Blame view
fs/exec.c
46.6 KB
1da177e4c
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 |
/* * 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. */ |
1da177e4c
|
24 25 |
#include <linux/slab.h> #include <linux/file.h> |
9f3acc314
|
26 |
#include <linux/fdtable.h> |
ba92a43db
|
27 |
#include <linux/mm.h> |
1da177e4c
|
28 29 |
#include <linux/stat.h> #include <linux/fcntl.h> |
ba92a43db
|
30 |
#include <linux/swap.h> |
74aadce98
|
31 |
#include <linux/string.h> |
1da177e4c
|
32 |
#include <linux/init.h> |
ca5b172bd
|
33 |
#include <linux/pagemap.h> |
cdd6c482c
|
34 |
#include <linux/perf_event.h> |
1da177e4c
|
35 36 37 38 39 |
#include <linux/highmem.h> #include <linux/spinlock.h> #include <linux/key.h> #include <linux/personality.h> #include <linux/binfmts.h> |
1da177e4c
|
40 |
#include <linux/utsname.h> |
84d737866
|
41 |
#include <linux/pid_namespace.h> |
1da177e4c
|
42 43 44 |
#include <linux/module.h> #include <linux/namei.h> #include <linux/proc_fs.h> |
1da177e4c
|
45 46 47 |
#include <linux/mount.h> #include <linux/security.h> #include <linux/syscalls.h> |
8f0ab5147
|
48 |
#include <linux/tsacct_kern.h> |
9f46080c4
|
49 |
#include <linux/cn_proc.h> |
473ae30bc
|
50 |
#include <linux/audit.h> |
6341c393f
|
51 |
#include <linux/tracehook.h> |
5f4123be3
|
52 |
#include <linux/kmod.h> |
6110e3abb
|
53 |
#include <linux/fsnotify.h> |
5ad4e53bd
|
54 |
#include <linux/fs_struct.h> |
61be228a0
|
55 |
#include <linux/pipe_fs_i.h> |
1da177e4c
|
56 57 58 |
#include <asm/uaccess.h> #include <asm/mmu_context.h> |
b6a2fea39
|
59 |
#include <asm/tlb.h> |
a6f76f23d
|
60 |
#include "internal.h" |
1da177e4c
|
61 |
|
1da177e4c
|
62 |
int core_uses_pid; |
71ce92f3f
|
63 |
char core_pattern[CORENAME_MAX_SIZE] = "core"; |
a293980c2
|
64 |
unsigned int core_pipe_limit; |
d6e711448
|
65 |
int suid_dumpable = 0; |
1da177e4c
|
66 |
/* The maximal length of core_pattern is also specified in sysctl.c */ |
e4dc1b14d
|
67 |
static LIST_HEAD(formats); |
1da177e4c
|
68 |
static DEFINE_RWLOCK(binfmt_lock); |
74641f584
|
69 |
int __register_binfmt(struct linux_binfmt * fmt, int insert) |
1da177e4c
|
70 |
{ |
1da177e4c
|
71 72 |
if (!fmt) return -EINVAL; |
1da177e4c
|
73 |
write_lock(&binfmt_lock); |
74641f584
|
74 75 |
insert ? list_add(&fmt->lh, &formats) : list_add_tail(&fmt->lh, &formats); |
1da177e4c
|
76 77 78 |
write_unlock(&binfmt_lock); return 0; } |
74641f584
|
79 |
EXPORT_SYMBOL(__register_binfmt); |
1da177e4c
|
80 |
|
f6b450d48
|
81 |
void unregister_binfmt(struct linux_binfmt * fmt) |
1da177e4c
|
82 |
{ |
1da177e4c
|
83 |
write_lock(&binfmt_lock); |
e4dc1b14d
|
84 |
list_del(&fmt->lh); |
1da177e4c
|
85 |
write_unlock(&binfmt_lock); |
1da177e4c
|
86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 |
} 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. */ |
1e7bfb213
|
101 |
SYSCALL_DEFINE1(uselib, const char __user *, library) |
1da177e4c
|
102 |
{ |
964bd1836
|
103 |
struct file *file; |
964bd1836
|
104 105 |
char *tmp = getname(library); int error = PTR_ERR(tmp); |
6e8341a11
|
106 107 108 109 110 111 112 113 114 |
if (IS_ERR(tmp)) goto out; file = do_filp_open(AT_FDCWD, tmp, O_LARGEFILE | O_RDONLY | FMODE_EXEC, 0, MAY_READ | MAY_EXEC | MAY_OPEN); putname(tmp); error = PTR_ERR(file); if (IS_ERR(file)) |
1da177e4c
|
115 116 117 |
goto out; error = -EINVAL; |
6e8341a11
|
118 |
if (!S_ISREG(file->f_path.dentry->d_inode->i_mode)) |
1da177e4c
|
119 |
goto exit; |
30524472c
|
120 |
error = -EACCES; |
6e8341a11
|
121 |
if (file->f_path.mnt->mnt_flags & MNT_NOEXEC) |
6146f0d5e
|
122 |
goto exit; |
1da177e4c
|
123 |
|
2a12a9d78
|
124 |
fsnotify_open(file); |
6110e3abb
|
125 |
|
1da177e4c
|
126 127 128 129 130 |
error = -ENOEXEC; if(file->f_op) { struct linux_binfmt * fmt; read_lock(&binfmt_lock); |
e4dc1b14d
|
131 |
list_for_each_entry(fmt, &formats, lh) { |
1da177e4c
|
132 133 134 135 136 137 138 139 140 141 142 143 144 |
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); } |
6e8341a11
|
145 |
exit: |
1da177e4c
|
146 147 148 |
fput(file); out: return error; |
1da177e4c
|
149 |
} |
b6a2fea39
|
150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 |
#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) { |
b6a2fea39
|
171 |
unsigned long size = bprm->vma->vm_end - bprm->vma->vm_start; |
a64e715fc
|
172 173 174 175 176 177 178 179 |
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; |
b6a2fea39
|
180 181 182 183 184 185 186 187 |
/* * 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. */ |
a64e715fc
|
188 |
rlim = current->signal->rlim; |
d554ed895
|
189 |
if (size > ACCESS_ONCE(rlim[RLIMIT_STACK].rlim_cur) / 4) { |
b6a2fea39
|
190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 |
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) { |
eaccbfa56
|
219 |
int err; |
b6a2fea39
|
220 221 222 223 224 |
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) |
eaccbfa56
|
225 |
return -ENOMEM; |
b6a2fea39
|
226 227 228 229 230 231 232 233 234 235 |
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. */ |
a8bef8ff6
|
236 |
BUG_ON(VM_STACK_FLAGS & VM_STACK_INCOMPLETE_SETUP); |
b6a2fea39
|
237 238 |
vma->vm_end = STACK_TOP_MAX; vma->vm_start = vma->vm_end - PAGE_SIZE; |
a8bef8ff6
|
239 |
vma->vm_flags = VM_STACK_FLAGS | VM_STACK_INCOMPLETE_SETUP; |
3ed75eb8f
|
240 |
vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); |
5beb49305
|
241 |
INIT_LIST_HEAD(&vma->anon_vma_chain); |
b6a2fea39
|
242 |
err = insert_vm_struct(mm, vma); |
eaccbfa56
|
243 |
if (err) |
b6a2fea39
|
244 |
goto err; |
b6a2fea39
|
245 246 247 |
mm->stack_vm = mm->total_vm = 1; up_write(&mm->mmap_sem); |
b6a2fea39
|
248 |
bprm->p = vma->vm_end - sizeof(void *); |
b6a2fea39
|
249 |
return 0; |
b6a2fea39
|
250 |
err: |
eaccbfa56
|
251 252 253 |
up_write(&mm->mmap_sem); bprm->vma = NULL; kmem_cache_free(vm_area_cachep, vma); |
b6a2fea39
|
254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 |
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; } |
1da177e4c
|
352 353 354 |
/* * count() counts the number of strings in array ARGV. */ |
d7627467b
|
355 |
