#include <linux/utsname.h>
42
#include <linux/pid_namespace.h>
L
Linus Torvalds 已提交
43 44 45 46 47 48
#include <linux/module.h>
#include <linux/namei.h>
#include <linux/proc_fs.h>
#include <linux/mount.h>
#include <linux/security.h>
#include <linux/syscalls.h>
49
#include <linux/tsacct_kern.h>
M
Matt Helsley 已提交
50
#include <linux/cn_proc.h>
A
Al Viro 已提交
51
#include <linux/audit.h>
R
Roland McGrath 已提交
52
#include <linux/tracehook.h>
J
Johannes Berg 已提交
53
#include <linux/kmod.h>
L
Linus Torvalds 已提交
54 55 56

#include <asm/uaccess.h>
#include <asm/mmu_context.h>
57
#include <asm/tlb.h>
L
Linus Torvalds 已提交
58

59 60 61 62 63
#ifdef __alpha__
/* for /sbin/loader handling in search_binary_handler() */
#include <linux/a.out.h>
#endif

L
Linus Torvalds 已提交
64
int core_uses_pid;
65
char core_pattern[CORENAME_MAX_SIZE] = "core";
A
Alan Cox 已提交
66 67
int suid_dumpable = 0;

L
Linus Torvalds 已提交
68 69
/* The maximal length of core_pattern is also specified in sysctl.c */

A
Alexey Dobriyan 已提交
70
static LIST_HEAD(formats);
L
Linus Torvalds 已提交
71 72 73 74 75 76 77
static DEFINE_RWLOCK(binfmt_lock);

int register_binfmt(struct linux_binfmt * fmt)
{
	if (!fmt)
		return -EINVAL;
	write_lock(&binfmt_lock);
A
Alexey Dobriyan 已提交
78
	list_add(&fmt->lh, &formats);
L
Linus Torvalds 已提交
79 80 81 82 83 84
	write_unlock(&binfmt_lock);
	return 0;	
}

EXPORT_SYMBOL(register_binfmt);

85
void unregister_binfmt(struct linux_binfmt * fmt)
L
Linus Torvalds 已提交
86 87
{
	write_lock(&binfmt_lock);
A
Alexey Dobriyan 已提交
88
	list_del(&fmt->lh);
L
Linus Torvalds 已提交
89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106
	write_unlock(&binfmt_lock);
}

EXPORT_SYMBOL(unregister_binfmt);

static inline void put_binfmt(struct linux_binfmt * fmt)
{
	module_put(fmt->module);
}

/*
 * Note that a shared library must be both readable and executable due to
 * security reasons.
 *
 * Also note that we take the address to load from from the file itself.
 */
asmlinkage long sys_uselib(const char __user * library)
{
107
	struct file *file;
L
Linus Torvalds 已提交
108
	struct nameidata nd;
109 110 111 112 113 114 115 116 117
	char *tmp = getname(library);
	int error = PTR_ERR(tmp);

	if (!IS_ERR(tmp)) {
		error = path_lookup_open(AT_FDCWD, tmp,
					 LOOKUP_FOLLOW, &nd,
					 FMODE_READ|FMODE_EXEC);
		putname(tmp);
	}
L
Linus Torvalds 已提交
118 119 120 121
	if (error)
		goto out;

	error = -EINVAL;
122
	if (!S_ISREG(nd.path.dentry->d_inode->i_mode))
L
Linus Torvalds 已提交
123 124
		goto exit;

125 126 127 128
	error = -EACCES;
	if (nd.path.mnt->mnt_flags & MNT_NOEXEC)
		goto exit;

129
	error = vfs_permission(&nd, MAY_READ | MAY_EXEC | MAY_OPEN);
L
Linus Torvalds 已提交
130 131 132
	if (error)
		goto exit;

A
Andi Kleen 已提交
133
	file = nameidata_to_filp(&nd, O_RDONLY|O_LARGEFILE);
L
Linus Torvalds 已提交
134 135 136 137 138 139 140 141 142
	error = PTR_ERR(file);
	if (IS_ERR(file))
		goto out;

	error = -ENOEXEC;
	if(file->f_op) {
		struct linux_binfmt * fmt;

		read_lock(&binfmt_lock);
A
Alexey Dobriyan 已提交
143
		list_for_each_entry(fmt, &formats, lh) {
L
Linus Torvalds 已提交
144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160
			if (!fmt->load_shlib)
				continue;
			if (!try_module_get(fmt->module))
				continue;
			read_unlock(&binfmt_lock);
			error = fmt->load_shlib(file);
			read_lock(&binfmt_lock);
			put_binfmt(fmt);
			if (error != -ENOEXEC)
				break;
		}
		read_unlock(&binfmt_lock);
	}
	fput(file);
out:
  	return error;
exit:
161
	release_open_intent(&nd);
J
Jan Blunck 已提交
162
	path_put(&nd.path);
L
Linus Torvalds 已提交
163 164 165
	goto out;
}

166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187
#ifdef CONFIG_MMU

static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos,
		int write)
{
	struct page *page;
	int ret;

#ifdef CONFIG_STACK_GROWSUP
	if (write) {
		ret = expand_stack_downwards(bprm->vma, pos);
		if (ret < 0)
			return NULL;
	}
#endif
	ret = get_user_pages(current, bprm->mm, pos,
			1, write, 1, &page, NULL);
	if (ret <= 0)
		return NULL;

	if (write) {
		unsigned long size = bprm->vma->vm_end - bprm->vma->vm_start;
188 189 190 191 192 193 194 195
		struct rlimit *rlim;

		/*
		 * We've historically supported up to 32 pages (ARG_MAX)
		 * of argument strings even with small stacks
		 */
		if (size <= ARG_MAX)
			return page;
196 197 198 199 200 201 202 203

		/*
		 * Limit to 1/4-th the stack size for the argv+env strings.
		 * This ensures that:
		 *  - the remaining binfmt code will not run out of stack space,
		 *  - the program will have a reasonable amount of stack left
		 *    to work from.
		 */
204
		rlim = current->signal->rlim;
205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255
		if (size > rlim[RLIMIT_STACK].rlim_cur / 4) {
			put_page(page);
			return NULL;
		}
	}

	return page;
}

static void put_arg_page(struct page *page)
{
	put_page(page);
}

static void free_arg_page(struct linux_binprm *bprm, int i)
{
}

static void free_arg_pages(struct linux_binprm *bprm)
{
}

static void flush_arg_page(struct linux_binprm *bprm, unsigned long pos,
		struct page *page)
{
	flush_cache_page(bprm->vma, pos, page_to_pfn(page));
}

static int __bprm_mm_init(struct linux_binprm *bprm)
{
	int err = -ENOMEM;
	struct vm_area_struct *vma = NULL;
	struct mm_struct *mm = bprm->mm;

	bprm->vma = vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL);
	if (!vma)
		goto err;

	down_write(&mm->mmap_sem);
	vma->vm_mm = mm;

	/*
	 * Place the stack at the largest stack address the architecture
	 * supports. Later, we'll move this to an appropriate place. We don't
	 * use STACK_TOP because that can depend on attributes which aren't
	 * configured yet.
	 */
	vma->vm_end = STACK_TOP_MAX;
	vma->vm_start = vma->vm_end - PAGE_SIZE;

	vma->vm_flags = VM_STACK_FLAGS;
256
	vma->vm_page_prot = vm_get_page_prot(vma->vm_flags);
257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374
	err = insert_vm_struct(mm, vma);
	if (err) {
		up_write(&mm->mmap_sem);
		goto err;
	}

	mm->stack_vm = mm->total_vm = 1;
	up_write(&mm->mmap_sem);

	bprm->p = vma->vm_end - sizeof(void *);

	return 0;

err:
	if (vma) {
		bprm->vma = NULL;
		kmem_cache_free(vm_area_cachep, vma);
	}

	return err;
}

static bool valid_arg_len(struct linux_binprm *bprm, long len)
{
	return len <= MAX_ARG_STRLEN;
}

#else

static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos,
		int write)
{
	struct page *page;

	page = bprm->page[pos / PAGE_SIZE];
	if (!page && write) {
		page = alloc_page(GFP_HIGHUSER|__GFP_ZERO);
		if (!page)
			return NULL;
		bprm->page[pos / PAGE_SIZE] = page;
	}

	return page;
}

static void put_arg_page(struct page *page)
{
}

static void free_arg_page(struct linux_binprm *bprm, int i)
{
	if (bprm->page[i]) {
		__free_page(bprm->page[i]);
		bprm->page[i] = NULL;
	}
}

static void free_arg_pages(struct linux_binprm *bprm)
{
	int i;

	for (i = 0; i < MAX_ARG_PAGES; i++)
		free_arg_page(bprm, i);
}

static void flush_arg_page(struct linux_binprm *bprm, unsigned long pos,
		struct page *page)
{
}

static int __bprm_mm_init(struct linux_binprm *bprm)
{
	bprm->p = PAGE_SIZE * MAX_ARG_PAGES - sizeof(void *);
	return 0;
}

static bool valid_arg_len(struct linux_binprm *bprm, long len)
{
	return len <= bprm->p;
}

#endif /* CONFIG_MMU */

/*
 * Create a new mm_struct and populate it with a temporary stack
 * vm_area_struct.  We don't have enough context at this point to set the stack
 * flags, permissions, and offset, so we use temporary values.  We'll update
 * them later in setup_arg_pages().
 */
int bprm_mm_init(struct linux_binprm *bprm)
{
	int err;
	struct mm_struct *mm = NULL;

	bprm->mm = mm = mm_alloc();
	err = -ENOMEM;
	if (!mm)
		goto err;

	err = init_new_context(current, mm);
	if (err)
		goto err;

	err = __bprm_mm_init(bprm);
	if (err)
		goto err;

	return 0;

err:
	if (mm) {
		bprm->mm = NULL;
		mmdrop(mm);
	}

	return err;
}

L
Linus Torvalds 已提交
375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390
/*
 * count() counts the number of strings in array ARGV.
 */
static int count(char __user * __user * argv, int max)
{
	int i = 0;

	if (argv != NULL) {
		for (;;) {
			char __user * p;

			if (get_user(p, argv))
				return -EFAULT;
			if (!p)
				break;
			argv++;
391
			if (i++ >= max)
L
Linus Torvalds 已提交
392 393 394 395 396 397 398 399
				return -E2BIG;
			cond_resched();
		}
	}
	return i;
}

/*
400 401 402
 * 'copy_strings()' copies argument/environment strings from the old
 * processes's memory to the new process's stack.  The call to get_user_pages()
 * ensures the destination page is created and not swapped out.
L
Linus Torvalds 已提交
403
 */
A
Adrian Bunk 已提交
404 405
static int copy_strings(int argc, char __user * __user * argv,
			struct linux_binprm *bprm)
L
Linus Torvalds 已提交
406 407 408
{
	struct page *kmapped_page = NULL;
	char *kaddr = NULL;
409
	unsigned long kpos = 0;
L
Linus Torvalds 已提交
410 411 412 413 414 415 416 417
	int ret;

	while (argc-- > 0) {
		char __user *str;
		int len;
		unsigned long pos;

		if (get_user(str, argv+argc) ||
418
				!(len = strnlen_user(str, MAX_ARG_STRLEN))) {
L
Linus Torvalds 已提交
419 420 421 422
			ret = -EFAULT;
			goto out;
		}

423
		if (!valid_arg_len(bprm, len)) {
L
Linus Torvalds 已提交
424 425 426 427
			ret = -E2BIG;
			goto out;
		}

428
		/* We're going to work our way backwords. */
L
Linus Torvalds 已提交
429
		pos = bprm->p;
430 431
		str += len;
		bprm->p -= len;
L
Linus Torvalds 已提交
432 433 434 435 436

		while (len > 0) {
			int offset, bytes_to_copy;

			offset = pos % PAGE_SIZE;
437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452
			if (offset == 0)
				offset = PAGE_SIZE;

			bytes_to_copy = offset;
			if (bytes_to_copy > len)
				bytes_to_copy = len;

			offset -= bytes_to_copy;
			pos -= bytes_to_copy;
			str -= bytes_to_copy;
			len -= bytes_to_copy;

			if (!kmapped_page || kpos != (pos & PAGE_MASK)) {
				struct page *page;

				page = get_arg_page(bprm, pos, 1);
L
Linus Torvalds 已提交
453
				if (!page) {
454
					ret = -E2BIG;
L
Linus Torvalds 已提交
455 456 457
					goto out;
				}

458 459
				if (kmapped_page) {
					flush_kernel_dcache_page(kmapped_page);
L
Linus Torvalds 已提交
460
					kunmap(kmapped_page);
461 462
					put_arg_page(kmapped_page);
				}
L
Linus Torvalds 已提交
463 464
				kmapped_page = page;
				kaddr = kmap(kmapped_page);
465 466
				kpos = pos & PAGE_MASK;
				flush_arg_page(bprm, kpos, kmapped_page);
L
Linus Torvalds 已提交
467
			}
468
			if (copy_from_user(kaddr+offset, str, bytes_to_copy)) {
L
Linus Torvalds 已提交
469 470 471 472 473 474 475
				ret = -EFAULT;
				goto out;
			}
		}
	}
	ret = 0;
out:
476 477
	if (kmapped_page) {
		flush_kernel_dcache_page(kmapped_page);
L
Linus Torvalds 已提交
478
		kunmap(kmapped_page);
479 480
		put_arg_page(kmapped_page);
	}
L
Linus Torvalds 已提交
481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498
	return ret;
}

/*
 * Like copy_strings, but get argv and its values from kernel memory.
 */
int copy_strings_kernel(int argc,char ** argv, struct linux_binprm *bprm)
{
	int r;
	mm_segment_t oldfs = get_fs();
	set_fs(KERNEL_DS);
	r = copy_strings(argc, (char __user * __user *)argv, bprm);
	set_fs(oldfs);
	return r;
}
EXPORT_SYMBOL(copy_strings_kernel);

#ifdef CONFIG_MMU
499

L
Linus Torvalds 已提交
500
/*
501 502 503
 * During bprm_mm_init(), we create a temporary stack at STACK_TOP_MAX.  Once
 * the binfmt code determines where the new stack should reside, we shift it to
 * its final location.  The process proceeds as follows:
L
Linus Torvalds 已提交
504
 *
505 506 507 508 509 510
 * 1) Use shift to calculate the new vma endpoints.
 * 2) Extend vma to cover both the old and new ranges.  This ensures the
 *    arguments passed to subsequent functions are consistent.
 * 3) Move vma's page tables to the new range.
 * 4) Free up any cleared pgd range.
 * 5) Shrink the vma to cover only the new range.
L
Linus Torvalds 已提交
511
 */
512
static int shift_arg_pages(struct vm_area_struct *vma, unsigned long shift)
L
Linus Torvalds 已提交
513 514
{
	struct mm_struct *mm = vma->vm_mm;
515 516 517 518 519 520
	unsigned long old_start = vma->vm_start;
	unsigned long old_end = vma->vm_end;
	unsigned long length = old_end - old_start;
	unsigned long new_start = old_start - shift;
	unsigned long new_end = old_end - shift;
	struct mmu_gather *tlb;
L
Linus Torvalds 已提交
521

522
	BUG_ON(new_start > new_end);
L
Linus Torvalds 已提交
523

524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549
	/*
	 * ensure there are no vmas between where we want to go
	 * and where we are
	 */
	if (vma != find_vma(mm, new_start))
		return -EFAULT;

	/*
	 * cover the whole range: [new_start, old_end)
	 */
	vma_adjust(vma, new_start, old_end, vma->vm_pgoff, NULL);