static int count(const char __user * const __user * argv, int max) |
1da177e4c
|
356 357 358 359 360 |
{ int i = 0; if (argv != NULL) { for (;;) { |
d7627467b
|
361 |
const char __user * p; |
1da177e4c
|
362 363 364 365 366 367 |
if (get_user(p, argv)) return -EFAULT; if (!p) break; argv++; |
362e6663e
|
368 |
if (i++ >= max) |
1da177e4c
|
369 |
return -E2BIG; |
9aea5a65a
|
370 371 372 |
if (fatal_signal_pending(current)) return -ERESTARTNOHAND; |
1da177e4c
|
373 374 375 376 377 378 379 |
cond_resched(); } } return i; } /* |
b6a2fea39
|
380 381 382 |
* '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. |
1da177e4c
|
383 |
*/ |
d7627467b
|
384 |
static int copy_strings(int argc, const char __user *const __user *argv, |
75c96f858
|
385 |
struct linux_binprm *bprm) |
1da177e4c
|
386 387 388 |
{ struct page *kmapped_page = NULL; char *kaddr = NULL; |
b6a2fea39
|
389 |
unsigned long kpos = 0; |
1da177e4c
|
390 391 392 |
int ret; while (argc-- > 0) { |
d7627467b
|
393 |
const char __user *str; |
1da177e4c
|
394 395 396 397 |
int len; unsigned long pos; if (get_user(str, argv+argc) || |
b6a2fea39
|
398 |
!(len = strnlen_user(str, MAX_ARG_STRLEN))) { |
1da177e4c
|
399 400 401 |
ret = -EFAULT; goto out; } |
b6a2fea39
|
402 |
if (!valid_arg_len(bprm, len)) { |
1da177e4c
|
403 404 405 |
ret = -E2BIG; goto out; } |
b6a2fea39
|
406 |
/* We're going to work our way backwords. */ |
1da177e4c
|
407 |
pos = bprm->p; |
b6a2fea39
|
408 409 |
str += len; bprm->p -= len; |
1da177e4c
|
410 411 |
while (len > 0) { |
1da177e4c
|
412 |
int offset, bytes_to_copy; |
1da177e4c
|
413 |
|
9aea5a65a
|
414 415 416 417 |
if (fatal_signal_pending(current)) { ret = -ERESTARTNOHAND; goto out; } |
7993bc1f4
|
418 |
cond_resched(); |
1da177e4c
|
419 |
offset = pos % PAGE_SIZE; |
b6a2fea39
|
420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 |
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); |
1da177e4c
|
436 |
if (!page) { |
b6a2fea39
|
437 |
ret = -E2BIG; |
1da177e4c
|
438 439 |
goto out; } |
1da177e4c
|
440 |
|
b6a2fea39
|
441 442 |
if (kmapped_page) { flush_kernel_dcache_page(kmapped_page); |
1da177e4c
|
443 |
kunmap(kmapped_page); |
b6a2fea39
|
444 445 |
put_arg_page(kmapped_page); } |
1da177e4c
|
446 447 |
kmapped_page = page; kaddr = kmap(kmapped_page); |
b6a2fea39
|
448 449 |
kpos = pos & PAGE_MASK; flush_arg_page(bprm, kpos, kmapped_page); |
1da177e4c
|
450 |
} |
b6a2fea39
|
451 |
if (copy_from_user(kaddr+offset, str, bytes_to_copy)) { |
1da177e4c
|
452 453 454 |
ret = -EFAULT; goto out; } |
1da177e4c
|
455 456 457 458 |
} } ret = 0; out: |
b6a2fea39
|
459 460 |
if (kmapped_page) { flush_kernel_dcache_page(kmapped_page); |
1da177e4c
|
461 |
kunmap(kmapped_page); |
b6a2fea39
|
462 463 |
put_arg_page(kmapped_page); } |
1da177e4c
|
464 465 466 467 468 469 |
return ret; } /* * Like copy_strings, but get argv and its values from kernel memory. */ |
d7627467b
|
470 471 |
int copy_strings_kernel(int argc, const char *const *argv, struct linux_binprm *bprm) |
1da177e4c
|
472 473 474 475 |
{ int r; mm_segment_t oldfs = get_fs(); set_fs(KERNEL_DS); |
d7627467b
|
476 |
r = copy_strings(argc, (const char __user *const __user *)argv, bprm); |
1da177e4c
|
477 478 479 |
set_fs(oldfs); return r; } |
1da177e4c
|
480 481 482 |
EXPORT_SYMBOL(copy_strings_kernel); #ifdef CONFIG_MMU |
b6a2fea39
|
483 |
|
1da177e4c
|
484 |
/* |
b6a2fea39
|
485 486 487 |
* 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: |
1da177e4c
|
488 |
* |
b6a2fea39
|
489 490 491 492 493 494 |
* 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. |
1da177e4c
|
495 |
*/ |
b6a2fea39
|
496 |
static int shift_arg_pages(struct vm_area_struct *vma, unsigned long shift) |
1da177e4c
|
497 498 |
{ struct mm_struct *mm = vma->vm_mm; |
b6a2fea39
|
499 500 501 502 503 504 |
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; |
1da177e4c
|
505 |
|
b6a2fea39
|
506 |
BUG_ON(new_start > new_end); |
1da177e4c
|
507 |
|
b6a2fea39
|
508 509 510 511 512 513 514 515 516 517 |
/* * 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) */ |
5beb49305
|
518 519 |
if (vma_adjust(vma, new_start, old_end, vma->vm_pgoff, NULL)) return -ENOMEM; |
b6a2fea39
|
520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 |
/* * 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. */ |
42b777281
|
535 |
free_pgd_range(tlb, new_end, old_end, new_end, |
b6a2fea39
|
536 537 538 539 540 541 542 543 |
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. */ |
42b777281
|
544 |
free_pgd_range(tlb, old_start, old_end, new_end, |
b6a2fea39
|
545 |
vma->vm_next ? vma->vm_next->vm_start : 0); |
1da177e4c
|
546 |
} |
b6a2fea39
|
547 548 549 |
tlb_finish_mmu(tlb, new_end, old_end); /* |
5beb49305
|
550 |
* Shrink the vma to just the new range. Always succeeds. |
b6a2fea39
|
551 552 553 554 |
*/ vma_adjust(vma, new_start, new_end, vma->vm_pgoff, NULL); return 0; |
1da177e4c
|
555 |
} |
b6a2fea39
|
556 557 558 559 |
/* * Finalizes the stack vm_area_struct. The flags and permissions are updated, * the stack is optionally relocated, and some extra space is added. */ |
1da177e4c
|
560 561 562 563 |
int setup_arg_pages(struct linux_binprm *bprm, unsigned long stack_top, int executable_stack) { |
b6a2fea39
|
564 565 |
unsigned long ret; unsigned long stack_shift; |
1da177e4c
|
566 |
struct mm_struct *mm = current->mm; |
b6a2fea39
|
567 568 569 570 |
struct vm_area_struct *vma = bprm->vma; struct vm_area_struct *prev = NULL; unsigned long vm_flags; unsigned long stack_base; |
803bf5ec2
|
571 572 573 |
unsigned long stack_size; unsigned long stack_expand; unsigned long rlim_stack; |
1da177e4c
|
574 575 |
#ifdef CONFIG_STACK_GROWSUP |
1da177e4c
|
576 |
/* Limit stack size to 1GB */ |
d554ed895
|
577 |
stack_base = rlimit_max(RLIMIT_STACK); |
1da177e4c
|
578 579 |
if (stack_base > (1 << 30)) stack_base = 1 << 30; |
1da177e4c
|
580 |
|
b6a2fea39
|
581 582 583 |
/* Make sure we didn't let the argument array grow too large. */ if (vma->vm_end - vma->vm_start > stack_base) return -ENOMEM; |
1da177e4c
|
584 |
|
b6a2fea39
|
585 |
stack_base = PAGE_ALIGN(stack_top - stack_base); |
1da177e4c
|
586 |
|
b6a2fea39
|