	/*
	 * move the page tables downwards, on failure we rely on
	 * process cleanup to remove whatever mess we made.
	 */
	if (length != move_page_tables(vma, old_start,
				       vma, new_start, length))
		return -ENOMEM;

	lru_add_drain();
	tlb = tlb_gather_mmu(mm, 0);
	if (new_end > old_start) {
		/*
		 * when the old and new regions overlap clear from new_end.
		 */
550
		free_pgd_range(tlb, new_end, old_end, new_end,
551 552 553 554 555 556 557 558
			vma->vm_next ? vma->vm_next->vm_start : 0);
	} else {
		/*
		 * otherwise, clean from old_start; this is done to not touch
		 * the address space in [new_end, old_start) some architectures
		 * have constraints on va-space that make this illegal (IA64) -
		 * for the others its just a little faster.
		 */
559
		free_pgd_range(tlb, old_start, old_end, new_end,
560
			vma->vm_next ? vma->vm_next->vm_start : 0);
L
Linus Torvalds 已提交
561
	}
562 563 564 565 566 567 568 569
	tlb_finish_mmu(tlb, new_end, old_end);

	/*
	 * shrink the vma to just the new range.
	 */
	vma_adjust(vma, new_start, new_end, vma->vm_pgoff, NULL);

	return 0;
L
Linus Torvalds 已提交
570 571 572 573
}

#define EXTRA_STACK_VM_PAGES	20	/* random */

574 575 576 577
/*
 * Finalizes the stack vm_area_struct. The flags and permissions are updated,
 * the stack is optionally relocated, and some extra space is added.
 */
L
Linus Torvalds 已提交
578 579 580 581
int setup_arg_pages(struct linux_binprm *bprm,
		    unsigned long stack_top,
		    int executable_stack)
{
582 583
	unsigned long ret;
	unsigned long stack_shift;
L
Linus Torvalds 已提交
584
	struct mm_struct *mm = current->mm;
585 586 587 588
	struct vm_area_struct *vma = bprm->vma;
	struct vm_area_struct *prev = NULL;
	unsigned long vm_flags;
	unsigned long stack_base;
L
Linus Torvalds 已提交
589 590 591 592 593 594 595

#ifdef CONFIG_STACK_GROWSUP
	/* Limit stack size to 1GB */
	stack_base = current->signal->rlim[RLIMIT_STACK].rlim_max;
	if (stack_base > (1 << 30))
		stack_base = 1 << 30;

596 597 598
	/* Make sure we didn't let the argument array grow too large. */
	if (vma->vm_end - vma->vm_start > stack_base)
		return -ENOMEM;
L
Linus Torvalds 已提交
599

600
	stack_base = PAGE_ALIGN(stack_top - stack_base);
L
Linus Torvalds 已提交
601

602 603 604
	stack_shift = vma->vm_start - stack_base;
	mm->arg_start = bprm->p - stack_shift;
	bprm->p = vma->vm_end - stack_shift;
L
Linus Torvalds 已提交
605
#else
606 607 608 609 610
	stack_top = arch_align_stack(stack_top);
	stack_top = PAGE_ALIGN(stack_top);
	stack_shift = vma->vm_end - stack_top;

	bprm->p -= stack_shift;
L
Linus Torvalds 已提交
611 612 613 614
	mm->arg_start = bprm->p;
#endif

	if (bprm->loader)
615 616
		bprm->loader -= stack_shift;
	bprm->exec -= stack_shift;
L
Linus Torvalds 已提交
617 618

	down_write(&mm->mmap_sem);
619
	vm_flags = VM_STACK_FLAGS;
620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641

	/*
	 * Adjust stack execute permissions; explicitly enable for
	 * EXSTACK_ENABLE_X, disable for EXSTACK_DISABLE_X and leave alone
	 * (arch default) otherwise.
	 */
	if (unlikely(executable_stack == EXSTACK_ENABLE_X))
		vm_flags |= VM_EXEC;
	else if (executable_stack == EXSTACK_DISABLE_X)
		vm_flags &= ~VM_EXEC;
	vm_flags |= mm->def_flags;

	ret = mprotect_fixup(vma, &prev, vma->vm_start, vma->vm_end,
			vm_flags);
	if (ret)
		goto out_unlock;
	BUG_ON(prev != vma);

	/* Move stack pages down in memory. */
	if (stack_shift) {
		ret = shift_arg_pages(vma, stack_shift);
		if (ret) {
L
Linus Torvalds 已提交
642 643 644 645 646
			up_write(&mm->mmap_sem);
			return ret;
		}
	}

647 648 649 650 651 652 653 654 655 656
#ifdef CONFIG_STACK_GROWSUP
	stack_base = vma->vm_end + EXTRA_STACK_VM_PAGES * PAGE_SIZE;
#else
	stack_base = vma->vm_start - EXTRA_STACK_VM_PAGES * PAGE_SIZE;
#endif
	ret = expand_stack(vma, stack_base);
	if (ret)
		ret = -EFAULT;

out_unlock:
L
Linus Torvalds 已提交
657 658 659 660 661 662 663 664 665 666 667
	up_write(&mm->mmap_sem);
	return 0;
}
EXPORT_SYMBOL(setup_arg_pages);

#endif /* CONFIG_MMU */

struct file *open_exec(const char *name)
{
	struct nameidata nd;
	struct file *file;
668
	int err;
L
Linus Torvalds 已提交
669

670 671 672 673 674 675 676 677 678
	err = path_lookup_open(AT_FDCWD, name, LOOKUP_FOLLOW, &nd,
				FMODE_READ|FMODE_EXEC);
	if (err)
		goto out;

	err = -EACCES;
	if (!S_ISREG(nd.path.dentry->d_inode->i_mode))
		goto out_path_put;

679 680 681
	if (nd.path.mnt->mnt_flags & MNT_NOEXEC)
		goto out_path_put;

682 683 684 685 686 687 688 689 690 691 692 693
	err = vfs_permission(&nd, MAY_EXEC | MAY_OPEN);
	if (err)
		goto out_path_put;

	file = nameidata_to_filp(&nd, O_RDONLY|O_LARGEFILE);
	if (IS_ERR(file))
		return file;

	err = deny_write_access(file);
	if (err) {
		fput(file);
		goto out;
L
Linus Torvalds 已提交
694 695
	}

696 697 698 699 700 701 702 703
	return file;

 out_path_put:
	release_open_intent(&nd);
	path_put(&nd.path);
 out:
	return ERR_PTR(err);
}
L
Linus Torvalds 已提交
704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737
EXPORT_SYMBOL(open_exec);

int kernel_read(struct file *file, unsigned long offset,
	char *addr, unsigned long count)
{
	mm_segment_t old_fs;
	loff_t pos = offset;
	int result;

	old_fs = get_fs();
	set_fs(get_ds());
	/* The cast to a user pointer is valid due to the set_fs() */
	result = vfs_read(file, (void __user *)addr, count, &pos);
	set_fs(old_fs);
	return result;
}

EXPORT_SYMBOL(kernel_read);

static int exec_mmap(struct mm_struct *mm)
{
	struct task_struct *tsk;
	struct mm_struct * old_mm, *active_mm;

	/* Notify parent that we're no longer interested in the old VM */
	tsk = current;
	old_mm = current->mm;
	mm_release(tsk, old_mm);

	if (old_mm) {
		/*
		 * Make sure that if there is a core dump in progress
		 * for the old mm, we get out and die instead of going
		 * through with the exec.  We must hold mmap_sem around
738
		 * checking core_state and changing tsk->mm.
L
Linus Torvalds 已提交
739 740
		 */
		down_read(&old_mm->mmap_sem);
741
		if (unlikely(old_mm->core_state)) {
L
Linus Torvalds 已提交
742 743 744 745 746 747 748 749 750 751 752 753 754
			up_read(&old_mm->mmap_sem);
			return -EINTR;
		}
	}
	task_lock(tsk);
	active_mm = tsk->active_mm;
	tsk->mm = mm;
	tsk->active_mm = mm;
	activate_mm(active_mm, mm);
	task_unlock(tsk);
	arch_pick_mmap_layout(mm);
	if (old_mm) {
		up_read(&old_mm->mmap_sem);
755
		BUG_ON(active_mm != old_mm);
756
		mm_update_next_owner(old_mm);
L
Linus Torvalds 已提交
757 758 759 760 761 762 763 764 765 766 767 768 769
		mmput(old_mm);
		return 0;
	}
	mmdrop(active_mm);
	return 0;
}

/*
 * This function makes sure the current process has its own signal table,
 * so that flush_signal_handlers can later reset the handlers without
 * disturbing other processes.  (Other processes might share the signal
 * table via the CLONE_SIGHAND option to clone().)
 */
770
static int de_thread(struct task_struct *tsk)
L
Linus Torvalds 已提交
771 772
{
	struct signal_struct *sig = tsk->signal;
773
	struct sighand_struct *oldsighand = tsk->sighand;
L
Linus Torvalds 已提交
774
	spinlock_t *lock = &oldsighand->siglock;
775
	struct task_struct *leader = NULL;
L
Linus Torvalds 已提交
776 777
	int count;

778
	if (thread_group_empty(tsk))
L
Linus Torvalds 已提交
779 780 781 782 783 784
		goto no_thread_group;

	/*
	 * Kill all other threads in the thread group.
	 */
	spin_lock_irq(lock);
785
	if (signal_group_exit(sig)) {
L
Linus Torvalds 已提交
786 787 788 789 790 791 792
		/*
		 * Another group action in progress, just
		 * return so that the signal is processed.
		 */
		spin_unlock_irq(lock);
		return -EAGAIN;
	}
793
	sig->group_exit_task = tsk;
794
	zap_other_threads(tsk);
L
Linus Torvalds 已提交
795

796 797
	/* Account for the thread group leader hanging around: */
	count = thread_group_leader(tsk) ? 1 : 2;
798
	sig->notify_count = count;
L
Linus Torvalds 已提交
799 800 801 802 803 804 805 806 807 808 809 810 811
	while (atomic_read(&sig->count) > count) {
		__set_current_state(TASK_UNINTERRUPTIBLE);
		spin_unlock_irq(lock);
		schedule();
		spin_lock_irq(lock);
	}
	spin_unlock_irq(lock);

	/*
	 * At this point all other threads have exited, all we have to
	 * do is to wait for the thread group leader to become inactive,
	 * and to assume its PID:
	 */
812 813
	if (!thread_group_leader(tsk)) {
		leader = tsk->group_leader;
814

815
		sig->notify_count = -1;	/* for exit_notify() */
816 817 818 819 820 821 822 823
		for (;;) {
			write_lock_irq(&tasklist_lock);
			if (likely(leader->exit_state))
				break;
			__set_current_state(TASK_UNINTERRUPTIBLE);
			write_unlock_irq(&tasklist_lock);
			schedule();
		}
L
Linus Torvalds 已提交
824

825 826 827 828 829 830 831 832 833 834
		/*
		 * The only record we have of the real-time age of a
		 * process, regardless of execs it's done, is start_time.
		 * All the past CPU time is accumulated in signal_struct
		 * from sister threads now dead.  But in this non-leader
		 * exec, nothing survives from the original leader thread,
		 * whose birth marks the true age of this process now.
		 * When we take on its identity by switching to its PID, we
		 * also take its birthdate (always earlier than our own).
		 */
835
		tsk->start_time = leader->start_time;
836

837 838
		BUG_ON(!same_thread_group(leader, tsk));
		BUG_ON(has_group_leader_pid(tsk));
L
Linus Torvalds 已提交
839 840 841 842 843 844
		/*
		 * An exec() starts a new thread group with the
		 * TGID of the previous thread group. Rehash the
		 * two threads with a switched PID, and release
		 * the former thread group leader:
		 */
845 846

		/* Become a process group leader with the old leader's pid.
847 848
		 * The old leader becomes a thread of the this thread group.
		 * Note: The old leader also uses this pid until release_task
849 850
		 *       is called.  Odd but simple and correct.
		 */
851 852
		detach_pid(tsk, PIDTYPE_PID);
		tsk->pid = leader->pid;
853
		attach_pid(tsk, PIDTYPE_PID,  task_pid(leader));
854 855 856
		transfer_pid(leader, tsk, PIDTYPE_PGID);
		transfer_pid(leader, tsk, PIDTYPE_SID);
		list_replace_rcu(&leader->tasks, &tsk->tasks);
L
Linus Torvalds 已提交
857

858 859
		tsk->group_leader = tsk;
		leader->group_leader = tsk;
860

861
		tsk->exit_signal = SIGCHLD;
862 863 864

		BUG_ON(leader->exit_state != EXIT_ZOMBIE);
		leader->exit_state = EXIT_DEAD;
L
Linus Torvalds 已提交
865 866

		write_unlock_irq(&tasklist_lock);
867
	}
L
Linus Torvalds 已提交
868

869 870
	sig->group_exit_task = NULL;
	sig->notify_count = 0;
L
Linus Torvalds 已提交
871 872 873

no_thread_group:
	exit_itimers(sig);
874
	flush_itimer_signals();
875 876 877
	if (leader)
		release_task(leader);

878 879
	if (atomic_read(&oldsighand->count) != 1) {
		struct sighand_struct *newsighand;
L
Linus Torvalds 已提交
880
		/*
881 882
		 * This ->sighand is shared with the CLONE_SIGHAND
		 * but not CLONE_THREAD task, switch to the new one.
L
Linus Torvalds 已提交
883
		 */
884 885 886 887
		newsighand = kmem_cache_alloc(sighand_cachep, GFP_KERNEL);
		if (!newsighand)
			return -ENOMEM;

L
Linus Torvalds 已提交
888 889 890 891 892 893
		atomic_set(&newsighand->count, 1);
		memcpy(newsighand->action, oldsighand->action,
		       sizeof(newsighand->action));

		write_lock_irq(&tasklist_lock);
		spin_lock(&oldsighand->siglock);
894
		rcu_assign_pointer(tsk->sighand, newsighand);
L
Linus Torvalds 已提交
895 896 897
		spin_unlock(&oldsighand->siglock);
		write_unlock_irq(&tasklist_lock);

898
		__cleanup_sighand(oldsighand);
L
Linus Torvalds 已提交
899 900
	}

901
	BUG_ON(!thread_group_leader(tsk));
L
Linus Torvalds 已提交
902 903
	return 0;
}
O
Oleg Nesterov 已提交
904

L
Linus Torvalds 已提交
905 906 907 908
/*
 * These functions flushes out all traces of the currently running executable
 * so that a new one can be started
 */
909
static void flush_old_files(struct files_struct * files)
L
Linus Torvalds 已提交
910 911
{
	long j = -1;
912
	struct fdtable *fdt;
L
Linus Torvalds 已提交
913 914 915 916 917 918 919

	spin_lock(&files->file_lock);
	for (;;) {
		unsigned long set, i;

		j++;
		i = j * __NFDBITS;
920
		fdt = files_fdtable(files);
921
		if (i >= fdt->max_fds)
L
Linus Torvalds 已提交
922
			break;
923
		set = fdt->close_on_exec->fds_bits[j];
L
Linus Torvalds 已提交
924 925
		if (!set)
			continue;
926
		fdt->close_on_exec->fds_bits[j] = 0;
L
Linus Torvalds 已提交
927 928 929 930 931 932 933 934 935 936 937 938
		spin_unlock(&files->file_lock);
		for ( ; set ; i++,set >>= 1) {
			if (set & 1) {
				sys_close(i);
			}
		}
		spin_lock(&files->file_lock);