587 588 589 |
stack_shift = vma->vm_start - stack_base; mm->arg_start = bprm->p - stack_shift; bprm->p = vma->vm_end - stack_shift; |
1da177e4c
|
590 |
#else |
b6a2fea39
|
591 592 |
stack_top = arch_align_stack(stack_top); stack_top = PAGE_ALIGN(stack_top); |
1b528181b
|
593 594 595 596 |
if (unlikely(stack_top < mmap_min_addr) || unlikely(vma->vm_end - vma->vm_start >= stack_top - mmap_min_addr)) return -ENOMEM; |
b6a2fea39
|
597 598 599 |
stack_shift = vma->vm_end - stack_top; bprm->p -= stack_shift; |
1da177e4c
|
600 |
mm->arg_start = bprm->p; |
1da177e4c
|
601 |
#endif |
1da177e4c
|
602 |
if (bprm->loader) |
b6a2fea39
|
603 604 |
bprm->loader -= stack_shift; bprm->exec -= stack_shift; |
1da177e4c
|
605 |
|
1da177e4c
|
606 |
down_write(&mm->mmap_sem); |
96a8e13ed
|
607 |
vm_flags = VM_STACK_FLAGS; |
b6a2fea39
|
608 609 610 611 612 613 614 615 616 617 618 |
/* * 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; |
a8bef8ff6
|
619 |
vm_flags |= VM_STACK_INCOMPLETE_SETUP; |
b6a2fea39
|
620 621 622 623 624 625 626 627 628 629 |
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); |
fc63cf237
|
630 631 |
if (ret) goto out_unlock; |
1da177e4c
|
632 |
} |
a8bef8ff6
|
633 634 |
/* mprotect_fixup is overkill to remove the temporary stack flags */ vma->vm_flags &= ~VM_STACK_INCOMPLETE_SETUP; |
5ef097dd7
|
635 |
stack_expand = 131072UL; /* randomly 32*4k (or 2*64k) pages */ |
803bf5ec2
|
636 637 638 639 640 641 |
stack_size = vma->vm_end - vma->vm_start; /* * Align this down to a page boundary as expand_stack * will align it up. */ rlim_stack = rlimit(RLIMIT_STACK) & PAGE_MASK; |
b6a2fea39
|
642 |
#ifdef CONFIG_STACK_GROWSUP |
803bf5ec2
|
643 644 645 646 |
if (stack_size + stack_expand > rlim_stack) stack_base = vma->vm_start + rlim_stack; else stack_base = vma->vm_end + stack_expand; |
b6a2fea39
|
647 |
#else |
803bf5ec2
|
648 649 650 651 |
if (stack_size + stack_expand > rlim_stack) stack_base = vma->vm_end - rlim_stack; else stack_base = vma->vm_start - stack_expand; |
b6a2fea39
|
652 |
#endif |
3af9e8592
|
653 |
current->mm->start_stack = bprm->p; |
b6a2fea39
|
654 655 656 657 658 |
ret = expand_stack(vma, stack_base); if (ret) ret = -EFAULT; out_unlock: |
1da177e4c
|
659 |
up_write(&mm->mmap_sem); |
fc63cf237
|
660 |
return ret; |
1da177e4c
|
661 |
} |
1da177e4c
|
662 |
EXPORT_SYMBOL(setup_arg_pages); |
1da177e4c
|
663 664 665 666 |
#endif /* CONFIG_MMU */ struct file *open_exec(const char *name) { |
1da177e4c
|
667 |
struct file *file; |
e56b6a5dd
|
668 |
int err; |
1da177e4c
|
669 |
|
6e8341a11
|
670 671 672 673 |
file = do_filp_open(AT_FDCWD, name, O_LARGEFILE | O_RDONLY | FMODE_EXEC, 0, MAY_EXEC | MAY_OPEN); if (IS_ERR(file)) |
e56b6a5dd
|
674 675 676 |
goto out; err = -EACCES; |
6e8341a11
|
677 678 |
if (!S_ISREG(file->f_path.dentry->d_inode->i_mode)) goto exit; |
e56b6a5dd
|
679 |
|
6e8341a11
|
680 681 |
if (file->f_path.mnt->mnt_flags & MNT_NOEXEC) goto exit; |
e56b6a5dd
|
682 |
|
2a12a9d78
|
683 |
fsnotify_open(file); |
6110e3abb
|
684 |
|
e56b6a5dd
|
685 |
err = deny_write_access(file); |
6e8341a11
|
686 687 |
if (err) goto exit; |
1da177e4c
|
688 |
|
6e8341a11
|
689 |
out: |
e56b6a5dd
|
690 |
return file; |
6e8341a11
|
691 692 |
exit: fput(file); |
e56b6a5dd
|
693 694 |
return ERR_PTR(err); } |
1da177e4c
|
695 |
EXPORT_SYMBOL(open_exec); |
6777d773a
|
696 697 |
int kernel_read(struct file *file, loff_t offset, char *addr, unsigned long count) |
1da177e4c
|
698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 |
{ 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; |
34e55232e
|
721 |
sync_mm_rss(tsk, old_mm); |
1da177e4c
|
722 723 724 725 726 727 728 |
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 |
999d9fc16
|
729 |
* checking core_state and changing tsk->mm. |
1da177e4c
|
730 731 |
*/ down_read(&old_mm->mmap_sem); |
999d9fc16
|
732 |
if (unlikely(old_mm->core_state)) { |
1da177e4c
|
733 734 735 736 737 738 739 740 741 742 743 744 745 |
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); |
7dddb12c6
|
746 |
BUG_ON(active_mm != old_mm); |
31a78f23b
|
747 |
mm_update_next_owner(old_mm); |
1da177e4c
|
748 749 750 751 752 753 754 755 756 757 758 759 760 |
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().) */ |
858119e15
|
761 |
static int de_thread(struct task_struct *tsk) |
1da177e4c
|
762 763 |
{ struct signal_struct *sig = tsk->signal; |
b2c903b87
|
764 |
struct sighand_struct *oldsighand = tsk->sighand; |
1da177e4c
|
765 |
spinlock_t *lock = &oldsighand->siglock; |
1da177e4c
|
766 |
|
aafe6c2a2
|
767 |
if (thread_group_empty(tsk)) |
1da177e4c
|
768 769 770 771 |
goto no_thread_group; /* * Kill all other threads in the thread group. |
1da177e4c
|
772 |
*/ |
1da177e4c
|
773 |
spin_lock_irq(lock); |
ed5d2cac1
|
774 |
if (signal_group_exit(sig)) { |
1da177e4c
|
775 776 777 778 779 |
/* * Another group action in progress, just * return so that the signal is processed. */ spin_unlock_irq(lock); |
1da177e4c
|
780 781 |
return -EAGAIN; } |
d344193a0
|
782 |
|
ed5d2cac1
|
783 |
sig->group_exit_task = tsk; |
d344193a0
|
784 785 786 |
sig->notify_count = zap_other_threads(tsk); if (!thread_group_leader(tsk)) sig->notify_count--; |
1da177e4c
|
787 |
|
d344193a0
|
788 |
while (sig->notify_count) { |
1da177e4c
|
789 790 791 792 793 |
__set_current_state(TASK_UNINTERRUPTIBLE); spin_unlock_irq(lock); schedule(); spin_lock_irq(lock); } |
1da177e4c
|
794 795 796 797 798 799 800 |
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: */ |
aafe6c2a2
|
801 |
if (!thread_group_leader(tsk)) { |
8187926bd
|
802 |
struct task_struct *leader = tsk->group_leader; |
6db840fa7
|
803 |
|
2800d8d19
|
804 |
sig->notify_count = -1; /* for exit_notify() */ |
6db840fa7
|
805 806 807 808 809 810 811 812 |
for (;;) { write_lock_irq(&tasklist_lock); if (likely(leader->exit_state)) break; __set_current_state(TASK_UNINTERRUPTIBLE); write_unlock_irq(&tasklist_lock); schedule(); } |
1da177e4c
|
813 |
|
f5e902817
|
814 815 816 817 818 819 820 821 822 823 |
/* * 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). */ |
aafe6c2a2
|
824 |
tsk->start_time = leader->start_time; |
f5e902817
|
825 |
|
bac0abd61
|
826 827 |
BUG_ON(!same_thread_group(leader, tsk)); BUG_ON(has_group_leader_pid(tsk)); |
1da177e4c
|
828 829 830 831 832 833 |
/* * 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: */ |
d73d65293
|
834 835 |
/* Become a process group leader with the old leader's pid. |
c18258c6f
|
836 837 |
* The old leader becomes a thread of the this thread group. * Note: The old leader also uses this pid until release_task |
d73d65293
|
838 839 |
* is called. Odd but simple and correct. */ |
aafe6c2a2
|
840 841 |
detach_pid(tsk, PIDTYPE_PID); tsk->pid = leader->pid; |
3743ca05f
|
842 |
attach_pid(tsk, PIDTYPE_PID, task_pid(leader)); |
aafe6c2a2
|
843 844 |
transfer_pid(leader, tsk, PIDTYPE_PGID); transfer_pid(leader, tsk, PIDTYPE_SID); |
9cd80bbb0
|
845 |
|
aafe6c2a2
|
846 |
list_replace_rcu(&leader->tasks, &tsk->tasks); |
9cd80bbb0
|
847 |
list_replace_init(&leader->sibling, &tsk->sibling); |
1da177e4c
|
848 |
|
aafe6c2a2
|
849 850 |
tsk->group_leader = tsk; leader->group_leader = tsk; |
de12a7878
|
851 |
|
aafe6c2a2
|
852 |
tsk->exit_signal = SIGCHLD; |
962b564cf
|
853 854 855 |
BUG_ON(leader->exit_state != EXIT_ZOMBIE); leader->exit_state = EXIT_DEAD; |
1da177e4c
|
856 |
write_unlock_irq(&tasklist_lock); |
8187926bd
|
857 858 |
release_task(leader); |
ed5d2cac1
|
859 |
} |
1da177e4c
|
860 |
|
6db840fa7
|
861 862 |
sig->group_exit_task = NULL; sig->notify_count = 0; |
1da177e4c
|
863 864 |
no_thread_group: |
1f10206cf
|
865 866 |
if (current->mm) setmax_mm_hiwater_rss(&sig->maxrss, current->mm); |
1da177e4c
|
867 |
exit_itimers(sig); |
cbaffba12
|
868 |
flush_itimer_signals(); |
329f7dba5
|
869 |
|
b2c903b87
|
870 871 |
if (atomic_read(&oldsighand->count) != 1) { struct sighand_struct *newsighand; |
1da177e4c
|
872 |
/* |
b2c903b87
|
873 874 |
* This ->sighand is shared with the CLONE_SIGHAND * but not CLONE_THREAD task, switch to the new one. |
1da177e4c
|
875 |
*/ |
b2c903b87
|
876 877 878 |
newsighand = kmem_cache_alloc(sighand_cachep, GFP_KERNEL); if (!newsighand) return -ENOMEM; |
1da177e4c
|
879 880 881 882 883 884 |
atomic_set(&newsighand->count, 1); memcpy(newsighand->action, oldsighand->action, sizeof(newsighand->action)); write_lock_irq(&tasklist_lock); spin_lock(&oldsighand->siglock); |
aafe6c2a2
|
885 |
rcu_assign_pointer(tsk->sighand, newsighand); |
1da177e4c
|
886 887 |
spin_unlock(&oldsighand->siglock); write_unlock_irq(&tasklist_lock); |
fba2afaae
|
888 |
__cleanup_sighand(oldsighand); |
1da177e4c
|
889 |
} |
aafe6c2a2
|
890 |
BUG_ON(!thread_group_leader(tsk)); |
1da177e4c
|
891 892 |
return 0; } |
0840a90d9
|
893 |
|
1da177e4c
|
894 895 896 897 |
/* * These functions flushes out all traces of the currently running executable * so that a new one can be started */ |
858119e15
|
898 |
static void flush_old_files(struct files_struct * files) |
1da177e4c
|
899 900 |
{ long j = -1; |
badf16621
|
901 |
struct fdtable *fdt; |
1da177e4c
|
902 903 904 905 906 907 908 |
spin_lock(&files->file_lock); for (;;) { unsigned long set, i; j++; i = j * __NFDBITS; |
badf16621
|
909 |
fdt = files_fdtable(files); |
bbea9f696
|
910 |
if (i >= fdt->max_fds) |
1da177e4c
|
911 |
break; |
badf16621
|
912 |
set = fdt->close_on_exec->fds_bits[j]; |
1da177e4c
|
913 914 |
if (!set) continue; |
badf16621
|
915 |
fdt->close_on_exec->fds_bits[j] = 0; |
1da177e4c
|
916 917 918 919 920 921 922 923 924 925 926 |
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); } |
59714d65d
|
927 |
char *get_task_comm(char *buf, struct task_struct *tsk) |
1da177e4c
|
928 929 930 931 932 |
{ /* buf must be at least sizeof(tsk->comm) in size */ task_lock(tsk); strncpy(buf, tsk->comm, sizeof(tsk->comm)); task_unlock(tsk); |
59714d65d
|
933 |
return buf; |
1da177e4c
|
934 935 936 937 938 |
} void set_task_comm(struct task_struct *tsk, char *buf) { task_lock(tsk); |
4614a696b
|
939 940 941 942 943 944 945 946 947 |
/* * Threads may access current->comm without holding * the task lock, so write the string carefully. * Readers without a lock may see incomplete new * names but are safe from non-terminating string reads. */ memset(tsk->comm, 0, TASK_COMM_LEN); wmb(); |
1da177e4c
|
948 949 |
strlcpy(tsk->comm, buf, sizeof(tsk->comm)); task_unlock(tsk); |
cdd6c482c
|
950 |
perf_event_comm(tsk); |
1da177e4c
|
951 952 953 954 |
} int flush_old_exec(struct linux_binprm * bprm) { |
221af7f87
|
955 |
int retval; |
1da177e4c
|
956 957 958 959 960 961 962 963 |
/* * 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; |
925d1c401
|
964 |
set_mm_exe_file(bprm->mm, bprm->file); |
1da177e4c
|
965 |
/* |
1da177e4c
|
966 967 968 969 |
* Release all of the old mmap stuff */ retval = exec_mmap(bprm->mm); if (retval) |
fd8328be8
|
970 |
goto out; |
1da177e4c
|
971 972 |
bprm->mm = NULL; /* We're using it now */ |
7ab02af42
|
973 974 975 976 |
current->flags &= ~PF_RANDOMIZE; flush_thread(); current->personality &= ~bprm->per_clear; |
221af7f87
|
977 978 979 980 981 982 983 984 985 986 |
return 0; out: return retval; } EXPORT_SYMBOL(flush_old_exec); void setup_new_exec(struct linux_binprm * bprm) { int i, ch; |
d7627467b
|
987 |
const char *name; |
221af7f87
|
988 989 990 |
char tcomm[sizeof(current->comm)]; arch_pick_mmap_layout(current->mm); |
1da177e4c
|
991 992 |
/* This is the point of no return */ |
1da177e4c
|
993 |
current->sas_ss_sp = current->sas_ss_size = 0; |
da9592ede
|
994 |
if (current_euid() == current_uid() && current_egid() == current_gid()) |
6c5d52382
|
995 |
set_dumpable(current->mm, 1); |
d6e711448
|
996 |
else |
6c5d52382
|
997 |
set_dumpable(current->mm, suid_dumpable); |
d6e711448
|
998 |
|
1da177e4c
|
999 |
name = bprm->filename; |
367720923
|
1000 1001 |
/* Copies the binary name from after last slash */ |
1da177e4c
|
1002 1003 |
for (i=0; (ch = *(name++)) != '\0';) { if (ch == '/') |
367720923
|
1004 |
i = 0; /* overwrite what we wrote */ |
1da177e4c
|
1005 1006 1007 1008 1009 1010 |
else if (i < (sizeof(tcomm) - 1)) tcomm[i++] = ch; } tcomm[i] = '\0'; set_task_comm(current, tcomm); |
0551fbd29
|
1011 1012 1013 1014 1015 |
/* 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; |