	}
	spin_unlock(&files->file_lock);
}

939
char *get_task_comm(char *buf, struct task_struct *tsk)
L
Linus Torvalds 已提交
940 941 942 943 944
{
	/* buf must be at least sizeof(tsk->comm) in size */
	task_lock(tsk);
	strncpy(buf, tsk->comm, sizeof(tsk->comm));
	task_unlock(tsk);
945
	return buf;
L
Linus Torvalds 已提交
946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968
}

void set_task_comm(struct task_struct *tsk, char *buf)
{
	task_lock(tsk);
	strlcpy(tsk->comm, buf, sizeof(tsk->comm));
	task_unlock(tsk);
}

int flush_old_exec(struct linux_binprm * bprm)
{
	char * name;
	int i, ch, retval;
	char tcomm[sizeof(current->comm)];

	/*
	 * Make sure we have a private signal table and that
	 * we are unassociated from the previous thread group.
	 */
	retval = de_thread(current);
	if (retval)
		goto out;

M
Matt Helsley 已提交
969 970
	set_mm_exe_file(bprm->mm, bprm->file);

L
Linus Torvalds 已提交
971 972 973 974 975
	/*
	 * Release all of the old mmap stuff
	 */
	retval = exec_mmap(bprm->mm);
	if (retval)
976
		goto out;
L
Linus Torvalds 已提交
977 978 979 980 981 982

	bprm->mm = NULL;		/* We're using it now */

	/* This is the point of no return */
	current->sas_ss_sp = current->sas_ss_size = 0;

983
	if (current_euid() == current_uid() && current_egid() == current_gid())
984
		set_dumpable(current->mm, 1);
A
Alan Cox 已提交
985
	else
986
		set_dumpable(current->mm, suid_dumpable);
A
Alan Cox 已提交
987

L
Linus Torvalds 已提交
988
	name = bprm->filename;
989 990

	/* Copies the binary name from after last slash */
L
Linus Torvalds 已提交
991 992
	for (i=0; (ch = *(name++)) != '\0';) {
		if (ch == '/')
993
			i = 0; /* overwrite what we wrote */
L
Linus Torvalds 已提交
994 995 996 997 998 999 1000 1001 1002 1003
		else
			if (i < (sizeof(tcomm) - 1))
				tcomm[i++] = ch;
	}
	tcomm[i] = '\0';
	set_task_comm(current, tcomm);

	current->flags &= ~PF_RANDOMIZE;
	flush_thread();

1004 1005 1006 1007 1008 1009
	/* Set the new mm task size. We have to do that late because it may
	 * depend on TIF_32BIT which is only updated in flush_thread() on
	 * some architectures like powerpc
	 */
	current->mm->task_size = TASK_SIZE;

1010
	if (bprm->e_uid != current_euid() || bprm->e_gid != current_egid()) {
1011 1012 1013 1014 1015
		suid_keys(current);
		set_dumpable(current->mm, suid_dumpable);
		current->pdeath_signal = 0;
	} else if (file_permission(bprm->file, MAY_READ) ||
			(bprm->interp_flags & BINPRM_FLAGS_ENFORCE_NONDUMP)) {
L
Linus Torvalds 已提交
1016
		suid_keys(current);
1017
		set_dumpable(current->mm, suid_dumpable);
L
Linus Torvalds 已提交
1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042
	}

	/* An exec changes our domain. We are no longer part of the thread
	   group */

	current->self_exec_id++;
			
	flush_signal_handlers(current, 0);
	flush_old_files(current->files);

	return 0;

out:
	return retval;
}

EXPORT_SYMBOL(flush_old_exec);

/* 
 * Fill the binprm structure from the inode. 
 * Check permissions, then read the first 128 (BINPRM_BUF_SIZE) bytes
 */
int prepare_binprm(struct linux_binprm *bprm)
{
	int mode;
1043
	struct inode * inode = bprm->file->f_path.dentry->d_inode;
L
Linus Torvalds 已提交
1044 1045 1046 1047 1048 1049
	int retval;

	mode = inode->i_mode;
	if (bprm->file->f_op == NULL)
		return -EACCES;

1050 1051
	bprm->e_uid = current_euid();
	bprm->e_gid = current_egid();
L
Linus Torvalds 已提交
1052

1053
	if(!(bprm->file->f_path.mnt->mnt_flags & MNT_NOSUID)) {
L
Linus Torvalds 已提交
1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082
		/* Set-uid? */
		if (mode & S_ISUID) {
			current->personality &= ~PER_CLEAR_ON_SETID;
			bprm->e_uid = inode->i_uid;
		}

		/* Set-gid? */
		/*
		 * If setgid is set but no group execute bit then this
		 * is a candidate for mandatory locking, not a setgid
		 * executable.
		 */
		if ((mode & (S_ISGID | S_IXGRP)) == (S_ISGID | S_IXGRP)) {
			current->personality &= ~PER_CLEAR_ON_SETID;
			bprm->e_gid = inode->i_gid;
		}
	}

	/* fill in binprm security blob */
	retval = security_bprm_set(bprm);
	if (retval)
		return retval;

	memset(bprm->buf,0,BINPRM_BUF_SIZE);
	return kernel_read(bprm->file,0,bprm->buf,BINPRM_BUF_SIZE);
}

EXPORT_SYMBOL(prepare_binprm);

1083
static int unsafe_exec(struct task_struct *p)
L
Linus Torvalds 已提交
1084
{
R
Roland McGrath 已提交
1085 1086
	int unsafe = tracehook_unsafe_exec(p);

L
Linus Torvalds 已提交
1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098
	if (atomic_read(&p->fs->count) > 1 ||
	    atomic_read(&p->files->count) > 1 ||
	    atomic_read(&p->sighand->count) > 1)
		unsafe |= LSM_UNSAFE_SHARE;

	return unsafe;
}

void compute_creds(struct linux_binprm *bprm)
{
	int unsafe;

1099
	if (bprm->e_uid != current_uid()) {
L
Linus Torvalds 已提交
1100
		suid_keys(current);
1101 1102
		current->pdeath_signal = 0;
	}
L
Linus Torvalds 已提交
1103 1104 1105 1106 1107 1108 1109 1110 1111 1112
	exec_keys(current);

	task_lock(current);
	unsafe = unsafe_exec(current);
	security_bprm_apply_creds(bprm, unsafe);
	task_unlock(current);
	security_bprm_post_apply_creds(bprm);
}
EXPORT_SYMBOL(compute_creds);

N
Nick Piggin 已提交
1113 1114 1115 1116 1117
/*
 * Arguments are '\0' separated strings found at the location bprm->p
 * points to; chop off the first by relocating brpm->p to right after
 * the first '\0' encountered.
 */
1118
int remove_arg_zero(struct linux_binprm *bprm)
L
Linus Torvalds 已提交
1119
{
1120 1121 1122 1123
	int ret = 0;
	unsigned long offset;
	char *kaddr;
	struct page *page;
N
Nick Piggin 已提交
1124

1125 1126
	if (!bprm->argc)
		return 0;
L
Linus Torvalds 已提交
1127

1128 1129 1130 1131 1132 1133 1134 1135
	do {
		offset = bprm->p & ~PAGE_MASK;
		page = get_arg_page(bprm, bprm->p, 0);
		if (!page) {
			ret = -EFAULT;
			goto out;
		}
		kaddr = kmap_atomic(page, KM_USER0);
N
Nick Piggin 已提交
1136

1137 1138 1139
		for (; offset < PAGE_SIZE && kaddr[offset];
				offset++, bprm->p++)
			;
N
Nick Piggin 已提交
1140

1141 1142
		kunmap_atomic(kaddr, KM_USER0);
		put_arg_page(page);
N
Nick Piggin 已提交
1143

1144 1145 1146
		if (offset == PAGE_SIZE)
			free_arg_page(bprm, (bprm->p >> PAGE_SHIFT) - 1);
	} while (offset == PAGE_SIZE);
N
Nick Piggin 已提交
1147

1148 1149 1150
	bprm->p++;
	bprm->argc--;
	ret = 0;
N
Nick Piggin 已提交
1151

1152 1153
out:
	return ret;
L
Linus Torvalds 已提交
1154 1155 1156 1157 1158 1159 1160 1161 1162 1163
}
EXPORT_SYMBOL(remove_arg_zero);