a6f76f23d
|
1016 1017 1018 |
/* install the new credentials */ if (bprm->cred->uid != current_euid() || bprm->cred->gid != current_egid()) { |
d2d56c5f5
|
1019 1020 |
current->pdeath_signal = 0; } else if (file_permission(bprm->file, MAY_READ) || |
a6f76f23d
|
1021 |
bprm->interp_flags & BINPRM_FLAGS_ENFORCE_NONDUMP) { |
6c5d52382
|
1022 |
set_dumpable(current->mm, suid_dumpable); |
1da177e4c
|
1023 |
} |
f65cb45cb
|
1024 1025 1026 1027 1028 |
/* * Flush performance counters when crossing a * security domain: */ if (!get_dumpable(current->mm)) |
cdd6c482c
|
1029 |
perf_event_exit_task(current); |
f65cb45cb
|
1030 |
|
1da177e4c
|
1031 1032 1033 1034 1035 1036 1037 |
/* 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); |
1da177e4c
|
1038 |
} |
221af7f87
|
1039 |
EXPORT_SYMBOL(setup_new_exec); |
1da177e4c
|
1040 |
|
a6f76f23d
|
1041 |
/* |
a2a8474c3
|
1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 |
* Prepare credentials and lock ->cred_guard_mutex. * install_exec_creds() commits the new creds and drops the lock. * Or, if exec fails before, free_bprm() should release ->cred and * and unlock. */ int prepare_bprm_creds(struct linux_binprm *bprm) { if (mutex_lock_interruptible(¤t->cred_guard_mutex)) return -ERESTARTNOINTR; bprm->cred = prepare_exec_creds(); if (likely(bprm->cred)) return 0; mutex_unlock(¤t->cred_guard_mutex); return -ENOMEM; } void free_bprm(struct linux_binprm *bprm) { free_arg_pages(bprm); if (bprm->cred) { mutex_unlock(¤t->cred_guard_mutex); abort_creds(bprm->cred); } kfree(bprm); } /* |
a6f76f23d
|
1071 1072 1073 1074 1075 1076 1077 1078 |
* install the new credentials for this executable */ void install_exec_creds(struct linux_binprm *bprm) { security_bprm_committing_creds(bprm); commit_creds(bprm->cred); bprm->cred = NULL; |
a2a8474c3
|
1079 1080 |
/* * cred_guard_mutex must be held at least to this point to prevent |
a6f76f23d
|
1081 |
* ptrace_attach() from altering our determination of the task's |
a2a8474c3
|
1082 1083 |
* credentials; any time after this it may be unlocked. */ |
a6f76f23d
|
1084 |
security_bprm_committed_creds(bprm); |
a2a8474c3
|
1085 |
mutex_unlock(¤t->cred_guard_mutex); |
a6f76f23d
|
1086 1087 1088 1089 1090 |
} EXPORT_SYMBOL(install_exec_creds); /* * determine how safe it is to execute the proposed program |
5e751e992
|
1091 |
* - the caller must hold current->cred_guard_mutex to protect against |
a6f76f23d
|
1092 1093 |
* PTRACE_ATTACH */ |
498052bba
|
1094 |
int check_unsafe_exec(struct linux_binprm *bprm) |
a6f76f23d
|
1095 |
{ |
0bf2f3aec
|
1096 |
struct task_struct *p = current, *t; |
f1191b50e
|
1097 |
unsigned n_fs; |
498052bba
|
1098 |
int res = 0; |
a6f76f23d
|
1099 1100 |
bprm->unsafe = tracehook_unsafe_exec(p); |
0bf2f3aec
|
1101 |
n_fs = 1; |
2a4419b5b
|
1102 |
spin_lock(&p->fs->lock); |
437f7fdb6
|
1103 |
rcu_read_lock(); |
0bf2f3aec
|
1104 1105 1106 |
for (t = next_thread(p); t != p; t = next_thread(t)) { if (t->fs == p->fs) n_fs++; |
0bf2f3aec
|
1107 |
} |
437f7fdb6
|
1108 |
rcu_read_unlock(); |
0bf2f3aec
|
1109 |
|
f1191b50e
|
1110 |
if (p->fs->users > n_fs) { |
a6f76f23d
|
1111 |
bprm->unsafe |= LSM_UNSAFE_SHARE; |
498052bba
|
1112 |
} else { |
8c652f96d
|
1113 1114 1115 1116 1117 |
res = -EAGAIN; if (!p->fs->in_exec) { p->fs->in_exec = 1; res = 1; } |
498052bba
|
1118 |
} |
2a4419b5b
|
1119 |
spin_unlock(&p->fs->lock); |
498052bba
|
1120 1121 |
return res; |
a6f76f23d
|
1122 |
} |
1da177e4c
|
1123 1124 1125 |
/* * Fill the binprm structure from the inode. * Check permissions, then read the first 128 (BINPRM_BUF_SIZE) bytes |
a6f76f23d
|
1126 1127 |
* * This may be called multiple times for binary chains (scripts for example). |
1da177e4c
|
1128 1129 1130 |
*/ int prepare_binprm(struct linux_binprm *bprm) { |
a6f76f23d
|
1131 |
umode_t mode; |
0f7fc9e4d
|
1132 |
struct inode * inode = bprm->file->f_path.dentry->d_inode; |
1da177e4c
|
1133 1134 1135 |
int retval; mode = inode->i_mode; |
1da177e4c
|
1136 1137 |
if (bprm->file->f_op == NULL) return -EACCES; |
a6f76f23d
|
1138 1139 1140 |
/* clear any previous set[ug]id data from a previous binary */ bprm->cred->euid = current_euid(); bprm->cred->egid = current_egid(); |
1da177e4c
|
1141 |
|
a6f76f23d
|
1142 |
if (!(bprm->file->f_path.mnt->mnt_flags & MNT_NOSUID)) { |
1da177e4c
|
1143 1144 |
/* Set-uid? */ if (mode & S_ISUID) { |
a6f76f23d
|
1145 1146 |
bprm->per_clear |= PER_CLEAR_ON_SETID; bprm->cred->euid = inode->i_uid; |
1da177e4c
|
1147 1148 1149 1150 1151 1152 1153 1154 1155 |
} /* 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)) { |
a6f76f23d
|
1156 1157 |
bprm->per_clear |= PER_CLEAR_ON_SETID; bprm->cred->egid = inode->i_gid; |
1da177e4c
|
1158 1159 1160 1161 |
} } /* fill in binprm security blob */ |
a6f76f23d
|
1162 |
retval = security_bprm_set_creds(bprm); |
1da177e4c
|
1163 1164 |
if (retval) return retval; |
a6f76f23d
|
1165 |
bprm->cred_prepared = 1; |
1da177e4c
|
1166 |
|
a6f76f23d
|
1167 1168 |
memset(bprm->buf, 0, BINPRM_BUF_SIZE); return kernel_read(bprm->file, 0, bprm->buf, BINPRM_BUF_SIZE); |
1da177e4c
|
1169 1170 1171 |
} EXPORT_SYMBOL(prepare_binprm); |
4fc75ff48
|
1172 1173 1174 1175 1176 |
/* * 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. */ |
b6a2fea39
|
1177 |
int remove_arg_zero(struct linux_binprm *bprm) |
1da177e4c
|
1178 |
{ |
b6a2fea39
|
1179 1180 1181 1182 |
int ret = 0; unsigned long offset; char *kaddr; struct page *page; |
4fc75ff48
|
1183 |
|
b6a2fea39
|
1184 1185 |
if (!bprm->argc) return 0; |
1da177e4c
|
1186 |
|
b6a2fea39
|
1187 1188 1189 1190 1191 1192 1193 1194 |
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); |
4fc75ff48
|
1195 |
|
b6a2fea39
|
1196 1197 1198 |
for (; offset < PAGE_SIZE && kaddr[offset]; offset++, bprm->p++) ; |
4fc75ff48
|
1199 |
|
b6a2fea39
|
1200 1201 |
kunmap_atomic(kaddr, KM_USER0); put_arg_page(page); |
4fc75ff48
|
1202 |
|
b6a2fea39
|
1203 1204 1205 |
if (offset == PAGE_SIZE) free_arg_page(bprm, (bprm->p >> PAGE_SHIFT) - 1); } while (offset == PAGE_SIZE); |
4fc75ff48
|
1206 |
|
b6a2fea39
|
1207 1208 1209 |
bprm->p++; bprm->argc--; ret = 0; |
4fc75ff48
|
1210 |
|
b6a2fea39
|
1211 1212 |
out: return ret; |
1da177e4c
|
1213 |
} |
1da177e4c
|
1214 1215 1216 1217 1218 1219 1220 |
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) { |
85f334666
|
1221 |
unsigned int depth = bprm->recursion_depth; |
1da177e4c
|
1222 1223 |
int try,retval; struct linux_binfmt *fmt; |
1da177e4c
|
1224 |
|
1da177e4c
|
1225 1226 1227 1228 1229 1230 1231 |
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); |
473ae30bc
|
1232 1233 1234 1235 |
retval = audit_bprm(bprm); if (retval) return retval; |
1da177e4c
|
1236 1237 1238 |
retval = -ENOENT; for (try=0; try<2; try++) { read_lock(&binfmt_lock); |
e4dc1b14d
|
1239 |
list_for_each_entry(fmt, &formats, lh) { |
1da177e4c
|
1240 1241 1242 1243 1244 1245 1246 |
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); |
85f334666
|
1247 1248 1249 1250 1251 1252 |
/* * Restore the depth counter to its starting value * in this call, so we don't have to rely on every * load_binary function to restore it on return. */ bprm->recursion_depth = depth; |
1da177e4c
|
1253 |
if (retval >= 0) { |
85f334666
|
1254 1255 |
if (depth == 0) tracehook_report_exec(fmt, bprm, regs); |
1da177e4c
|
1256 1257 1258 1259 1260 1261 |
put_binfmt(fmt); allow_write_access(bprm->file); if (bprm->file) fput(bprm->file); bprm->file = NULL; current->did_exec = 1; |
9f46080c4
|
1262 |
proc_exec_connector(current); |
1da177e4c
|
1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 |
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; |
5f4123be3
|
1277 1278 |
#ifdef CONFIG_MODULES } else { |
1da177e4c
|
1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 |
#define printable(c) (((c)=='\t') || ((c)==' ') || (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); /* * sys_execve() executes a new program. */ |
d7627467b
|
1298 1299 1300 |
int do_execve(const char * filename, const char __user *const __user *argv, const char __user *const __user *envp, |
1da177e4c
|
1301 1302 1303 1304 |
struct pt_regs * regs) { struct linux_binprm *bprm; struct file *file; |
3b1253880
|
1305 |
struct files_struct *displaced; |
8c652f96d
|
1306 |
bool clear_in_exec; |
1da177e4c
|
1307 |
int retval; |
1da177e4c
|
1308 |
|
3b1253880
|
1309 |
retval = unshare_files(&displaced); |
fd8328be8
|
1310 1311 |
if (retval) goto out_ret; |
1da177e4c
|
1312 |
retval = -ENOMEM; |
11b0b5abb
|
1313 |
bprm = kzalloc(sizeof(*bprm), GFP_KERNEL); |
1da177e4c
|
1314 |
if (!bprm) |
fd8328be8
|
1315 |
goto out_files; |
1da177e4c
|
1316 |
|
a2a8474c3
|
1317 1318 |
retval = prepare_bprm_creds(bprm); if (retval) |
a6f76f23d
|
1319 |
goto out_free; |
498052bba
|
1320 1321 |
retval = check_unsafe_exec(bprm); |
8c652f96d
|
1322 |
if (retval < 0) |
a2a8474c3
|
1323 |
goto out_free; |
8c652f96d
|
1324 |
clear_in_exec = retval; |
a2a8474c3
|
1325 |
current->in_execve = 1; |
a6f76f23d
|
1326 |
|
1da177e4c
|
1327 1328 1329 |
file = open_exec(filename); retval = PTR_ERR(file); if (IS_ERR(file)) |
498052bba
|
1330 |
goto out_unmark; |
1da177e4c
|
1331 1332 |
sched_exec(); |
1da177e4c
|
1333 1334 1335 |
bprm->file = file; bprm->filename = filename; bprm->interp = filename; |
1da177e4c
|
1336 |
|
b6a2fea39
|
1337 1338 1339 |
retval = bprm_mm_init(bprm); if (retval) goto out_file; |
1da177e4c
|
1340 |
|
b6a2fea39
|
1341 |
bprm->argc = count(argv, MAX_ARG_STRINGS); |
1da177e4c
|
1342 |
if ((retval = bprm->argc) < 0) |
a6f76f23d
|
1343 |
goto out; |
1da177e4c
|
1344 |
|
b6a2fea39
|
1345 |
bprm->envc = count(envp, MAX_ARG_STRINGS); |
1da177e4c
|
1346 |
if ((retval = bprm->envc) < 0) |
1da177e4c
|
1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 |
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; |
7b34e4283
|
1365 |
current->flags &= ~PF_KTHREAD; |
1da177e4c
|
1366 |
retval = search_binary_handler(bprm,regs); |
a6f76f23d
|
1367 1368 |
if (retval < 0) goto out; |
1da177e4c
|
1369 |
|
a6f76f23d
|
1370 |
/* execve succeeded */ |
498052bba
|
1371 |
current->fs->in_exec = 0; |
f9ce1f1cd
|
1372 |
current->in_execve = 0; |
a6f76f23d
|
1373 1374 1375 1376 1377 |
acct_update_integrals(current); free_bprm(bprm); if (displaced) put_files_struct(displaced); return retval; |
1da177e4c
|
1378 |
|
a6f76f23d
|
1379 |
out: |
1da177e4c
|
1380 |
if (bprm->mm) |
b6a2fea39
|
1381 |
mmput (bprm->mm); |
1da177e4c
|
1382 1383 1384 1385 1386 1387 |
out_file: if (bprm->file) { allow_write_access(bprm->file); fput(bprm->file); } |
a6f76f23d
|
1388 |
|
498052bba
|
1389 |
out_unmark: |
8c652f96d
|
1390 1391 |
if (clear_in_exec) current->fs->in_exec = 0; |
f9ce1f1cd
|
1392 |
current->in_execve = 0; |
a6f76f23d
|
1393 1394 |
out_free: |
08a6fac1c
|
1395 |
free_bprm(bprm); |
1da177e4c
|
1396 |
|
fd8328be8
|
1397 |
out_files: |
3b1253880
|
1398 1399 |
if (displaced) reset_files_struct(displaced); |
1da177e4c
|
1400 1401 1402 |
out_ret: return retval; } |
964ee7df9
|
1403 |
void set_binfmt(struct linux_binfmt *new) |
1da177e4c
|
1404 |
{ |
801460d0c
|
1405 1406 1407 1408 |
struct mm_struct *mm = current->mm; if (mm->binfmt) module_put(mm->binfmt->module); |
1da177e4c
|
1409 |
|
801460d0c
|
1410 |
mm->binfmt = new; |
964ee7df9
|
1411 1412 |
if (new) __module_get(new->module); |
1da177e4c
|
1413 1414 1415 |
} EXPORT_SYMBOL(set_binfmt); |
1da177e4c
|
1416 1417 1418 1419 |
/* 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. */ |
6409324b3
|
1420 |
static int format_corename(char *corename, long signr) |
1da177e4c
|
1421 |
{ |
86a264abe
|
1422 |
const struct cred *cred = current_cred(); |
565b9b14e
|
1423 1424 |
const char *pat_ptr = core_pattern; int ispipe = (*pat_ptr == '|'); |
1da177e4c
|
1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 |
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, |
b488893a3
|
1451 |
"%d", task_tgid_vnr(current)); |
1da177e4c
|
1452 1453 1454 1455 1456 1457 1458 |
if (rc > out_end - out_ptr) goto out; out_ptr += rc; break; /* uid */ case 'u': rc = snprintf(out_ptr, out_end - out_ptr, |
86a264abe
|
1459 |
"%d", cred->uid); |
1da177e4c
|
1460 1461 1462 1463 1464 1465 1466 |
if (rc > out_end - out_ptr) goto out; out_ptr += rc; break; /* gid */ case 'g': rc = snprintf(out_ptr, out_end - out_ptr, |