/*
 * cycle the list of binary formats handler, until one recognizes the image
 */
int search_binary_handler(struct linux_binprm *bprm,struct pt_regs *regs)
{
	int try,retval;
	struct linux_binfmt *fmt;
1164
#ifdef __alpha__
L
Linus Torvalds 已提交
1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178
	/* handle /sbin/loader.. */
	{
	    struct exec * eh = (struct exec *) bprm->buf;

	    if (!bprm->loader && eh->fh.f_magic == 0x183 &&
		(eh->fh.f_flags & 0x3000) == 0x3000)
	    {
		struct file * file;
		unsigned long loader;

		allow_write_access(bprm->file);
		fput(bprm->file);
		bprm->file = NULL;

1179
		loader = bprm->vma->vm_end - sizeof(void *);
L
Linus Torvalds 已提交
1180 1181 1182 1183 1184 1185 1186

		file = open_exec("/sbin/loader");
		retval = PTR_ERR(file);
		if (IS_ERR(file))
			return retval;

		/* Remember if the application is TASO.  */
1187
		bprm->taso = eh->ah.entry < 0x100000000UL;
L
Linus Torvalds 已提交
1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205

		bprm->file = file;
		bprm->loader = loader;
		retval = prepare_binprm(bprm);
		if (retval<0)
			return retval;
		/* should call search_binary_handler recursively here,
		   but it does not matter */
	    }
	}
#endif
	retval = security_bprm_check(bprm);
	if (retval)
		return retval;

	/* kernel module loader fixup */
	/* so we don't try to load run modprobe in kernel space. */
	set_fs(USER_DS);
A
Al Viro 已提交
1206 1207 1208 1209 1210

	retval = audit_bprm(bprm);
	if (retval)
		return retval;

L
Linus Torvalds 已提交
1211 1212 1213
	retval = -ENOENT;
	for (try=0; try<2; try++) {
		read_lock(&binfmt_lock);
A
Alexey Dobriyan 已提交
1214
		list_for_each_entry(fmt, &formats, lh) {
L
Linus Torvalds 已提交
1215 1216 1217 1218 1219 1220 1221 1222
			int (*fn)(struct linux_binprm *, struct pt_regs *) = fmt->load_binary;
			if (!fn)
				continue;
			if (!try_module_get(fmt->module))
				continue;
			read_unlock(&binfmt_lock);
			retval = fn(bprm, regs);
			if (retval >= 0) {
R
Roland McGrath 已提交
1223
				tracehook_report_exec(fmt, bprm, regs);
L
Linus Torvalds 已提交
1224 1225 1226 1227 1228 1229
				put_binfmt(fmt);
				allow_write_access(bprm->file);
				if (bprm->file)
					fput(bprm->file);
				bprm->file = NULL;
				current->did_exec = 1;
M
Matt Helsley 已提交
1230
				proc_exec_connector(current);
L
Linus Torvalds 已提交
1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244
				return retval;
			}
			read_lock(&binfmt_lock);
			put_binfmt(fmt);
			if (retval != -ENOEXEC || bprm->mm == NULL)
				break;
			if (!bprm->file) {
				read_unlock(&binfmt_lock);
				return retval;
			}
		}
		read_unlock(&binfmt_lock);
		if (retval != -ENOEXEC || bprm->mm == NULL) {
			break;
J
Johannes Berg 已提交
1245 1246
#ifdef CONFIG_MODULES
		} else {
L
Linus Torvalds 已提交
1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261
#define printable(c) (((c)=='\t') || ((c)=='\n') || (0x20<=(c) && (c)<=0x7e))
			if (printable(bprm->buf[0]) &&
			    printable(bprm->buf[1]) &&
			    printable(bprm->buf[2]) &&
			    printable(bprm->buf[3]))
				break; /* -ENOEXEC */
			request_module("binfmt-%04x", *(unsigned short *)(&bprm->buf[2]));
#endif
		}
	}
	return retval;
}

EXPORT_SYMBOL(search_binary_handler);

1262 1263 1264 1265 1266 1267
void free_bprm(struct linux_binprm *bprm)
{
	free_arg_pages(bprm);
	kfree(bprm);
}

L
Linus Torvalds 已提交
1268 1269 1270 1271 1272 1273 1274 1275 1276 1277
/*
 * sys_execve() executes a new program.
 */
int do_execve(char * filename,
	char __user *__user *argv,
	char __user *__user *envp,
	struct pt_regs * regs)
{
	struct linux_binprm *bprm;
	struct file *file;
1278
	struct files_struct *displaced;
L
Linus Torvalds 已提交
1279 1280
	int retval;

1281
	retval = unshare_files(&displaced);
1282 1283 1284
	if (retval)
		goto out_ret;

L
Linus Torvalds 已提交
1285
	retval = -ENOMEM;
1286
	bprm = kzalloc(sizeof(*bprm), GFP_KERNEL);
L
Linus Torvalds 已提交
1287
	if (!bprm)
1288
		goto out_files;
L
Linus Torvalds 已提交
1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300

	file = open_exec(filename);
	retval = PTR_ERR(file);
	if (IS_ERR(file))
		goto out_kfree;

	sched_exec();

	bprm->file = file;
	bprm->filename = filename;
	bprm->interp = filename;

1301 1302 1303
	retval = bprm_mm_init(bprm);
	if (retval)
		goto out_file;
L
Linus Torvalds 已提交
1304

1305
	bprm->argc = count(argv, MAX_ARG_STRINGS);
L
Linus Torvalds 已提交
1306 1307 1308
	if ((retval = bprm->argc) < 0)
		goto out_mm;

1309
	bprm->envc = count(envp, MAX_ARG_STRINGS);
L
Linus Torvalds 已提交
1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333
	if ((retval = bprm->envc) < 0)
		goto out_mm;

	retval = security_bprm_alloc(bprm);
	if (retval)
		goto out;

	retval = prepare_binprm(bprm);
	if (retval < 0)
		goto out;

	retval = copy_strings_kernel(1, &bprm->filename, bprm);
	if (retval < 0)
		goto out;

	bprm->exec = bprm->p;
	retval = copy_strings(bprm->envc, envp, bprm);
	if (retval < 0)
		goto out;

	retval = copy_strings(bprm->argc, argv, bprm);
	if (retval < 0)
		goto out;

O
Oleg Nesterov 已提交
1334
	current->flags &= ~PF_KTHREAD;
L
Linus Torvalds 已提交
1335 1336 1337 1338 1339
	retval = search_binary_handler(bprm,regs);
	if (retval >= 0) {
		/* execve success */
		security_bprm_free(bprm);
		acct_update_integrals(current);
1340
		free_bprm(bprm);
1341 1342
		if (displaced)
			put_files_struct(displaced);
L
Linus Torvalds 已提交
1343 1344 1345 1346 1347 1348 1349 1350 1351
		return retval;
	}

out:
	if (bprm->security)
		security_bprm_free(bprm);

out_mm:
	if (bprm->mm)
1352
		mmput (bprm->mm);
L
Linus Torvalds 已提交
1353 1354 1355 1356 1357 1358 1359

out_file:
	if (bprm->file) {
		allow_write_access(bprm->file);
		fput(bprm->file);
	}
out_kfree:
1360
	free_bprm(bprm);
L
Linus Torvalds 已提交
1361

1362
out_files:
1363 1364
	if (displaced)
		reset_files_struct(displaced);
L
Linus Torvalds 已提交
1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388
out_ret:
	return retval;
}

int set_binfmt(struct linux_binfmt *new)
{
	struct linux_binfmt *old = current->binfmt;

	if (new) {
		if (!try_module_get(new->module))
			return -1;
	}
	current->binfmt = new;
	if (old)
		module_put(old->module);
	return 0;
}

EXPORT_SYMBOL(set_binfmt);