86a264abe
|
1467 |
"%d", cred->gid); |
1da177e4c
|
1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 |
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, |
e9ff3990f
|
1495 |
"%s", utsname()->nodename); |
1da177e4c
|
1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 |
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; |
74aadce98
|
1509 1510 1511 |
/* core limit size */ case 'c': rc = snprintf(out_ptr, out_end - out_ptr, |
d554ed895
|
1512 |
"%lu", rlimit(RLIMIT_CORE)); |
74aadce98
|
1513 1514 1515 1516 |
if (rc > out_end - out_ptr) goto out; out_ptr += rc; break; |
1da177e4c
|
1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 |
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 |
c4bbafda7
|
1527 |
* the filename. Do not do this for piped commands. */ |
6409324b3
|
1528 |
if (!ispipe && !pid_in_pattern && core_uses_pid) { |
1da177e4c
|
1529 |
rc = snprintf(out_ptr, out_end - out_ptr, |
b488893a3
|
1530 |
".%d", task_tgid_vnr(current)); |
1da177e4c
|
1531 1532 1533 1534 |
if (rc > out_end - out_ptr) goto out; out_ptr += rc; } |
c4bbafda7
|
1535 |
out: |
1da177e4c
|
1536 |
*out_ptr = 0; |
c4bbafda7
|
1537 |
return ispipe; |
1da177e4c
|
1538 |
} |
5c99cbf49
|
1539 |
static int zap_process(struct task_struct *start, int exit_code) |
aceecc041
|
1540 1541 |
{ struct task_struct *t; |
8cd9c2491
|
1542 |
int nr = 0; |
281de339c
|
1543 |
|
d5f70c00a
|
1544 |
start->signal->flags = SIGNAL_GROUP_EXIT; |
5c99cbf49
|
1545 |
start->signal->group_exit_code = exit_code; |
d5f70c00a
|
1546 |
start->signal->group_stop_count = 0; |
aceecc041
|
1547 1548 1549 1550 |
t = start; do { if (t != current && t->mm) { |
281de339c
|
1551 1552 |
sigaddset(&t->pending.signal, SIGKILL); signal_wake_up(t, 1); |
8cd9c2491
|
1553 |
nr++; |
aceecc041
|
1554 |
} |
e4901f92a
|
1555 |
} while_each_thread(start, t); |
8cd9c2491
|
1556 1557 |
return nr; |
aceecc041
|
1558 |
} |
dcf560c59
|
1559 |
static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm, |
8cd9c2491
|
1560 |
struct core_state *core_state, int exit_code) |
1da177e4c
|
1561 1562 |
{ struct task_struct *g, *p; |
5debfa6da
|
1563 |
unsigned long flags; |
8cd9c2491
|
1564 |
int nr = -EAGAIN; |
dcf560c59
|
1565 1566 |
spin_lock_irq(&tsk->sighand->siglock); |
ed5d2cac1
|
1567 |
if (!signal_group_exit(tsk->signal)) { |
8cd9c2491
|
1568 |
mm->core_state = core_state; |
5c99cbf49
|
1569 |
nr = zap_process(tsk, exit_code); |
1da177e4c
|
1570 |
} |
dcf560c59
|
1571 |
spin_unlock_irq(&tsk->sighand->siglock); |
8cd9c2491
|
1572 1573 |
if (unlikely(nr < 0)) return nr; |
1da177e4c
|
1574 |
|
8cd9c2491
|
1575 |
if (atomic_read(&mm->mm_users) == nr + 1) |
5debfa6da
|
1576 |
goto done; |
e4901f92a
|
1577 1578 |
/* * We should find and kill all tasks which use this mm, and we should |
999d9fc16
|
1579 |
* count them correctly into ->nr_threads. We don't take tasklist |
e4901f92a
|
1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 |
* 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, ¤t->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(). */ |
7b1c6154f
|
1607 |
rcu_read_lock(); |
aceecc041
|
1608 |
for_each_process(g) { |
5debfa6da
|
1609 1610 |
if (g == tsk->group_leader) continue; |
15b9f360c
|
1611 1612 |
if (g->flags & PF_KTHREAD) continue; |
aceecc041
|
1613 1614 1615 |
p = g; do { if (p->mm) { |
15b9f360c
|
1616 |
if (unlikely(p->mm == mm)) { |
5debfa6da
|
1617 |
lock_task_sighand(p, &flags); |
5c99cbf49
|
1618 |
nr += zap_process(p, exit_code); |
5debfa6da
|
1619 1620 |
unlock_task_sighand(p, &flags); } |
aceecc041
|
1621 1622 |
break; } |
e4901f92a
|
1623 |
} while_each_thread(g, p); |
aceecc041
|
1624 |
} |
7b1c6154f
|
1625 |
rcu_read_unlock(); |
5debfa6da
|
1626 |
done: |
c5f1cc8c1
|
1627 |
atomic_set(&core_state->nr_threads, nr); |
8cd9c2491
|
1628 |
return nr; |
1da177e4c
|
1629 |
} |
9d5b327bf
|
1630 |
static int coredump_wait(int exit_code, struct core_state *core_state) |
1da177e4c
|
1631 |
{ |
dcf560c59
|
1632 1633 |
struct task_struct *tsk = current; struct mm_struct *mm = tsk->mm; |
dcf560c59
|
1634 |
struct completion *vfork_done; |
269b005a2
|
1635 |
int core_waiters = -EBUSY; |
1da177e4c
|
1636 |
|
9d5b327bf
|
1637 |
init_completion(&core_state->startup); |
b564daf80
|
1638 1639 |
core_state->dumper.task = tsk; core_state->dumper.next = NULL; |
269b005a2
|
1640 1641 1642 1643 |
down_write(&mm->mmap_sem); if (!mm->core_state) core_waiters = zap_threads(tsk, mm, core_state, exit_code); |
2384f55f8
|
1644 |
up_write(&mm->mmap_sem); |
dcf560c59
|
1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 |
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); } |
2384f55f8
|
1657 |
if (core_waiters) |
9d5b327bf
|
1658 |
wait_for_completion(&core_state->startup); |
dcf560c59
|
1659 |
fail: |
dcf560c59
|
1660 |
return core_waiters; |
1da177e4c
|
1661 |
} |
a94e2d408
|
1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 |
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; } |
6c5d52382
|
1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 |
/* * 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; } } |
6c5d52382
|
1722 |
|
30736a4d4
|
1723 |
static int __get_dumpable(unsigned long mm_flags) |
6c5d52382
|
1724 1725 |
{ int ret; |
30736a4d4
|
1726 |
ret = mm_flags & MMF_DUMPABLE_MASK; |
6c5d52382
|
1727 1728 |
return (ret >= 2) ? 2 : ret; } |
30736a4d4
|
1729 1730 1731 1732 |
int get_dumpable(struct mm_struct *mm) { return __get_dumpable(mm->flags); } |
61be228a0
|
1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 |
static void wait_for_dump_helpers(struct file *file) { struct pipe_inode_info *pipe; pipe = file->f_path.dentry->d_inode->i_pipe; pipe_lock(pipe); pipe->readers++; pipe->writers--; while ((pipe->readers > 1) && (!signal_pending(current))) { wake_up_interruptible_sync(&pipe->wait); kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN); pipe_wait(pipe); } pipe->readers--; pipe->writers++; pipe_unlock(pipe); } |
898b374af
|
1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 |
/* * uhm_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 file *rp, *wp; struct fdtable *fdt; struct coredump_params *cp = (struct coredump_params *)info->data; struct files_struct *cf = current->files; wp = create_write_pipe(0); if (IS_ERR(wp)) return PTR_ERR(wp); rp = create_read_pipe(wp, 0); if (IS_ERR(rp)) { free_write_pipe(wp); return PTR_ERR(rp); } cp->file = wp; sys_close(0); fd_install(0, rp); spin_lock(&cf->file_lock); fdt = files_fdtable(cf); FD_SET(0, fdt->open_fds); FD_CLR(0, fdt->close_on_exec); spin_unlock(&cf->file_lock); /* and disallow core files too */ current->signal->rlim[RLIMIT_CORE] = (struct rlimit){1, 1}; return 0; } |
8cd3ac3ac
|
1797 |
void do_coredump(long signr, int exit_code, struct pt_regs *regs) |
1da177e4c
|
1798 |
{ |
9d5b327bf
|
1799 |
struct core_state core_state; |
1da177e4c
|
1800 1801 1802 |
char corename[CORENAME_MAX_SIZE + 1]; struct mm_struct *mm = current->mm; struct linux_binfmt * binfmt; |
d84f4f992
|
1803 1804 |
const struct cred *old_cred; struct cred *cred; |
1da177e4c
|
1805 |
int retval = 0; |
d6e711448
|
1806 |
int flag = 0; |
d5bf4c4f5
|
1807 |
int ispipe; |
a293980c2
|
1808 |
static atomic_t core_dump_count = ATOMIC_INIT(0); |
f6151dfea
|
1809 1810 1811 |
struct coredump_params cprm = { .signr = signr, .regs = regs, |
d554ed895
|
1812 |
.limit = rlimit(RLIMIT_CORE), |
30736a4d4
|
1813 1814 1815 1816 1817 1818 |
/* * 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, |
f6151dfea
|
1819 |
}; |
1da177e4c
|
1820 |
|
0a4ff8c25
|
1821 |
audit_core_dumps(signr); |
801460d0c
|
1822 |
binfmt = mm->binfmt; |
1da177e4c
|
1823 1824 |
if (!binfmt || !binfmt->core_dump) goto fail; |
269b005a2
|
1825 1826 |
if (!__get_dumpable(cprm.mm_flags)) goto fail; |
d84f4f992
|
1827 1828 |
cred = prepare_creds(); |
5e43aef53
|
1829 |
if (!cred) |
d84f4f992
|
1830 |
goto fail; |
d6e711448
|
1831 1832 1833 1834 1835 |
/* * 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. */ |
30736a4d4
|
1836 1837 |
if (__get_dumpable(cprm.mm_flags) == 2) { /* Setuid core dump mode */ |
d6e711448
|
1838 |
flag = O_EXCL; /* Stop rewrite attacks */ |
d84f4f992
|
1839 |
cred->fsuid = 0; /* Dump root private */ |
d6e711448
|
1840 |
} |
1291cf416
|
1841 |
|
9d5b327bf
|
1842 |
retval = coredump_wait(exit_code, &core_state); |
5e43aef53
|
1843 1844 |
if (retval < 0) goto fail_creds; |
d84f4f992
|
1845 1846 |
old_cred = override_creds(cred); |
1da177e4c
|
1847 1848 1849 1850 1851 |
/* * Clear any false indication of pending signals that might * be seen by the filesystem code called to write the core file. */ |
1da177e4c
|
1852 |
clear_thread_flag(TIF_SIGPENDING); |
6409324b3
|
1853 |
ispipe = format_corename(corename, signr); |
725eae32d
|
1854 |
|
c4bbafda7
|
1855 |
if (ispipe) { |
d5bf4c4f5
|
1856 1857 |
int dump_count; char **helper_argv; |
898b374af
|
1858 |
if (cprm.limit == 1) { |
725eae32d
|
1859 1860 1861 |
/* * Normally core limits are irrelevant to pipes, since * we're not writing to the file system, but we use |
898b374af
|
1862 1863 |
* cprm.limit of 1 here as a speacial value. Any * non-1 limit gets set to RLIM_INFINITY below, but |
725eae32d
|
1864 1865 |
* a limit of 0 skips the dump. This is a consistent * way to catch recursive crashes. We can still crash |
898b374af
|
1866 |
* if the core_pattern binary sets RLIM_CORE = !1 |
725eae32d
|
1867 1868 1869 1870 1871 1872 1873 |
* 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 |
898b374af
|
1874 1875 |
"Process %d(%s) has RLIMIT_CORE set to 1 ", |
725eae32d
|
1876 1877 1878 1879 1880 |
task_tgid_vnr(current), current->comm); printk(KERN_WARNING "Aborting core "); goto fail_unlock; } |
d5bf4c4f5
|
1881 |
cprm.limit = RLIM_INFINITY; |
725eae32d
|
1882 |
|
a293980c2
|
1883 1884 1885 1886 1887 1888 1889 1890 1891 |
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 ", task_tgid_vnr(current), current->comm); printk(KERN_WARNING "Skipping core dump "); goto fail_dropcount; } |
d5bf4c4f5
|
1892 |
helper_argv = argv_split(GFP_KERNEL, corename+1, NULL); |
350eaf791
|
1893 1894 1895 1896 |
if (!helper_argv) { printk(KERN_WARNING "%s failed to allocate memory ", __func__); |
a293980c2
|
1897 |
goto fail_dropcount; |
350eaf791
|
1898 |
} |
323211371
|
1899 |
|
d5bf4c4f5
|
1900 1901 1902 1903 1904 |
retval = call_usermodehelper_fns(helper_argv[0], helper_argv, NULL, UMH_WAIT_EXEC, umh_pipe_setup, NULL, &cprm); argv_free(helper_argv); if (retval) { |
d025c9db7
|
1905 1906 1907 |
printk(KERN_INFO "Core dump to %s pipe failed ", corename); |
d5bf4c4f5
|
1908 |
goto close_fail; |
d025c9db7
|
1909 |
} |
c71354112
|
1910 1911 1912 1913 1914 |
} else { struct inode *inode; if (cprm.limit < binfmt->min_coredump) goto fail_unlock; |
f6151dfea
|
1915 |
cprm.file = filp_open(corename, |
6d4df677f
|
1916 1917 |
O_CREAT | 2 | O_NOFOLLOW | O_LARGEFILE | flag, 0600); |
c71354112
|
1918 1919 |
if (IS_ERR(cprm.file)) goto fail_unlock; |
1da177e4c
|
1920 |
|
c71354112
|
1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 |
inode = cprm.file->f_path.dentry->d_inode; 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; /* * Dont allow local users get cute and trick others to coredump * into their pre-created files. */ if (inode->i_uid != current_fsuid()) goto close_fail; if (!cprm.file->f_op || !cprm.file->f_op->write) goto close_fail; if (do_truncate(cprm.file->f_path.dentry, 0, 0, cprm.file)) goto close_fail; } |
1da177e4c
|
1943 |
|
c71354112
|
1944 |
retval = binfmt->core_dump(&cprm); |
1da177e4c
|
1945 1946 |
if (retval) current->signal->group_exit_code |= 0x80; |
d5bf4c4f5
|
1947 |
|
61be228a0
|
1948 |
if (ispipe && core_pipe_limit) |
f6151dfea
|
1949 |
wait_for_dump_helpers(cprm.file); |
d5bf4c4f5
|
1950 1951 1952 |
close_fail: if (cprm.file) filp_close(cprm.file, NULL); |
a293980c2
|
1953 |
fail_dropcount: |
d5bf4c4f5
|
1954 |
if (ispipe) |
a293980c2
|
1955 |
atomic_dec(&core_dump_count); |
1da177e4c
|
1956 |
fail_unlock: |
5e43aef53
|
1957 |
coredump_finish(mm); |
d84f4f992
|
1958 |
revert_creds(old_cred); |
5e43aef53
|
1959 |
fail_creds: |
d84f4f992
|
1960 |
put_cred(cred); |
1da177e4c
|
1961 |
fail: |
8cd3ac3ac
|
1962 |
return; |
1da177e4c
|
1963 |
} |