/* 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.
 */
1389
static int format_corename(char *corename, long signr)
L
Linus Torvalds 已提交
1390
{
1391 1392
	const char *pat_ptr = core_pattern;
	int ispipe = (*pat_ptr == '|');
L
Linus Torvalds 已提交
1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418
	char *out_ptr = corename;
	char *const out_end = corename + CORENAME_MAX_SIZE;
	int rc;
	int pid_in_pattern = 0;

	/* Repeat as long as we have more pattern to process and more output
	   space */
	while (*pat_ptr) {
		if (*pat_ptr != '%') {
			if (out_ptr == out_end)
				goto out;
			*out_ptr++ = *pat_ptr++;
		} else {
			switch (*++pat_ptr) {
			case 0:
				goto out;
			/* Double percent, output one percent */
			case '%':
				if (out_ptr == out_end)
					goto out;
				*out_ptr++ = '%';
				break;
			/* pid */
			case 'p':
				pid_in_pattern = 1;
				rc = snprintf(out_ptr, out_end - out_ptr,
1419
					      "%d", task_tgid_vnr(current));
L
Linus Torvalds 已提交
1420 1421 1422 1423 1424 1425 1426
				if (rc > out_end - out_ptr)
					goto out;
				out_ptr += rc;
				break;
			/* uid */
			case 'u':
				rc = snprintf(out_ptr, out_end - out_ptr,
1427
					      "%d", current_uid());
L
Linus Torvalds 已提交
1428 1429 1430 1431 1432 1433 1434
				if (rc > out_end - out_ptr)
					goto out;
				out_ptr += rc;
				break;
			/* gid */
			case 'g':
				rc = snprintf(out_ptr, out_end - out_ptr,
1435
					      "%d", current_gid());
L
Linus Torvalds 已提交
1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462
				if (rc > out_end - out_ptr)
					goto out;
				out_ptr += rc;
				break;
			/* signal that caused the coredump */
			case 's':
				rc = snprintf(out_ptr, out_end - out_ptr,
					      "%ld", signr);
				if (rc > out_end - out_ptr)
					goto out;
				out_ptr += rc;
				break;
			/* UNIX time of coredump */
			case 't': {
				struct timeval tv;
				do_gettimeofday(&tv);
				rc = snprintf(out_ptr, out_end - out_ptr,
					      "%lu", tv.tv_sec);
				if (rc > out_end - out_ptr)
					goto out;
				out_ptr += rc;
				break;
			}
			/* hostname */
			case 'h':
				down_read(&uts_sem);
				rc = snprintf(out_ptr, out_end - out_ptr,
1463
					      "%s", utsname()->nodename);
L
Linus Torvalds 已提交
1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476
				up_read(&uts_sem);
				if (rc > out_end - out_ptr)
					goto out;
				out_ptr += rc;
				break;
			/* executable */
			case 'e':
				rc = snprintf(out_ptr, out_end - out_ptr,
					      "%s", current->comm);
				if (rc > out_end - out_ptr)
					goto out;
				out_ptr += rc;
				break;
1477 1478 1479 1480 1481 1482 1483 1484
			/* core limit size */
			case 'c':
				rc = snprintf(out_ptr, out_end - out_ptr,
					      "%lu", current->signal->rlim[RLIMIT_CORE].rlim_cur);
				if (rc > out_end - out_ptr)
					goto out;
				out_ptr += rc;
				break;
L
Linus Torvalds 已提交
1485 1486 1487 1488 1489 1490 1491 1492 1493 1494
			default:
				break;
			}
			++pat_ptr;
		}
	}
	/* 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
1495
	 * the filename. Do not do this for piped commands. */
1496
	if (!ispipe && !pid_in_pattern && core_uses_pid) {
L
Linus Torvalds 已提交
1497
		rc = snprintf(out_ptr, out_end - out_ptr,
1498
			      ".%d", task_tgid_vnr(current));
L
Linus Torvalds 已提交
1499 1500 1501 1502
		if (rc > out_end - out_ptr)
			goto out;
		out_ptr += rc;
	}
1503
out:
L
Linus Torvalds 已提交
1504
	*out_ptr = 0;
1505
	return ispipe;
L
Linus Torvalds 已提交
1506 1507
}

1508
static int zap_process(struct task_struct *start)
1509 1510
{
	struct task_struct *t;
1511
	int nr = 0;
1512

1513 1514
	start->signal->flags = SIGNAL_GROUP_EXIT;
	start->signal->group_stop_count = 0;
1515 1516 1517 1518

	t = start;
	do {
		if (t != current && t->mm) {
1519 1520
			sigaddset(&t->pending.signal, SIGKILL);
			signal_wake_up(t, 1);
1521
			nr++;
1522
		}
1523
	} while_each_thread(start, t);
1524 1525

	return nr;
1526 1527
}

1528
static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm,
1529
				struct core_state *core_state, int exit_code)
L
Linus Torvalds 已提交
1530 1531
{
	struct task_struct *g, *p;
1532
	unsigned long flags;
1533
	int nr = -EAGAIN;
1534 1535

	spin_lock_irq(&tsk->sighand->siglock);
1536
	if (!signal_group_exit(tsk->signal)) {
1537
		mm->core_state = core_state;
1538
		tsk->signal->group_exit_code = exit_code;
1539
		nr = zap_process(tsk);
L
Linus Torvalds 已提交
1540
	}
1541
	spin_unlock_irq(&tsk->sighand->siglock);
1542 1543
	if (unlikely(nr < 0))
		return nr;
L
Linus Torvalds 已提交
1544

1545
	if (atomic_read(&mm->mm_users) == nr + 1)
1546
		goto done;
1547 1548
	/*
	 * We should find and kill all tasks which use this mm, and we should
1549
	 * count them correctly into ->nr_threads. We don't take tasklist
1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576
	 * 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().
	 */
1577
	rcu_read_lock();
1578
	for_each_process(g) {
1579 1580
		if (g == tsk->group_leader)
			continue;
1581 1582
		if (g->flags & PF_KTHREAD)
			continue;
1583 1584 1585
		p = g;
		do {
			if (p->mm) {
1586
				if (unlikely(p->mm == mm)) {
1587
					lock_task_sighand(p, &flags);
1588
					nr += zap_process(p);
1589 1590
					unlock_task_sighand(p, &flags);
				}
1591 1592
				break;
			}
1593
		} while_each_thread(g, p);
1594
	}
1595
	rcu_read_unlock();
1596
done:
1597
	atomic_set(&core_state->nr_threads, nr);
1598
	return nr;
L
Linus Torvalds 已提交
1599 1600
}

1601
static int coredump_wait(int exit_code, struct core_state *core_state)
L
Linus Torvalds 已提交
1602
{
1603 1604 1605
	struct task_struct *tsk = current;
	struct mm_struct *mm = tsk->mm;
	struct completion *vfork_done;
O
Oleg Nesterov 已提交
1606
	int core_waiters;
L
Linus Torvalds 已提交
1607

1608
	init_completion(&core_state->startup);
1609 1610
	core_state->dumper.task = tsk;
	core_state->dumper.next = NULL;
1611
	core_waiters = zap_threads(tsk, mm, core_state, exit_code);
O
Oleg Nesterov 已提交
1612 1613
	up_write(&mm->mmap_sem);

1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626
	if (unlikely(core_waiters < 0))
		goto fail;

	/*
	 * Make sure nobody is waiting for us to release the VM,
	 * otherwise we can deadlock when we wait on each other
	 */
	vfork_done = tsk->vfork_done;
	if (vfork_done) {
		tsk->vfork_done = NULL;
		complete(vfork_done);
	}

O
Oleg Nesterov 已提交
1627
	if (core_waiters)
1628
		wait_for_completion(&core_state->startup);
1629 1630
fail:
	return core_waiters;
L
Linus Torvalds 已提交
1631 1632
}

O
Oleg Nesterov 已提交
1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653
static void coredump_finish(struct mm_struct *mm)
{
	struct core_thread *curr, *next;
	struct task_struct *task;

	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;
}

1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702
/*
 * set_dumpable converts traditional three-value dumpable to two flags and
 * stores them into mm->flags.  It modifies lower two bits of mm->flags, but
 * these bits are not changed atomically.  So get_dumpable can observe the
 * intermediate state.  To avoid doing unexpected behavior, get get_dumpable
 * return either old dumpable or new one by paying attention to the order of
 * modifying the bits.
 *
 * dumpable |   mm->flags (binary)
 * old  new | initial interim  final
 * ---------+-----------------------
 *  0    1  |   00      01      01
 *  0    2  |   00      10(*)   11
 *  1    0  |   01      00      00
 *  1    2  |   01      11      11
 *  2    0  |   11      10(*)   00
 *  2    1  |   11      11      01
 *
 * (*) get_dumpable regards interim value of 10 as 11.
 */
void set_dumpable(struct mm_struct *mm, int value)
{
	switch (value) {
	case 0:
		clear_bit(MMF_DUMPABLE, &mm->flags);
		smp_wmb();
		clear_bit(MMF_DUMP_SECURELY, &mm->flags);
		break;
	case 1:
		set_bit(MMF_DUMPABLE, &mm->flags);
		smp_wmb();
		clear_bit(MMF_DUMP_SECURELY, &mm->flags);
		break;
	case 2:
		set_bit(MMF_DUMP_SECURELY, &mm->flags);
		smp_wmb();
		set_bit(MMF_DUMPABLE, &mm->flags);
		break;
	}
}

int get_dumpable(struct mm_struct *mm)
{
	int ret;

	ret = mm->flags & 0x3;
	return (ret >= 2) ? 2 : ret;
}

L
Linus Torvalds 已提交
1703 1704
int do_coredump(long signr, int exit_code, struct pt_regs * regs)
{
1705
	struct core_state core_state;
L
Linus Torvalds 已提交
1706 1707 1708 1709 1710 1711
	char corename[CORENAME_MAX_SIZE + 1];
	struct mm_struct *mm = current->mm;
	struct linux_binfmt * binfmt;
	struct inode * inode;
	struct file * file;
	int retval = 0;
1712
	int fsuid = current_fsuid();
A
Alan Cox 已提交
1713
	int flag = 0;
1714
	int ispipe = 0;
1715
	unsigned long core_limit = current->signal->rlim[RLIMIT_CORE].rlim_cur;
1716 1717 1718
	char **helper_argv = NULL;
	int helper_argc = 0;
	char *delimit;
L
Linus Torvalds 已提交
1719

S
Steve Grubb 已提交
1720 1721
	audit_core_dumps(signr);

L
Linus Torvalds 已提交
1722 1723 1724 1725
	binfmt = current->binfmt;
	if (!binfmt || !binfmt->core_dump)
		goto fail;
	down_write(&mm->mmap_sem);
R
Roland McGrath 已提交
1726 1727 1728
	/*
	 * If another thread got here first, or we are not dumpable, bail out.
	 */
1729
	if (mm->core_state || !get_dumpable(mm)) {
L
Linus Torvalds 已提交
1730 1731 1732
		up_write(&mm->mmap_sem);
		goto fail;
	}
A
Alan Cox 已提交
1733 1734 1735 1736 1737 1738

	/*
	 *	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.
	 */
1739
	if (get_dumpable(mm) == 2) {	/* Setuid core dump mode */
A
Alan Cox 已提交
1740
		flag = O_EXCL;		/* Stop rewrite attacks */
1741
		current->cred->fsuid = 0;	/* Dump root private */
A
Alan Cox 已提交
1742
	}
1743

1744
	retval = coredump_wait(exit_code, &core_state);
1745
	if (retval < 0)
1746
		goto fail;
L
Linus Torvalds 已提交
1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758

	/*
	 * Clear any false indication of pending signals that might
	 * be seen by the filesystem code called to write the core file.
	 */
	clear_thread_flag(TIF_SIGPENDING);

	/*
	 * lock_kernel() because format_corename() is controlled by sysctl, which
	 * uses lock_kernel()
	 */
 	lock_kernel();
1759
	ispipe = format_corename(corename, signr);
L
Linus Torvalds 已提交
1760
	unlock_kernel();
1761 1762 1763 1764 1765 1766 1767 1768
	/*
	 * Don't bother to check the RLIMIT_CORE value if core_pattern points
	 * to a pipe.  Since we're not writing directly to the filesystem
	 * RLIMIT_CORE doesn't really apply, as no actual core file will be
	 * created unless the pipe reader choses to write out the core file
	 * at which point file size limits and permissions will be imposed
	 * as it does with any other process
	 */
1769
	if ((!ispipe) && (core_limit < binfmt->min_coredump))
1770 1771
		goto fail_unlock;

1772
 	if (ispipe) {
1773 1774 1775 1776 1777
		helper_argv = argv_split(GFP_KERNEL, corename+1, &helper_argc);
		/* Terminate the string before the first option */
		delimit = strchr(corename, ' ');
		if (delimit)
			*delimit = '\0';
1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790
		delimit = strrchr(helper_argv[0], '/');
		if (delimit)
			delimit++;
		else
			delimit = helper_argv[0];
		if (!strcmp(delimit, current->comm)) {
			printk(KERN_NOTICE "Recursive core dump detected, "
					"aborting\n");
			goto fail_unlock;
		}

		core_limit = RLIM_INFINITY;

1791
		/* SIGPIPE can happen, but it's just never processed */
1792 1793
 		if (call_usermodehelper_pipe(corename+1, helper_argv, NULL,
				&file)) {
1794 1795 1796 1797 1798 1799
 			printk(KERN_INFO "Core dump to %s pipe failed\n",
			       corename);
 			goto fail_unlock;
 		}
 	} else
 		file = filp_open(corename,
1800 1801
				 O_CREAT | 2 | O_NOFOLLOW | O_LARGEFILE | flag,
				 0600);
L
Linus Torvalds 已提交
1802 1803
	if (IS_ERR(file))
		goto fail_unlock;
1804
	inode = file->f_path.dentry->d_inode;
L
Linus Torvalds 已提交
1805 1806
	if (inode->i_nlink > 1)
		goto close_fail;	/* multiple links - don't dump */
1807
	if (!ispipe && d_unhashed(file->f_path.dentry))
L
Linus Torvalds 已提交
1808 1809
		goto close_fail;

1810 1811 1812
	/* AK: actually i see no reason to not allow this for named pipes etc.,
	   but keep the previous behaviour for now. */
	if (!ispipe && !S_ISREG(inode->i_mode))
L
Linus Torvalds 已提交
1813
		goto close_fail;
I
Ingo Molnar 已提交
1814 1815 1816 1817
	/*
	 * Dont allow local users get cute and trick others to coredump
	 * into their pre-created files:
	 */
1818
	if (inode->i_uid != current_fsuid())
I
Ingo Molnar 已提交
1819
		goto close_fail;
L
Linus Torvalds 已提交
1820 1821 1822 1823
	if (!file->f_op)
		goto close_fail;
	if (!file->f_op->write)
		goto close_fail;
1824
	if (!ispipe && do_truncate(file->f_path.dentry, 0, 0, file) != 0)
L
Linus Torvalds 已提交
1825 1826
		goto close_fail;

1827
	retval = binfmt->core_dump(signr, regs, file, core_limit);
L
Linus Torvalds 已提交
1828 1829 1830 1831 1832 1833

	if (retval)
		current->signal->group_exit_code |= 0x80;
close_fail:
	filp_close(file, NULL);
fail_unlock:
1834 1835 1836
	if (helper_argv)
		argv_free(helper_argv);

1837
	current->cred->fsuid = fsuid;
O
Oleg Nesterov 已提交
1838
	coredump_finish(mm);
L
Linus Torvalds 已提交
1839 1840 1841
fail:
	return retval;
}